An assessment to derive the essential dates of AWiFS and MODIS 16-day NDVI composite data for the identification of corn and soybean fields was conducted by the USDA’s National Agricultural Statistics Service (NASS). The goal was to determine the essential dates of AWiFS and MODIS data needed for the separation of corn and soybean crops over twelve discrete study areas. Site selection was based on the availability of ground truth, concentrations of corn and soybean acreage, and the availability of spring and summer AWiFS and MODIS data. Ground truth was provided by the USDA’s Farm Service Agency (FSA). Single date May, June, July and August AWiFS scenes and spring/summer MODIS data were obtained. Eleven classifications were produced for each study site. Initial classifications were generated using AWiFS data alone. Corn and soybean accuracies were compared for all classifications produced using four dates (May, June, July and August) of AWiFS data followed by three (May, June, July), two (May and June) and one month (May) of AWiFS data. A second series of classifications were produced using the same dates of AWiFS data combined with MODIS data. Corn, soybean and overall accuracies were compared. Remote sensing estimates are more valuable to NASS the earlier they can be accurately produced. The goal of this assessment was to determine if corn and soybean fields could be identified with a degree of accuracy such that estimates for these major commodities could be provided to the NASS Agricultural Statistics Board (ASB) by an August 1 deadline.
[1]
Russell G. Congalton,et al.
Assessing the accuracy of remotely sensed data : principles and practices
,
1998
.
[2]
C. Brodley,et al.
Decision tree classification of land cover from remotely sensed data
,
1997
.
[3]
Rick L. Lawrence,et al.
Classification of remotely sensed imagery using stochastic gradient boosting as a refinement of classification tree analysis
,
2004
.
[4]
J. Ross Quinlan,et al.
Bagging, Boosting, and C4.5
,
1996,
AAAI/IAAI, Vol. 1.
[5]
R. DeFries,et al.
Classification trees: an alternative to traditional land cover classifiers
,
1996
.
[6]
David M. Johnson.
A Comparison of Coincident Landsat-5 TM and Resourcesat-1 AWiFS Imagery for Classifying Croplands
,
2008
.
[7]
Claire G. Boryan,et al.
MULTIRESOLUTION LANDSAT TM AND AWIFS SENSOR ASSESSMENT FOR CROP AREA ESTIMATION IN NEBRASKA
,
2005
.
[8]
Limin Yang,et al.
Development of a 2001 National land-cover database for the United States
,
2004
.