Skip to main content
ARS Home » Southeast Area » Mississippi State, Mississippi » Crop Science Research Laboratory » Genetics and Sustainable Agriculture Research » Research » Publications at this Location » Publication #170023

Title: REMOTE SENSING, SAMPLING AND SIMULATION APPLICATIONS IN ANALYSES OF INSECT DISPERSION AND ABUNDANCE IN COTTON

Author
item Willers, Jeffrey
item McKinion, James
item Jenkins, Johnie

Submitted to: Symposium Proceedings
Publication Type: Proceedings
Publication Acceptance Date: 3/1/2005
Publication Date: 7/1/2006
Citation: Willers, J.L., McKinion, J.M., Jenkins, J.N. 2006. Remote sensing, sampling and simulation applications in analyses of insect dispersion and abundance in cotton. USDA Forest Service Proceedings RMRS-P-42CD. p. 879-885.

Interpretive Summary: A computer simulation model was employed to analyze the impact of pest density and search area size on how pests may spatially distribute themselves. The results indicate that assuming cotton insect pests are randomly dispersed within distinct crop habitats occupied at different characteristic densities is useful for commercial scouting and decision making purposes. However, large enough areas of a field must be searched. The computer findings were supported by field studies, which indicate large areas can be effectively searched with small investments of time if remote sensing images of cotton fields are available. The images can be processed into a map, which shows the different areas of crop growth in fields. Cotton scouts can use simplistic sampling methods within the regions of growth (or habitats) shown on the map. The pest density estimate for a habitat stratum geographically applies to each class of growth shown on the map if conditions such as planting date, variety, or other agronomic influences are also similar.

Technical Abstract: Simulation was employed to create stratified simple random samples of different sample unit sizes to represent tarnished plant bug abundance at different densities within various habitats of simulated cotton fields. These samples were used to investigate dispersion patterns of this cotton insect. It was found that the assessment of spatial pattern varied as a function of sample unit size and was independent of large pest densities. Using this knowledge, it was demonstrated how remote sensing assists field scouts in estimating pest abundance and the pattern of dispersion within commercial cotton fields. For the majority of pest management decisions in commercial fields, both the simulation data and field results supported the robust assumption that cotton pests are randomly dispersed at different densities within homogenous habitats. Therefore, it was possible to estimate the boundaries of different pest densities by mapping habitat variability within cotton fields using high resolution, geo-referenced, remotely sensed imagery. Once boundaries for various densities of the pest were established, site-specific prescriptions of pesticides were implemented, reducing pest management costs and the amount of pesticide applied to the environment.