Here - ECN

Transcription

Here - ECN
VRIJE UNIVERSITEIT
Methods for observation and quantification
of trace gas emissions from diffuse sources
ACADEMISCH PROEFSCHRIFT
ter verkrijging van de graad Doctor aan
de Vrije Universiteit Amsterdam,
op gezag van de rector magnificus
prof.dr. L.M. Bouter,
in het openbaar te verdedigen
ten overstaan van de promotiecommissie
van de faculteit der Aard- en Levenswetenschappen
op donderdag 12 januari 2012 om 15.45 uur
in de aula van de universiteit,
De Boelelaan 1105
door
Adrianus Hensen
geboren te Hilversum
1
promotor:
copromotor:
prof.dr. A.J. Dolman
prof.dr. J.W. Erisman
Members of the thesis comittee
Dr. J. Van Huissteden
VU Amsterdam
Prof. Dr. T Röckmann
Universiteit Utrecht
Prof. Dr. K Butterbach Bahl University of Freiburg & IMK-IFU
Dr. A. Freibauer
Johann Heinrich von Thünen-Institut
Prof.Dr. D. Fowler
Centre for Ecology and Hydrology, UK
This research was funded by:
Energy research Centre of the Netherlands (ECN)
The ministry of Environmental Affairs in the Netherlands
GRAMMINAE EU project
MIDAIR EU project
NITROEUROPE EU project
Afvalzorg N.V.
Colophon:
Printed by Ipskamp Drukkers B.V. on FCS certified paper.
Copyright © Chapter 3 Phys. Chem. Earth
Copyright © Chapter 4 Biogeosciences
Copyright © Chapter 5 Water,Air and Soil pollution
Copyright © Chapter 6 Biogeosciences
Copyright © Chapter 7 Agriculture, Ecosystems and Environment
Photo’s without reference:
Photo’s figure 2.6 & 2.7 (left) Alex Vermeulen
Photo’s figure 2.14 (left & mid) Afvalzorg N.V.
Photo’s page 53 & 82: Jan Willem Erisman
Other photo’s: Arjan Hensen
2
Summary
Methods for observation and quantification
of trace gas emissions from diffuse sources
This thesis is a synthesis of experimental studies that aims to contribute to the improvement of estimates of trace gas emissions from diffuse sources into the atmosphere. The term diffuse source is used to distinguish from sources that do have well
defined exhausts like for example chimneys or ventilators. Diffuse sources include:
• soils, either natural, agricultural or industrial (landfills);
• groups of sources within a confined area (farm with multiple animal houses, manure storages or industrial plants with a large number of point sources;
• regions: a group of farms; farms and industries, urban areas.
Work is presented for the main greenhouse gasses Carbon dioxide (CO2), Methane
(CH4) and Nitrous oxide (N2O). Ammonia (NH3) is subject for research, as an important
air pollutant that is mainly emitted from agricultural diffuse sources.
About 20% of the total GHG emission to the atmosphere is emitted from diffuse
sources either as CO2, CH4 or N2O. The uncertainty in the emission level from these
sources is a dominant component in the uncertainty of the national emission level for
the Netherlands. For ammonia 88% of all emissions in the Netherlands originate diffuse
source systems, mainly in the agronomic sector. In the 27 EU countries the rate is even
higher, 93%. Similar to the greenhouse gas case, the NH3 emission level on a national
and European scale is still far from certain.
The main reason for the large uncertainties in the emissions from diffuse sources is the
difficulty to quantitatively measure these emissions that can show large variations both
in time and in space. Depending on the trace gas species different sensors and different measurement methods are required to evaluate the emission levels. This thesis
gives an overview of the sources, the emission levels and the associated uncertainties.
Also currently available measurement sensors and methods are described.
Examples of studies that aimed to develop measurement strategies for diffuse source
systems are provided in the five publications that are part of this thesis. This set starts
with two papers on the micrometeorological technique known as the relaxed eddy accumulation technique (REA) which was evaluated for carbon dioxide (CO2) and for
ammonia (NH3). This technique measures the net vertical gas transport above the surface. The emission from a diffuse source system can also be evaluated remotely by
measurement of gas that is transported away from the source area with the wind in a
gas plume. Two papers discuss the use of this type of measurements for evaluation of
methane emissions from landfills and for the assessment of NH3 emissions from a farm
site in Germany. The last paper describes how the plume technique with mobile measurements was applied to quantify methane emissions from farms. After these examples
a synthesis is made both from the technical perspective and from the perspective of different source types.
General remarks
Diffuse source systems have challenged scientist for several decades. Although our
knowledge has improved considerably, science has still far from unravelled how diffuse
source systems work at the process level and how to quantitatively measure the net
emission on an annual basis on representative spatial scales (for example > 1ha ecosystem, full farm or full landfill). This holds for example for the net CO2 equivalent exchange after land-use change, for the direct and indirect N2O emissions from fields and
3
water bodies, for methane emissions from landfills and both for methane and NH3
emissions from livestock activities.
The uncertainty of net gas exchange between diffuse sources and the atmosphere is
partially explained by the large number of individual source systems in combination
with the fact that different geographical position, environment, weather and management can have a strong impact on the emission processes. The multitude of source locations makes it inevitable to extrapolate emission measurements. Results obtained at
a limited set of locations have to be used, to calculate the emission over the full extent
of the source type. This will require a combination of data and parameterizations in
computer models. Meteorological data, land use and management data statistics as
well as distributions in space and time of emission driving parameters are needed to do
this extrapolation.
The parameterizations used in the computer models have to be generated (or validated) with experimental work. Zooming in on the emission measurements, there are
two parts, the gas concentration measurement itself and the method used to translate
these measurements into a flux. These two parts are interlinked. Some methods, like
for example micrometeorological techniques, can only be carried out with fast response
sensors that can show small changes in ambient concentration levels. The slow but
continuous improvement of sensors makes currently used methods more accurate.
Apart from that also enables the development of innovative measurement strategies.
Conclusions from a technical perspective
With the current state of sensor techniques, the conclusion is that fundamental methodological issues for several measurement methods used are still not resolved. Two
examples are discussed:
• Mass balance measurements that are both underlying the national emission estimates for CH4 from landfills and NH3 from fields treated with manure have provided
overestimates (probably between 5-15%) in the results thus far. The reason is that
the available sensors used for this method cannot resolve the so called backward
turbulence correction needed for this method. When implemented, this correction
will always provide reduced flux levels.
• A similar reasoning holds for the box measurements that in general provide underestimates of the real flux by between 0-30%. An explanation for this is that the
available sensors are too slow. Most experiments are carried out at a time resolution that prevents the detection of non-linearity in the concentration increase under
a box. Consequently the bulk of the box measurements reported in literature used
linear interpolation to calculate the emission level. And these estimates are very
likely to be underestimations of the real flux.
These two “errors” have a very similar background. The need for correction in both
cases was already discussed 20 years ago. With available sensors however the correction can hardly be carried out. So almost no one uses it. This would not be much of
a problem when these errors were random. But unfortunately the corrections needed
lead to a systematic shift in the results. New sensors can now show and quantify these
errors. There are several recent publications discussing the box measurements. For
the mass balance method a similar discussion will have to start.
4
In this thesis three improvements to better understand emissions from diffuse source
systems were explicitly demonstrated.
• The fast box measurement that makes a combination of a simple box method with
an advanced concentration sensor (with high resolution both in concentration and
in time). This method helps to tackle spatial variability of the emissions within a
given source area. It also helps to show the non linearity of the concentration increase in box measurements discussed above.
• The plume measurement that was used to provide source integrated emission data
both for individual farms and for landfills.
• The relaxed eddy accumulation for flux measurements offers a temporal solution
for components were fast eddy covariance measurements are still not possible.
Conclusions from a source perspective
Important diffuse source systems that feature in this thesis are landfills, agricultural
fields and farms. For landfills, the data obtained with plume measurements can be considered important internationally. Thus far emission accounting methods for these
sources are not standardized. Within landfill research there is a hypothesis that the use
of different methods leads to a factor four difference between the methane emission
estimates for an average UK landfill compared with a similar Dutch landfill.
For agricultural fields, there is still a large uncertainty both in direct and indirect emissions of N2O. For these source systems both the fast box method and new eddy covariance data should provide better emission estimates. To evaluate the net greenhouse gas balance, there is an evident need to perform simultaneous and co-located
measurements for CO2, CH4 and N2O. This is now possible. It is remarkable that in the
Netherlands there is virtually no data with co-located NH3 and greenhouse gas measurements. This is strange because our country has the highest emission densities for
both greenhouse gases and NH3 in Europe.
As for farms the use of farm integrated emission measurements are still hardly used.
The plume measurement data obtained thus far did however contribute to the debate
on the CH4 emission factor for dairy cows. The result was a significant update of the
emission calculation method for this source.
There is no such thing as “the best method” to perform the emission evaluations for
any of these source types. There is even a potential danger of systematic biases when
using a single method and rely completely on more and more measurements of the
same type to reduce the uncertainty. On the other hand, comparability of different experiments is an advantage of using the “standard” methods. The best option at the
moment seems the use of a combination of techniques in order to better quantify the
emissions from diffuse source systems. Emphasis should be on further development of
the more integrating (micrometeorological) techniques to enable a significant reduction
in the uncertainty to whatever is agreed on as an acceptable level.
Finally
This thesis shows the slow but steady progress over almost two decennia in this field of
science. In the decades to come the key to further improvement is in new, improved or
significantly cheaper sensors that will enable further improvement available strategies
or even new measurement strategies. As for the national and global scale combination
of ground truth measurements and satellite retrievals will be a crucial step. For individual sources, further development of the plume methodology towards fence line monitoring systems is a likely development. The latter implies further development of low cost
sensors, lidar (light detection and ranging) systems or other remote techniques. The
fundamental errors for box and mass balance method of course have to be resolved as
5
soon as possible and if possible correction algorithms for historic data have to be developed.
This thesis wants to emphasise the importance or measurements and innovation of
measurement techniques. Measurements alone do not provide all the answers. Activity
data, modelling tools and measurements from a trinity needed to evaluate diffuse
source emission levels. Direct observations, however, are leading within this trinity. Improvement of sensors and measurement techniques are therefore important to better
quantify the emissions of trace gasses from diffuse source systems.
Arjan Hensen
December 2011
6
Samenvatting
Methoden voor waarneming en kwantificering van
emissies van sporegassen vanuit diffuse bronnen
Dit proefschrift is synthese op basis van een aantal experimentele studies die wil bijdragen aan de verbetering van schattingen van emissies van sporegassen vanuit diffuse bronnen naar de atmosfeer. De term diffuse bron wordt gebruikt voor bronnen die
geen goed gedefinieerde uitlaat hebben zoals schoorstenen of ventilatoren. Diffuse
bronnen zijn onder andere:
• bodems, hetzij natuurlijk, agrarisch of industrieel (stortplaatsen);
• groepen van bronnen binnen een beperkt gebied (boerderij met meerdere stallen,
mestopslagen of industriële installaties met een groot aantal puntbronnen;
• regio’s: een groep bedrijven, boerderijen en industrie, stedelijke gebieden.
Er wordt werk gepresenteerd voor de belangrijkste broeikasgassen kooldioxide (CO2),
methaan (CH4) en lachgas (N2O). Daarnaast is Ammoniak (NH3) onderwerp van onderzoek als een belangrijke component in luchtverontreiniging die voornamelijk wordt
uitgestoten door agrarische diffuse bronnen.
Ongeveer 20% van de totale broeikasgasuitstoot naar de atmosfeer komt uit diffuse
bronnen die CO2, CH4 en N2O uitstoten. De onzekerheid in de emissie vanuit deze
bronnen is een dominante component in de onzekerheid van het nationale emissieniveau voor Nederland. Voor ammoniak is 88% van de totale uitstoot in Nederland afkomstig van diffuse bronsystemen, voornamelijk in de agrarische sector. In de 27 EUlanden is dit percentage met 93% zelfs hoger. Net als in het geval van de broeikasgassen is ook het NH3 emissieniveau op nationale en Europese schaal nog steeds verre
van zeker.
De belangrijkste reden voor de grote onzekerheden in de emissies vanuit diffuse bronnen is dat het moeilijk is om goede metingen uit te voeren waarmee deze emissies, die
zowel in tijd als in ruimte sterk kunnen variëren, kwantitatief bepaald kunnen worden.
Afhankelijk van het type sporegas zijn verschillende sensoren en meetmethoden nodig
om het emissieniveau te beoordelen. Dit proefschrift bespreekt de bronnen, de emissieniveaus en de bijbehorende onzekerheden. Ook de sensoren en methoden die momenteel beschikbaar zijn worden kort beschreven.
Voorbeelden van studies die gericht zijn op de ontwikkeling van meetstrategieën voor
diffuse bronsystemen, worden besproken in de vijf publicaties die deel uitmaken van dit
proefschrift. Deze set start met de micrometeorologische techniek die bekend staat als
de “relaxed eddy accumulation”(REA) techniek. Deze werd geëvalueerd voor koolstofdioxide (CO2) en ammoniak (NH3). De REA techniek meet het netto vertikaal transport
van gas in de atmosfeer boven een oppervlak. De emissie vanuit een diffuse bron kan
ook op afstand bepaald worden door aan de gaspluim te meten die vanaf de bron met
de wind wordt meegevoerd. Twee publicaties beschrijven inzet van dit type metingen
voor evaluatie van de methaanemissie van stortplaatsen en voor de bepaling van de
NH3-emissie van een boerderij in Duitsland.
De laatste publicatie beschrijft hoe de pluimmeettechniek met de mobiele metingen
werd toegepast om de uitstoot van methaan uit boerderijen te bepalen. Na deze voorbeelden wordt een synthese gemaakt, zowel vanuit technisch oogpunt als vanuit de het
perspectief van verschillende soorten bronnen.
7
Algemene opmerkingen
Diffuse bronsystemen bieden al meerdere decennia een uitdaging voor wetenschappers. Hoewel onze kennis aanzienlijk is verbeterd, heeft de wetenschap nog lang niet
ontrafeld hoe diffuse bronsystemen op procesniveau werken en hoe de netto-uitstoot
op jaarbasis kwantitatief gemeten kan worden op een representatieve ruimtelijke
schaal (bijvoorbeeld >1 ha ecosysteem, een volledige boerderij of een volledig stortplaats). Dit geldt bijvoorbeeld voor de netto CO2-equivalent uitwisseling bij veranderingen in landgebruik, voor directe en indirecte N2O-emissies uit velden en water, voor
methaanemissies uit stortplaatsen en voor zowel CH4 als NH3 emissies vanuit veeteelt.
De onzekerheid in de netto uitwisseling van gassen tussen diffuse bronnen en de atmosfeer wordt gedeeltelijk verklaard door het grote aantal individuele bronsystemen in
combinatie met het feit dat verschillen in geografische ligging, milieu, weercondities en
beheer een sterke invloed kunnen hebben op de emissieprocessen. De veelheid van
bronlocaties maakt het onvermijdelijk dat emissiemetingen geëxtrapoleerd moeten
worden. Metingen uitgevoerd op een beperkte set locaties worden hierbij gebruikt om
de emissie uit te rekenen voor de volle omvang van een broncategorie. Dit vereist een
combinatie van (meet)gegevens en parameterisaties in computermodellen. Meteorologische gegevens, landgebruik en het managementgegevens maar ook verdeling in
ruimte en tijd van parameters die de emissie sturen zijn nodig om deze extrapolatie uit
te kunnen voeren.
De parameterisaties gebruikt in de computermodellen moeten worden gegenereerd (of
gevalideerd) met experimenteel werk (metingen). Wat betreft de emissiemetingen, zijn
er twee onderdelen, het meten van de gasconcentratie zelf en methode om die metingen te vertalen naar een netto uitwisseling. Deze twee onderdelen zijn met elkaar verbonden. Sommige methoden, zoals micrometeorologische technieken kunnen alleen
worden uitgevoerd met snelle sensoren die kleine veranderingen kunnen laten zien in
concentratieniveaus in de buitenlucht. De langzame maar continue verbetering van
sensoren maakt thans beschikbare methoden steeds nauwkeuriger. Daarnaast maakt
het ook mogelijk ontwikkeling van innovatieve meetstrategieën mogelijk.
Conclusies vanuit een technisch perspectief
Geconcludeerd kan worden dat met de huidige stand van de sensortechnieken, de
fundamentele methodologische problemen voor een aantal gebruikte meetmethoden
nog steeds niet zijn opgelost. Twee voorbeelden komen aan bod:
• Massabalansmetingen die ten grondslag liggen aan zowel de nationale emissieschattingen voor CH4 uit stortplaatsen, als voor NH3 uit bemeste velden hebben de
emissie overschat (waarschijnlijk tussen de 5-15%). De reden hiervoor is dat de
beschikbare sensoren die worden gebruikt voor deze methode te traag zijn om de
zogenaamde terugwaardse turbulentiecorrectie uit te kunnen voeren die bij deze
methode nodig is. Indien deze correctie wordt toegepast zal de uitwisseling altijd
kleiner zijn.
• Een soortgelijk verhaal geldt voor de boxmetingen maar in dit geval gaat het om
een onderschatting van de feitelijke flux van tussen de 0-30%. De reden hiervoor
is, dat de tot nog toe beschikbare sensoren te langzaam en /of te ongevoelig zijn.
Experimenten die worden uitgevoerd met een te lage tijdresolutie of een te klein
aantal metingen lijken een lineaire toename van de concentratie in de meetbox
aan te tonen terwijl de toename feitelijk steeds verder afgeremd wordt. Voor het
overgrote deel van de boxmetingen beschreven in de literatuur is de emissie met
lineaire interpolatie van de metingen uitgerekend.
Deze twee "fouten" hebben een zeer vergelijkbare achtergrond. De behoefte aan correctie in beide gevallen is al 20 jaar geleden besproken. Maar met de beschikbare sensoren kan correctie eigenlijk nauwelijks worden uitgevoerd. Dus bijna niemand gebruikt
8
de correcties. Dat zou geen probleem hoeven zijn als het om een toevallige (random)
fout ging, maar in beide gevallen leiden de benodigde correcties tot een systematische
verschuiving van de resultaten. Nieuwe sensoren kunnen deze fouten laten zien en
kwantificeren. Over de boxmetingen wordt inmiddels volop gediscussieerd, voor de
massabalansmethode moeten de discussie nog starten.
In dit proefschrift worden drie verbeteringen gedemonstreerd om de uitstoot vanuit diffuse bron-systemen beter te begrijpen:
• De snelle boxmeting, die een combinatie maakt van een eenvoudige box meetmethode met een geavanceerde concentratiesensor (met hoge resolutie in zowel
concentratie als in tijd). Deze methode helpt om ruimtelijke variabiliteit van de
emissies in beeld te brengen binnen een bepaald brongebied. Het helpt ook de
niet lineaire toename van de concentratie bij box metingen te demonstreren.
• De pluimmeting die werd gebruikt voor brongeïntegreerde emissiedata, zowel voor
individuele boerderijen als voor stortplaatsen.
• De “relaxed eddy accumulatie” fluxmeting die een tijdelijke oplossing biedt voor die
gassen waarvoor de snelle eddy covariantie-metingen nog steeds niet mogelijk
zijn.
Conclusies vanuit een bronperspectief
Belangrijke diffuse bronsystemen die een rol spelen in dit proefschrift zijn stortplaatsen,
agrarische velden en boerderijen. Voor stortplaatsen zijn de gegevens die werden verkregen met de pluimmetingen internationaal gezien van belang. Tot nu toe zijn emissie
berekeningsmethoden voor deze bronnen niet gestandaardiseerd. Binnen het stortplaatsonderzoek is een hypothese dat het gebruik van verschillende methoden leidt tot
een factor vier verschil tussen de schatting van de methaanemissie voor een gemiddelde Britse stortplaats in vergelijking tot een soortgelijke Nederlandse stortplaats.
Voor agrarische velden is er nog steeds een grote onzekerheid in zowel de directe en
indirecte emissie van N2O. De snelle boxmeting en de nieuwe eddy covariantie gegevens zullen helpen om tot een betere emissieraming te komen voor deze bronnen.
Voor het bepalen van de netto broeikasgasbalans is het noodzakelijk om simultaan op
dezelfde plek metingen van zowel CO2, CH4 als N2O uit te voeren. Dit is nu mogelijk.
Het is opmerkelijk dat er in Nederland vrijwel geen datasets zijn met simultane metingen van NH3 en broeikasgassen. Dit is vreemd omdat ons land voor alle vier de gassen
de hoogste emissie dichtheden in Europa laat zien.
Voor wat betreft boerderijen is het gebruik van de boerderijgeïntegreerde emissiemetingen nog steeds niet in zwang. De pluim meetgegevens tot nu hebben wel al een
bijdrage kunnen leveren aan het debat over de CH4-emissiefactor voor melkkoeien.
Het resultaat was een significante verbetering van de emissie berekeningswijze voor
deze bron.
Er is niet zoiets als “de beste methode” om de emissie voor elk van deze brontypes te
evalueren. Er is zelfs het potentieel gevaar van systematische fouten indien steeds dezelfde methode wordt gebruikt en er vertrouwd wordt op meer en meer metingen van
hetzelfde type om onzekerheden te verminderen. Aan de andere kant is de vergelijkbaarheid tussen verschillende experimenten een evident voordeel van het gebruik van
"norm" methoden.
De beste optie nu lijkt simultane inzet van meerdere technieken om een betere emissieschatting voor diffuse bronnen te krijgen. De nadruk zou moeten liggen op de verdere ontwikkeling van meer integrerende (micrometeorologische) meettechnieken om een
significante vermindering van de onzekerheid naar wat is overeengekomen als een
aanvaardbaar niveau mogelijk te maken.
9
Tot slot
Dit proefschrift toont de langzame, maar gestage vooruitgang in dit veld van wetenschap gedurende bijna twee decennia. In de komende decennia ligt de sleutel tot verdere verbetering in nieuwe, verbeterde of aanzienlijk goedkopere sensoren die het mogelijk zullen maken beschikbare meetstrategieën te verbeteren of zelfs om nieuwe
meetstrategieën te ontwikkelen. Wat betreft de nationale en mondiale schaal zal de
combinatie van metingen op het aardoppervlak met satellietmetingen een cruciale stap
zijn. Bij de individuele bron is de verdere ontwikkeling van de pluimmethodologie richting “fence-line” monitoring systemen een waarschijnlijke ontwikkeling. Dit betreft bijvoorbeeld de verdere ontwikkeling van goedkope sensoren, lidar (light detection and
ranging) systemen of andere technieken op afstand. De fundamentele fouten voor de
box- en massabalansmethode moeten natuurlijk zo snel mogelijk worden opgelost en
indien mogelijk, moeten correctie algoritmen voor historische gegevens worden ontwikkeld.
Dit proefschrift benadrukt het belang van metingen en het verbeteren van meettechnieken. Metingen alleen geven echter niet alle antwoorden. Metingen, activiteitgegevens
en modellen vormen een drieëenheid die nodig is om de emissieniveaus van diffuse
bronnen te evalueren. Directe metingen, echter, zijn leidend binnen deze drieëenheid.
Verbetering van de sensoren en meettechnieken zijn daarom van belang om de emissies van sporengassen van diffuse bronsystemen beter te kwantificeren.
Arjan Hensen
December 2011
10
Contents
Chapter 1
1.1
1.2
1.3
Outline of the thesis
The introduction
The scientific developments
Synthesis and conclusions
15
15
15
15
Chapter 2
2.1
2.2
2.3
Introduction
Aim of this thesis
Diffuse emission sources
Trace gas focus
2.3.1 Sources and emissions of Carbon dioxide (CO2)
2.3.2 Sources and emissions of Methane (CH4)
2.3.3 Sources and emissions of Nitrous oxide (N2O)
2.3.4 Sources and emissions of Ammonia (NH3)
Impacts of the emissions
Uncertainties in greenhouse gas and ammonia emissions
2.5.1 Uncertainties in CO2 emissions
2.5.2 Uncertainties in CH4 emissions
2.5.3 Uncertainties in N2O emissions
2.5.4 Uncertainty in NH3 emissions
Measurement techniques
2.6.1 Emission measurement strategies for diffuse sources
2.6.2 Sensors
17
17
17
17
18
19
20
21
22
24
24
25
25
26
27
27
28
2.6.2.1 Gas Chromatography.
28
2.6.2.2 Infra red techniques
28
2.6.2.3 Wet chemical techniques
2.6.3 Measurement methods
29
30
2.6.3.1 Box measurement method
30
2.6.3.2 Eddy covariance
34
2.6.3.3 The relaxed eddy accumulation method
35
2.6.3.4 Mass balance method
35
2.6.3.5 Gradient method
35
2.6.3.6 Plume measurements method
Trace gas challenges
Different questions, different methods
36
38
40
2.4
2.5
2.6
2.7
2.8
Chapter 3
Eddy Correlation and Relaxed Eddy Accumulation measurements of
CO2 fluxes over grassland
3.1
Introduction
3.2
Eddy Correlation Setup
3.2.1 Eddy Correlation: Technical description
3.2.2 Eddy Correlation Results
3.3
Open and Closed Path Intercomparison Experiment
3.3.1 Technical Description
3.3.2 Results of the Intercomparison Experiment
3.4
A Combined E.C. and R.E.A. System
3.4.1 Technical Description
3.4.2 Results of the combined E.C. - R.E.A. system
3.5
Conclusions
43
44
44
44
46
47
47
47
49
49
49
50
11
Chapter 4
Inter-comparison of ammonia fluxes obtained using the Relaxed Eddy
Accumulation technique
53
4.1
Introduction
54
4.2
Material and methods
55
4.2.1 Micrometeorological theory
55
4.2.2 Measurement site
56
4.2.3 Measurement set-up
57
4.3
4.4
4.5
4.2.3.1 Detection and response characteristics
57
4.2.3.2 Reference mode sampling
58
4.2.3.3 Use of a dead-band
59
4.2.3.4 Gradient reference
Results
4.3.1 Data coverage
4.3.2 Micrometeorological estimates and deadband
4.3.3 Reference sampling mode.
4.3.4 Comparison of ammonia concentration estimates
4.3.5 Comparison of the flux estimates
Discussion
4.4.1 Estimation of the REA scaling factor β
4.4.2 Comparison of measured ammonia concentrations and fluxes
by the REA and AGM systems
4.4.3 Design features of the different REA systems
59
60
60
60
60
61
62
66
66
66
67
4.4.3.1 Sampling inlet and fast response switching
67
4.4.3.2 Sampling height
68
4.4.3.3 Chemical detection systems
68
4.4.3.4 Use of the reference sampling mode
4.4.4 Potential for measurement of NH3 flux divergence using REA
Conclusions & recommendations
68
69
69
Chapter 5
Methane emission estimates from landfills obtained with dynamic
plume measurements
5.1
Introduction
5.2
Materials and Methods
5.2.1 Description of the measurement sited and methodology
5.2.2 Emission modelling
5.2.3 The landfill gas production model
5.3
Results
5.3.1 Experiments at Nauerna
5.3.2 Experiments at Hollandse Brug
5.3.3 Experiments at Braambergen
5.3.4 Comparison between measured emission levels and
production estimates
5.4
Conclusions
79
80
Chapter 6
Estimation of NH3 emissions from a naturally ventilated livestock farm
using local-scale atmospheric dispersion modelling
6.1
Introduction
6.2
Materials and methods
6.2.1 Measurement site
6.2.2 Concentration measurements
6.2.3 Micrometeorological measurements
6.3
Inference of the source strength from NH3 concentration
83
84
85
85
86
86
86
12
73
74
74
74
76
77
77
77
79
79
6.3.1
6.3.2
6.3.3
6.3.4
6.3.5
6.4
6.5
6.6
General approach
86
The Gaussian-3D Model
88
The Huang-3D plage source model
88
The FIDES-2D model
89
Spatial distribution of the farm emissions with the Gaussian 3D
model
89
6.3.6 Time course of the farm emissions
90
Results
91
6.4.1 Spatial variability of the farm emissions inferred with the fast
response NH3 sensor
91
6.4.2 Temporal variability of the farm emissions estimated with the
Huang 3d model
92
Discussion
93
6.5.1 Emissions from the farm buildings estimated with different
techniques and sensitivity analysis
93
6.5.2 Daily variability of the emissions from a naturally ventilated
building
98
Conclusions
100
Chapter 7
Dairy farm CH4 and N2O emissions, from one square metre to the full
farm scale
103
7.1
Introduction
104
7.2
Materials and Methods
104
7.2.1 Measurement locations
104
7.2.2 CH4 and N2O concentration measurements.
104
7.2.3 Gaussian plume method
104
7.2.4 The fast box measurement technique.
105
7.2.5 "Under cover" measurements.
105
7.2.6 Theoretical emission calculation
106
7.3
Results and Discussion
107
7.3.1 Fast box measurements on the farms
107
7.3.2 Measurements at manure storages
108
7.3.3 Whole farm experiments
109
7.4
Conclusions
112
Chapter 8
8.1
8.2
8.3
Synthesis & Conclusions
General comments
The technical perspective
8.2.1 Chamber measurements
8.2.2 Micrometeorological techniques
8.2.3 Plume technique
The source perspective
8.3.1 Landfills
8.3.2 Agricultural fields
8.3.3 Farms
8.3.4 Final remarks
115
115
115
115
116
118
119
119
121
123
124
Bibliography
127
Acknowledgements
141
13
Figure 1.1
14
CH4 and N2O emission measurements on a peat meadow near
Reeuwijk.
Chapter 1 Outline of the thesis
1.1
The introduction
This thesis is a compilation of experiments and studies that aimed to contribute to the
improvement of estimates of trace gas emissions from diffuse sources into the atmosphere. The thesis starts with a short introduction on the importance of diffuse sources
in different environmental issues. The main questions that initiated the measurement
activities are listed to put the actual work described in the several chapters in perspective. Emphasis is on the complexity of diffuse sources, their variation in time and space
and the role of diffuse sources in the different environmental issues, for which emission
measurements still present a major scientific challenge. The source types and their
characteristics are also outlined.
1.2
The scientific developments
Chapters 3-7 are copies of peer reviewed papers that were published previously. The
first two papers describe a micrometeorological technique known as the relaxed eddy
accumulation technique (REA). The first paper describes how this technique was
tested for carbon dioxide (CO2). In the second paper an intercomparison is made for
REA measurement systems for ammonia (NH3). The second set of papers starts with
the paper on methane plume measurements performed at landfills. The next paper discusses how this technique with mobile measurements was applied to evaluate methane emissions from farms. The last paper of this series describes how stationary
measurement stations were used to evaluate the NH3 emission of a farm site in Germany.
1.3
Synthesis and conclusions
The last part of the thesis discusses these results and provides conclusions. It also assesses the lessons that were learned, the uncertainty in measurement methods and
how to deal with this in future measurements. It also discusses possible and required
developments in the field of trace gas measurements for the near future.
15
NH3
Figure 2.1
16
N2O
CO2
CH4
Agricultural fields are clear examples of diffuse source systems.
Chapter 2 Introduction
2.1
Aim of this thesis
After the introduction of diffuse sources and the trace gas components that are the subject of this thesis, this chapter will explain which environmental issues require the
knowledge that is obtained through the type of measurements that feature in this thesis. The trace gasses that are subject to the case studies are documented and the
various source types are discussed. Finally the measurement methods that can be
used to evaluate several combinations of source type and gas species are discussed.
2.2
Diffuse emission sources
This thesis discusses the progress in quantification of diffuse sources for the emissions
of trace gasses into the atmosphere. The term diffuse source is used to distinguish
from sources that do have well defined outlets, like for example chimneys or ventilators. In general, for those sources, a flow and concentration measurement can be used
to evaluate the net source strength. However, when there are many outlets from one
source or when there are multiple sources within a certain area that whole area can be
considered as a diffuse source as well. So the definition used is:
Diffuse sources emit gas without a well defined outlet.
Diffuse sources include:
• Soils, either natural, agricultural of industrial (landfills),
• Groups of sources within a confined area (farm with multiple animal houses, manure storages or industrial plants with a large number of point sources,
• Regions, a group of farms; farms and industries, households; urban areas.
2.3
Trace gas focus
The atmosphere has three main components, nitrogen, oxygen and water. Next to
these there is a whole suite of less abundant trace gasses. Trace gasses are defined
as those gases that make up less than 1% of the atmosphere. In spite of the low concentrations in which these gasses occur they play several leading roles in different environmental issues, including climate change and air pollution. In this thesis work is
presented for a subset of the relevant trace gasses: Carbon dioxide (CO2), Methane
(CH4) and Nitrous oxide (N2O) as the main greenhouse gasses with significant contributions from diffuse sources. For air pollution, ammonia (NH3) is featuring as the important trace gas from diffuse agricultural sources.
This thesis focuses on emission from diffuse sources of CO2, CH4, N2O and NH3.
17
2.3.1 Sources and emissions of Carbon dioxide (CO2)
The main net source of CO2 emitted to the atmosphere is fossil fuel burning. The natural exchange of CO2 between the atmosphere and ecosystems is bi-directional. Photosynthetic uptake of CO2 from the atmosphere occurs in presence of sunlight by plants
and algae. Meanwhile respiration of CO2 occurs continuously through decay of organic
material from the soil or litter. During the night the plants themselves also respire part
of the CO2 that was stored during daytime (Winegardner, 1995). The CO2 exchange
between ecosystems and atmosphere therefore almost always shows a significant diurnal pattern that might be superimposed on a more continuous net emission from decomposing soil material. The amplitude of the diurnal pattern of uptake and plant respiration is a function of the amount of plants available (leaf area index, LAI) whereas the
soil emission is a function of available organic material and its composition. Increasing
temperatures in general lead to increased decomposition. Apart from temperature, humidity, rainfall and water (table) management will also influence decomposition the
process.
The Dutch National emission inventory report (NIR) provides both the annual emission
levels and the trend in the emission since 1990. In spite of the emission reduction policies for CO2, the emission of the main greenhouse gas increased since 1990 as shown
in the graph below (NIR, Maas et al., 2009). Emissions are split up in the most important sectors, as defined by IPCC (IPPC AR4). According to the NIR 2008, the trend in
total national CO2 emissions shows an increase in the period 1990–2006 with 8% (from
159.4 to 172.2 Tg CO2/y).This number is not including the impact of land use, land use
change and forestry (LULUCF). The Energy sector is by far the largest contributor to
CO2 emissions in the Netherlands (36%), with the categories 1A1 ‘Energy industries’
(36%) and 1A4 ‘Other Sectors’ (22%) as largest contributors in 2006.
Figure 2.2
18
CO2 emissions in the Netherlands (Maas et al., 2009).
2.3.2 Sources and emissions of Methane (CH4)
Methane is produced when carbohydrates, organic material, are exposed to anaerobic
conditions (no oxygen available). Temperature increase in general speeds up CH4 formation and water is the main driver to get the anaerobic conditions. CH4 uptake in soils
occurs as well through methane consuming bacteria (Winegardner, 1995; Hendriks
2009). In general, in the Netherlands this uptake is small compared to the emission
levels. Under temperate conditions large CH4 emissions occur from enteric fermentation, in particular from ruminant animals, and from anaerobic storage of manures.
There are however, also other sources of methane on livestock farms, such as ditches
and feed storage.
Within Europe the Netherlands have the highest emission density for methane of about
around 16 ton CH4 per km-2 per year (www.Compendiumvoordeleefomgeving.nl, 2009).
The most relevant anthropogenic emissions originate from ruminants (with dairy cows
as the most important category in the Netherlands) and landfills. Landfill emission estimates have gone down with a factor 3 between 1990 and 2010 due to a shift from landfilling to recycling and incineration (Maas et al., 2009). Although dairy numbers have
decreased by 20 % since 1990 the emission reported for enteric fermentation has only
decreased by 12% due to changes in animal diet (Bannink et al., 2011).
Methane leakage also occurs from oil and gas exploration and domestic or industrial
use. Landfills, herds of cows or farm sites with multiple sources in a confined area are
all clear examples of diffuse sources. Industrial installations will in theory have well defined gas exhausts. But when a complex industrial installation has multiple outlets and
possible leaking on various places this too can be considered to be a diffuse source.
In total, 90% of all CH4 emissions in the Netherlands originate from diffuse source systems. According to the NIR 2008, the national CH4 emissions decreased by 36%, from
1,211 Gg in 1990 to 775 Gg in 2006 (25.4 to 16.3 Tg CO2 -eq.). The Agriculture and
Waste sector (54 % and 36%) are the largest contributors in 2006. Compared to 2005,
national CH4 emissions dropped by 3% in 2006 (-0.7 Tg CO2 -eq.), due to the further
decrease of CH4 emissions mainly in category 6A: ‘Solid waste disposal on land’.
Figure 2.3
CH4 emissions in the Netherlands (Maas et al., 2009).
19
2.3.3 Sources and emissions of Nitrous oxide (N2O)
Nitrous oxide is primarily emitted through agricultural activities and industrial processes
such as fertilizer production and fuel combustion. Changes in land use such as switching from a forest to arable or grassland also affect N2O emissions. Natural emitters of
nitrous oxide such as forests, aquifers, rivers and estuaries are enhanced by nitrogen
leaking and deposition. For N2O about half of the emissions originate from diffuse agricultural sources. Nitrous oxide emissions occur from all parts of the agricultural nitrogen cycle (Mosier et al., 1998), and there are many different diffuse sources of N2O on
livestock farms. Industrial N2O is used in hospitals as anaesthetic, in discos as a party
drug and in racing cars (Hensen, 2000a). The production of fertilizer is the main source
of N2O from industry. Transportation also is a source of N2O, produced in the three way
catalytic converters used in cars that are installed to reduce the emission of NOx to the
atmosphere.
On the third place as a contributor to anthropogenically-enhanced greenhouse effect,
N2O also plays an important role in the nitrogen cycle where it has a strong link with
emissions of NH3 (see below). The implementation of new legislation for manure
spreading around 2004 has had a negative effect on the amount of N2O emitted. Above
ground spreading of manure is forbidden on grassfields but, in order to reduce NH3
emissions, manure has to be either incorporated into the soil (injection) or applied directly on the surface (trailing hose/shoe techniques). Doing so more N becomes available for the soil microbes that produce N2O as an intermediate product in nitrification
and denitrification processes (Wrage et al., 2001). The term indirect N2O emission is
used for those emissions of N2O that occur in ditches, rivers, lakes and seas or in terrestrial systems either due to NO3 and NH4 runoff into the aquatic systems or due to
NH3 and NOx deposition.
According to the NIR 2008, the total national inventory of N2O emissions decreased by
about 15%,from 64.3 Gg in 1990 to 54.7 Gg in 2006 (from 19.9 to 16.9 Tg CO2-eq.).
Sectors contributing the most to this decrease in N2O emissions are sectors ‘Agriculture’ (–18%) and ‘Industrial Processes’ (–13%). During the same period N2O emissions
from fossil fuel combustion increased. This latter trend can be largely explained by increased emissions from Transport. Compared to 2005, the total N2O emissions decreased by 1% in 2006 (-0.2 Tg CO2-eq.).
Figure 2.4
20
N2O emissions in the Netherlands (Maas et al., 2009).
2.3.4 Sources and emissions of Ammonia (NH3)
Agricultural activities and more specific decomposition and volatisation of animal
wastes are the main source of NH3. In manure, dissolved ammonia (NH3) and ammonium (NH4+) are in equilibrium. When manure is exposed to ambient air ammonia can
escape and available ammonium will provide new ammonia. The volatisation process
depends on pH of the manure, temperature and of course on the structure of the manure. Excreta left by grazing animals in a field, or manure that is spread are definitely
diffuse source systems. There are some point sources for NH3 like forced ventilated
barns for pigs of poultry and some industrial sources. Housing systems with natural
ventilation however also behave as diffuse source systems. Emissions from catalytic
converters in traffic can be considered to be diffuse as well.
According to NIR 2008, the NH3 emissions decreased by about 46%, from 250 Gg in
1990 to 135 Gg in 2007. The main decrease occurs in the episode 2000-2004 after
which the decrease stops. The emission reduction measures for manure application in
agriculture are the main cause of this decrease. A further reduction of the national
emission level was related to the decrease of livestock. (Milieucompendium, 2008)
NH3 emissions
kton NH3
300
250
200
150
100
50
0
1990 1995 2000 2001 2002 2003 2004 2005 2006 2007
Figure 2.5
NH3 emissions in the Netherlands (NIR, 2008).
21
2.4
Impacts of the emissions
Ammonia and the three greenhouse gasses have major sources in terrestrial and/or
aquatic ecosystems besides those from industrial and fossil fuel related emissions.
These gasses fulfil a prominent role in the N and C cycles. Anthropogenic activities
have direct and indirect effects on these cycles. For the C cycle, this disturbance is
relatively straightforward, in essence it comes down to the addition of extra CO2 or CO2
equivalent emissions into the atmosphere from fossil fuel use and deforestation. CO2
plays an important role in climate change (IPCC, AR4).
The disturbance of the N cycle is more difficult with the large set of interacting N-trace
gas species. Almost all nitrogen is available in the atmosphere as non-reactive nitrogen
gas (N2). Only about 0.1% is nitrous oxide (N2O) (Wallace and Hobbes, 2006) and an
even smaller fraction is filled with a suite of reactive trace gas species, including NH3
and NOx. All the non N2-gasses, are being termed “reactive nitrogen” (Nr) (Galloway et
al., 2003, 2008). The global cycling of that reactive nitrogen is estimated to have more
than doubled, (Vitousek 1997, Galloway 2004) where the perturbation of the C cycle is
about 10%. The two cycles interact, additional input of nitrogen in the N cycle leads to
more C uptake and changes in landuse also lead to changes in the carbon exchange.
