A Clean Water Act (CWA) Section 319 Nonpoint Source Pollution Management Program grant was issued by the Illinois Environmental Protection Agency (IEPA) to the city of Aurora to promote increased water quality in stormwater runoff draining into the municipality’s separate storm sewer system (MS4), which eventually discharges to the Fox River. As part of the project, five hydrologically connected biofiltration facilities (commonly known as rain gardens) were constructed along Spring Street within the south parkway, between Fourth Street and Root Street. One biofiltration facility is 60 feet long, and the remaining four are 50 feet long. All the biofiltration facilities are 11 feet wide with an 8-foot gap between two consecutive facilities, with the exception that there is a driveway in between two of the biofiltration facilities. The biofiltration facilities were placed parallel to and in between the road curbside and a sidewalk. The facilities were designed to take runoff from the street as well as from the sidewalk and also from the open grassy area beyond the sidewalk. In case of large storm events, runoff overflows from one facility to another until it reaches the storm sewer constructed beneath the rain gardens as part of a separate project via the catch-basins located in two facilities.
These biofiltration facilities were designed with an appropriate substrate to provide stormwater infiltration, removal of constituents, and a planting medium for native shrubs, grasses, and flowers. The plantings in the biofiltration facilities consist of mesic prairie with flowers (live plants and seed mix) and a cover crop of seed oats, Virginia wild rye, and annual rye. The pedestrian walk-through and buffer zone along the sidewalk lawn consist of 50% Kentucky bluegrass 98/85. The shrubs in the biofiltration facilities consisted of grey dogwood and nannyberry.
Modeling
The nonpoint-source pollutants to be removed were particulate solids, total solids, particulate phosphorus, total Kjeldahl nitrogen, metals (total copper, total lead, and total zinc), filterable ammonia, and fecal coliform bacteria. Before the rain gardens were constructed, they had to be designed and modeled. The efficiency of the biofiltration facilities in removing pollutants from runoff was modeled using two approaches and the results were compared. The first approach was the Source Loading and Management Model (SLAMM) and the second one was the Urban Runoff worksheet developed by IEPA. SLAMM was the primary model for determining the pollutant reduction rates of the biofiltration facilities. SLAMM is an urban water management tool that can be used to model pollutant loadings in surface runoff water resulting from rainfall events, and it can also be used to study the efficiency of control devices in removing the pollutants.
For modeling purposes, the biofiltration facilities were grouped into two systems based on their flow. The two systems are separated by an existing driveway. For convenience, the biofiltration facilities were numbered as 1, 2, 3, 4, and 5 from west to east, beginning at Fourth Street. The first three biofiltration facilities (1, 2, and 3) are connected in series and were modeled as System 1. Runoff from catchment areas 1, 2, and 3 flows to biofiltration facilities numbered 1, 2, and 3, respectively, and then to the catch basin located in biofiltration facility 3. The total area modeled for System 1 was approximately 0.49 acre. Biofiltration facility 4 is connected in series to biofiltration facility 5, and both were modeled as System 2. Runoff from catchment areas 4 and 5 flows to biofiltration facilities 4 and 5, respectively, and then to the two catch basins located in biofiltration facility 5. The total area modeled for System 2 was approximately 0.75 acre.
For modeling purposes, the biofiltration facilities were simulated as biofiltration control devices. Each biofiltration facility has two curb cuts, which were modeled as a single 3-foot-long broad crested weir on each biofiltration control device. Figure 7 shows the schematic of the biofilter geometry. As can be seen in Figure 7, 1 foot of engineered soil (compost-sand) was used at the bottom of each biofiltration facility. Each biofiltration control device had a set of parameters that served as inputs for the model.
According to the Natural Resources Conservation Service (NRCS) soil database, the native soil is silt loam. The event-based rainfall data were from Moline, IL, for the period December 27, 1997, through December 27, 1999. The maximum rainfall event recorded during this period was 4.13 inches. The justification behind using the Moline, IL, rainfall data is that no rainfall data for the SLAMM model was available for the Chicago metropolitan area at the time of modeling. Additionally, Internet sources show that Moline has an average annual rainfall (3.17 inches) similar to Aurora, IL (3.20 inches), suggesting that Moline has a similar precipitation climate as Aurora. The maximum monthly average rainfall for Moline and Aurora is also very comparable (4.63 inches for Moline and 4.39 inches for Aurora).
Results
SLAMM modeling of System 1 showed that the runoff volume flowing into the three rain gardens was 29,083 cubic feet (during the two-year modeling period), of which only 5,333 cubic feet overflowed to the catch basin located in biofiltration facility 3. The percentage of water lost due to infiltration in the rain gardens was 82%. The particulate solids reduction by the biofiltration facilities was 88%. The second column of Table 1 shows the various constituent removal rates based on pounds per year entering System 1. For System 2, the runoff volume flowing into the two rain gardens was 36,855 cubic feet (during the two-year modeling period), of which only 12,817 cubic feet overflowed to the catch basins located in biofiltration facility 5. The water loss due to infiltration in the rain gardens was 65%. The particulate solids reduction by the
biofiltration facilities was 85%. The third column of Table 1 shows the various constituent removal rates based on pounds per year entering System 2.
According to the SLAMM model results, the “Receiving water impacts due to stormwater runoff” was “good” with the rain gardens as compared to “poor” without the rain gardens. The average of the total ponding duration in each biofiltration facility for System 1 was 8.95 hours and for System 2 was 10.78 hours. The maximum total ponding duration in System 1 was 61.3 hours and in System 2 was 62.7 hours.
The IEPA worksheet was used as a comparison chart against the SLAMM model results. The IEPA worksheet is an Urban Runoff worksheet developed by IEPA. IEPA requires that this worksheet be used in conjunction with the best management practices (BMP) application form for each proposed BMP funded through Section 319 of the CWA. This is a very simple worksheet that calculates the particulate solids reduction rates based on inputs. The comparison of SLAMM model results with the IEPA worksheet showed that there was 88% reduction of particulate solids for the SLAMM model and 76% reduction of same for the IEPA model in System 1. For System 2, there was 85% reduction of particulate solids for the SLAMM model and 75% reduction of same for the IEPA model. The SLAMM model results showed higher particulate solids reduction rates compared to the IEPA model results. This is understood given the different modeling approaches followed by the two models.
The rain gardens were constructed to be beneficial in reducing the total surface runoff to the Fox River. The City of Aurora is committed to conducting regular inspections and maintenance of the Spring Street rain gardens for a period of at least 10 years.
Acknowledgements
The author acknowledges the contributions of Shaw Environmental Inc. staff who assisted in the project, and the city of Aurora for support. Special thanks to Gerald DeMers of Shaw Environmental Inc. for his support. All project work (except construction) mentioned on this article was done by Shaw Environmental Inc.
