CONSTRUCTING AND EVALUATING LARGE-SCALE INFILTRATION SWALES FOR RETAINING AND INFILTRATING ROADWAY STORMWATER RUNOFF

Abstract

Urbanization, characterized by the increase of impervious surfaces including roads, parking lots, and buildings, presents challenges for stormwater management. The expansion of impervious surfaces disrupts natural infiltration processes, leading to increased volumes and peak flow rates of stormwater runoff. These changes necessitate effective management strategies and practices to mitigate the negative consequences such as flooding, streambank erosion, and pollution of waterways. In response, post-construction stormwater control measures (SCMs) that integrate Low Impact Development (LID) and Green Infrastructure (GI) principles are gaining traction. LID and GI SCMs utilize various sustainable processes, such as evapotranspiration, infiltration, filtration, and water reuse, to mimic pre-development hydrology and manage the quantity and quality of stormwater runoff. The Alabama Department of Transportation (ALDOT) has embraced an infiltration-based LID and GI SCM called infiltration swales for highway runoff management. Infiltration swales are linear vegetated channels that employ an engineered soil media and check dams to minimize surface discharge and reduce peak flow rates. These vegetated systems mimic predevelopment hydrology by promoting infiltration of stormwater runoff into their media and the underlying native soils, potentially replenishing the local groundwater table. This infiltration process is facilitated by the engineered soil media, a component designed to optimize water permeability while utilizing natural materials. While infiltration swales are a common stormwater management practice in Alabama, their performance can vary. To address this variability, ALDOT partnered with Auburn University on a two-phase research project to develop a modified infiltration swale design based on the existing ALDOT standard, and to evaluate both swale’s infiltration performance. The first phase focused on developing a modified swale design based on the existing ALDOT standard. The second phase, which forms the basis of this thesis, involved large-scale field testing to compare the performance of the ALDOT and modified swale designs at the Auburn University Stormwater Research Facility. This research aims to evaluate and compare the infiltration performance of the standard ALDOT infiltration swale design to the newly developed modified swale design. Large-scale infiltration swales were constructed, side-by-side, to facilitate controlled experimentation and monitoring. The evaluation focused on infiltration rates and drawdown times under various scenarios designed to assess the influence of external factors on infiltration performance. These factors included variations in rainfall frequency, underdrain valve settings (open vs. closed), initial soil moisture conditions (wet vs. drier), and seasonal variations. Moisture content sensors were also installed within the swale media and surrounding soil at different depths and locations to track the movement of infiltrated water. Additionally, settlement of the swales was monitored from construction completion, and surface storage volumes were measured. This comprehensive approach, incorporating various tests and measurements, allowed for a robust comparison of the overall infiltration performance between the ALDOT and modified swale designs. The evaluation revealed significant performance differences between the ALDOT and modified infiltration swales. The ALDOT swale displayed a notably lower average infiltration rate of 1.6 ft/day (0.49 m/day) and a longer average drawdown time of 12.25 hours than the modified swale. The modified swale exhibited a higher average infiltration rate of 5.2 ft/day (1.6 m/day) and a considerably faster average drawdown time of 5.06 hours. These results indicate a statistically significant difference in performance between the two infiltration swale designs. Further analysis concluded that decreased rainfall frequency increased infiltration rates by 1.6 times more for the ALDOT swale and 2.4 times more for the modified swale compared to increased rainfall frequency. Drier soils increased infiltration rates by 1.5 times more for the ALDOT swale and 2.3 times more for the modified swale compared to wetter soils. Closed valve underdrain tests outperformed open valve tests for both swales potentially due to seasonal variation. Colder months were associated with slower infiltration rates while warmer months were associated with enhanced infiltration rates. Warmer months showed improved infiltration rates by 1.7 times more for the ALDOT swale and 2.7 times more for the modified swale compared to colder months. Moisture sensor data revealed an interesting contrast. While the ALDOT swale held 10% more surface water volume compared to the modified swale, the infiltrated water traversed the modified swale media significantly faster, reaching the bottom interface with the native soil in an average of only 0.13 hours (7.6 minutes). In contrast, the ALDOT swale exhibited a much slower travel time, with infiltrated water taking an average of 1.8 hours to reach the bottom. Further evaluation showed that there was no settlement observed in both infiltration swales. Beyond the core findings on infiltration performance, this thesis accomplished several key milestones. A comprehensive literature review explored factors influencing infiltration-based SCMs, a study on the different infiltration-based SCMs used across the United States by different Departments of Transportation, and studies on grass swales and their average infiltration rates. The project then progressed to a detailed field-scale construction phase. The construction process included geotechnical and soil investigation, site selection, excavation, media material placement, moisture content installation, and site stabilization for both infiltration swale designs. Another major milestone involved constructing and calibrating the testing apparatus used for the subsequent experimentation and evaluation of the two swales. These combined efforts ultimately served two primary objectives: to inform ALDOT on the modified infiltration swale design created, and to provide robust evidence from large-scale experimentation to support the findings on the modified swale's infiltration performance compared to the existing ALDOT swale design.