I'm trying to plan a wetland preservation. This is the DEM I have:

enter image description here

My aim is to hold as much rainwater as possible in my area of interest (red poly) by building (as few as possible) new and raising existing dams. As the soils are (mostly) impermeable to water I only have to analyse the surface runoff (and plant transpiration, this is done separately). I already tried calculating the watersheds of surrounding pour points (Different watersheds for same area because of "burned streams"). Since I'm not directly interested in where the water comes from which reaches the pour points but concerned where the water is flowing to which falls as rain, I'm asking myself if this is the right way to handle my task or if there are better solutions.

What would be the correct and best practice way to solve this problem?

  • Hi Simon, hope my or IYDKGJG's answer has helped you out. Please accept one of them as the answer to your question. Dec 23, 2015 at 12:36
  • Hi HDR, I'm without GIS over Christmas time so I wasn't able to do any further research. But I printed out the mentioned JAWRA article and I'm going to read it tommorow when visiting my parents-in-law ;-)
    – Simon
    Dec 24, 2015 at 10:21

2 Answers 2


Yes I believe HSR's approach is a good start. This newly published paper does pretty much as you desire by identifying concentrated flow paths.

It is in JAWRA, the Journal of the American Water Resources Association.

A Geospatial Methodology to Identify Locations of Concentrated Runoff from Agricultural Fields.

DOI: 10.1111/1752-1688.12345

Good news one, is they paid they open access fee so it is public.

Good news two, is they provide their ArcGIS model as a supplement.

Good news three, is they do not gloss over the ArcGIS methods.

Your polygon would be their field margin. This and the DEM are all you need to follow their process. Ensure all data is projected before you start. I would advise your UTM zone in NAD 83 in USA or WGS 84 elsewhere.

This process will identify the exact locations of n concentrated flow paths. Your levees well extend in a linear manner connecting these points. The most high priority areas are the paths with highest values. Do not block any negative values, this is your input water.

My approach when you have the levees' lines would be a zonal max for the levee height unless you have massive relief which is unlikely if this is a wetland.

You see how they have burned through a "levee" using zonal minimum is this answer (Account for bridge overpasses in a ground DEM to allow for water flow under the overpass), you would create the barrier using zonal max.


Is it an idea to sample your DEM along the red line every 10 meter or so (depending on how large your area is)? Then select the point with the lowest elevation, which should be the first point where water will drain from your AOI. From there you can prioritize building or raising dikes by repeatedly selecting the next lowest point.

You can use the Interpolate Shape from the ArcGIS 3D toolbox to sample the DEM to your line.

update: you can subsequently use the Feature Vertices to Points tool. Verify that the DEM value of each vertex is carried to each point as an attribute. Then you can symbolize the point featureclass along its elevation attribute. This should instantly highlight the low and high areas of your areas perimeter.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.