Phenological Parameters Estimation Tool (PPET)

About this technology

Overview

The Phenological Parameters Estimation Tool (PPET) is a set of algorithms implemented in MATLAB that estimates key vegetative phenological parameters. For a given year, the PPET software package takes in temporally processed vegetation index data (3-D spatio-temporal arrays) generated by the Time Series Product Tool (TSPT) and outputs spatial grids (2-D arrays) of vegetation phenological parameters. As a precursor to PPET, the TSPT uses quality information for each pixel of each date to remove bad or suspect data, and then interpolates and digitally filters the time series to produce a continuous, smoothed vegetation index product in which data voids are eliminated. Both the TSPT and PPET use Moderate Resolution Imaging Spectroradiometer (MODIS) satellite multi-spectral data as a default, but each software package is easily modifiable and could be used with any high-temporal-rate remote sensing system that is capable of producing vegetation indices.Unlike other known plant phenological parameter estimation software, the PPET produces not only common phenological parameters but also real-time and custom parameters without a priori assumptions about the shape of the phenological cycle. Common phenological parameters, like those produced in PPET, are associated with the annual vegetation growth cycle. They quantitatively describe vegetative states related to annual cyclical growing seasons, such as green-up, maturity, senescence, and dormancy, by analyzing the temporal shape of given vegetation index time series. The real-time phenological and custom parameters are formed from a cumulative sum (integral) produced at a fixed temporal interval. In addition, a cumulative vegetation index and time-specific/pest-specific phenological parameters can be designed to optimize the detection of vegetation damage from specific pests and diseases. These problem-specific phenological parameters have the potential to be integrated into near real-time, predictive surveillance systems (i.e. early warning systems) and, with improved vegetative state information, could assist decision makers in making intelligent vegetation and associated land resource management choices.MATLAB, MATLAB Runtime Library and ERDAS IMAGINE are required to run the software.