The carbon cycle is strongly linked with climate change, the nitrogen cycle features
both in air pollution and in the climate issue.
Diffuse source systems and the climate change issue
The majority of the anthropogenic emissions of the most important greenhouse gas,
CO2 originate from relatively well defined (fossil fuel) point sources (chimneys or exhausts). However, diffuse sources that emit CO2 or CH4 and N2O in the Netherlands
are estimated to contribute about 20% of the total GHG emission in kg CO2 equivalents
(Milieubalans, 2009) to the atmosphere.
The CO2 exchange between the atmosphere and diffuse source systems is highly relevant to climate. According to Janssens et al. (2003), Europe’s terrestrial biosphere captures the equivalent of 7 to 12% of anthropogenic carbon emissions each year. When
separated into uptake and emission fluxes these are of the same order of magnitude as
the fossil fuel contributions. Furthermore, these CO2 exchange fluxes are expected to
change both with changing management and with changing climate conditions. The potential changes in the net CO2 exchange will be reflected in changes in agricultural productivity and thus have implications for food security.
Still the net contribution of different ecosystems is hard to quantify. Regarding CO2 it is
hard to quantify the anthropogenic net effect on top of the large amounts of CO2 that
leave and enter the atmosphere in the “undisturbed C cycle”. However, in the Netherlands, forest are expected to be a net GHG sink (e.g., Dolman et al., 2002), grasslands
a minor sink (Soussana et al., 2007), arable land a source (Moors et al., 2010). Due to
past drainage in Dutch fen meadow areas, these peat soil ecosystems have been a
strong net source of carbon dioxide over a long period with increased peat oxidation as
the primary source (Dirks & Goudriaan 1994; Langeveld et al. 1997). Peat oxidation
can be reduced and fen meadows can be turned into sinks of CO2 if water levels are
increased as suggested from literature from mostly more natural (i.e. less exploited) fen
ecosystems (Burgerhart, 2001). The total GHG emission reduction through the increase of water levels could be of considerable magnitude (5 – 15 Tg CO2 equivalent
ha-1 yr-1). This is similar to the carbon gain that can be achieved in (re-) growing temperate forests (4 – 11 Tg equivalent CO2 ha-1 yr-1), (Dolman et al. 2002).
22
Methane emission reduction options are considered to be an attractive short term option to avoid climate change. The lifetime of CH4 in the atmosphere (estimated to be
about 20 years, IPCC, AR4) is relatively short as compared to those of CO2 and N2O.
Emission reduction measures therefore will be reflected in atmospheric concentration
changes within decades after implementation. For CO2 and N2O the time lag between
emission reduction implementation and atmospheric response is more in the order of a
century or more (IPCC, AR4). Another good reason to capture methane that would
otherwise escape into the atmosphere is that methane is a fuel. Burning a kg of methane and converting it to CO2 (2.2 kg) reduces the net GHG contribution by a factor of 8.
Landfills play a prominent role here because at landfills methane extraction and use is
relatively easy.
Enhanced fertilization in deforested areas increases the N2O emission from soils. This
is strongly linked to the important development of a rising worldwide demand for biomass. Bio fuel and biomass should replace a significant percentage of fossil fuel use in
the decades to come. The changes in use of agricultural and set aside areas and the
potential loss of natural area will however feed back into the C and N cycle. In a paper
by Crutzen et al. (2007) it is suggested that the net gain of biomass/biofuel use in terms
of avoided fossil fuel carbon emissions could suffer severely from enhanced CO2
equivalent emissions related to N2O escaping into the atmosphere (see below). Extra
fertilizer use and changed soil management both lead to these higher N2O emissions.
NH3 plays an indirect role in climate change. In the atmosphere, part of the NH3 reacts
to form ammonium nitrate and ammonium sulphate aerosol (IPPC, 2001, 2007). These
constituents play an important role in cloud formation and in direct backscattering of
sunlight, thus leading to a cooling effect. (Ten Brink 1996; Veefkind, 1996). A point of
discussion is to what extent ammonia emissions and subsequent deposition, for example to forests, might enhance C uptake in ecosystems that normally do not receive fertilizer. (Magnani et al., 2007; De Vries et al., 2009)
Diffuse source systems and air pollution
The most prominent role in air pollution from diffuse sources is played by NH3. Most of
the NH3 comes from diffuse source systems. In the Netherlands, 88% of all NH3 emissions originate from diffuse agricultural systems (Milieubalans, 2009). In the EU 27 this
value is a staggering 93% (EMEP, 2009) After emission and transport NH3 will be deposited to the surface where it can lead to acidification, eutrophication of natural ecosystems and loss of biodiversity (Galloway 1998, Erisman et al., 2008).
The average N deposition level in the Netherlands related to both NH3 and NOx emissions adds up to about 30-40 kgN/ha/year (Milieubalans, 2009). This is considered to
be too high in vulnerable ecosystems, leading to loss of biodiversity and loss of other
ecosystem services (Wamelink et al., 2008)
CO2 CH4 and N2O are not considered air pollutants that have an adverse effect on
health. However, both CH4 and N2O can play a role affecting the ozone (O3) concentration, thus connecting climate change and air pollution. Emitted CH4 has an indirect air
pollution effect since it can lead to higher tropospheric ozone concentrations through
removal of the OH radical (Hansen et al., 2000; Fiore et al. 2008). The O3 formed acts
both as a greenhouse gas and as an air pollutant. N2O also interacts with O3 but this
occurs in the stratosphere. Elevated N2O levels lead to O3 destruction and therefore
contribute to the Ozone Hole (Crutzen, 1970).
23
2.5
Uncertainties in greenhouse gas and ammonia emissions
The uncertainty in the national emission budget for the Netherlands is illustrated in Tables 2.1 and 2.2. These tables are copied from the national emission inventory report
(Maas et al. 2009). The first table 2.1 shows that the uncertainty in the emissions of
methane and N2O are significantly higher as compared to those of CO2 (when the
LULUFC contribution is ignored). For CO2 the national emission level is dominated by
the relatively well constrained contributions from fossil fuel combustion. The table
shows there are still significant uncertainties in the emissions of CH4 and N2O.
Table 2.2 shows the top ten sources (excluding LULUCF) contributing most to the annual uncertainty in 2008 after ranking the sources according to their calculated contribution to the uncertainty in the total national emissions (using the column ‘Combined
uncertainty as a percentage of the national emissions in 2008 in table A7.1 (from Maas
et al. 2009) It also provides an estimate of the contribution to the total uncertainty in the
national emission estimate for the Netherlands. Six out of the top ten of uncertainty
contributors are in fact diffuse sources. Most of these are linked with agriculture. The
most important one is the N2O emission from nitrogen use in agriculture. But also
emission from manure management, enteric fermentation (cows) and landfills have
significant uncertainties.
Table 2.1
Uncertainty of the total annual emissions (excl. LULUCF)
(from Maas et al. 2009)
CO2
CH4
N2O
Total GHG
Table 2.2
3%
HFCs
50%
25%
60%
PFCs
SF6
50%
50%
5%
Top ten sources contributing to total annual uncertainty in 2008
IPCC
category
Category
Gas
4D3
Indirect N2O emissions from nitrogen use in
agriculture
Direct N2O emissions from agricultural soils
Stationary combustion: Other sectors commercial/Institutional gases
Emissions from solid waste disposal sites
Emissions from manure management: cattle
Stationary combustion: Petroleum Refining:
liquids
Emissions from manure management : swine
Mobile combustion : road vehicles: diesel oil
CH4 emissions from enteric fermentation in
domestic livestock: cattle
Emissions from manure management
N2O
Combined uncertainty as a
percentage of the total national
emission in 2008
*)
3.18%
N2O
CO2
1.4%
1.1%
CH4
CH4
CO2
0.8%
0.7%
0.6%
CH4
CO2
CH4
0.5%
0.5%
0.4%
N2O
0.4%
4D1
1A4a
6A1
4B1
1A1b
4B8
1A3b
4A!
4B
2.5.1 Uncertainties in CO2 emissions
In the national inventories, the contribution of diffuse systems (LULUCF = landuse, land
use change and forestry) were left out of the GHG budget calculations for quite some
time (see the graph in 2.2.1). The contributions of landuse, land use change and forestry were reported for the first time in 2003. At that time it only covered the sink in forests. In 2005 all LULUCF categories were included for the first time in the inventory
with a net source as result. As stated in the previous paragraph, a proper assessment
of the net effect of the large bi-directional exchange is the main source of uncertainty
24
for the CO2 budget. Dolman et al., 2008, provide a comprehensive overview of all aspects of getting an overview of the greenhouse gas balance on the European scale.
2.5.2 Uncertainties in CH4 emissions
Emissions of methane from landfills have a high uncertainty. Both commercial recovery
projects and the national emission calculations rely on first order kinetic equation models to predict CH4 generation at landfills (Scheepers and van Zanten, 1994; Coops et
al., 1995; Blaha et al., 1999) Inputs to these models are waste inputs, climate variables, and other factors. For individual landfills the uncertainty in the gas production
and emission estimates is also a problem. According to the European Pollutant Release and Transfer Register (E-PRTR) (CEC, 2006), large landfills have to report their
CH4 emissions. But substantial differences in methane emission estimates of different
models that are allowed were demonstrated by Scharff et al. (2006) and Ogor (2005).
The highest estimates obtained with the models can be five to seven times higher than
the lowest estimates. These variations are unacceptable for E-PRTR reporting purposes. At the international level, the uncertainty in landfill CH4 production and subsequent emission can safely be assumed to be higher, as compared with the Dutch situation. Several landfill operators in the Netherlands have reasonable to good inventories
of what was landfilled. These data is lacking in a lot of countries. Bogner and Matthews
(2003) therefore propose a proxy for landfill gas emissions that is based on population
and energy consumption data.
Emissions from ruminants are affected by their diet, bodyweight and milk production.
(Kirchgessner et al.,1995) Problem is that emission measurements are always carried
out on small groups of animals after which extrapolation to the whole herd of 4 million
animals (in the case of dairy cows in the Netherlands) is needed (Van Amstel, 2003;
Corre 2002). Significant differences between management at individual farms and including dietary differences introduce the large uncertainty in the upscaling to a national
level. (Veen, 2000)
Natural methane sources are wetlands, ditches, lakes and soils. Anthropogenic impact
on these systems has however modified these emissions but it is difficult to quantify to
what extend this has happened. The emissions from industrial processes, transport
and oil and gas treatment are believed to be small (Maas et al., 2009).
2.5.3 Uncertainties in N2O emissions
The contribution of N2O from fertilized agricultural fields to the total GHG balance for
the Netherlands still has an uncertainty in the order of 50% (Maas et al. 2009). Table
2.3 shows the EU budget and its uncertainties (Freibauer, 2008). The reasons for the
high uncertainty levels are the large spatial variability of the emission over a field. In
particular grazed field with urine patches show a hudge spatial variability in the emission. Furthermore the temporal structure of the emission pattern is complicated. Significant peaks in the emission can typically occur between 0 and about 14 days after
fertilising, often triggered by rain. This emission “scheme”, in combination with the
measurement methods available, leads to uncertain emission estimates for the measured field. On top of that, extra uncertainty will be added in the process of upscaling to
a national level. This upscaling has to rely on the representativity of a limited set of
measured fields for other similar fields. Reviews of the spatial and temporal variability
are found in Freibauer and Kaltschmitt (2003) Smith et al. (1998) or Velthof et al,
(1996). Emissions from waste water treatment facilities are unknown but believed to be
small and emissions from industrial processes are believed to be accurately known.
25
Table 2.3
EU Estimated budget of anthropogenic N2O for the European Community Emissions are derived from the national communications and uncertainties. Table
copied from Freibauer (2008)
Eu-27 in 2005
N2O Emission
Gg year
Agricultural soils, “natural background”
(Bouwman, 1996)
-1
EU-27: 200
EU-15: 130
Anthropogenic N2O emissions from national communications
Soils anthropogenic total
EU-27: 499
EU-15: 346
Agricultural soils direct
267
Agricultural soils indirect
169
Manure management
60
LULUCF
5
Industrial processes
128
Transport
52
Other energy
74
Waste
27
Other sources
9
−1
Anthropogenic N2O (Gg N year )
EU-27: 861
EU-15: 687
Anthropogenic N2O in EU-27 (Tg Ceq
91
−1
year )
Soilborne N2O emissions in EU-15 from models
a
Agricultural soils in EU-15
442
Forest soils in EU-15
Soil N2O emissions in EU-15
Uncertainty
Gg year
-1
%
160
104
80
400
277
80
214
135
45
4
13
26
22
27
9
551
440
80
80
75
80
10
50
30
100
100
64
64
58
a
116
26
b
78–87
525
LULUCF Land-Use, Land-Use Change and Forestry
a
Freibauer (2003)
b
Kesik et al. (2005)
2.5.4 Uncertainty in NH3 emissions
Beusen et al. (2008) performed an analysis of the global NH3 emission from agricultural
activities and assigned uncertainty for different parts of the budget calculation. Table
2.4 shows a copy of the table from their article. The animal stocks are the most certain
part of the budget calculations with 10% uncertainty worldwide. The highest uncertainty
is assigned to the emission factors for NH3 from fields and housing systems both with a
50% uncertainty range. For the Netherlands the total uncertainty in the NH3 emission
level is estimated to be about 17% (in 2000, Milieubalans 2009, PBL)
The preceding paragraphs show that there are still significant uncertainties in diffuse
source emission levels. And that the emission levels are substantial enough to try and
improve that situation. Parts of the uncertainties are linked to shortcomings in available
measurement techniques or methods. Another reason is that measurements are usually done only at a subset of the sources in a source category. Both are reason enough
to elaborate on what measurement techniques are available in the next part of this
chapter.
26
Table 2.4
This table, adapted from Beusen et al. (2008) shows the typical uncertainties used in a modelling uncertainty analysis. The table shows selected parameters with the method and assumption for the uncertainty
analysis.
a
Parameter description
Method
Animal stocks
A
N fertilizer use
A
Maximum manure application rate
A
Excretion in pastoral systems
Fraction NH3 volatilization from grazing animals in
A
mixed and landless systems
Fraction NH3 volatilization from grazing animals in
A
pastoral systems
Fraction NH3 volatilization from housing and storage in mixed A
and landless systems
Fraction NH3 volatilization from housing and storage in pastoral A
systems
Fraction production in mixed and landless systems
B
Fraction grazing in mixed and landless systems
B
Fraction grazing in pastoral systems
B
Fraction manure ending outside the agricultural system in mixed B
and landless systems
Fraction manure ending outside the agricultural system in pas- B
toral systems
Fraction stored manure applied to grassland in mixed
B
and landless systems
Fraction stored manure applied to grassland in pastoral systems
B
a
A, the range around the default value X is calculated as [(1.0–R)_X,
(1.0+R)_X]; B, the range around the default value X is calculated as
[max(0.0,(X_R)), min(1.0,(X+R))] (in Beusen et al. 2008)
2.6
a
R
0.10
0.10
0.25
0.25
0.5
0.5
0.5
0.5
0.25
0.25
0.25
0.25
0.25
0.25
0.25
Measurement techniques
2.6.1 Emission measurement strategies for diffuse sources
There is a wide range of potential measurement strategies that can be used to measure the concentrations and emission levels. This section provides an overview of the
available techniques relevant to this thesis. In essence, a measurement strategy covers a combination of one or more sensors that provide data on the concentration level
for a trace gas and a method to translate these data into flux data. As for sensors, the
gas chromatographic technique, infra red detection and wet chemical sample analyses
are relevant for this thesis.
The methods section describes box measurements, micro meteorological techniques,
like eddy covariance and relaxed eddy accumulation, gradient and mass balance technique and plume measurements.
Finally there is the scheme to decide when and where to carry out measurements.
Sometimes models or parameterizations are needed to translate the measurement into
emissions. Models are also essential tools to upscale emissions to spatial scales that
extend beyond the footprint of the measurements themselves.
27
2.6.2 Sensors
2.6.2.1 Gas Chromatography.
A Gas Chromatograph (GC) system uses the molecular properties of the gas species
to split a sample into the different components. Conceptually all GC’s inject a few ml air
sample at one side of a tube (column) with a small diameter and use a detector at the
other end of the tube. Different gas species will have different travelling times through
the tube depending for example on molecular size or dipole behaviour. Different delay
times can be used to split the mixed gas sample into its individual components, the
concentration of which can then be evaluated with the detectors. The detector shows
the gasses for which it is sensitive as a set of subsequent peaks. For ambient concentration levels of greenhouse gasses two detectors are used, the flame ionisation detector (FID) for methane and other hydrocarbons is in fact sensitive for the number of C
atoms that can be oxidized. The electron capture detector (ECD) is used for CO2 and
N2O. The ECD detects molecules that have an affinity with free electrons. For our applications a single GC measurement generally lasts for about 6 minutes. In order to increase the sensitivity a number of subsequent analyses can be done. Regular (every
30 minutes) calibration of the system is required to enable corrections for system drift.
A proper GC system with a skilled operator can measure CH4 concentrations in the
near background conditions at about 2000 ppb range with an accuracy of 0.2 ppb max
and 310 ppb N2O with a 0.2 ppb accuracy.
Figure 2.6
Gas Chromatography (GC-FID & ECD systems at Cabauw).
2.6.2.2 Infra red techniques
There are different infrared detection (IR) techniques available at present. In fact all
these options use the greenhouse gas effect on a small scale. An infrared light source
sends a beam into a confined space with the sample gas. When this sample contains
gasses have absorption bands in the infrared (H2O, CO2, CH4, N2O and NH3) a part of
the light send out is absorbed and does not reach the detector. An increased loss indicates a higher concentration level.
28
Sampling
The sample gas can either be pumped into a measurement cell where the IR beam is
illuminating the sample or the IR beam can be used in the outside air. We refer to
closed path and open path systems for the two respective methods. In this thesis only
closed path measurement systems were used. The main reason to choose these systems is that calibration can be done more frequently (i.e. on a daily basis).
IR source & detector
The IR source can either be a broadband thermal system or a laser with a specified
wavelength. A broadband system generally uses interference filters to select a wavelength band where the gas of interest has absorption lines. Fourier transform infra red
spectrometry (FTIR) uses a broadband IR source and makes a scan though the spectrum. A laser-based system can be tuned to a unique absorption line of a specific trace
gas. Tuneable diode laser (TDL), Quantum cascade laser spectrometers (QCL) and
cavity ringdown systems (CRD) all use the latter whereas the NDIR (Licor 6262, Siemens Ultramat 5 and the opto acoustic B&K1403/Innova) systems use broadband IR
sources. Opto acoustic systems can also be combined with a laser source.
Figure 2.7
Different IR systems. Left non dispersive infra red monitor, (Licor 7000),
right Quantum Cascade Laser spectrometer (ARI) for CH4, N2O and,
NH3 with in-field alignment tool.
2.6.2.3 Wet chemical techniques
Measurement of NH3 is a lot more difficult compared with measurement of CH4, N2O
and CO2. The gas is highly reactive and adheres to surfaces easily, with a Henry coefficient well over 104, adsorption into small water layers on sampling lines is a big problem for non-open path optical systems. Ammonia is lost at high concentration levels
and reemitted from water and surface when the concentration level decreases. (Neftel
et al., 2011). This is visible as a significant damping (variations are smoothed out) and
tailing when concentration changes occur. Inlet filters can hardly be used for this reason, which is a problem for optical systems that have dust-sensitive mirrors. Virtual impactors, heated inlet lines and high sample flows can reduce the problems for closed
path systems but these make the measurement setups bulky and power intensive. The
disadvantage of NH3 is turned into an advantage when using wet chemical techniques.
Several types of denuder systems were developed over the last decades, and they all
29
rely on the absorbance of NH3 into water where it is in equilibrium with OH- and NH4+. At
ECN the rotating wet annular denuder (AMANDA) was developed and deployed extensively in the 90’s. (Wyers et al., 1993). These systems were further improved and
adapted for gradient measurements. Typical sensitivity of the wet chemical systems is
about 0.05 µg/m3 which is sufficient to resolve gradients over ecosystems for both
emission and deposition studies. (For example Milford et al., 2009)
In spite of this, the optical systems are becoming more popular because of their user
friendliness. Furthermore the wet chemical systems are unable to provide the 10 Hz
resolutions in the concentration reading that are required to enable eddy covariance
studies (see below). Trebs et al., (2004) give an overview of sensor systems available
for NH3 and other reactive gasses and aerosols.
Figure 2.8
Airrmonia wet chemical NH3 monitor.
2.6.3 Measurement methods
2.6.3.1 Box measurement method
Putting a box upside down on an emitting area and measuring how the concentration in
the box in- or decreases. That is in essence the box or chamber measurement technique. Box type measurements have the main advantage that the concentration signal
is amplified significantly so that smaller fluxes can be evaluated with a given, low instrumental precision. The sensors used for chamber measurements also need not to
be fast response sensors. GC systems and opto-acoustical instruments (Innova/BK)
are used everywhere for all three greenhouse gas species. The advantage of box
measurements is that a single m2 of the ecosystem can be studied without interference
from other sources. The measured flux level can be linked to other on the spot data like
soil temperature, nitrate availability, water table depth or pH which facilitates the development of process models. Box measurements do have a problem covering the fast
temporal and spatial inhomogenity of, for example, N2O emissions from a grazed field
(Flechard et al., 2007) or CH4 emissions from landfills (Bogner et al., 1997; Spokas et
al., 2003). Manual chamber studies have are in danger of missing main peak events
after rain, for example for N2O, because the experimentalists are not in the field on a
particular day. Furthermore many datasets with box measurements are underestimating the emission level because a linear increase in the concentration in the box is assumed after closure. As discussed in a number of recent papers, (Kroon et al., 2008,
Kutzbach et al., 2007) this underestimation might be in the order of 20-40%. In spite of
30
the uncertainty in the emissions measured with this technique the box measurements
are still widely used and underlying the emission factors used for N2O from soils. A
good reason for this is that the box technique amplifies the concentration signal related
to the flux. This is needed since until recently the N2O sensors were not sensitive
enough to use other methods like the once described below.
Automated box systems (for example Smith & Dobbie, 2001) that are in the field continuously can solve part of the problem but in general these systems only have a limited spatial coverage (a limited number of automated chambers on a field). These limitations and the problem that interpretation of the box measurements can be underestimating by 20-40% when assuming that the concentration will increase in a linear way,
is underlying part of the large uncertainty both in direct and indirect in N2O emissions
as listed in the table shown in table 2.2.
560
0
40 minuten
30 cm
Figure 2.9
Box measurement methods, measure the increasing (or decreasing)
concentration inside a box, preferably a number of times after closing
the box.
CH4
N2O
TDL
AS
//
/
/
tracer
Position A
Position B
Air to the GC
0.8
Figure 2.10
Two box measurement setups used. The fast box measurement setup
(left using the TDL as a sensor, and right the automatic chamber that alternates between two positions and is connected to a GC.
31
Figure 2.11
Box measurement setups for N2O and CH4 at various experiments.
Above (left to right), N2O emission from manured grassland on a peat
soil (Reeuwijk), emission measurements on a potato field, Automatic box
measurements on a grassfield at Cabauw. Below: Farm yard manure,
dung heap Dronten, Ditch, Lelystad.
In Chapter 7 of this thesis the combination of the box measurement technique with the
fast response TDL is described (see fig 2.10). The advantage of using the 10 Hz instruments for this application is that, in general, only about 20 seconds to a minute per
flux measurement is required. With this method, a high temporal resolution of the emission landscape can be obtained (see figure 2.11).
32
Example of chamber measurements at landfills
In the Netherlands chamber measurements are used to evaluate the emission level at
specific parts of the landfill. An example of box measurements is shown below at the
Braambergen landfill near Almere. Here a methane oxidation system was applied on a
part of the landfill. The plot shows the emission pattern in space derived from fast box
measurement. An emission level for this part was derived by classification of the cover
soil in different areas (bars soil, dense vegetation, cracked soil etc.) and using a
weighted average of the emission measurements taking into account this classification.
North
N or d -S
u d (m)
(m )
-49650
350
Emission l.m -1.h -1
0 .5
1
5
-49700
300
0.5
1
5
Em
is s ionafternoon
M it t a g18-6-2
18-6-2
Emission
-49750
250
Em
is s ionmorning
M or ge18-6-2
n 18-6-2
Emission
s aMeas.
m p linlocation
g s t a t18-6-2
ion s
c
-49800
200
MMeas.
e s s . loc
14-8-1
location
14-8-1
E Emission
m is s ion14-8-1
14-8-1
150
-49850
-49900
100
-49950
50
-50000
0
0
40500
50
40550
100
40600
150
40650
200
40700
250
40750
300
40800
350
40850
400
40900
450
40950
500
41000
550
41050
Ost -W est
(m(m)
)
East
Figure 2.11
The Braambergen landfill (top left) and the air injection
system in the topsoil of the landfill that alternates between pumping out
gas and injecting air (oxygen) to optimise oxidation in the cover soil.
Photo top right the fast response box measurement system, (see figure
2.10). Below: result of fast box measurement system. Most of the measurements were carried out on the slope that had the Smell well system
implemented (orange part). Some measurements were done on top of
the landfill (green part). The size of the circle indicates the emission
level.
33
41100
2.6.3.2 Eddy covariance
Eddies in the atmosphere transport air up and down and thereby mix high concentration air from nearby the source with background air (e.g.Stull, 1988). The up and down
movement of turbulent eddies is monitored using ultra sonic anemometry. This technique uses the travel time of ultra sonic signals between three sets of emitters/transceivers to compose the 3D wind vector at frequencies of 20-100 Hz. Above a
field a combination of the vertical wind speed data with the sonic and the concentration
measures or samples close by this wind sensor will show a positive correlation when
the field is emitting gas and a anti-correlation when an uptake occurs. The eddy covariance (EC) method relies on the possibility to measure rapid changes in the concentration at frequencies up to at least 1 but preferably 10 Hz. Fast response sensors are
available for H2O and CO2 now for almost two decades. After fierce debates in the 80’s
on the comparison of box and atmospheric measurement techniques, now almost all
ecosystem research teams are using CO2 EC measurements for hectare scale exchange measurements. For CH4 and N2O this discussion is still ongoing, and the box
measurements are still the reference technique. This is mainly due to the fact that
available CH4 and N2O flux measurement systems (for example TDL’s) are wonderful
machines but very complicated to use as compared to the CO2 /H2O sensors. Over the
last 5 years this is rapidly changing with more turnkey cavity ringdown systems and
with the switch from TDL to the much more stable QCL techniques. For NH3 EC systems are available as well, based on QCL, opto-acoustic and proton transfer mass
spectrometry (PTRMS) but the main problem for these systems is the inlet as already
discussed in the NH3 paragraph above.
Figure 2.12
Eddy covariance measurements at Zegveld over a manured field
Overviews of the EC technique are provided in Baldocchi (2003), or Aubinet et al.,
(2000). There are numerous papers on EC applications for CO2. Examples of measurements for CH4 and N2O with TDLs are described in Verma et al. (1992), Hargreaves
et al. (2001), Kormann et al. (2001), or Laville et al. (1997). QCL experiments and a
discussion on uncertainties in CH4 and N2O EC measurements can be found in Kroon
et al. (2007, 2010).
For methane, Hendriks et al. (2008) presented the first data for an off axis integrated
cavity output instrument in a peat system, Smeets at al. (2009), used the same instrument over a forest. EC measurements are also performed over landfills (Hovde 1995,
Laurila et al., 2005). N2O EC measurements have been mainly carried out in agricultural soil ecosystems (for example Wienhold et al., 1994; Laville et al., 1997; Neftel et
al., 2009; Kroon et al., 2007), and forest soil ecosystems (Pihlatie et al., 2005; Eugster
et al., 2007) Shaw et al. (1998), Famulari et al. (2004) and Whitehead et al. (2008)
have reported eddy covariance measurements of NH3 fluxes
34
2.6.3.3 The relaxed eddy accumulation method
The relaxed eddy accumulation technique is the “cheap version” of the eddy correlation
technique. These sensors are expensive and not available for all trace gas species.
Here the Relaxed Eddy accumulation method provides a way out. This technique uses
the same sonic anemometer that can measure the vertical windspeed for the EC
measurement but samples the up and down drafts into two separate sampling lines.
Gasbags can be used to collect 30 minute averaged up and down going air when CO2,
CH4, N2O or other non reactive gasses are concerned (see chapter 3). For NH3 a wet
chemical sensor or trap can be used. (See chapter 4).
2.6.3.4 Mass balance method
Similar to the gradient technique, this method uses concentration measurements versus height, in combination with the vertical windspeed gradient. But instead of using
the flux gradient parameterization as well, the mass balance method simply calculates
u*c for all heights and integrates this. (Fowler & Duyzer, 1987; Denmead et al. 1998)
This method is applied for finite sources that only stretch out about 4-5 times the height
of the measurement tower in the upwind direction. This method is the bridge between
the micrometeorological techniques and the plume measurements that relies on advection as well. The method is underlying the emission factors for NH3 emissions from manure and CH4 emission from landfills in the Netherlands. (Coops et al., 1995; Oonk &
Boom, 1995; Huijsmans et al., 2001, 2003). Scharff et al., (2003) report on CH4 emission measurements at four landfills using this technique as well as the dynamic and
stationary plume measurement technique described below.
2.6.3.5 Gradient method
This technique uses atmospheric stability depended parameterizations to convert vertical gradients of the concentration measured above an ecosystem into fluxes. In chapter 4 the REA measurements are compared with gradient measurements. The aerodynamic gradient technique uses measurements at different heights in the atmosphere to
determine the vertical gradients of temperature, windspeed and the trace gas concentration. These gradients are related to the degree of turbulence of the atmosphere.
The trace gas concentration gradient is further determined by the exchange that occurs
between the vegetation and the atmosphere. In a very turbulent atmosphere (due to
strong winds or convection during sunny days) the lower part of the atmosphere is well
mixed and the gradients will be small. In a very stable atmosphere with hardly any turbulence (for example during nights with low wind speed) the differences between the
measurement heights will be much larger. A detailed description of this can be found in
Stull (1988) or Fowler & Duyzer (1989). This method is still the reference method for
NH3 deposition measurements over fields and forests (for example Milford et al., 2009,
Mosquera et al., 2001). For methane, Judd et al. (1999) used a flux-gradient technique
to measure CH4 emissions from grazing sheep. Denmead et al. (2000) used this method to measure emissions from grazed areas several km2 in extent.
35
Lopik
Polder area
Grassland
Orchards and
Grassland
200m tower
KNMI
1 kilometer
ECN
River Lek
Amsterdam
The hague
Utrecht
Cabauw
Rotterdam
KNMI 200 m & 20 m tower
387
3000
51
49
98
74
55
63
84
/m 2
1000
500
5
6
57
77
60
14
1500
-6
0
-126
-1000
-168
-166
-126
-105
-1500
-221
-234
-184
-127
-225
-500
-61
-60
-38
-54
-35
-50
Cu m u la t ive
-2000
Best E st im a t e
-2500
P a r a m et r isa t ion on ly
Figure 2.13
9611
9609
9607
9605
9603
9601
9511
9509
9507
9505
9503
9501
9411
9409
9407
9405
9403
9401
9311
9309
9307
9305
9303
-3000
9301
-400
Seasonal variation and net annual emission of CO2 from peat meadow
area in the Netherlands derived from CO2 gradient measurements at
Cabauw (Hensen et al., 1997)
2.6.3.6 Plume measurements method
Sources that have a complex spatial distribution of the source strength within a confined area can be evaluated with plume measurement techniques. This technique
evaluates the concentration plume that originates at the source and is transported with
the wind (Czepiel et al., 1996, 2003, Tregoures et al. 1999). At a transect crossing the
wind direction the concentration is measured. (See fig 2.14). Meteorological data (wind
speed, wind direction, turbulence) and a transport model are used to calculate the
emission level.
36
Accumulated Emission in grCO
139 162
2
208
225
246
224
70
109
129
70
120
63
43
2
Flux in grCO
2000
171
238
165
153
/m 2
125
265
279
2500
300
The plume method can either be executed using a mobile measurement system (dynamic plume measurements, DPM), or using an array of stationary measurement systems or samplers (static plume measurement, SPM). In both cases an atmospheric
transport model is used to evaluate the source strength within site meteorological data
(windspeed and winddirection) as additional input. If possible a tracer will be released
from within the source area, either to enable a dual tracer measurement or to calibrate
the transport model.
Wind
Figure 2.14
Plume method, a mobile measurement system detects the gas plume
that originated from a particular source. Typical distances, depending on
the source strength between measurement transect and plume is in the
range 50m to 2 km.
When the spatial distribution of a tracer source can be made “sufficiently” similar to the
source distribution of the actual source, the dispersion function in both the equation for
the tracer and for the source are equal. The limitation of this method lies in the term
“sufficiently similar”. If the tracer only originates from a particular place within the
source other parts of that source, not in the vicinity of the tracer might experience different dispersion effects. This problem decreases when the distance between source
and measurement transect increases. The atmospheric dispersion model can help to
provide a correction for a non-ideal tracer source distribution on the site.
The plume technique can evaluate the emission for integrated diffuse sources where
the micrometeorological techniques cannot be applied. Landfills are a perfect example
for this. With hills and slopes the variation in the source strength easily shows a factor
10^4 difference in fluxdata when sampling with a chamber method. EC and gradient
measurements cannot deal with the steep slopes and patchy structure of the emission
from different compartments within a landfill and most landfills are so big that the mass
balance method will need a mast that well over 50 m high. Drawback of the plume
technique is that meteorological circumstances and logistics (availability of a measurement transect in a specific wind direction) dictate when measurement can or cannot
take place the wind direction. Plume measurements are covered in chapters 5, 6 and 7.
These case studies subsequently cover landfill measurements; measurements of NH3
plumes at a farm and methane emission measurements from farm sites.
Plume experiments for landfills are described in Czepiel et al., (1996a and 1996b), Savanne et al., (1997), Galle et al., (2001), Scharff et al., (2003, 2009), For NH3 experiment are discussed in Berkhout (2008),
37
2.7
Trace gas challenges
The main complexity of emission evaluation of these sources lies in the spatial and
temporal variability of the emission within the source area. The research questions are
similar for different source types. Examples are grasslands (Flechard et al., 2007), full
farm evaluations (Flesh et al., 2007) or landfills, (Bogner et al., 2003). Somehow emission measurements have to aggregate the emission of various sub sources in the
source area and provide a representative value for the total emission. Each of the trace
gas components has its own specific challenges when emission measurements are
concerned:
CO2 measurements.
Difficulties
• For CO2 the main challenge is to separate the effect from the targeted source from
the effect of other sources on the measured concentration or flux level. However,
the signal from a source is in general a small perturbation on top of this high background value.
• Also, in ecosystems net annual uptake or emission estimates are the result of
much larger amounts of CO2 that are either taken up or emitted in a seasonal and
diurnal pattern. Even if the uncertainty in these two big terms might be relatively
small, the effect on the net exchange is still significant.
• When CO2 uptake measurements are carried out above vegetation the measurement technique should preferably not affect the available solar radiation to that
vegetation. This almost rules out chamber measurement techniques for this application.
Advantages
• With the currently available instrumentations, CO2 concentrations in the atmosphere are relatively easy to measure at a background level that usually varies in
the range 380-500 ppm.
• CO2 is inert, although solution in water can occur, but samples can be stored in
gasbags. Soil respiration measurements can be carried out with chamber techniques.
CH4 measurements.
Challenges
• The CH4 concentration level in the atmosphere is a factor 300 lower as compared
to CO2. This implies that more sensitive (and thus more expensive) sensors are
required.
• Variability in time: CH4 is formed in anaerobic conditions i.e. in the soil, below water, within a manure or compost heap or inside a landfill. Although this production
is normally rather constant (with a modulation through temperature and water
availability) the actual emission at the surface–atmosphere interface occurs both
diffusion controlled and in bubbles.
• System disturbance: On wetlands and peat areas a scientist placing a measurement chamber, animals or farm vehicles can cause vibration of the soil, thereby
triggering the release of methane bubbles that have accumulated in the soil or under sediment in a ditch.
• Variability in space: On landfills emissions occurs through preferential channels,
i.e. cracks in the landfill body. This leads to an enormous spatial variability of the
emission pattern over the landfill area. Changing atmospheric pressure can trigger
or reduce out gassing.
38
Advantages
• In general the changes in the concentration level are relatively large compared to
the background level. Background values in the air are in the range 1.8-5 ppm and
near sources (landfills, cows, manure etc, concentration levels can easily be 10100 ppm.
• CH4 emissions are significantly more abundant than the sparse sites with significant CH4 oxidation, which makes the process evaluation easier as compared with
CO2. Both chamber and micrometeorological techniques can be used.
• CH4 is inert and samples can be stored in gasbags if needed.
N2O measurements.
Challenges
• N2O has the same drawback as CO2 with a large background concentration on top
of which the relevant concentration variations are to be detected. But the concentration level is a factor 1000 lower compared with CO2. For these low concentration
levels, affordable and reliable techniques are still lacking. Due to this, most studies
rely on chamber measurement techniques.
• Variability in time: Emission levels in ecosystems in general consist of very low
background emissions on top of which high spikes occur some time after N fertilization has taken place.
• Variability in space: In grazed fields urine patches are significant sources for N2O
as compared to similar parts of the field without urine.
Advantages
• Like for CH4 only emissions are relevant from a budget point of view although the
occurrence of N2O uptake fluxes is reported occasionally.
• N2O is inert; storage of samples in gasbags however might show artefacts of increasing N2O concentration levels when NOx is present as well.
NH3 measurements
Challenges
• All other gasses are inert trace gasses, NH3 is not inert. Furthermore it is highly
soluble and it quickly adheres to any surface especially when the surface is wet.
• Annual ecosystem exchange will show episodes with significant deposition levels
as well as emission episodes (for example when fertilizer or manure is applied).
• Similar to N2O, easy to use, precise measurement techniques for NH3 for low concentration levels are lacking.
Advantages
• The background of NH3 is almost zero to a few ppb and very low compared to the
concentration levels close to sources (up to 1000 ppb). For spectroscopic measurements however background concentration levels that approach zero concentrations are a problem for line locking.
39
2.8
Different questions, different methods
As demonstrated above, various measurement methods and sensors can be used for
emission studies. The data that is produced by these observational systems is used for
several evaluations:
• what sources are important
• what is the relative contribution of these sources
• where can measures be implemented to reduce emissions (or enhance uptake)
• how can source contributions be added up to national emission levels
Importantly, it is not the source type alone that determines what methods and sensors
should be used, but what type of knowledge we want to obtain is leading when selecting the appropriate methods and sensors.
Process studies
There are studies that provide insight into the processes that lead to or affect trace gas
exchange patterns. This can be studied for example in the laboratory, or on small plots
in a field. Key is that the conditions associated with the flux are as well documented to
facilitate the use of the data for parameterisation or for mechanistic modelling of the
exchange process. Box measurement techniques are well suited for this since for a
small plot the conditions in the soil can be defined. Techniques that are based on spatial integration, for example micrometeorological techniques, are by definition integrating over various sub plots that might have different emission characteristics and are
therefore less suitable for certain process studies.
Budget and scenario studies
These studies aim to get an annual exchange level or an emission factor. Exact understanding on how the emission is generated is less relevant. The flux levels are linked to
average ambient management and meteorological parameters. Spatial and temporal
integrative techniques for this type of evaluation are aiming directly for this data.
Emission verification studies
This type of study includes multi source evaluation. The aim of these studies is to obtain independent information of the cumulative contribution of different source (types)
within an area.
Figure 2.15 shows how the different “questions” (rectangular blocs) need information
on different spatial and temporal scales. The available measurement methods provide
the knowledge at these different scales in time and space.
In the following chapters a selection of studies demonstrates experimental work with
box, micrometeorological or plume techniques that range from m2 to the multi hectare
scale. In Chapter 8 a synthesis will be made on the progress in knowledge in this field.
40
Single source
year
season
Inventories
&
Activity data
registration
Timescale
week
Proces
Studies
Groups of sources
Budget &
Scenario
Studies
Micro meteo
Technique
Gradient
Eddy
day
All
National Emission
Inventory report
Emission
Verification
Studies
Tall
Towers
Plume
hour
Box
min
m2
Figure 2.15
hectare
Spatial scale
regio
land
EU
The role of different methods used in view of the spatial and temporal
variability.
41
Licor CO2 monitor
42
Chapter 3 Eddy Correlation and Relaxed Eddy Accumulation
measurements of CO2 fluxes over grassland
A. Hensen 1, A.T. Vermeulen 1, G.P. Wyers 1 and Y. Zhang 2
1
Netherlands Energy Research Foundation, P.O. Box 1, 1755 ZG Petten, The Netherlands
2
Centre of Environmental Sciences, Peking University, 10087 Beijing, P.R. of China
Received 29 August 1996; accepted 18 November 1996
Abstract. Eddy correlation measurements of CO2 fluxes over grassland on peat soil
were performed from April to October 1994. In this article the measurement system is
described and some results are discussed. The closed path system was compared with
an open path eddy correlation system developed at the Royal Netherlands Meteorological Institute (KNMI) in December 1994. Fluxes deduced from the results of the two
systems agree within 10%. The eddy correlation system was adapted to perform eddy
correlation and relaxed eddy accumulation at the same time. Measurements over a
pasture in the Netherlands showed an agreement between eddy correlation and relaxed eddy accumulation CO2 fluxes within 10%.
