Watershed-scale impacts of stormwater green infrastructure on hydrology, nutrient fluxes, and combined sewer overflows in the mid-Atlantic region

Area characterisation: 

Three areas were selected in the Mid-Atlantic region of the United States: Baltimore County, MD, Montgomery County, MD, and Washington, DC (referred to as DC herein after). These locations were chosen based on their geographic proximity and the abundance of available data. From each region, public and private data was compiled on the locations of SGI facilities, streamflow data, and long-term stream chemistry, and for DC only long-term combined sewer overflow data was obtained. Watershed sizes ranged from 0.5 to 34.3 km2 with a median area of 6 km2. For CSO analysis, individual sewersheds that drain to each CSO were selected. The sewershed size ranged from 12 m2 to 5.6 km2, with a median area of 0.2 km2. All sites in this study, are within the Piedmont physiographic region and have had approximately the same average annual rainfall since 1984 of 39.4 in. in DC and 40.8 in. in Baltimore.


Use of vegetation in urban areas in the Mid-Atlantic watersheds of Washington DC, Montgomery County and Baltimore County MD, has reduced flooding and nutrient runoff. The ‘stormwater green infrastructure’ in these municipalities includes green roofs, bioswales, rain gardens and stormwater ponds. By increasing infiltration and groundwater recharge, and/or evaporation, they reduce the volume of run-off that can contribute to floods.

Potential impacts/benefits: 

Watersheds with more stormwater green infrastructure were found to have their peak runoff being 44% lower (mm/day), and runoff events being 26% less frequent and 26% less variable. Taken together this suggests that the green infrastructure reduced the potential for flooding. Of the areas studied, DC had the greatest proportion of its landscape drained through green infrastructure, at 12.7%, whilst Baltimore County had the lowest, at 7.9%. Additionally, the green infrastructure was shown to decrease nitrate export by 44%, hence likely decreasing nutrient pollution of rivers and consequent harm to wildlife.

The green infrastructure was found to have not caused a significant reduction in phosphorous exports or sewer overflows. This was likely because the scale of implementation was too small. Indeed, to reduce flooding during heavy rainfall , the area of stormwater green infrastructure would need to be much larger.  To maximise the benefit from increasing green infrastructure area,  more research is needed into the differential effects of the types of stormwater green infrastructure and the impact of their location.

NbS benefits 
  • Developing climate change adaptation; improving risk management and resilience
  • Flood peak reduction
  • Increase infiltration / Water storage
  • Increasing infiltration
  • Reduce flood risk
Lessons learned: 

When comparing individual watersheds over time, increases in stormwater green infrastructure (SGI) corresponded to non-significant reductions in hydrologic flashiness compared to watersheds with no change in SGI. While the implementation of SGI is somewhat in its infancy in some regions, cities are beginning to have a scale of SGI where there are statistically significant differences in hydrologic patterns and water quality.


The NatureNet Science Fellowship Program provided funding for this research through a donation to The Nature Conservancy

Further information