The relationship between primary production and precipitation in grasslands has long been an issue of great interest both to ecologists and managers who graze livestock and/or maintain conservation properties. Data from remote sensors are now obtained at higher spatial resolutions on a near daily basis providing new methods to explore broad scale relationships between grassland production and moisture availability. To promote remote sensing-based analysis of grasslands in North America, I developed the Prairie Spatial Database (PSD) for the northern Great Plains. To develop the PSD, I obtained geospatial coordinates and ecological data from conservation managers for 261 prairies. Sites were filtered for suitability for remote sensing studies, and 221 were selected for further analysis; representing a mix of shortgrass, mixedgrass, and tallgrass prairie types and different C3/C4 dominances. Using time series weather data (PRISM), I next derived Palmer's Drought Severity Index (PDSI) values and produced a time series of normalized difference vegetation index (NDVI) values for each site for 2000 - 2008 from the Moderate Resolution Imaging Spectroradiometer (MODIS). Mixed modeling techniques allowed me next to compare the influence of soil moisture on growing season time integrated NDVI (TINDVI) and maximum NDVI across prairie types. Selection of best-fit models highlighted the contingency of biomass production on soil moisture availability and a hidden relationship between TINDVI and soil moisture within tallgrass prairies.
Read
