Stormwater infiltration and groundwater mounding

Schematic illustrating groundwater mounding beneath an infiltration BMP (Source: CDM Smith)

The localized groundwater surface may temporarily rise below an infiltration BMP, creating a condition termed groundwater mounding. Mounding can occur in areas where the volume of water entering the subsurface is greater than the soil is able to convey away (Susilo, 2009). Infiltration BMPs in particular have the potential to cause a groundwater mound, given the right subsurface conditions, because they direct the recharge to a specific area (Machusick and Traver, 2009).

When is Groundwater Mounding a Problem?

The vadose zone is the unsaturated depth of in-situ (existing) soil below the infiltration BMP. Contaminants that are not captured within the BMP may be attenuated within this zone. The temporary rise of the groundwater elevation caused by mounding will decrease the available vadose zone which may decrease the removal of certain pollutants. If the mound reaches the base of the infiltration BMP, then the hydraulic gradient (direction of water movement) shifts from vertical to horizontal and significantly slows the movement of water through the soil. Groundwater mounds that significantly widen below the base of an infiltration BMP may damage underground utilities, basements and building structures if the mounds are high enough and there is not enough separation between the mound and structures (Machusick and Traver, 2009).

When Should a Mounding Analysis be Conducted?

A mounding analysis is a way to determine the likelihood that a groundwater mound will occur. A mounding analysis should be performed if one or more of the following conditions exist:

• Low saturated hydraulic conductivity. Soils with low saturated hydraulic conductivity do not dissipate the infiltrating water quickly, allowing for the water to accumulate below the base of the infiltration BMP.
• Thin vadose zone. Mounding is more likely to occur in areas with a shallow groundwater table (thin vadose zone).
• Low aquifer thickness. Groundwater aquifers that are thick are able to dissipate infiltrating water. Alternatively, aquifers with low thickness are less able to dissipate the water. The maximum height of a mound tends to decrease as the thickness of the aquifer increases (Carleton, 2010).
• Large and/or non-rectangular basin. Under the same conditions, circular, hexagonal and triangular basins have been found to have higher groundwater mound than rectangular basins (Domey, 2006).
• Close proximity to a polluted site. If a mound on one site extends to an adjacent property it will raise the groundwater elevation at that property. This is of particular concern if the adjacent site is contaminated since the mound could facilitate the movement of the contaminant plume (Nimmer et al., 2010).
• Close Proximity to a Building or Structure or Underground Utility. When infiltration basins are just vertically above or within 10 horizontal feet of a building, structure, or underground utility, a mounding analysis should be conducted.
• Close Proximity to Other Infiltration BMPs. The presence of multiple BMPs can compound the effect of mounding (Maimone et al., 2011). For example, a study in Syracuse, New York modeled the impact of multiple infiltration BMPs and found that there was the potential for a 0.2 to 0.7 meter mound to form due to close proximity. The height depended on the arrangement of the basins and the soil characteristics.

How to Predict the Extent of a Mound

Both analytical and numerical methods exist that can predict the extent of a groundwater mound. The most widely known and accepted analytical method is based on the work by Hantush (1967). A simple Excel spreadsheet of the Hantush Method was created by the USGS. This method requires the user to input information on the recharge rate, specific yield, horizontal hydraulic conductivity, dimensions of the infiltration BMP, and the initial thickness of the unsaturated zone. The result is considered by experts to be a simplified version of the actual site conditions. The Hantush method can be limited by the assumptions which include no storage loss, uniform and horizontal infiltration, and vertical sides to the BMP. If these assumptions are violated then a more robust numerical method should be used.

A common numerical method is to model site conditions using computer simulations. MODFLOW is the most widely used among the many programs that exist. While numerical modeling can provide a more accurate representation of site conditions, it also requires the user to have considerable training in order to develop the model and run simulations, and interpret the results.

Case Studies

• Machusick and Traver, 2009. The authors studied the effect of stormwater infiltration from an infiltration BMP at the Villanova campus on the shallow unconfined aquifer below. The study was conducted from November 2007 to August 2008. The BMP was a vegetated basin with a 0.53 hectare drainage area and was designed to infiltrate the first 2.5 cm of stormwater runoff. Four monitoring wells were installed to provide continuous monitoring of the groundwater below and around the basin. Monitoring well (MW) 1 was located upgradient, MW-3 was located downgradient, and MW-2 and MW-4 were located adjacent to the site. The results of the experiment show that for storms smaller than 1.9 cm, the upgradient well (MW1) exhibited a larger increase in the groundwater table than the wells closer to the site (i.e., MW2 and MW3). For storms that were larger than 1.9 cm, the reverse was observed. One important note is that the rise in groundwater at MW-2 was attenuated prior to MW-3, meaning the mound did not extend that far laterally. The observed rise in the groundwater at the different monitoring wells was not found to correlate with the amount of rainfall. Infiltration rate was found to be the primary factor that affected the amount of groundwater mounding that occurred, with temperature being the factor that most influenced the change in infiltration rates between stormwater events.
• Thomas and Vogel, 2012. The City of Boston, MA has been periodically experiencing a decline in groundwater elevations over the past century. To try to combat this, the City enacted a code that requires that any new development or redevelopment project install stormwater recharge BMPs. Since the implementation of the code in 2006, 69 recharge BMPs have been installed which has resulted in an estimated 163,450 gallons of recharge per 1-inch storm event. In order to estimate the effect that these recharge BMPs would have on the groundwater level, regional multivariate regression models were developed. The regional groundwater model was developed through the use of groundwater level measurement at 234 observation wells which started between 1999 and 2005 and went through 2009. The final results of the study indicate that the recharge BMPs led to an increase in groundwater elevations. In addition, the model developed during the study can be used by the City to determine the potential impact of future BMPs. While not directly related to groundwater mounding, this study provides some information on the impact of multiple BMPs in one area.