A Soil-up Approach to Precision Agriculture.

There is an unpreceded increase in data quantity and quality in agriculture through the advances in unmanned aerial vehicle (UAV) technology and sensor systems. One of the most important issues in production agriculture is water management. Crop water use is driven by the crop genetics and the local environment, including the soil and the microclimate. By linking soil information with sensor-data, especially the soil properties that drive water storage, it may be possible to separate out the impacts of genetics and environment on water use. The objective of this study was to measure crop water use and crop stress using UAV-based sensors, as well as evaluate how differences in soil properties affect the uncertainty of UAV-based crop sensing. Corn and cotton were evaluated for water use across multiple soil types in Texas using both UAV- and ground-based sensors. The soils across the study site ranged from fine sandy loam to clay textures, giving us a large variety of local conditions. Multi-spectral and thermal sensors were flown using a fixed-wing UAV. Ground truth data includes soil moisture, soil nutrients, leaf area index, plant height, and canopy temperature. Eddy covariance flux towers were also installed on two soil types in the cotton field to validate estimates of crop transpiration. Ground-truth data will be used to validate UAV-sensors, and a framework for water management decisions that integrates soil information with sensor-date will be developed.