Does the common flow accumulation techniques still work over flat area?

Question

My terrain is a highly developed area--which means that

1. It involves lots of flat area ( gradient=0) because they have been properly paved.
2. And it involves sudden drop, such as retaining wall, or drain drop.

These two features make me wonder whether the usual flow accumulation algorithms ( like D8) can actually work on such terrain?

I want to do streamline analysis on it, but I am quite unsure whether the usual flow accumulation algorithms ( like D8) can actually work on such terrain? Or, as my previous question frames it, is there any open source flow accumulation algorithm in C++/C# on Windows that handles these features nicely?

Note: I would have to conduct watershed analysis at modelling stage ( because we need to study the waterflow path even before the development begins, to avoid flooding at unwanted area), which means that the DEM is based on my modelling and not something out there on the actual terrain (yet).

Answer 1

Check out SAGA GIS, it's C++ based, and has tons of options for hydrological analysis, including preprocessing your data (e.g. filling sinks and detecting flat areas). It also has the most common flow accumulation algorithms implemented. I would think twice about using the D8 in any case.

EDIT: More about flow accumulation algorithms: D8 is very sensitive to errors in the input digital elevation model. All the accumulated flow is directed to one neighbor, and sometimes the pixel is erroneous, which mean all the flow accumulation can be directed to a wrong pixel. Other algorithms, which distribute flow to more than one neighbor, are more robust. For example consider an area of 10 by 10 m (10 m resolution), quite often the water is not flowing to one direction, but to two or even more. I'd prefer multiple flow direction algorithm (by Quinn, Freeman), or the Tarboton's D infinity. The choice of algorithm also depends on your data. How accurate is it? Does it have noise, or other errors? What resolution is the data?

The SAGA Wetness Index, which uses the neighboring maximum flow accumulation values can also be useful with high-resolution data (< 2 m, in my personal experience,) which can have noise. It gives smoother flow accumulation patterns, so it is not that sensitive to errors. I have tested these algorithms with LiDAR based DEMs and field-quantified data (top-soil moisture), so I am writing from that perspective, although I must say I have not tested these algorithms in urban environments.

For those who can access, this publication has good, short description about the most common flow accumulation algorithms: https://onlinelibrary.wiley.com/doi/pdf/10.1111/j.1654-109X.2010.01083.x

Answer 2

Water falling on an infinite flat surface should accumulate in place. with a single flow algorithm such as G8, there will be no gradient. Multiple flow direction should distribute the water in all directions, so this could help you. This is available in, e.g., GRASS

On the other hand, a simple workaround would be to "artificially" create a gradient from your drain drops based on the distance to the drains. Indeed, adding a few millimeter in height when you move away from the drain will "imitate" what happens when you "empty" a flat surface (e.g. a dam overflow)