You are here:
Mapping Cropland and Major Crop Types Across the Great Lakes Basin Using MODIS-NDVI Data
Citation:
Shao, Y., R. S. LUNETTA, J. Ediriwickrema, AND J. S. IIAMES. Mapping Cropland and Major Crop Types Across the Great Lakes Basin Using MODIS-NDVI Data. PHOTOGRAMMETRIC ENGINEERING AND REMOTE SENSING. American Society for Photogrammetry and Remote Sensing, Bethesda, MD, 75(1):73-84, (2010).
Impact/Purpose:
Healthy Communities and Ecosystems - by providing new approaches to characterize landscape features, conditions, and change.
Description:
This research evaluated the potential for using the MODIS Normalized Difference Vegetation Index (NDVI) 16-day composite (MOD13Q) 250-m time-series data to develop a cropland mapping capability throughout the 480 000 km2 Great Lakes Basin (GLB). Cropland mapping was conducted using a two-step processing approach that included an initial differentiation of cropland versus non-cropland and subsequent identification individual crop types. Training data for individual crop type identification were developed using the National Agricultural Statistics Service (NASS) progress report data, MODIS-NDVI cluster analysis, and MODIS-NDVI derived phenology data. Three classification algorithms were examined to support cropland mapping, including a maximum likelihood, decision tree, and multi-layer perceptron neural network classifiers. The classification results were compared with the NASS agricultural statistics data to assess relative performance. Results indicated that the neural network classifier produced best overall performance for mapping cropland resulting in an overall coefficient of determination (r2) of 0.94 (SE = -2.46 and RMSE = 77.70). However, there were significant variations in performances across the study region. A stratification of the GLB by ecoregion was found to significantly improve the results of individual crop type classifications. A validation using the NASS county level census data indicated that the MODIS-NDVI classification generated significantly better results for the identification of soybean (r2 =0.94) and corn (r2 =0.89) than hay (r2 = 0.83) and wheat (r2 = 0.82).