Reference: Hensen, A., Vermeulen, A.T., Wyers, G.P., Zhang, Y., 1996: Eddy correlation and
relaxed eddy accumulation measurements of CO2 fluxes over grassland. Phys. Chem.
Earth 21, (5-6) 383-388, 1996.
Licor 6262 used for the Zegveld experiments
43
3.1
Introduction
The direct measurement of the vertical flux of a trace gas by correlating the vertical
wind speed with concentration fluctuations has been described extensively in literature.
The eddy correlation technique requires fast response sensors for the constituent of
interest. Such instruments are available for H2O, CO2 and various other gasses. For
VOC's or pesticides however no fast response sensors are available. For these constituents eddy accumulation is a potentially useful method. With this technique air is
sampled in a separate container depending on the direction of the vertical wind velocity. The difference in concentration in the two air samples is related to the flux. This
concept was proposed by in Desjardins (1972). For eddy accumulation however sampling speed must be proportional to the vertical wind velocity. This is technically difficult. Businger and Oncley (1990) proposed a "relaxed" version of eddy accumulation
using a constant sampling speed. The relation between the concentration in the downdrafts (Cdown) and in the updrafts (Cup) is related to the flux (F) by
F = βοσw (C↑ − C↓)
(1)
In this formula bo is an empirical constant with a value of about 0.6 (Businger and Oncley, 1990). In order to increase the difference in Cup and Cdown sampling only takes
place whenever the absolute value of the vertical windspeed exceeds a certain deadband velocity wo. Using a zero deadband value both concentrations will approach the
mean concentration. Normally wo is set between 0 and σw. When choosing wo too large
only the large eddies will be sampled, if any air is sampled at all. The factor b has to be
corrected for the choice of the deadband velocity. This correction was derived by Businger and Oncley to be:
b = bo. exp(-0.75 * σw/wo)
3.2
(2)
Eddy Correlation Setup
3.2.1 Eddy Correlation: Technical description
A schematic overview of the eddy correlation system is shown in figure 3.1. This setup
was used from April to October 1994 at (Regionaal Onderzoek Centrum) R.O.C. Zegveld, an agricultural research centre in the centre of the Netherlands. CO2 fluxes over
grassland on a peat soil were monitored continuously during this period (Hensen et al,
1995).
A sonic anemometer (A.T.I. SWS-211/3K) was mounted on an open frame tower at 4
m above ground level. The anemometer was facing south and the fetch was >400 m for
the wind sector 60°-330°. For the wind sector 330°- 60° the measurement tower, which
was also equipped with other instruments, disturbed the flow.
44
Windspeed & direction
Temperature
Sonical
Anemometer
Air inlet
Windspeed
Temperature
Span
Zero
Cal.gas
Sonic &
NDIR data
MFC
ADC
MFC
10Hz
Data
measure
Auto cal
Figure 3.1
NDIR
reference
p
Experimental setup for eddy correlation measurements, used at Zegveld
(NL)
The sampling inlet was located near the sonic. Air was transported to the base of the
tower through 4m insulated (PE) tubing. The flow was kept at 7.5 l.min-1 using a mass
flow controller. This flow resulted in a delay time of 1.9 seconds between the sonic data
and the CO2 data. This delay time was checked to be constant during the experimental
period by reprocessing of the raw data. This check is necessary since a mismatch between CO2 and sonic data of about 0.1 second can reduce the calculated flux by 1530%.
The CO2 concentration measurements were performed using a fast response nondispersive infrared monitor (LICOR, Li-6262). N2 was used as reference gas with a 50
ml/min flow. Atmospheric pressure, temperature and H2O concentration were measured simultaneously. The CO2 concentrations were corrected for density fluctuations
and pressure effects.
Automatic calibration using N2 for zero calibration and CO2 in air for span calibration
was performed every day at 9:00 GMT in the morning. Zero drift of the monitor was
generally below 1 ppm/day. Span drift was below 0.5 ppm/day. In the design of the system the pump is located in front of the NDIR. The outlet of the measurement cell is
open to the atmosphere. Therefore pressure in the cell is unaffected when switching
from measurement to calibration operation or vice versa. In both cases the atmospheric
pressure can be used for correction.
Other parameters measured at Zegveld.
Temperature and windspeed measurements took place at 1, 2 and 5 m height. Both
photo synthetically active radiation (PAR) and global radiation were measured at the 1,
45
0.5
0.0
-0.5
-1.0
25
300
10
200
5
100
0
CO2 flux
Figure 3.2
Temperature at 4m
15-5-94
14-5-94
13-5-94
12-5-94
11-5-94
10-5-94
9-5-94
8-5-94
7-5-94
6-5-94
5-5-94
4-5-94
0
PAR at 1.5m
Example of eddy correlation results for May 1994
850
900
950
3000
1000
050
1
1100
1150
Open path w’CO2 ’
2000
Rawdata
datafor
forboth
bothsensors
sensors(mV.m.s
(mV.m.s-1)
Raw
1000
0
-1000
1000
-2000
Closed path w’CO2 ’
1500
1000
500
0
-500
-1000
850
850
Figure 3.3
900
900
950
950
1000
1000
1050
050
1
1100
1100
1150
1150
Example of raw data obtained with open path and closed path systems.
Time in data points (20 data points = 1 sec).
3.2.2 Eddy Correlation Results
An example of the performance of the E.C. system is shown in figure 3.2. The CO2
fluxes over grassland showed a clear diurnal variation. Daytime uptakes of CO2
reached up to 0.7 mgCO2.m-2s-1 and night time emission up to 0.4 mgCO2.m-2s-1. On
most days the CO2 fluxes anti-correlated with available photosynthetical active radiation (PAR). This can be seen clearly for May 3-12. On May 13 and 14 however high
PAR levels were observed whereas the CO2 uptake was significantly reduced. The reduced uptake was found to be related to harvesting on upwind plots. Similar events
46
PAR (Wm-2 )
400
15
3-5-94
Temperature (ºC)
CO2 flux (mg m-2 s-1 )
2 m height south of the main mast. Furthermore detailed information was available on
fertilizer programmes, harvesting periods and grazing. The ground water table was at 60 cm in the wind sector between 90° and 180° between 90° and 180° degrees wind
direction. In the sector 180°-270° the ground water table was at -30 cm.
were detected showing emission of CO2 immediately after manure spreading. A more
detailed analysis of the data obtained at Zegveld is beyond the scope of this article and
will be presented elsewhere. (Hensen et al. 1995), (Villoria et al.1996).
3.3
Open and Closed Path Intercomparison Experiment
3.3.1 Technical Description
In order to check whether the eddy correlation measurements had to be corrected for
loss of high frequency CO2 variations, an intercomparison experiment was performed.
This experiment took place at Cabauw (experimental site of the Royal Netherlands Meteorological Institute, KNMI) in the centre of the Netherlands, about 30 km south of
Zegveld. The experiment took place in December 1994. Both open and closed path
measurement systems were mounted on a 4 m mast. KNMI used a 20 cm open path
sensor for CO2 and 1-I20 measurements (IFM, KNMI) (Kohsiek, 1991) and a sonic
anemometer (DAT-300, TR61A Kaijo Denki).
The ECN measurement system was located just behind the KNMI sensors using a
sonic anemometer (Gill Solent). Both units had their own data acquisition system. The
analogue signals of the open path sensor (KNMI) were also logged by the ECN system
in order to check for differences in processing of the data.
Additional meteorological data (wind, temperature, humidity profiles, and radiation) is
available from the routine measurements at the site (Monna and v.d.Vliet, 1987).
3.3.2 Results of the Intercomparison Experiment
Both the open and closed path systems worked well, apart from periods with rain, when
the open path sensor does not work. An example of raw data from the two CO2 sensors
is shown in figure 3.3. From this figure it can be concluded that the closed path system
shows no significant loss of information as compared to the open path system.
0.20
0.15
Fluxes in mgCO2/m 2s
0.10
0.05
0.00
-0.05
-0.10
-0.15
-0.20
-0.25
-0.30
1-12-94 0:00
Figure 3.4
1-12-94 12:00
2-12-94 0:00
2-12-94 12:00
3-12-94 0:00
Date
Closed path(ECN)
3-12-94 12:00
4-12-94 00:00
Open Path (KNMI)
Simultaneous recorded CO2 fluxes at Cabauw
47
0.25
Flux mgCO2 m-2 s-1 open path system (KNMI)
0.20
0.15
0.10
0.05
0.00
-0.05
-0.10
-0.15
-0.20
-0.25
-0.30
-0.30
-0.20
-0.10
0.00
0.10
0.20
0.30
Flux mgCO2 m-2 s-1 closed path system (ECN)
Figure 3.5
Linear regression obtained for 272 data points
The flux measurements for the period December 4-7 are shown in figure 3.4. During
the measurement period the weather was sunny and mild. Therefore the grass still
showed photosynthetical activity. Uptake during daytime reaches up to 0.2 mgCO2.m2 -1
s . From November 24 to December 5,272 data blocks of 30 minutes were obtained
for which both systems operated simultaneously. The results of open versus closed
path measurements of the flux are plotted in figure 3.5. Linear regression shows good
agreement between the two methods:
Fopen path KNMI = 0.99 * Fclosed path ECN
r2=0.78 (n=272)
The good result of the regression analysis is mainly determined by the high uptake values on December 2-5 and 18-19. For individual days the correlation can be lower, especially when the fluxes are small. Generally the two systems agree within 10%. This is
well within the expected accuracy of the individual systems. For a single flux value
w'CO2' the error was estimated from the resolutions of both the sonical anemometer
and the NDIR. For w' the error is typically 20% and for CO2 the error was between
100% and 10% for the small fluxes and the large fluxes respectively. Therefore the error in an individual flux value was 120% for the small flux values. In summer conditions
with high fluxes the error will go down to about 20%. For 30 minutes of averaging, 600
individual points are accumulated. The measurement error in the averaged flux will
then be about 4%. The variation of the flux within a 30 minute period will be larger than
that. Three 10-minute-averaged flux values will show 30-50% variability within a 30
minute period. Compared to this variability the agreement between the two sensors
was very good.
The conclusion from this experiment was that the damping in the tubing system of the
closed path system is probably only affecting those frequencies that represent eddies
with a diameter smaller than 20 cm. Smaller eddies will also be neglected in both the
open path CO2 sensor and in the sonic anemometers. During the experiment wind
48
speed was below 6m/s. In order to determine the damping factor for situations with
more turbulent conditions (when the eddy spectrum shifts to higher frequencies) an intercomparison in summer time would be useful.
We did not assess the corrections for loss of the flux due to sensor response, sensor
separation etc., as described in Moore (1986), which are needed to obtain the actual
flux. For the intercomparison we assume these errors will be the same (within 10%) for
both systems.
3.4
A Combined E.C. and R.E.A. System
3.4.1 Technical Description
The combined E.C. and R.E.A. setup is shown in figure 3.6. The sonic anemometer
data was used to compute a running mean value for the vertical windspeed (w). Whenever the difference between the running mean and the actual vertical windspeed exceeded the deadband velocity, the up or downward valve was selected. This procedure
including sonic anemometer measurement, data transfer to the computer, processing
by the program and actuating the valves, took to be about 80 ms.
The two valves were connected to the main sample flow that was used for the eddy
correlation measurements at 54 cm downstream of the inlet. The transport time for the
gas to travel from the inlet to that point was also 80 ms. In this way the correct air volume was sampled and there was no dead volume in the inlet line of the r.e.a, system.
Eddy correlation measurements were performed for 20 minutes. Meanwhile conditional
sampling was done and two bags were filled with 7-8 litres of sample air each. The
NDIR was used to measure the concentrations in both gas bags and then the two bags
were evacuated. This procedure takes 10 minutes in total, after which the eddy correlation measurements were resumed. The revised inlet changed the delay time for the
eddy correlation measurements by 0.1 second to 2.0 seconds (see section 2.1).
Figure 3.6
Scheme of the eddy accumulation setup
3.4.2 Results of the combined E.C. - R.E.A. system
The relaxed eddy accumulation setup was used over a grass pasture. Experiments
were performed with two setups. During the first period a Solent anemometer was used
on a 4 m high mast. Data was obtained on 4 days, May 22-23 and May 30 -June 1,
49
showing agreement of both systems within 20%. In the second period the system was
slightly modified and an ATI sonic was used at 2.5 m height. Just after the start of the
experiment the grass at the site was mowed. Therefore CO2 fluxes showed a small
emission only. Undisturbed measurements were obtained when the grass was removed. The results for these last 3 days of the experiment are shown in figure 3.7.
Both methods seem to agree well. A linear regression for the data from June 4-7 is
shown in figure 3.8. The result is:
r2=0.9
Frea =Fcc* 0.9 - 0.01
For these measurements a deadband velocity of 3 cm/s was used and the factor b was
corrected using (2). The concentration difference between the two bags was between
0.1 and 1.5 ppm. The error in the R.E.A. results is mainly deter-mined by the error in
the concentration difference. This difference can be determined within 0.1 ppm, which
will give an error of about 100% for the night time emission values and about 20% for
the day time uptake values. The 10% difference between the eddy correlation and relaxed eddy accumulation results is therefore not very significant. Further improvement
of the result will require a higher accuracy for the measurement of the CO z concentration difference.
CO2 Flux (mgCO2 m-2s-1)
0.4
0.3
0.2
0.1
0
-0.1
-.2
-0.3
-0.4
5-7-95 12:00
Figure 3.7
6-7-95 0:00
6-7-95 12:00
7-7-95 0:00
EC flux
7-7-95 12:00
R.E.AC flux
Example of simultaneously obtained eddy correlation and eddy accumulation CO2 fluxes
The main reason for the good agreement between both methods is the configuration of
the sampling inlet system which avoids otherwise necessary corrections for delay between sonic anemometer and CO2 signal while no dead volume is introduced.
3.5
Conclusions
Both the comparison between the open and closed path eddy correlation measurements and the comparison between eddy correlation and relaxed eddy accumulation
measurements were successful. The combination for CO2 flux measurements showed
that the relaxed eddy accumulation method worked satisfactorily. The described system can also be used for any other constituent that can be analyzed in the two sample
bags. Simultaneous measurement of E.C. and R.E.A. fluxes for one constituent (for
example CO2) provides a powerful tool for quality control of the system.
50
Acknowledgement.
We want to thank Wim Kohsiek, Fred Bosveld and Frans Rennes of KNMI for their help
in the intercomparison experiments, both with the setup on the Cabauw site, and the
data analysis. We also want to thank Han Möls for his useful suggestions in the development of the eddy accumulation setup.
0.4
0.3
R.E.A. CO2 Flux (mgCO2 m-2 s-1)
0.2
0.1
0
-0.1
-0.2
Fit results
FREA=FEC* 0.90-0.01
R2=0.9
-0.3
-0.4
-0.4
-0.2
0
0.2
0.4
Eddy Correlation CO2 Flux (mgCO2 m-2 s-1)
Figure 3.8
Linear regression applied to data July 5-7, 1995
51
ECN REA system at Braunschweig 2000
52
Chapter 4 Inter-comparison of ammonia fluxes obtained using the
Relaxed Eddy Accumulation technique
A. Hensen1, E. Nemitz2, M.J. Flynn3, A. Blatter4, S.K. Jones4, L.L. Sørensen5,6,
B. Hensen5, S. C. Pryor5,7 ,B. Jensen6, R.P. Otjes1, J. Cobussen1, B. Loubet8,
J.W. Erisman1, M.W. Gallagher3, A. Neftel4 and M.A. Sutton2
1
ECN, Petten, Netherlands
Centre for Ecology and Hydrology (CEH), Edinburgh, UK
3
University of Manchester, Centre for Atmospheric Science, Manchester, UK.
4
Agroscope Reckenholz-Tänikon Research Station ART, Zürich CH.
5
RISOE National Laboratory, Roskilde, DK
6
National Environmental Research Institute, Roskilde, DK
7
Indiana University, Bloomington, Indiana, U.S.A
8
INRA Thiverval-Grignon, FR
2
Abstract. The exchange of Ammonia (NH3) between grassland and the atmosphere
was determined using Relaxed Eddy Accumulation (REA) measurements. The use of
REA is of special interest for NH3, since the determination of fluxes at one height permits multiple systems to be deployed to quantify vertical flux divergence (either due to
effects of chemical production or advection). During the Braunschweig integrated experiment four different continuous-sampling REA systems were operated during a period of about 10 days and were compared against a reference provided by independent
application of the Aerodynamic Gradient Method (AGM). The experiment covered episodes before and after both cutting and fertilizing and provided a wide range of fluxes 60 – 3600 ng NH3 m-2 s-1 for testing the REA systems. The REA systems showed moderate to good correlation with the AGM estimates, with r2 values for the linear regressions between 0.3 and 0.82. For the period immediately after fertilization, the REA systems showed average fluxes 20% to 70% lower than the reference. At periods with low
fluxes REA and AGM can agree within a few %. Overall, the results show that the continuous REA technique can now be used to measure NH3 surface exchange fluxes.
While REA requires greater analytical precision in NH3 measurement than the AGM, a
key advantage of REA is that reference sampling periods can be introduced to remove
bias between sampling inlets. However, while the data here indicate differences consistent with advection effects, significant improvements in sampling precision are essential
to allow robust determination of flux divergence in future studies. Wet chemical techniques will be developed further since they use the adsorptive and reactive properties
of NH3 that impedes development of cheaper optical systems.
Braunschweig REA experiments 2000
Reference: Inter-comparison of ammonia fluxes obtained using the relaxed eddy accumulation
technique. Hensen, A., E. Nemitz, M. J. Flynn, A. Blatter, S. K. Jones, L. L. Sørensen, B.
Hensen, S. Pryor, B. Jensen, R. P. Otjes, J. Cobussen, B. Loubet, J. W. Erisman, M. W.
Gallagher, A. Neftel, and M. A. Sutton. Biogeosciences, 6, 2575-2588, 2009.
53
4.1
Introduction
Eddy covariance (EC) measurement of trace gas fluxes is becoming more and more
popular, and is in principle preferred to gradient methods due to the fundamental robustness and lack of empiricism in the EC approach. For CO2, H2O, N2O, CH4 and O3,
for example, fast sensors are available that enable EC flux measurements (Moncrieff et
al., 1997; Pattey et al., 2006, Mészáros et al. 2009). For a number of trace gases,
however, such as ammonia (NH3), sensors that can provide concentrations at the required sampling rates of above 1 Hz and with sufficient detection limit (<0.1 µg m-3 for
NH3) are becoming available but remain expensive (Shaw et al., 1998; Fehsenfeld et
al., 2002, Famulari et al., 2004; Whitehead et al., 2008). Especially for these components, the Relaxed Eddy Accumulation (REA) technique (Businger & Oncley, 1990;
Zhu et al., 2000) can provide a means to determine vertical fluxes without requiring the
measurement of vertical concentration profiles. Just like the EC technique, REA does
not require the stability corrections that are needed for the aerodynamic gradient
method (AGM), although it still relies on an empirical parameterization. In addition, as
with EC, REA derives the vertical flux from measurements at a single height above the
surface. This avoids the potential errors in gradient methods arising from the fact that
the flux-footprint (e.g. Korman and Meixner, 2001) is different at different heights. Flux
measurement at one height also has the advantage that REA measurements at several
heights could theoretically be used to investigate the potential for divergence in the vertical flux, occurring for example due to chemical production/consumption in the surface
layer or due to advection from nearby sources. These issues are of topical interest for
NH3 (Sutton et al., 2001), with the REA technique providing the prospect for direct
measurement of these effects. In addition, REA has the potential to be applied in logistically difficult situations, such as on aircraft (Delon et al. 2000; Zhu et al. 1999) or urban flux towers.
Measurements of ammonia are hampered by the fact that the NH3 molecule quickly
adheres to surfaces of different kinds. Especially wet surfaces in inlet-tubes, sample
cells etc. can be a sink or, when a wet layer evaporates, a source of NH3. The high
solubility of NH3 in water is related to the high polarity of the NH3 molecule. The amount
of NH3 that can dissolve in water is further increased because NH3 reacts into NH4+.
Until now, chemical adsorption techniques have therefore been most often used to determine atmospheric NH3 concentrations for flux measurements, with the most advanced being continuous wet-denuder methods (Kruit et al. 2007; Neftel et al., 1998;
Blatter et al., 1993; Wyers et al., 1993). The time response of these chemical systems
is improving, but it is unlikely that it will become sufficient to enable EC measurements.
However, the sampling frequency required for REA is much less than for EC, with the
result that several research teams have developed REA systems for NH3 that use either denuder filterpack combinations (Myles et al., 2007; Meyers et al., 2006; Ham &
Baum, 2007) or denuder continuous wet chemical detection systems (Neftel et al.,
1999; Erisman et al., 2001; Nemitz et al., 2001; Sørensen et al. 2003). Furthermore,
inlet losses can significantly limit the response time for EC NH3 systems resulting in
uncertainties in flux estimates in the range of 0 - 40% that are comparable to those
from REA (Whitehead et al., 2008; von Bobrutzki et al., 2009). The present study provides an inter-comparison based on these four REA implementations made within the
context of the Braunschweig Integrated Experiment which was conducted as part of the
EU GRAMINAE project. An overview of the experiment is given by Sutton et al.
(2009a). Reference estimates for the time-course in NH3 concentrations and fluxes
were provided by the AGM from a parallel inter-comparison of systems measuring vertical concentration profiles of NH3 Milford et al. (2009). The inter-comparison presented
here provides a basis to analyze the robustness of the REA approaches for continuous
NH3 flux measurement, as well as to identify recommendations for further improve-
54
ment, while the magnitude of the fluxes is discussed elsewhere (Milford et al., 2009;
Sutton et al., 2009b).
4.2
Material and methods
4.2.1 Micrometeorological theory
In the EC approach the instantaneous fluctuation of the concentration (χ') of each eddy
of air is related to the instantaneous vertical velocity (w') of the eddy, such that, at its
simplest within the atmospheric turbulent boundary layer over an extensive, uniform
surface, the vertical flux (Fz) at a location may be given as (see Lee et al., 2004 for details):
Fz = χ '.w'
(1)
It is the requirement to sample each eddy of air contributing to the flux that leads to fast
instrument response times being necessary for EC measurements. By contrast, the
REA approach is based on the relationship between Fz and the difference between the
average trace gas concentration of upward and downward moving eddies of air. This
requires fast response switching between air sampling of the up- and down-drafts, but
only slow response (~15 minute to 2 hours) sampling of the trace gas concentrations,
such that:
Fz= βσw (χ↑ − χ↓)
(2)
where χ↑ and χ↓ are the average concentrations in the up- and down-drafts, respectively, σw is the standard deviation of the vertical wind velocity (derived from fast response measurements of w'), and β is an empirical dimensionless parameter which
may be assumed to be a constant or estimated from measurements of fluxes of other
scalars (Pattey et al., 1993). Values of β coefficients are shown to vary between 0.40 to
0.63 (Milne et al., 1999). When applying the REA methodology, the β factor can be estimated from turbulent measurements of temperature (T) or momentum. In that case
values of, for example, T↑ and T↓ are calculated in an on-line simulation. Then β is calculated by analogy to equation [2] setting Fz equal to the value of the sensible heat flux
(H) that is obtained from the covariance of T' and w'.
The difference in concentration (χ↑ − χ↓) can be increased by introduction of a “deadband”, such that air is not sampled when w' is near zero, with sampling only taking
place when |w'| > D. In case a constant β coefficient is used in the flux calculation a
correction should be applied to the empirical uncorrected β factor (βo) to account for
this deadband (after Pattey et al., 1993):
βcorrected = βo {1-βo[1- exp(-b1* D/σw)] }
(3)
With D the deadband value, σw the standard deviation of the vertical wind speed,
βo=0.4 and b1=1.9.
55
A
D
H
Ma
Ml
P
S
V
W
air pump
detector
heater
air mass flow meter
liquid mass flow meter
peristaltic pump
sonic anemometer
valve
wet effluent diffusion denuder
1 absorption solution
2 standard
3 reagent
4 waste
A
B
D
Ma
Ml
P
S
V
FAL
Ml
A
A
Ml
air
Ma
air
V
V
T:
C:
Ml:
PPD:
S:
AD:
Vl:
Va:
CO:
CB:
2
1
2
3
4
W
V
PP
V
H
H
P:
D:
temperature detector
conductivity detector
liquid mass flow meter
parallel plate denuder
ultrasonic anemometer
active debubbler
3-way liquid valve
3-way air valve
critical orifice
control box of AMANDA
analyzer
peristaltic pump
detector block
1
2
3
4
NaHSO4 solution
NaOH + NH4+
deionized water
waste
ECN
V
2
M: Fluormonitor
air
CEH
RISOE
H: Heating coil
CO
VVa VVa
M
P
I: Air inlet
R: reagents
Vl
I I
V
L: sampling liquid
4
AA
Ml
S
S
D
A
W
A: Air pumps
PPD
P
H
V: Vials
AD
AD
W: WEDD
S: sonic anemometer
3
S
4
P: peristaltic pump
CO
T
C
2
P
D
3
DD
Vl
1
B
V
absorption solution
waste
hydroxide solution
deminerized water
1
S
3
V
Ma
A
air
Ml
4
1
air pump
membrane sampler
detector
air mass flow meter
liquid mass flow meter
peristaltic pump
sonic anemometer
valve
R
R
L
S
CB
Figure 4.1
Schematic diagrams of the four continuous relaxed eddy accumulation
systems applied in the present intercomparison, as developed by: FAL
(Neftel et al., 1999), ECN (Erisman et al., 2001), CEH/UMIST (Nemitz et
al., 2001) and RISOE (Sørensen and Jensen, 2004).
4.2.2 Measurement site
The flux measurement site was an intensively managed experimental grassland of 12
ha situated on the grounds of the Federal Agricultural Research Centre (FAL), Braunschweig, Germany (N 52° 18' E 10° 26'). Directly adjacent to the field are an experimental dairy farm of the FAL and a station of the German Weather Service (Deutscher
Wetterdienst). The measurements reported here were made during the GRAMINAE Integrated Experiment (21 May – 15 June 2000). Canopy height before the cut was
0.75m. The grass of the field was cut to 0.07 m canopy height on 29 May and the grass
was removed on May 31.
Fertilisation of the field with calcium ammonium nitrate took place on 5 June, being applied with a dose of 108 kg N ha-1. The farmer irrigated the fields on 15–17 May, before
the experiment started. Weather conditions were mixed during the experiment, with in
excess of 5mm precipitation on 7 days, and maximum temperatures ranging from 14°C
on 29 May to 31°C on 10 June. Mean daily windspeeds during the experiment ranged
for 2.0–5.4 m s−1 (overall mean 3.4 m s−1), with mean daily relative humidity in the
range 49–83%.
A plan of the experimental site and other details of site conditions are provided by Sutton et al.(2009a). The REA systems were located near the centre of the experimental
field at the main sampling site of the Integrated Experiment (Site 1 of Sutton et al.
2009a), where the fetch exceeded 150 m for wind sectors 20-180 ° and 190-360°. A
detailed assessment of vertical flux divergence at the site has been provided else-
56
where, covering both storage and advection effects (Loubet et al., 2009; Milford et al.,
2009), as well as the effects of chemical production and consumption in the surface
layer (Nemitz et al., 2009a), with an overall synthesis of the different issues being provided by Sutton et al. (2009b).
4.2.3 Measurement set-up
The positions of the four REA systems spanned a total range of 15 m on a north-south
line at Site 1. Fetch conditions were similar for all samplers, with the exception of obstruction of fetch in the immediate vicinity by other equipment, for which account was
taken in filtering the results. The schematic outlines of each of the four systems used
and a summary of the differences and similarities of the systems are shown in Fig.4.1
and Table 4.1 respectively. In all systems an ultra-sonic anemometer was used to determine σw and drive the sample switching between χ↑ and χ↓. An evaluation of the friction velocity (u*) estimates and sensible heat fluxes (H) derived by EC from each of
these sonic-anemometer systems during the experiment and comparison with other
parallel estimates has been provided by Nemitz al. (2009b).
4.2.3.1 Detection and response characteristics
The CEH/UMIST and ECN systems use the same analytical principle as the AMANDA
(Ammonia Measurement by Annular Denuder sampler with Analysis, ECN Netherlands) (Wyers et al., 1993). In this approach, NaOH is added to a continuous flow of an
aqueous solution containing NH4+ from NH3 in the air-sampling stream. The aqueous
sample is passed over a hydrophobic PTFE membrane behind which is passed a
counter-flow of deionised water. At the pH of the sample, NHx is present in solution as
NH3(g), which passes the membrane where it reverts to NH4+. The NH4+ ions are then
measured by electrical conductivity of the counter-flow solution. In the implementation
of the CEH/UMIST system, both the air-flow and the liquid flow are a factor ten higher
than the ECN system. The result is that resolution in concentration is similar, with a detection limit of about 50 ng m-3.
The FAL and RISOE systems use a fluorescence detection technique (Blatter et al.,
1993; Sørensen et al., 1994). The analytical detection is based on the reaction between o-phthaldialdehyde, a reducing agent and ammonia to produce a fluorescent
compound (Rapsomanikis et al., 1988). With this system a detection limit of 10 ng m-3
NH3 can be obtained. The fluorescence detector used in the FAL system is sufficiently
stable to allow two separate detectors to be used to determine the concentrations simultaneously in the up- and down-channels. The CEH/UMIST, RISOE and ECN system use one common detector to obtain the concentration for both the up and the down
sampling system, in order to avoid differences between individual detectors. In the
case of the ECN system the concentration collected in the ”up” sampler is measured in
a 10 min period, followed by subsequent analysis of the sample obtained in the “down”
sampler. The liquid sample of the latter is passed through a delay-loop so both concentrations correspond to the same air mass. A similar approach is taken with the
CEH/UMIST detection system, although, due to the higher flow rates, this has a faster
response time compared with the ECN system, permitting a switching interval between
up- and down-air samples of 2.5 minutes. The liquid-phase transport time between the
denuders and the detector of 14 minutes in the ECN system is taken into account when
the concentrations are related to the turbulence. In the CEH/UMIST system the results
of 3 switching cycles are block-averaged to provide 15-minute fluxes. In the RISOE
system the liquid from the two WEDDS are sampled in two valves from where it is injected with a switching interval of 7.5 minutes into the fluorescence detector.
57
Table 4.1
System specifications of the different REA sampling methods.
CEH/UMIST
ECN
FAL*
RISOE
Sampling Inlet
Vertical parallel-plate de- Membrane chambers
nuders
Mini-WEDDs
Measurement height
above ground (m)
Inlet Distance to ultrasonic anemometer
(mm)
Ultra-sonic anemometer
type
Inlet system and length
2.09
1.4
1.15
Stainless
steel
inlets with subsampling to vertical
wet denuders
1.95
300
50
50
500
Gill Solent 1012RA
Gill Solent 1012RA
Deadband value of w’
(m s-1)
Switch delay to account
for common inlet line
Air sampling rate
(l min-1)
Liquid flow rate to detector (ml min-1)
Switching delay time for
liquid phase transport
(min)
Detector
Detection limit (ng m-3)
Chemical detector response time (min)
Chemical detector cycle
of up- and down air
samples (min, min)
Calibration type
Switching pattern for
up-down in reference
sampling mode
Cycle of sampling and
reference sampling
modes for automatic
bias correction
Data acquisition software
Manual performance
check
* FAL-CH is now ART
Separate inlets for χ↑ and Common
mm
χ↓, 10 mm
inlet,
Gill
Solent
1012RA
150 Separate inlets
for χ↑ and χ↓, 80
mm
none
Metek.
Separate inlets for
χ↑ and χ↓, 500 mm
0.5 σw
0.05
0.05
None
1.5 s
None
None
12
1.25
0.7
0.8
-
0.1
10 (denuders subsample at 1 l min-1)
-
14
15
Selective membrane diffu- Selective membrane
sion & conductivity
diffusion & conductivity
50
50
2
5
ophthaldialdehyde
fluorescence
10
5
up 2.5, down 2.5; block- up 10, down 10
up 10, down 10
averaged over 3 cycles to
provide 15 min fluxes
Aqueous standard [NH4+] Aqueous
standard Aqueous stansolutions
[NH4+] solutions
dard [NH4+] solutions
2 Hz
W’ time shift
Random similar
to w
7 hr, 1 hr
10
5
up7.5, down 7.5
Aqueous standard
[NH4+] solutions
1 Hz
7 hr, 1 hr
LabView (National Instru- Delphi (Borland)
ments, Austin, TX)
6h
o-phthaldialdehyde
fluorescence
12 h
Lab view (National
Instruments, Austin,
TX
6h
DAQ-SYS
National
tory)
(Risoe
Labora-
6h
4.2.3.2 Reference mode sampling
In order to increase the sensitivity of the systems, the ECN, FAL and CEH/UMIST systems were operated in a reference sampling mode automatically for 0.5 – 1 hour every
6 or 7 hours to establish minor systematic biases between the two channels. The
RISOE system was operated in a reference mode for 3 hours at the end of the field
campaign. In this reference mode, the air is sampled in a manner that is uncorrelated
with vertical windspeed, so that the NH3 concentrations should be equal in both the upand down-draught sampling inlets. These measurements provide useful information on
the performance of the system. The difference found between the channels is used to
correct for biases between the up- and down-draft sampling inlets.
For switching in reference mode, the CEH/UMIST system uses a fixed 2 Hz switching,
opening and closing both the up- and the down-inlet at the same time. The RISOE system uses a 1 Hz switching between up and down inlet. By contrast, the ECN system
58
uses a delayed windspeed signal so that the random structure of sampling is maintained. The concentrations obtained with respectively up- and down-sampling systems
are then corrected so both end up at the average of the two levels:
χcorrected↑ = χraw↑ * (χref↓ + χref↑)/(2χref↑)
χcorrected↓ = χraw↓ * (χref↓ + χref↑)/(2χref↓)
(4a)
(4b)
where χraw and χref are the uncorrected concentration level and the concentration level
obtained in reference mode for either the upward or down-inlet, respectively.
4.2.3.3 Use of a dead-band
Since a very small deadband of |w'| < 0.05 m s-1 was chosen for the CEH/UMIST system, only a small fraction of time was spent on the deadband and the concentration in
this denuder tended to be below the detection limit. Thus the deadband denuder was
deactivated for this study. Similar, the ECN system did not sample whenever |w'| < 0.05
m s-1. By contrast, a “dynamic” deadband (|w’| < 0.5 σw) deadband was used by the
RISOE system. The benefit of using a deadband as a function of σw is that the β coefficient found from the measurements does not need an empirical correction (Businger
and Oncley, 1990). Furthermore this ensures that a sufficient amount of NH3 will always be sampled on the up- and down-WEDDs.
The overall sampling cycle for the ECN system was 20 minutes and for the RISOE and
CEH/UMIST system of 15 minutes. The FAL system used no deadband and a 10 minute cycle.
4.2.3.4 Gradient reference
The measurements by four NH3 vertical profile measurement systems were used with
the AGM approach to obtain the concentration at 1 m height (z) above the zero plane
displacement (d) and the reference flux (Milford et al. 2009). Values of u* for the AGM
flux estimates were provided by the inter-comparison of Nemitz et al. (2009b) . When
comparing the best profile estimate of NH3 concentration with the REA estimates, the
difference in height needs to be accounted for. Therefore the gradient estimates of (zref)
= 1 m above d and the gradient flux estimate were used to obtain height-corrected gradient-reference concentrations (χAGM) at the heights of the various REA inlets:
χAGM(zREA) = χ(1 m) - Fz, AGM .[Ra(zREA) - Ra(1 m)]
(5)
Here the term Ra(zREA) - Ra(1 m) represents the aerodynamic resistance (Ra) between
the level zref and zREA for each REA system, with the Ra values calculated from the consensus micrometeorological estimates of Nemitz et al. (2009b). The measured concentration levels obtained from each REA system were either obtained by simply averaging χ↑ and χ↓ or, for the CEH/UMIST dataset, as a weighted average of χ↑ and χ↓ using
the actual time spent on sampling the up- and down-directions.
The comparison of the REA flux estimates and the AGM flux estimates was made in
two stages. Firstly, the measured concentration values of the REA systems were compared with the AGM reference at the respective REA heights. The AGM reference flux
estimates were then compared with the REA flux estimates that were normalized for
the difference in REA and corresponding AGM concentrations. The purpose of this
second stage was to distinguish methodological errors in the flux due to the REA approach, from those that were actually due to propagated errors arising from the basic
determination of (zREA). Since the four gradient measurement systems were not consistent about the concentration levels on June 3,8,9 and 10, alternative estimates of
59
both the AGM flux and AGM concentration were proposed by Milford et al. (2009) and
these are discussed in the following sections.
4.3
Results
4.3.1 Data coverage
The data coverage of the REA measurements is listed in Table 4.2. The CEH/UMIST
system measured fluxes continuously from May 25 to June 14, while the ECN and FAL
systems were used for other purposes prior to cutting of the field. The FAL system unfortunately malfunctioned during a substantial part of the campaign. The ECN system
stopped earlier than foreseen due to data acquisition problems. The table indicates the
number of 15-minute intervals for which measurements were foreseen (maximum) and
the number of 15-minute intervals with data. The data selection excluded data with
windspeeds below 1 m s-1 and with wind directions outside the sector 180-350°. Especially when using a deadband, (CEH/UMIST, ECN and RISOE), the amount of air sampled during these periods is very small, leading to very low concentrations in the liquid
flow and large uncertainties in the calculated concentration level.
Table 4.2
Data Coverage of the different REA sampling methods.
CEH/UMIST
FAL
RISOE
ECN
Start of measurements (GMT)
25/5 18:45
7/6 22:00 31/5 12:30
30/5 11:20
End of measurements (GMT)
14/6 16:00
10/6 22:00 7/6 04:00
8/6 05:40
Equipment operating period
477 / 1909
72 / 288
160 / 638
210 / 841
(hours / no. of 15 minute estimates)
Available flux estimates
(no. 15 min estimates / % of operat1160 / 61%
27 / 9%
304 / 48%
380 / 45%
ing period)
Flux estimates passing micromet
criteria * (no. 15 min estimates / %
725 /38%
32/11%
187/29%
250/30%
of operating period)
-1
* Micrometeorological criteria as defined by Nemitz et al.(2009b): u* > 0.2 m s and |L| > 5 m.
4.3.2 Micrometeorological estimates and deadband
Comparison of w’ data of the different ultra-sonic anemometers showed almost identical patterns. The CEH/UMIST and RISOE systems calculated the β factor on-line using
the measured temperature and sensible heat flux data. The same calculation of β was
made for the FAL system, but in a post-processing of the data, rather than on-line. The
ECN system used a deadband of 0.05 m s-1 and a correction was applied to the data
according to Eq. (3).
4.3.3 Reference sampling mode.
The data for the CEH/UMIST,FAL and RISOE systems were corrected using Eq. (4)
with a linear interpolation of the reference sampling mode data before and after each 7
hour sampling period. The ECN results showed significant scatter, when applying this
correction using each interval between reference sampling data. In subsequent postprocessing analysis, the ensemble of reference sampling measurements from the ECN
system showed that the concentrations measured in the downward channel had a good
linear relation to the concentrations in the upward channel (r2 = 0.9), but with a difference of 15%. The ECN fluxes were therefore calculated using this fixed correction factor. This means that, while the relative difference was constant in the ECN system, it
was variable in the other three systems. The difference in the ECN system was most
60
likely due to a slight difference in collection efficiency of the membrane samplers. This
implies that the lower of the two concentrations in reference mode was underestimating
the actual concentration.
4.3.4 Comparison of ammonia concentration estimates
The NH3 concentrations estimated by the four REA systems and the reference estimates from the profile measurements (χAGM) ranged from 1 to 5 µg NH3 m-3, in the period before mowing and fertilizing (Fig. 2). Larger NH3 emissions following N fertilization
resulted in peaks of up to 22 µg NH3 m-3, with largest values on the day of fertilization.
The linear regressions of the REA concentration data versus the reference concentrations χAGM (Fig. 4.3.) showed a good correlation with r2 values ranging from 0.66 to
0.79. The regression results of the comparison are summarized in Table 4.3, for all
data.
For the small concentration ranges both the CEH/UMIST and ECN sensors gave similar levels compared with the gradient measurements, while the RISOE system tended
to give lower concentrations. For the higher concentration range above 5 µg NH3 m-3,
after the fertilisation event, all systems gave lower concentration levels compared with
the gradient estimates. Overall the ECN system showed best agreement with the AGM
concentrations with only 8% difference.
The χAGM in Figs. 4.2 and 4.3 is the best estimate, while the alternative estimate has
lower concentration levels on the four days of 3, 8-10 June. On June 3 the ECN and
CEH/UMIST data both show a good agreement with the "best estimate" concentration,
while on June 8,9 and 10 both the CEH/UMIST system and the FAL system show concentration levels that are in better agreement with the alternative estimate.
61
25
AGM
20
AGM
REA-FAL
20
14-06
12-06
10-06
8-06
11-06
6-06
0
4-06
0
2-06
5
31-05
5
29-05
10
25-05
10
9-06
15
27-05
REA UMIST/CEH
15
7-06
25
15
10
10
5
5
0
0
7-06
5-06
3-06
1-06
30-05
Figure 4.2
REA-ECN
9-06
15
AGM
30-05
20
NH3 µg.m-3
20
7-06
REA-RISOE
5-06
AGM
3-06
25
1-06
NH3 µg.m-3
25
Time series of the NH3 concentrations measured by the REA systems
and the gradient reference concentration extrapolated to each REA inlet
height as explained in the material and method section. All y-axis in
µg NH3 m-3. Fertilisation occurs on 5 June, 06:00 – 07:00 a.m.
