PML_ChangeLog.txt

Pierre Cazenave <pica@pml.ac.uk> fvcom-toolbox ChangeLog

20150319
--------

Add support for writing nesting files to force FVCOM. Also bugfix and
streamline existing functions.

examples:
    example_fvcom_inputs.m
        * New example script to generate FVCOM inputs from TPXO tides, NCEP
        surface heat and HYCOM open boundary forcing.

fvcom_prepro:
    add_obc_nodes_list.m
        * Output number of nodes if the verbose flag is set. Minor coding style
        changes.
    add_stations_list.m
        * Add support for exporting the element ID closest to the station of
        interest (as well as the node ID which was already there). Fix some
        formatting issues.
    find_nearest_pt.m
        * Tidy up the code a bit and add extra information to the help.
    get_NCEP_forcing.m
        * Fix off by one error with the air pressure data from NCEP.
    read_fvcom_obc.m
        * Add new function to read in open boundary node IDs from a FVCOM
        casename_obc.dat file.
    read_sms_mesh.m
        * Add spherical coordinates on element centres.
    write_FVCOM_bath.m
        * Minor coding style changes.
    write_FVCOM_nested_forcing.m
        * New function to output nesting data (either interpolated from a
        structured grid/other data source or from a larger FVCOM model) to
        netCDF.
    write_FVCOM_probes.m
        * Fix export of the u and v locations to use the closest element
        instead of the node ID. Using the node ID for the u and v data will
        yield velocity time series miles away from the actual location of
        interest.
    write_FVCOM_restart.m
        * Move variable creation outside of a loop as it doesn't change between
        iterations.

utilities:
    read_netCDF_FVCOM.m
        * Fix verbose output and change indenting here and there. Add a BUGS
        section.
    surrounders.m
        * New function to find the nodes connected to a specified node.


20141017
--------

Fairly major update to fix a number of bugs and improve functionality. This
includes additional functionality to work with POLCOMS model output (mainly the
input files, configuration files and flat binary outputs). There are also some
new functions to work with FVCOM-ERSEM.

fvcom_prepro:
    get_EA_river_climatology.m
        * Another fix for leap year times.
    get_EHYPE_rivers.m
        * Add option to exclude rivers by name. Add new option to use an
        alternatively formatted input (two column time,flux).
    get_HYCOM_tsobc.m
        * Update interp1 function to use pchip instead of csap as the latter
        will be removed in a future version of MATLAB and the innumerable
        warnings were doing my nut in.
    get_POLCOMS_tsobc_gcoms.m
        * Add support for clipping in time.
    get_POLPRED_spectide.m
        * Update help.
    init_lag.m
        * Add new (largely untested) function to create the initial Lagrangian
        netCDF file.
    interp_HYCOM2FVCOM.m
        * Fix bug when interpolating velocity data due to incorrectly sized
        preallocated array.
        * Make the way the data to be interpolated onto the element centres
        (i.e. the velocity data), as opposed to the element nodes, more
        understandable.
    read_MetUM_forcing.m
        * Change warning to a simple printed statement.
    read_sigma.m
        * Add the sigma levels for the element centres in additions to the
        element nodes.
    read_sms_mesh.m
        * Add Rory as an author.
        * Fix bugs when not using bathymetry (i.e. only reading the grid data
        in).
        * Also make a depth array for the element centres.
    write_FVCOM_elevtide.m
        * Add new flags to control which time variables to use. FVCOM reads the
        'Times' variable first if present, then falls back to 'Itime' and
        'Itime2' and finally 'time'.
        * Also reinstate the original version of the calculation of Itime2 as
        the rounding effect was smoothing out the data too much, affecting its
        precision.
        * Update help accordingly.
    write_FVCOM_forcing.m
        * Add new flags to control which time variables to use. FVCOM reads the
        'Times' variable first if present, then falls back to 'Itime' and
        'Itime2' and finally 'time'.
    write_FVCOM_restart.m
        * Update the help to better describe how the optional 'out_date'
        argument works.
    write_FVCOM_river_ERSEM.m
        * Add new function to output river data for FVCOM-ERSEM.
    write_FVCOM_tsobc.m
        * Add new flags to control which time variables to use. FVCOM reads the
        'Times' variable first if present, then falls back to 'Itime' and
        'Itime2' and finally 'time'.
        * Add better check for the size of the input arrays (works with scalars).
    write_FVCOM_tsobcERSEM.m
        * Add new function to output temperature and salinity open boundary
        data for FVCOM-ERSEM.

