Investigating the application of artificial hot and cold reference surfaces for improved ETc estimation using the UAS-METRIC energy balance model

Abstract

The METRIC energy balance model requires the temperature of a hot (dry bare agricultural soil) and cold (fully transpiring crop) surfaces for the internal calibration and estimation of crop evapotranspiration (ETc). A previous study indicated that because of the limited field-of-view of UAS-based imagery, the application of unmanned aerial systems (UAS)-based METRIC could not guarantee the availability of representative hot and cold pixels in the field-of-view of the flight mission, even if the pixels were manually chosen. A study was conducted for two years (2017 and 2018) to evaluate the suitability of various artificial and portable hot (dry red mulch, dry terra cotta saucer plate, coconut fiber, wheat straw) and cold (surface of water, wet terra cotta saucer plate, green sponge, and white granite tile on top of polystyrene foam floating on water) artificial surfaces for use in the calibration of the UAS-METRIC model. The temperature of the surfaces was measured with a thermal infrared sensor mounted on a ground vehicle with a retractable mast (up to 7-m) facing downward to capture images at around the nadir hours (11 AM to 3 PM). The results showed that a green sponge floating in a water tray and a dry terra cotta saucer plate full of water best matched (rwet sponge= 0.93, rwet saucer =0.86) the temperature dynamics of fully irrigated short grass. Dry red mulch and a dry terra cotta saucer plate best matched (rdry red mulch= 0.98, rdry saucer =0.97) the temperature dynamics of bare and dry agricultural soil. These artificial hot and cold reference surfaces were then located within a field of view of nine different UAS flight missions over a spearmint field during the summer of 2021. Although the temperatures of the cold artificial surfaces correlated well with fully irrigated short grass planted near the spearmint field, they were 1.9–8.3 °C warmer than the taller spearmint crop canopy. The temperature of a dry red mulch/wet sponge and dry/wet terra cotta saucer plate used as artificial hot/cold anchor temperatures for ET estimation by the UAS-METRIC model resulted in overestimating the ETc obtained using field hot/cold pixels (dry bare soil/fully irrigated spearmint), or the ETc estimation based on a soil water balance measured by neutron probe and TEROS-10 sensors. The best hot/cold temperatures for internal calibration of the UAS-based imagery ET model are from dry soil and a fully irrigated canopy located in the field of view. In the absence of dry bare soil within the image, the use of dry red mulch is suggested. Additional research is needed to find artificial cold surfaces with temperatures closer to the temperature of fully irrigated canopies, which can be conveniently located for auto-selection within the images. In the meantime, for farm-scale applications, maintaining a field section guaranteed to be well irrigated is recommended.