An official website of the United States government.
<ol> <LI> Validate the Spatially Explicit Russian Wheat Aphid Density (SERD) Model on winter wheat, using both artificially infested and naturally infested treatments.<LI> Validate the SERD model for Russian wheat aphid on spring irrigated barley, using both artificially infested and naturally infested treatments.<LI> Generating loss functions between Russian wheat aphid density data and harvested wheat and barley yield.<LI> Reparameterize the SERD model using previously collected data combined with current data including the new variables, Crop Type and Infestation Level.<LI> Create a software product for use by pest managers in the forecasting of within-field Russian wheat aphid densities.
Non-Technical Summary: Russian wheat aphid is a key pest of winter wheat. Scouting to determine the need for control is difficult because of cost and large acreages. The purpose of this study is to validate a computer model that replaces conventional scouting by predicting the need to treat Russian wheat aphid based on satellite images and weather data. <P> Approach: Objective 1. Landsat 7 ETM+ imagery will be collected for each site in both the winter (approximately December 1st of each season), and the spring (approximately April 1st). A USGS 30-meter grid Digital Elevation Map for Colorado, which produced the Slope surface is already in our GIS, detailed soil survey maps will be produced where necessary by digitizing existing hard-copy soil surveys; registering, rectifying, and converting to the appropriate map projection. Determine Russian wheat aphid (RWA) densities in field plots at Lamar and Akron, Colorado, near Allison-Pike suction traps. SERD model predictions will be generated for each treatment plot for each sample period. Prediction errors will be generated between SERD model predictions and measured RWA densities. Prediction errors will be used to validate the SERD model predictions. Objective 2. Similar methods will be used for barley field sites located near Briggsdale, and Akron, Colorado. Independent data layers will be collected and inputted into our GIS as needed for our SERD model predictions. Objective 3. To quantify the loss function between RWA density data and yield loss, artificially and naturally infested wheat and barley plots will be harvested at the locations indicated above. These data will be used to examine the correlation of early spring RWA densities to areas of later RWA population growth, and to subsequent yield loss. Control plot yields will be spatially modeled, using NDVI from the spring as a covariate, to generate a predicted yield surface. Yield loss will be calculated by differences between expected yield and measured yield per treatment plot. RWA densities per time period will be calculated by the mean RWA density per plot per sample date using the subplots sampled in Objective1. Loss functions will be calculated by proportion of yield lost per RWA per sample date. Sample date will be correlated with growth stage. This will result in loss functions across growth stages through the growing season. Objective 4. Infestation Level and Crop Species will be parameterized and added into the existing SERD model. Also, data collected through this project will be integrated with previously collected data to reparameterize the existing SERD model. Objective 5. Build a software program in ArcGIS 9.2 ESRI (1995-2007) that will for quick retrieval of RWA prediction surfaces, including spatially delineated areas where control efforts would be advised and, conversely, not advisable (i.e., Risk Assessment Maps). Programming will require yearly input of Landsat imagery and linked input of CoAgMet data. User would input temperature, precipitation and locations of interest, and the program will generate a RWA density map and a Risk Assessment Map for the requested locations. A fact sheet detailing the use of the online program and interpretation of results will be written and disseminated through appropriate channels.