I'm building a model of a city in the Netherlands and i would like to simulate different intensities of rainfall (40 mm/h, 80 mm/h, 120 mm/h etc). I would like to see the differences these rainfalls make in the lower parts of the city, where most off the water accumulates. I've tried pretty much every hydrology tool from the toolbox, but i can't figure it out.
I've tried using the toolbox provided in this ArcMap tutorial (Find Areas at Risk of Flooding in a Cloudburst | Learn ArcGIS) but I'm not getting any decent results.
Does anybody have any idea on how to solve this problem?
What did you get? What do you expect? Perhaps their model is as best as it gets given the inputs?
I'm the author of the Learn ArcGIS Cloudburst exercise, but what's available there is the first component, only, of a series of steps to estimate where and how much stormwater accumulates on terrain when asuming Horton flow conditions. At the Esri UC 2017 I presented a paper on the next steps: http://proceedings.esri.com/library/userconf/proc17/papers/2250_701.pdf . Let me know if this points in the right direction.
So far my workflow is implemented in Desktop, only, as I'm using the Geometric Network component not available in Pro. Apparently, a new Pro extension (Utility Network Analyst) will have the tools necessary in the future to programme a similar custom trace tool as mentioned in my presentation.
Thomas Balstroem, Assoc.Prof.
Dept. of Geoscience, Univ. of Copenhagen, Denmark
As you probably have heard ArcGIS Pro 2.6 introduced Trace Networks, which have similar capabilities of the geometric networks in ArcMap. There is a tool available to convert a geometric network to a trace network.
So this is probably a long shot, but would you be willing to update the Find areas at risk of flooding in a cloudburst lesson for ArcGIS Pro?
Ewout ter Hoeven
Thanks for asking.
Actually, it's done, so a new extended Cloudburst lesson will be available shortly from Esri's Learn site. The only thing missing is the copy-editing of my stuff, so I expect the new tools to be available before the end of May. The new lesson will model the accumulated downstream spillover from any bluespot present within a drainage basin for several rain events as originally requested by Jochemvan der Kop who strated this thread.
I tested the new Trace Network but a tracing component was missing. So, having consulted the developers, I decided to encode my own traceable network apart from that environment based on a Python plugin called NetworkX.
Awesome, thanks for the effort, I look forward to it!
Is there any chance I could get a preview of the new Cloudburst lesson? We're currently doing a project at the TU Delft researching flood risk in Helsinki, and I really would like to try if your process and methods are portable to our data set (we have generated a DEM of Helsinki).
If you find it useful, I could also provide some feedback on the lesson from a students point of view.
I would also greatly appreciate a preview of the new Cloudburst lesson! The (older) tutorial has been of great help so far and is really unique in its use of the custom ModelBuilder when compared to other flood risk related tutorials.
As Ewout mentioned, if there is any possibility of a preview of the newer version that would be amazing! Feel free to contact me if necessary Thomas or Ewout. I'll be following this thread closely.
Thanks once again!
Hi guys and thanks for asking. I forwarded your inquiry to Colin Childs at Esri who is doing the copy/editing of the new Cloudburst lesson. I expect to hear from him shortly.
Hi @Anonymous User !
Cian and I are currently working on research projects on urban resilience, for which we need to map flood risks in Helsinki. The model Thomas build in the Cloudburst lesson is exactly what we're looking for. We understand it will be made available for ArcGis Pro in the coming weeks, but we're on a bit of a tight timeline. Is there any chance we could get a preview a bit earlier? Doesn't need to be polished, a rough draft is more than enough.
If that's not possible that's perfectly fine, but then we know we need to take a different approach or reverse engineer it ourselves.
Ewout ter Hoeven