Thermal crop sensing technologies have potential as tools for monitoring and mapping crop water status, improving water use efficiency and precisely managing irrigation. As thermal sensors and imagers became more affordable, various platforms were examined to allow for canopy- and field-scale acquisitions of canopy temperature and to extract maps of water status variability. Various canopy temperature statistics and crop water stress index (CWSI) were used to estimate water status in different field and orchard crops. Despite the potential reflected from these studies, currently, thermally-based water status measures and maps are scarcely utilized for irrigation management. In the current study thermal-based irrigation management was compared with a commercial practice in cotton.

Currently, the best practice to assess water status of a commercial cotton field in the boll-filling period, leaf water potential (LWP) of few selected plants is measured which do not necessarily represent the entire field or its variability. Despite its proven efficiency the LWP-based irrigation management is not the common practice because of its complication.

In recent years a multi-year CWSI-LWP regression model was built for the boll-filling period in cotton using ground-based thermal images. On the last cotton season (2013), the CWSI-LWP model was used to test its applicability for irrigation management. Two irrigation regimes were conducted: 1) commercial irrigation decision making based on few direct LWP measurements conducted twice a week; 2) thermal-based irrigation decision making based on calculated LWP values extracted from thermal images acquired once a week. The two irrigation regimes had six replications, each in size of 18X19 square meters. Diagonal thermal images were acquired above the experimental plots using an uncooled infrared thermal camera (SC655, FLIR systems, Oregon, USA). Three steps were required to calculate LWP using thermal images. First, canopy temperature average was extracted for each replicate-plot. Secondly, CWSI was calculated using the canopy temperature average and two reference temperature values: Twet and Tdry. For the Twet the canopy temperature of over-irrigated plots was used and for Tdry the air temperature plus 5⁰C was used. Finally, the LWP was calculated using the linear CWSI-LWP model that was built in previous years. At the end of the season the two irrigation regimes were compared in terms of yield and irrigation amounts. No significant differences were found between the two irrigation regimes in both yield and irrigation amounts. The thermal-based irrigation regime obtained slightly higher yield with slightly lower irrigation amounts. The results of this study showed that thermal images can be used for irrigation management in cotton. For its assimilation the cost effective of the thermal-based irrigation management should be examined in commercial scales. Future study should focus on the applicability of this approach for variable rate irrigation.