UPDATES: June 2013

UPDATES is a monthly email newsletter on stormwater management, assessment (including monitoring), and maintenance research at St. Anthony Falls Laboratory and the University of Minnesota.

Click HERE to subscribe to UPDATES Newsletter

Or Text SWUPDATES to 22828 to Join. Message and data rates may apply.

For Email Marketing you can trust


Importance of Storm Drain Baseflow to Nutrient Loading

June 2013 (volume 8 - issue 6)

Contributed by Benjamin D. Janke, Jacques C. Finlay, and Sarah E. Hobbie (Ecology, Evolution, and Behavior); and Larry A. Baker (Bioproducts and Biosystems Engineering), University of Minnesota.

UPDATES would like to congratulate Sarah E. Hobbie for her election to the National Academy of Sciences. This UPDATES article is the second by her team and colleagues, so we would also like to welcome all of them to the urban hydrology and water quality field.


Nonpoint nitrogen (N) and phosphorus (P) pollution is the primary driver of eutrophication of urban surface waters, and remains an obstacle to the effective management of urban water quality. Due in part to the efficiency of the urban drainage network in removing runoff from the land surface, much previous research has focused on understanding the loading of nutrients to urban lakes and streams from stormwater. By contrast, the contribution of baseflow – defined here as the water moving through storm drains between rainfall events due mostly to groundwater leakage into storm drains and to outflow from drain-connected surface water – has rarely been addressed in previous work despite its potential impacts on receiving waters and the unique challenge it would present from a management perspective.

In this study, we investigated the importance of storm drain baseflow for water and nutrient loading in a large urban watershed using an extensive monitoring data set provided by the Capitol Region Watershed District (CRWD; http://www.capitolregionwd.org). This data set, collected by CRWD from 2005 – 2011 as part of its stormwater monitoring program, included continuous flow measurements and nutrient chemistry of water samples collected during both stormflow and baseflow periods throughout each year at sites installed primarily in storm drains and at outlets of best management practices. From these data we calculated stormflow and baseflow loads of total phosphorus (TP), soluble reactive phosphorus (SRP), total nitrogen (TN), nitrate-nitrite-nitrogen (NO3-), ammonium nitrogen (NH4+), and total organic nitrogen (TON) for the 6 largest CRWD sub-watersheds. These sites, which are located mainly within St. Paul, MN, include East Kittsondale (EK), Phalen Creek (PC), St. Anthony Park (SAP), Trout Brook East Branch (TBEB), Trout Brook West Branch (TBWB), and Trout Brook Outlet (TBO) (Figure 1), and range in size from 327 ha (808 ac) at TBEB to 2038 ha (5036 ac) at TBO and are relatively similar in terms of urban land use.

Figure 1: Monitoring sites and corresponding sub-watersheds in the Capitol Region Watershed District. Note that the Trout Brook Outlet sub-watershed includes both the West Branch and East Branch Trout Brook sub-watersheds.

How important is storm drain baseflow to water yields?

We found that baseflow in CRWD storm drains contributed more than half of the warm season (April – October) water loads, ranging from 56% to 66% of total seasonal water yield (inches/season) at 4 of the 6 sites (Figure 2). In addition, average baseflow water yields were more variable across sites than stormwater yields, varying from 8.4 to 36.7 cm/season (compared to 11.0 to 24.9 cm/season for stormflow).

Figure 2. Mean seasonal (Apr-Oct) water yields of baseflow and stormflow as observed in 6 drains in CRWD from 2005-2011.

The magnitude and variability of baseflow water yields suggests a variety of baseflow sources may be present across sub-watersheds. In CRWD, baseflow in storm drains likely results from two primary sources: (1) groundwater that seeps in through joints and cracks in aging stormwater pipes constructed at or below the water table, and (2) outflow from surface waters, such as lakes, ponds, and wetlands, that are connected to storm drains.

With these sources in mind, some of the variation in water yields may be explained by the developmental history of the watershed. For example, the main storm drains in the Trout Brook and Phalen Creek watersheds (which include the TBWB, TBO, and PC sites) were constructed in stream channels beginning in the late 1800’s. Therefore these drains are located in areas with concentrated shallow groundwater flow and high water tables, likely enhancing seepage rates of groundwater into the drains. Accordingly, these sites had the highest baseflow water yields.