4.3.5 Comparison of the flux estimates
The resulting fluxes as calculated for the four REA systems before and after fertilizing
were compared with the "best estimate" AGM reference data (Figs. 4.4 & 4.5 ). In Table 4.4 the results of linear regression for the uncorrected data versus the best estimate AGM data are shown in the first row. Over the whole period, the REA fluxes uncorrected for concentration are at least 50% lower than the AMG ‘best estimate’ flux.
The best correlation is observed for the CEH/UMIST data set with r2=0.72.
Table 4.3
Results of the linear regression for the concentration measurements observed by the four REA systems versus the AGM results
-3
Comparison
CEH/UMIST
ECN
FAL
RISOE
-3
χAGM (µg NH3 m )
Linear Fit
Slope
Intercept
R
0.69
0.93
0.48
0.95
0.74
0.79
0.77
0.77
1.3
0.45
1.2
-1.6
#
2
χAGM (µg NH3 m )
Fit forced through zero
2
Slope
R
No.
of
points
n
0.87
0.97
0.57
0.74
1626
415
134
199
0.66
0.79
0.74
0.72
Using Alternate AGM estimate
CEH/UMIST
0.70
1.5
0.68
0.94
0.55
FAL
0.67
1.1
0.76
0.80
0.72
#
In the episode June 3,8,9,10 the gradient data instruments are not unanimous so an “alternative gradient estimate” was proposed for these days.
62
25
25
y = 0.48x + 1.19
R2 = 0.77
y = 0.69x + 1.32
R2 = 0.74
20
20
y = 0.57x
R2 = 0.74
y = 0.87x
R2 = 0.66
15
15
10
10
`
5
5
CEH
FAL
1:1
1:1
0
0
0
5
10
15
20
25
25
0
5
10
15
20
25
25
y = 0.93x + 0.45
R2 = 0.79
y = 0.95x - 1.57
R2 = 0.77
20
20
y = 0.74x
R2 = 0.72
y = 0.97x
R2 = 0.79
15
15
10
10
5
5
RISOE
ECN
1:1
1:1
0
0
0
Figure 4.3
5
10
15
20
25
0
5
10
15
20
25
Linear regression of the NH3 concentrations measured by the REA systems (y-axis) versus the AGM gradient reference concentration (x-axis)
extrapolated to the each REA inlet height. All axes in µg NH3 m-3. Fitting
functions are shown both forced though zero (in the box) and with zero
offset.
The REA flux corrected for the concentration difference (χREA / χAGM ) are also compared with the "best estimate" and the “alternative estimate” AGM fluxes. (Table 4) The
CEH/UMIST and FAL REA fluxes compare better with the “alternate estimate” AGM
flux. The r2 value for the CEH/UMIST data is smaller than without the correction, suggesting that the correction with χREA/χAGM introduces extra noise. The RISOE and ECN
data only show a small change since the normalisation correction is small and these
systems have no data on June 8-10 where the two AGM estimates are different.
63
Table 4.4
Summary of the comparison of the REA fluxes with the “best estimate”
and the alternative estimate fluxes obtained with the Aerodynamic Gradient Method (AGM). Intercepts are in ng NH3 m-2 s-1.
Comparison
with AGM flux
CEH/UMIST
FAL
RISOE
2
ECN
Slope / Interc
r
Slope / Interc
r
Slope / Interc
r
Slope / Interc
R2
0.52 / 43
0.73
0.29 / -5
0.30
0.50 / 1.3
0.52
0.50 / 165
0.48
0.41 / 169
0.30
0.46 / -30
0.45
0.49 /145
0.59
0.65 / 56
0.65 0.69 / 78
AGM
Selected data. u(1m) > 1 m s-1, 180° < Wd <350°
0.40
0.45 / -23
0.49
0.50 / 145
0.60
Best
AGM
All data
Best est. AGM
2
2
All data. Fluxes corrected for χREA / χAGM *
Best
AGM
estimate
0.62 / 50
0.65
Alternate est.#
estimate
0.68 / 33
0.74
0.31 / 188
0.42
0.33 / -43
0.52
0.64 / 140
0.72
Alternate AGM”
0.75 / 22
0.80
0.5 / 86
0.47
0.33 / -43
0.52
0.65 / 141
0.72
Before june 5
1.03 / -7
0.55
no data
0.56 / -98
0.13
1.22 / 90
0.28
0.71 / 45
0.79 0.5 / 86
0.47
0.21 / 126**
0.85 0.71 / 50
0.81
After june 5
* For each 15-minute period the REA flux was multiplied by χREA/χAGM to distinguish errors due to REA flux measurement
from errors in concentration measurements.
** only 9 data points available
#
In the episode June 3, 8,9,10 the gradient data instruments are not unanimous so an “alternative gradient estimate”
was proposed for these days.
The last part of Table 4.4 shows the effect of selection of the proper meteorological
conditions, i.e. optimal fetch conditions and u >1 m s-1. In can be argued that data that
does not meet the micrometeorological criteria should not be shown at all. However it is
clear that this selection will preferentially remove fluxes at for example low windspeed
i.e. at night. Table 4 therefore starts with the full data set and shows that while the selection seems important for one instrument it might be less sensitive for another (for
example depending on the position on the field). The improvement for the CEH/UMIST
and ECN data is most significant. The results for the FAL system do not improve, due
to the small data set left after selection (33 data points). Finally, the comparison can be
done either for the period before or after fertilisation on June 5.
The fit for the CEH/UMIST REA data versus the alternate AGM data shows a slope of
1.03 with r2 = 0.55 for the first period. The overall underestimation of the REA fluxes as
compared to the AGM flux is mainly explained by the underestimation during the postfertilisation period, although during this period, the correlation is good as indicated by
the large r2. The FAL system only has post-fertilisation data available, and the RISOE
system has only 9 data points during this period, with selected meteorological conditions. The ECN and the CEH/UMIST system give comparable fluxes after fertilisation,
which were about 30% below the AGM estimate flux, with an r2 of 0.8. In the period before fertilisation the ECN system gives flux estimates that are 20% above the AGM estimates.
64
900
800
700
600
500
400
300
200
100
0
-100
5-06
4-06
3-06
2-06
5-06
4-06
3-06
2-06
1-06
31-05
29-05
5-06
4-06
REA-ECN
Selected
AGM
Ammonia fluxes by the Relaxed Eddy Accumulation (REA) and the
Aerodynamic Gradient Method (AGM) during the first part of the experiment. Cutting of the field took place 0800-1000 on 29 May. The REA
fluxes are corrected by χAGM / χREA for each 15-minute period. The
RISOE and ECN systems started after cutting of the field.
15-06
0
14-06
0
13-06
500
12-06
500
11-06
1000
10-06
1000
9-06
1500
5-06
1500
8-06
2000
7-06
2000
6-06
REA UMIST/CEH
Selected
AGM
2500
8-06
2500
3000
7-06
REA-RISOE
Selected
AGM
6-06
3000
5-06
1-06
31-05
30-05
29-05
28-05
3-06
2-06
1-06
31-05
ng NH3 m-2.s-1
Figure 4.4
3000
500
500
0
0
7-06
1000
12-06
1000
11-06
1500
10-06
1500
9-06
2000
8-06
2000
Figure 4.5
REA-ECN
Selected
AGM
2500
6-06
ng NH3 m-2.s-1
2500
9-06
REA-FAL
Selected
AGM
8-06
3000
5-06
ng NH3 m-2.s-1
900
800
700
600
500
400
300
200
100
0
-100
REA-RISOE
Selected
AGM
30-05
900
800
700
600
500
400
300
200
100
0
-100
27-05
26-05
25-05
ng NH3 m-2.s-1
REA UMIST/CEH
Selected
AGM
Ammonia fluxes by the Relaxed Eddy Accumulation (REA) and the
Aerodynamic Gradient Method (AGM) following N fertilization of the field
0800-1000 on 5 June. The REA fluxes are corrected χAGM / χREA for each
15-minute period.
65
4.4
Discussion
4.4.1 Estimation of the REA scaling factor β
The β factor estimated from T and H was, as expected, ~0.6 (Pattey et al., 1993; Gao,
1995). A systematic investigation by Ammann et al. (2002) showed that the βfactor increases weakly for stable conditions. However, a significant number of outliers and
even negative values of β were derived when H was small and its direction poorly determined. Therefore most systems (CEH/UMIST, FAL & RISOE) used only acceptable
β factors, for example in the range 0.1-1, together with a fixed β factor for periods for
which the experimental value was unreasonable. For the CEH/UMIST dataset a comparison of 1162 flux values calculated both with a fixed β factor and with measured
(and acceptable) β factor gives almost the same result. Linear regression between
these two data sets showed an r2 of 0.98 and a slope of 1.03 (the variable β gives similar, though slightly larger fluxes on average than the fixed β). The ECN system used a
theoretical β factor (Eq.3) throughout. Milne et al. (2001) have shown that the empirical
value of β can theoretically be derived from the combined frequency distribution of w’
and χ’, if this follows a Gram-Charlier equation. This frequency distribution will depend
on the spatial distribution of sources and sinks of the different tracers and may be different for highly reactive species such as NH3 or O3 and non-reactive species like N2O
and CO2. Andreas et al. (1998) found β to be 0.63 for momentum fluxes at near neutral
conditions, but β for scalars was only 0.52.
Another possibility is that the β-factor used should actually be larger for reactive gases.
In order to assess this, a set of O3 covariance flux measurements obtained during the
Braunschweig experiment (Mészáros et al.1) was processed to estimate β. The average β estimated over a day with 40 flux measurements was β = 0.62 ± 0.09 for H, and
β = 0.58 ± 0.06 for O3 fluxes. The difference is thus not significant suggesting that reactive gases have the same β as heat flux.
These small discrepancies need further assessment from a theoretical point of view, as
the choice of the β factor has a linear effect on the flux levels that are reported. However, the scale of these discrepancies is much smaller than those observed between
the REA and AGM ammonia fluxes found in this study. This suggests that other reasons must account for the differences observed here.
4.4.2 Comparison of measured ammonia concentrations and fluxes by
the REA and AGM systems
The concentration levels observed in the REA systems are generally less than the
AGM values. For the FAL, RISOE and CEH/UMIST systems that use wet denuders,
this can be caused by periods during which the denuders were not fully wet, therefore
under-sampling. The ECN system with the membrane detector avoids this problem and
indeed shows the best agreement in terms of average concentration levels. Incomplete
wetting of the denuder walls was observed to be a problem with the CEH/UMIST system during episodes with high air temperatures (up to > 36 °C) experienced during the
campaign. High temperatures also coincided with the periods of the largest NH3 emissions and thus largest air concentration. While the concentration ratios between the
two denuders of the CEH/UMIST system were typically in the range 0.95 to 1.05 in the
referencing periods (bias of ±5%), during some warm episodes, biases of more than
30% were observed, indicating that the capture efficiency of at least one denuder was
reduced.
66
During the first half of the experiment, after correction for differences in concentration,
the REA fluxes obtained with the CEH/UMIST system agree remarkably well with the
AGM best estimate flux, within the range of variability found between individual gradient systems (Milford et al., 2009). The ECN system shows a higher noise level and
shows flux levels that are 20 % above the AGM estimates. This could be related to
evaporation of dew on inlets in the early morning, as suggested by the higher emission
peak of the ECN system in the morning of June 1, for example. The CEH/UMIST system with a much larger air flow is less exposed to such problems. The RISOE fluxes
seem to show an offset compared to the AGM fluxes, with deposition in the evening, or
even during the day on June 4. This could be due to a delay in the RISOE system
caused by the automatic sampling in vials prior to injection into the analytical system.
After fertilising during the episode with high concentrations and large fluxes, all REA
systems underestimate the flux by 30-50 %. Some days show an excellent agreement,
for example on June 6. The emission level on June 5, on the fertilisation day (fertilised
between 6:00 and 7:00 AM), is significantly underestimated. This can be due to the meteorological conditions which lie out of the selection for flux calculation (low windspeed
from the east). For this period the correction for the concentration difference by
REA/AGM is not sufficient to explain the difference between AGM and REA fluxes. This
could be caused by an ineffective sampling of the REA systems, although the reference sampling mode should account for this effect. Crossover contamination between
subsequent analyses of the up and down draft sample with a single detector could explain a systematic underestimation of the flux, since it will tend to bring the concentration levels closer together. Moreover, such a bias would not be detected by either the
absolute concentration calibration or by the referencing mode. However, laboratory
tests were performed on all systems with the detector alternating between two liquid
solutions, and no such a problem was found.
Another problem may arise from the delay for the air flow to establish in the detectors
after the air valve has been switched on. This delay, which can be as important as
0.5 s for system with large pressure drop (Pattey et al., 1993), causes a de-correlation
between w’ and χ’, which reduces the REA flux estimate. It would also reduce the REA
average concentration by reducing the actual flow rate through the denuder, which is in
agreement with the systematic underestimation of the REA concentrations.
Finally, the AGM flux estimates might actually show levels that are too high. Sutton et
al. (2009b) show that while there is evidence for this in selected days (3, 8, 9, 10 June)
using the "best estimates", the "alternative estimates" and results for other days give no
evidence of over-estimation.
4.4.3 Design features of the different REA systems
The results show that the REA principle for NH3 flux measurement permits overall good
performance, but the quality of the data still varies between implementations and in
time. Improvement of the instrumental setup is still needed to obtain a robust continuous NH3 monitoring system.
4.4.3.1 Sampling inlet and fast response switching
There is always a time delay between the eddy occurring in the sonic anemometer and
the opening of the valve that selects either up or downward sampling. This takes about
100 milliseconds depending on the valves used. The common inlet used in the ECN
system enables a correction for this delay. This principle was demonstrated earlier for
CO2 (Hensen et al., 1996). On the other hand in wet conditions, adsorption of NH4+ to
the walls of the tube can occur, introducing adsorption / desorption effects which can
67
smooth-out concentration differences. The trade-off between these two effects needs
further evaluation in future campaigns.
Compared with the small inlet systems used in the FAL and ECN systems, the RISOE
inlet system is large. In theory this might cause a flow disturbance. Therefore a test
was performed comparing the spectral output of the ultra-sonic anemometer used in
the REA configuration and a 'stand- alone sonic anemometer'. This exercise showed
that there is no effect of the inlet systems for the relevant wind directions.
At increasing wind speeds and at decreasing measurement heights, a large fraction of
the variance of w is carried by high frequency eddies. This can be a problem when
REA systems are operated too close to the ground at higher roughness since fast fluctuations in w will cause the valves to open and close at very high frequencies (1-10 Hz)
in order to measure the large variance.
A general recommendation for the inlet system is to make this as small as possible,
because a smaller inlet can be positioned closer to the sonic without flow distortion.
With the RISOE system, however, it was shown that even with a rather large inlet configuration, no disturbance on the sonic could be detected.
4.4.3.2 Sampling height
Requirements for high sampling rates and fast response inlets increase as the height
above ground decreases. The spectral analysis of the data can be used to evaluate
which part of the spectrum is contributing to vertical transport. This frequency depends
on wind speed, roughness, atmospheric stability and measurement height (e.g. Eugster
et al. 1995). The technical specifications of the system (for example the maximum
switching frequency) that can be obtained with the valves therefore indirectly determine
the minimum height at which the REA system will operate properly. Moreover, an upper
limit to this height is set either by the fetch or by the minimum resolution of the concentration measurement. At large heights transport shifts to lower frequencies. So a longer
averaging time is required to obtain representative sample for these frequencies.
Longer average times will also reduce concentration differences between updrafts and
downdrafts. The systems used here can be applied at heights above 1 m, and below
30 m. This will in general mean evaluation of fluxes in an area with a spatial scale of
homogeneous fetch between 100 and 3000 m is possible.
4.4.3.3 Chemical detection systems
All the methods were sufficiently accurate to obtain the concentration differences between up and down samples. The reference sampling mode and the calibration tests
showed that the systems worked according to specifications listed in Table 1. On three
occasions an aqueous ammonia standard sample was provided blind to all systems.
This showed agreement of the wet chemical analyses within 10%.
4.4.3.4 Use of the reference sampling mode
The use of a reference sampling mode on a regular basis (e.g. for 1 hour every 6-8
hours), during which the sampling is uncorrelated with the wind-speed, should provide
equal concentrations in the up and down-drafts sampling lines. The reference sampling
method corrects for bias between the sampler inlets for up and down-drafts. It can,
however, not correct for absolute errors in NH3 concentrations. These measurements
provide also useful information on the performance of the system. Two methods were
used: delayed wind-speed data set or synchronous switching. With the first method
sampling conditions are identical to measurement conditions. However, crosscorrelation spectra between w and c sometimes show secondary peaks. Potentially a
68
shift could be selected that leads to a non-zero correlation and therefore still a difference in concentrations. Large changes in concentrations during the episode of reference sampling can pose a problem, especially for the systems that have a relatively
large response time (RISOE and ECN system). Correction of the measurement data by
scaling up and down concentrations, so that reference sampling concentrations in both
channels are identical, works well most of the time.
4.4.4 Potential for measurement of NH3 flux divergence using REA
As noted in the introduction, one of the interests in developing REA systems to measure vertical fluxes of NH3 is that it allows flux measurement at one height. As a consequence, it becomes possible in principle to deploy several REA systems at different
heights to measure divergence in the vertical flux, either due to local advection sources
or chemical production / consumption in the surface layer. Nemitz et al. (2009a) demonstrate that NH3 emissions following fertilization lead to formation of particulate NH4+,
and correspondingly a consumption of NH3 in the air within and above the canopy.
While the effect on particulate NH4+ dynamics is very large, NH3 gradients are the driving potential so that the relative effect on NH3 fluxes is rather small. Nemitz et
al.(2009a) estimate that during daytime conditions in the days after fertilization (5-10
June) about 2% of the ground NH3 flux (Fz(zo)) would be lost to NH4+ aerosol by a
height of 2 m. Due to the large NH3 concentrations close to the surface most of this
conversion is estimated to happen close to the ground (< 0.5 m). Between 0.5 m and 2
m the effect is minor (< 1%), while above 2 m volatile NH4+ aerosol species (NH4NO3,
NH4Cl) are more likely to re-evaporate. The magnitude of these effects is below the
resolution in the REA systems when they perform best: the differences between the
REA systems are larger, showing that further improvements are needed before REA
can be applied as a robust tool for assessment of vertical flux divergence due to
chemical production / consumption at the rates observed in this experiment.
There is a greater potential to apply REA to determine vertical flux divergence due to
advection (∆Fz,adv). Loubet et al. (2009) analysed advection errors during the
GRAMINAE Integrated Experiment, accounting for both advection from a local farm
and advection caused by emissions from the field itself. For the REA sampling position
(Site 1) in the middle of a large field, ∆Fz,adv at 1 m was in general from +50 ng NH3 m-2
s-1 (advection dominated by the farm, prior to cutting of the field), to –50 ng NH3 m-2 s-1
(advection dominated by the field following cutting and fertilization). The errors at the
sampling heights of the REA systems (~2 m) would be approximately double these values. Following cutting, the errors mostly represent < 5% of the emission fluxes and are
not resolvable by the current REA implementations. However, the errors prior to cutting
are of the same magnitude as the fluxes, indicating the potential for measurement of
∆Fz,adv.
4.5
Conclusions & recommendations
Overall, the conclusion of this inter-comparison is that REA measurements for NH3
fluxes are possible, but they require substantial care in running the analytical equipment. For monitoring NH3 fluxes over the long-term (e.g. continuous operation over
many months), these prototype systems are not yet sufficiently robust. Further attention
is therefore needed to develop the methods for robust-automated operation. The wet
chemical techniques will never be fast enough to enable full eddy covariance measurements of NH3. Optical techniques for NH3 detection are improving and flux measurements are becoming possible (Whitehead et al., 2008, Fischer and Littlejohn, 2007,
Emmeneger, 2004). The accuracy of these systems is now sufficient to detect larger
emission fluxes and the detection of deposition fluxes could become possible over the
69
next years. Where the wet chemical systems make use of the adsorptive behaviour of
the NH3 molecule and it’s affinity with water this is an important problem for any closed
path optical system. Open path optical systems that can evaluate NH3 with enough
sensitivity do not have inlet problems but need sufficient path length and can therefore
only detect larger eddies. So the wet chemical REA approach therefore remains a useful option as a research tool for intensive flux measurement over periods of days and
weeks. In principle, it is attractive as a means to determine divergence in the vertical
NH3 flux due to advection or chemical production/consumption. In practice, however,
the scale of flux divergence due to chemistry (e.g. 10%) is smaller than that which can
currently be resolved, within the practical height restrictions of REA measurements.
Advection errors due to emissions from the field itself are similarly too small to determine. By contrast, advection errors from nearby farm sources may be detectable
through the application of REA measurements at several heights.
Acknowledgements
The work presented here was partly funded by the EU GRAMINAE Project (ENV4CT98-0722) and the NitroEurope IP. For the experimental work for ECN we acknowledge the colleagues Julio Mosquera and Piet Jongejan. Additional support for this work
was received from EU MEAD Project ENV4 (RISOE), the UK Department for Environment, Food and Rural Affairs (DEFRA), the UK Natural Environment Research Council
(NERC) and the ministry of EZ and VROM (NL). NinE and COST 729 are acknowledged for travel funds. We are grateful for the experimental field site and facilities provided by Prof U. Daemmgen of the FAL, Braunschweig.
70
71
Landfill emission measurements
72
Chapter 5 Methane emission estimates from landfills obtained with
dynamic plume measurements1
A. Hensen1 and H. Scharff2
1
ECN, Petten, the Netherlands
2
Afvalzorg Deponie, Haarlem, The Netherlands
Abstract.
Methane emissions from 3 different landfills in the Netherlands were estimated using a
mobile Tuneable Diode Laser system (TDL). The methane concentration in the cross
section of the plume is measured downwind of the source on a transect perpendicular
to the wind direction. A Gaussian plume model was used to simulate the concentration
levels at the transect. The emission from the source is calculated from the measured
and modelled concentration levels. Calibration of the plume dispersion model is done
using a tracer (N20) that is released from the landfill and measured simultaneously with
the TDL system. The emission estimates for the different locations ranged from 3.6 to
16 m3 ha-I hr-1 for the different sites. The emission levels were compared to emission
estimates based on the landfill gas production modeIs. This comparison suggests oxidation rates that are up to 50% in spring and negligible in November. At one of the
three sites measurements were performed in campaigns in 3 consecutive years. Comparison of the emission levels in the first and second year showed a reduction of the
methane emission of about 50% due to implementation of a gas extraction system.
From the second to the third year emissions increased by a factor of 4 due to new landfilling. Furthermore measurements were performed in winter when oxidation efficiency
was reduced. This paper describes the measurement technique used, and discusses
the results of the experimental sessions that were performed.
Keywords: landfills, methane emissions plume dispersion tuneable diode laser
1
Reference: Hensen, A., Scharff, H., Methane emission estimates from landfills obtained with
dynamic plume measurements, Water,Air and Soil pollution, Kluwer, focus1: 455-464, 2001.
73
5.1
Introduction
In the Kyoto protocol, the Netherlands committed themselves to a substantial reduction
of greenhouse gas emissions. Within Europe the Netherlands have the highest emission density for methane of about 24 ton2 CH4 per km2 per year. The Dutch 1997 emissions of methane from landfills of about 445 ktonne (Coenen et al., 2000) counts for
about 7% of the total national emission level in CO2 equivalents. As a result of the current waste policy, methane emissions are expected to have decreased by a factor of 3
by 2010, thus contributing significantly to the total Dutch effort to reduce greenhouse
gas emissions. Since landfills play such an important role in the Dutch effort to meet
the Kyoto target, this source should be weIl monitored.
The mass balance of landfill gas is shown in Figure 5.1. The production inside the landfill is estimated from the total volume, composition and age of the stored waste. A part
of the produced gas can be recovered using extraction systems, the remaining methane escapes to the atmosphere or is oxidised on its way out of the landfill body. In the
emission inventories the oxidation level is set to 10% as an IPCC default, but in fact
oxidation is both variable in space and time (Czepiel et al., 1996).
Extraction
CO2 Emission
CH4 Emission
CH4 Emission
oxidation
CH4 Formation
migration
??
Formation = Emission + Oxidation + Extraction + Accumulation
Figure 5.1
Methane emission as a function of production, recovery and oxidation.
Emission reduction at landfill sites can be achieved using a gas recovery system or by
improving the oxidizing capacity of the landfill cover. Apart from the CH4 contribution to
global warming, other landfill gas components can cause local environmental problems. Odorous emissions in particular can be an incentive to implement emission reduction measures.
5.2
Materials and Methods
5.2.1 Description of the measurement sited and methodology
Measurements were performed at Nauema (52°27'N, 4°45'E), a 72 ha landfill situated
15 km west of Amsterdam, at Hollandse Brug (52°19'N, 5°08'E), a 27 ha landfill situated 20 km south east of Amsterdam and at Braambergen (52°20'N,5°16'E) near the
2
There was an error in the original paper with the unit kg instead of ton.
74
city of Almere. Characteristics of the locations are listed in Table 5.1. All these landfills
are located in flat terrain at a height of 15-20 m.
Downwind of the landfills, the CH4 concentration was measured on a transect through
the plume that originates from the source. This was done using a Tuneable Diode Laser absorption spectrometer (TDLAS, Aerodyne Res. Inc. Billerica, Mass. U.S.A.)
mounted in a van. Absorption of the light emitted from the diode laser occurs in a 30 m
astigmatic multipass cell. The 1270 and 1271 cm-1 absorption lines are used for methane and nitrous oxide respectively. Ambient air is sampled from the roof of the van and
lead through the cell. Each plume transect takes approximately 5-7 min. Figure 5.2
shows the measurement method. The TDL showed a 10 ppb resolution for CH4 and 20
ppb for N2O. In these campaigns 1 Hz concentration data was used. Calibration took
place while driving, applying standards before and after each transect measurement.
The position of the van was obtained using a GPS system (Survey II, Garmin Inc.).
Table 5.1
Location
Overview of the measurement locations
Waste Characteristics
Area
C-Content
Estimated
(ha)
(kg/tonne)
formation in 2010
3
(Mm )
Hollandse Brug
Demolition waste
Nauerna
Polluted soil, industrial waste
Sludge
Braambergen
Polluted soil, Domestic,industrial and
demolition waste
Satellite
positioning
15
70
0.87
60
50
5.25
30
80
3.5
10*10 punt sources
Each point one gauss plume
Meteorology determines plume dispersion
Figure 5.2
Measurement method, using the methane plume downwind of the landfill
75
5.2.2 Emission modelling
The measured concentrations in the plume transects were compared with the output of
a multiple gauss plume model. This model uses a grid of 10 x 10 points over the landfill
(Figure 5.2). For each grid point of the map a Gaussian plume is used, taking reflection
of the plume at the ground level into account. No correction was applied for reflection at
an inversion layer, the distance between source and receptor is not more than 2000 m
and at that distance the effect of an inversion layer is not expected in daytime. For
each source-receptor combination the receptor concentration is obtained:
C ( X , Y , z) =
Q
−Y 2
⋅e
2π ⋅ u ⋅ .σ y ⋅ σ z
with
( 2⋅σ y )2
⋅  e −( z−hs )

2
( 2⋅σ z ) 2
+ e −( z +hs )
2
( 2⋅σ z )2


(2)
σ y = A ⋅ X B ⋅ z o0.2 ⋅ T 0.35
σ z = C ⋅ X D ⋅ (10 ⋅ z o ) 0.53⋅E
E = X −0.22
Where X is the distance along the plume axis, Y the axis perpendicular to the plume
axis, z the height above ground level, Q the source strength, u the wind speed measured on top of the landfill, and hs the height of the emission (top of the landfill). σy and
σz are dispersion parameters that depend on distance to the source, on the degree of
turbulence of the atmosphere, the roughness length of the surface zo, and on the timescale T used for averaging. A, B, C and D are dependent on the stability class (Pasquill, 1974).
The differences in source strength of the different parts of the landfill are accounted for
by using a weighting index on different parts of the 10 x 10 matrix. Using this index
map the total emission Q is distributed over landfill. The emission strength of the landfill
is equal to the source strength needed in the model to obtain an agreement between
the integral of the concentration along the transect for the modelled and the measured
plume. In general, there are two extremes to consider when choosing the matrix. A
minimum estimate for the emission strength is obtained when all sources are close to
the measurement transect. With sources far away, a higher total source strength is
needed to obtain the measured concentration levels at the plume transect. The shape
of the modelled curve will show more spatial structure with sources close to the transect. The best estimate will be in between these two levels with source locations chosen to obtain a similar structure for the model and measured plume. Tests with individual plumes for the Nauema site showed that the minimum and maximum levels show a
range of ±15% compared to the average level.
The meteorological data (heatflux, wind speed, cloud cover) indicate the Pasquill stability class, but a check on this choice is recommended. Therefore N2O was released
from a gas flask on top of the landfill and the TDL measured the N2O and CH4 plume
simultaneously. For N2O source, bath position and source strength are known, and the
horizontal dispersion σy is obtained. The model calculation for this plume was used to
check and if necessary to adapt, the dispersion parameters.
The uncertainty in the CH4 concentration measurements is about 1-5% due to instrument noise, drift of the laser and uncertainty in the background concentration level.
Changes in wind direction (on a timescale of 5-10 min) were the main cause for variation in the set of emission estimates. Furthermore, the emission of the landfill site will
76
not be constant with time and in order to obtain an annual average emission level,
more measurements must be taken over the year.
5.2.3 The landfill gas production model
The landfill gas production model is based on a multi-phase first order degradation
model used extensively in this field. In this case the model as described by the Dutch
Landfill Gas Advisory Centre (Scheepers, 1994) was used. This model and its different
parameters were optimised and validated on the basis of results from nine Dutch landfill gas projects (Coops et al., 1995).
Table 5.2
Information on the measurement campaigns
Location
Date
Wind
speed
-1
(m s )
Wind
direction
Stability
Class
Cloud
cover
Hollandse Brug
09/09/97
3
NE
B
Nauerna
15/04/97
04/05/98
19/11/99
10/11/99
7
5
4-6
5-4
N
N
N
NE
D/C
B
D/E
D
Braambergen
Table 5.3
No
of
transects
7/8
Distance
to transect
(m)
750
2/8
2/8
6/8
5/8
780
780
780
1000
8
14
10
13
7
Emission levels during the measurement campaigns
Location
Year
No
of
transects
Emission
-1
gCH4.s
Landfill
area
(ha)
Remarks
Hollandse Brug
1997
8
11±4
15
No gas extraction
Nauerna
1997
1998
1999
1999
7
11
10
13
62±12
31±13
109±33
48±8
60
60
60
30
No gas extraction
Gas extraction at 20 ha
Gas extraction at 20 ha
Gas extraction at 14 ha
Braambergen
5.3
Results
Five campaigns were carried out in total at the three sites. Table 2 summarises the
measurement conditions. Table 3 shows the emission levels that were derived for the
different sites.
5.3.1 Experiments at Nauerna
Both the experiments in 1997 and in 1998 took place in spring. In 1997 the average
windspeed was 7 m s-1. Stability class D and zo = 0.1 were used. During the experiment
in 1998 the wind speed was lower (about 3 m.s-1) and the wind direction was more
variable. Stability class B and a roughness length zo of 0.10 m were used in the model.
Figure 5.3 shows an example of a plume transect as obtained at Nauerna in 1998. The
line shows the measured excess concentration. The markers indicate the model calculation. For this plume the emission was estimated to be 70 gCH4 s-1.
77
East
1600
1400
Model 70 gCH4 s -1
Measured plume
1 15 15 15 15 15 15 15 15 2
1 3 3 3 3 3 3 3 3 2
1 2 2 3 3 3 3 3 2 2
1200
0 0 1 2 2 3 3 3 2 2
0 0 0 0 1 2 1 1 2 1
0 0 0 0 0 1 2 1 2 1
0 0 0 0 0 1 1 1 2 1
1000
0 0 0 0 0 0 0 1 1 0
0 0 0 0 0 0 0 0 1 0
800
West
600
400
200
0
-200
-1000
Figure 5.3
N2O concentration above background (ppb)
0 0 2 2 2 2 2 2 2 1
North
CH4 concentration above background (ppb)
1800
500
0
500
1000
1500
2000
Distance along the traject (m)
Example of a transect measurement, the line shows the measured concentration, the markers the modelled concentration along the measurement transect.
120
100
Model plume at
Q=3.3 grN2O s-1
Measured plume
80
60
40
20
0
-20
-100
100
300
500
700
900
1000
Distance along the traject (m)
Figure 5.4
78
The N2O plume measured at the transect together with the modelled
plume.
The matrix in the graph shows the source distribution used. In April 1997 8 plume
measurements were obtained. One of the plumes was rejected because it showed an
emission level that was a factor of two above the average of the other plumes. The average emission level obtained using the 7 plumes was 62 ± 12 gCH4 s-1. One year later,
after the implementation of a gas extraction system on about 20 ha of the landfill, 11
plume transects were obtained. For this experiment the estimated emissions ranged
from 19 g CH4 s-1 to 40 g CH4 s-1 with one outlier of 65 g CH4 s-1 The average emission
level was 31 ± 13 g CH4 s-1.
In 1999 the average emission level was significantly higher 109 ± 30 g CH4 s-1. Most of
this emission originated from an area on the landfill where relatively fresh material was
landfilled. Measurements look place in November at a temperature of about 2 °C,
which will have reduced the oxidation efficiency of the landfill cover. The N2O plume indicated that stability class E and roughness length zo = 0.5 m give the best results for
the model calculations. The N2O release in 1999 was 8-10 higher than in 1998 which
resulted in a peak concentration of 800 ppb and an improved S/N ratio compared to
1997 and 1998 (Figure 5.4). The standard deviation in the set of ten emission estimates was relatively large. Such large variation might have arisen due to the construction of a harbour, which meant that part of the measurement transect had to be carried
out on a road parallel to the plume axis.
5.3.2 Experiments at Hollandse Brug
An old part of the landfill of about 7 ha is used as a recreation area. Because of the
shape of the landfill, the south west part of the site was only 200 m away from the
measurement transect, whereas the north east part was at a distance of 1500 m upwind. For a set of 7 transects a net emission level of 11 ± 4 gCH4 s-1 was obtained. Meteorological data obtained on the site could not be used to derive stability due to disturbances from wind along the landfill slope. Stability data were obtained from a meteo
station 25 km upwind of the site. This induces an extra uncertainty, which is reflected in
the standard deviation reported above.
5.3.3 Experiments at Braambergen
At Braambergen, measurements took place on a road at about 1 km southwest of the
landfill. Windspeed decreased during the experiment from about 6 to 3.5 m s-1. Stability
class D in combination with a relatively large zo of 0.3 m showed the best agreement
for the N2O plumes. The plume data were corrected for the contribution of a dairy farm
along this road that also showed a clear methane emission of about 1.8 g CH4 s-1.
5.3.4 Comparison between measured emission levels and production
estimates
The emission levels obtained from mobile measurements were compared with those
based on the landfill gas production modeIs. Gas extraction levels are monitored continuously (TabIe 5.4). In theory the sum of recovered and emitted methane must be below the production level estimate. The difference between these two levels is an estimate of the oxidation.
The gas production model predicted a methane production of 110 ± 10 m3 CH4 hr-1 at
Hollandse Brug in 1997. The measured emission is equivalent to about half that value.
This also happened at Nauerna in 1997 where the gas production model predicted a
methane production of 590 ± 60 m3CH4 hr-1. In 1998 the gas production estimate at
Nauerna decreased to 540 ± 54 m3 CH4 hr-1.
79
This decrease, combined with the gas extraction system that was in operation lead to a
50% emission reduction compared to the 1997 level. In both years the oxidation levels
during the experiments must have been 50-60% in order to account for the relatively
low emission levels. This is high compared to the 10% IPCC default value. It was a
surprise to find that the gas extraction system on 20% of the landfill reduced the emission level by 50%. We assume this is caused by an increase in oxidation efficiency.
Also under-pressure induced by the gas extraction will reduce gas emission through
cracks in the top cover.
The 1999 production level estimate of 500 ± 50 m3 CH4 h-1 at Nauerna was lower compared to the sum of recovered CH4 and the measured emission level. We do not have
a good explanation for this. Probably fresh waste material that was landfilled shortly before the measurement day was not accounted for well enough in the production estimate. A production estimate that lies below the sum of measured emission and extraction was also observed at Braambergen in 1999. One month after the emission measurements the landfill gas extraction system was extended from 14 to 24 ha of the total
landfill (30 ha). It was found that the extraction system recovered an amount of CH4
that was equal to the estimated production level. Since a 100% efficiency of the system
over the whole site is not expected, the production level estimate must have been too
low.
Table 5.4
Emission data and production model compared (m3 hr-1)
Location
Date
Production
Recovery
Emission
Oxidation
Hollandse Brug
1997
110
0
55
50%
Nauerna
April 1997
April1998
Nov.1999
Nov. 1999
Dec. 1999
590
540
500
500
500
0
110
110
200
490(!!)
310
155
450
500
See text
50%
60%
??
Braambergen
5.4
??
Conclusions
With favourable meteorological conditions the TDL plume measurements only showed
a standard deviation of 20% between individual estimates over one day. The TDL
measurement technique is convenient for the landfill operator since it does not interfere
with the landfill operation.
Experiments in 1997 and 1998 at Nauerna and Hollandse Brug showed that emission
levels obtained from the measurements correspond with those obtained from the landfill gas production model when an oxidation level in the top soil of 50% is assumed.
This is a factor of 5 above the 10% level used in the national emission inventories in
the Netherlands. This finding is not unique Bergamaschi et al. (1998) also showed 3040% oxidation. At Nauerna a 50% reduction of the emission occurred between 1997
and 1998. This emission reduction was larger than expected from the model calculations. An increased oxidation efficiency of the landfill cover probably caused this. It
showed however that even assumptions for oxidation efficiencies based on previous
measurements at the same site need not be representative when conditions at the
landfill have changed or when measurements are carried out in a different season. The
experiments in 1999 showed that both for Braambergen and for Nauerna, the produc-
80
tion model leads to under predicted emission levels. The emission level of a landfill is
variable both in space and time. Czepiel et al. (1996) report an increase in emission of
almost a factor of 2, associated with a decrease in the atmospheric pressure over 5
days. For the campaigns reported here air pressures were similar and stable at about
1025 mbar. This confirms that emission levels reported here can be considered representative for the measurement day. More measurements in a year are needed to obtain an annual emission level. Landfills world-wide are considered to be an important
anthropogenic methane source. The estimated emission levels however have a significant uncertainty. Most landfills do not have the detailed documentation on the composition of the deposited waste which is available in the Netherlands. Furthermore crude
assumptions are used with respect to top cover oxidation. The importance of more
measurements at more landfills of different types is therefore obvious. For the Netherlands, in 2010 a limited number of landfills will be responsible for the major part of the
methane emissions: over 95% will be caused by a group of 25 landfills. This implies
that an inventory of Dutch methane emissions from landfills might be based on direct
measurements from this group, on the condition that an accurate and affordable measurement methodology is available.
Acknowledgements
Various colleagues co-operated during the measurement campaigns described here.
We therefore thank W. C. M. v. d. Bulk, H. Möls, P. Fonteijn, B. Liang and R. Meestert
or their assistance A.T. Vermeulen and J. W. Erisman helped with the data evaluation
and reporting on the individual projects.
81
NH3 plume measurements Braunschweig 2000
82
Chapter 6 Estimation of NH3 emissions from a naturally ventilated
livestock farm using local-scale atmospheric dispersion modelling3
A. Hensen1, B. Loubet2, J. Mosquera1,*, W.C.M van den Bulk1, J. W. Erisman1, U.
Dämmgen3, C. Milford4, F.J. Löpmeier3, P. Cellier2, P. Mikuška5 and M. A. Sutton4
1
ECN, Petten, The Netherlands
Institut National de la Recherche Agronomique (INRA), Thiverval-Grignon, France
3
Federal Agricultural Research Centre, Braunschweig (FAL), Germany
4
Centre for Ecology and Hydrology (CEH), Edinburgh, UK
5
Institute of Analytical Chemistry, ASCR,v.v.i., Brno, Czech Republic
*
now at: Animal Sciences Group (ASG), Wageningen
2
Abstract. Agricultural livestock represents the main source of ammonia (NH3) in
Europe. In recent years, reduction policies have been applied to reduce NH3 emissions.
In order to estimate the impacts of these policies, robust estimates of the emissions
from the main sources, i.e. livestock farms are needed. In this paper, the NH3 emissions were estimated from a naturally ventilated livestock farm in Braunschweig, Germany during a joint field experiment of the GRAMINAE European project. An inference
method was used with a Gaussian-3D plume model and with the Huang 3D model. NH3
concentrations downwind of the source were used together with micrometeorological
data to estimate the source strength over time. Mobile NH3 concentration measurements provided information on the spatial distribution of source strength. The estimated
emission strength ranged between 6.4 ± 0.18 kg NH3 d-1 (Huang 3D model) and 9.2 ±
0.7 kg NH3 d-1 (Gaussian-3D model). These estimates were 94% and 63% of what was
obtained using emission factors from the German national inventory (9.6 kg d-1 NH3).
