Development of a broad landscape monitoring system using hyperspectral imagery to detect pest infestation

Abstract

Comprised of 28 million hectares in 2007, the United States (US) maize agricultural landscape was valued at $52 billion. Any threat to the continued dependability of this crop, now important as both a food and fuel source, becomes an important economic and ecological resource management factor for the global economy. Many genetically modified maize varieties contain toxins that target insects, reducing insecticide applications with positive ecological and human health consequences. Crop monitoring is required by US law to manage the development of insect resistance. To assess the onset of resistance, a proactive monitoring system must be able to identify maize infestation across the broad agricultural landscape. Non-destructive monitoring tools are necessary at a scale and definition required to discern insect infestation effects. A joint USEPA and NASA hyperspectral imagery and decision support system project has been successfully evaluated for its ability to distinguish insect infestations in several locations.