The presence of surface water in the sub-watersheds may also explain some of the variation in baseflow water yields. Several sub-watersheds, including TBWB, TBO, and SAP, have a relatively large number of lakes, ponds, and wetlands that may supplement baseflow. By contrast, EK has almost no surface water and a smaller storm drain network that may be mostly above the water table, which could explain its relatively small baseflow water yield.

How does baseflow in storm drains impact total nutrient loading?

In the 6 major CRWD storm drains, baseflow delivered a significant proportion of the total seasonal TN load, ranging from 33% to 68% across sites (51% on average; Table 1). Baseflow was less important for TP, contributing 8% to 34% of total seasonal TP, or 21% on average (Table 1). TN, and to a lesser extent TP, showed considerable site-to-site variability in baseflow, suggesting a range in baseflow sources to storm drains.

Table 1: Mean seasonal loads of Total Phosphorus and Total Nitrogen in baseflow and stormflow (in kg/ha and as a percent of the combined seasonal load) at six sub-watersheds in CRWD.


Total Phosphorus

Total Nitrogen

















East Kittsondale (EK)











Phalen Creek (PC)











St. Anthony Park (SAP)











Trout Brook East Branch (TBEB)











Trout Brook West Branch (TBWB)











Trout Brook Outlet (TBO)











Concentrations of various forms of N, in particular NO3- and TON, may be useful for explaining whether baseflow was coming from groundwater vs. surface water for the six CRWD storm drains (Figure 3). For example, separate observations show relatively high concentrations of NO3- in shallow springs in and around CRWD (data not shown). Therefore the high NO3- concentrations observed for PC and EK, which are both sub-watersheds with very little surface water connections to storm drains, may be evidence of shallow groundwater as the primary baseflow source in these sub-watersheds. The source of high NO3- in groundwater is not known, however.

Figure 3. Flow-weighted average concentrations of nitrogen forms in baseflow at the six main CRWD sites. NH4+ = ammonium N, NO3- = nitrate-nitrite N, and TON = total organic N (particulate and dissolved). Total Nitrogen (TN) = TON + NO3- + NH4+.

TON, which includes both particulate and dissolved forms of organic N, made up the majority of baseflow TN in several sub-watersheds. Most of these sub-watersheds have surface water, so the presence in storm drains of organic N, which may come from algae and detritus in surface water, could be evidence of outflow from lakes and detention ponds connected to the storm drains of these watersheds.

What are the implications of baseflow loading by storm drains, especially for nitrogen?

Stormwater dominated loading of P by CRWD storm drains, consistent with the expectation that most P is mobilized and transported in surface runoff processes. By contrast, baseflow contributed substantially to N loading because of the prevalence of TON and NO3- in baseflow at most sites.

These results have implications for water quality management, in particular where N loading is a concern. The presence of TON in baseflow suggests that the impacts of outflow from lakes and detention ponds during dry periods may need to be considered in the development of management practices.

In addition, the results suggest that infiltration practices may have the potential to pollute shallow groundwater, specifically in places where storm drains intersect shallow aquifers and provide a “short circuit” to natural sub-surface flow paths. However, while high NO3- concentrations in baseflow may be an indicator of groundwater input to the storm drains due to the mobility of NO3- in groundwater, the source of NO3- is not known. One major N source is thought to be leaching from lawns, which can be sites of large inputs of fertilizer, pet waste, leaves and lawn clippings.

Therefore in watersheds with a significant amount of infrastructure below the water table and where high rates of sub-surface water movement may be present (due to buried streams, sandy soils, etc.), such as in CRWD, new approaches to water quality management may be necessary. These approaches could include taking steps to ensure that existing BMPs are efficient N traps through denitrification, or reducing N (and P) sources to the land surface in upstream watersheds, such as through appropriate lawn fertilizer applications, lawn clipping and leaf disposal, etc.

We want to hear from you!!!

Let us know your thoughts, experiences, and questions by posting a comment. To get you thinking, here are a few questions:

  • If infiltration practices can cause pollution of groundwater in some locations, what is the trade-off in terms of water quality impacts between keeping stormwater on the landscape vs. allowing it to become runoff? In other words, is it possible for enough infiltration BMPs to be present in a watershed to have a measureable impact on nutrient export?
  • What alternatives to infiltration BMPs exist for reducing runoff?
  • What are some other approaches that might be employed to control N and P at the source (e.g., through creative land or water management)?