Transcript Effects of 20th Century Climate Change on Mountain Watersheds in

Effects of 20th Century Climate
Change on Mountain Watersheds
in the Western U.S.
Center for Science in the Earth System
Climate Impacts Group
and Department of Civil and Environmental Engineering
University of Washington
March, 2005
http://www.hydro.washington.edu/Lettenmaier/Presentations/2005/hamlet_cee_seminar_mar_2005.ppt
Alan F. Hamlet Philip W. Mote
Martyn Clark
Dennis P. Lettenmaier
Changes in Simulated
April 1 Snowpack for the
Cascade Range in
Washington and Oregon
Current Climate
“2020s” (+1.7 C)
-44%
April 1 SWE (mm)
“2040s” (+ 2.5 C)
-58%
Effects to the Cedar River (Seattle Water Supply)
for “Middle-of-the-Road” Scenarios
9000
8000
+1.7 C
6000
Simulated 20th
Century Climate
2020s Climate
Change Scenario
2040s Climate
Change Scenario
5000
4000
3000
2000
+2.5 C
1000
Date
9/2
8/5
7/8
6/10
5/13
4/15
3/18
2/18
1/21
12/24
11/26
10/29
0
10/1
Inflow (acre-ft)
7000
Obs. Summer Water Availability is Declining
0.7
Cedar River: -30.7%
May-Sept frac
Linear (May-Sept
frac)
0.5
0.4
0.3
0.2
0.6
SFTolt River: -15.7%
May-Sept frac
y = -0.0020x + 4.3416
0.1
0
1945
1955
1965
1975
55 years
1985
1995
2005
May-Sept fraction of annual flow
May-Sept fraction of annual flow
0.6
Linear (May-Sept
frac)
0.5
0.4
0.3
0.2
y = -0.0010x + 2.2890
0.1
0
1945
1955
1965
1975
1985
1995
2005
Figures courtesy of Matt Wiley and Richard Palmer at CEE, UW
Water Year 2005: Crazy weather or
climate change?
SWE Mar 1, 2005
http://www.hydro.washington.edu/Lettenmaier/Projects/fcst/
Observed Climate Change
and Hydrologic Impacts for the West
Physical Characteristics of the Mountain West
Elevation (m)
DJF Temp (C)
NDJFM PCP (mm)
Schematic of VIC Hydrologic Model and Energy Balance Snow Model
PNW
GB
CA
CR
B
Snow Model
Trends in April 1 SWE 1950-1997
Source: Mote et al. , 2005, Declining Snowpack in the Western U.S.,BAMS 86 (1)
Why Do We Need Model Simulations of
the Historic Record?
•Longer Record (Avoids problems with PDO from 1950-1997)
•Spatial Coverage (high and low elevations not in the
observations)
•Temporal Resolution (daily time step)
•Consistency between SWE and streamflow
•Explicit sensitivity analysis for effects of temperature and
precipitation
Effects of the PDO and ENSO on Columbia River
Summer Streamflows
PDO
450000
Cool
Cool
Warm
Warm
350000
300000
250000
200000
Red = Warm ENSO Green = ENSO neut. Blue = Cool ENSO
2000
1990
1980
1970
1960
1950
1940
1930
1920
1910
150000
1900
Apr-Sept Flow (cfs)
400000
DJF AVG T (C)
1916-1997
Relative Trend in April 1 SWE
(% per year)
DJF AVG T (C)
1916-1997
Effects of Temp
Relative Trend in April 1 SWE
(% per year)
DJF AVG T (C)
1916-1997
Effects of Precip
Relative Trend in April 1 SWE
(% per year)
Decadal Climate Variability Doesn’t Explain the Loss of
SWE Due to Warming
1916-97
1947-97
1925-46
with 1977-95
Relative SWE Trends Due to Temperature Effects Alone (% per year)
Decadal Climate Variability Does Explain Some of the
Changes of SWE Due to Precipitation Changes
1916-97
1925-76
1947-97
Relative SWE Trends Due to Precipitation Effects Alone (% per year)
Figure 7
Trends from 1916-1997
Region 1
Region 1 (Coastal)
Region 2 (Inland)
Region 3 (Interior)
Trend %/yr
djf avg T (C)
1
2
Region 2
3
Trend %/yr
Trend %/yr
Region 3
Effects due to
precip trends
only
Trend %/yr
a) 10 % Accumulation
b) Max Accumulation
c) 90 % Melt
Trends
in
SWE
19161997
Change in Date
Change in Date
Change in Date
DJF Temp (C)
DJF Temp (C)
DJF Temp (C)
Change in Date
Change in Date
Change in Date
Trends from 1916-1997
DJF Temp (C)
Change in Date
DJF Temp (C)
Effects
of Temp.
