We developed an irrigation mapping methodology that relies on remotely sensed inputs from the MODerate Resolution Imaging Spectroradiometer (MODIS) instrument, globally extensive ancillary sources of gridded climate and agricultural data and on an advanced image classification algorithms. The methodology involves four major steps. First, we use climate-based indices of surface moisture status and a map of cultivated areas to generate a potential irrigation index. Second, we identify temporal and spectral signatures in the remotely sensed data that are associated with presence of irrigation. Here, the temporal indices are based on the difference in annual evolution of greenness between irrigated and non-irrigated crops, while spectral indices are based on the reflectance in the green and near-infrared and are sensitive to vegetation chlorophyll content associated with moisture stress. In the third step, we combine the climate-based potential irrigation index, remotely sensed indices, and learning samples within a decision tree supervised classification tool to make a binary irrigated/non-irrigated map. In the last step, we apply a tree-based regression algorithm to derive the fraction of irrigated area within each pixel that has been identified earlier as irrigated. Application of the procedure to the continental US produced an objective and comprehensive map of irrigated areas for the year 2001. The map exhibits expected patterns of irrigation across the continental US such as a strong east-west divide with most irrigated areas concentrated in the arid west along dry lowland valleys. Qualitative assessment of the map across different climatic and agricultural zones reveals a high quality product with sufficient detail as compared to existing large area irrigation databases. Accuracy assessment indicates that the large area map is highly accurate in the western US but less accurate in the east. Comparison of area estimates made with the new procedure to those reported at the state and county levels show a strong correlation with a small bias with an estimated RMSE of 2,500 square kilometers, or little over two percent of the total irrigated area in the US. The future application of the new procedure at a global scale may require a better potential irrigation index as well as the use of remotely sensed skin temperature measurements.