Moisture transport during the inland penetrating atmospheric river of

Moisture transport during the inland penetrating atmospheric river of

Moisture transport during the inland penetrating atmospheric
river of early November 2006 in the Pacific Northwest:
A high-resolution model-based study
Michael J. Mueller1 and Kelly Mahoney2
1 Cooperative
Institute for Research in Environmental Sciences (CIRES) CU-Boulder
2 National Oceanic and Atmospheric Administration
International Atmospheric Rivers Conference 2016
11 August 2016
Case Background & Motivation
• Series of storms in Pacific Northwest early
November 2006
• 5-7 November: Record-breaking rainfall in Cascades
and interior mountains (Neiman et al. 2008)
• Extreme rainfall and snowmelt caused destructive
flooding at Montana’s Glacier National Park
(Bernhardt 2006)
• Fraction of heavy precipitation events attributable
to atmospheric rivers (“AR fractions”) largest along
Pacific coast
• Elevated AR fractions extend to the interior of North
America north and south of the High Sierras (Rutz et
al. 2015) – including the Pacific Northwest
• Investigate how and where water vapor
penetrated the Pacific Northwest using a highresolution weather modeling system
Glacier National Park
Courtesy: National Park Service
Case Background & Motivation
• Series of storms in Pacific Northwest early
November 2006
• 5-7 November: Record-breaking rainfall in Cascades
and interior mountains (Neiman et al. 2008)
• Extreme rainfall and snowmelt caused destructive
flooding at Montana’s Glacier National Park
(Bernhardt 2006)
• Fraction of heavy precipitation events attributable
to atmospheric rivers (“AR fractions”) largest along
Pacific coast
• Elevated AR fractions extend to the interior of North
America north and south of the High Sierras (Rutz et
al. 2015) – including the Pacific Northwest
• Investigate how and where water vapor
penetrated the Pacific Northwest using a highresolution weather modeling system
Glacier National Park
Courtesy: National Park Service
Model Configuration
Model System
Weather Research & Forecasting (WRF-ARW)
Version 3.6
Domain
•
•
•
•
Single domain (2400km x 2400km)
4km grid spacing
53 terrain-following vertical levels
No cumulus parameterization scheme
Simulation
•
00 UTC 3 November 2006 – 00 UTC 9
November 2006 (144 hours)
• LBC/ICs: Climate Forecast System Reanalysis
Verification
• Precipitation: Livneh et al. (2013) dataset
based on NOAA COOP data in CONUS
Glacier
National Park
C.R.
Basin
Columbia
River Gorge
Model Terrain Height (x1000 feet)
•
•
WRF Model Terrain Height
WRF Verification: Precipitation
WRF
4
GNP
1 2
•
•
•
•
3
General agreement in distribution and magnitude
Major regions of precipitation align with mountains
Coastal Ranges: 10-15” ; 20”+ high ridges and mountains
Interior Ranges: 5-7” ; 10”+ high ridges
Accumulated Precipitation (inches)
Livneh
WRF Precipitation Plumes
2
1
10
8
6
3
4
4
GNP
2
• Antecedent precipitation 1-4”
• Main event: 00 UTC 6 Nov – 12 UTC 7 Nov (Coast)
12 UTC 6 Nov – 00 UTC 8 Nov (Interior)
00z N9
00z N8
00z N7
00z N6
0
00z N5
3
12
00z N3
1 2
GNP
14
00z N4
4
Accumulated Precipitation (inches)
Area-averaged Accumulated Precipitation
WRF Synoptic Overview: Upper Air
300 hPa
700 hPa
500 hPa
12 UTC 6 Nov
00 UTC 7 Nov
12 UTC 6 Nov
00 UTC 7 Nov
