The following datasets are climate projections for the state of Alaska based on downscaled outputs from five IPCC Global Circulation models. They include average monthly air temperature (degrees Celsius) and total monthly precipitation (millimeters; SWE) grids for the state of Alaska, as well as additional grids derived from these data.

These datasets are formatted as zipped folders, which when unzipped yield delimited text files for ASCII (GIS) use. Each file contains grids for the entire state for the years 1980-2099. The three emission scenarios (A2, A1B and B2) refer to the IPCC emission scenarios used in our modeling, where A1B is the midrange scenario, A2 is more pessimistic (higher emissions) and B1 is more optimistic (lower emissions).

Due to the fine resolution of the maps, these zipped files are very large. Please allow adequate download time. We are working to make smaller subsets of these datasets and more user-friendly file types available. If you would like to browse these maps, we suggest viewing them in our GoogleEarth format. If you need a small subset of the data or are having trouble downloading, please contact us at ffnlf@uaf.edu

Historical datasets are derived from Climate Research Unit (CRU) data and include years 1901-2008 for temperature and 1901-2006 for precipitation (coming soon). They are downscaled to 2km grid cells.

Future projections are derived from General Circulation Models (GCMs) used by the Intergovernmental Panel on Climate Change (IPCC). These datasets include years 1980-2099 and are downscaled to 2km grid cells.

The 5 models available have been ranked as the best (out of 15 used by the IPCC) models for Alaska. Of these five, the rank of the models for overall accuracy in Alaska and the far north was as follows:
1)ECHAM5
2)GFDL21
3)MIROC
4)HAD
5)CCCMA
The five-model composite uses mean values from the outputs of all five models.

As listed below, there is one set of temperature data and one set of precipitation data available for each model plus the 5-model composite under each of the three climate scenarios. For example, had.prec.a1b.zip refers to the down-scaled precipitation data generated by the Hadley model.

Derived layers are defined as follows:
Date of thaw and date of freeze were derived by creating a running average of mean monthly values and interpolating on what ordinal day these values crossed the freezing point. Length of growing season is the number of days between these two derived dates. Air temperature and precipitation refer to decadal mean monthly values, e.g. the mean value of May temperatures for 2030-2039. All derived values are based on the five-model composite. Standard deviations reflect the variability between the five model outputs.

All data should be projected as Alaska Albers Equal Area Conic with the North American Datum 1983.

Please contact us at ffnlf@uaf.edu if you have questions or encounter any problems.