Estimating daily root-zone soil moisture in snow-dominated regions using an empirical soil moisture diagnostic equation

Abstract

Soil moisture in snow-dominated regions has many important applications including evapotranspiration estimation, flood forecasting, water resource and ecosystem services management, weather prediction and climate modeling, and quantification of denudation processes. A simple and robust empirical approach to estimate root-zone soil moisture in snow-dominated regions using a soil moisture diagnostic equation that incorporates snowfall and snowmelt processes is suggested and tested. A five-water-year dataset (10/1/2010–9/30/2015) of daily precipitation, air temperature, snow water equivalent and soil moistures at three depths (i.e., 5cm, 20cm, and 50cm) at each of 12 Snow Telemetry (SNOTEL) sites across Utah (37.583°N–41.883°N, 110.183°W–112.9°W), is applied to test the proposed method. The first three water years are designated as the parameter-estimation period (PEP) and the last two water years are chosen as the model-testing period (MTP). Applying the estimated soil moisture loss function parameters and other empirical parameters in the soil moisture diagnostic equation in the PEP, soil moistures in three soil columns (0–5cm, 0–20cm, and 0–50cm) are estimated in the MTP. The relatively accurate soil moisture estimations compared to the observations at 12 SNOTEL sites (RMSE⩽6.23 (%V/V), average RMSE=4.28 (%V/V), correlation coefficient ⩾0.75, average correlation coefficient =0.89, the Nash-Sutcliffe efficient coefficient Ec⩾0.24, average Ec=0.72) indicate that the soil moisture diagnostic equation is capable of accurately estimating soil moisture in snow-dominated regions after the snowfall and snowmelt processes are included in the soil moisture diagnostic equation.