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.

 

Effective Impervious Area in Urban Stormwater Management

July 2014 (volume 9 - issue 2)

Contributed by Ali Ebrahimian, PhD student, St. Anthony Falls Laboratory, Department of Civil Engineering, University of Minnesota; John Gulliver, Professor, St. Anthony Falls Laboratory, Department of Civil Engineering, University of Minnesota; and Bruce Wilson, Professor, Department of Bioproducts and Biosystems Engineering, University of Minnesota.

Funded by the Local Road Research Board of Minnesota.

Introduction

Impervious surfaces have been identified as an indicator of the impacts of urbanization on water resources. Some of the affected characteristics of a watershed due to the increase of impervious surfaces include hydrological impacts (the amount of runoff, peak discharge rates, and base-flow are altered), physical impacts (stream morphology and temperature are changed), water quality impacts (nutrient and pollutant loads increase), and biological impacts (stream biodiversity decreases) (Chabaeva et al. 2009). Although total impervious area (TIA) has been traditionally used as an indicator of urban disturbance, recent studies suggest that a better indicator of urban runoff is the “effective” impervious area (EIA), or the portion of total impervious area that is hydraulically connected to the storm sewer system. Impervious area is hydraulically connected if water travels over an entirely impervious pathway to a stormwater drainage system inlet. EIA is often considerably less than TIA and can vary with rainfall characteristics (e.g., depth and intensity) but as the urban density increases EIA approaches TIA.

Figures 1-a (left) and 1-b (right): Examples of effective impervious area and total impervious area. Fig. 1-a (left): The street is an example of EIA; Fig. 1-b (right): The sidewalk and roof gutters are incorporated into TIA, but may not contribute to EIA except for high intensity and depth of rainfall.

Current and developing management techniques, such as rain gardens, infiltration basins, or pervious pavements, show awareness of the need to reduce EIA, or ‘disconnect’ impervious areas from the drainage system. However, there are no standard methods to assess the impact of these disconnection practices, partly because the connectedness of the existing watershed is not well known. Methods to improve estimates of EIA are not highly researched, and need further investigation.

Importance of EIA

EIA is the most important parameter in determining urban runoff. Knowledge of EIA is therefore critical in rainfall-runoff modeling. The incorrect use of TIA instead of EIA in urban hydrologic modeling leads to an overestimation of runoff volumes and rates (Alley & Veenhuis, 1983). This overestimation results in the overdesign of associated hydraulic structures. Also, directly connected impervious areas are the primary contributing area for smaller storms and therefore, the main concern for water quality (Lee and Heaney, 2003). Stormwater control measures (SCMs), to improve water quality should therefore use EIA in design.

The outcome and applications of this project will eventually lead to the design of a more sustainable urban stormwater infrastructure. Proper identification of EIA will result in more effective planning, locating and design of SCMs, in identifying stormwater runoff pollution sources and environmental pollution control, in cost saving, and in more public consent due to decreasing projects’ size. A wide range of organizations involved in the design of stormwater management, pollution prevention, and transportation structures will benefit from this project. The end users of this research will be cities, counties, watershed districts, watershed management organizations, state departments of transportation, and the consultants who work for these entities in computing and modeling runoff from urban watersheds.

Methods for determining EIA

Impervious area is typically fitted to measured runoff in calibration of hydrologic models. However, it is subject to large errors because the response is correlated with infiltrations parameters, the other important fitted parameter. Currently, EIA can be estimated by analyzing rainfall-runoff data (Boyd et al., 1993), using aircraft or satellite-derived spatial data such as land cover and elevation and GIS techniques (Han and Burian, 2009), empirical equations developed from regression analysis conducted on field calculations (Alley and Veenhuis, 1983), or by conducting field surveys of study sites such as inspection of downspout connectivity, watershed delineation during rainfall events, and identification of street connectivity to drainage system as with or without curb and gutter (Lee and Heaney, 2003). Without a good comparison to EIA determined from rainfall and runoff data, the other techniques to measure EIA cannot be verified. In this article, the statistical rainfall-runoff method for determining EIA will be improved and applied to several urban watersheds in Twin Cities metro area and the results will be presented and discussed.

