there is a weird at elevation 22. Also, maybe include negative elevation for oceans ?

Ok thank you. It was because the first point in x was not the sea. It's corrected.

hi @neocartocnrs, very cool notebook! I had some fun flipping the axis! https://observablehq.com/d/c2aa797f5f7eae14. I'm not sure if it's correct ! would it make sense to offer an option to switch x for y?

That's cool :-) But I suggest tu use dem2.csv instead of dem.csv (dem.csv gives the value of the max altitude per pixel while dem.csv gives the average value)

ohh yes, I guess I've forked you project before your mean/max update