The datasets included in this ftp site contain the projected weekly water temperatures for over 100 sites in Washington State. This project was completed as part of the Washington State Climate Change Impacts Assessment by the Climate Impacts Group (http://cses.washington.edu/db/pdf/wacciareport681.pdf).

These stream temperture projections are derived from a non-linear regression model that uses air temperatures to estimate water temperatures at specific sites. The data presented here are estimates for summer weeks (weeks 15 - 40 of the 52 in a year) projected through the 21st century. These estimates are based on projected weekly air temperatures from ten statistically downscaled global climate models under two scenarios (A1B and B1) included in the CMIP3 results.

The columns for each model run include the following headers:

Model_scenario Data_source Region Site Latitude Longitude Year Week Modeled_Water_Temp(C) Modeled_Air_Temp(C)

The composite A1B and B1 datasets contain the weekly estimates for air and water temperatures averaged over all models for each scenario.

The data sources are as follows:
USGS United States Geological Survey
USACE United States Army Corps of Engineers
DOE Washington State Department of Ecology

The regions are as follows:
UC upper Columbia upstream of Bonneville Dam
LC lower Columbia downstream of Bonneville Dam
PS Puget Sound
OS Olympic Penninsula

A key to the site names, locations and regression parameter estimates can be found in the "sites_and_parameters.pdf" document.

Any questions can be directed to Guillaume Mauger (gmauger@uw.edu)

---

Data format (space delimited):
     Model_scenario Data_source Region Site Latitude Longitude Year Week Modeled_Water_Temp(C) Modeled_Air_Temp(C)
     e.g.: ccsm3_A1B USACE UC ALFI 48.15625 -117.09375 2010 15 6.51 6.29

There areÊseparate files for each greenhouse gas scenario (B1, A1B) and global climate model (GCM).Ê

For A1B there are 10ÊGCMs:
     ccsm3, cgcm3.1_t47, cnrm_cm3, echam5, echo_g, hadcm, hadgem1, ipsl_cm4, miroc_3.2, pcm1

For B1 there are 9 GCMs -- the same list as for A1B except hadgem1:
     ccsm3, cgcm3.1_t47, cnrm_cm3, echam5, echo_g, hadcm, ipsl_cm4, miroc_3.2, pcm1

File naming format is: <GCM>_<scenario>_temps_pub.txt

There arealso "composite" files for each scenario, which give the results for the average change in temperature/precipitation among all 10 GCMs. Naming:
     A1B_composite_temps_pub.txt
     B1_composite_temps_pub.txt