Using Wireless-Based Thermal Radiometers for Monitoring Crop Water Use At Maricopa, Arizona.

Water availability for irrigated agriculture in the U.S. Southwest is under threat due to competition from urban users and reduced supplies due to drought. This reduced availability means that farmers and irrigation district water managers will need tools and systems that can significantly improve water conservation and water use efficiency. One of these is a wireless-based sensor network system which can provide near real time water use information at numerous sampling locations across a farm. Samples collected can include plant surface temperature, plant spectral reflectances, and soil moisture profiles and provide observations closely related to plant water status, plant stress, and root zone soil moisture. By combining these observations with other data sets, one can estimate the spatial distribution of current and near-future daily plant water use. To evaluate their feasibility and limitations we have incorporated a commercially available wireless sensor system into our irrigation studies since 2007. These include experiments in cotton, wheat, and the oil-seed crop camelina.  The system consists of a laboratory sited base-station, 24 wireless solar-powered nodes, infrared thermometers, soil moisture probes, visible/near infrared proximal sensors, web-based software, and an ancillary air temperature/humidity probe. Data are collected asynchronously via the Zigbee protocol, which provides a robust mechanism for collecting and transmitting low data volumes with low power requirements. Our studies have focused upon the use of thermal infrared radiometry since this observation type is sensitive to both plant water status and root zone soil moisture at short and long-term time steps. When combined with spatial estimates of near surface meteorology and fractional vegetative cover, the resulting surface temperatures can be used to model the surface energy balance, evapotranspiration (ET) and root-zone soil moisture.  In retrospective analyses for field experiments in wheat and cotton, wireless-derived datasets were used to independently estimate surface heat fluxes. When these were integrated to daily time steps, crop water use estimates sometimes showed agreement within 1 mm/d ET.  Details of the wireless deployments and resulting ET estimates will be discussed.