Percent impervious


Do we need to define the "percentage impervious (%)" attribute to account for the runoff coefficient? 

Or SewerGEMS would automatically pick up the net/scaled flow by using the Curve Number?

  • Percent Impervious field is applicable to EPA-SWMM method only, you don't need to consider that for unit hydrograph runoff method.

    The proper hiding of the percent impervious field was implemented in a patch for version If your organization has a SELECT subscription, Technical Support can provide you with a link to the patch for your IT/Helpdesk to install. However, this will only simply hide the field to reduce confusion. This change will also be included in the next new version release.

    Percent impervious field - which implies percent of land area which is impervious, please search for Percent impervious in help documentation. File > Help >SewerGEMS Help. 

    Assumptions behind Subarea Routing with the EPA-SWMM Runoff Method 


    Sushma Choure

    Bentley Technical Suppport

  • I disagree that you "don't need to consider" percent impervious for calculations involving the NRCS unit hydrographic method.  It very much depends on the rainfall amounts being considered and the purpose for which the hydrologic calculations are being performed.  The NRCS runoff equation is non-linear.  For rainfall events with small rainfall amounts, if percent impervious is not accounted for by separate calculation of runoff for pervious and non-pervious areas, the resulting calculation of total runoff volume can be significantly underestimated.  This is not much of a concern for the larger storm events for which we typically size drainage infrastructure, which are sized based on discharges anyway.  However, it is certainly a concern for many volume-based water quality BMPs or SCMs that are being designed these days.  If you are designing a volume-based water quality BMP for a 1.25" rainfall amount, and you are using the NRCS runoff equation, and your contributing catchment has a mix of both pervious and impervious areas, you better take into account percent impervious or you are going to significantly undersize your BMP.

    The original post mentioned a "runoff coefficient", which sounds like Rational Method.  However, based on the screen capture it's clear that the NRCS method is being used.

  • In response to Karl's points: although the field called "Percent Impervious" is only applicable to the EPA-SWMM runoff method, one can still account for impervious area with other runoff methods. If there is a need to use the Unit Hydrograph Method with the SCS Loss method and a portion of the catchment area is impervious, there are a few options to account for the impervious area:

    1. Model it as two separate catchments both with the same Outflow Element, one configured to model the pervious area and corresponding CN and the other to model the impervious area (for example with a SCS CN of 98). 
    2. Use the "Subareas" option for specify the component parts (the portion that is impervious vs. pervious) and have it calculate the resulting composite CN.
    3. Adjust the catchment's SCS CN to account for impervious area

    See: Multiple CN numbers for a catchment and multiple catchments with a single outlet node

    However as seen in the article Sushma provided, the routing of the impervious and pervious areas can be more complex. The engineer may need to decide on an appropriate and/or conservative approach or consider using the EPA-SWMM runoff method with the "Subarea Routing" options.

    are you saying that you find method 1 to provide more accurate runoff volume for smaller storms?

    Note: the term "runoff coefficient" could also refer to the separate "Runoff Coefficient" Loss Method which is available for some runoff methods like Time-Area and is different than the Rational C or SCS CN. However from the screenshot provided, it appears that the SCS CN loss method is being used.


    Jesse Dringoli
    Technical Support Manager, OpenFlows
    Bentley Communities Site Administrator
    Bentley Systems, Inc.

  • Thank you for the response Jesse.  Yes, Method 1 in your post is the accurate method to use.  The issue with Methods 2 and 3 is that the losses that occur for the pervious portion of the catchment are "averaged" into the impervious portion, and this leads to error in the calculation of runoff volume, especially for small rainfall amounts.  To illustrate, consider a 2-acre catchment that is 50% impervious, with 1.25" rainfall:

    Pervious portion: 1 acre @ CN = 65, Volume of runoff = 20 CF
    Impervious portion: 1 acre @ CN = 98, Volume of runoff =  3755 CF
    Total Volume of runoff = 3775 CF

    Using composite CN:
    2 acre @ CN = 81.5, Volume of runoff = 1500 CF.

    That's a 60% underestimate of volume.  If rainfall amount increased to 9.0", the difference is negligible.  

    This is due to the non-linear nature of the NRCS runoff equations.  For Curve Number < 62, the Initial Abstraction is greater than a rainfall amount of 1.25", and there is zero runoff.  This condition should not be "averaged" into the impervious portion of the catchment, which has nearly zero losses.  It's been a while since I used Pondpack, but I remember that you could specify a % impervious for a catchment and it would calculate the runoff separately for the pervious and impervious portions - it would show you the separate calculations in the output.  I was surprised to read that this is now available only with EPA-SWMM method.  But perhaps I misunderstood the post.

    My point is not that Methods 2 and 3 above are improper.  They may be appropriate in most cases.  It just depends on the purpose of the analysis and the specific characteristics of the catchment being analyzed.  When designing stormwater control structures for small rainfall events, many jurisdictions will require that estimation of runoff include only the contributing impervious areas in order to avoid this problem.