Cotton management using crop modeling and precision farming technologies

Cotton (Gossypium hirsutum L) crop yields have been shown to vary within fields. Spatial variability of yield and plant growth may be caused by pests, weather, soil characteristics, and slope. Precision farming technologies such as Global Positioning System (GPS) and variable rate technology (VRT) allow for site-specific management of crops. With the combination of soil characteristics, plant growth monitoring, and accurate measurements of crop yield, improved site-specific management strategies may be found compared to whole field management.Cotton plant mapping allows for a continual record of production of vegetative and fruiting sites throughout the season. One system commonly used to acquire such data is the COTMAN Expert System. This system involves plant mapping and current and long-term weather information to determine plant growth; applications of plant growth regulators (PGRs), insecticides, harvest aids; and harvest scheduling. The overall objective of this thesis was to evaluate management options for cotton production using the correct amount and optimal timing of certain inputs. Specific objectives for this project were to (i) evaluate crop physiological development in management zones that were developed from soil characteristics and slope, (ii) compare insect scouting and the COTMAN model for specific decision management of each zone versus the whole field management, and (iii) evaluate yield consistency across years in each management zone.COTMAN data were collected in four managment zones delineated by soil mapping units for two consecutive years (2000 and 2001). These zones contained precision plots, where management could be adjusted if necessary in response to COTMAN and other analyses, and plots using whole field management. Precision and whole-field management plots were sampled for insects and COTMAN analysis weekly. Along with COTMAN data, insect scouting, nodes above cracked boll (NACB) information, and various boll measurements, (mainly number and average weights), were used as precision tools to determine specific management versus whole field management.Yields were significantly different within zones in 2000, and yield differences in plots in 2000 and 2001 were observed. No differences in crop management were implemented between precision and field managed plots in 2000. In 2001, precise management in one of the four management zones resulted in lower yield, while another resulted in increased yield compared to the adjacent field management plots. Overall, coefficients of variation data suggests that zone 1, poorly drained zone, consistently had the highest amount of yield variability, and zone 4 had the lowest amount of yield variability. Further refinement of management zone delineation with the use of precision farming technologies should produce a better understanding of yield variability.