Evapotranspiration and Infiltration in Rain Garden Systems

Abstract

Quantification of evapotranspiration (ET) and infiltration from vegetated stormwater control measures (SCMs), such as rain gardens, is necessary to properly assess their volume reduction potential. Weighing lysimeters at Villanova University mimic three bioretention rain garden designs and measure water budget parameters to determine how design elements impact ET and infiltration (percolation). The designs compare two soil medias: a loamy sand (72% sand, 12% silt, 1% clay) and sand (96% sand, 4% silt, 0% clay), and different drainage systems: a controlled valve and internal water storage (IWS) outflow. The controlled valve outflow allows a user defined flow rate. The IWS outflow maintains a storage volume within the media. A custom distribution system was built to simulate excess rain delivered to a rain garden during natural storm events. In Pennsylvania, SCMs are often sized to control a multiple (typically 5:1 for rain gardens) of a selected rain event. Trials were performed in April and August 2014 for half open controlled valve configurations. Both loamy sand and sand soils with a controlled valve configuration showed an average ET of 3.1 mm per day for 7 days after simulated 5:1 events. Sand soil with an IWS showed an average of 6.0 mm per day for 7 days after simulated 5:1 events. Simulated storm events produced larger ET rates, on average, than that of a 1:1 ratio. Comparisons of predictive equations show the ASCE Penman-Monteith (uncorrected for water stress or crop type) under-predicting and Hargreaves (corrected for water availability) over-predicting observed ET, but follows the trend of each lysimeter.