Delineating catchment on our data fails

[mvhulten]

When delineating the catchment, and clipping, using our HSURF-Sphere.tiff (not burnt), according to https://mattbartos.com/pysheds/views.html the output is a $2\times 2$ grid. I expected a larger grid.

Following the example with the elevation.tiff input is without problems.

Flow directions and slopes, on the other hand, look on first inspection realistic, though.

[mvhulten]

I tried this on the HSURFBurned-Sphere.tiff (figure below) as well, but it complains about the coordinate out of bounds:

>>> catch = grid.catchment(x=100, y=100, fdir=fdir, xytype='coordinate')
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/p/home/jusers/vanhulten1/jureca/.local/lib/python3.12/site-packages/pysheds/sgrid.py", line 727, in catchment
    raise ValueError('Pour point ({}, {}) is out of bounds for dataset with bbox {}.'
ValueError: Pour point (100, 100) is out of bounds for dataset with bbox (0.0, 432.0, 444.0, 0.0).

[mvhulten]

For comparison, here is original HSURF-Sphere.tiff and HSURFBurnedAndMod-Sphere.tiff: