SNAPverse R package ecosystem development frozen.

Development of the SNAPverse R package ecosystem has been frozen indefinitely. This project was never directly funded. I began it out of personal interest, but no longer work for SNAP. It will remain archived here for historical reference.

The snapclim package provides access to curated collections of public climate data sets offered by Scenarios Network for Alaska and Arctic Planning (SNAP) at the University of Alaska Fairbanks. snapclim interfaces with SNAP’s Amazon Web Services cloud storage to retrieve specific climate data. Available data includes historical observation-based climate data as well as historical and projected climate model outputs. Available climate variables include total precipitation and minimum, mean and maximum temperature.

Time and space

Regional climate summaries and climate data at point locations are available, stretching over Alaska and western Canada. Regions include the state of Alaska and several Canadian provinces, ecological regions, fire management zones, terrestrial protected areas under jurisdiction and management of various governmental agencies and more. Point locations include cities, towns, villages and other municipal units and locations of interest.

Daily, monthly, seasonal, annual and decadal temporal resolutions are planned for inclusion. At this time, available data include AR5/CMIP5 2-km downscaled outputs over Alaska and western Canada for 86 regions and 3,967 point locations at all but the daily temporal resolution. Please not that not all combinations of temporal and spatial resolution exist, e.g., daily point location climate projections.

SNAPverse context

snapclim is a member package in the data sector of the SNAPverse collection of R packages. Data packages typically include raw data sets in support of other R packages. snapclim is technically more like a typical R package. Instead of storing local copies of data sets, it contains functions for accessing external data sets that would be too large to store conveniently even in an explicit data package, especially considering that any given user session would likely utilize only a small fraction of the total available data. However, snapclim does not offer functionality beyond accessing data and is therefore still best conceptualized as a data package. Functions for statistical analysis and modeling of SNAP data are already encompassed in packages like snapstat.


You can install snapclim from github with:

# install.packages('devtools')