Rain on Buildings in the 2D Domain
Do you want Green Eggs on Toast? in the car?
After writing last week's post about Buildings in the 2D domain, I feel like Dr. Suess’s Sam-I-am asking you how you like your Green Eggs and Ham. CoPilot turned the ham green, and the eggs are sunny side up.
RECAP
Buildings are an important component in the hydraulic flood model. They are the primary reason for Flood Insurance and the metric by which we attempt to reduce flood damage. However, there are several different ways buildings get modeled, both hydraulically and hydrologically.
In the Buildings in 2D post, I discussed some hydraulic methods for representing buildings in the 2D domain and highlighted a paper that TUFLOW’s Bill Syme published in 2008. I would encourage you all to read that paper.
Rain on Grid
In this post, let’s discuss Rain-on-Grid and how buildings interact.
Let’s start with the question.
How might one model a building when rain is applied to the mesh or grid?
With the giant piece of toast in the image above, it is hard to suggest how rain runs off the surface. However, I see downspouts on the three visible sides of the building, so we can assume a diagonal sloped roof to each building edge.
IF, we use the high Manning’s roughness of 10, water fails to runoff this surface at the speed to which it should. After reading the Haris County Flood Control research papers didn’t discuss any criteria for using 10 as the roughness coefficient.
IF, we use use the Digital Surface Model approach, the velocity of the water falling becomes an issue.
Other Papers
The Engineers in Australia organization updated their modeling practices several years ago after devastating floods and published updated standards in the Australian Rainfall and Runoff guidelines. The TUFLOW wiki references the following information from the Australian Rainfall and Runoff Guideline, Project 15 (Representation of Buildings in 2D Numerical Flood paper.
Using depth-varying Manning's n over the area of the building footprint, with the application of a lower value (n = 0.02) at shallow depths (d < 0.1 ft) and a higher value (n > 0.3) at more significant depths (d > 0.33 ft):
This is a very common and easy-to-implement option. The low Manning's n value mimics the quick runoff response associated with drainage from the roof. The higher Manning's n value represents the losses associated with deeper floodwater impacting the side of the building.
Raise the building footprint elevation using TUFLOW's topographic update features (eg. Read GIS Z Shape):
Raising the model topography creates an obstruction to flow. It prevents floodwater from passing through buildings (as is the case with Manning's n approach)
However, applying rainfall on top of the building can produce some undesirable results that require further consideration. Water falling from the building roof to the ground can require a reduced model timestep to maintain model stability, slowing the simulation speed. Also, depending on the Map Cutoff Depth assumptions, water may be present in the results on the building rooftops. This may not be desired for mapping purposes.
InfoWorks ICM
In Autodesk's first quarter last year, they introduced a building object in the 2024 release. This object collected network results, runoff, roughness, and porous polygons into a single object. This should simplify model construction with the buildings in the rain-on-mesh scenario with an easy-to-use runoff model and discharge point or link.
The runoff portion of the polygon suggests a slope and storage lag, along with the ability to limit drain capacity to mimic a downspout system with excess flows off to the mesh or lost to the unknown.
It is rare to see a building dataset with this type of information here in the States. However, I know a few cities that have modeled to this level of detail.
The 2025 release added the ability to send excess flows to the perimeter, which requires marking to limit drain capacity. To get what I requested in the feature, I now have to mark the limit and set the drain capacity to zero to model buildings with no downspouts. This appears in an interesting direction but is still on the bleeding edge and not yet ready for gutterless buildings.
Here is the Autodesk training on Buildings.
HEC-HMS
I haven’t seen an HMS model with buildings, as I don’t use HEC-HMS often. They likely exist but would be highly impervious elements with lower runoff numbers and little ability to send flow to anything but the mesh.
It will be interesting to see where this may go when RAS can run 2D runoff with the pipes.
TUFLOW & XPSWMM
With XPSWMM running the TUFLOW engine, we can give the buildings a depth-varying roughness. The TUFLOW wiki describes a few alternative scenarios for applying rain outside buildings or using Read GIS SA RF or Read GIS SA RF PITS to spread rainfall off to a pit or to the 2D grid.
Maybe we can convince TUFLOW to build a 2D Building object that contains similar features: a single polygon with roof characteristics (slope) that initially allows low roughness, together with either a blocking/permeable and trigger workflows that enable the building to leak slowly before failing in a single polygon.
Maybe next week, I can build the test model.