The effect of deposition was evaluated with the FIDES-2D model. This increased the
emission estimate to 11.7 kg NH3 d-1, showing that deposition can explain the observed
difference. The daily pattern of the source was correlated with net radiation and with
the temperature inside the animal houses. The daily pattern resulted from a combination of a temperature effect on the source concentration together with an effect of variations in free and forced convection of the building ventilation rate. Further development
of the plume technique is especially relevant for naturally ventilated farms, since the
variable ventilation rate makes other emission measurements difficult.
3
Reference: Hensen, A., Nemitz, E., Flynn, M. J., Blatter, A., Jones, S. K., Sørensen, L. L.,
Hensen, B., Pryor, S. C., Jensen, B., Otjes, R. P., Cobussen, J., Loubet, B., Erisman, J. W.,
Gallagher, M. W., Neftel, A., and Sutton, M. A., 2009: Inter-comparison of ammonia fluxes obtained using the Relaxed Eddy Accumulation technique, Biogeosciences, 6, 2575-2588,
doi:10.5194/bg-6-2575-2009.
83
6.1
Introduction
Atmospheric ammonia (NH3) is recognized as a major pollutant impacting on sensitive
ecosystems (Bobbink et al., 1992). NH3 deposition may indeed cause soil acidification
through nitrification processes (van Breemen et al., 1982), although this depends upon
the biological and chemical status of the soil on which it is deposited, and upon the
form of NHx deposited (NH3 or NH4+) (Galloway, 1995). Furthermore, atmospheric inputs of NHx may induce eutrophication of sensitive ecosystems, as well as decrease
their biodiversity (Heij and Schneider, 1991; Bobbink et al., 1992). Agriculture is the
main source of NH3 in Europe (Asman, 1992, Heij and Schneider, 1995), and represents more than 80% of the total anthropogenic input at the global scale (Bouwman et
al., 1997). The inventory studies of Jarvis and Pain (1990), of Pain et al. (1998) and
Döhler et al. (2002) show that cattle represents the largest source of agricultural NH3
emissions in Europe. These studies as well as Bussink et al. (1998) also showed that
between 40% and 50% of the emissions from cattle arise from housing and waste storage, housing itself representing about 7% to10%. There are also important emissions
from pig and poultry production, although the emission factor per animal is smaller for
these animals.
1500
1300
Buildings
Sources
Other Grassland
Measurement locations
Arable land
Main sampling site
Wood & Sub-urban
Main Field (grass)
Roads
DWD meteo field
N
Site 6
Background
Tower
Transect measurement
Grass field II
1100
900
Grass field III
PTB
Site 3
Farm area
Site 5
K
J
I
F
H GE
C
500
A
Site 2
300
B
100
Test
crop
-300
Site 4
Site 1
700
-100
100
300
500
Grass field IV
Site 7
Meteo data
700
900
Main Field
“Kleinkamp”
1100
1300
sub-urban housing
1500
1700
1900
Distance (m)
Figure 6.1
Overview of the measurement site. The three locations of NH3 air concentration measurement (grey circles) used in this analysis are indicated
as Site 1, Site 2 and Site 3. The plume transect measurements were
made along the track at Site 3. The sources buildings are labelled from
A to K (See Table 1). Distances are shown on the axes of the map in
meters (For further description of the site, see Sutton1).
Measurement of NH3 emissions from naturally ventilated animal houses is technically
complex, expensive, and labour intensive (Phillips et al., 2001; Scholtens et al., 2004;
84
Welch et al., 2005a, 2005b). This is primarily due to difficulties in determining the ventilation rate, which varies according to temperature, wind speed, building design, orientation to the wind and animal occupancy (Zhang et al., 2005; Welch et al., 2005b), but
also to difficulties inherent in accurately measuring the ammonia concentration on
short-time basis (Phillips et al., 2005). However, methods such as the internal or external tracer-ratio techniques or the flux-sampling technique have been successfully
tested under real conditions (Demmers et al., 2001; Dore et al., 2004).
Dispersion models can be used as tools to estimate ammonia emissions from animal
houses. Among them, the Gaussian-3D model, assuming constant wind-speed (U) and
diffusivity (Kz) with z is widely used because of its simplicity (Gash, 1985). Analytical
models that include variation of U and Kz with height also exist (e.g., Smith, 1957;
Philip, 1959; Yeh and Huang, 1975; Huang, 1979; Wilson et al., 1982). The technique
using Gaussian models to infer NH3 source from a farm building has been reported in
Mosquera et al. (2004). Lagrangian stochastic models have also been tested to infer
sources of tracers from concentration measurements downwind of sources such as
animal houses (Flesch et al., 2005) cattle feedlots (Flesch et al., 2007), and fields
(Loubet et al. 2006, Sommer et al., 2005). More complex Eulerian models have also
been used and validated against extensive measurements for estimating emissions of
NH3 from buildings (Welch et al., 2005b).
In this paper, a Gaussian-3D plume model and a 2D dispersion model (FIDES-2D,
Loubet et al., 2001) are used to estimate the source strength of a set of farm buildings
in an experimental field site in Braunschweig (Germany). The stationary measurements
of ammonia concentrations were used to infer the source strength of several farm
buildings upwind. The inferred emissions are compared with inventory emission factors. The daily variability of the emissions is analyzed in comparison with environmental
factors such as indoor temperature, external radiation and wind speed. The inferred
emission strengths are used in a companion paper (Loubet et al., 2006) as inputs for
quantifying the local advection errors induced by the plume coming from the farm on
NH3 fluxes measured over a grassland site nearby. This study was performed within
the framework of the European project GRAMINAE.
6.2
Materials and methods
6.2.1 Measurement site
The field site is a 12 ha experimental grassland situated in the grounds of the Federal
Agricultural Research Centre (FAL), Braunschweig, Germany. Directly adjacent to the
field are an experimental farm of the FAL and a station of the German Weather Service
(Deutscher Wetterdienst). A more detailed description of the site can be found in Sutton et al. (2009a)
The main source of ammonia in the area is the set of farm buildings A-K (Figure 6.1).
The distance from west to east is named x, whereas the distance from south to north is
called y, and height above ground is z. The distance between the downwind edge of
the farm building area and the different sites are: 230 m for site 3 where plume measurements were done, and 610 m for site 1, were micrometeorological measurements
were done. The farm buildings themselves occupy an area of approximately 180 m (EW) × 300 m (S-N). Table 1 gives an estimate of the yearly NH3 emission for each building identified in Figure 6.1, based on the emission factors of Döhler et al. (2002) which
can be compared to measurement-based estimates emissions factors from Demmers
et al. (2001) in Great-Britain, which are 3.5 and 8.9 kg NH3 animal-1 year-1, for beef and
dairy cattle, respectively.
85
6.2.2 Concentration measurements
The locations of ammonia concentration measurements are shown as grey dots in Figure 6.1. Three AMANDA rotating wet denuders (Wyers et al., 1993) were placed along
the N-S transect at Site 3, which gave NH3 concentrations on a 15 min averaging period at 1 m height. These instruments were also used to calibrate a fast-response mobile NH3 analyser, itself being moved along the North-West line on the adjacent track
(marked as Site 3 in Figure 6.1), to measure the plume cross-section at 1.5 m height.
The background NH3 concentrations were measured with a batch denuder system, located at 42 m height on the top of a tower (Site 6 in Figure 6.1), which was located at
approximately 1600 m East-North-East of the farm and 800 North-East of the grassland. Mean and standard deviation of the concentration was estimated for Site 3 over
the three measurement systems.
For the mobile measurements, a fast response AMANDA sensor was used, with a time
resolution of 30 seconds. This system, described in Erisman et al. (2001), is essentially
similar to an AMANDA, but the liquid flow is higher (10 ml min-1 instead of 1.6 ml min-1).
However, it does not give an absolute concentration, and therefore requires regular
calibrations against a reference. The fast response sensor was placed on a trolley and
moved along a track through the plume cross-section, and calibrated against the three
AMANDA sensors on the same transect (Figure 6.1). Note that the calibration was
done on differing integration time, as the AMANDA integrates over 15 minutes. Four
subsequent transects were used to evaluate the emission strength with these data.
6.2.3 Micrometeorological measurements
Micrometeorological measurements were performed at Sites 1 and 2. The consensus
micrometeorological database derived from the range of measurements performed and
detailed in Nemitz et al. (2009) was used in this study. The wind-speed (U), winddirection (WD), friction velocity (u*), Monin-Obukhov length (L), and sensible heat fluxes
(H), were derived from several ultrasonic anemometer measurements. The latent heat
flux (LE) was derived from eddy covariance measurements using the ultrasonic anemometers in combination with a close-path H2O analyser (Licor 6262) or open-path
H2O analyser (KH2O). Air temperature (Ta), relative Humidity (HR) and global (Rg) and
net radiation (Rn) were also used.
6.3
Inference of the source strength from NH3 concentration
6.3.1 General approach
In this paper, we compare two models: a 3D Gaussian plume model that assumes a
constant wind speed (U) and diffusivity (Kz) everywhere and the Huang (1979) 3D
model (Huang-3D) that assumes that both U and Kz are power low functions of height,
and a plage source. Moreover, the FIDES-2D model, which is also based on Huang
(1979) but assumes cross-wind homogeneity of the source (see Loubet et al., 2001, for
details) is also used to estimate the potential error made by neglecting deposition or
emission of NH3 from nearby fields.
86
Three inference methods were used to estimate the source strength variability in space
and time:
1. The Gaussian-3D model was used in conjunction with mobile NH3 measurements
to estimate the source strength, based on a source distribution equivalent to the
emission inventory.
2. The Gaussian-3D model was used in conjunction with concentrations measured at
the three locations of Site 3 to estimate the spatial distribution of the source using
wind-direction changes.
3. The Huang-3D and the Gaussian-3D model were used in conjunction with 15 min
NH3 concentration measured at Site 3 to estimate the time course of the emissions, knowing its spatial variability.
Table 6.1
Description of the farm animal houses, along with their potential emission estimated using German national emission factors (Döhler et al.,
2002) and inventoried number of animals. The total yearly emission of
the source is 3.5 103 kg year-1 NH3 and the total number of animals is
553. The corresponding average daily total emission is 9.6 kg d-1 NH3.
Source number
on map in Fig.1
Animals
Nr. of
animals
A
B
C
D
E
F
G
H
I
J
K
Cattle
Cattle
Cattle
Cattle
Bulls
Cattle
Calves
Calves
Pigs
Pigs
Pigs
60
60
11
48
64
80
60
0
80
45
45
Total
-
-
Emission factor
Emission
per animal
per building
kg NH3 animal-1
kg NH3
year-1
year-1
10.35
621
10.35
621
10.35
114
10.35
497
3.89
249
10.35
828
3.89
233
3.89
0
1.97
158
1.97
89
1.97
89
-
3.5 103
Percentage
contribution
%
18%
18%
3%
14%
7%
24%
7%
0%
5%
3%
3%
100%
The basic hypothesis made in these approaches are: (a) unless specified, no NH3
deposition is included, as it is considered to be of secondary importance compared with
the effect of dispersion on the concentration (e.g., Loubet and Cellier, 2002); (b) no
chemical reactions are envisaged, as again the effect on the overall NH3 concentration
is not expected to be large on such a small scale (Nemitz et al.; 2009); and (c) the surface characteristics are considered to be homogeneous (i.e., z0, d), since the models
used cannot deal with varying surface roughness, though this is considered to be of
minor importance when compared with the potential perturbations induced by the buildings.
With each of the models, the approach is based on the general superposition principle
(e.g., Thomson, 1987; Raupach, 1989), which relates the concentration at a location (x,
y, z), χ(x, y, z), to the source strength at another location (xs, ys, zs), Sfarm(xs, ys, zs), with
the use of a dispersion function D(x, y, z / xs ,ys, zs) (in s m-3):
c(x, y, z) = cbgd +
⌠ Sfarm(xs, ys, zs) D(x, y, z | xs, ys, zs) dxs
⌡
(6.1)
all xs and ys
87
where χbgd is the background concentration, assumed to be constant with height. The
three models are different in the way they calculate the dispersion function D(x, y, z
| xs, ys, zs) and the different approaches are detailed in the following. In the following
the dispersion function will be written D(x, y, z) for simplicity.
6.3.2 The Gaussian-3D Model
The Gaussian-3D model is based on the assumptions that U and Kz are constant in the
whole domain, which implies that χ(x, y, z) is a function of two independent Gaussian
distributions in the horizontal and the vertical planes (see e.g. Lin and Hildemann,
1997). The contribution from a single source located at (xs ,ys, zs) to a point receptor
located at (x, y, z) is the dispersion function D(x, y, z / xs ,ys, zs), which was calculated
using X = x - xs and Y = y - ys, as:
D( X , Y , z) =
Q
−Y 2
⋅e
2π ⋅ u ⋅ σ y ⋅ σ z
( 2⋅σ y )2
⋅  e −( z −hs )

2
( 2⋅σ z ) 2
+ e −( z +hs )
2
( 2⋅σ z )2


(2)
σ y = g ⋅ X h ⋅ zo0.2 ⋅ T 0.35
σ z = c ⋅ X d ⋅ (10 ⋅ zo )0.53⋅e
(3)
e = X −0.22
where Q is the source strength (in g s-1 NH3), hs the height of the source (animal house)
(in m), σy and σz are the standard deviation of the lateral and vertical concentration distribution respectively (in m), zo is the roughness length (in m), T is the averaging time,
and the parameters g, h, c and d are dependent on the stability classes as detailed in
Pasquill (1974). The model used is a multiple plume model based on the superposition
of several Gaussian plumes each described by equation (2-3). The concentration at
each receptor is the sum of the contribution of all the sources according to (1). The
downwind (X) and crosswind (Y) distances are calculated for each source-receptor
couple as follows:
X = − ( x ( R ) − x ( S ) ) ⋅ sin(WD ) − ( y ( R ) − y ( S ) ) ⋅ cos(WD )
Y = (x ( R ) − x ( S ) ) ⋅ cos(WD ) − ( y ( R ) − y ( S ) ) ⋅ sin(WD )
(4)
where (x(R), y(R)) and (x(S), y(S)) are the receptor and source coordinates, respectively.
6.3.3 The Huang-3D plage source model
In this model based on Huang (1979), D(x, z/xs, zs) is evaluated from a solution of the
advection-diffusion equation obtained assuming power law functions for U(z) and Kz(z):
U ( z ) = az p
Kz ( z ) = bz n
 (Y ) 2  ( zhs ) (1−n ) / 2
 a( z α + hs α ) 
 2a( zhs )α / 2 
Q





D( X , Y , z ) =
exp −
×
× exp −
× I −υ 
2 
2
bαX
bα 2 X 
σy 2π
 2σy 

 bα X 
1
σy =
Cyx
2
88
2−m
2
(5)
Where α = 2 + p + n, ν = (1 - n) / α, and I-ν is the modified Bessel function of the first
kind of order -ν, and Cy and m were determined from Sutton (1932). The values of a, b,
p and n were inferred by linear regression between ln(U), ln(Kz) and ln(z), over the
height 2×z0 – 20 m, using U(z) and Kz(z) estimated from the Monin-Obukhov similarity
theory (see e.g. Kaimal and Finnigan, 1994). Since the model uses stability corrections,
its use is limited to values of the Monin Obukhov length (L) and the friction velocity u*
such that | L | > 5 m and u* > 0.2 m s-1. In this model, the source is assumed to be a
plage source of size 180 m in x and 300 m in y located at 230 m upwind from Site 3
(where concentration is measured).
6.3.4 The FIDES-2D model
The FIDES-2D model was used to evaluate the influence of deposition and emission of
NH3 from downwind fields on the estimation of the source strength. FIDES-2D is based
on the two dimensional dispersion model of Huang (1979), which is hence an integration of eq. (5) over y. The dispersion model is coupled with a surface resistance model
as detailed in Loubet et al. (2001). The equivalent 2D source was considered 180 m
long in x and infinite in y. The source strength per unit area inferred with FIDES-2D has
been multiplied by the equivalent surface of the source: 180 m × 300 m.
6.3.5 Spatial distribution of the farm emissions with the Gaussian 3D
model
The Gaussian-3D model was used to infer the source strength of the farm buildings,
using the ammonia concentration measured at the three fixed points at Site 3 (AM1..3).
The Gaussian-3D model was run in forward mode to estimate the contribution of the
eleven sources in Table 1 to the three receptor concentration levels. The sum of the
eleven time series is compared with the measured data. The eleven emission levels
were modified using the EXCEL (Microsoft) solver function optimizing the correlation
coefficient between the measured and modelled concentration time series. The emission strength of the whole farm area Sfarm was calculated for each 15 minute measurement interval t at the individual receptor locations (AM1..3) using:
Sfarm(t, AMi) = Qinput ×
cmeasured(AMi) – cbgd
cmodel(AMi)
(6.6)
where Qinput is the initial source strength used to estimate χmodel, AMi = (AM1, AM2 or
AM3) are the receptor locations at Site 3. The time-average and standard deviation of
Sfarm was then estimated at each receptor locations AM1..3. The emission from the farm
site was evaluated using all measurements with a wind direction between 200° and
340°.
89
North to South (12:00)
NH3 concentrations (µ g m-3)
10.0
Model 12:00
AM1-3 12:00
9.0
North to South (14:00)
8.0
Model calc 14:00
7.0
AM1-3 14:00
6.0
5.0
4.0
3.0
2.0
1.0
<--- South
Figure 6.2
Space (m)
-60
-10
40
90
140
190
0.0
North ---->
Two fast-sensor measurements of the NH3 plume cross-section at Site 3
the 12/06/2000 at 12:00 and 14:00. The simultaneous concentrations
measurements obtained with the 3 stationary AMANDA systems (AM1..3)
are shown at their corresponding locations in the plume. The figure also
shows the plume modelled with the Gaussian-3D model using the
source distribution and strength as given in Table 1. The difference between the fast sensor and the stationary AMANDA concentrations is
mainly due to the time response of the AMANDA being 15 minutes while
the fast sensor has a 30 time response. For the N-S axis, 0 m corresponds to the intersection with the main E-W track N of Kleinkamp
(Fig.1).
6.3.6 Time course of the farm emissions
To evaluate the time course of the farm emissions, the average concentration at Site 3
(over the three sensors) χmeas(Site3) was used. The Huang-3D, FIDES-2D and the
Gaussian-3D model were all used in the same manner to evaluate the source strength
Sfarm simply as :
(cmeas(Site3) – cbgd)
Sfarm(t) = D(X , Y
Site3
Site3 , z = 1 m)
(6.7)
Where XSite3 and YSite3 are the coordinates of Site 3, and D is the dispersion function. In
the Huang-3D and in the FIDES-2D model the source was a plage-source considered
homogeneous, whereas in the Gaussian-3D model the source distribution was set to
that in Table 1 The wind sector considered was 240° - 300°, which corresponds to a
sector of ± 30° around the average wind direction as discussed in 3.3.
90
Table 6.2
Plume 1
Plume 2
Emission estimates using 4 plume transects measured with the fast response sensor (12 June 2000). Uncertainty of the mean is given as
standard error.
Max Dispersion
Min Dispersion
Average
-1
-1
kg NH3 d
kg NH3 d
kg NH3 d-1
8.4
6.1
7.3
11.4
9.3
10.4
Plume 3
Plume 4
Inventory
6.4
11.1
10.7
9.3
10.4
10.2
10.6
9.6 ± 0.78
Results
6.4.1 Spatial variability of the farm emissions inferred with the fast response NH3 sensor
Figure 6.2 shows two NH3 concentration transects across the plume coming from the
farm, obtained with the fast response sensor at Site 3. The transects were obtained at
midday from 12:00 to 14:00 on June 12. The absolute concentration differs slightly between the two periods, which could be explained on the basis of differences in windspeed.
Four plumes obtained with the mobile measurements system were used to evaluate
the emission of the farm site. A set of two model runs was performed using the source
distribution in Table 1. The potential for large and small initial mixing close to the farm
area was evaluated by setting an initial horizontal dispersion at the farm source of either 50 or 10 metres. Initial dispersion in the vertical direction was set to 5 m. The two
model runs provide concentration patterns along the transect that were compared with
the measured concentration levels. The source estimates for the four plumes and two
model runs are shown in Table 2. The emission needed in the model to make the output fit with the measurements is about 10 % higher for the plumes 2-4 compared with
the inventory source strength. Including plume 1, the emission estimate equals the inventory estimate at 9.6 kg NH3 d-1.
The mobile plume measurements show the shape of the plume providing information
on the lateral dispersion. The model runs for these plumes showed that the lateral dispersion in the model corresponds with the measurement data at a roughness length
zo= 0.2 m. The mobile measurements only covered a small time-period and were hard
to do during the night. Therefore, the spatial distribution of the sources within the farm
area was evaluated further using the three AMANDA stations AM1-AM3 at Site 3. Using
the source distribution of Table 1 (emission inventory), and tuning the overall source
strength Sfam, the modelled and measured concentration patterns for the three
AMANDA locations showed low correlation coefficients of R=0.36, 0.30 and 0.20 respectively (Table 3). The model calculations were used to obtain insight in the sourcereceptor relation for each measurement location. Sources A, B, D & F give the highest
contribution to the modelled signal at the three receptor locations. The time series of
the estimated Sfarm (formula 5) were used to estimate the average, standard deviation
and median (Table 3). The average emission level obtained in this way was 9.8 ± 0.8
kg NH3 d-1, which is close to the expected level of 9.6 kg NH3 d-1. The uncertainty range
of 0.8 kg NH3 d-1 was obtained from the standard errors in table 3. The standard deviation of the source estimate is about 100% this is a combination of uncertainty in the
method and of the actual temporal variation of the source strength (discussed below).
91
In a second step, the correlation between the measured and modelled concentration
pattern at Site 3 was optimised by modifying the source distribution. The results are
shown in Table 4. The source distribution estimated by this means shows a decreased
contribution from sources A, E and F, and an increase in the emission from B and C.
The correlation between measured and modelled concentration patterns increased to
R=0.5 and 0.49 for stations AM1 and AM2. No significant effect is observed for AM3.
The revised total source strength estimate obtained from the time-series at this source
distribution is 9.2 kg NH3 d-1. Accounting for simple uncertainty in estimates by system
AM1-3 (n=3) gives a standard error of 1.6 kg NH3 d-1, which gives a conservative estimate of the uncertainty. Conversely, given the large sample sizes (n = 286 to 441),
standard errors for the individual AMANDA systems are in the range 0.2 – 0.6 kg NH3
d-1.
Figure 6.3 shows an illustration of the measured and modelled concentrations using
the Gaussian-3D model for AM1-3 at Site 3 over a two-day period using the adjusted
source distribution. The correlation coefficient in table 4 shows that 10-25% of the temporal variation can be explained. The differences are expected due to the assumption
of a constant source-strength with time. This shows that the combination of time variation in meteorological conditions and the spatial variability of the source can explain
part of the variability in the concentration, but not all.
Table 5 gives the estimated source strength using the Gaussian-3D model, as well as
the source strength estimated using the emission factors of Döhler et al. (2002). The
modification in the source distribution, which was done in order to increase the correlation between measurements and model, suggests that the sources in the south have a
low emission, whereas the sources in the north are a factor 3-5 higher. For building
A,E,G the Gaussian-3D model significantly underestimates the relative source strength,
whereas it overestimates for buildings C,D, I, and K.
6.4.2 Temporal variability of the farm emissions estimated with the
Huang 3d model
Figure 6.4 shows an example result of inferred source strength (kg NH3 day-1) using the
Huang-3D model for the windsector 240°-300°. Corresponding hourly source strengths
range from 0 to 0.5 kg h-1 NH3. Figure 6.4 shows a clear diurnal variation with a maximum emission at midday and a minimum during the night. This daily variability is not
always due to concentration changes, as shown in Figure 6.4 (bottom graph) where
concentration difference χ - χbgd is constant and the source strength varies. The variability is also due to the turbulent diffusivity increasing during the day, therefore requiring a larger source to generate a similar concentration at a given distance. Averaging
the inferred hourly emission rates over all the available data (for westerly winds), and
converting to daily emission rates lead to an average emission of 6.4 kg d-1 NH3, with a
standard error of 0.18 kg d-1 NH3 (N=495).
The inferred emission strength was averaged to hourly values in order to estimate a
mean daily pattern (Figure 6.5). The daily pattern was observed throughout the period,
with a minimum between 01:00 and 02:00 GMT in the morning and a maximum between 07:00 and 09:00 GMT in the morning, then decreasing towards the end of the
day. This pattern is partly reflected by the concentration difference between Site 3 and
the background concentration, also shown in Figure 6.5. Note that there are not many
data at night, due to u* or | L | being below the model requirement under stable conditions, so that the results obtained for night-time conditions are based on less data than
for the day.
92
Table 6.3
Source-receptor relation and source estimate obtained with the Gaussian-3D model using the spatial source distribution as obtained using the
inventory data
Contribution in %
Q kg NH3 d-1
St.
St.
Corr
A B C D E F
G&H I J K Mean
N
dev
error* (R)
AM1 26 30 3 9
4 13 3
6 2 2 7
4
456 0.19
0.36
AM2 22 26 4 11 6 15 4
7 3 3 10
10
568 0.42
0.30
8 3 4 12
14
459 0.65
0.20
AM3 18 23 4 10 6 19 5
Mean for 3 stations
9.8
*standard error =st. deviation/sqrt(n)
Table 6.4
Source-receptor relation and source estimate obtained with the Gaussian-3D model after modification of the spatial distributions of sources
within the farm area.
Contribution in %
Q kgNH3 d-1
St.
St.
Corr
A B C D E F G&H I
J K
Mean
n
Dev
error* (R)
AM1 0 36 17 25 0 0 0
15 1 6
6
4
286 0.24
0.50
AM2 0 29 22 24 0 0 0
18 1 6
10
8
441 0.38
0.49
AM3 0 23 21 19 0 0 0
26 1 11
11
11
396 0.55
0.24
Average for 3 stations
9.2
*standard error =st. deviation/sqrt(n)
6.5
Discussion
6.5.1 Emissions from the farm buildings estimated with different techniques and sensitivity analysis
The averaged daily emission estimated with the Gaussian-3D model was found to be
9.2 ± 0.7 kg d-1 NH3, and 6.4 ± 0.18 kg d-1 NH3 (± standard errors) with the Huang-3D
model, while it was estimated at 9.6 kg d-1 NH3 with the emission factors approach. The
inferred emission strength with the Gaussian-3D and the Huang-3D model would represent 3.3 tonnes NH3 year-1 and 2.3 tonnes NH3 year-1, respectively. Although, the inventory emission factors do not include formal uncertainty estimates, the comparison
with the estimates of Demmers et al. (2001) for dairy cattle of 8.9 kg NH3 animal-1 yr-1
compared with 10.35 kg animal-1 yr-1 of Döhler et al. (2002) suggests that the difference
between the Gaussian-3D estimates and the inventory (5%) is within the range of uncertainties. Conversely, the Huang-3D model was 37% and 33% lower than the inventory and Gaussian-3D model estimates, respectively. The difference may result from
the fact that the Gaussian-3D model uses a constant wind speed and diffusivity profile,
and a source height of 4 m while in the Huang-3D model, the wind speed increases
with height and the source is at 1 m height. It may also be due to the difference in the
source treatment: the Gaussian-3D model considers point sources.
93
Measured
AM2
AM3
100
90
80
70
60
50
40
30
20
10
0
AM3
AM1
Figure 6.3
17-06
16-06
15-06
14-06
13-06
12-06
11-06
10-06
09-06
08-06
07-06
06-06
05-06
04-06
03-06
02-06
01-06
31-05
30-05
29-05
28-05
27-05
26-05
25-05
24-05
21-05
20-05
AM2
23-05
-3
-3
Source strength kgNH3/day
NH3 bkg
AM1
22-05
40
35
30
25
20
15
10
5
0
-3
NH3 (µg m )
NH3 (µg m )
40
35
30
25
20
15
10
5
0
NH3 (µg m )
bkg+modelruns
40
35
30
25
20
15
10
5
0
Example of measured and modelled concentration with the Gaussian-3D
model by AM1-3 at Site 3. The source-strengths of the farm buildings
were estimated by minimizing the difference between the measured and
the modelled concentrations at all locations.
In order to better understand the uncertainty in the different emission estimates a sensitivity analyses was done. The result of this exercise is summarized in Table 6 with the
major parameters that might affect the estimation of the emission strength using the
two models 3D dispersion models, as well as the effect of taking into account dry
deposition downwind of the source using FIDES-2D.
It can be seen that the Huang-3D and the FIDES-2D models are sensitive to the height
of the source and the roughness length with a maximum error around 20%. The Gaussian-3D model calculation showed a similar result for the zo change from 0.1-1. The
Gaussian-3D model uses different zo values per source with larger values for stables at
the back of the farm. A change of 50% of all these values had a 10% effect on the average emission level.
The difference in source height, already mentioned above, can have a 20% effect for
the Huang-3D model with a lower emission estimates when the source height in-
94
creases. This will bring the Huang-3D model and the Gaussian-3D model closer together. The source height range in Table 6 only shows a small (2%) effect on the
Gaussian-3D model estimate. Only when changing the source height further for example from 4 to 10 m this model would decrease the emission estimate by 8%. For the
Gaussian-3D model changing the stability classification at a 15 minute interval from
Pasquill class D neutral to C or E would increase or decrease the emission estimate by
40% respectively. Furthermore, changing the initial dispersion in the wake of the buildings from 5m to 10m reduced the emission by 20%.
20
Whole farm emission
15
10
5
0
Concentration
(µg NH3 m-3)
22/05/00
29/05/00
25
GMT
05/06/00
12/06/00
25
Concentration
Whole farm emission
20
15
Concentration
(µg m-3 NH3)
40
35
30
25
20
15
10
5
0
Concentration
20
15
10
10
5
5
0
0
06/06 00:00
Figure 6.4
Emission
(kg NH3 day-1)
Emissions
(kg day-1 NH3)
25
06/06 12:00
07/06 00:00
GMT
07/06 12:00
08/06 00:00
(top graph) time course of the NH3 emission strength as inferred with the
Huang-3D model using the averaged concentration measured for the
wind direction 240°-300° at Site 3 and the background concentration
measured at Site 6. Also shown is the concentration difference Site 3 –
Site 6. (bottom graph) magnification for the period 6-8 June.
The Huang-3D model is very sensitive to the length (in x direction) of the source (26%
decrease for a 50 m source and 151% increase for a 400 m source). The measured
NH3 concentration at 230 m downwind from the farm will be lowered by the dry deposition taking place between the farm buildings and Site 3. This was studied with the
FIDES 2D model and the effect was in the order of 40%. When correcting both the
Gauss model and the Huang 3D model, the latter will be close to the emission inventory result and the Gauss model estimate will be above that level.
Neglecting the deposition issue for the Gauss model, the different sensitivity runs provide a set of emission estimates that have a standard deviation of 25 % around the
emission level of 9.2 kg NH3 d-1. This uncertainty range is probably more accurate compared to the calculated standard errors bases on the large number of observations.
The freedom of choice in the input parameters for the model would be reduced significantly with data available on the vertical distribution of NH3 downwind of the source.
Additional data along a vertical profile up to about 10 m height would have provided
better constraints on the parameters that set the vertical dispersion in the model.
95
Table 6.5
Ammonia source-strength estimated with the Gaussian-3D model, as
compared to that estimated from the emission factors of Döhler et al.
(2002). The total amount of ammonia emitted per year estimated using
the Gaussian-3D model is 3.3 103 kg year-1 NH3 as compared to the inventory estimate of 3.5 103 kg year-1 NH3. These values correspond to
averaged daily emissions of 9.2 kg d-1 NH3 and 9.6 kg d-1 NH3, respectively.
Estimated from emission
factors (Döhler et al.,
2002)
Building
(Figure 6.1)
A
B
C
D
E
F
G
H
I
J
K
Total
Animals
Nr
Emission
of
per building
animals
kgNH3
-1
year
Cattle
60
621
Cattle
60
621
Cattle
11
114
Cattle
48
497
Bulls
64
249
Cattle
80
828
Calves
60
233
Calves
0
0
Pigs
80
158
Pigs
45
89
Pigs
45
89
3
3.5 10
Estimated with the
AMANDA measurements
(Site 3) and
Gaussian model (this study)
Percentage
contribution
Emission
per building
Percentage
contribution
%
kg NH3
-1
year
0
729
559
598
0
4
0
0
926
43
483
3
3.3 10
%
18%
18%
3%
14%
7%
24%
7%
0%
5%
3%
3%
100%
1000
0%
22%
17%
18%
0%
0%
0%
0%
28%
1%
14%
100%
10
Emission strength
-1
(g NH3 hour )
800
8
C - Cbgd
600
6
400
4
200
2
0
0:00
4:00
8:00
12:00
16:00
20:00
Concentration difference
-3
(µg NH3 m )
Srce strength mean
0
0:00
GMT
Figure 6.5
96
Diurnal variability of the farm NH3 source strength as estimated with the
Huang-3D model for a wind sector of 240°-300°. The concentration difference between Site 3 and the background at Site 6 (C – Cbgd) is also
given. The dark circles represent the source strength during the whole
period that has been hourly averaged to give the mean source strength
for each hour. The error bars are the standard errors.
Table. 6.6
Methodology
Emission factors
Inverse
Huang-3D
Inverse
Gaussian-3D
Inverse
FIDES-2D
*
Daily emission of the source estimated using the emission factor and the
modelling approach, and sensitivity of each estimate to micrometeorological parameters. Standard parameters used were: hsrc = 1.0 m,
z0 = 0.10 m, Source width = 180 m, and no dry deposition. In the sensitivity analysis, hsrc have been set to 0 and 5 m height, z0 has been set to
0.01 m and 1.0 m, the source width has been set to 50 and 400 m, and
in the deposition sensitivity analysis, the stomatal compensation point s
has been set to 0 and 1 µg m-3 NH3. The cuticular Rw and stomatal resiswere
estimated
as
Rw = 7 exp((100 - RH) / 12),
and
tance
Rs
Rs = 30 × {1 + 200 / max(0.01, St)}, where RH is the relative humidity at
z0' and St is the global solar radiation.
Daily
Sens. to
Sens. to
Source width
hsrc
emission
z0
(± std.
(0 – 5 m) (0.01 – 1.0 (50 –400 m)
err.)
m)
-1
-1
-1
-1
kg d NH3 Kg d NH3 kg d NH3
kg d NH3
(%
(%
(% change) (% change)
change)
change)
9.6
-
-
-
*
6.0 - 7.5
-3%
+23%
5.9 – 7.7
16.8 – 4.2
-6% +23% +151% -26%
9.2 ± 2.7
9-9.3
1% -2%
6-10.9
-35% 18%
5.8 - 7.2
-4%
+20%
5.6 - 7.4
-6% +23%
6.4 ± 0.18
5.7 ± 0.16
*
Deposition
s
(0 – 2 µg m
3
NH3)
-1
kg d NH3
(% change)
Stability
class
C-E
-1
kg d NH3
(%
change)
-
5.8-10.3
-36% 11%
4.9 - 7.6
-19% +27%
8.7 - 8.5
45% 41%
standard error estimated with 632 data.
The analysis shows that deposition as well as the source geometry can really influence
the emission strength inferred with inverse modeling techniques. However of these two
effects, deposition is probably the more problematic in the case of emissions from farm
building, since there is a much larger uncertainty on the deposition parameters (Rs and
particularly Rw and s) than on the source geometry. Indeed, the fields in between the
source and Site 3 were patches of small crop trials of varying Rs, Rw and s, which
makes it difficult to define a unique surface characteristic. Moreover, Rw is a very uncertain parameter that has a major influence on local deposition (Flechard et al., 1999;
Loubet and Cellier 2002; Burkhardt et al. 2009). This constitutes a major uncertainty in
the dispersion model approach, which is difficult to overcome. One way would be to
perform the measurements closer to the source and at higher levels (such as Welch et
al. 2005b). However close to the farm, the influence of the farm buildings on the flow
would prohibit the use of Gaussian-like models, and would require more sophisticated
approaches. This study also suggests that the use of a 3D model relies on a precise
description of the size of the source.
Finally, apart from the model uncertainties there are additional possibilities that can
cause a difference between the emission factor calculation and the model result:
1. The small sampling time due to the concentration measurements being done only
on the east of the farm, which may hence induce a bias if wind directions show a
daily pattern. This was indeed the case as the wind was blowing from the west 7%
of the time at 15:00 and only 1.5% of the time at 5:00. However, this would have
lead to a bias toward higher emission estimates.
97
2.
The experimental farm seemed especially well managed with surfaces regularly
cleaned of manure, so that the emission factors estimated from national inventories might indeed be an overestimate for the actual emission level at this site.
Emission strength (g NH3 hour-1)
500
450
400
350
300
250
200
150
100
50
0
14
16
18
20
22
24
Indoor temperature (°C)
Figure 6.6
Hourly averaged emission strength for a wind sector of 240°-300° as a
function of hourly mean indoor temperature (measured for the main cattle building A). The bold line is a regression with equation: Source [g h-1
NH3] = 2.03 exp(0.2342 × T[°C]), R2 = 0.771. The error-bars are ± standard deviation.
In literature the underestimation of emissions by the dispersion models has been observed before. For example, Welch et al. (2005b) found a collection efficiency of 80%
using the ADMS model in a controlled NH3 release experiment. However, Flesch et al.
(2004) found a much better collection efficiency with a methane tracer using a backward Lagrangian Stochastic model under flat terrain conditions. Flesch et al. (2004)
also showed that the modelling approach was not reliable under strongly stratified conditions. Michorius et al. (1997) evaluated that the underestimation of NH3 emission by a
farm building using a Gaussian-3D approach and concentrations measured at 100 m
varied from 47% to 68% of the expected emission.
6.5.2 Daily variability of the emissions from a naturally ventilated building
Figures 6.4 and 6.5 show that the emission strength followed a clear diurnal pattern,
with a maximum in mid-morning and a minimum at night. Although some of this variability is reflected in the concentration change (Figure 6.5), this is not always the case,
because of the daily pattern of the turbulent diffusivity. The question arises whether this
variability in emission is real or is a bias linked with the inference method. Indeed, the
emission from farm houses with forced ventilating systems can be rather constant with
time, due to a rather constant indoor temperature regulated by the ventilation system.
However, at the FAL farm studied here, most of the buildings are naturally ventilated,
which implies that the indoor temperatures fluctuate and that the flow rate through the
buildings also changes with environmental conditions (Welch et al. 2005b, Zhang et al.,
2005). This seems to be confirmed by Figure 6.5, where one can see a morning maxi-
98
mum, which might for example be due to a flushing out of NH3 accumulated during the
night.
800
1.0
(a)
Srce strength mean
Emission strength
(g NH3 hour-1)
Forced convection velocity
600
0.8
0.7
400
0.6
0.5
200
Convection veclocity
(mm s-1)
0.9
Free convection veclocity
0.4
0
0.3
800
1.9
Srce strength mean
Modelled convection velocity
600
1.7
1.5
400
1.3
200
Convection velocity
(mm s-1)
Emission strength
(g NH3 hour-1)
(b)
1.1
0
0:00
4:00
8:00
12:00
16:00
20:00
0.9
0:00
GMT
Figure 6.7
Hourly averaged emission strength for a wind sector of 240°-300°, (a)
free and forced convection velocities calculated as 1 / Rb, and (b) resultant total convection velocity (sum of free and forced convection velocities). The error-bars are ± standard errors. A linear regression between
the convection velocity and the emission strength gives an R2 = 0.97
Figure 6.6 shows the hourly-mean emission strength as a function of the indoor temperature of the main cattle building (A in Figure 6.1). This suggests that the inferred
daily variability of the emission strength is real, since NH3 emission is expected to vary
as the exponential function of the temperature according to the Clausius-Clapeyron law
(e.g., with a doubling of emission for every 5 °C increase, Sutton et al. 2001). This dependence of NH3 emissions to indoor temperature was experimentally demonstrated by
Zhang et al. (2005) over a range of temperature 5-23°C. However Zhang et al. found a
maximum increase by a factor of 3 over the range of temperature observed here (1423°C), whereas Figure 6.6 shows an increase of up to a factor of 7. On the basis of the
solubility equilibria, a factor 4 increase would have been expected. This suggests that
the response of Figure 6.6 may include effects of temperature as well as correlated effects linked with changes in the ventilation rate.
99
To better understand the factors that cause the observed daily emission pattern, a free
and a forced convection velocity were calculated by adapting Monteith and Unsworth
(1990) and Murphy et al. (1977) approaches, respectively. These velocities were computed as the inverse of the transfer resistance Rb, using u* as a velocity scale, the difference between indoor and outdoor temperatures, and a characteristic size of the
building of 10 m was taken. Although the expressions from Murphy et al. (1977) are not
adapted to free and forced convection in buildings, they can give a good qualitative information on the daily variability of the ventilation rate. They are shown in Figure 6.7,
calculated from hourly averages, along with the source strength, and the resultant convective velocity (the sum of the two). The daily pattern of free convection (7a) shows
two maxima, one between 4h and 8h, and the other around 22h. This pattern is due to
a time de-correlation between the indoor and the outdoor temperatures, which might be
explained by the building being heated more rapidly than the air in the morning (because of the solar radiation onto the roof and small ventilation), and a longer decrease
at night due to the naturally forced ventilation being small at that time. On the opposite,
the modelled forced convection velocity follows directly the daily pattern of u*.
