Roulette,
As someone with a degree in surveying an mapping and who worked in that industry for years before I started flying for a living, I probably have a better understanding than most of the limitations of Google earth. I agree with your comments, especially that the accuracy varies widely in different locations. Certainly there are areas where even the 8 meters +/- accuracy in elevation would be optimistic. And again you're correct that there is no guaranteed level of accuracy anywhere, as here would be in a professionally prepared topographic product.
But given that the elevations at the airport match the published ones very well, it's probable the the errors are in the range of tens of feet, and not hundreds of feet. Which is useful for estimating things like: Did the crash happen above or below the airport elevation, or MDA.
In any case, if as it seems to be that the aircraft was simply way too low at that point, the issues you raise may, or may not, in fact be relevant or significant.
Completely agree with that. I pointed out the PAPI obstacle clearance issue, merely as a point of interest. I never meant to suggest that it was a a factor in this accident, clearly they were far below the PAPI when they started hitting trees.
And it is worth noting that the image of the USGS topographic map posted by aterpster, which *is* a professional topographic product with a specified level of accuracy, shows elevations of around 840 MSL at the high point, and about 750 where the trees were first contacted.