This research compares three different classification algorithms for mapping crops in Pinal County, Arizona, using both present and historical image data. The study area lacked past crop maps, and farmers were dealing with the risk of evolution of resistance to insecticides in the whitefly, a global pest of cotton, fruits, and vegetables. The ability to create historical crop maps without concurrent training data is an invaluable tool for historical integrated pest management research. Comparison of maximum likelihood, object-oriented, and regression tree classifiers was done with Landsat Thematic Mapper imagery and high quality crop maps. Classification outputs for the three years in this research all achieved overall accuracies above the traditional benchmark of 85%.Comparison of the classification results shows that the classification and regression tree technique clearly outperformed the other classifiers. Using training data from one year and applying that data to another year for classification purposes demonstrated that overall accuracies from 71% to 83% are achievable, although accuracies were consistently greater than 85% for some crops.

Bt cotton is used to control pink bollworm, Pectinophora gossypiella (Saunders), in much of the southwestern U.S. Where growers choose non-Bt varieties, early fruiting (squaring) Bt cotton might be used as a trap crop for pink bollworm control. Field tests were conducted in 1997 and 1998 to determine if early planting, varietal selection, and very low planting rates could increase early squaring. Cotton planted 4 April produced 6 mm squares approximately 11 days earlier than cotton planted 7 May. Low plant populations (4.3, 2.5 plants/m) did not result in more early squares, compared to recommended populations with 7.5 plants/m. Very dense populations with 18.4 vs. 9.4 plants/m produced more early squares/m despite fewer squares/plant. However, yield was reduced with planting rates above 7.5 plants/m. Some varieties produced more early-season squares than others, but differences were not always consistent. Paymaster 1215 produced significantly more early squares compared to two other short-season varieties (Deltapine 20B and Deltapine 50B) and the standard Acala 1517-95 in 1997, but not more than Acala 1517-95 in 1998. A mid-season variety (Paymaster 1560) produced the highest number of early squares in a comparison of ten Bt varieties in 1998. A non-Bt very short-season variety, Xpress, produced the earliest squares in 1997, more than 11 days earlier than Paymaster 1215 (a short-season variety), indicating that earlier squaring varieties of Bt cotton could be developed. These results indicate that early planting will provide the earliest squares but that variety and planting rate should be considered in developing a trap crop for pink bollworm control. Planting rate should be at least 7.5 plants/m but if yield is not a consideration could be as high as 18 plants/m.