It can be seen that the resultant convective velocity follows very well the source
strength pattern (Figure 6.7a). A linear regression between the convective velocity and
the emission strength gives an R2 = 0.67. This suggests that the emission pattern observed in Figure 6.5 is probably caused by (i) the indoor concentration increasing with
indoor temperature, as suggested by Figure 6.6, and (ii) the ventilation rate increasing
also during the day as resulting from the combination of natural convection (indoor
temperature change), and forced convection (external wind).
6.6
Conclusions
Within the framework of the European GRAMINAE project, an intensive joint field experiment was performed at the FAL research station in Braunschweig (Germany), during May and June 2000. This experiment, summarized in Sutton et al. (2009b) has
brought together many atmospheric NH3 concentration measurements techniques located at several sites around a cluster of farm buildings. This gave a great opportunity
to use the measured NH3 concentration, as well as mobile fast sensor measurements
to infer the emission from the farm building with inverse modelling technique. Three
models were used, a Gaussian-3D plume model, the local dispersion model of Huang
(1979), and the FIDES-2D model to account for NH3 deposition downwind from the
source.
The inferred emission strength was on average 6.4 kg d-1 NH3 for the Huang-3D model
and 9.2 ± 0.7 kg d-1 NH3 for the Gaussian-3D plume model. The mobile NH3 measurements provided valuable data on the horizontal dispersion of the NH3 plume form the
farm houses. These data were used to constrain the dispersion model parameters.
Concentration measurements of the vertical distribution of NH3 that could be used to
evaluate the vertical dispersion of the NH3 plume were not available but are recommended for similar experiments in future.
A sensitivity analysis showed that the inference method was very sensitive to the
deposition scheme used, and when a maximum deposition was applied, the farm emission strength could be increased by 45%. The height and size of the source, the surface roughness, the Pasquill stability classes were found to influence the emission
strength estimates both with the Huang-3D and the Gauss model by up to 50%. The
source-strength exhibited a clear diurnal cycle with a maximum in the morning (07:00–
08:00 GMT) and a minimum at night. This variability can be fully explained by changes
in the indoor temperature and the ventilation rate.
100
In the context of remaining uncertainty in the inventory estimates and lack of independent measurement of NH3 emissions from the buildings, the present work does not fully
validate the approach used. However, the application of dispersion models combined
with NH3 concentrations measured at large distances downwind, provides an approach
in close agreement with the inventory, while this study demonstrates the need to consider other interacting factors, such as dry deposition between the source and the
measurement location.
Acknowledgements
The work presented here was partly funded by the EU FP5 GRAMINAE Project (EU
contract ENV4-CT98-0722). Final synthesis of this paper was conducted as part of the
NitroEurope Integrated Project. We thank the Spanish Commission of Advanced Education and Scientific Research (Dirección General de Enseñanza Superior e Investigación Científica) who provided funding for J. Mosquera to come to the Energy Research
Center of the Netherlands (ECN) as a postdoctoral fellow, and the UK Department for
Environment Food and Rural Affairs (Defra).
101
CH4 from dairy cows, Dronrijp 2007
102
Chapter 7 Dairy farm CH4 and N2O emissions, from one square
metre to the full farm scale4
A. Hensen*1, T.T. Groot1, W.C.M. van den Bulk1, A.T. Vermeulen1. J.E. Olesen2,
K. Schelde2.
1
ECN, Petten, P.O.Box 1, 1755 ZG, The Netherlands
2
Danish Institute of Agricultural Sciences, P.O. Box 50 DK-8830 Tjele, Denmark
*
Corresponding author: E-mail: hensen@ecn.nl
Abstract.
The greenhouse gas emissions from agricultural systems contribute significantly to the
national budgets for most countries in Europe. Measurement techniques that can identify and quantify emissions are essential in order to improve the selection process of
emission reduction options and to enable quantification of the effect of such options.
Fast box emission measurements and mobile plume measurements were used to
evaluate greenhouse gas emissions from farm sites. The box measurement technique
was used to evaluate emissions from farmyard manure and several other potential
source areas within the farm. Significant emissions (up to 250 gCH4.m-2day-1and 0.4
gN2O.m-2day-1) from ditches close to stables on the farm site were found.
Plume emission measurements from individual manure storages were performed at
three sites. For a manure storage with 1200 m3 dairy slurry in Wageningen emission
factors of 11 ± 5 gCH4.m-3manure.day-1 and 14 ± 8 mgN2O.m3manure.day-1 were obtained in February 2002.
Mobile plume measurements were carried out during 4 days at distances between 30
and 300 m downwind of 20 different farms. Total farm emissions levels ranged from 14
to 95 kg CH4 day-1 for these sites. Expressed as emission per animal the levels were
0.7 ± 0.4 kgCH4.animal-1.day-1 for conventional farms. For three farms that used straw
bedding for the animals1.4 ± 0.2 kgCH4.animal-1.day-1 was obtained. These factors include both respired methane and emission from manure in the stable and the outside
storages.
For a subset of these farms the CH4 emission was compared with monthly averaged
model emission calculations using FARMGHG. This model calculates imports, exports
and flows of all products through the internal chains on the farm using daily time steps.
The fit of modelled versus measured data has a slope of 0.97 but r2= 0.27. Measurements and model emission estimates agree well on average, for large farms within
30%. For small farms the differences can be up to a factor of 3. CH4 emissions during
winter seem to be underestimated.
Keywords: farm, greenhouse gas emissions, nitrous oxide, methane, N2O, CH4.
4
Reference: Hensen, A., Groot, T.T., Van Den Bulk, W.C.M., Vermeulen, A.T., Olesen, J.E., Schelde, K.
Dairy farm CH4 and N2O emissions, from one square metre to the full farm scale (2006) Agriculture,
Ecosystems and Environment, 112 (2-3), pp. 146-152.
103
7.1
Introduction
Many agricultural activities contribute to emissions of the greenhouse gases, methane
(CH4) and nitrous oxide (N2O), to the atmosphere (Oenema et al., 2001). Methane is
produced where there is a readily available source of carbohydrates and conditions are
anaerobic. Under temperate conditions large CH4 emissions occur from enteric fermentation, in particular from ruminant animals, and from anaerobic storage of manures.
There may, however, also be other sources of methane on livestock farms, such as
ditches and feed storage. Nitrous oxide emissions occur from all parts of the agricultural nitrogen cycle (Mosier et al., 1998), and there may therefore be many diverse
sources of N2O emissions from livestock farms.
In this study fast box measurements were used on a small spatial scale to search for
unknown sources of CH4 and N2O emissions on the farm site. Gaussian plume measurements were tested both for the evaluation of the emission from manure storage and
on the total-farm scale (at 30-300 m away from the farm). The measured emission levels are used to derive emission factors that can be used in farm emission models. This
was exemplified by comparison of the farm CH4 measurements with the FARMGHG
model (Olesen et al., 2005).
7.2
Materials and Methods
7.2.1 Measurement locations
Emission measurements were carried out on multiple farms in the Netherlands. The
two main experimental regions were North Holland (52º46 N, 5º40’E) and Brabant
(51°40'N, 5°30'E,). Fast box measurements, manure storage measurements and the
first plume measurements were carried out in Lelystad (52º31’N 5º33’E), Wageningen
(N51°58', E5°38') and Hardinxveld (51º50’N 4º49’E) respectively.
7.2.2 CH4 and N2O concentration measurements.
Both for Gaussian plume measurements and the fast box measurements a Tuneable
Diode Laser absorption spectrometer (TDLAS, Aerodyne Res. Inc., Billerica, Mass.,
USA) was used to measure the CH4 and N2O concentrations. Gas was sampled into a
30 m closed path multiple reflection sample cell. Light absorption at the 1270 and 1271
cm-1 infrared absorption lines were used for CH4 and N2O, respectively. The system
has a 10 ppb resolution for CH4 and 20 ppb for N2O at 2 Hz data acquisition.
7.2.3 Gaussian plume method
The mobile TDL (Zahniser et al., 1995) was used to measure CH4 and N2O concentrations downwind of a source. A transect was measured through the emitted plume of
the source (Fig. 7.1). While driving, air was collected at the front of the van at 3m
height. Also a handheld inlet system of 60 m was used when no road is available. A
GPS system (GPS 76, Garmin Inc.) logged position along the transect. Concentration
measurements on the transect were compared with Gaussian plume model calculations (Hensen and Scharff, 2000). The farm emission was equal to the source strength
needed in the Gaussian plume model achieve an agreement between the integral of
the modelled and measured concentration pattern along the transect. Meteorological
data (wind speed and wind direction) were obtained either from the van or from a separate mast. The reported emission level at a particular farm is the average of a set of
emission estimates for individual plume transects. In general 3-5 plumes were available
at each farm. The standard deviation of the estimates is reported and is generally
104
about 25% with a range between 10 and 50% depending on the reproducibility of the
plume measurement.
Wind
Figure 7.1
The concept of the plume measurement technique. The emitted gas
from various sources on the farm is moving along with the wind. On a
road downwind the increased concentration is measured.
7.2.4 The fast box measurement technique.
The TDL measurements were also used in the fast box method configuration (Figure
7.2) Gas from an emitting surface was trapped in a 0.8×0.8×0.5 m box. Gas sampling
from the box took place at a flow rate of 4l/minute using a 60m 1/4" PTFE tube. The
gas flows continuously to the TDL and to the box to avoid over or under pressure in the
box. Air in the box is mixed using two fans. When the box is placed on the soil the concentration starts to increase. The increase in concentration within 0.5-1 minute was
used to calculate the emission level. A tracer (N2O was used in this study) can be injected into the box to evaluate the delay-time caused by the 60m sampling line and potential leaking in the system. The uncertainty in an individual box measurement is estimated to be between 10-20%. This is a combination of the uncertainty in the concentration measurement and the volume of the box (variability in height of the emitting surface under the box). When the concentration measurement shows strong variations the
box does not seal properly. These measurements are not used.
7.2.5 "Under cover" measurements.
The N2O emission from covered manure storages was too small to be evaluated using
the plume technique with the current laser setup. Therefore the N2O emission level was
evaluated indirectly using the CH4 emission level. We assume that the emissions (Q) of
both N2O and CH4 are related to the concentration difference between the storage interior and the ambient concentration. The emission can be calculated using:
Q=[Conc.inside -Conc.ambient ] * k
(7.1)
105
Assuming that the mixing factor k is identical for both CH4 and N2O we can write:
Q(N2O) = {[N2O]inside-[N2O]ambient} / {[CH4]inside-[CH4]ambient} * Q(CH4) plume
(7.2)
The methane plume measurements provide Q[CH4] and the ambient concentration levels. “Under-cover” measurements in the headspace of the covered storage provide
[N2O]inside and [CH4]inside . Measurements of the N2O and CH4 concentration for this
method were obtained with an Opto Acoustical instrument (Bruel & Kjaer 1304). The
uncertainty in CH4 emission level (about 20-30%) based on the plume measurement is
the main source of uncertainty for this evaluation. The concentrations in the storage are
relatively high and easy to measure within 2-5 %. This means a 20-30% uncertainty for
the CO2 emission and 30-40% for the N2O emission.
CH4
N2O
(tracer)
TDLAS
//
//
tracer
0.8 m
Figure. 7.2
The setup for the fast box measurements. Air is pumped from the box to
the TDL and back. A tracer can be applied in the return flow to test the
system. A ventilator inside the box mixes the air.
7.2.6 Theoretical emission calculation
The methane emissions from each farm level were calculated using the FarmGHG
model (Olesen et al., 2005). FarmGHG is a model of C and N flows on dairy farms,
which quantifies all direct and indirect gaseous emissions from dairy farms, so that it
can be used for assessment of mitigation measures and strategies. The imports, exports and flows of all products through the internal chains on the farm are modelled using daily time steps. The model includes the N balance, and it allows calculation of environmental effect balances for GHG emissions (CO2, CH4 and N2O) and eutrophication
(NO3 and NH3). The methane emission from enteric fermentation is calculated using
the empirical equation of Kirchgessner et al. (1995). The methane emission from stored
slurry is calculated from the amount of volatile solids in the manure using a temperature dependence as described by Sommer et al. (2004). The manure management in
the model was adjusted to match the observed manure stock during the month of
measurements, and the methane emissions were taken from simulated values of the
livestock and manure handling from the respective month of measurements. For comparison with the results of the FarmGHG model, methane emissions were also calculated using simple emission factors (Table 7.1).
106
Using these emission factors, the CH4 emission (Q) from a farm was calculated as:
Q = (No.Dairy)*Ed+(No.Young)*Ey+(No.Calves)*Ec +(m3 Slurry)*Es+(m3 FYM)*EF (7.3)
Table 7.1
Emission factors used for simple methane emission calculation.
Source
Label
Emission factor
Dairy cows (adult)
Ed
274 g CH4 day animal
Young cows
Ey
170 g CH4 day animal
Calves
Ec
48 g CH4 day animal
Manure (Slurry )
Es
53 g CH4 day m
Farmyard Manure
FYM
40 g CH4-C day m
-1
-1 *1
-1
-1 *1
-1
-1 *1
-1
-3
-1
*1
Annual average emission factors (van Amstel et al., 2003)
2004)
7.3
-3 *2
*2
Levels obtained in summer (Sneath et al.,
Results and Discussion
7.3.1 Fast box measurements on the farms
Emission measurements were performed on several farms using a fast box measurement technique. In this report we show results of a short survey for unknown GHG
sources on an experimental farm in Lelystad (52°31'N, 5°33'E) with 400 dairy cows.
(Table 7.2). Measurements were performed during two mornings. The main finding was
that a ditch close to the farmhouse showed significantly higher emission levels for both
N2O and CH4 compared to a ditch between two perennial ryegrass (Lolium perenne)
grasslands.
Table 7.2
Time
Results of the survey measurements of CH4 and N2O on different parts
of the farm. (All emission levels have an uncertainty range of 10-20%)
Label
CH4
slope
-1
ppb.s
N2O
slope
-1
ppb.s
gCH4.
-2
-1
m day
gN2O
-2
-1
m day
area
2*
m
kgCH4. gN2O. CO2 eq
-1
-1
-1
day
day
kg.day
(CH4) (N2O)
8-10-2002 9:35
Ditch 1
10-10-2002 12:52 Ditch 2
#
10-10-2002 13:02 Ditch 2b
8-10-2002 9:50
Grass
leftover
10-10-2002 13:20 Silage
10-10-2002 13:40 Silage
9.7
-
0.3
8248
176
0.7
4.59
0.14
0.66
257
5
27
24
0.72
3.19
0.9
0.8
1000
0.31
0
6
0
0.40
20
5
8
108
2
0.06
50
0.00
3
0
1
0.06
0.28
50
50
0.04
0.04
3
14
1
1
1
4
Enteric Fermentation estimate:
no. Cows 150
40.5
851
*Rough estimate of the area for which the emission was representative.
#
Ditch 2b (farm yard) showed a slow increase after initial flux. This was the minimum estimate !!
Figure 7.3 (right) shows the concentration increase observed for both N2O and CH4 after the box was placed on the water surface of a ditch close to a farm. The CH4 emission was estimated to be equivalent to 250 gCH4.m-2.day-1. This is three orders or
magnitude above the emission observed at the ditch in the field.
107
Ditch next to a grass field
2800
Ditch close to farm
400
200000
350
180000
650
600
2600
160000
300
200
CH4 (ppb)
2200
N2 O (ppb)
250
150
2000
100000
450
80000
400
60000
100
1800
500
120000
CH4
50
350
40000
CH4
20000
N2O
300
N2O
1600
0
9:32:00 9:34:00 9:36:00 9:38:00 9:40:00 9:42:00 9:44:00
Figure 7.3
0
12:51:00
12:52:00
12:53:00
12:54:00
250
12:55:00
Methane and nitrous oxide concentrations measured using a fast box in
a ditch between two grasslands (left) and in a ditch close to a farm
(right). The increase in the curve starts when the box is put on the surface. Lifting the box causes the decrease in concentration.
A clear emission of N2O and CH4 was observed from a fresh heap of maize silage.
Older silage heaps did not show significant emissions. Table 2 summarises the locations where emissions were found. An estimate was made of the area on the farm for
which these measurements were representative. The results can be put into perspective by comparing these emission estimates with the CH4 emitted due to enteric fermentation by the 150 cows present on this farm (40 kg CH4.day-1, using the emission
factor in Table 7.1). Table 7.2 shows that on the day of measurements an area of 20
m2 "active" Ditch has an emission that is equivalent to 10% of the level expected from
enteric fermentation.
7.3.2 Measurements at manure storages
Plume measurements downwind of a cattle slurry storage took place to estimate the
CH4 emission on a farm in Wageningen (N51°58', E5°38') in February 2003. A set of 8
plumes was obtained at 50 m distance from the storage. From these data a source
strength of 0.15 ± 0.03 gCH4 s-1 was derived. The uncertainty in the estimate was
mainly due to the fact that the manure storage plume was on the edge of the plume
that originates from the animal house. The concentrations of CO2, N2O and CH4 in the
headspace of the covered storage were measured using an opto acoustic device (Bruel
& Kjaer 1305). Using these concentrations, the N2O emission from the manure storage
can be calculated using eqn [2] (Table 7.3). This gives a source-strength in the order of
200 µg N2O s-1 for the manure storage. The total amount of manure stored was 1200
m3 so the emission factors were 11 ± 5 gCH4 m-3 manure day-1 and 14 ± 8 mgN2O.m3
.manure.day-1. The CH4 emissions were in the same order of magnitude as the emission rate of about 19 gCH4 m-3 manure day-1 obtained for a covered cattle slurry tank in
the Netherlands (Hilhorst et al., 2003).
108
N2 O (ppb)
2400
CH4 (ppb)
550
140000
Table 7.3
Component
Concentration of CH4, N2O and CO2 under the cover of the manure storage at Wageningen. CH4 flux based on measurements downwind and
indirectly derived fluxes for CO2 and N2O.
Concentration (ppm)
Source strength
-1
CH4
CO2
N2 O
#1
g.s
#1
0.15 ± 0.03
#2
0.44 ± 0.9
-4 #2
(2.1±0.6) x 10
5800
6200
3
-1
CO2 eq. kg.day
272 ± 50
38 ± 8
6±2
Emission based on plumes (uncertainty range 20%)
scaled with in-storage concentration levels.
#2
Emission calculated using CH4 emission level
7.3.3 Whole farm experiments
In March 2002 emission measurements were made at a set of 7 farms near Hardinxveld (N51°50', E4°49'). Plume measurements took place along a road at 40 to 100 m
downwind of the sources on the farms. Fig. 7.4 shows an example of the measurement
transect running east-west and the CH4 plumes plotted on this transect. Calibration of
the dispersion model took place using a release of about 5 g N2O s-1 from a gas flask.
Animal numbers at the 7 locations were available, but manure data were not available.
At the end of January however, all farms had their winter manure still in store. With a
manure production of 12 m3 manure cow-1 (obtained at farm 1 in the regional experiment) the amount of manure in store was estimated. For (biological) farm no.4 more
detailed information was available. The results are listed in Table 7.4.
m East
-1150
820
-650
route
Model CH4
CH4
Farms
350
850
Farms Hardinxveld 29/01/03
farm 3
farm 4*
farm 2
320
m North
-150
farm 1
farm 6 farm 5
-180
farm 7
-680
-1180
Figure 7.4
CH4 plume plotted with the wind direction along a transect with 7 farm
sites. The circles indicate the locations of manure storages, stables etc.
109
5000
ECN
0
Schagen
Petten
North Holland
Amsterdam
-5000
Amsterdam
Utrecht
Hardinxveld
Hardinxveld
-10000
Brabant
-15000
-20000
-25000
-5000
Figure 7.5
Alkmaar
OK
Unwilling
No meas.
d
Win
0
5000
10000
15000
20000
The combined measurement transect for the two days with
measurements in the Noord Holland region. The top left corner or the
figure is the North Sea. Measurements were carried out with a
southwest direction.
In June 2003, measurements were performed in two regions in the Netherlands (figure
7.5) in North Holland (N 52° 42', E 4°42') and Brabant (N 51° 40',E 5°27'). The farmers
were asked to provide activity data on manure management, animal numbers and food
intake. Useful plume data were obtained for 20 farms. Table 7.4 shows the set with
measured and calculated emissions. The standard deviation of a set of emission
measurements was generally in the order of 20-30% with up to 5 emission estimates
per farm. The linear regression of the measured versus FarmGHG estimates for the set
shown in Fig. 7.6 showed a slope of 0.97 with r2= 0.27. More data points are available
for a comparison with simple emission calculation with a similar slope and r2=0.6. The
activity data for all the "A" farms was based on an interview with the farmer of A4. The
emission estimated for farm A1 is a factor 4 different from the calculated level. For all
other "A" farms the agreement is much better. When rejecting the data from Farm A1
(circle in graph 5), the comparison with the simple emission calculation r2 improves further to 0.77 with a slope of 0.94. The slope of 0.94 -0.97 indicates that on average the
measured emission levels are slightly above the calculated levels.
110
-1
FARMGHG&&Simple
Simplemod.
mod.kgCH
Kg4.day
CH4 day-1
FARMGHG
120
Simple *
Meas< Model
Simple
100
B5
FARMGHG
B1
80
FARMGHG
FARMGHG
0.98x
yy == 0.98x
A1
60
B4
B1
R2 == 0.28
0.28
Simple all
Simple
all
0.96x
yy == 0.96x
40
B2
2
0.60
R == 0.60
C2
20
Meas> Model
C1
B3
Simple-A1
Simple-A1
0.94x
yy == 0.94x
2
0
R == 0.78
0.78
R
0
20
40
60
80
100
120
-1
Measured
Measured
Emission
Emission
levellevel
kg CH
kgCH
day -1
44.day
Figure 7.6
Modelled emission plotted versus measured emission per farm. Error
bars are left out. Each point has about 30% uncertainty in both measured and modelled estimate. The labels show the farm codes for the
FARMGHG model data.
In-stable emission measurements in the Netherlands were reported by Huis in 't Veld
and Monteney (2003). These data showed average emission levels (the sum of emissions from manure and from the animals) of 0.5 kgCH4animal-1day-1 for conventional
dairy systems and 1 kgCH4 animal-1day-1 for cow houses with straw bedding. In our
study the average emission for 7 conventional farms with all animals inside the stable
was 0.7 ± 0.4 kgCH4 animal-1day-1.Three farms (A4, B3, B6) use straw bedding for the
animals. They were evaluated separately and give 1.4 ± 0.2 kgCH4 animal-1day-1.
These numbers are similar to the levels reported by Huis in’t Veld and Monteney. In
their experiments emissions from the stable were measured during a longer time period
including periods when the animals were outside the stable for grazing. The plume
measurements reported here also include the emissions from manure storages outside
the stable. So the emission factor per animal should indeed be somewhat higher.
The emission factors for dairy cow housing systems reported in other recent literature
are lower. These are expressed as emission per Livestock Unit (LU=500kg weight, a
dairy cow in the Netherlands is about 520 kg). Experiments in Germany (Jungbluth et
al., 2001) and Austria (Amon et al., 2001) showed emission factors of 0.23 and 0.19
kgCH4 LU-1day-1 respectively.
Farm C1 has 600 calves on the site. For this farm the measured emission was much
larger than the FarmGHG level, primarily because the FarmGHG model does not specifically consider methane emissions from calves.
111
Table 7.4
Farm
Characteristics of farms for measurements of CH4 emissions using the
Gaussian plume method and corresponding measured and estimated
emissions using the simple emission factor approach or using the
FarmGHG model.
Young
Dairy
>1/2 y
>2y
E=274
Calves
*
No.
E=170
<1/2y
*
No.
E=48
m
3
Slurry
E=53
*
Simple Calcu- Model
FYM
E=1.3
*
*
lated
sion
m
3
m
3
emis- FARM-
&
kgCH4.d
: Measured
GHG
-1
emission
&
kgCH4.d
-1
-1
kgCH4. d
Hardinxveld March 2002
A1
65
-
-
780
#
-
59 ± 24
16 ± 4
A2
-
780
#
-
59 ± 24
55 ± 16
480
#
-
36 ± 15
47 ± 24
955
***
75 ± 29
82 ± 10
-
78 ± 31
96 ± 28
12
43 ± 17
78 ± 16
42 ± 13
65
A3
40
A4
53
A5
85
20
-
20
-
1020
#
Regional experiment North Holland (June 2003)
B1
30 (95)
(60)
0
675
B1
**
146
***
100
43
0
1500
20
118 ± 47
64 ± 13
94 ± 15
B2
0 (50)
0(50)
15
300
27
17 ± 7
29 ± 6
23 ± 5
B3
3 (20)
0
6
200
60
12 ± 5
14 ± 3
35 ± 6
B4
3 (128)
60
5
600
0
43 ± 17
54 ± 11
42 ± 5
B5
0 (150)
0 (20)
0
700
100
41 ± 16
93 ± 19
65 ± 13
B6
18
0
0
20
150
13 ± 5
Regional Experiment Brabant (June 2003)
C1
0
0
600
400
0
50 ± 20
C2
100
60
0
100
?
40 ± 16
C3
85
8
0
700
5
62 ± 25
*
Remark
Straw
Jan ‘04
Straw
23± 5
Straw
18 ± 4
42 ± 8
Calves
34 ± 7
32 ± 8
66 ± 17
-1
Emission factor in the unit E =gCH4.d . The numbers in the first column show the actual nr of animals inside the farm area on the day of
measurement. The numbers between brackets give the total herd, including the animals that were grazing and not included in the plume
**
***
#
measurements. Same farm as B1 but measured in January 2004.
Estimated from storage capacity, no exact data available. Calcu3
lated from herd size and a production of 12 m manure/animal for the whole winter period. The uncertainty estimate for calculated emission from farms A1-A6 was estimated to be about 40%. The uncertainty in the measured estimates was about 30%, For Farm A4 more
&
information was available, the uncertainty in the measured emission level was about 15 %. The uncertainty in the emission estimate
using the simple calculation was assumed to be 40%, for the FarmGHG we used 20%.
At farm B1, measurements were made both in June 2003 and in January 2004. The
measured emission level doubled compared to the summer level. This was not the
case for the FARMGHG emission level. Methane emission from slurry stores have
been found to increase exponentially with increasing slurry temperature (Khan et al.,
1997; Sommer et al., 2000), and the FarmGHG model includes this temperature effect.
There were, however, no available information on the actual temperature of the inside
slurry and on the exact times of removal of the slurry from the inside to the outside
slurry store. A better prediction of methane emissions from manure handling systems
with both inside and outside storages of slurry requires better information on storage
times and storage conditions (temperature and cover). A more detailed overview of
these experiments is available as an ECN report (Hensen et al., 2005)
7.4
Conclusions
With the fast box method emissions can be evaluated on a large number of locations
within a few hours. Silage heaps and ditches around the farms should be studied further. Plume measurements within a farm site (for example to evaluate the emissions
from manure storage) work well when the wind direction was such that plumes of indi-
112
vidual sources on the farm are separated in space. Plume measurements from manure
storages can best be performed at a distance between 30 and 100 m away from the
storage. Emission evaluation based on plume measurements for complete farms work
well at distances of 30-300 m depending on the farm size. Co-operation of the farmer is
crucial to obtain the information required for emission modelling. The modelled and
measured emission levels can be different by a factor of 3 especially for small farms.
For larger farms the measurements and modelled emission levels generally agree
within 30 %. The emission levels for a farm with calves only was underestimated using
FarmGHG, and the results show that attention should be paid to modelling the effects
of storage temperature on methane emissions from slurry storages.
Acknowledgement.
The study was part of the MIDAIR project funded by EC contract EVK2-CT-200000096. Additional support for this work was received from the ministry of VROM. Our
colleagues H. Möls, P. Fonteijn, M. Blom and T. Schrijver are acknowledged for assistance in the experiments. Field site facilities were provided by ROC Zegveld, PV Lelystad, the farms at Hardinxveld and Wageningen. Chris Flechard is acknowledged for his
support for software development. Last but not least we thank all the families at the
farms that helped with contribution of activity data.
113
Emission measurements of N2O and CH4 at Reeuwijk
114
Chapter 8 Synthesis & Conclusions
8.1
General comments
The previous chapters of this thesis discussed the quantification of emissions from diffuse sources using different measurement techniques. From a technical perspective
Chapters 3-7 have a significant overlap and measurements and data evaluation methods can be compared regardless of the evaluated sources. Using the type of diffuse
source as a starting point, the chapters deal with a variety of systems with different dynamics. The source systems relevant to this thesis are grasslands, landfills and animal
housing systems.
This final chapter starts with the technical perspective and then discusses the knowledge gained for the three different source types.
8.2
The technical perspective
8.2.1 Chamber measurements
The fast box measurement system used at farms in chapter 7 was used for a number
of years at the landfills. Making a combination of a very sophisticated sensor (the TDL)
with the relatively simple box method does not immediately make sense. However, with
this method at least a factor 4 more measurements can be done at a single day as
compared with the methods used thus far, that use either a GC or opto-acoustic instruments. Moreover, the difference in output, turned out to be more like comparing a
movie with a set of instantaneous photographs. The high time resolution shows the
variability of the increasing concentration. An example is figure 8.1 that shows ebolution (bubbles) events in a ditch when measuring over water. With slow techniques the
difference between ebolution and diffusive emission is very difficult to establish. With
the decreasing price of optical systems that allow >1Hz data acquisition, an increasing
number of teams now use this fast box measurement system which helps to improve
our understanding of the emission processes (Flechard et al 2007).
The fast box technique also helped to demonstrate the non-linearity of the concentration increase that is well documented in fundamental physics, but is hardly ever used in
environmental and agronomic science (Kroon et al., 2009). The implications of this
non-linearity reach however far beyond the technical interest. All emission studies that
still rely, or have relied on, either linear interpolation of concentration measurements in
a static chamber or on only a start and end concentration measurement, are in danger
of showing underestimated emission levels. (Kroon et al., 2009; Kutzbach et al., 2007)
115
Measurements on october 8 2002
2.5
CH4 measurements
12000
CH4 jump in mlCH4
2
10000
1.5
8000
position
1
6000
position
2
position
3
1
CH 4 jump ip in m lCH4
CH4 concentration in ppb
14000
4000
0.5
2000
0
8-10-2002
10:04
Figure 8.1
8-10-2002
10:12
8-10-2002
10:19
8-10-2002
10:26
8-10-2002
10:33
8-10-2002
10:40
8-10-2002
10:48
8-10-2002
10:55
0
8-10-2002
11:02
Ebolution events in a ditch. The box measurements on a ditch show the
upward jumps in the methane concentration whenever a bubble of
methane enters the box. At position 1&2 apparently the box is not sealing properly and the methane concentration decays after each peak. At
position 3 the concentration remains more constant and increases after
every subsequent ebolution event.
8.2.2 Micrometeorological techniques
Greenhouse gasses EC & REA
Eddy correlation and eddy accumulation measurements presented in chapter 3 were
innovative when this paper was published in 1996. At present (2011), the error analyses presented in 3.3. is considered to be too simple when compared with the correction
algorithms that are currently in use. The corrections for high frequency loss and sensor
separation (Kroon et al., 2009) were not applied to the data shown in the paper. The
good comparison between the open path and closed path sensors was “explained” by
the fact that apparently, these corrections were not really relevant with this instrumental
setup. It was also suggested that the frequency loss in open and closed path setup
might have been similar. Since the ‘90’s potential problems with damping of higher frequency information in the tubing and closed path cell were an important incentive to
further develop the open path sensors that are now widely used. With these systems,
there is the need to perform a Webb correction (Webb et al., 1980) and data loss can
occur at rain and snow events. This is one of the reasons for the comeback of improved versions of the closed cell system used in chapter 3. The main reason to use
closed path sensors is that they allow for regular calibration to “absolute standards”.
This is important when the system should not only produce flux (fluctuations around the
mean) data, but also provide simultaneous results in a network of CO2 concentration
monitoring stations.
One innovation introduced in chapter 3 is still missing in several REA designs. This is
the combined inlet with a fixed flow that allows correction for the 20-100 ms that are required to actuate the valve after an up or downwind eddy is observed. Since a mismatch in timing will always lead to a decreasing correlation between concentration and
windspeed data, this inevitably leads to a decreased flux amplitude so both a lower uptake and lower emission.
116
The REA setup was never considered as an option to measure CO2 fluxes itself, having
a superior EC measurement method available. The use of simultaneous EC-REA
measurements for CO2, however offers a correction option for other components that
can be analysed in the gas bags. The setup demonstrated in chapter 3 was used by
Vroom (1997) to evaluate DMS emissions along the Wadden Sea.
sonic data
Vertical windspeed
PC
up
down
down
N2O concentration
12 port valves
up
10 l
Tedlar
bag
10 l
Tedlar
bag
Figure 8.2 Schematic view of a relaxed eddy accumulation setup
Ammonia EC & REA
The wet chemical REA setup for NH3 in chapter 4 is completely different from the one
for CO2 in Chapter 3, since for NH3 the use of gasbags is impossible.
The main conclusion of the intercomparison of different REA systems for NH3 was that
the wet chemical systems are in principle capable of providing useful flux data, but
regular maintenance and a well-trained operator are essential to keep the systems
running. The S/N ratio is still not sufficient to enable the evaluation of flux divergence
for NH3 which was one of the aims of the study in chapter 4. This is especially relevant
for NH3 as it reacts with other atmospheric components.
Subsequent campaigns in this field of research focussed on testing optical devices that
can be used for direct eddy covariance measurements rather than relaxed eddy accumulation. Famulari et al. (2004) performed NH3 EC measurements by means of Tuneable Diode Laser Absorption Spectrometry (TDLAS). Whitehead et al. (2008) intercompared two fast optical systems (TDLAS and Quantum Cascade Laser Absorption Spectrometer, QCLAS) in EC configuration to determine NH3 fluxes associated with slurry
spreading. They tested these against the AMANDA wet chemistry gradient system and
found a systematic underestimation of the EC fluxes of 47%, which they could not correct nor explain. Recently Sintermann et al., (2011) have published EC NH3 flux measurements using a proton transfer mass spectrometer (PTRMS) that were very successful, as judged by the spectral analyses of the data. A drawback of the Swiss setup is
that a long 150°C heated inlet line had to be used.
The quality of EC measurements with optical techniques is improving every year. The
drawback is that in order to prevent degradation of the optical components, inlet filters
for aerosol are required. These cause attenuation of the high frequency components in
the EC signal. As mentioned above, a heated inlet line to the optical or mass spectrometry systems can help diminish this problem. Such a setup with a high power consumption does, however, impose severe constraints on instrument deployment. So, in
117
spite of their “operator-sensitivity” the wet chemical systems with a low power consumption and small size are still a very attractive option to measure NH3 fluxes.
8.2.3 Plume technique
The first mobile measurements were carried out with the TDL in February 1996 and the
measurements described in Chapter 3 are among the first studies worldwide with a
mobile TDL to evaluate methane emissions. The first study of this kind was reported by
Lamb (1995) and Czepiel et al. (1996). Also in 1996 the first mobile emission measurements in the Netherlands were done in September 1996 at landfills in Linne (Limburg, NL) and Geldern-Pont (Germany). That data was reported by Bergamaschi et al.
(1998). Similar measurements were done by Tregoures et al., (1999), Galle et al.,
(2001) and Czepiel et al., (2003). These experiments all used SF6 as a tracer (about
0.6 gr/sec, GWP=22800) which is not very good from an environmental perspective.
The setup presented here used N2O (3 g/sec, GWP=300) as a tracer which has a factor 10 lower net CO2 equivalent emission. The other advantage of N2O versus SF6 is
that the instrument that evaluates the methane emission is also used to evaluate the
tracer emission. Therefore, there is no difference in response characteristic.
Chapter 6 is important in this thesis because of the comparison between the Huang
model and the Gaussian model. The latter is also used in chapters 5 & 7. The two
models use the same measurements and provide results that agree within 30%. Or,
equally fair to state, provide estimates that are different by 30%. Unfortunately this is
the current state of play. In face of this uncertainty the agreement between the Gaussian plume modelling results and the inventory based estimate in chapter 6 is encouraging. Both in chapter 6 and 7 the emission level calculated with inventory emission
factors is treated as a reference level for the measurement results. Since it is the only
reference available. It is clear, that these “national factors” will by definition not be representative for individual farm sites and should be used with an uncertainty range similar to the measurement results.
There are more studies that report differences between modelled and expected emission levels (for example Michorius et al., 1997). In fact only an artificial release, that is
carefully designed to match the source distribution of the site, can offer a real emission
reference level (Flesch 2004). In fact, we should have used an independent tracer during the Braunschweig experiments (like for example N2O). For the landfill studies like
that in Chapter 5 an N2O tracer was used.
In particular for NH3, improvement of this method with both better measurement systems and better models is needed. As for the measurements themselves, the plume
measurement carried out with the wet chemical mobile sensor are somewhat outdated
with the current availability of fast optical sensors (for example the Quantum cascade
laser) that have a 1 second response time versus 30 seconds for the wet chemical device. Berkhout et al. (2009) report on measurements at manured fields that were performed in 2007-2008. The mobile plume measurements with the TDL were compared
with single point (static plume) Amanda system. The emission levels were compared
with both LIDAR measurements (Berkhout et al., 2007) and mass balance measurements (Huijsmans, 2001).
The fast response time is needed to enable mobile measurements or to allow for interpretation of stationary plume data. A detection limit below 0.1 µgNH3.m-3 is needed in
order to allow for measurements at distances over 100 m away from the animal
houses. This significantly reduces the sensitivity for dispersion modelling. Backward la-
118
grangian models are now available that will bring improvement on the modelling part of
this type of emission evaluations. (Loubet et al., 2010)
8.3
The source perspective
The previous chapters show how different techniques were used at different source
systems. In the following paragraphs lessons learned for the different source systems
we studied will be summarized.
8.3.1 Landfills
Measurements of emissions from landfills in the Netherlands are scarce. With a plausible reason that after the ‘90’s of the last century the policy in the Netherlands has been
to minimize land filling and to reduce the number of active landfills.
18
Composting
Incineration
16
Landfilling
Amount of waste (Mton)
14
12
10
8
6
4
2
Figure 8.3
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994
1993
1992
0
The net decrease of landfilled waste
Chapter 5 stated that the landfill emissions were expected to decrease towards 2010
and that 95% of the emissions would come from only 25 landfills. This happened indeed as shown in figure 8.3. So, whatever the level is, it is going down. However, as
discussed in paragraph 2.5 the total emission level from landfills still has a significant
uncertainty and the table from Maas et al. (2009) in figure 2.7 shows that with 0.8%,
the landfills are number four in the top ten of contributors to the total uncertainty in the
national GHG emission level for the Netherlands.
Landfill owners are also faced with an uncertainty when designing landfill gas extraction systems. As mentioned before in section 2.5, when using the models from different
countries for the same landfill the emission estimated obtained vary with a factor 7
(Scharff et al., 2006; Ogor, 2005). This ultimately results in large differences in emissions listed for landfill sites in the European pollutant register (E-PRTR).
For landfills, chamber measurements and for example isotope studies can provide
valuable information to decide what emission reduction measures are effective. (Bergamashi et al., 1998; Börjesson et al., 2007; Chanton et al., 2008) Those tests will
normally only be implemented on small plots on the landfill. However, the box measurements will never be able to produce useful annual average emission data for a
whole landfill. Therefore, to obtain an annual emission level we suggest that landfill
119
emission measurements for an annual emission level should be performed with remote
techniques (off site).
It is not just the chamber measurement technique that is problematic at a landfill site.
Any method that will try to estimate the landfill emission within the landfill area (mass
balance, eddy correlation, open path systems) will provide data that has a large inherent uncertainty in view of the difficult and uneven terrain on these sites. For landfills in
the Netherlands, the mass balance method with a 10m high tower on 21 sites was
used to validate the emission calculation models (Coops et al., 1995). This method has
drawbacks that can lead both to over and underestimation of the actual emission level
from landfills. This uncertainty contributes to the significant uncertainty at the national
emission level.
A fundamental problem of the MBM technique could imply that emissions from landfills
in the Netherlands have always (since the ‘90’s) been overestimated by 5 to 20 %. In
the MBM method the vertically integrated advected concentration is calculated from the
product of average concentration and windspeed. A correction is needed to account for
the turbulent transport that also takes place, which is causing a net reduction of the
emission estimate. This would not be such a problem if the error always works in one
direction, e.g. towards lower emissions. The need for this correction was already reported in Wilson (1984), Denmead, (1983) but is hardly ever used by teams doing the
bulk of MBM measurements (either ammonia emission measurements from fields or
methane emissions from landfills). An empirical correction of 15% has been suggested
(Denmead, 1995), although larger corrections would be needed in very small plots and
over very rough surface (Wilson and Shum, 1992).
For landfills there is a possible compensating error. When methane emitted from the
landfill surface is mixed up to a height that surpasses the measurement height of the
MBM tower, an underestimation of the emission occurs. Coops et al. (1995) report on
this problem and state that the side of the landfill should not be more than 150 m away
from the 10 m high tower. This implies a maximum scale of the landfill of about 7 hectares. In the same report the method is applied to landfills of >20 ha. Only with the raw
measurement data available, this effect can be quantified. The net effect of multiple ‘errors’ probably shows up as noise in the (relatively small) data set of landfill emission
estimates.
