Q about Mt Hood dataset

[accidental_PhD]

(11-08-2014 03:24 PM)mcarter Wrote:  I'm curious about the Mt Hood data set. I'm wondering what the values represent.

It seems to be some variant of altitude, but it can't be altitude alone. If it were, the shape of the mountain would be like a mountain till altitude 128, then, everywhere just inside the region enclosed by the 128 isocontour would drop to -127 altitude, which rises back up to 0 around the center of the enclosed 128 isocontour. However, traditional wisdom (as well as looking at a picture of Mt Hood) tells us that Mt Hood's shape should be similar to most other mountains.

I guess I'm just curious whether this is a common method for ranging altitude data, and if so, what benefit does ranging the data like this serve (perhaps topological maps, where each band starts at a value of a signed byte MIN_VALUE and goes to MAX_VALUE)?

If my guess about use in topo maps is correct, the reason for doing this would be to save space in the representation of the altitude data. However, if this is the case, to be able to find the actual altitude at a point, you would need an algorithm that would count the bands traversed to go from one at sea level to the band the current point is a part of. This seems a rather convoluted way of representing the data.

Can you give some insight into this data set?

This is an excellent question, and a great example of what we encounter every day working with domain scientists and their data! Miriah or Hitesh may have more detail, but I don't know much about its source – it's been a while since I did a variant of this assignment, and I had actually forgotten about this anomaly.

I think your inference about storage is correct. We won't require you to repair the dataset, but this is an important thing to discuss in your report (e.g. describe how you discovered the anomaly, what you might do to fix it with more time, questions you would want to ask whomever generated the dataset, etc.).