12 UTC 6 Nov
00 UTC 7 Nov
12 UTC 7 Nov
00 UTC 8 Nov
12 UTC 7 Nov
00 UTC 8 Nov
12 UTC 7 Nov
00 UTC 8 Nov
70
90 110 130 150
Wind Speed (knots)
20
25
30
Abs. Vorticity
35
40
(x10-5 s-1)
70
80
90
95
Relative Humidity (%)
WRF Synoptic Overview: Upper Air
300 hPa
700 hPa
500 hPa
12 UTC 6 Nov
00 UTC 7 Nov
12 UTC 6 Nov
00 UTC 7 Nov
12 UTC 6 Nov
00 UTC 7 Nov
12 UTC 7 Nov
00 UTC 8 Nov
12 UTC 7 Nov
00 UTC 8 Nov
12 UTC 7 Nov
00 UTC 8 Nov
70
90 110 130 150
Wind Speed (knots)
20
25
30
Abs. Vorticity
35
40
(x10-5 s-1)
70
80
90
95
Relative Humidity (%)
WRF Synoptic Overview: Upper Air
300 hPa
700 hPa
500 hPa
12 UTC 6 Nov
00 UTC 7 Nov
12 UTC 6 Nov
00 UTC 7 Nov
12 UTC 6 Nov
00 UTC 7 Nov
12 UTC 7 Nov
00 UTC 8 Nov
12 UTC 7 Nov
00 UTC 8 Nov
12 UTC 7 Nov
00 UTC 8 Nov
70
90 110 130 150
Wind Speed (knots)
20
25
30
Abs. Vorticity
35
40
(x10-5 s-1)
70
80
90
95
Relative Humidity (%)
WRF Synoptic Overview: Surface
12 UTC 5 Nov
00 UTC 6 Nov
12 UTC 6 Nov
20 kts
12
10
00 UTC 7 Nov
12 UTC 7 Nov
00 UTC 8 Nov
8
6
4
2
0
Interior precip: 12 UTC 6 Nov – 00 UTC 8 Nov
2-Meter Water Vapor Mixing Ratio (g/kg)
14
Moisture Transport Diagnostics
Integrated Water Vapor (IWV)
• Total water vapor in column
• Summed over each model level
• Units: centimeters
Integrated Vapor Transport (IVT)
•
•
•
•
Horizontal water vapor flux
Calculated as 50-hPa layer averages
Summed over layers
Units: kg m-1 s-1
Water Vapor Mass Flux (QFLUX)
• Total water vapor mass moving
through vertical unit cross section
• Calculated at each model level
• Units: kg s-1
Moisture Transport Diagnostics
Integrated Water Vapor (IWV)
• Total water vapor in column
• Summed over each model level
• Units: centimeters
Integrated Vapor Transport (IVT)
•
•
•
•
Horizontal water vapor flux
Calculated as 50-hPa layer averages
Summed over layers
Units: kg m-1 s-1
Water Vapor Mass Flux (QFLUX)
• Total water vapor mass moving
through vertical unit cross section
• Calculated at each model level
• Units: kg s-1
Moisture Transport Diagnostics
Integrated Water Vapor (IWV)
• Total water vapor in column
• Summed over each model level
• Units: centimeters
Integrated Vapor Transport (IVT)
•
•
•
•
Horizontal water vapor flux
Calculated as 50-hPa layer averages
Summed over layers
Units: kg m-1 s-1
Water Vapor Mass Flux (QFLUX)
• Total water vapor mass moving
through vertical unit cross section
• Calculated at each model level
• Units: kg s-1
WRF Integrated Water Vapor
12 UTC 5 Nov
12 UTC 6 Nov
00 UTC 6 Nov
• Approach, landfall, penetration, and
decay of AR
• Oriented WSW to ENE
• Most intense phase:
00 UTC 7 Nov
12 UTC 7 Nov
00 UTC 8 Nov
•
•
•
Centered on Ore./Wa. border
4-5cm IWV at coast
3-4cm IWV in C.R. Basin
• Decaying phase:
•
•
•
Moving south to N. Cal.
3-4cm IWV at coast
2cm IWV in NE Ore.
• Location of moisture, not transport
1
2
3
IWV (cm)
4
5
Interior precip: 12 UTC 6 Nov – 00 UTC 8 Nov
WRF Integrated Vapor Transport: Plan View
12 UTC 5 Nov
00 UTC 6 Nov
12 UTC 6 Nov
• Landfall at Ore./Wa. border
• Penetration occurred across
length of Cascades
• Maximum penetration
through C. R. Gorge and over
terrain immediately south of it
00 UTC 7 Nov
12 UTC 7 Nov
00 UTC 8 Nov
• Follow this corridor with time
series
Interior precip: 12 UTC 6 Nov – 00 UTC 8 Nov
IVT (kg m-1 s-1)
600 kg m-1 s-1
• Offshore, C. R. Gorge, C. R.