Remote measurements like the plume measurement technique circumvent the problems faced by on site measurements. But, in general the remote techniques provide
intermittent data. The mobile plume measurements discussed in chapter 3 provide data
for a single day. The temporal variability of emissions from a landfill is linked to
changes in temperature, rainfall and atmospheric pressure. Furthermore, maintenance
on a site can cause peak emission event. The resulting day-to-day variation in the
emission level does not allow the extrapolation of a single day with plume emission
data to an annual emission level. A statistical analysis shows that in order to reach a
5% uncertainty level of the annual emission, about 20 measurements carried out over a
year are required (Scharff et al., 2003). Similar strategies are already in use in the procedures to obtain NH3 emission factors for animal housing systems. There too, sets of
about 20 measurements per housing system are used. (Mosquera & Ogink, 2008).
For the Netherlands, with the bulk of the emissions originating from 25 landfills, a (preferably low cost) measurement technique would provide a significant better annual CH4
emission estimate as compared to the value used now.
In order to get to the “relatively low cost measurement technique”, innovative measurement setups that allow fence line emission evaluation are still missing. There are
120
lidar measurement setups that can perform such evaluations, but these measurement
systems are still extremely expensive and bulky. Forward looking infra read (FLIR)
analyses systems are promising; they can make videos that reveal leaks on complex
sites. The main problem for these systems is calibration or the concentration measurements to absolute levels and the translation of line integrated concentration levels
into emissions. Low cost solid-state sensors for methane would be useful for continuous monitoring. Several sensor companies have claimed that this should be straightforward, but still they did not deliver a simple sensor that actually does the job. The
only method at hand is a simple gasbag collection system, which has as main drawback, that a tracer release on the site is needed to optimize or avoid dispersion modeling in complex terrain.
8.3.2 Agricultural fields
Greenhouse gases
The paper in chapter 3 did not make statements on the flux levels that were actually
observed. However, the data provided with this EC setup was used to evaluate the net
CO2 exchange at the Zegveld site in 1993-1995. The outcome was reported in the paper by Langeveld et al., (1997), one of the first studies to document simultaneous and
co-located budget measurements for all three GHG’s. The paper showed the net GHG
emission from the oxidation of peat in the top-soil on peat meadows. A further analysis
of the CO2 data in relation to management was done by Villoria et al., (1996).
At Zegveld a net CO2 eq. emission from oxidizing peat is observed. The same oxidation effect was documented for the Cabauw site, about 30 km to the southeast of Zegveld. There the atmospheric gradient technique was used to evaluate the CO2 exchange during the episode 1993-1996. That data was used to obtain parameterisations
for the CO2 exchange (Dirks et al., 1994; Hensen et al., 1997). From 2003-2005, a configuration similar to the one in chapter 3 was used to perform eddy correlation measurements at a grassland site on clay soil at Lelystad in the Greengrass EU-project.
There too, CO2, CH4 and N2O were measured simultaneously to obtain a full GHG balance (Gilmanov et al., 2007; Sousanna et al., 2007; Jacobs et al., 2007). Flux measurements all over Europe quantitatively show what farmers already know for centuries:
grassfield are very productive. However, the measurements also show that disturbances of the soil in grassland ecosystems in general cause significant CO2 emissions.
Undisturbed grassfield might be able to store (sequester) large amounts of carbon and
therefore could play a role in a European CO2 mitigation strategy. It should be realized
that when nitrogen input is the basis for this C uptake, the N2O emissions linked to that
fertilizer use may offset the net CO2 equivalent uptake (Neftel et al., 2011, in prep).
Combined CO2, N2O and CH4 experimental experience was gained with the BSIK programme (2006-2010, Moors & Dolman 2011). In this episode, measurements were carried out again in the peat meadow area in the centre of the Netherlands. Hendriks
(2009), Kroon (2010), and Schrier (2010) provided three subsequent PhD theses covering the total greenhouse gas balance at three different ecosystems in intensively
managed, extensively managed and natural grassland ecosystems, respectively. Their
conclusions were similar to those from the studies at Zegveld and Cabauw, but now
underpinned with significantly improved data sets. Conclusions from these analyses
were that with new micromet instruments the budgets for N2O and CH4 can be significantly improved. The ditches in the peat meadow landscape showed high and probably
anthropogenic triggered CH4 emissions (Schrier et al., 2010). This is probably related
to influx of organic material after manure application at the intensively manages site.
The comparison of the three sites shows that in spite of the increased methane emis-
121
sion levels, inundation of peat meadow fields can turn these areas from a source into a
CO2 equivalent sink.
Ammonia
The ammonia exchange between grasslands and the atmosphere was subject of the
Braunschweig experiment that took place in the EU-GRAMINAE project in 2000. The
aim was to understand the dynamic of nitrogen within the grassland ecosystem in relation to management and weather conditions. The REA technique for ammonia that features in chapter 4 delivered only a small part of the total dataset acquired in this campaign. With REA providing a flux at a single height, it would, in theory, be possible to
evaluate the flux divergence. The latter is of interest for NH3, since it might react to, for
example, NH3NO3 while moving away from the soil surface. However, the scale of flux
divergence due to chemistry (e.g. 10%) is smaller than that which can currently be resolved. This also holds for advection errors due to emissions from the field itself. Advection errors from nearby farm sources may be detectable through the application of
REA measurements at several heights.
Figuur 8.1
The Braunschweig experiment just after mowing around the line of instruments. (Sutton et al. 2009)
The improved knowledge on the time pattern of the NH3 flux after cutting, harvesting
and fertilising a field helped to understand how plants use the N that is provided. These
conclusions and lessons learned are documented in the other papers of the special issue (see Sutton, 2001) of which both the chapters 4 and 6 are part.
For ammonia exchange (emission and deposition) over fields the atmospheric gradient
method is still state of the art. This can be done either with wet chemical techniques
(Trebs et al., 2004; Kruit Wichink, 2007) or with optical techniques (Kruit Wichink,
2009). Attempts are being made to switch to EC systems based on TDL, QCL techniques (Nowak et al., 2006; Whitehead et al., 2008; Norman et al., 2009) and CRD systems. Soon longer time series with these techniques will become available. For emission evaluation at manured fields, the mobile plume method was compared with LIDAR
and MBM measurements in a recent study in the Netherlands (Berkhout et al., 2007).
122
Sinterman et al. (2011). have pointed out striking differences between the NH3 emission measurements from manured fields as evaluated by different measurement methods. There are several reasons to believe that the reference method used thus far for
these emissions, leads to overestimation of the NH3 emission by almost a factor of two.
This will definitely be a subject for discussion, since about 30% of the national NH3
emission level is involved. Measurements that can show the difference between full
field manure application and test plot manure application will have to provide the answers to this question...
8.3.3 Farms
Farms site emissions feature in chapters 6 and 7, where different measurement techniques were used to evaluate emissions within the farm-yard, on surrounding ditches
and from the farm-yard as a whole. Chapter 6 has a focus on the NH3 emission, chapter 7 on the greenhouse gas emission.
The ammonia emission measurements in chapter 6 were part of the GRAMMINAE intercomparison campaign that took place at Braunschweig. The plume measurements
were also used in Loubet et al. (2009) to assess the effect of horizontal flux divergence
on the gradient and REA measurement positions that were discussed in chapter 4. The
diurnal variation of the emission from the housing system is pronounced in this paper. It
should be noted that this variation is typical for a dairy system. In pig and poultry housing systems the emission level has a smaller variation because there the temperature
level is kept relatively constant.
Chapter 7 shows the full farm emission measurements of greenhouse gasses carried
out in the MIDAIR project. They were done in combination with a questionnaire. Measurements reported by Huis in’ t Veld et al. (2003) and the measurements performed in
the Greengrass experiments at Lelystad (Sousanna et al. 2007) showed that national
inventory emission levels of 100 kg CH4 per dairy cow per year were clearly below the
observed emission levels. The CH4 emissions from ruminants depend on the diet, on
the animal weight and milk production. A re-evaluation of the calculation methodology
revealed that the increase in animal weight and changes in diet, which occurred over
the years, had not been properly accounted for in the inventory calculations.
The agreement between the FARMGHG model calculations and the full farm measurements were reasonable but, as discussed in the chapter, with significant scatter.
The high emissions from ditches adjacent to the farm were remarkable. The results obtained there are in fact supporting the results in section 8.3.2. discussing that there
might be an anthropogenic component in the greenhouse gas emission level from
ditches adjacent to intensively managed fields.
This is an emission that is not accounted for in the national inventory. The question
now surfaces why it took from 2003 to 2008 to draw the same conclusion again based
on new measurements. One reason is that the MIDAIR dataset is relatively small with a
few events at a small number of farms. This raises the discussion on representativity of
the data. The findings in the BSIK project are fortunately based on an improved dataset
in terms of temporal and spatial coverage and hopefully this issue will now be addressed with new measurement campaigns. Also the improved capabilities of EC systems enable new experiments over water surfaces. (Eugster et al. 2011) Also on the
European scale lakes and reservoirs are not yet accounted for (Cole et al., 2007; Tranvik et al., 2009), they only cover a small fraction of the European surface area, but they
could potentially be substantial local sources of methane (DelSontro et al., 2010).
123
As for the use of the plume method for full farm emission evaluation, there is a good
chance that more plume studies will be carried out in the years to come. One reason is
that there is a tendency to make dairy farm housing systems more open. Emission
quantification for these new systems using current methods, like releasing SF6 inside
the stable (Mosquera et al., 2005) is increasingly difficult. Especially in these cases, the
plume method can be an attractive option. In contrast to landfills however, the number
of farms in the Netherland is far beyond the limit where individual source strength
evaluation makes sense. The measurements will always have to be used in combination with models/parameterisations to do the upscaling to a regional or national level.
8.3.4 Final remarks
Integration of greenhouse gas and ammonia research.
In this thesis both work on emissions from farms and from fields is discussed for
greenhouse gases and NH3 in separate projects. Simultaneous co-located measurements of NH3 and the GHG’s are hard to find. Only few groups in Europe are doing this
(UK, CH) but in the Netherlands that type of data sets is very scarce. Unfortunately
funding almost always aims at either GHG or N research and teams working in these
different fields are not always successful in making joined experiments. Positive exception and hopefully the start of a trend was the EU Nitro-Europe project that did cover
both.
Systematic errors in GHG emissions
In section 8.2.1. the technical discussion on linear or non linear fitting to measurements
obtained in chamber measurements was addressed. In section 8.3.1. and 8.3.2. the
potential errors in the mass balance technique were addressed. These two issues are
very similar. In either case, corrections were formulated 20-30 years ago and these are
well understood. In the decades after that, the communities that have done the bulk of
these measurements did not use these corrections. A good reason is that the corrections are not easy to implement with the measurements available. And the assumption
is that the corrections are probably small as compared with other uncertainties. However, in both cases the problem is that the correction always works into one direction
and, importantly, is not random. Therefore, unlike other random uncertainties, the impact will not decrease with the square root of the number of measurements.
The implications of this should not be underestimated. A large part of all N2O emission
measurements and several other emission measurements based on chamber measurements reported in literature thus far, might need corrections that will increase emissions in the order of 20-40%. This can have effect all the way up to the IPCC emission
factor level. NH3 emissions from fields and CH4 emissions from landfills might both
have been overestimated by anything between 5 and 20% due to the missing backward turbulence correction for mass balance measurements that almost no one has
used. Desjardins et al. 2004 demonstrate the use of a fast response sensor for MBM
measurement. In their experiment the backward turbulence correction was found to be
5% so hopefully the correction is in this order of magnitude rather than the 20% level.
This correction factor would hold for the CH4 emissions from landfills in the Netherlands
and for 30% of the national NH3 emissions in the Netherlands.
Eddy correlation versus chambers
The use of the eddy covariance technique over fields is becoming more and more
popular with decreasing prices of the sensors required. Software for interpretation is
getting standardised making the EC setups more and more turn-key systems. For CO2
it is safe to say that chamber measurements are hardly used anymore for net emission
evaluation studies. However they are still useful for the evaluation of soil respiration.
124
For methane and N2O the tide is now turning and within the next decade the micromet
techniques will be implemented throughout Europe, taking over from the box measurements. Recent studies (Kroon et al., 2007; Hendriks et al., 2008) demonstrated the
operational use of QCL and CRD based EC studies. These studies also demonstrated
that the chamber measurements can still play a vital role in showing why the EC measurements are what they are. The annual coverage and the net exchange levels calculated with the EC estimates were significantly more accurate as compared to the
weekly to bi weekly manual chamber evaluation. On the other hand, the chambers
could be used to show which part of the fields actually provided the highest emission
levels.
Relevance of landfill emissions
The uncertainty in CH4 emissions from landfills and the use of different measurement
setups and models is underlying significant differences between national emission estimates even within the EU for these sources. This is not only a problem for GHG emission accounting, but also for emission trading and for quantification of the effect of
emission reduction measures.
Diffuse sources need multiple methods
In section 2.8 the need for different methods for different questions was raised. In general, the question sequence for diffuse sources is:
1. how big are the emissions for a source on an annual basis
2. what measures can reduce these emissions
3. are these measures effective
Figure 8.4 shows how the answer to these questions is translated into the questions as
to what method and which sensor to use. The conditions and specifications of a source
are input for the design of a measurement strategy. Models that can estimate the emission behaviour are valuable tools to decide on what method–sensor combination
should be used. Models are also often used in the interpretation, generalisation or upscaling of the measurement data.
Emission Question
Answer
Measurement strategy
Conditions
Where ?
Spatial scale ?
When ?
How often ?
Logistics?
Budget?
Method
(Box,MBM,
EC,REA,
ATM,
Plume etc)
Sensor
Output frequency
Sneitivity
Accuracy
Data
Design
Model
Interpretation
Figure 8.4
Upscale
Model
Development of a measurement strategy
125
Different questions require different methods to be used. Methods that provide the best
annual estimate results do not necessarily need to improve the understanding of the
internal processes in the source and vice versa.
Getting annual emission factors for important sources should ideally be done with
methods that cover the whole source area at all points in time. The micrometeorological techniques like the EC measurements that are used for CO2 exchange in ecosystems, are examples where this is indeed possible. And this is now also possible for
N2O and CH4. These “first option” techniques however, are not yet available for all
source types.
The second option is to either sample continuously in time, but not in space, or vice
versa. What to choose depends on the source under consideration. For example for
landfills, the spatial variability is the main problem, whereas for a manured field the
temporal variability is the biggest problem. So interval sampling at a landfill with a
method that accounts for the full spatial emission pattern is probably better than the
use of an automatic chamber system somewhere on the landfill. For manured fields,
this would be the other way round.
The third option is sub sampling both in time and space. In fact these kind of data are
underlying most emission factors for diffuse source systems at the moment. It is the
gradual technical and methodological improvement and decreasing costs per instrument that has allowed steps towards these first option methods in recent years.
This is a clear story for annual budget measurements. The “third option” is not ideal for
this, but might be the only option available. And these techniques have merits that
should not be underestimated. The techniques that provide detailed information for the
emission at a certain location and time within the source area can provide vital information of the underlying processes and determinants for the emission level at a specific
part within the total source area. This knowledge is crucial to develop further emission
reduction options.
The conclusion is that there is no such thing as the single best method to perform the
emission evaluations for these source types. Diffuse source systems cover very different spatial and temporal scales and the associated variations of the source strength
require dedicated measurement designs. The table in chapter 2 showed the uncertainty
in diffuse source system emissions that dominate for the uncertainty in emission levels
in the Netherlands. A combination of different study types with a continuous shift towards the more integrating techniques will be the best way to reduce this uncertainty to
whatever is agreed on as an “acceptable level”.
This thesis documents the slow but steady progress over almost two decades in this
field of science. In the decades to come, the key to further improvement is in new or
improved sensors that will enable new measurement strategies or further improvement
available strategies. As for the national and global scale combination of ground thruthing measurements and satellite retrievals will be a crucial step. At the individual source
scale further development of the plume methodology towards fence line monitoring
systems is a likely development. The latter implies further development of low cost lidar
systems or similar techniques. The fundamental errors for box and mass balance
method of course have to be resolved as soon as possible and correction algorithms
for historic data have to be developed.
In the end, the conclusion of this work is that indeed, more measurements are needed
to better observe and quantify trace gas emissions from diffuse sources.
126
Bibliography
Ammann, C., and Meixner, F.,2002: Stability dependence of the relaxed eddy accumulation coefficient for various scalar quantities, J. Geophys. Res., 107(D8), 4071,
doi:10.1029/2001JD000649.
Amon, B., Amon, Th., Boxberger, J.,2001: Alt, C., Emissions of NH3, N2O and CH4 from dairy
cows housed in a farmyard manure tying stall (Housing, manure storage, manure
spreading). Nutrient Cycling in Agro ecosystems 60, 103-113.
Amstel, A. van, Kroeze, C., van Eerdt, M., Dumont, M., Both, D., 2003: National emissions of
methane and nitrous oxide from agriculture in the Netherlands: current estimates and future trends. Workshop on Inventories and Projections of Greenhouse Gas and Ammonia
Emissions from Agriculture. 27-28 February 2003, European Environmental Agency,
Copenhagen.
Andreas, E.L., Hill, A.J., Gosz, J.R., Moore, D.I., Otto, W.D. and Sarma, A.D., 1998: Stability
dependence of the eddy-accumulation coefficients for momentum and scalars, Boundary-layer Meteorol., 86, 409-420.
Asman, W. A. H., 1992: Ammonia emission in Europe: updated emission and emission variations. Report 228471008, National Institute of Public Health and Environmental Protection (RIVM), Bilthoven, The Netherlands.
Aubinet, M., Grelle A., Ibrom, A. et al. 2000: Estimates of the annual net carbon and water exchange of forests: The EUROFLUX methodology. Advances in Ecological Research, 30:
113-175.
Baldocchi, D. D., 2003: Assessing the eddy covariance technique for evaluating carbon dioxide
exchange rates of ecosystems: past, present and future. Global Change Biology, 9, 114.
Bannink, A., Schijndel, M.W.van, Dijkstra, J., 2011: A model of enteric fermentation in dairy
cows to estimate methane emission for the Dutch National Inventory Report using the
IPCC Tier 3 Approach, Animal Feed Science and Technology.(in Press)
Bergamaschi, P., C. Lubina, R. Königstedt, H. Fischer, A. C. Veltkamp, and O. Zwaagstra,
13
1998: Stable isotopic signatures (δ C, δD) of methane from European landfill sites, J.
Geophys. Res., 103(D7), 8251–8265, doi:10.1029/98JD00105.
Berkhout A.J.C., Hoff G.R. van der, Bergwerff J.B., Swart D.P.J., Hensen A., Kraai A., Bleeker
A., Huijsmans J.F.M., Mosquera J., Pul W.A.J. van, 2008: Measuring ammonia emissions from manured fields, RIVM rapport 680150003
Beusen, A.H.W., Bouwman, A.F.,Heuberger P.S.C., Van Drecht G., Van Der Hoek K.W.,2008:
Bottom-up uncertainty estimates of global ammonia emissions from global agricultural
production systems. Atmospheric Environment 42, 6067– 6077.
Börjesson, G., Samuelsson J., Chanton J., 2007: Methane oxidation in Swedish landfills quantified with the stable carbon isotope technique in combination with an optical method for
emitted methane, Envron. Sci. Technol., 41, 6684‐6690.
Blaha, D., Bartlett, K., Czepiel, P., Harris, R., Crill, P., 1999: Natural and anthropogenic methane sources in new England. Atmospheric Environment 33, 243–255.
Blatter, A., Neftel, A., Dasgupta, P.K. and Simon P.K., 1993: A combined wet effluent denuder
+
and mist chamber system for deposition measurements of NH3, NH4 , HNO3 and NO3 .
th
Presented at the 6 European Symposium of Physico-Chemical behavior of atmospheric
Pollutants, Varese, Italy.
Bobbink, R., Boxman, D., Fremstad, E., Heil, G., Houdijk, A., Roelofs, J., 1992: Critical loads for
nitrogen eutrophication of terrestrial and wetland ecosystems based upon changes in
vegetation and fauna. In: Critical loads for nitrogen (Grennfelt, P. and Thörnelöf, E.,
eds.), pp. 41. Nordic Council of Ministers, Copenhagen.
127
Bobrutzki, K. von, Braban, C. F., Famulari, D., Jones, S. K., Blackall, T., Smith, T. E. L.,
Blom, M., Coe, H., Gallagher, M., Ghalaieny, M., McGillen, M. R., Percival, C. J., Whitehead, J. D., Ellis, R., Murphy, J., Mohacsi, A., Junninen, H., Pogany, A., Rantanen, S.,
Sutton, M. A., and Nemitz, E., 2009: Field inter-comparison of eleven atmospheric ammonia measurement techniques, Atmos. Meas. Tech. Discuss., 2, 1783-1836.
Bogner, J., Meadows, M., Czepiel, P., 1997: Fluxes of methane between landfills and the atmosphere: natural and engineered controls. Soil Use and Management 13, 268–277.
Bogner, J., and E. Matthews, 2003: Global methane emissions from landfills: New methodology
and annual estimates 1980–1996, Global Biogeochem. Cycles, 17(2), 1065,
doi:10.1029/2002GB001913.
Bouwman, A. F., Lee, D. S., Asman, W. A. H., Dentener, J. F., van de Hoek, K. W., Olivier, J. J.
G., 1997: A global emission inventory for ammonia. Global Biogeochemical Cycles, 11,
561-587.
Breemen, N. van, Burrough, P.A., Velthorst, E.J., van Dobben, H.F., de Wit, T., Ridder, T.B.,
Reijnders, H.F.R., 1982: Soil acidification from ammonium sulphate in forest canopy
throughfall. Nature 288, 548-550.
Brink, H. M. ten, J. P. Veefkind, A. Waijers-Ijpelaan, and J. C. van der Hage, 1996: Aerosol
light-scattering in the Netherlands, Atmos. Environ., 30, 4251-4261.
Burgerhart N., 2001: Mogelijkheden voor Koolstofopslag in Nederlandse ecosystemen. Rapport
voor Staatsbosbeheer. WUR/NCP (in Dutch).
Burkhardt J., Flechard C.R., Gresens F., Mattsson M., Jongejan, P.A.C., Erisman J.W., Weidinger T., Meszaros R., Nemitz E. and Sutton M.A., 2009: Modelling the dynamic chemical interactions of atmospheric ammonia with leaf surface wetness in a managed grassland canopy, Biogeosciences, 6, 67-83.
Businger, J.A., and Oncley, S.P., 1990: Flux Measurements with Conditional Sampling. J. Atmos. Oceanic Technol. 7(2), 349-352.
Bussink, D. W., Oenema O., 1998: Ammonia volatilization from dairy farming systems in temperate areas: a review. Nutrient Cycling in Agroecosystems, 51, 1352-2310.
CEC (2006) Regulation (EC) 166/2006 of the European Parliament and of the Council of 18
January 2006 concerning the establishment of a European Pollutant Release and Transfer Register and amending Council Directives 91/689/EEC and 96/61/EC, 4 February
2006, Brussels, Belgium
Coenen, P. W. H. G., Harmelen, A. K., Draaijers, G. P. J. and Van Grootveld, G., 2000: Emission Data for The Netherlands 1997 and Estimates for 1998, VROM, The Hague.
Cole, J. J., Prairie, Y. T., Caraco, N. F., McDowell, W. H., Tranvik, L. J., Striegl, R. G., Duarte,
C. M., Kortelainen, P., Downing, J. A., Middelburg, J. J., and Melack, J., 2007: Plumbing
the global carbon cycle: Integrating inland waters into the terrestrial carbon budget, Ecosystems, 10,171–184.
Coops, O., L. Luning, H. Oonk, and A. Weenk, 1995: Validation of landfill gas formation models,
in Proceedings Sardinia 95 Fifth International Landfill Symposium, vol. I, edited by T. H.
Christensen, R. Cossu, and R. Stegmann, pp. 635– 646, Environ. Sanit. Eng. Cent.
(CISA), Univ. of Cagliari, Sardinia, Italy.
Corré, W.J., 2002: Agricultural land use and emissions of methane and nitrous oxide in Europe.
Report 40, Plant Research International, Wageningen.
Crutzen P. J., Mosier A. R., Smith K. A., Winiwarter W., 2007: N2O release from agro-biofuel
production negates global warming reduction by replacing fossil fuels, Atmos. Chem.
Phys. Discuss., 7, 11191–11205.
Crutzen, P. J., 1970: The influence of nitrogen oxides on the atmospheric ozone content, Q. J.
Roy. Meteor. Soc., 96, 320–325.
Czepiel, P.M., Mosher, B., Crill, P.M., Harriss, R.C., 1996a: Quantifying the effect of oxidation
on landfill methane emissions. Journal of Geophysical Research 101, 16721–16729.
128
Czepiel,P. M., Mosher,B ., Harriss,C ., ShorterJ, . H., McManus.J. B., Kolb, C. E., Allewine,E.
and Lamb, B. K., 1996b: Landfill methane emissions measured by enclosure and atmospheric tracer methods, J . Geophys Res.1 01, 16711-16719.
Czepiel, P.M., Shorter, J.H., Mosher, B., Allwine, E., McManus, J.B.,Harriss, R.C., Kolb, C.E.,
Lamb, B.K., 2003: The influence of atmospheric pressure on landfill methane emissions.
Waste Management 23, 593–598.
Delon, C., Druilhet, A., Delmas, R. and Greenberg, J., 2000: Aircraft assessment of trace compound fluxes in the atmosphere with Relaxed Eddy Accumulation: Sensitivity to the conditions of selection, J. Geophys. Res., 105(D16), 20,461/20,472.
DelSontro, T., McGinnis, D. F., Sobek, S., Ostrovsky, I., and Wehrli, B., 2010: Extreme methane
emissions from a Swiss hydropower reservoir: contribution from bubbling sediments,
Environ. Sci. Technol., 44, 2419–2425, doi:10.1021/es9031369.
Demmers, T. G. M., Burgess, L. R., Short, J. L., Phillips, V. R., Clark, J. A., Wathes, C. M.,
1999: Ammonia emissions from two mechanically ventilated UK livestock buildings. Atmospheric Environment, 33,107-116.
Denmead O.T., 1983, Mircometeorological methods for measuring gaseous losses of nitrogen
in the field. In J.R. Freney and J.R. Simpson, editors, Gaseous loss of nitrogen from
plant-soil systems. Martinus Nijhoff, The Hague, The Netherlands.
Denmead, O.T., Raupach, M.R., 1993: Methods for measuring atmospheric gas transport in agricultural and forest systems. In: Agricultural Ecosystem Effects on Trace Gases and
Global Climate Change.
Denmead, O.T., 1995: Novel meteorological methods for measuring trace gas fluxes. Phil.
Trans. R. Soc. London Ser. A: Math. Phys. Sci. 351, 383–396.
Denmead, O.T., Harper, L.A., Freney, J.R., Griffith, D.W.T., Leuning, R., Sharpe, R.R., 1998: A
mass balance method for non-intrusive measurements of surface–air trace gas exchange. Atmos. Environ. 32, 3679–3688.
Denmead, O.T., Leuning, R., Griffith, D.W.T., Jamie, I.M., Esler, M.B., Harper, L.A., Freney,
J.R., 2000: Verifying inventory predictions of animal methane emissions with meteorological measurements. Boundary-Layer Meteorology96, 187–209.
Desjardins R.L., Ph.D. Dissertation, Comcll University, 1972.
Desjardins, R.L., Denmead O.T., Harperc L., McBaina M., Masse´d D., Kaharabataa S., 2004:
Evaluation of a micrometeorological mass balance method employing an open-path laser for measuring methane emissions, Atmospheric Environment 38, 6855–6866
Dirks, B.O.M. & J. Goudriaan J., 1994: Diurnal and seasonal CO2 fluxes between grassland
ecosystems and the atmospheric boundary layer in The Netherlands. Dutch National
Research Programme on Global Air Pollution and Climate Change Report 410 100 043.
Department of Theoretical Production Ecology, Wageningen Agricultural University,
Wageningen.
Döhler, H., Dämmgen, U., Berg, W., Bergschmidt, A., Brunsch, R., Eurich-Menden, B., Lüttich,
M., Osterburg, B., 2002: Adaptation of the German emission calculation methodology to
international guidelines, determination and forecasting of ammonia emissions from German agriculture, and scenarios for reducing them by 2010 (in German, summary in English), in print at Umweltbundesamt (Berlin).
Dolman, A.J., E.J. Moors & J.A. Elbers, 2002: The carbon uptake of a mid latitude pine forest
growing on sandy soil. Agric. and Forest. Met., 111 (3): 157-170.
Dolman A.J., Valentini R., Freibauer, A., 2008: The Continental-Scale Greenhouse Gas Balance
of Europe, Ecological studies, Volume 203, 2008, DOI: 10.1007/978-0-387-76570-9
129
Dore, C. J., Jones, B. M. R., Scholtens, R., Huis in'T Veld, J. W. H., Burgess, L. R., Phillips, V.
R., 2004: Measuring ammonia emission rates from livestock buildings and manure
stores - Part 2: Comparative demonstrations of three methods on the farm. Atmospheric
Environment, 38, 3017-3024.
EMEP Webdab emission data hosted by the Centre on Emission Inventories and Projections
(CEIP): http://www.ceip.at/, last access: 20 August 2009.
Emmeneger, L., Mohn, J., Sig, M., 2004: Measurement of ammonia emissions using various
techniques in a comparative tunnel study. Int. J. Environment and Pollution. 22, 326-341.
Erisman, J.W., Otjes, R., Hensen, A., Jongejan, P., van den Bulk, P., Khlystov, A., Möls, H.,
Slanina, S., 2001: Instrument development and application in studies and monitoring of
ambient ammonia. Atmos. Environ. 35, 1913-1922.
Erisman, J. W., Sutton, M. A., Galloway, J. N., Klimont, Z., and Winiwarter, W., 2008: 100 years
of ammonia synthesis: how a single patent changed the world, Nature Geosci., 1, 636–
639.
Eugster, W. and Senn, W., 1995: A Cospectral Correction Model for Measurement of Turbulent
NO2 Flux. Boundary-Layer Meteorology, 74(4): 321-340.
Eugster W., DelSontro T. and Sobek S., 2011: Eddy covariance flux measurements confirm extreme CH4 emissions from a Swiss hydropower reservoir and resolve their short-term
variability, Biogeosciences Discuss., 8, 5019–5055, 2011.
Famulari, D., Fowler, D., Hargreaves, K., Milford, C., Sutton M.A., Nemitz, E. and Weston,
K.,2004: Measuring eddy-covariance fluxes of ammonia using tunable diode laser absorption spectroscopy. Water, Air and Soil Pollution Focus 4 (6) 151-158.
Fehsenfeld, F.C., Huey, L.G., Leibrock, E., Dissly, R., Williams, E., Ryerson, T.B., Norton, R.,
Sueper, D.T., Hartsell, B., 2002: Results from an informal intercomparison of ammonia
measurement techniques. J. Geophys. Res. 107(D24), 20001JD001327.
Fiore, A. M., West J. J., Horowitz L. W., Naik V., and Schwarzkopf M. D., 2008: Characterizing
the tropospheric ozone response to methane emission controls and the benefits to climate and air quality, J. Geophys. Res., 113, D08307, doi:10.1029/2007JD009162.
Fischer, M. L. and Littlejohn, D., 2007: Ammonia at Blodgett Forest, Sierra Nevada, USA, Atmos. Chem. Phys. Discuss., 7, 14139–14169.
Flechard, C., Fowler, D., Sutton, M. A. and Cape J. N., 1999: A dynamic chemical model of bidirectional ammonia exchange between semi-natural vegetation and the atmosphere.
Quart. J. R. Met. Soc. 125, 2611-2641.
Flechard, C.R., Ambus, P., Skiba, U., Rees, R.M., Hensen, A., van Amstel, A., van den Pol-van
Dasselaar, A., Soussana, J.F., Jones MClifton-Brown, J., Raschi, A.,Horvath, L., Neftel,
A., Jocher, M., Ammann, C., Leifeld, J., Fuhrer, J., Calanca, P., Thalman, E., Pilegaard,
K., Di Marco, C., Campbell, C., Nemitz, E., Hargreaves, K.J., Levy, P.E., Ball, B.C.,
Jones, S.K., van de Bulk, W.C.M., Groot, T., Blom, M., Domingues, R., Kasper, G.,
Allard, V., Ceschia, E., Cellier, P., Laville, P., Henault, C., Bizouard, F., Abdalla, M., Williams, M., Baronti, S., Berretti, F., Grosz, B., 2007, Effects of climate and management
intensity on nitrous oxide emissions in grassland systems across Europe. Agric. Ecosyst.
Environ. 121, 135–152.
Flesch, T. K., Wilson, J. D., Harper, L. A., Crenna, B. P., 2005: Estimating gas emissions from a
farm with an inverse-dispersion technique. Atmospheric Environment, 39:4863-4874.
Flesch, T. K., Wilson, J. D., Harper, L. A., Todd, R. W., Cole N. A., 2007: Determining ammonia
emissions from a cattle feedlot with an inverse dispersion technique. Agricultural and
Forest Meteorology, 144:139-155.
Fowler D., Duyzer J.H., 1989: Micrometeorological techniques for the measurement of trace gas
exchange, Exchange of trace gases between terrestrial ecosystems and the atmosphere, Andreae M.O., Schimel D.S. eds., John Wiley and Sons, 1989, pp. 189-207.
Freibauer, A. 2003: Regionalised inventory of biogenic greenhouse gas emissions from European agriculture. European Journal of Agronomy, 19(2): 135–160.
130
Freibauer, A. and Kaltschmitt, M. 2003: Controls and models for estimating direct nitrous oxide
emissions from temperate and boreal agricultural mineral soils in Europe. Biogeochemistry, 63(1): 93–115.
Freibauer A., 2008: Designing an observation strategy for N2O. In: Dolman AJ; Valentini R
(Eds.): The continental-scale greenhouse gas balance of europe. Ecological Studies.
Vol. 203. Springer, New York [u.a.]. pp. 135-151
Galle, B., Samuelsson, J., Svensson, B.H., Börjesson, G., 2001: Measurements of methane
emissions from landfills using a time correlation tracer method based on FTIR absorption
spectroscopy. Environmental Science & Technology 35 (1): 21-25.
Galle, B., Klemedtsson, L., Griffith, D.W.T., 1994: Application of a Fourier transform IR system
for measurements of N2O fluxes using micrometeorological methods, an ultralarge
chamber system, and conventional field chambers. J. Geophys. Res. 99 (D8), 16575–
16583.
Galloway, J. N., Aber, J. D., Erisman, J. W., Seitzinger, S. P., W., H. R., Cowling, E. B., and
Cosby, B. J., 2003: The nitrogen cascade, BioScience, 53, 341–356.
Galloway, J. N., Dentener, F. J., Capone, D. G., Boyer, E. W.,Howarth, R. W., Seitzinger, S. P.,
Asner, G. P., Cleveland, C.C., Green, P. A., Holland, E. A., Karl, D. M., Michaels, A.
F.,Porter, J. H., Townsend, A. R., and Vorosmarty, C. J., 2004: Nitrogen cycles: past,
present and future, Biogeochemistry, 70, 153–226.
Galloway, J. N., Townsend, A. R., Erisman, J. W., Bekunda, M.,Cai, Z., Freney, J. R., Martinelli,
L. A., Seitzinger, S. P., and Sutton, M. A., 2008: Transformation of the Nitrogen Cycle:
Recent Trends, Questions and Potential Solutions, Science, 320, 889–892.
Galloway, J. N., 1998: The global nitrogen cycle: changes and consequences, Environ. Poll.,
102, 15–24.
Galloway, J.N., 1995: Acid deposition: perspectives in time and space. Water, Air and Soil Pollution 85, 15-24.
Gao, W., 1995: The vertical change of coefficient b used in the relaxed eddy accumulation
method for flux measurements above and within a forest canopy. Atmos. Environ.
29(17): 2339-2347.
Gash, J. H. C., 1985: A note on estimating the effect of a limited fetch on micrometeorological
evaporation measurements, Boundary Layer Meteorology, 35, 409-413.
Gilmanov, T. G., Soussana, J.-F., Allards, A. L., Ammann, C., Balzarolo, M., Barza, Z., Bernhofer, C., Campbell, C. L., Cescatti, A., Clifton-Brown, J., Dirks, B. O. M., Dore, S., Eugster, W., Fuhrer, J., Gimenco, C., Gruenwald, C., Haszpra, L., Hensen, A., Ibrom, A., Jacobs, A. F. G., Jones, M. B., Laurila, G., Lohila, A., Manca, G., Marcolla, B., Nagy, Z., Pilegaard, K., Pinter, K., Pio, C., Raschi, A., Rogiers, N., Sanz, M. J., Stefani, P., Sutton,
M., Tuba, Z., Valentini, R., Williams, M. L., and Wohlfahrt, G., 2007: Partitioning European grassland net ecosystem CO2 exchange into gross primary productivity and ecosystem respiration using light response function analysis, Agric. Ecosyst. Environ., 121,
93–120.
Ham, J.M. and Baum, J.M., 2007: Measuring Ammonia Fluxes from Cattle Feedlots using TimeAveraged Relaxed Eddy Accumulation. International Symp. on Air Quality and Waste
Management for Agriculture, 16-19 Sept 2007, Broomfield, Colorado.
Hargreaves, K. J., Fowler, D., Pitcairn, C. E. R. et al., 2001: Annual methane emission from
Finnish mires estimated from eddy covariance campaign measurements. Theoretical
and Applied Climatology 70(1-4): 203-213.
Heij, G.J., Schneider, T., eds., 1991: Acidification research in the Netherlands, pp. 3-24. Studies
in Environmental Science 46, Elsevier, Amsterdam.
Heij, G.J., Schneider, T., eds., 1995: Dutch Priority Programme on Acidification. Final report
No. 300-05, National Institute of Public Health and Environmental Protection (RIVM),
Bilthoven, The Netherlands.
131
Hendriks, D. M. D., Dolman, A. J., van der Molen, M. K., and van Huissteden, J., 2008: A compact and stable eddy covariance set-up for methane measurements using off-axis integrated cavity output spectroscopy, Atmos. Chem. Phys., 8, 431–443.
Hendriks, D. M. D., 2009: Integrated observations of greenhouse gas budgets at the ecosystem
level. Changing environment and management pratices in peat meadows, Dissertation
VU Amsterdam.
Hensen, A., Kieskamp, W.M., Vermeulen, A.T., v.d.Bulk W.C.M., Bakker, D.F., Beemsterboer,
B., Möls, J.M., Veltkamp, A.C., Wyers, G.P., 1995: Determination of the relative importance of sources and sinks of carbon dioxide, ECN-C-95-035.
Hensen, A., Vermeulen, A.T., Wyers, G.P., Zhang, Y., 1996: Eddy correlation and relaxed eddy
accumulation measurements of CO2 fluxes over grassland. Phys. Chem. Earth 21, (5-6)
383-388, 1996.
Hensen, A., 2000, Op zoek naar onbekende bronnen van broeikasgassen, ECN report ECN-C00-12 (in Dutch)
Hensen, A., Scharff, H., 2001: Methane emission estimates from landfills obtained with dynamic
plume measurements, Water,Air and Soil pollution, Kluwer, focus1: 455-464.
Hensen, A., Groot, T.T., van den Bulk, W.C.M., Blom, M.J., Flechard, C., Vermeulen, A.T.,
2005: Greenhouse gas emissions from dairy farms in the Netherlands, MIDAIR WP4 and
WP2.1 contribution report. ECN-C--05-017.
Hilhorst, M., Monteney, G.J., Mosquera, J., 2003. Monitoring GHG emissions from manure storages at organic and conventional dairy farms: development of a method for monitoring
emissions from covered slurry storages. Report from MIDAIR-project. IMAG, the Netherlands.
Hovde, D. C., Stanton, A. C., Meyers, T. P., and Matt, D. R., 1995. Methane emissions from a
landfill measured by eddy correlation using a fast response diode laser sensor. J. Atmos.
Chem. 20(2), 141–162.
Huang, C. H., A theory of dispersion in turbulent shear flow, Atmospheric Environment, 13, 453463, 1979.
Huijsmans, J. F. M., Hol, J. M. G. and Hendriks, M. M. W.,2001: Effect of application technique,
manure characteristics, weather and field conditions on ammonia volatilization from manure applied to grassland, Neth. J. Agr. Sci., 49(4), 323-342.
Huijsmans, J. F. M., Hol, J. M. G. and Vermeulen, G. D., 2003: Effect of application method,
manure characteristics, weather and field conditions on ammonia volatilization from manure applied to arable land, Atmos. Environ., 37(26), 3669-3680.
Huis in ’t Veld, J.W.H., Monteney, G.J., 2003: Methane emission from cubicle housing systems
for dairy cows. Report 2003-01. IMAG, the Netherlands.
IPCC AR3, Third Assessment Report, Climate Change, 2001.
IPCC AR4, Forth Assessment Report, Climate Change, 2007.
Jacobs J., Scharff H., Hensen A., Kraai A., Scheutz C., Samuelsson J., 2007: Testing a simple
and low cost methane emission measurement method, Sardinia 2007, Eleventh International Waste Management and Landfill Symposium.
Jacobs, C. M. J. , Jacobs, A. F. G., Bosveld F. C. , Hendriks, D. M. D., Hensen A., Kroon P. S.,
Moors E. J. , Nol L. , Schrier-Uijl A., Veenendaal, E. M., 2007: Variability of annual CO2
exchange from Dutch grasslandsBiogeosciences, 4, 803–816.
