Transcript hamlet_b_USACOE_mar_2002

Water Resources Planning for an
Uncertain Future Climate
Center for Science of the Earth System and
the Department of Civil Engineering
University of Washington
March, 2002
Alan F. Hamlet
Dennis P. Lettenmaier
Overview
•Why include climate change information in water resources
planning studies at all?
•What criteria can help identify the kinds of planning studies
should include climate change information?
•What techniques can be used for including climate change
information in hydrologic studies?
•Use of linked climate and hydrologic models
•Use of historic analogues
•Some issues in the Columbia basin and general considerations
for generating alternate operating policies.
Why include climate change information in water resources
planning studies at all?
Most water systems cannot respond quickly to changes in climate. Significant
physical or operational changes to water systems frequently requires 25 years or
more to implement, the same time scale as expected changes in climate. Avoiding
impacts will require innovative and flexible long-term planning.
Many water systems in the PNW are vulnerable to loss of snowpack and
shortages of water in summer, the most likely effect of climate change.
Having a plan is not the same as implementing a plan. Long term planning
always includes uncertain information, and including climate change information
is prudent, does not cost a lot, and makes the planning process more robust to
climate issues in general, even if contingency plans are never implemented.
Addressing public and legislative concern. Being asked to redo a planning
study because climate change was not considered is costly and embarrassing.
What criteria can help identify the kinds of planning studies
that should include climate change information?
•Piggybacking: Planning is in progress or required for other
reasons (climate change assessment adds relatively little cost to an
existing planning process)
•Rare opportunity: The planning arena is unlikely to be revisited in
the next several decades due to cost or other considerations
•Sensitivity: the water system in question is highly sensitive to
reductions in snowpack and summer streamflow, or to other changes
in streamflow timing.
•Durability: High costs and/or long economic life span associated
with decisions addressed by a particular planning process
•Irreversibility: planning decisions made now that may
permanently and irreversibly increase future vulnerability
•Inflexibility: The importance of long-term planning is elevated
because planning at shorter time scales would be ineffective
What techniques can be used for including climate change
information in hydrologic studies?
Long term planning for climate change may include a stronger
emphasis on drought contingency planning, testing of preferred
planning alternatives for robustness under various climate change
scenarios, and increased flexibility and adaptation to streamflow
uncertainty.
Observed Streamflows
Planning Study
Altered Streamflows
Climate Change Scenarios
Changes in Mean
Temperature and
Precipitation or Bias
Corrected Output
from GCMs
VIC
Hydrology Model
ColSim
Reservoir
Model
Changes to Snow Extent and Naturalized Streamflow at The Dalles
April 1 Snow Extent
Estimated Range of
Natural Flow
With 2040’s Warming
Current
20th Century
Natural Flows
~2045
Changes to Mean Hydrographs Columbia Basin 2045
CHIEF JOSEPH
20000
100000
50000
0
aug
jun
DALLES
40000
20000
aug
jun
apr
feb
dec
0
300000
HC
MPI
200000
100000
0
aug
MPI
Base
jun
HC
60000
400000
apr
Base
80000
500000
dec
100000
600000
oct
120000
Average Flow (cfs)
140000
oct
Average Flow (cfs)
ICE HARBOR
feb
apr
feb
dec
0
MPI
aug
MPI
40000
HC
150000
jun
HC
Base
200000
apr
60000
250000
feb
Base
300000
dec
80000
350000
oct
100000
Average Flow (cfs)
120000
oct
Average Flow (cfs)
CORRA
Historic Analogues: El Niño and drought years as a surrogate
for climate change
Water Year 2001
Water Year 1992
Years with Unusually High Winter Flows
800000
1996 CC 3
700000
Forecast2
Forecast3
600000
High Climatology
500000
Low Climatology
400000
Observed Virgin
Flow
300000
200000
100000
Month
sep
aug
jul
jun
may
apr
mar
feb
jan
dec
nov
0
oct
Streamflow (cfs)
Forecast1
Years with Unusually Early or Rapid Snowmelt
800000
Forecast1
700000
Forecast2
600000
Forecast3
500000
Forecast4
400000
Forecast5
Forecast6
300000
Forecast7
200000
Forecast8
100000
High Climatology
Month
sep
aug
jul
jun
may
apr
mar
feb
jan
dec
nov
0
oct
Streamflow (cfs)
1998 CC 4
Low Climatology
Observed Virgin
Flow
Some Water Management Issues in the Columbia Basin
Affecting Potential Climate Change Impacts
Conflicts between irrigation, hydropower, flood control, and
instream flows for fish. Reductions in summer streamflow are
likely to exacerbate these conflicts. The relatively fragmented 2001
drought response demonstrates that we are not well prepared for this
eventuality in the PNW.
Columbia River Treaty--~50% of the system storage is located in
and controlled by Canada. For the future climate a larger proportion
of the snowpack and late summer streamflow may also be in Canada.
That water no longer flows across the border unimpeded.
Lack of Drought Contingency Planning. The situation in the
Klamath basin in 2001 is an example of what can happen if
fundamental climate vulnerabilities are ignored until major impacts
occur.
Implications for Flood Control in the Columbia Basin
in a Warmer Climate
Current flood control operations may provide insufficient
flood protection in late fall and early winter (higher natural
streamflows in winter due to warming)
Flood evacuation quantity and timing may need to be adjusted
to achieve appropriate flood protection in early spring AND
acceptable reliability of reservoir refill and instream flow in
spring and summer. (earlier snowmelt and reduced snowpack)
Other changes in the system may significantly alter the current
management framework in the basin. Flood control will need
to be flexible enough to mesh with these changes. (E.g. move
dominant hydropower production to summer.)
Qualitative Changes to Flood Evacuation Requirements
to Adjust to Warmer Conditions
7000000
5000000
4000000
3000000
July
June
May
April
March
December
November
October
September
0
February
1000000
Altered RC
(Wet Year)
Altered RC
(Dry Year)
January
2000000
August
Storage (acre- ft)
6000000
Conclusions
The PNW’s vulnerability to reduced snowpack, and the time frame needed to make
substantive changes in water systems places a strong emphasis on long-term
planning as a means to successfully re-designing our water systems to be robust to
the potential impacts of climate change.
Piggy backing climate change adaptation studies with planning studies already
proposed for other reasons is a good way to include climate change information
without adding large costs.
A number of straight-forward techniques are currently available for including
climate change scenarios in planning studies, including integrated hydrologic
modeling, and use of unusual historic water years as analogues for flows in a
warmer PNW.
The implications for flood control include the potential for increased risk of
flooding in late fall and early winter (more run-off in winter), and changes in the
timing and quantity of peak flows in spring (earlier and less spring and summer
runoff) . Operational flexibility will be essential in coping with uncertain changes
as they unfold.