ASPRS 2006 Annual Conference Reno, Nevada  May 1-5, 2006 STRATEGIES FOR ALTERNATE APPROACHES FOR VEGETATION INDICES COMPOSITING USING PARALLEL TEMPORAL MAP ALGEBRA Bijay Shrestha Charles G. O’Hara Nicolas H. Younan GeoResources Institute Mississippi State University ERC 2, Research Blvd. Starkville, MS 39759. bijay@GRI.MsS-tate.edu cgohara@GRI.MsState.edu younan@ECE.MsState.edu ABSTRACT Remotely sensed images from satellite sensors such as MODIS and AVHRR provide high temporal resolution and wide area coverage. Unfortunately, these images frequently include undesired cloud and water cover. Areas of cloud or water cover preclude analysis and interpretation of terrestrial land cover, vegetation vigor, and/or analysis of change. Multi-temporal image compositing techniques may be employed to create a synthetic cloud free image that includes representative values derived from a set of possibly cloudy satellite images collected during a given time period of interest. However, spatio-temporal analytical processing methods that utilize moderate spatial resolution satellite imageries with high temporal resolution to create multi-temporal composites are data intensive and computationally intensive. Therefore, a study of the compositing strategies using high performance parallel solutions based on their computation and IO characteristics is required. This research focuses on analyzing alternate compositing strategies for vegetation indices using parallel temporal map algebra. The report provides objective findings on computational expense, IO complexity, and the relative benefits observed from various analysis methods and parallelization strategies. INTRODUCTION Remotely sensed images from satellite sensors such as MODIS and AVHRR provide high temporal resolution and wide area coverage, and therefore are highly useful in performing land use analysis. Unfortunately, these images frequently include undesired cloud and water cover. Areas of cloud or water cover preclude analysis and interpretation of terrestrial land cover, vegetation vigor, and/or analysis of change. Multi-temporal image compositing techniques may be employed to create a synthetic cloud free image that includes representative values derived from a set of possibly cloudy satellite images collected during a given time period of interest (Cihlar et al., 1994). This study is basically concerned with vegetation indices. Therefore, all the references to composites would implicitly refer to composites of vegetation indices. Spatio-temporal satellite image analysis is a technique for monitoring spatial and temporal changes of land cover and oceanic locations on earth. Temporal Map Algebra (TMA) is a novel technique developed by Jeremy Mennis and Roland Viger for analyzing a time series of satellite imagery using simple algebraic operators that treats time series of imagery as a three-dimensional data set, where two dimensions encode planimetric position on earth surface and the third dimension encodes time (Mennis and Viger, 2004). The high dimensionality of raster data leads to high computational cost, which is why parallel computation is attractive. This paper describes the design, implementation, and performance evaluation of parallel compositing of vegetation indices derived from MODIS datasets using TMA.