Developing a near real-time system for monitoring the foliar phenology of the terrestrial biosphere

Foliar phenology significantly influences the exchanges of energy, mass and momentum between the land and the atmosphere. The timing of phenological events such as the flushing or senescence of leaves is highly variable in space and time and these events are very sensitive to changes in climate. Computer models have been developed to predict foliar phenology in response to the prevailing climatic conditions but none of these models were sufficiently generalized to provide estimates of phenology anywhere on the globe. The work of this dissertation sought to bridge that gap by developing a generalized phenology model that could adequately predict the time course of vegetation foliar development and maintenance throughout the season. This model adequately predicted the timing and duration of foliage as compared to satellite-derived Normalized Difference Vegetation Index values.; I then coupled this model with gridded surface weather observations produced using statistical interpolation techniques to create a system that predicts foliar phenology anywhere within a region of interest. The gridded surface weather data is generated as part of a near real-time system which automatically retrieves point-source weather data, stores these data in a relational database, and interpolates the data onto a user-specified grid. I performed extensive methodological crossvalidation of interpolated results to ensure that these datasets are accurate and unbiased.; Gridded phenology predictions are produced automatically each day using the surface weather data and served to the public over the World Wide Web. This interface provides the ability to extract a point of interest from the gridded dataset as either an ascii text file or an automatically generated graph in addition to creating spatial maps of vegetation foliar status throughout the entire landscape.