- 20 May, 2015 6 commits
-
-
Pierre Cazenave authored
-
Pierre Cazenave authored
Move all the additional steps outside the month loop since they only need occur at the end once we have all the dependent data downloaded. We also do the de-averaging here.
-
Pierre Cazenave authored
Force the code to download from within the current month only. Some of the files returned data from outside the current month (which is weird). Since the variables which have the longer returned time series seem to be inconsistent (i.e. uwnd might, but prate might not), I thought it would be easier to force everything to be contained within the current month. This is probably slightly less efficient as we have to do multiple calls (one for each month), but makes it cleaner to handle.
-
Pierre Cazenave authored
-
Pierre Cazenave authored
-
Pierre Cazenave authored
Rename the temporary struct to make it clearer what is temporary and what is permanent. This also requires better management of the preallocation so as to preserve the output from previous iterations.
-
- 19 May, 2015 5 commits
-
-
Pierre Cazenave authored
-
Pierre Cazenave authored
-
Pierre Cazenave authored
Fix the way the years and months are handled so they are in sync. The nested loop approach does not work well across year boundaries as you end up with a December to January on the inner loop but you do not increment the year in between.
-
Pierre Cazenave authored
-
Pierre Cazenave authored
-
- 19 Mar, 2015 4 commits
-
-
Pierre Cazenave authored
-
Pierre Cazenave authored
-
Pierre Cazenave authored
-
Pierre Cazenave authored
-
- 27 Feb, 2015 2 commits
-
-
Pierre Cazenave authored
-
Pierre Cazenave authored
-
- 23 Feb, 2015 5 commits
-
-
Pierre Cazenave authored
Add new function to export a nesting forcing file from external model output (either interpolated from a regular model grid (type 3 nesting) or from a bigger FVCOM model (type 1 and 2)).
-
Pierre Cazenave authored
-
Pierre Cazenave authored
-
Pierre Cazenave authored
-
Pierre Cazenave authored
-
- 14 Jan, 2015 9 commits
-
-
Pierre Cazenave authored
Remove the verbose flag as this function is sometimes called in a loop and this makes the on screen output very messy.
-
Pierre Cazenave authored
-
Pierre Cazenave authored
Apparently something has changed with the air data from NCEP. This try/catch (quite inelegantly) fixes it.
-
Pierre Cazenave authored
-
Pierre Cazenave authored
-
Pierre Cazenave authored
-
Pierre Cazenave authored
-
Pierre Cazenave authored
-
Pierre Cazenave authored
relevant element ID instead of the node ID.
-
- 25 Nov, 2014 1 commit
-
-
Pierre Cazenave authored
Reinstate the preallocation of the output array in the case for which no valid stations are found. Not having this causes the if ~isempty(out) a bit later to fail, bringing the whole thing crashing down (or at least stopping the script).
-
- 17 Oct, 2014 1 commit
-
-
Pierre Cazenave authored
-
- 16 Oct, 2014 1 commit
-
-
Pierre Cazenave authored
-
- 18 Sep, 2014 1 commit
-
-
Pierre Cazenave authored
Fix the way the number of node strings is calculated to work on older MATLAB versions (specifically v2010a).
-
- 02 Sep, 2014 1 commit
-
-
Pierre Cazenave authored
-
- 19 Aug, 2014 1 commit
-
-
Pierre Cazenave authored
-
- 14 Aug, 2014 3 commits
-
-
Pierre Cazenave authored
Add the ability to control which time variables are output when writing the netCDF file. The script defaults to Times since this is the easiest to control from a precision perspective. Also includes an update to the format of the Times output to better store fractions of a second.
-
Pierre Cazenave authored
Add the ability to control which time variables are output when writing the netCDF file. The script defaults to Times since this is the easiest to control from a precision perspective. Also includes an update to the format of the Times output to better store fractions of a second.
-
Pierre Cazenave authored
-