Basin, Glacier Nat’l Park
4
1400
3
1200
• Lesser AR before main event
2
1
1000
• Progressively less transport
from Cascades eastward
1
2
800
3
Maximum IVT at Point
4
600
1400
400
1000
600
200
00z N9
00z N8
00z N7
00z N6
00z N5
00z N4
200
00z N3
Integrated Vapor Transport (in kg m-1 s-1)
WRF Integrated Vapor Transport: Temporal Plumes
1
2
3
4
WRF Water Vapor Mass Flux: Plan View
Water Vapor Flux (00 UTC 6 Nov – 00 UTC 8 Nov)
• Isolates main AR from initial
landfall to decay
• Corridor of penetration
~150km C.R. Gorge and south
• Terrain south of Gorge 3-4k
feet and ~50km wide
7
6
5
4
3
• Terrain north of Gorge 4-6k
feet and ~100km wide
2
1
Vertically Integrated Water Vapor Flux (x 109 kg)
• 48-hour total water vapor flux
Model Terrain Height (x1000 feet)
12
7
10
6
8
4
2
5
4
6
3
2
1
Vertically Integrated Water Vapor Flux (x 109 kg)
WRF Water Vapor Mass Flux: Plan View
Model Terrain Height (x1000 feet)
12
7
10
6
W R E
8
4
2
W’ R’ E’
Cross-Sections
5
4
6
3
2
1
Vertically Integrated Water Vapor Flux (x 109 kg)
WRF Water Vapor Mass Flux: Plan View
WRF Water Vapor Mass Flux: Cross-Sections
West of Cascades
•
•
Height (km)
5
W
Cascades Ridge
East of Cascades
Core flux under 3km
Northern half of OR
WA OR
W’ R
R’ E
WA OR
4
3
2
1
C. R. Gorge
100
200
300
400
48-hour Total Q Flux (kg m-2)
500
WA OR
E’
WRF Water Vapor Mass Flux: Cross-Sections
West of Cascades
•
•
Height (km)
5
W
Cascades Ridge
•
•
•
Core flux under 3km
Northern half of OR
WA OR
East of Cascades
Through C.R. Gorge
Over Cascades, esp. in OR
Flux through Gorge similar
to that west of Cascades
W’ R
R’ E
WA OR
4
3
2
1
C. R. Gorge
100
200
300
400
48-hour Total Q Flux (kg m-2)
500
WA OR
E’
WRF Water Vapor Mass Flux: Cross-Sections
West of Cascades
•
•
Height (km)
5
W
Cascades Ridge
•
•
•
Core flux under 3km
Northern half of OR
WA OR
East of Cascades
•
Through C.R. Gorge
Over Cascades, esp. in OR
Flux through Gorge similar
to that west of Cascades
W’ R
•
R’ E
WA OR
4
3
2
1
C. R. Gorge
100
200
300
400
48-hour Total Q Flux (kg m-2)
500
Core of flux downwind of
Gorge
Notable weakening of flux
WA OR
E’
WRF Water Vapor Loss Over Cascades
Total Water Vapor Mass Flux (x 1012 kg)
2.0
1.5
31%
reduction
25%
reduction
1.0
0.5
West
Ridge
East
WRF Water Vapor Loss: 200 km Segments
Total Water Vapor Mass Flux (x 1012 kg)
0.6
18% reduction
Northern
Cascades
Downstream
Upstream
Downstream
Upstream
0.2
Gorge
Corridor
Downstream
18%
reduction
0.4
Upstream
25%
reduction
Southern
Cascades
Summary
• Intense inland penetrating AR in Pacific Northwest early November 2006
• Record rainfall and flooding over interior mountains (Glacier Nat’l Park)
• Water vapor breached length of Cascades (3-7k foot ridge)
• 25% water vapor transport reduction across Cascades
• Main corridor for water vapor penetration ~150 km wide from Columbia
River Gorge south
• Greater reduction north of Gorge (25%) vs. south of Gorge (18%)
Next Steps
• Follow moisture transport to interior mountains
• Calculate moisture transport including hydrometeor mass
• Modify terrain to close Gorge, increase barrier height
Acknowledgements
•
•
•
•
•
Funding from the Bureau of Reclamation
Cooperative Agreement with CIRES
Kelly Mahoney: co-author at NOAA
University of Colorado’s Research Computing (Janus supercomputer)
Weather Research and Forecasting modeling system