Beneath the Surface: Soil Particle Size Analysis of Green Stormwater Infrastructure in Washington, D.C.

Researcher(s)

  • Emily Berger, Environmental Engineering, University of Delaware

Faculty Mentor(s)

  • Carolyn Voter, Civil, Construction, and Environmental Engineering, University of Delaware

Abstract

Green stormwater infrastructure (GSI) has been used for over a decade; however, its long-term functionality is not well studied. An important factor in GSI’s functionality is the soil composition. Standard soil particle-size analysis methods have been extensively studied, but few studies have used rain garden soils. Due to the highly variable levels of organic matter in raingarden soils, it must first be removed before traditional soil particle-size analysis methods can be used. To determine the particle sizes of a large number of soil samples from raingardens, a fast method that works on soils with varying levels of organic matter is necessary.
Soil samples were collected at the inlets and low points from 146 rain gardens installed by RiverSmart Homes in Washington, D.C. Using raingarden soils collected from campus, sieve and hydrometer tests were completed with and without LOI and compared. These results demonstrated that the LOI procedure was not necessary before sieving the raingarden soil; however, it is necessary before performing the hydrometer test. Next, the organic matter of the D.C. samples was burned off in the furnace at 550 degrees Celsius using the loss-on-ignition (LOI) procedure. Results from the LOI tests indicated that the rain gardens have highly variable organic matter levels ranging from about 2 to 40 percent. Due to the large number of soil samples from Washington, D.C. and the time-consuming nature of the loss-on-ignition and sieve-hydrometer procedures, a faster method was needed to analyze all 292 samples. Using a combination of wet sieving to obtain the sand fraction and laser diffraction to obtain the fines fraction, it is possible to determine the soil composition. Preliminary results from this method indicate that the soils contain a high fraction of sand particles, with about 70 – 80 percent sand.