Bioclimatic Predictors for Supporting Ecological Applications in the Conterminous United States
[Folder View]
Abstract: The data in this data series represent a set of multi-band rasters, each containing 20 bioclimatic variables for the continental United States for the years 1980 - 2009. Staff at the U.S. Geological Survey, Fort Collins Science Center developed these raster layers using 2-km time series data created by Climate Source, Inc. and Spatial Climate Analysis Service (SCAS) at Oregon State University. The original climate data for these rasters came from multiple climate station point measurements. Using these station measurements, Climate Source/SCAS produced interpolated climate grids which became the inputs in the algorithm used by the USGS to derive the bioclimatic variables. The values within each raster band were produced using an algorithmic calculation that is performed on a cell-by-cell basis. Calculations consider only the relevant inputs for the individual raster cell in question. The climate inputs used to calculate these products constitute an improvement upon the original 4-km PRISM climate data developed by the PRISM Climate Group at Oregon State University. Climate Source/SCAS applied a Gaussian filter to increase the resolution of the original grids from the base resolution of 4 kilometers to 2 kilometers. For details on this process, see 'Processing Notes.' Staff at the U.S. Geological Survey, Fort Collins Science Center purchased this higher resolution data from Climate Source/SCAS and maintain the data in-house (inquiries about the data/data derivatives should be directed to the USGS Fort Collins Science Center Information Science Branch GIS group). The data sets derived from this data represent a higher value product because of the higher resolution inputs used to calculate them. Climate data can be represented and disseminated in different ways: as time-series data or as normalized data. Time-series data represent static recordings for a particular period (i.e., average highest temperature for a particular month from a particular year). For example, time-series climate data for the month of January over a 10 year period will consist of 10 separate data sets-- one for each year. Normalized data (or 'normals') represent averaged climatic measurements for a specified period (i.e., the average highest temperature for a particular month is recorded every year for the time period in question, and all of those values are subsequently averaged-- or normalized). For example, normalized climate data for the month of January over a 10 year period will consist of only one data set. The difference between time-series and normalized climate data should be noted and taken into consideration when using and interpreting climate data sets. *THE INFORMATION CONTAINED IN THIS DATA SERIES WAS DERIVED FROM TIME SERIES CLIMATE INPUTS.*