Hi Johnny,
Okay, attached is a project with a text input dataset in the data directory. These 2 dataset are the exact same grid spacing, value and number of items with the notable exception that one was created ~74N and the other ~33N.
If you run the test app, you will find 2 buttons. Button 1 loads the Northern dataset and button 2 loads the southern dataset.
Because both datasets use a static 500m depth value, immediately, you will be able to tell that the northern datasets good data covers much less area than the central dataset. If you examine the .grd files you will see that there are many more cells in the Northern dataset than the southern, but yet both grids are supposedly covering the same space. In fact, the Northern grid file is much larger size than the southern. My first indication that something is wrong.
If you then go into the code to the procedure called GetBathyPoints, you will see a block of code with my comments. Disable the code that converts the input coordinates to spherical mercator and run the analysis again. You will see that the grids now cover an identical area and the file sizes are equivalent; however, my environment doesn't put them where they should be because I want to use spherical mercator as the base grid.
So, the problem I have is that the coordinate locations of the grid are only correct in my spherical mercator environment when I convert the UTM point coordinates to spherical mercator. But, the interpolation is incorrect and doesn't show 500m depths over the required area. If I process the grid in UTM, the interpolation is correct, but the coordinates are not showing where they should because I have setup my environment in spherical mercator. Since GridFeatureLayers write their data to disk, I would guess the solution would be to have a Projection property for GridFeatureLayer and IsolineLayer that can handle reprojecting this, but so far my attempts at doing this have caused hangups in the software.
Best Regards,
Damian
SphericalMercatorGridTest.zip (87.5 KB)