pinga-lab/paper-moho-inversion-tesseroids ? reverse-engineered prompt

Build me a small research project that reproduces the main results from a paper about estimating the South American Moho from gravity data in spherical coordinates.

I want a Python based setup with Jupyter notebooks that can run the inversion, generate the figures, and save the final Moho depth model as an ASCII xyz file. The core should include the non linear gravity inversion code, using a spherical Earth approximation and tesseroid based modeling, plus a few notebooks for a simple synthetic example, the paper figures, and the South America case. Include the needed data files and make it easy to run from a notebook after setting up the environment. If some parts take too long to run, it’s fine to keep them as longer workflow notebooks and make the lighter example easy to test first.

Please also include a clean README with clear steps for setting up the environment and running the notebooks, and make sure the project can be reproduced end to end. If you need to check current notebook or conda docs online while wiring it up, go ahead.