search

ESCOMP / CTSM

Community Terrestrial Systems Model (includes the Community Land Model of CESM)
http://www.cesm.ucar.edu/models/cesm2.0/land/
Other
305 stars 307 forks source link

lake depth raw dataset uses the wrong mask #8

Closed billsacks closed 1 year ago

billsacks commented 6 years ago

Bill Sacks < sacks@ucar.edu > - 2013-03-14 15:22:31 -0600 Bugzilla Id: 1642 Bugzilla CC: rfisher@ucar.edu, subin@berkeley.edu,

Currently, we have a single 3x3min rawdata file that contains both percent lake and lake depth. This file has two masks: one that is a standard landmask, and one that gives a mask of where we have real lake depth data ("LAKEDATAMASK"). Currently, we are using the same mapping files for both of these fields, which (I assume) use the LANDMASK for a mask. This is a problem because it means that the fill value (10m) currently gets averaged with the real data when doing the regridding.

I believe that the right thing to do would be to create a new set of mapping files that uses LAKEDATAMASK as a mask. Once we do that, we might want to split the raw data file into two separate files, so we maintain the notion of a single mask per raw data file. However, apparently Zack Subin has said that this rigorous handling of the mask isn't totally necessary at this point.

Incidentally: I consider this to be a good example of what a file's mask variable should provide: a mask telling you where there are valid data.

billsacks commented 5 years ago

Blocked by #286 - Let's wait to tackle this until #286 is resolved.

billsacks commented 1 year ago

This is probably resolved with the new lake dataset.