I have 3 m resolution bathymetric data of a lake that I want to use to make a volume estimate of a landslide deposit that is on the lake floor. Basically, I want to interpolate the bathy data to make a "pre landslide lake floor" surface in order to get a volume estimate of the deposit.

Here's my thinking:

  1. Draw a polygon slightly larger than the landslide.
  2. Feature vertices to points.
  3. Extract values to points of the bathy data
  4. Interpolate the bathy values of the points using kriging
  5. Extract by mask the interpolated bathy raster to the polygon shape
  6. Extract by mask the bathy raster to the polygon shape
  7. Raster calculator: bathy raster - interpolated bathy raster
  8. Raster calculator: subtracted raster from step 7 * cell resolution
  9. Zonal Statistics table to calculate sum which equals volume.

Does my method seem logical?

Also, my 3 m resolution bathy data is such that the high value (max) is 176 m and the low value (min) is 1.5 m. Should I invert the bathy data?

  • 1
    What happens when you try this? You seem to be asking for feedback on a methodology without told us whether it appears to work or not. As a new user be sure to take the Tour to learn about the site and its protocols. – PolyGeo May 12 '16 at 21:55

You are drawing a wedge through the hummocky splooge. The methodology is fairly crude, but without a "before".. you are just guessing so why waste a bunch of time.

The points you proposed to create will be fine for the deep end of the wedge, but you will need a shoreline from old imagery (pre-slide) or getting something out of some vector data to use as your upper extent (0 depth). This is really all you will need for this.. forget about modifying your existing bathy data. Make a TIN from the points and the "shoreline".

Positive/negative doesn't really matter as long as it is clear in your head. It would likely be easiest to keep things straight if you made it negative and thought of it as depth (unless you are near sea level and this represents elevation).

There is lots of newer tools to do this, but I have always done this with the Cut Fill tool in Spatial Analyst. You will need to convert your Tin to a raster and then just input the 2 surfaces. I do try and do volume calculations with TIN's whenever possible (because arc has a glitch where it discludes external cells with rasters) but with two irregular surfaces I would go another route if I was suspect of the numbers. Always confirm your 2D Area results with the polygonal area of the feature.

Also.. keep in mind that a volume calculation BETWEEN 0 and -100 meters does not include 0 depth (which can account for a fair bit of volume in some cases) or 100. If you want to know FROM 0 you would need to ask for BETWEEN 0.00001 and 100.000001.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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