UC Davis

Evapotranspiration (ET) is a critical variable in understanding soil-plant-atmosphere interactions and has been widely used as a parameter to inform irrigation. Common micrometeorological approaches to measuring ET are the Eddy Covariance (EC) and Surface Renewal (SR) methods; however, these methods are either not affordable, inaccessible, or rely on instrumentation that is too fragile to withstand environmental conditions and hinder agricultural operations. Like air temperature high-frequency time-series data, water vapor shows ramp-like patterns that can be used to estimate latent heat (LE) fluxes more directly. We applied a novel wavelet analysis to high-frequency (20 Hz) and semi-high frequency (1 Hz) water vapor data using a multilevel one-dimension wavelet decomposition based on the symlet wavelet to detect the amplitude and duration of ramp-like features over 30-minute intervals. The ET was computed based on this new method and compared to the EC and SR methods. ET estimates using the wavelet analysis approach strongly correlate with the estimates derived from EC and SR. Additionally, a more robust response at lower frequencies when using the wavelet analysis method suggests a more affordable, accessible, and direct method for estimating ET and offering growers a more cost-efficient and effective way to manage irrigation.