Open climate data and services to serve landscape modeling and decision making at the USGS: Past, present and future.

Friday, 19 December 2014
David L Blodgett, USGS Office of Water Information, Center for Integrated Data Analytics, Middleton, WI, United States, Nate Booth, USGS, Baltimore, MD, United States and Jordan I Walker, USGS Center for Integrated Data Analytics, Madison, WI, United States
High-resolution gridded data and model output is among the most massive information used in environmental analysis and modeling. Gridded historical weather and downscaled climate projections are now available for the Conterminous US at 800 meter monthly and 12 kilometer daily resolution and even finer resolution regionally. Combined, the high volume and unfamiliar file formats of these data make using it a challenge for all but the most determined or technologically savvy users. The U.S. Geological Survey’s (USGS) Center for Integrated Data Analytics (CIDA) in cooperation with the many federal, academic, and open-source software partners, has been working to make base datasets and useful summaries available in formats that are readily usable by scientists and managers familiar with GIS and modeling of landscape-phenomena.

When an analyst needs information such as decadal average growing degree day estimates for historical and future periods, she shouldn’t have to download and process terabytes of historical and projected data to produce a few summary values, or a simple map. A USGS project, known as the Geo Data Portal (GDP), has assembled a catalog of web-service available gridded climate datasets at the USGS, NASA, NOAA and several universities. GDP processing services provide model-ready spatially summarized gridded time series data for user-submitted polygons for any dataset in the catalog or any dataset published using supported open standards. Recently, progress has been made toward providing annual climate indices from monthly and daily data in common GIS formats.

Using the GDP system, a person can execute processing tasks that run on USGS servers and use custom datasets, statistic types, and statistic thresholds. This work has been made possible by numerous organizations committed to publishing software and data that scale well, use standards, and are freely available for anyone to use. A high-level overview of assembling the Geo Data Portal system’s capabilities will be presented, reflecting on technical and strategic difficulties experienced, and the future of open data and processing between the climate and landscape domains.