Can pesticide use be reduced by managing landscape complexity?

Pesticide use reduces crop losses but can result in significant negative externalities. To reduce pesticide use, farmers might take advantage of ecosystem services like pest suppression provided by agricultural landscape complexity. I hypothesize that temporal and spatial landscape complexity will reduce pest pressure, lowering pesticide application rates. I develop measures of landscape complexity using primary productivity calculated from remotely sensed data to detect patterns of vegetation diversity from 2008 - 2012. I incorporate this information and important covariates in an econometric model describing pesticide use in Midwestern and Southeastern United States. Results indicate that landscape pattern metrics reflecting composition, configuration, and connectivity influence application rates at various points during the growing season. Inclusion of these metrics strengthens model function significantly, increasing adjusted R2 from 0.66 to 0.79. Additional information regarding landscape effects could improve farmers ability to reduce risk by controlling harmful pests while reducing the need for pesticide application.