As I am writing this I have spent almost a month in CHI’s office in Guelph getting my master thesis project off the ground.

The project aims to somehow improve how SWMM models the underdrain in LIDs. I will soon have an introduction uploaded as a research paper here on openSWMM.
I have found several research papers (and posts here on openSWMM) mentioning the lack of physical representation of some processes in LIDs that relate to the underdrain flow. I have also looked at design manuals which often suggest design practices that SWMM cannot be configured for without workarounds.

I have already been discussing the topic with CHI’s staff as well as Ryan Winston and Jay Dorsey who has taken the time to share insights from their research on LIDs. Now I am asking you for additional input.

The following is what I have found so far:

• SWMM has a simplified way of e.g. treating water movement and moisture distribution in the soil layer (especially unsaturated part) compared to models like DRAINMOD which accounts for a range of parameters. I am trying to figure out how much I should look into what parts of DRAINMOD that could be applied to SWMM. As DRAINMOD is built for agricultural modeling, much of what it accounts for could be negligible if the drain is laid in gravel, especially Hooghoudt flow toward the drain. However, I have been informed that there are cases where underdrains are placed in the soil layer which definitely could make the DRAINMOD approach preferable.
• In this forum, Bill Lucas in mentioning that head losses for driving flow through the soil media should be accounted for, then there should also be head losses to drive water through the perforations/slots of the drain pipe.
• The reference manual advices how to set the current Ch^n equation for underdrains when utilizing manning equation or orifice flow. The settings however require separate calculations by the user. SWMM already has built in functions for a more comprehensive orifice and conduit flow. It seems to me that both the interface could be more user friendly and the calculations be more representative. Storage nodes seem to be a common workaround that should be possible to integrate in LID’s in a similar form (both for user input and calculations).
• Underdrain outlets placed above the storage layer (commonly as an elevated outlet for IWS) cannot be represented in LIDs.
• Overflow devices connected to the underdrain also cannot be modelled in LIDs. Sometimes overflow is even dealt with as an extra underdrain at a higher elevation, which is not an option in SWMM either.

All this being said, I am aware that increased complexity can increase computation time. However, I believe that there should still be room for improvements and I am all ears to your opinions on what you think is important or more applicable to SWMM.

I have access to some data to calibrate with but the more the merrier, so if either you are able to share LID field measurements that would be greatly appreciated.