I am wondering if it is possible to delineate and create small polygons closer to the points of interest and then perform your tilting procedure on each small polygons polygon. This'll give the maximum elevation for each little polygon.

Interesting but the first step is to discover the "points of interest" as you call them. They are just where you have the most relative relief (height) above the surrounding seafloor in the polygon (as it is for Bathymetry).

Dear David,

As far as I understand, what you need is trend removal, that is make your terrain dominantly horizontal, where you have just to identify the location with maximum elevation. I never tried to remove trend from a polygon, but converting it to raster may help you.

Good luck,

László

Hi David,

From yours and László's ideas, here is my suggestion:

Use the points on the perimeter to compute the linear trend - which you call slope.

For each elevation point within your polygon, calculate its distance to the trend plane.

I haven't tried the following solution, so you might need to tweak it.

Let's say you use ArcGIS, with a raster DEM (bathymetry) and a polygon vector. Make sure that the DEM has a projected coordinate system with unit length=meter, like the elevation unit.

1) convert polygon to polyline (to get the perimeter)

2) extract raster values under polyline (Extract by mask: https://m.youtube.com/watch?v=2Fs5LBC9T_U)

3) convert extracted raster to points

4) compute trend interpolation (Trend Tool, polynomial order 1) with extracted points, and get the plane equation from the rms file: that file gives coefficients c0, c1, c2

z = c0 + c1 x + c2 y

Or

c1 x + c2 y - z + c0 = 0

5) convert DEM raster to points (RasterToPoint), the field value being Z coordinate

6) add fields X and Y to converted feature class, and calculate the X and Y coordinates

7) add 1 field d for the distance to trend plane, and calculate that distance like this:

https://mathinsight.org/distance_point_plane

Leave out the absolute value in the expression of d because you want to know if the point is above or below the plane.

Good luck!

Interested ,Try zonal stat

What you are describing is 'Local Dominance', but there are also quite a number of other functions similar to that. In Qgis and ArcMap, zonal statistcs can give you a crude answer, I as a rule use the Relief Visualization Toolbox (RVT) https://iaps.zrc-sazu.si/en/rvt#v

Thanks