Improving methods of landscape analysis for pest management in agroecosystems

Monday, November 11, 2024

8:36 AM – 8:48 AM MST

Location: Phoenix Convention Center, 128 AB, PCC

Presenting Author(s)

TM

Timothy Martin

Graduate Student
Texas A&M University
College Station, Texas

Co-Author(s)

TC

Thomas Chappell

Texas A&M University
College Station, Texas

Agroecological processes such as pest and allelic migration, disease spread, and chemical drift are spatial. The spatial configuration of sources and sinks relevant to these processes determines their influence on points in space. For example, the abundance of insect herbivores at the location of an emerging annual food source can be predicted based on the abundance and configuration of habitat surrounding the location. Spatial agroecological modeling can enhance precision pest management by translating GIS data into site-specific predictions, and depicting spatially distributed risk of disease transmission, pest migration, or biological invasion. Current methods useful for analyzing spatial processes in agroecosystems have been applied to identify unknown sources and sinks and to estimate dispersal ranges. These methods form a basis for developing quantitative models for prediction of pest abundance, and for empirically estimating dispersal kernels. Through a simulation approach that emulates differential insect dispersal from various habitat types in space, we demonstrate site-specific prediction of abundance, as would be conducted to estimate risk of infestation in agroecological context. Our simulation facilitates comparison with related approaches that focus on identifying source habitats, and that will enable us to add accounting for habitat patch size and fragmentation as additional factors influencing insect abundance and dispersal. The differential effects of habitat types on the variations in dispersal and phenology of pests is an important consideration that requires more investigation. Models are needed that represent these spatial processes quantitatively and realistically to render accurate and useful site-specific predictions in agroecosystems that vary in spatial configuration.