nmt28 / C-SHELPh

Apache License 2.0
13 stars 6 forks source link

Use of bathymetry model for Inland water-bodies #6

Open amanbagrecha opened 1 year ago

amanbagrecha commented 1 year ago

Hello @nmt28 Thanks for the scripts and thoughtful documentation. I got to run the example you provided but was wondering if the same could be applied to inland water-bodies.

The example you've illustrated is of the ocean. If the same is applied to inland water-bodies, what changes would be needed, if any.

I know for the fact that water bodies (inland or otherwise) for which bathymetry from icesat-2 is to be derived, has to be clear. On what parameters can we confidently say that the derived bathymetric output is accurate enough.

Thank you!

nmt28 commented 1 year ago

Hi.

I'm glad you found it useful. I would suggest only one change for inland water bodies, related to the refraction correction of the laser at the air/surface interface. As sea water is saline, it has a different refraction coefficient than fresh water.

This should be helpful: http://research.engr.oregonstate.edu/parrish/index-refraction-seawater-and-freshwater-function-wavelength-and-temperature#overlay-context=research

Otherwise, no other changes are needed. Inn terms of what is considered accurate, this is often subjective. From the example you can see that the bathymetric surface is very clearly defined. If this is not the case then it is much harder for the code to find a surface.

nmt28 commented 1 year ago

This looks satisfied so I'm closing it out

nmt28 commented 1 year ago

I wanted to reopen this as I have realized that additional changes would be needed to be made.

C-SHELPh does an orthometric correction on the data so that the sea surface is effectively at 0m. Thresholds are then set to help the algorithm find the sea surface. If you are studying inland lakes, these are likely at a higher elevation than sea surface so the thresholds would need to be modified