Stream Condition Databases

Hi Everyone,
Just sending word that the stream temperature database and scenarios for the Central Valley & remaining parts of California are now available on the NorWeST website ( Also attached to this email is a poster showing the statewide stream thermalscape that’s now been interpolated from data at 2,663 sites that were contributed by biologists and hydrologists working for 15 state, federal, tribal, university, and private resource organizations. The NorWeST project is a grassroots effort to develop an open access, comprehensive stream temperature database and set of high-resolution climate scenarios for the American West. The database currently houses >200,000,000 hourly records from >20,000 unique sites that were contributed by >100 agencies & resource groups. NorWeST was funded for California by the USFWS & the California Fish Passage Forum (, with complimentary grants to do the project elsewhere from the Landscape Conservation Cooperatives, NASA, NFWF, USFS, & USGS.

NorWeST data products consist of three basic things (described below), all of which can be summarized, displayed, or queried in ArcGIS and other digital formats for ease of use. You can also view & query the stream temperature scenarios dynamically from your desktop using this nifty GoogleMap tool: (zoom in until the streams appear).

Geospatial products:
#1) ArcGIS shapefiles of mean August stream temperature predictions from an accurate model fit to all the data (r2 ~ 0.90; Average prediction error ~ 1.00°C; Average observed error ~ 0.70°C) and used to develop a consistent set of historical and future climate scenario maps at 1-kilometer resolution. The historical scenarios consist of composite averages over multi-year periods (Scenarios 1 & 2), each individual year’s mean August stream temperature (from 1993 ~ 2015), and 10 different future scenarios (Scenarios 23 – 32; metadata describing the scenarios and stream temperature model are available at the project website). The NorWeST stream temperature model predicts an August mean temperature because it’s the one month that’s been monitored most consistently across the region (which allowed us to use the largest proportion of everyone’s data in the model), because August is a thermally stressful period for many fishes, and because a monthly mean can be easily linked to outputs from global climate models to create an integrated system for downscaling climate effects to local stream temperatures.

#2) ArcGIS shapefile showing the precision associated with the stream temperature model predictions (Scenario 22), which is useful for identifying areas in streams that are not redundant with existing temperature measurements and could aid in designing efficient monitoring strategies.

#3) Daily summaries (min/max/mean) of all temperature data (including non-August days) and georeferenced locations of monitoring sites to facilitate coordinated monitoring efforts and new temperature research. Note that we can only distribute those data we are given permission to distribute, but the great majority of people (>95%) do give their permission.