GitLab

Allow the input temperature and salinity values to be either single values (for uniform grids) or otherwise 3D arrays of unstructured grid points with sigma layers and time as the two other dimensions. This should maintain backwards compatibilty with the previous invocation

Add code to try to parse the time attribute such that a Modified Julian Day time series for the POLCOMS data can be calcualted. Also means slightly changing the format of the pc struct so that each variables has a data and units array, the latter being used to find the reference time for the seconds in pc.time.data

Rename the POLCOMS temperature and salinity function to better match what it does (the forcing I typically think of as being the wind forcing)

Take the workings out of the shelf model get_POLCOMS_forcing.m script and make a new function from it. This reads in a POLCOMS output NetCDF file and interpolates (not in time, in space only, but horizontally and vertically) to create boundary inputs for the open boundaries in a given FVCOM grid

Attempt at generating a 3D array without resorting to a couple of loops. Not yet tested (hence it is commented out)

New function (based on the new code in show_sigma.m) to extract sigma levels for every node in a mesh structure

Fix code indentation. Also fix bug in MATLAB display of two different types of plot (e.g. patch and plot) by setting the renderer to OpenGL.

Add more robust code to read in a sigma.dat file (skipping comments, blank lines etc.). Also add optional argument to sample a larger number of points from the bathymetry

Add script to create the files needed by FVCOM when outputting a set of time series at given locations

Fix a weird issue with the printing of the status information. This way should be more robust (if ftbverbose is empty, for example)

Script to parse the POLPRED ASCII files and extract a specified list of harmonics. Based off the relevant functions in tide_tools.py from fvcom-py (https://bitbucket.org/pwcazenave/fvcom-py)

Fix two problems: 1. The date strings in Times were completely wrong (because it was assumed the values in MJD were MATLAB dates (not Modified Julian Days). Fixing that problem highlighted some possible limitations in mjulian2greg in which some second values are output as 60 (which is obviously wrong). So, some additional checks on the values of the seconds are performed, and times adjusted as necessary. 2. The Code to get the list of open boundary node IDs was a bit ugly, so I rewrote it to be a bit prettier. The old version is left in as a comment so that it can be reverted easily if necessary

Change the boundary file generation code to expect Modified Julian Day (rather than doing the conversion blindly in the function). Also fold in the code from set_elevtide.m into this function as it was somewhat pointless having set_elevtide call write_FVCOM_elevtide

Add additional variables in order to match the format specified in the FVCOM manual. Also fix the dimensions of the U and V wind vectors

Extend the search for the source data for the interpolation along the x-axis to two grid cells away from the SMS grid extent. This should hopefully stop any NaNs creeping in to the interpolation

Add check to make sure the interpolated grid is NaN-free (otherwise FVCOM will crash trying to run a model using this data)

Rename the nVerts variable to nNodes to make the nomenclature match that of SMS. Also fix the dimensions of the wind vectors, heating flux and a few other variables to match the description in the FVCOM manual

Fix the creation of the boundary temperature and salinity array to be the length of the time series rather than a single value

Fix bug in creation of the NetCDF spectide file which was only revealed by running it on MATLAB 2012. The error lay in setting a two dimensional dimension when a NetCDF dimension is necessarily 1D only. This may or may not have been causing FVCOM v3.1.6 to crash

Write out the proper surface heat flux by combining the three relevant parameters from the NCEP data

Fix bug whereby the variables which do not require interpolation were not being correctly transferred

Add the FVCOM mesh object to the function input arguments so that the number of elements and nodes can be obtained automatically