Data obtained from 59 Landsat scenes were used to estimate the areal extent of irrigated agriculture over the High Plains region of the United States for a ground-water flow model. This model provides information on current trends in the amount and distribution of water used for irrigation. The analysis and reduction process required that each Landsat scene be ratioed, interpreted, and aggregated. Data reduction by aggregation was an efficient technique for handling the volume of data analyzed. This process bypassed problems inherent in geometrically correcting and mosaicking the data at pixel resolution and combined the individual Landsat classification into one comprehensive data set.