Download www.cwemf.org

Water Supply and
Flood Forecasting with
Climate Change
Michael Anderson, PhD
California Department of Water Resources
Division of Flood Management
Acknowledgments
Thanks to David Rizzardo, Matt Winston,
Boone Lek, BG Heiland, Tawnly Pranger,
Aaron Miller, Maury Roos, John King, and
Steve Nemeth of DWR-DFM and Jamie
Anderson of DWR Delta Modeling.
Presentation Outline
• Introduction
• Historical Trends
• Climate Change Information
• Impacts
• Future Work
How much of the New Year’s flooding
was due to climate change?
Photo by Ralph Finch
Introduction
• Climate vs. Weather – Climate is what you
expect and weather is what you get
• Global Circulation Models aim to predict
climate change
• Individual weather events may vary greatly
• New Year’s event within historical observed
variability
Introduction
• Climate Change impacts of note for
flood/water supply forecasting:
– Temperature
– Precipitation type (snow vs. rain)
– Seasonal shifts in precipitation or snowmelt
• Examine historical record for trends
• Look at climate change data for possible
future scenarios
Historical Trends - Temperature
• Statewide average observed warming of about 0.5 °C (1 °F)
Temperature (deg F)
80
70
60
50
40
1900
1910
1920
1930
Maximum
1940
1950
Average
1960
1970
Minimum
1980
1990
2000
Historical Trends Precipitation
• Slight Increase in statewide average precipitation
Precipitation (inches)
50
40
1980-2000:
25.6 inches
1890-2000:
23.9 inches
30
20
10
0
1915-1935:
20.8 inches
1890 1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000
Year
Changes in Peak Runoff Statistics
Pre/Post
1955
Feather
Tuolumne
Eel
Mean
42/52
12/17
93/123
Standard
Deviation
33/50
11/19
48/84
145/232
52/91
165/489
Range
Values in 1000 cfs for annual peaks of 3-day average flows
1904-2004 data used for analysis
Range is maximum-minimum values for time period
Historical Trends – Runoff Timing
Sacramento River System
April-July Runoff
Runoff Volume (tAF)
16,000
14,000
12,000
10,000
8,000
6,000
4,000
2,000
0
1901
1911
1921
1931
1941
1951
1961
1971
Water Year (October 1 - September 30)
1981
1991
2001
Historical Trends – Runoff Timing
Sacramento River System
Water Year Runoff
Runoff Volume (tAF)
40,000
32,000
24,000
16,000
8,000
0
1901
1911
1921
1931
1941
1951
1961
1971
1981
Water Year (October 1 - September 30)
1991
2001
Historical Trends – Runoff Timing
San Joaquin River System
April - July Runoff
Runoff Volume (tAF)
10000
8000
6000
4000
2000
0
1901
1911
1921
1931
1941
1951
1961
1971
1981
Water Year (October 1 - September 30)
1991
2001
Historical Trends – Runoff Timing
San Joaquin River System
Runoff Volume (tAF)
Water Year
16,000
14,000
12,000
10,000
8,000
6,000
4,000
2,000
0
1901 1911 1921 1931 1941 1951 1961 1971 1981 1991 2001
Water Year (October 1 - September 30)
A look ahead…
Climate change model data from
GCMs statistically downscaled
over California
Projected Changes Temperature
Some Uncertainty
From Dettinger, 2005
Projected Changes Precipitation
Lots of Uncertainty !
From Dettinger, 2005
Analysis of Projected Changes
• There is less variability in air temperature
projections vs. precipitation projections
• GCMs need refinement before flood/
drought frequency analysis appropriate
• Analysis focuses on impacts of potential
temperature changes
Snowpack Reduction Impacts
• A 3° C increase could result in a 33%
decrease in Sierra Nevada snowpack
(~5 MAF loss in snow water storage)
• North more sensitive to change than south
• Feather basin snowpack area decreases from
72% to 20% for a 3° C warming and to
2% for a 5° C warming
Values assume a 500 ft increase in snow line
for each degree Celsius warming.
Percent Snow Free
Snowpack Reduction Impacts
100
90
80
70
60
50
40
30
20
0
1
2
3
4
5
Avg. Atm. Temp. Increase ( C)
Above Shasta
Above Oroville
Above Folsom
Storm Runoff Impacts
Storm Runoff Impacts
• Higher snow levels yield more direct runoff per storm
Base Case
1 degree increase
3 degree increase
5 degree increase
Scaled Runoff
2.5
2.0
1.5
1.0
0.5
0.0
0
12
24
36
48
60
72
Hour
84
96
8
0
1
0
2
1
2
3
1
4
4
1
Water Supply/Flood
Forecasting Issues
• Will climate change move snowmelt into
the March transition period?
• Will there be more/fewer storms?
• Will the character of storms change and
how will that affect forecasting?
Future Directions
• Continue analysis of historical data
• Evaluate new GCM data as available
• Study flood producing atmospheric
circulations (historical and GCM data)
• Determine probability of occurrence of
potential impacts
• Identify mitigation measures
Questions ?