Data Driven Farming: Delivering the Benefits of Remotely Sensed Data and Decision Support Tools to Farmers

Abstract:

Web-based agricultural management software allows for the delivery of previously hard to access soil, weather and remotely sensed data to growers. While access to these data sources is beneficial, growers can realize large gains by leveraging field level data and integrating decision support tools that have been presented in the literature. Using a previously developed model for estimating the growth stage of maize (Sakamoto et al. 2010), we examine how remotely sensed data can be integrated into a web-based agricultural management tool, informing grower management decisions by providing near real-time estimates of crop growth stage and sub-field level variability in growing conditions. When combined with field-level soil and weather data, growers can use remote sensing based models to tailor management activities, taking variable (climate related) and invariant (site characteristic) yield determining factors into account.

Time series of Wide Dynamic Range Vegetation Index (WDRVI) derived from Landsat observations were linked to 97 fields growing maize across the Mid-western region of the United States in 2013. Crop growth stage day of year (DOY) estimates (V2.5, R1, R5, and R6) were predicted for each field based on the WDRVI profile and compared to a growing degree day based estimate. These estimates aid in scheduling growth stage specific management activities and allow farmers to more efficiently monitor geographically remote fields. Within-field variation in growing conditions is presented on an ordinal scale (below average, average, above average) based on the field-level distribution of WDRVI values at each observation. Monitoring of sub-field level conditions allows growers to calibrate field-level yield estimates, prioritize field scouting activities and plan targeted interventions. Integration of these tools into existing web-based agricultural management tools allows growers to easily incorporate remotely sensed data into their decision making process, facilitating the exchange of information between researchers and practitioners.