Richard Palmer

1
Water Supply and Allocation Issues in the Puget
Sound

• Richard Palmer
• Michael Miller
• University of Washington
• Department of Civil and Environmental Engineering

2
Objective and Outline

• What will be the climate and hydrology of the
  Puget Sound in 2020/2040 and what are the impacts
  of climate change on water supply issues?
• How can mid-term forecasts improve management of
  water supplies for people and fish?

3
(No Transcript)
4
(No Transcript)
5
Objective and Outline

• What will be the climate and hydrology of the
  Puget Sound in 2020/2040 and what are the impacts
  of climate change on water supply issues?
• How can mid-term forecasts improve management of
  water supplies for people and fish?

6
(No Transcript)
7
Current state of climate modeling

• Climate models are currently capable of credibly
  simulating present climate at the continental
  scale.
• Models are continually improving, yet key
  physical relationships remain poorly understood,
  the water vapor/cloud formation and feedback
  process being the most significant.
• Greater resolution and more complex
  parameterization of physical processes will
  continue as computing power increases and study
  continues.
• Models are not predictions of future, but can be
  considered as credible simulations of a multitude
  of possible futures.

8
GCMs - General Circulation Models

• IPCC discusses 34 GCMs
• Coupled Model Intercomparison Study examines 29
  in more detail
• Compares GCMs via historical observations for air
  temperature, precipitation, sea temperature, air
  pressure, ice extent.
• We have selected nine of the more prominent
  models to demonstrate GCM selection process

9
(No Transcript)
10
Downscaling Methods

• Dynamical downscaling
• Regional climate models Difficult and computer
  intensive. Not yet proven to be any more
  reliable.
• Statistical downscaling
• Transfer functions Based on observed empirical
  relationships.
• Weather generators Extension of stochastic
  hydrology
• Weather typing Use of historic patterns,
  predicated on observed climactic variables.

11
Downscaling Methods
12
Downscaling Methods
13
Evaluation of Climate Change
Climate Shift
Meteorological Data
Hydrology Model
Demand Model
Operations Model
14
Temperatures will increase by 2 C by 2040, with
higher temperatures in the summer
Precipitation will increase in the winter and
decrease in the summer.
15
DHSVMDistributed Hydrology-Soil-Vegetation Model
16
Sultan River Inflows into Spada Reservoir Average
Annual Hydrograph
17
Tolt River Inflows into Tolt Reservoir Average
Annual Hydrograph
18
Cedar River Inflows into Chester Morse Reservoir
Average Annual Hydrograph
19
Green River Inflows into Howard Hansen Reservoir
Average Annual Hydrograph
20
Ranked Cumulative Winter Flow (JFM) 2040
cfs-weeks
32
43
21
Ranked Cumulative Spring (AMJ) Flow 2040
cfs-weeks
-30
22
Results Impacts on HydrologyPercent difference
from current climate cumulative seasonal flows
23
Conclusions

• Climate impacts on the four basins hydrology are
  similar
• Average percent difference in seasonal flows
• 2020 Winter 28 2020 Spring
  -20
• 2040 Winter 37 2040 Spring
  -31
• Absolute average percent difference
• Sultan 31
• Tolt 18
• Cedar 36
• Green 32
• Average supply system impact is 15-17 increase
  in System Use (surface storage, groundwater
  and/or system shortfalls)

24
Climate Impact on Water Supply Average climate
impact on Supply Used, Percent Difference from
Current Climate
25
Possible Reactions to Climate Change Information

• Supply
• Tacoma to Seattle Connection (2nd Supply Project)
  
• Seattle to Everett Connection
• Water Reuse
• Demand
• Conservation Measures
• Pricing
• Change Service Base

26
Objective and Outline

• What will be the climate and hydrology of the
  Puget Sound in 2020/2040 and what are the impacts
  of climate change on water supply issues?
• How can mid-term forecasts improve management of
  water supplies for people and fish?

27
Why do a Forecast?

• 6-month forecast applied to the PRISM models
• Usefulness of forecasts
• Why forecasts are useful
• Who could use the forecasts
• How are the forecasts developed
• Examples of the forecasts
• Future direction with the forecasts

28
(No Transcript)
29
Applying 6-month Forecast

• Prior to a forecast
• Water management decisions
• 50 years of meteorological records
• 73 years water supply and demand records
• With a forecast
• Water management decisions based on potential
  future conditions
• Forecast continue using DHSVM and CRYSTAL for
  water supply and management

30
PRISM
31
Usefulness of Forecast

• For policymakers
• MI Demands
• During below average conditions
• Improve timing of water restrictions
• Provide more information as to the type of
  restriction
• HCPs
• During above average conditions
• Determine amount and length of large flow
  releases
• During below average conditions
• Revise timing of releases to minimize habitat
  damage

32
Usefulness of the Forecast

• For water managers
• During average and above average flow
• Forecast potential of these resources
• Discharge necessary to meet future flood control
• During below average flows
• Forecast initial drought conditions a couple
  months sooner
• In the early summer months, forecasts could
  indicate when fall and winter flows will increase

33
Forecast Development

• Developed by Andy Wood,
• Edwin Maurer, Arun Kumar, and
• Dennis Lettenmaier
• NCEP Data
• Bias Correction
• Downscaling
• DHSVM

34
NCEP Data

• National Center for Environmental Prediction
  (NCEP)
• Global Spectral Models (GSMs)
• Hindcasts
• Temperature and precipitation
• 10 initial conditions
• 21 years (79 99)
• Forecasts
• 20 ensembles
• 6-month forecast

35
Bias Correction
36
Downscaling

• Forecasted meteorological data
• Based on month from the historic 21-year record
  (79-99), most similar precipitation
• Precipitation is scaled (multiplicative process)
• Temperature is shifted (additive process)
• Preformed to each month of each ensemble

37
DHSVM

• Distributed Hydrologic Soil Vegetation Model
  (DHSVM)
• Most recent year of actual data run prior to the
  forecast
• To have the model set for the forecasted data.
• Keep model run time reasonable

38
DHSVM Output

• 20 forecasts of stream flow
• Forecasts are compared to historic average flows
• Comparison used to forecast higher or lower then
  average flow.

39
June Forecast Cedar
40
(No Transcript)
41
(No Transcript)
42
Summary

• Climate Change
• Initial results suggest significant impacts on
  water supply
• Lower summer flows will challenge releases for
  both fish and folks
• New management strategies are necessary
• Future planning should include this impact

43
Summary on Forecasting

• Forecasting with longer-range climate indicators
  offers promise
• Past forecasts did not provide sufficient lead
  time for certain times of year
• Will have on line forecasts this quarter