SUSHI - Snohomish Utility Salmon Habitat Improvements Plan

Overview
Background
Approach
Results

Overview

One of the most important impacts on water resources will be climate change. Those responsible for managing water resources must consider the potential impacts of climate change on any long-term management plan. In the Pacific Northwest, climate change will have a significant effect on the hydrologic cycle. The temperature increases will alter the type of precipitation that will fall (i.e. snow versus rain), the timing of melt of the snow, and the flows in the streams. These hydrologic changes could have a significant impact on some endangered species in the region, especially anadromous fish.

NOAA (National Oceanic and Atmospheric Administration) has created several Chinook salmon recovery plan scenarios for the Snohomish River basin. Recovery plans generally identify a suite of habitat restoration actions designed to increase population sizes to some target level. The salmon recovery targets have been set for a 50 year horizon. Developing a salmon recovery plan involves making estimates of the effects of habitat condition and functioning, hatchery and harvest management, interactions with other species, and other environmental factors on salmon population status. These potential impacts of the recovery plans are modeled using SHIRAZ, a population model created by Ray Hilborn at the School of Aquatic and Fisheries Sciences at the University of Washington (UW).

The NWFSC (Northwest Fisheries Science Center) at NOAA is sponsoring research in the Department of Civil Engineering at the UW to explore the potential impacts that climate change may have on their recovery plan alternatives. The researchers at UW will work with both the NWFSC and the Puget Sound Technical Recovery Team (TRT) to create information and perform technical analysis that will be useful in the recovery of endangered species in the Snohomish River Basin.  Specifically the research will develop and apply hydrologic and fish modeling tools to evaluate the impacts of climate change and potential management actions on endangered fish species. To examine the effectiveness of the planned restoration actions under different future scenarios (for both climate and land use) three process-based models will be used (Figure 1). General Circulation Models (GCMs) will drive the hydrologic model ( DHSVM- Distributed Hydrology Soil- Vegetation Model) that predicts the streamflow and temperature based on land cover, topography, climate, and meteorology. The outputs from DHSVM will then be run through a salmon life cycle model (SHIRAZ) that uses information on water, temperature, flow, and sediment, as well as habitat quality  indicators, to predict salmon population size.

Figure 1. Model Flowchart

Background - Snohomish River Basin

The Snohomish River basin drains about 5,000  km²  and lies just northeast of Seattle, Washington.  The Snoqualmie - Skykomish Watershed drains to the Snohomish River and out into Puget Sound.  The greatest monthly flows occur between November and February.

Figure 2. Location of the Snohomish River Basin

Figure 3. Map of Snohomish River Basin

Approach

The hydrologic models and statistical analyses of model outputs are aimed at addressing the following questions:

1. How do alternative assumptions about future climate and land use conditions affect hydrology and temperature in Pacific Northwest watersheds?

2. How do alternative assumptions about future climate and land use conditions affect predictions of fish-habitat dynamic models?

3. How robust are the alternative sets of habitat recovery actions to assumptions about future conditions (e.g., climate change, human population growth, ocean conditions)? To alternative modeling approaches?

Task 1 - Collect Data

To run a hydrologic model the following data is needed:

Task 2 - Develop a Watershed Model

DHSVM will be used as the hydrologic model. DHSVM uses GIS derived representations o f elevation, soil type, soil thickness, and vegetation. These representations are used in conjunction with meteorological forcing data to simulate water and energy fluxes at and below the land surface. DHSVM has been used successfully in other watersheds in the Pacific Northwest.

Figure 4. The layers required for DHSVM excluding the stream layer:

Each layer incorporated are modeled as 450m X 450m grid. The data sets are georeferenced using the NAD27 UTM Zone 10 coordinate system.

Task 2A - Calibrate the Watershed Model: The initial watershed model constructed in DHSVM is calibrated using available historic records.  This involves comparing the model results (streamflows) generated with historic meteorological data to recorded streamflows.  In this process, model parameters are varied until recorded runoff patterns are accurately simulated. 

Task 2B - Validate the Watershed Model: In the validation process, data not used in the calibration are used to evaluate model accuracy. 

Task 3- Running the DHSVM: The model will be applied for several different scenarios of future climate and land use.  The climate scenarios are based on the outputs of different GCM models, which predict different levels of warming and precipitation. A series of climate change impacts will be evaluated using four or more commonly used CGMs with typical greenhouse gas forcing scenarios. Future land use scenarios will be based on projections made by Snohomish County.  A rigorous sensitivity analysis will be performed in DHSVM to capture the range of management options available.

Task 3 - Develop SHIRAZ model

The outputs from DHSVM will be used as part of the input for NOAA to run the SHIRAZ model. SHIRAZ is population model that links habitat effects to fish population dynamics that is run on an annual time-step. Several life-stage transitions may occur within a given year, but by defining a beginning and an end year, the model allows for forward projections of population size by stock, life stage, and location.

Understanding all of the “H” factors (harvest, habitat, hydroelectric, and hatcheries) in a salmon life cycle can help to identify those factors whose improvements are likely to have the greatest effect on salmon recovery.  A rigorous sensitivity analysis will be performed in SHIRAZ to capture the range of management options available. NOAA will choose the recovery plan alternatives that will have the greatest effect on salmon survival.

Updated 05/13/2005