T
T
T

Stream Temperature Modeling and Monitoring

Status
Ongoing

Thermal regimes are important to aquatic ecosystems because they strongly dictate species distributions, productivity, and abundance. Inexpensive digital temperature loggers, geographic information systems (GIS), remote sensing technologies, and new spatial analyses are facilitating the development of temperature models and monitoring networks applicable at broad spatial scales.

This stream temperature modeling and monitoring web site provides resources to help those in the western United States organize temperature monitoring efforts, describes techniques for measuring stream temperatures, and describes several statistical models for predicting stream temperatures and thermally suitable fish habitats from temperature data. The site also has useful links to other stream temperature resources such as publications, videos, and presentations on topics relating to thermal regimes in streams. By accurately portraying and making available a comprehensive and updated set of stream temperature sites monitored by several agencies, we hope to facilitate data sharing and avoid redundancies as new monitoring sites are added to the regional network. If you are interested in obtaining temperature data or have sites to add to stream temperature monitoring maps, please contact the point of contact under the People tab.

Air Temperature Based Thermal Stream Habitat Model

This model was developed to predict air temperatures based on elevation, and geographic latitude and longitude with good success (R2 ~ 0.89).

Spatial Statistical Stream Temperature Model

This model uses thermograph data in conjunction with predictor variables (climatic, vegetation, geomorphic) to predict stream temperature. These models are more accurate than non-spatial models and offer far greater predictive ability (R2 ~ 0.90).

Multiple Regression Models

Multiple Regression Models use thermograph data and geomorphic predictor variables to predict stream temperature metrics with moderate accuracy (R2 ~ 0.65)

NorWeST

NorWeST is a comprehensive database of stream temperatures for all streams across the Northwest. Temperature data is being used with spatial statistical stream network models to develop an accurate and consistent set of future climate scenarios.

SSN & STARS  

SSN & STARS is a set of tools used to accurately model stream networks based on spatial statistical stream models. 

Data Downloads & Maps

Air Temperature Based Thermal Stream Habitat Model

Data or information used from this section should be cited as: Rieman, B.E., D.J. Isaak, D. Myers, S. Adams, C. Luce, D. Horan, D. Nagel. 2007. Anticipated effects of climate warming on bull trout within the Interior Columbia River Basin. Transactions of the American Fisheries Society. 136: 1552-1565. 

  • Project Summary
  • Methods and Metadata 

  • Maps - Includes various stream temperature scenarios

    MapDescription
    Study Area ExtentStudy area extent for the publication
    Fish Sample LocationsFish sample locations and study area extent    
    Thermally Suitable HabitatCurrent and projected thermally suitable habitat in the Columbia River basin    
  • Data 
  • GIS Data and Metadata - Interior Columbia River Basin      
    • These GIS data are intended to assist users in understanding general patterns in bull trout distributions and inferring potential alteration of these distributions with changes in future mean annual air temperatures.  The data have been developed specifically for bull trout and are not intended for use with other aquatic organisms unless similar linkages with air temperatures can be established.  The data are most appropriate for broad scale displays and inference (i.e., map scales ~ 1:1,000,000) and should not be applied at finer scales, where local conditions may cause significant deviations from model predictions.  The lower limit bull trout model predicts accurately (R2 = 0.74) across the Columbia River basin, but has a RMSE of 180 m (RMSE = average elevation difference between observed values and model predictions). This amount of error would easily obscure meaningful inference when applied to an individual stream.
TitleDescriptionDownloads Available Up Request from Project Contacts
Documentation and MetadataDetailed process steps, documentation and metadata Documentation and Metadata
Model DevelopmentFish sample and air temperature table data Model Development
WatershedsStudy area hydrologic units (HUCs)Watersheds
Lower Limit FilesLower limit elevation line files Lower Limit Files
Stream LinesTauDEM stream lines Stream Lines
Intersection TablesLower limit elevation and stream line intersection tables Intersection Tables

Spatial Statistical Stream Temperature Model

Data or information used from this section should be cited as: Rieman, B.E., D.J. Isaak, D. Myers, S. Adams, C. Luce, D. Horan, D. Nagel. 2007. Anticipated effects of climate warming on bull trout within the Interior Columbia River Basin. Transactions of the American Fisheries Society. 136: 1552-1565. 

  • Project Summary
    • Project Summary, Data Analysis and Model Development
  • Methods and Metadata 

  • Maps - Includes various stream temperature scenarios

    MapDescription
    Study Area ExtentStudy area extent for the upper Boise River basin in central Idaho 
    Study Area Dams and ReservoirsDam and Reservoir locations for the upper Boise River basin in central Idaho
    Fire PerimetersFire perimeters/burn areas within the upper Boise River basin in central Idaho from 1992-2003
  • Data 
  • GIS Data and Metadata - Interior Columbia River Basin      
    • These GIS data are intended to assist users in understanding general patterns in bull trout distributions and inferring potential alteration of these distributions with changes in future mean annual air temperatures.  The data have been developed specifically for bull trout and are not intended for use with other aquatic organisms unless similar linkages with air temperatures can be established.  The data are most appropriate for broad scale displays and inference (i.e., map scales ~ 1:1,000,000) and should not be applied at finer scales, where local conditions may cause significant deviations from model predictions.  The lower limit bull trout model predicts accurately (R2 = 0.74) across the Columbia River basin, but has a RMSE of 180 m (RMSE = average elevation difference between observed values and model predictions). This amount of error would easily obscure meaningful inference when applied to an individual stream.
TitleDescriptionDownloads Available Up Request from Project Contacts
Documentation and MetadataDetailed process steps, documentation and metadata Documentation and Metadata
Model DevelopmentFish sample and air temperature table data Model Development
WatershedsStudy area hydrologic units (HUCs)Watersheds
Lower Limit FilesLower limit elevation line files Lower Limit Files
Stream LinesTauDEM stream lines Stream Lines
Intersection TablesLower limit elevation and stream line intersection tables Intersection Tables

Multiple Regression Models

Data or information used from this page should be cited as: Isaak, Dan; Rieman, Bruce; Horan, Dona. 2009. A watershed-scale monitoring protocol for bull trout. Gen. Tech. Rep. RMRS-GTR-224. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. 25 p.

Key Personnel

Science Contact

Geospatial Data Contact

  • Person
    David E. Nagel

    David E. Nagel

    Physical Scientist/Spatial Analyst
    https://research.fs.usda.gov/about/people/david.nagel
Last updated July 7, 2023
Return to top