GSA Annual Meeting in Seattle, Washington, USA - 2017

Paper No. 5-5
Presentation Time: 9:20 AM

MAPPING WATER INFILTRATION OF URBAN SOILS USING ELECTROMAGNETIC INDUCTION AND TIME-LAPSE ELECTRICAL RESISTIVITY TOMOGRAPHY


CARSILLO, Vince, NYQUIST, Jonathan, WONG, Tyler and TORAN, Laura, Earth and Environmental Science, Temple University, Philadelphia, PA 19122, vince.carsillo@temple.edu

A major issue in urban hydrology is stormwater runoff generated from impervious land cover. In Philadelphia, PA, there are 43,000 derelict properties, many of them vacant vegetated lots that potentially contribute to the infiltration of stormwater. However, urban soils are often heterogeneous in composition, compaction and macropore distribution, all of which are difficult to characterize. Methods are needed to improve predictions of urban infiltration despite this heterogeneity. We compared two geophysical methods and double-ring infiltrometer measurements to characterize a site where a house had been demolished and the property seeded with grass. Electrical resistivity tomography (ERT) can survey a line repeatedly to provide time-lapse cross sections of the subsurface. Changes in soil moisture appear as a decrease in resistivity in the time-lapse profile. Electromagnetic (EM) induction can detect changes in soil moisture by walk-over surveys, that cover larger areas in less time. However, data from EM surveys that is not calibated may lead to unreliable interpretations of changes in resistivity with depth. Our ERT surveys served as this calibration. By determining the relationship between ERT and EM induction, a larger scale grid can quickly be measured using the EM profiler to map infiltration patterns following a rain event. At this site, the surveys were conducted using a sprinkler to create a reproducible artificial infiltration event. During preliminary experiments, ambient temperature drift skewed EM profiler readings. This effect was eliminated by taking measurements at base stations outside of the irrigation range and normalizing our data against these controlled base measurements. The ERT line showed highly heterogeneous infiltration influenced by the house’s former driveway and building foundation, which appeared as anomalies in EM surveys of the site’s background heterogeneity and were later identified in historical aerial photographs. The variable infiltration rates that we analyzed with these two methods were compared to infiltrometer measurements. Applying the EM profiler method in an urban setting may result in a cost-effective way to monitor heterogeneous infiltration of urban basins and provide data to better parameterize stormwater runoff models.