Document Type

Thesis - University Access Only

Award Date


Degree Name

Master of Science (MS)

Department / School



Variability in plant phenology has been studied for centuries. Early studies relied almost entirely on ground observations and often focused on a single plant species. Although phenology networks improved the spatial coverage of this approach, it remained limited to small area studies. The ability to monitor vegetation conditions over large areas became possible with the advent of satellite remote sensing. Regional phenology data is important in ecosystem model and coupled biosphere/atmosphere simulations. The precision of such modeling efforts will have tremendous impacts on the accuracy of assessing global climate change. This research utilized data from the Advanced Very High Resolution Radiometer (AVHRR) sensor, mounted aboard meteorological satellites, to identify vegetation characteristics and relate them to variability in the regional climate record. A Normalized Difference Vegetation Index (NDVI) time-series of data (1990-1998) was used to derive seasonal characteristics for sixty-eight sites in the northern and central Great Plains. Using a stepwise linear regression approach, the e data were correlated with climate variables obtained from the Automated Weather Data Network (A WDN) for the same region. A significance level of 0.05 was necessary for a particular variable to remain in the model. Land cover type were distinguished based on the Seasonal Land Cover Regions (SLCR) data base available for the contiguous U.S. The purpose of this research was to identify which climate variables most influenced a particular suite of seasonal metrics. Results showed two metric variables, start of season time and time integrated NDVI, to be most consistently explained by climate. Furthermore, variations in precipitation and temperature represented the mo. t significant controls in the productivity of both grasslands and croplands. Although this research focused primarily on the environmental controls influencing productivity, it also identifies the need for more ancillary data in assessing cropland management practices. This study suggests that these methods are viable for assessing the influence of climatic variation on vegetation dynamics. The ability to provide more accurate inputs to climate models will assist in better understanding the relationship between the landscape and atmosphere.

Library of Congress Subject Headings

Vegetation monitoring -- Great Plains -- Remote sensing Vegetation and climate -- Great Plains Climatic changes -- Great Plains



Number of Pages



South Dakota State University