Trends
Alone
DJF Temp (C)
Change in Date
c) 90 % Melt
Change in Date
DJF Temp (C)
DJF Temp (C)
Effects
of Temp.
and
Precip.
b) Max Accumulation
DJF Temp (C)
a) 10 % Accumulation
600
Area Average Water
(depth in mm)
Seasonal
Water Balance
Naches River
700
500
precipitation
400
swe
runoff+baseflow
soil storage
300
evapotranspiration
200
100
Current Climate
sep
aug
jul
jun
may
apr
mar
feb
jan
dec
600
500
precipitation
400
swe
runoff+baseflow
soil storage
300
evapotranspiration
200
100
sep
aug
jul
jun
may
apr
mar
feb
jan
dec
nov
0
oct
2040s Scenario
(+ 2.5 C)
700
Area Average Water
(depth in mm)
More runoff in
winter and
early spring,
less in
summer
nov
oct
0
Basin Averaged Runoff (mm)
Changes in Average Runoff Timing in the Naches River
180.00
160.00
140.00
120.00
+ 2.5 C
100.00
current climate
composite 2040
80.00
60.00
40.00
20.00
0.00
oct
nov dec
jan
feb mar apr may jun
jul
aug sep
Current Climate
Water Balance from April-September
(depth in mm)
700
600
500
precipitation
400
snowmelt
soil drainage
300
streamflow
ET
200
100
0
1
Climate Change
Scenario + 2.5 C
Water Balance from April-September
(depth in mm)
700
600
500
400
-110
-220
precipitation
snowmelt
soil drainage
300
streamflow
ET
200
100
0
1
As the West warms,
winter flows rise
and summer flows
drop
I.T. Stewart, D.R. Cayan, M.D.
Dettinger, 2004, Changes toward
earlier streamflow timing across
western North America, J. Climate
(in review)
Figure courtesy of Iris
Stewart, Scripps Inst. of
Oceanog. (UC San Diego)
Trends in fraction of annual runoff 1947-2003 (cells > 50 mm of SWE on April 1)
March
Relative Trend (% per year)
June
Effects of temp and precip
Runoff
Effects of temp only
June
June
19162003
Relative Trend (% per year)
Effects of temp and precip
Runoff
Effects of temp only
Sept
Sept
19162003
Relative Trend (% per year)
Soil Moisture
Effects of temp and precip
Effects of temp only
Apr 1
Apr 1
19252003
Relative Trend (% per year)
Soil Moisture
Effects of temp and precip
Effects of temp only
Sep 1
Sep 1
19252003
Relative Trend (% per year)
Cascades Sub Domain
Elevation (m)
Trends in April 1 SWE for the WA and OR Cascades
600
-19%
500
400
Effects of Temperature
And Precipitation
1-Apr
300
Linear (1-Apr)
200
-2.15% per decade
100
y = -0.5851x + 295.29
2001
1996
1991
1986
1981
1976
1971
1966
1961
1956
1951
1946
1941
1936
1931
1926
1921
1916
0
500
-25%
450
400
350
300
Effects of Temperature
Alone
1-Apr
250
Linear (1-Apr)
200
150
100
50
2001
1996
1991
1986
1981
1976
1971
1966
1961
1956
1951
1946
1941
1936
1931
1926
1921
0
1916
-2.84% per decade
y = -0.7553x + 301.86
Trends in April 1 SWE for the WA and OR Cascades
600
-35%
500
400
Effects of Temperature
And Precipitation
1-Apr
Linear (1-Apr)
300
200
-6.48% per decade
100
y = -1.739x + 313.96
2001
1998
1995
1992
1989
1986
1983
1980
1977
1974
1971
1968
1965
1962
1959
1956
1953
1950
0
400
-23%
350
300
250
Effects of Temperature
Alone
1-Apr
Linear (1-Apr)
200
150
y = -1.1264x + 288.63
100
50
2001
1998
1995
1992
1989
1986
1983
1980
1977
1974
1971
1968
1965
1962
1959
1956
1953
0
1950
-4.25% per decade
Combined Cedar-Tolt basin wide average April 1 SWE
Simulated from HadCM3
Simulated from Observed Climate
Linear (Simulated from HadCM3)
Linear (Simulated from Observed Climate)
KAF
100
50
0
1935
1955
1975
1995
2015
2035
2055
• Transient SWE simulation from HadCM3 (A2)
GCM run (with running 10 year average smoothing)
• Simulated from observed climate shows a
declining trend of ~3KAF per decade (19352000)
• HadCM3 simulated declines ~4KAF per decade
Figure courtesy of Matt Wiley and Richard Palmer at CEE, UW
2075
What about WY 2005? Is this climate change?
It would appear that 2005 has more to do with very unusual
weather patterns than a systematic long-term change in the
climate system. Storm track redirection, for example, is
typical of warm ENSO years.
That said, we show that the risk of unusually
warm years like 2005 has been increasing
over time. These unusually warm years also
provide valuable information about the
impacts of what may prove to be typical
conditions forty years from now.
If 2005 proves to be warm and dry in
summer, Seattle Public Utility may have
difficulty in meeting demands. This kind of
experience can have important implications
for long-term planning and policy.
Conclusions
•Large-scale changes in the seasonal dynamics of snow
accumulation and melt have occurred in the West as a result of
increasing regional temperatures.
•The most sensitive areas are coastal mountain ranges with
relatively warm winter temperatures (e.g. the Cascades)
•Hydrologic changes include earlier and reduced peak
snowpack, more runoff in March, less runoff in June, and
corresponding increases in simulated spring soil moisture and
decreases in late summer and fall soil moisture.
•Because these effects are shown to be predominantly due to
temperature changes, we expect that they will both continue
and increase in intensity as global warming progresses in the
21st century.