Location: Temperate Tree Fruit and Vegetable Research
Project Number: 2092-21220-003-032-S
Project Type: Non-Assistance Cooperative Agreement
Start Date: Aug 1, 2024
End Date: Aug 31, 2025
Objective:
Objective 1; Use satellite imagery to estimate “greenness” across landscapes.
Objective 2; Build forecasting models that predict vector migration into potato.
Approach:
Objective. 1; In recent years, availability of real-time satellite imagery for landscape analyses has grown rapidly. For example, the MODIS (Moderate Resolution Imaging Spectroradiometer) instrument has been installed in several NASA satellites and provides daily satellite imagery throughout the US. We will gather MODIS satellite imagery across the PNW landscapes where we monitored leafhoppers and psyllids since 2020 (four years of data). We will use the Google Earth engine to process daily satellite images and calculate NDVI (normalized difference vegetation index), which is a widely-used metric for quantifying the health and density of vegetation from sensor data. NDVI is a metric than can assess “greenness” of a landscape and can indicate when patches of habitat are turning from green to brown.
Objective 2; Once we have processed satellite imagery for the past four years, we will build models that can estimate the migration of psyllids into potato fields. We will use the package MigClim in R, which was designed to predict dispersal of organisms across spatially-fragmented landscape. To build models, we will assume that patches with low NDVI values (indicating brown vegetation) are likely sources of migrating leafhoppers or psyllids, while patches with high NDVI values (indicating lush green irrigated crops) will serve as sinks. Our models will also consider how factors such as wind direction and temperature affect vector dispersal, given that most movement should occur in the direction of prevailing winds and more vectors will disperse with higher ambient temperatures. We will validate our models with 5-years of monitoring data we have already obtained on population dynamics of both vectors and pathogens in potato crop fields. Once models have been built and validated, we will incorporate predictions into our decision aid system (potatoes.decisionaid.systems). Our expected outputs will include maps that show areas of high dispersal potential, which will be linked with recommendations that growers can follow to better anticipate when vectors will arrive in their fields.
