Alan F. Hamlet

1
Weekly and Daily Climate Change Streamflow
Scenarios and Estimates of Changing Crop Water
Demand
JISAO Center for Science in the Earth System
Climate Impacts Group and Department of Civil
and Environmental Engineering University of
Washington October, 2004

http//www.hydro.washington.edu/Lettenmaier/Presen
tations/2004/hamlet_daily_flow_crop_et_2004.ppt
Alan F. Hamlet Dennis P. Lettenmaier
2

• Problem
• 1) Monthly naturalized streamflow observations
  are frequently available for a large number of
  sites over long periods of time, but availability
  of weekly and daily observations is typically
  very limited.
• 2) Monthly climate change scenarios are useful,
  but for many studies weekly or daily flows are
  required (e.g. flood control).
• Methods are needed to
• Produce weekly and daily observed records that
  are consistent from monthly naturalized data.
• Produce climate change scenarios at weekly and
  daily time step that are consistent with observed
  data sets.

3
Short Time Step Streamflow Reconstruction Process
Monthly average value comes from observed data.
The daily and weekly time history come from the
simulations.
Observed Monthly Average Streamflow Data
Reconstructed Observed Daily Record
Simulated Daily Streamflow Data
VIC Hydrologic Model Driven by Observed Temperatur
e and Precipitation Data
Reconstructed Observed Daily Record
Simulated Weekly Streamflow Data
4
Reconstructed Naturalized Weekly and Daily Flows
at Palisades Dam for 1958-1992
Weekly Flow 1958-1992
Streamflow (cfs)
Daily Flow 1958-1962
5
Short Time Step Climate Change Streamflow Scenario
Bias Corrected Monthly or Weekly Climate
Change Streamflow
Monthly or weekly data comes from bias corrected
simulation. The daily time history used to
construct the daily data comes from the
simulations.
Daily Time Step Streamflow Scenario
VIC Hydrologic Model Driven by Climate
Change Temperature and Precipitation Scenario
Simulated Daily Streamflow Data
6
Weekly Climate Change Scenario for Palisades (MPI
2040 warm and dry)
7
Long-Term Estimates of Potential
Evapotranspiration from a Reference Crop
8

• Problem
• Quantitative, spatially-explicit estimates of
  evaporation from irrigated crops are needed for
• Estimates of future surface water diversions and
  return flows as a function of climate, irrigation
  technology, crop type, etc.
• Estimating losses from aquifers due to
  groundwater pumping for irrigation and aquifer
  recharge due to surface water application.

9
Methods A well-tested and frequently used
method is to estimate the potential evaporation
(PotET) from a well-watered reference crop (e.g.
mature alfalfa), and then relate this to the
PotET for other crops using linear factors that
vary with crop type and season Actual Crop
PotET Kc (PotETref) (where Kc varies with
date and actual crop) PotETref is often
estimated by the Penman Monteith equation. See
e.g. http//www.cprl.ars.usda.gov/wmru/pdfs/982
123.pdf
10
Conceptual Diagram of the Penman Monteith Approach
Aerodynamic Resistance Wind Speed Crop Height
Potential Evapotranspiration
Canopy Resistance Stomotal Resistance Leaf Area
Index
Surface Energy Incoming Solar
Radiation Outgoing Longwave
Vapor Pressure Deficit Temperature Relative
Humidity
11
Schematic Diagram of Simulation Tool for
Producing Long Records of PotET
Gridded Daily Precipitation and Temperature
Records 1915-2002
Daily Time Series of Estimated Reference Crop
PotET 1915-2002
12
Seasonal Cycle of PotET for a Single Grid Cell in
the Snake River Plain
13
Average July PotET for Alfalfa Reference Crop
Potential ET (mm)
14
Four Delta Method Climate Change Scenarios for
the PNW
2.5 C
1.7 C
Somewhat wetter winters and perhaps somewhat
dryer summers
15
Average July PotET over the Southern Plain Region
Current Climate vs. MPI2040 scenario
Current Climate
MPI2040
PotET (mm/day)
16
Trends in July Avg PotET over the Southern Plain
Region from 1915-2002
17
Conclusions Long-term gridded temperature and
precipitation records can be used to drive
hydrologic models to simulate potential ET for a
reference crop. Simple experiments in which the
temperature is perturbed while other explanatory
variables remain about the same suggest that crop
water demand ought to be going up over time as
the region warms. The long term historic
simulations, however, show that the trends are
downward over time. One possible explanation for
these trends is associated with increasing night
time temperatures, which indicate that
atmospheric moisture content is systematically
increasing. This reduces the vapor pressure
deficit and the incoming solar radiation. These
results suggest that changes in relative
humidity, cloudiness, and wind may play a more
dominant role than temperature alone in
controlling ET. If so, more sophisticated
methods for evaluating the effects of changing
climate on these variables will be needed to
better assess the potential changes.