Rainfall-runoff data analysis method

The most common method for determining EIA using rainfall-runoff data is the method of Boyd et al. (1993). In this method, the runoff depth is plotted versus rainfall depth for each storm in the record. A regression line is then fitted to this data, where the slope of the line is the fraction of total watershed area contributing to runoff. If all events are assumed to involve only impervious runoff (i.e., runoff that is generated from impervious surfaces) (EIA events), then this slope is the EIA fraction (i.e., the ratio of EIA over total watersheds area). Naturally, some points plot well above the regression line, and thus involve a greater contributing area, perhaps from pervious or unconnected impervious surfaces (Janke et al. 2011). Boyd et al. (1993) recommends discarding points that are more than 1 mm above the line and recalculating the regression line. This process is repeated until all points are within 1 mm of the line, at which point the slope is assumed to reflect the fraction of effective impervious area (f_EIA). In the final step of the method, the x-intercept of the regression line represents the initial abstraction of the impervious area (i.e., the depth of water stored on the surface prior to the onset of runoff). The slope of a regression line fit to the excluded points approximates the contributing area of the combined impervious and pervious runoff events (i.e., combined events); significant scatter in these excluded points is generally an indication that the contributing area outside of the effective impervious area (source area) is not consistent, and the initial losses are not fixed. Variable source area and initial losses can be explained by factors like antecedent wetness of the watershed, rainfall intensity, and rainfall duration (Boyd et al., 1993).

We improved the rainfall-runoff data analysis method to address the issues of spatial variation of rainfall and runoff measurement error. As an attempt to include more uniform rainfall data in the analysis, we used a weighted modified coefficient of variation (WMCOV) as a measure of spatial variability of rainfall in watersheds with more than one rain gauge. Then, we excluded the storm data with high spatial variability (i.e., high WMCOV). Part of the scatter of data around the regression line is due to runoff measurement errors that have not been considered in the method of Boyd et al. (1993). In order to address this issue, we used a new EIA criterion (i.e., criterion for omitting combined runoff events from the data in each step of the method) as “standard error + 1 mm” rather than 1 mm in the original method. The improved method was then applied to several urban watersheds mostly in the Twin Cities metro area of Minnesota. Figures 2-a and 2-b show the application of both the original and improved method to a small watershed (MG1) in the City of Maple Grove, MN. The improved method changed the EIA from 17.8% to 16.4% of the watershed area and reduced the number of events that were classified as combined at lower rainfall values.

Figure 2-a (top) and 2-b (bottom): Application of the rainfall-runoff data analysis method to the MG1 watershed. Fig. 2-a (top): The original method by Boyd et al. (1993); Fig. 2-b (bottom): The improved method

Results

The results are summarized in Table 1, where f_TIA (fraction of TIA) is the ratio of TIA over total watershed area. In order to see goodness of fit in linear regressions and evaluate the amount of scatter especially in combined events, the final regression plots are presented for all the watersheds in appendix 1.

Table 1. Results in different watersheds

Row	Monitoring Site Name	Location	Area (ha)	f TIA	f EIA	EIA/TIA
Capitol Region Watershed District, MN
1	Arlington-Hamline Facility (AHUG)	Ramsey County, MN	20.2	0.495	0.181	0.37
2	Como Park Regional Pond inlet (GCP)	Ramsey County, MN	51.8	0.438	0.240	0.55
3	Como 3	Ramsey County, MN	185.8	0.405	0.122	0.30
4	Sarita (inlet)	Ramsey County, MN	376.0	0.367	0.071	0.19
5	Trout Brook - East Branch (TBEB)	Ramsey County, MN	377.2	0.447	0.198	0.44
6	East Kittsondale	Ramsey County, MN	451.6	0.562	0.391	0.70
7	Phalen Creek	Ramsey County, MN	579.9	0.587	0.305	0.52
8	St. Anthony Park (SAP)	Ramsey County, MN	1,383.2	0.548	0.165	0.30
9	Trout Brook Outlet (TBO)	Ramsey County, MN	2,034.8	0.473	0.265	0.56
Three Rivers Park District, MN
1	MG1	Maple Grove, MN	5.5	0.41	0.165	0.41
2	MG2	Maple Grove, MN	3.5	0.39	0.245	0.63
3	P1	Plymouth, MN	5.1	0.38	0.208	0.55
4	P2	Plymouth, MN	6.8	0.35	0.114	0.33
5	P3	Plymouth, MN	5.6	0.27	0.104	0.38
City of Minnetonka, MN
1	Hedburg Drive	Minnetonka, MN	2.8	0.88	0.549	0.63
2	Mayflower Ave (Tapestry)	Minnetonka, MN	11.1	0.24	0.180	0.76
City of Bloomington, MN
1	Smith Pond	Bloomington, MN	55		0.136
2	Mall of America (MOA)	Bloomington, MN	202		0.094
City of Madison, WI
1	Harper Basin	Madison, WI	16.4		0.293
2	Monroe Basin	Madison, WI	92.9		0.232

