Since the launch of the first earth observation satellites, monitoring and mapping of U.S. croplands has been a primary goal of many data users. The advantages of using low spatial and high temporal resolution data are i) as the temporal frequency becomes finer, the ability to monitor phenological change of crop plants during growing season increases and ii) large area crop mapping is available to identify specific crop planting/harvesting areas and to estimate crop yields before harvesting. The objective of this study was to investigate the potential of 500-m MODIS (Moderate Resolution Imaging Spectroradiometer) data in estimating corn and soybean areas for the dominant production areas of the United States. The MODIS annual metrics were generated by land data standard product (MOD09). To avoid cloud cover, the MODIS daily acquisitions were converted to 32-day composites covering March 2002 to February 2003 resulting in 11 roughly monthly composites. The USDA NASS 2002 Cropland Data Layers (CDL) based on Landsat were used to provide sub-pixel training for the 500-m MODIS data. For regression tree analysis, the S-Plus statistical package was used. Based on the regression tree models, corn and soybean area mapping using 500-m MODIS has been conducted for the dominant production areas. The validation was done at the national, state, and county levels using county-level statistics from the NASS 2002 Census data. When the MODIS estimates were compared with NASS Census, r² of corn, soybean, and corn and soybean areas were 0.957, 0.949, and 0.984 in state level, respectively. As a result, the estimates of corn and soybean areas by 500-m MODIS metrics were significantly close to NASS Census data.