Evapotranspiration estimation in rain gardens using soil moisture sensors

Abstract

Rain gardens are increasing in use as the shift from gray to green infrastructure continues. Water that enters a rain garden is removed by three mechanisms: overflow to an outlet, percolation to the underlying soil, and evapotranspiration (ET) to the atmosphere. Despite the importance of ET in aiding a rain garden to recover void space during the time between storm events, it is not often measured or calculated because it is so difficult to do so. This paper explores the use of soil moisture sensors to estimate ET, since they are relatively inexpensive to purchase and install. Three rain garden weighing lysimeters in Villanova, PA, were used for this study. Over a 3-yr study period daily ET was calculated each dry day and summed during the time in between storm events >25 mm. The cumulative changes of soil moisture readings during the interevent time at three depths were compared with the concurrent cumulative changes in the lysimeter weight readings. Cumulative soil moisture change was found to be strongly correlated to the cumulative ET for the different lysimeter media and drainage types. Using the two soil moisture sensors at the top and bottom of the soil column provided a similar result to using all soil moisture reading depths. The best single soil moisture reading depth was at the bottom of the lysimeter. The ET estimates via soil moisture tracking performed similar to uncalibrated potential ET estimates, but not as well as calibrated ET estimates.