Abstract: The USDA-NASS has developed a system to use medium resolution (Landsat TM) satellite imagery and Agency collected field data to produce the Cropland Data Layer (CDL) every year, a classified image showing cropland land use. This product has been released into the public domain since crop year 1999. It details the acreage of major crops of rice, cotton and soybeans grown throughout the state. ArcGIS® was utilized to create a multiyear crop overlay of the cotton land use. The overlay shows the cumulative cotton acreage for the 7 year period. The range and extent of the cotton growing region is defined and researched in this paper. We have also used the CDL to examine field-level land use for all crops. For example, in Bolivar County we have compared crop rotation patterns for cotton over the 1999-2005 period for 15,203 fields. Pixel counts for each classification type allowed crop assignments per field to be made automatically. Thus the CDL product is not only used to define the spatial extent of the cotton growing region, but to help describe the actual cropping practices and improve crop estimates. By reducing the classified imagery to a multi-year database on a field level we can use improved algorithms to enhance estimates, for example using multiyear regression to compare pixel counts with actual acres.

Keywords: Mississippi, remote sensing, agriculture, cotton, crop rotation.