Discussion

In order to evaluate the usefulness and limitations of the statistical rainfall-runoff method and provide a basis for comparing to outcome of the other EIA estimation methods (e.g., GIS method), 20 monitored watersheds were analyzed in this study. Eighteen are located in the Twin Cities metro area, and the remaining two watersheds are located in the City of Madison, WI. In order to provide a better understanding of the urban runoff mechanisms, the analysis has been performed on a wide range of watershed areas from less than 3 ha to 2,035 ha. f_EIA values ranged from 0.07 to 0.55. At one extreme is Sarita watershed in Ramsey County, MN with a f_EIA of only 0.07, and at the other extreme is Hedburg in Minnetonka, MN with a f_EIA of 0.55. The main land use within the Sarita watershed is institutional. The fraction of TIA in the watershed is about 0.37 and it encompasses the Minnesota State Fair grounds and open spaces in the University of Minnesota St. Paul Campus. In contrast, Hedburg is a watershed with commercial land use and high density of roadways, sidewalks, and parking lots. Total impervious areas form about 88 percent of the watershed. Excluding these two extremes from the results, the average and standard deviation of f_EIA are 0.2 and 0.08, respectively. These are primarily residential watersheds. Combined runoff events were absent for 6 watersheds including Sarita, Harper, Monroe, MOA, Hedburg and Tapestry. This can be explained by the land cover or the small rainfall depths available in their data set. Table 1 shows the average and median of the EIA/TIA ratio as 0.48, with a standard deviation of 0.16.

Conclusion

The rainfall-runoff data analysis method has the advantage of being quick and relatively simple to implement, as it does not require familiarity with specialized software tools (e.g., ArcGIS). While the results of this method provide a better understanding of the urban runoff mechanisms in the watersheds of study, they can be used as accurate estimations of EIA that would benefit a wide range of organizations involved in the design of stormwater control measures. They can also be used for verification of other EIA estimation methods. A major down side of this method is that it is not applicable to un-gauged watersheds. To address this issue, we are developing a new GIS based method for determination of the fraction of EIA in urban watersheds. This method will be based on the integration of GIS and the NRCS-Curve Number (CN) method.

References

• Alley, WM, and JE Veenhuis. (1983). Effective impervious area in urban runoff modeling. Journal of Hydraulic Engineering, 109(2), 313-319.
• Boyd, M. J., Bufill, M. C., & Knee, R. M. (1993). Pervious and impervious runoff in urban catchments. Hydrological Sciences Journal, 38(6), 463-478.
• Chabaeva, A., Civco, D. L., & Hurd, J. D. (2009). Assessment of impervious surface estimation techniques. Journal of Hydrologic Engineering, 14(4), 377-387.
• Han, W. S., & Burian, S. J. (2009). Determining effective impervious area for urban hydrologic modeling. Journal of Hydrologic Engineering, 14(2), 111-120.
• Janke, B., Gulliver, J. S., & Wilson, B. N. (2011). Development of Techniques to Quantify Effective Impervious Cover (No. CTS 11-20). St. Anthony Falls Laboratory.
• Lee, J. G., & Heaney, J. P. (2003). Estimation of urban imperviousness and its impacts on storm water systems. Journal of Water Resources Planning and Management, 129(5), 419-426.

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:

• Have you ever calculated EIA in your area or used it in hydrologic modeling? What method did you use for EIA determination?
• Do you have qualified data in your watershed to use the rainfall-runoff data analysis method for EIA determination?
• Disconnecting impervious surfaces to reduce EIA is a key SCM to control runoff. Are such actions feasible in your watershed?

Appendix 1

Capitol Region Watershed District, MN

Three Rivers Park District, MN

City of Bloomington, MN

City of Minnetonka, MN

City of Madison, WI