Janssens, I.A., Freibauer, A., Ciais, P., Smith, P., Nabuurs, G.-J., Folberth, G., Schlamadinger,
B., Hutjes, R.W.A., Ceulemans, R., Schulze, E.-D., Valentini, R., Dolman, A.J., 2003:
Europe's Terrestrial Biosphere Absorbs 7 to 12% of European Anthropogenic CO2
Emissions. Science 300, 1538-1542; Published online May 22, 2003;
10.1126/science.1083592.
Jarvis S.C., Pain, B.F., 1990:Ammonia volatilisation from agricultural land. Proceedings of the
Fertilizer Society 298, 35 pp. The Fertilizer Society, Peterborough.
132
Judd, M.J., Kelliher, F.M., Ulyatt, M.J., Lassey, K.R., Tate, K.R., Shelton, I.D., Harvey, M.J.,
Walker, C.F., 1999: Net methane emissions from grazing sheep. Global Change Biology,
647–657.
Jungbluth, T., Hartung, E., Brose, G., 2001: Greenhouse gas emissions from animal houses
and manure stores. Nutrient Cycling in Agro ecosystems 60, 133-145.
Khan, R.Z., Müeller, C., Sommer, S.G., 1997: Micrometeorological mass balance technique for
measuring CH4 emission from stored cattle slurry. Biol. Fertil. Soils 24, 442-444.
Kesik, M. et al. 2005: Inventories of N2O and NO emissions from European forest soils. Biogeosciences, 2(4): 353–375.
Kirchgessner, M., Windisch, W., Müller, H.L., 1995: Nutritional factors for the quantification of
methane production. In: Engelhardt, W.V., Leonhard-Marek, S., Breves, G., Giesecke,
D. (Eds.), Ruminant Physiology: Digestion, Metabolism, Growth and Reproduction. Proceedings VIII International Symposium on Ruminant Physiology, pp. 333-348.
Kohsiek, W.,1991: Infra Red H2O-CO2 sensor with fibre optics, 7th AMS symposium on Meteorological observations and instrumentation, January 13-18 1991, New Orleans, USA.
Kormann, R., Meixner, F.X., 2001: An analytical footprint model for nonneutral stratification,
Boundary Layer Meteorology, 99, 207-224.
Kroon, PS, Hensen, A, Jonker, HJJ, Zahniser, MS, Veen, WH van 't & Vermeulen, A., 2007:
Suitability of quantum cascade laser spectroscopy for CH4 and N2O eddy covariance flux
measurements. Biogeosciences, 4, 715-728.
Kroon, P.S., Hensen, A., van den Bulk, W.C.M., Jongejan, P.A.C., Vermeulen, A.T., 2008: The
importance of reducing the systematic error due to non-linearity in N2O flux measurements by static chambers. Nutr. Cycl. Agroecosyst. 82, 175–186, doi:10.1007/s10705008-9179-x.
Kroon, P. S., Hensen, A., Jonker, H. J. J., Ouwersloot, H. G., Vermeulen, A. T., and Bosveld, F.
C., 2010: Uncertainties in eddy covariance flux measurements assessed from CH4 and
N2O observations, Agr. Ecosyst. Environ., 150, 806–816..
Kroon, P. S., Schrier-Uijl A. P., Hensen A., Veenendaal E. M., Jonker H. J. J., 2010: Annual
balances of CH4 and N2O from a managed fen meadow using eddy covariance flux
measurements. Eur. J. Soil Sc. 61,2010
Kroon-van Loon, P. S., 2010: Eddy covariance observations of methane and nitrous oxide
emissions: Towards more accurate estimates from ecosystems. Dissertation TU Delft,
Juni 2010.
Kruit Wichink, R. J., van Pul, W. A. J., Otjes, R. P., Hofschreuder, P., Jacobs, A. F. G. and Holtslag A. A. M., 2007: Ammonia fluxes and derived canopy compensation points over nonfertilized agricultural grassland in The Netherlands using the new gradient ammonia*high
accuracy*monitor (GRAHAM) Atmos.Environ. 41, 1271-1287.
Wichink Kruit, R.J., Volten H., Haaima M., Swart D.P.J., Zanten M.C. van., Pul W.A.J. van,
2009: Ammonia exchange measurements over a corn field in Lelystad, the Netherlands
in 2009, RIVM Bilthoven, Report 680180002/2010
Kutzbach, L., J. Schneider, Sachs, T. E. A., 2007: CO2 flux determination by closed-chamber
methods can be seriously biased by inappropriate application of linear regression. Biogeosciences, 4, pp. 1005–1025.
Lamb, B., McManus, B., Shorter, J., Kolb, C., Mosher, B., Harriss,R., Allwine, E., Blaha, D.,
Howard, T., Guenther, A., Lott, R.,Siverson, R., Westberg, H., 1995: Development of
atmospheric tracer methods to measure methane emissions from natural gas facilities
and urban areas. Environmental Science & Technology 29,1468–1479.
Langeveld, C.A., R. Segers, B.O.M. Dirks, A. van Den Pol - van Dasselaar, G. Velthof & A.
Hensen, 1997: Emissions of CO2, CH4 and N2O from pasture on drained peat soils in the
Netherlands. European Journal of Agronomy, 7, 35-42.
133
Laurila, T., Tuovinen, J.-P., Lohila, A., Hatakka, J., Aurela, M., Thum, T., Pihlatie, M.,Rinne, J.,
and Vesala, T., 2005: Measuring methane emissions from a landfill using a cost-effective
micrometeorological method. Geophys. Research Letters 32, L19808.
Laville, P., Henault, C., Renault, P., Cellier, P., Oriol, A., Devis, X., Flura, D., and Germon, J. C.,
1997: Field comparison of nitrous oxide emission measurements using micrometeorological and chamber methods, Agronomie, 17, 375–388.
Lee, X., Massman W., Law, B. (Eds.) 2004: Handbook of Micrometeorology. A Guide for Surface Flux Measurement and Analysis. Series: Atmospheric and Oceanographic Sciences
Library, Vol. 29, XIV, 250 p.
Lin, J. S., Hildemann, L. M., 1997: A Generalized mathematical scheme to analytically solve the
atmospheric diffusion equation with dry deposition. Atmospheric Environment, 31, 59-71.
Loubet, B., Milford, C., Sutton, M.A., Cellier, P., 2001: Investigation of the interaction between
sources and sinks of atmospheric ammonia in an upland landscape using a simplified
dispersion-exchange model, Journal of Geophysical Research., 106(D20), 24183-24196.
Loubet, B. and Cellier,P.,2002: Experimental assessment of atmospheric ammonia dispersion
and short-range dry deposition in a maize canopy. Water Air and Soil Pollution, Vol
1(5/6), 157-166.
Loubet, B., Cellier, P., Milford, C. and Sutton, M.A., 2006: A coupled dispersion and exchange
model for short-range dry deposition of atmospheric ammonia. Quart. J. Royal. Meteor.
Soc. 132, 1733-1763.
Loubet, B., Milford, C., Hensen, A., Daemmgen, U., Erisman, J.-W., Cellier, P., and Sutton, M.
A., 2009: Advection of NH3 over a pasture field and its effect on gradient flux measurements, Biogeosciences, 6, 1295–1309.
Loubet, B., Génermont, S., Ferrara, R., Bedos, C., Decuq, C., Personne, E., Fanucci, O., Durand, B., Rana, G. and Cellier, P., 2010. An inverse model to estimate ammonia emissions from fields. European Journal of Soil Science, 61: 793-805.
Maas, C.W.M. van der, Coenen, P.W.H.G., Zijlema, P.J., Brandes, L.J., Baas, K., van den
Berghe, G., van den Born, G.J., Guis, B., Geilenkirchen, G., te Molder, R., Nijdam, D.S.,
Olivier, J.G.J., Peek, C.J., van Schijndel, M.W. en van der Sluis, S.M., 2009: Greenhouse Gas Emissions in the Netherlands 1990-2007. Publicatienummer 50008012, Planbureau voor de Leefomgeving, Bilthoven/Den Haag.
Magnani F, Mencuccini M, Borghetti M, Berbigier P, Berninger F, Delzon S, Grelle A, Hari P,
Jarvis PG, Kolari P, Kowalski AS, Lankreijer H, Law BE, Lindroth A, Loustau D, Manca
G, Moncrieff JB, Rayment M, Tedeschi V, Valentini R, Grace J., 2007: The human footprint in the carbon cycle of temperate and boreal forests. Nature 447: 848-850
Mészáros, R. , Horváth, L. , Weidinger, T. , Neftel, A. , Nemitz, E. , Dämmgen, U. , Cellier, P. ,
Loubet, B., 2009: Measurement and modelling ozone fluxes over a cut and fertilized
grassland, Biogeosciences Discuss., 6, 1069-1089.
Meyers, T. P., Luke, W. T., Meisinger, J. J., 2006: Fluxes of ammonia and sulfate over maize
using relaxed eddy accumulation. Agric. For. Meteor. 136, 203-213.
Michorius, J. A. T., Hartog, K. D., Scholtens, R., Harssema, H., 1997: Measuring ammonia
emissions from building complexes using the flux frame method and the Gaussian plume
model: a feasibility study, Translation - Silsoe Research Institute.
Milford C., Theobald M.R.., Nemitz E., Hargreaves K.J., Horvath L., Raso J., Dämmgen U.,
Neftel A., Jones, S.K., Hensen A., Loubet B., Sutton M.A.., 2009: Ammonia fluxes in relation to cutting and fertilization of an intensively managed grassland derived from an inter-comparison of gradient measurements. Biogeosciences, 6, 819-834.
Milieubalans 2009: report 500051015, Planbureau voor de leefomgeviong Bilthoven,The
Netherlands, 2009. (In Dutch)
Milne, R., Beverland, I.J., Hargreaves, K., Moncrieff, J.B., 1999: Variation of the beta coefficient
in the relaxed eddy accumulation method. Boundary-Layer Meteorol. 93(2), 211-225.
134
Milne, R., Mennim, A., Hargreaves K., 2001: The value of the coefficient in the relaxed eddy
accumulation method in terms of fourth order moments. Boundary-Layer Meteorol. Volume 101, Number 3, pp. 359-373(15).
Moncrieff, J.B., Massheder, J.M., de Bruin, H., Elbers, J., Friborg, T., Heusinkveld, B., Kabat, P.,
Scott, S., Sogaard, H., Verhoef, A., 1997: A system to measure surface fluxes of momentum, sensible heat, water vapour and carbon dioxide. Journal of Hydrology 188–
189, 589–611.
Monna, W.A.A., v.d.Vliet, J.G., 1987: Facilities for Research and weather observations on the
213 m tower at Cabauw and remote locations, scientific report WR-nr87-5, KNMI.
Monteith J. L., Unsworth M. H., 1990: Principles of Environmental Physics, 2nd ed., 291 pp., Arnold, New York.
Moore C.J., 1986: Frequency response corrections for eddy correlation systems, Boundary
Layer Meteorology, 37.
Moors E.J., J.A. Elbers, C.M.J. Jacobs, W.W.P. Jans, B. Kruijt, S. Supit, et al., 2009: Variability
in carbon exchange of European croplands. Agriculture, Ecosystem & Environment doi:
10.1016/j.agee.2010.04.013.
Mosier, A., Kroeze, C., Nevison, C., Oenema, O., Seitzinger, S., van Cleemput, O., 1998: Closing the global N2O budget: nitrous oxide emissions through the agricultural nitrogen cycle. Nutrient Cycling in Agro ecosystems 52, 225-248.
Mosquera J., Hensen,A., van den Bulk,W. C. M., Vermeulen A. T., Erisman J. W., 2001: Long
Term NH3 Flux Measurements above Grasslands in The Netherlands Water, Air, and
Soil Pollution: Focus 1: 203–212.
Mosquera, J., Monteny G. J., Erisman J. W., 2005: Overview and assessment of techniques to
measure ammonia emissions from animal houses: the case of the Netherlands. Environmental Pollution, 135, 381–388.
Mosquera, J., Ogink, N.W.M., 2008: Analyse ammoniakemissieniveaus van praktijkbedrijven in de
varkenshouderij (1990-2003). Lelystad, Rapport / Animal Sciences Group 135.( in Dutch)
Myles, L., Meyers, T.P., Robinson, L, 2007.: Relaxed eddy accumulation measurements of ammonia, nitric acid, sulfur dioxide and particulate sulfate dry deposition near Tampa, FL,
USA., Environ. Res. Lett. 2 034004 (8pp) doi: 10.1088/1748-9326/2/3/034004.
Neftel, A., Blatter, A., Gut, A., Hoegger, D., Meixner, F. X., Ammann C., and Nathaus F. J.,1998:
NH3 soil and soil surface gas measurements in a triticale wheat field. Atmos. Environ.
(Ammonia Special Issue) 32 (3), 499-506.
Neftel, A., Blatter, A., Otjes R., Erisman, J. W. and Hansen, A.,1999: State of the art REA NH3
th
flux measurements. In: Proc. 10 Nitrogen Workshop, Copenhagen, August 1999. II.49.
Royal Veterinary and Agricultural University, Copenhagen.
Neftel, A., Ammann, C., Fischer, C., Spirig, C., Conen, F., Emmenegger, L., Tuzson, B., Wahlen, S., 2009: N2O exchange over managed grassland: Application of a quantum cascade laser spectrometer for micrometeorological flux measurements, Agr. Forest Meteorol., in press, doi:10.1016/j.agrformet.2009.07.013.
Nemitz, E., Flynn, M., Williams, P.I., Milford, C., Theobald, M.R., Blatter, A., Gallagher, M.W.,
and Sutton, M.A., 2001: A Relaxed Eddy Accumulation system for the automated measurement of atmospheric ammonia fluxes. Water Air Soil Poll. Focus 1(5-6), 189-202.
Nemitz, E. , Dorsey, J. R., Flynn, M. J. , Gallagher, M. W. , Hensen, A., Erisman, J.-W., Owen,
S. M., Dämmgen, U., and Sutton M. A., 2009a: Aerosol fluxes and particle growth above
managed grassland, Biogeosciences Disc Atmos. Chem. Phys., 9, 2635–2645.
Nemitz, E., Hargreaves, K. J., Neftel, A., Loubet, B., Cellier, P., Dorsey, . R., Flynn, M., Hensen,
A., Weidinger, T., Meszaros, R., Horvath, L., Dämmgen, U., Frühauf, C., Löpmeier, F. J.,
Gallagher, M. W., Sutton M. A.,2009b: Intercomparison and assessment of turbulent and
physiological exchange parameters of grassland, Biogeosciences, 6, 1445-1466.
135
Norman M., Spirig C., Wolff V., Trebs I., Flechard C., Wisthaler A., Schnitzhofer R., Hansel A.,
Neftel A. , 2009, Intercomparison of ammonia measurement techniques at an intensively
managed grassland site (Oensingen, Switzerland) Atmos. Chem. Phys., 9, 2635–2645,
2009
Nowak, J. B., Huey, L. G., Russell, A. G., Tian, D., Neuman, J. A., Orsini, D., Sjostedt, S. J.,
Sullivan, A. P., Tanner, D. J., Weber, R. J., Nenes, A., Edgerton, E., and Fehsenfeld, F.
C., 2006: Analysis of urban gas phase ammonia measurements from the 2002 Atlanta
Aerosol Nucleation and Real-Time Characterization Experiment (ANARChE), J. Geophys. Res., 111, D17308, doi:10.1029/2006JD007113.
Oenema, O., Velthof, G., Kuikman, P., 2001: Technical and policy aspects of strategies to decrease greenhouse gas emissions from agriculture. Nutrient Cycling in Agro ecosystems
60, 301-315.
th
Ogor, Y., 2005: Comparison of landfill methane emission models: a case study, 10 International waste management and landfill symposium, Sardinia.
Olesen, J.E., Schelde, K., Weiske, A., Weisberg, M.R., Asman, W.A.H., Durhuus, J., 2005:
Modelling greenhouse gas emissions from European conventional and organic dairy
farms. Agric. Ecosyst. Environ.
Oonk H., Boom T., 1995: Landfill gas formation, recovery and emission, TNO-report 95-203,
TNO, Apeldoorn, the Netherlands.
Oonk, J., Scheepers, M., Takke, J.W., 1993: Overzicht Stortgasprojecten in Nederland 1983–
1991. Adviescentrum Stortgas. Apeldoorn, Netherlands.
Pain, B.F., van der Weerden, T.J., Chambers, B.J., Phillips, V.R., Jarvis, S.C., 1998: A new inventory for ammonia emissions from UK agriculture. Atmospheric Environment 32, 309313.
Pasquill, F. 1974: Atmospheric Diffusion, 2nd ed., J. Wiley & Sons, New York.
Pattey, E., Desjardins, R.L., and Rochette, P., 1993: Accuracy of the relaxed eddy-accumulation
technique, evaluated using CO2 flux measurements. Boundary-Layer Meteorol. 66, 341355..
Pattey, E., Strachan, I.B., Desjardins, R.L., Edwards, G.C., Dow D., MacPherson, J.I., 2006:
Application of a tunable diode laser to the measurement of CH4 and N2O fluxes from
field to landscape scale using several micrometeorological techniques, Agricultural and
Forest Meteorology 136 (2006) 222–236.
Philip, J. R., 1959: The theory of local advection: 1, J. Meteorol., 16, 535-547.
Phillips, V. R., Lee, D. S., Scholtens, R., Garland, J. A., and Sneath, R. W., 2001: A review of
methods for measuring emission rates of ammonia from livestock buildings and slurry or
manure stores, part 2: monitoring flux rates, concentrations and airflow rates. Journal of
Agricultural Engineering Research, 78, 1-14.
Pihlatie, M., Rinne, J., Ambus, P., Pilegaard, K., Dorsey, J. R., Rannik, U¨ ., Markkanen, T.,
Launiainen, S., and Vesala, T., 2005: Nitrous oxide emissions from a beech forest floor
measured by eddy covariance and soil enclosure techniques, Biogeosciences, 2, 377–
387.
Rapsomanikis, S., Wake, M.T., Kitto, A.-M. and Harrison, R.M. 1998: Analysis of atmospheric
ammonia and particulate ammonium by a sensitive fluorescence method. Envir. Sci.
Technol. 22, pp. 948–952.
Raupach, M. R., 1989: Stand overstorey processes, Philos. Trans. R. Soc. London, Ser. B, 324,
175-190.
Savanne D., Arnaud A., Beneito A., Berne P., Burkhalter R., Cellier P., Gonze M.A., Laville P.,
Levy F., Milward R., Pokryszka Z., Sabroux J.C., Tauziede C., Tregoures A., 1997:
Comparison of different methods for measuring landfill methane emissions. Sardinia ’97
Sixth International Landfill Symposium. Vol. IV, pp. 81‐85.
136
Schrier-Uijl, A.R., 2010: “Flushing meadows. The influence of management alternatives on the
greenhouse gas balance of fen meadow areas”, Dissertation, Wageningen University,
Netherlands.
Schulze, E. D., Luyssaert, S., Ciais, P., Freibauer, A., Janssens, I. A., Soussana, J. F., Smith,
P., Grace, J., Levin, I., Thiruchittampalam, B., Heimann, M., Dolman, A. J., Valentini, R.,
Bousquet, P., Peylin, P., Peters, W., R¨odenbeck, C., Etiope, G., Vuichard, N., Wattenbach, M., Nabuurs, G. J., Poussi, Z., Nieschulze, J., Gash, J. H., and the CarboEurope
Team, 2009: Importance of methane and nitrous oxide for Europe’s terrestrial greenhouse-gas balance, Nat. Geosci., 2, 842–850 + Corrigendum, doi:10.1038/ngeo686.
Scharff, H., Oonk, J., Vroon, R., Hensen, A., de Visscher, A., Boeckx, P., 2003: A comparison of
measurement methods to determine landfill methane emissions. NOVEM Programme
Reduction of Other Greenhouse Gases (ROB), projectnumber 0373-01-01-04-001,
Utrecht, Netherlands, http://www.robklimaat.nl/docs/ 3730040010.pdf
Scharff, H., Jacobs, J., 2006: Applying guidance for methane emission estimation for landfills,Waste Management, Vol. 26, Issue 4, pp 417-429.
Scharff H., Hensen A., 2009: Further development of a cheap and simple methane emission
measurement method, Proceedings Sardinia 2009, Twelfth International Waste Management and Landfill Symposium, S. Margherita di Pula, Cagliari, Italy; 5 ‐9 October
2009.
Scheepers M.J.J.,van Zanten,B., 1994: Handleiding Stortgaswinning Adviescentrum Stortgas,
Utrecht, The Netherlands (in Dutch).
Scholtens, R., Dore, C. J., Jones, B. M. R., Lee, D. S., Phillips, V. R., 2004: Measuring ammonia emission rates from livestock buildings and manure stores - Part 1: Development and
validation of external tracer ratio, internal tracer ratio and passive flux sampling methods.
Atmospheric Environment, 38, 3003-3015.
Shaw, W. J., Spicer, C.W, and Kenny, D.V., 1998: Eddy correlation fluxes of trace gases using
a tandem mass spectrometer, Atmos. Environ., 32, 2887-2898.
Sintermann J., Spirig C., Jordan A., Kuhn U., Ammann C., Neftel A.,2011: Eddy covariance flux
measurements of ammonia by high temperature chemical ionisation mass spectrometry,
Atmos. Meas. Tech., 4, 599–616.
Sinterman BGD 2011
Smeets C. J. P. P., Holzinger R., Vigano I., Goldstein A. H., and Röckmann T., 2009: Eddy covariance methane measurements at a Ponderosa pine plantation in California Atmos.
Chem. Phys., 9, 8365–8375.
Smith, F. B., 1957: The diffusion of smoke from a continuous elevated point source into a turbulent atmosphere, Journal of Fluid Mechanics, 2, 49-76.
Smith, K.A., Clayton H., Arah J. R. M., Christensen S., Ambus P., Fowler D., Hargreaves K. J.,
Skiba U., Harris G. W., Wienhold F. G., Klemedtsson L., Galle B., 1994. Micrometeorological and chamber methods for measurement of nitrous oxide fluxes between soils and
the atmosphere: Overview and conclusions. Journal of Geophysical Research, 99(D8):
16541–16548.
Smith, K. A. and Dobbie, K. E., 2001: The impact of sampling frequency and sampling times on
chamber based measurements of N2O emissions from fertilized soils. Global Change
Biology, 7: 933–945. doi: 10.1046/j.1354-1013.2001.00450.x
Sneath, W., Beline, F., Hilhorst, A., Peu, P., 2005: Monitoring GHG from manure stores on organic and conventional dairy farms. Agric. Ecosyst. Environ.
Sommer, S. G., McGinn, S. M., Flesch, T. K., 2005: Simple use of the backwards Lagrangian
stochastic dispersion technique for measuring ammonia emission from small field-plots.
European Journal of Agronomy, 23:1-7.
Sommer, S.G., Petersen, S.O., Møller, H.B., 2004: Algorithms for calculating methane and nitrous oxide emissions from manure management. Nutrient Cycling in Agro ecosystems
60, 133-145.
137
Sommer, S.G., Petersen, S.O., Søgaard, H.T., 2000: Greenhouse gas emission from stored
livestock slurry. J. Environ. Qual. 29, 744-751.
Sørensen, L. L, Hertel, O., Ambelas Skjøth, C., Lund, M., Pedersen, B., 2003: Fluxes of ammonia in the coastal marine boundary layer. Atmos. Environ. 37, Suppl. 1, 167-177.
Sørensen, L.L., Granby, K., Nielsen, H. and Asman, W.A.H., 1994: Diffusion scrubber technique
used for measurements of atmospheric ammonia. Atmos. Environ., 28, 3637-3645.
Soussana, J.-F., Allard, V., Pilegaard, K., Ambus, P., Ammann, C., Campbell, C., Ceschia, E.,
Clifton-Brown, J., Dominques, R., Flechard, C., Fuhrer, J., Hensen, A., Horvath, L.,
Jones, M., Kasper, G., Martin, C., Nagy, Z., Neftel, A., Raschi, A., Rees, R. M., Skiba,
U., Manca, G., Sutton, M., Tuba, Z., and Valentini, R., 2007: A full accounting of the
greenhouse gas (CO2, N2O,CH4) budget of nine European grassland sites, Agric. Ecosyst. Environ., 121, 121–134
Spokas, K., Graff, C., Morcet, M., Aran, C., 2003: Implications of the spatial variability of landfill
emission rates on geospatial analyses. Waste Management 23, 599–607
Stull, R. B., 1988: An introduction to Boundary Layer Meteorology. Kluwer Academic Publishers.
Sutton M.A., Milford C., Nemitz E., Theobald M.R., Hill P.W., Fowler D., Schjoerring J.K., Mattsson M.E., Nielsen K.H., Husted S., Erisman J.W., Otjes R., Hensen A., Mosquera J.,
Cellier P., Loubet B., David M., Genermont S., Neftel A., Blatter A., Herrmann B., Jones
S.K., Horvath L., Führer E., Mantzanas K., Koukoura Z., Gallagher M., Williams P., Flynn
M. and Riedo M., 2001: Biosphere-atmosphere interactions of ammonia with grasslands:
experimental strategy and results from a new European initiative. Plant and Soil 228
(1):131-145.
Sutton, M. A., Nemitz, E., Theobald, M. R., Milford, C., Dorsey, J. R., Gallagher, M. W., Hensen,
A., Jongejan, P. A. C., Erisman, J. W., Mattsson, M. E., Schjoerring, J. K., Cellier, P.,
Loubet, B., Roche, R., Neftel, A., Hermann, B., Jones, S., Lehman, B. E., Horvath, L.,
Weidinger, T., Rajkai, K., Burkhardt, J., Lopmeier, F. J., and Daemmgen, U., 2009a: Dynamics of ammonia exchange with cut grassland: strategy and implementation of the
GRAMINAE Integrated Experiment, Biogeosciences, 6, 309-331.
Sutton, M. A., Nemitz, E., Milford, C., Erisman, J. W., Hensen, A., Cellier, P., David, M., Loubet,
B., Personne, E., Schjoerring, J. K., Mattsson, M. E., Dorsey, J., Gallagher, M., Horvath,
L., Weidinger, T., D¨ammgen, U., Neftel, A., Herrmann, B., Lehman, B., and Burkhardt,
J., 2009b: Dynamics of ammonia exchange with cut grassland: Synthesis of results and
conclusions, Biogeosciences, 6, 1121-1184.
Sutton, O. G., 1932: A theory of eddy diffusion in the atmosphere. Proc. R. Soc .A, 135: 143165.
Thomson, D. J. 1987: Criteria for the selection of stochastic models of particle trajectories in
turbulent flows, Journal of Fluid Mechanics, 180, 529-556.
Tranvik, L. J., Downing, J. A., Cotner, J. B., Loiselle, S. A., Striegl, R. G., Ballatore, T. J., Dillon,
P., Finlay, K., Fortino, K., Knoll, L. B., Kortelainen, P. L., Kuster, T., Larsen, S., Laurion,
I., Leech, D. M., McCallister, S. L., McKnight, D. M., Melack, J. M., Overholt, E., Porter,
J. A., Prairie, Y., Renwick, W. H., Roland, F., Sherman, B. S., Schindler, D.W., Sobek,
S., Tremblay, A., Vanni, M. J., Verschoor, A. M., van Wachenfeldt, E., and Weyhenmeyer, G. A., 2009: Lakes and reservoirs as regulators of carbon cycling and climate, Limnol. Oceanogr., 54, 2298–2314.
Trebs, I., Meixner, F., Slanina, J., Otjes, R., Jongejan, P., & Andreae, M., 2004: Real-time
measurements of ammonia, acidic trace gases and water-soluble inorganic aerosol species at a rural site in the Amazon Basin. Atmospheric Chemistry and Physics, 4(1), 967987.
Trégourès A., Beneito A.,Berne P., Gonze M.A., Sabroux J.C., Savanne D., Pokryszka Z., Tauziède C., Cellier P., Laville P., Milward R., Arnaud A., Levy F., Burkhalter R.,1999: Comparison of seven methods for measuring methane flux at a municipal solid waste landfill
site, Waste Management & Research, Vol. 17, No. 6, 453‐458
138
Veefkind, J. P., J. C. H. van der Hage, and H. M. ten Brink, 1996: Nephelometer derived and
directly measured aerosol optical depth of the atmospheric boundary layer. Atmos. Res.,
41, 217–228.
Veen, W.A.G., 2000: Veevoedermaatregelen ter vermindering van methaanproductie door herkauwers, Schothorst Feed Research, Lelystad (in Dutch).
Velthof G. L.,, Jarvis S. C., Stein A., Allend A. G., Oenema O., 1996: Spatial variability of nitrous
oxide fluxes in mown and grazed grasslands on a poorly drained clay soil Soil Biology
and Biochemistry, Volume 28, Issue 9, September 1996, Pages 1215-1225
Verma, S.B., Ullman, F.G., Billesbach, D. et al., 1992: Eddy correlation measurements of methane flux in a northern peatland ecosystem. Boundary-Layer meteorology 58: 289-304.
Villoria, D, Hensen A., Wyers G.P., 1996: The effect of Agricultural Management on the CO z
exchange between atmosphere and grassland: ECN report. ECN-C-96-071.
Vitousek, P. M., Aber, J. D., Howarth, R. W., Likens, G. E., Matson, P. A., Schindler, D. W.,
Schlesinger, W. H., Tilman. D. G., 1997: Human alteration of the global nitrogen cycle:
sources and consequences. Ecol. Appl. 7(3):737-750
Vroom, C. 1997: Direct flux measurement of DMS and other volatile sulphur gases from marine
systems, RUG, Groningen
Wallace, J. M. and Hobbes, P. V. 2006: Atmospheric Science – An Introductory Survey. 2nd
Edition, Elsevier Publishers.
Wamelink et al., 2008: Effects of nitrogen deposition on biodiversity and carbon sequestration,
Forest Ecol. Manage.
Webb, E. K., G. I. Pearman, R. Leuning, 1980: Correction of flux measurement for density effects due to heat and water vapour transfer. Q. J. Roy. Meteor.Soc., 106, pp. 85–100.
Welch, D. C., Colls, J. J., Demmers, T. G. M., Wathes, C. M. A., 2005a: methodology for the
measurement of distributed agricultural sources of ammonia outdoors--Part 1: validation
in a controlled environment. Atmospheric Environment, 39, 663-672.
Welch, D. C., Colls, J. J., Demmers, T. G. M., Wathes, C. M., 2005b: A methodology for the
measurement of distributed agricultural sources of ammonia outdoors - Part 2: Field
validation and farm measurements. Atmospheric Environment, 39, 673-684,.
Whitehead, J.D., Twigg, M., Famulari, D., Nemitz, E., Sutton, M.A., Gallagher, M.W., and
Fowler, D., 2008: Evaluation of Laser Absorption Spectroscopic Techniques for Eddy
Covariance Flux Measurements of Ammonia, Environ. Sci. Technol., 42 (6), 2041–2046.
Wienhold, F. G., Frahm, H., and Harris, G. W. , 1994: Measurements of N2O fluxes from fertilized grassland using a fast response tunable diode laser spectrometer, J. Geophys.
Res., 99(D8), 16557– 16567.
Wilson, J.D., Shum,W.K.N., 1992: A re-examination of the integrated horizontal flux method for
estimating volatilization from circular plots. Agric. For. Meteorol. 57, 281–295.
Wilson, J. D., Thurtell, G. W., Kidd, G. E., Beauchamp, E. G., 1994: Estimation of the rate of
gaseous mass transfer from a surface source plot to the atmosphere, Atmospheric Environment, 16, 1861-1867.
Winegardner, D. L., 1995: An introduction to soils for environmental professionals. Lewis publishers, Florida.
Wrage N, Velthof GL, Beusichem ML van, Oenema O., 2001: Role of nitrifier denitrification in
the production of nitrous oxide. Soil Biology and Biochemistry 33: 1723-1732.
Wyers, G.P., Otjes, R.P. and Slanina, J., A., 1993: continuous-flow denuder for the measurement of ambient concentrations and surface-exchange fluxes of ammonia. Atmospheric
Environment, 27A(13), 2085-2090, 1993.
Yeh, G. T., and C. H. Huang, 1975: Three-dimensional air pollutant modeling in the lower atmosphere. Boundary Layer Meteorology., 9, 381-390.
139
Zahniser, M.S., Nelson, M.S., McManus, J.B., Kebabian, P.L., 1995: Measurement of Trace
Gas Fluxes Using Tunable Diode Laser Spectroscopy, Phil. Trans. Roy. Soc. Lond. A.,
351, 371-382.
Zhang, G., Strom, J. S., Li, B., Rom, H. B., Morsing, S., Dahl, P. and Wang, C., 2005: Emission
of ammonia and other contaminant gases from naturally ventilated dairy cattle buildings.
Biosystems Engineering, 92, 355–364.
Zhu T., Pattey, E. and Desjardins, R.L., 2000: Relaxed eddy-accumulation for measuring ammonia volatilization. Environ. Sci. Technol. 34(1): 199-203.
Zhu, T., Wang, D., Desjardins, R.L., and Macpherson, J.I., 1999: Aircraft-based volatile organic
compounds flux measurements with relaxed eddy accumulation, Atmos. Environ.
33,1969-1979.
140
Acknowledgements
Doing measurements is great. If you liked to play with LEGO-technic when you were a
child or when you fancy a bit of computer programming, try this: playing around with
state of the art equipment amidst geese and cows on a peatland site on a nice summer’s day. It probably takes a mental deficit to enjoy trimming a laser spectrometer to
single mode operation and to get exited of a green scattering line appearing on the PC
screen. On the other hand, there are addictions with more negative social and economic impacts. This type of work is rewarding because it aims to better understand
how mother nature deals with what we humans impose on her. Furthermore we like to
get a grip on what we can do to ease that burden and avoid potential negative consequences. Having been able to do this work now for two decades is valuable. Having
been able to do this work in a team of nice colleagues who share my ‘mental deficits’ to
a large extend, is priceless.
In chronological order I have to start to thank my mentor at Utrecht University, Jeroen
van der Hage, who learned and showed me the fun of doing experimental work both in
the lab and on the Atlantic ocean. With him and with my “grandpa Hensen” I share the
hamster habit of preserving (and if necessary retrieving from a container) all kind of
seemingly useless things that are key tools for an experimentalist.
Sjaak Slanina (), Pim Kieskamp and Kees van der Klein hired me at ECN and I thank
them for giving me the opportunity to work in the dunes. Sjaak once promised me that if
he would catch me holding a screwdriver he would chop my hand off. Fortunately that
never happened in spite of all the tools I used. With Pim Kieskamp, Pim van der Bulk
and Alex Vermeulen we started installation of the monitoring system at the Cabauw
tower. In 1992 we made a sketch of a network of tall towers to monitor greenhouse
gasses. Now, in 2011, the ICOS NL proposal (co submitted by Alex), asks funding for
that same idea. Pim and Alex have both contributed significantly to this booklet and I
thank them for all the nice work we did together. Paul Wyers, Jan Willem Erisman and
Gerard de Groot were my supervisors in the subsequent years. With Paul I was introduced to ammonia experiments and to some aerosol work. I really liked the MEMORA
project in which Gerard Kos and Harry ten Brink provided me with a re-introduction to
the aerosol work I did with Jeroen.
With Jan Willem I had lots of nitrogen-fun, ranging from ammonia plume measurements to designing energy-friendly roads, to making the Nitrogenius game. The latter
still provides me with an annual highlight: giving a lecture on Nitrogenius and evaluation of the games played at Wageningen University, with Carolien Kroeze. Jan Willem
is still leading the integrated Nitrogen work, although (unfortunately) at a somewhat larger distance. In this work Albert Bleeker fortunately works just around the corner and
we have spend many hours together designing working on the N-visualization tool and
on all kind of other Nitrogen projects. In the substantial carbon and nitrogen related
projects like LIFE, GRAMMINA, GREENGRASS and NITROEUROPE I also worked
closely with the other team members. Piet Jongejan, Han Möls, Mark Blom, Peter Fonteijn, Hans van ’t Veen and Arnoud Frumau also provided valuable input for this thesis.
Lots of experiments were also successful thanks to Theo Schrijver who provided all
sorts of the handy tools and adaptations for our instruments. Gerard de Groot supervised us over the last years and I really appreciate him helping me out with project administrative activities.
I have had the pleasure to work together with several trainees and guests both from
Dutch and international origin. Yuanhang Zhang worked with me on the REA paper. A
Spanish delegation with Antonio Diegues, Luis Amor and Julio Mosquera thought me
how to prepare tortilla and they were very nice colleagues to work with. Julio played an
141
important role for the Braunschweig papers and is still a good friend. So is Chris Flechard, who came after Julio left. These two have taught me to read more literature before doing something yourself. The positive aspects of proper theoretical underpinning
of the experimental work I also learned from Petra Kroon, who I supervised during her
thesis work at ECN. With her and Aline Kraai we carried out very nice projects in recent
years looking at greenhouse gas emissions and for example emissions from ships.
Great fun as long as your equipment is not washed away from the cay (life on Radio-1).
In many projects we worked together with the teams at KNMI, RIVM, RUG, TNO,
WUR, UU, VU and international teams. I especially enjoyed working with Daan Swart,
and his Lidar team and with the TNO colleagues Jan Duyzer and Hilbrant Weststrate
whom I met in several projects. Marc Zahniser and the US Aerodyne team have meant
a lot to me, providing ECN with the beautiful laser spectrometers that are key instruments in this thesis. Albrecht Neftel and his team, Pierre Cellier, Benjamin Loubet and
Patricia Laville are part of the Nitrogen research community and also active with these
instruments. This also goes for the CEH Edinburgh crew. I am honoured to have David
Fowler in the thesis committee. He and his team at CEH with Mark Sutton, Eiko Nemitz, Ute Skiba, Celia Milford and Daniela Famuliari et. al., have done and are doing
lots of innovative experiments that were an example for me and may others. From the
same scene I know Klaus Butterbach Bahl who is a bit of a modeler but that should by
no means be interpreted as a disqualification. Annette Freibauer is the lady on the
committee and she was co-coordinating the MIDAIR project that features in chapter 7
and fortunately we’ve kept in contact after that in the EU carbon projects. The other two
committee members Thomas Röckmann and Ko van Huissteden are Dutch university
partners in several projects where we cooperate either at the Cabauw tower or at the
Horstermeer site.
I thank both Han Dolman and Jan Willem Erisman as my promoter and co promoter.
Han and I have quite a long relation in time but intermittent in intensity. Jan Willem is a
nearby and stimulating colleague. Both of them helped me to get this work done. For all
layout work and corrections needed, Marloes Krijnen and Albert Bleeker provided significant input for the thesis. And there are many other colleagues in and outside ECN
that I should thank but I will not do that here to keep within the 12*1*12=144 pages
which seems appropriate given the defence day date.
Last but certainly not least I owe great and many thanks to Marije. She helped me al lot
to keep me work on this booklet, providing both time at home and comments to the final version. I thank Emma and Timo to make me realize how important this work really
is……
I hope you, the reader of this, will like what you see in this booklet and if you have any
questions left, do not hesitate to contact me.
Arjan
142
CV Author
Arjan Hensen
1966
Born in Hilversum the Netherlands
1979 -1985 VWO Willem de Zwijger College Bussum
1985 -1990 Physics University of Utrecht
Joined a JGOFS leg from Iceland to the Azores working on aerosol formation and atmospheric electricity.
1990 - now
Energy research Centre of the Netherlands
Arjan Hensen is working at ECN since 1990. He has extensive experience with development and implementation of measurement techniques for air quality and climate
change research. He worked in projects related to greenhouse gas, nitrogen issues
and particulate matter. In the same fields of research Arjan has worked on development of several knowledge dissemination tools.
Key Activities:
• Greenhouse gas measurements
o Cabauw 200m tower observations
o CO2 Flux measurements
Gradient measurements, eddy coveriace and rea measurements
o CH4 emission measurements from landfills and agricultural sources
Box and plume measurements for Afvalzorg & National ROB
projects, BSIK project ( documented in thesis Petra Kroon)
Individual landfill studies for ESSENT and SMINK
o N2O emission measurement for waste water treatment & agricultural
sources
o Mobile emission measurements
MIDAIR & GREENGRASS EU projects and survey project for the
ministry of environment
o GHG studies for biomass projects
Emission measurements from a laguna in Costa Rica
Assessment studies for Colombia, Sierra Leone and Indonesia
• Ammonia exchange between atmosphere and biosphere
o Measurement of NH3 emission and deposition
STOP programme, National projects, GRAMMINAE & Nitro
Europe EU projects
o Emission measurements housing systems
RAV project emissions from a chicken housing system
o Deposition monitoring Speuld
Life & Nitroeurope EU projects
o Landscape scale emission monitoring
Nitro Europe EU project, National NFW project
o Emissions from manured fields
Plume measurements compared with mass balance and lidar.
• Aerosol and radiation measurements
o closure experiments
MEMORA project
o PM, SO2 & NOx Emissions from ships
National projects for sea going and inland shipping
o Urban air quality measurements
Project for the municipality of Den Bosch
o Design of a energy-collecting road
143
•
Dissemination tools
o Trilemma energy game
o Nitrogenius N game
o N-Visualisation Instrument
o Clean = Cool tool
Teaching Experience:
'94, '97 Beijing University lectures
'03 – ‘11 Annual Nitrogenius seminar Wageningen University
Internship coaching for guests from Spain, Philipines, France, China and Switzerland
Internship coaching for Dutch students (Aline Kraai, Dorien Lolkema)
PhD supervisor for Petra Kroon
144