ASABE Technical Library - Abstract

Abstract. Evapotranspiration (ET) controllers have the potential to produce water savings when implemented in scenarios with frequent excessive irrigation, but not all controllers have the same capabilities. The Irrigation Association’s Smart Water Application Technologies Testing Protocol for Weather-Based Irrigation Controllers (SWAT test) evaluates ET controllers using a daily soil water balance as a way to quantify performance. The manufacturers have been known to manipulate the landscape settings into non-representative descriptions to achieve good SWAT test scores. The United States Environmental Protection Agency (USEPA) WaterSense program adopted the SWAT test to certify products as water-saving devices with chosen thresholds of 80% for irrigation adequacy and 95% for scheduling efficiency as efficacy measures for under- and over-irrigation, respectively. The objective of this research was to determine if using an hourly soil water balance (hourly SWB) would result in better correlation of efficacy measures to irrigation application and efficiency compared to the SWAT test. Two data sets were utilized for this analysis. The first was from a bench test using ET controllers programmed to pass the SWAT test (not selected based on the landscape description) and the second was from a field plot study with landscape settings matching the plot descriptions. Correlations between irrigation application and efficacy measures were consistent for the hourly SWB, making it the preferential model due to the more realistic representation of soil water movement. The SWAT test scores rarely correlated with irrigation application, indicating a lack of connection to real world performance. When program settings were not realistic for the described landscape, there was the tendency of under-irrigation to occur. This resulted in an increased failure rate of achieving the WaterSense label at the 80% threshold using the hourly SWB. When program settings were related to the landscape characteristics, as in the field plot study, all treatments tended to have over-irrigation. The WaterSense threshold of 95% for scheduling efficiency is already so high that both models frequently resulted in failing scores.