HR: 1340h
AN: IN43B-1190 [Abstracts]
TI: A Comparison of the Hypertemporal Crop Mapping Abilities of NDVI Products Derived from both MODIS Surface Reflectance and Calibrated Geolocated Radiance Data
AU: * Hilbert, K
EM: khilbert@iftd.org
AF: Institute for Technology Development, Building 1103, Suite 118, Stennis Space Center, MS
39529, United States
AU: Lewis, D
EM: dlewis@iftd.org
AF: Institute for Technology Development, Building 1103, Suite 118, Stennis Space Center, MS
39529, United States
AU: O'Hara, C
EM: cgohara@gri.msstate.edu
AF: GeoResources Institute, Mississippi State University, Mississippi State, MS 39762, United
States
AB:
This project examined the usefulness of Moderate Resolution Imaging Spectroradiometer (MODIS)-derived
hypertemporal, curve-fitted vegetation indices for agricultural mapping. The analysis focused on accurately
mapping corn and soybean crops cultivated during the 2004 / 2005 growing season within Argentina's Córdoba
and Santa Fe provinces—one of the world's most productive agricultural regions. Crop fields were selected
according to field size in order to allow for the spatial resolution constraints of the MODIS sensor. Hypertemporal
data sets were created by temporally filtering multiple dates of customized MODIS NDVI products. Hypertemporal
data sets provide the ability to track phenology growth curves of different crop types and therefore provide
discriminating temporal signatures as input for classification algorithms. Several supervised classification
algorithms were used to investigate the ability of MODIS data to identify crop types. Conclusions were drawn from
post classification confusion matrices derived from representative fields. Results will provide computational
methods for furthering the informatics and operational capability of the USDA Production Estimates and Crop
Assessment Division.
DE: 0402 Agricultural systems
DE: 0480 Remote sensing
DE: 0540 Image processing
DE: 6314 Demand estimation
SC: Earth and Space Science Informatics [IN]
MN: 2007 Fall Meeting