LANDSAT AUTOMATED SCENE SELECTION TOOLBOX

Abstract

The manual web-based download and use of United States Geological Survey (USGS) Landsat imagery to perform raster-based geoprocessing analysis is often time-consuming and repetitive in nature. Searching, sorting, and downloading images to cover an area of interest is overly complicated and error-prone due to the naming conventions of the output files and the excellent level of data management skills required. The purposes of this project are to automate web-based searches based upon date and coverage area and the download of Landsat imagery so that the Geographic Information Systems (GIS) analyst can focus on geoprocessing analysis and spend less time on data management. The solution to this problem is in the form of an ArcGIS toolbox developed and written in Python. Inputs to the toolbox include a selected feature (point, line, or polygon), date of interest, percent maximum cloud coverage, and a buffer length used to calculate an extent that buffers the input feature. Given the user’s input, the toolbox searches the online Landsat USGS database for the date and study area of interest to locate the necessary Landsat imagery. The toolbox output is the downloaded Landsat imagery in the form of an ArcGIS mosaic dataset that encompasses the user’s input extent. The toolbox typically completes this automated process within three to five minutes as compared to a manual process that may take hours if not days, depending on the input extent complexity. This toolbox will provide time-saving benefits to any analyst interested in utilizing Landsat imagery in their geoprocessing analysis.