Water Use in Turfgrass Using Sensor-Driven Decisions.

Residential and commercial landscapes are routinely overwatered to ensure plants remain vibrant and visually pleasing.  Plant and soil instrumentation now allow real-time detection of in situ and remotely sensed parameters to better inform models like the classic Penman-Monteith evapotranspiration calculation. However, these approaches are routinely discounted, preferring simple user generated model inputs; a choice that has the potential to waste precious freshwater resources.  The objective of this study was to combine inexpensive but research-grade instrumentation into an actively managed turf farm to optimize water use and reach a net-zero water loss below zone. The study utilized a three-pronged approach to ensure all aspects of plant growth were considered. Soil water potential and water content sensors were collocated in and below the turf route zone to assess both available and applied water and unused water seeping to the ground water. Spectral reflectance of both normalized difference vegetation index (NDVI) and photochemical reflectance index (PRI) were taken over the plant canopy to sense the onset of stress and estimate crop coefficients. And, a microenvironment monitoring station was placed next to the turf to evaluate meteorological conditions and determine water loss.  Results from the ongoing experiment show basic turf management applies water well in excess of plant needs; an expected result.  Unexpectedly, collocation of soil sensors provided a rich picture of water availability, soil type, and hydraulic conductivity and has the potential to improved water use models by avoiding user based inputs.  Spectral reflectance measurements proved more challenging as data required intense filtering.  Still, key inputs of crop coefficient estimates again offer the possibility of removing this user generated values.  Local ET estimates from the microenvironment monitor showed a much-improved estimate of latent heat flux due to the challenging advective conditions that are common in urban areas. In all, this study shows an exciting opportunity to save scarce freshwater resources using sensor-driven decision making backed by plant and soil water relations.