utilites:
    calc_rad.m
        * New function for calculating instantaneous irradiance from a given
        latitude and time.
    calc_scoord.m
        * Calculate POLCOMS s-coordinates.
    distance_to_coast.m
        * Calculate the distance to the model coastline from a given position.
        Requires additional third-party tools (HJB Solver Package - the path to
        this must be edited in the function (not ideal)).
    do_surface_plotMatlabMap.m
        * Add script to plot a surface using the MATLAB mapping toolbox.
    do_surface_plotVel.m
        * Add script to plot a velocity surface using the third-party m_map
        mapping toolbox.
    do_surface_plotVelMatlabMap.m
        * Add script to plot a velocity surface using the MATLAB mapping
        toolbox.
    do_vector_plot_MatlabMap.m
        * Add script to plot velocity vectors with the MATLAB mapping toolbox.
    find_relaxation_boundary.m
        * Add function to find the elements which fall along the boundary.
    get_polcoms_fixed_vars.m
        * Add function to get POLCOMS fixed variables from a flat binary output
        file.
    get_polcoms_timeseriesv1.m
        * Add function to extract time series from POLCOMS outputs.
    greg2julian.m
        * Fix input arguments check.
    grid_vert_interp.m
        * Add new argument to allow specifying different coordinates for the
        extrapolation. This allows us to interpolate data which belongs either
        to the element centres or element nodes (defaults to element nodes).
        This is only really important when the model grid falls outside the
        coverage of the supplied data and we're extrapolating data.
    interpolate_near_coast.m
        * Add new function to interpolate a dataset towards a coastline.
    interpolate_polcoms_timeseries_zet.m
        * Add new function to interpolate POLCOMS surface elevation data
        towards a coastline.
    read_netCDF_FVCOM.m
        * Complete the functionality to be able to slice the data along any
        dimension (siglay, time, node etc.).
        * Fix ability to slice with any combination of space (horizontal and
        vertical) and time.
    read_polcoms_params.m
        * New function to read polcoms parameters.
    read_scoord_param.m
        * New function to read hc, cc, theta and bb from POLCOMS
        scoord_params.dat file
    readzetUBVB.m
        * New function to read POLCOMS output binary files (dailymean).


20140728
--------

Minor update to fix a number of niggling bugs.

fvcom_prepro:
    add_river_nodes_graphic.m
        * Set the river flag in the Mobj when loading river data.
    add_river_nodes_list.m
        * Set the river flag in the Mobj when loading river data.
    get_EA_river_climatology.m
        * Fix (I think) bug with leap years and climatology across multiple
        years.
    get_EHYPE_rivers.m
        * Add optional arguments to exclude specific rivers (by name). Also add
        new flag to change the assumed format of the input data to be
        "time,flux" rather than the EHYPE-specific raw data format.
    get_FVCOM_rivers.m
        * Set the river flag in the Mobj when loading river data.
    get_HYCOM_tsobc.m
        * Use more stable interpolation (pchip instead of csaps) for the depth
        interpolation of the data for each open boundary node.
    get_POLCOMS_river_discharge.m
        * Set the river flat in the Mobj when loading river data.
    interp_HYCOM2FVCOM.m
        * Add support for the latest syntax for the parallel processing tools
        in MATLAB.
    make_blank_mesh.m
        * Add new flag to the Mobj to indicate whether we have rivers in the
        model.
    write_FVCOM_restart.m
        * Fix the calculation of Itime2 in the restart file (previously always
        set to 0).

utilities:
    grid_vert_interp.m
        * Fix bug in interpolating in the vertical when HYCOM data has only
        a single depth bin.


20140423
--------

Relatively minor update to bring in some improvements in consistency in the
toolbox (i.e. remove duplicated functions) and fix some bugs in the function to
read the FVCOM outputs (read_netCDF_FVCOM.m).

README.md
    * Flesh out the README with links to the tagged releases as well as a bit
    of history of the toolbox.

fvcom_prepro:
    nodes2elems.m
        * Made the calculation array-based instead of loop-based to increase
        performance.
    smoothfield.m
        * Minor cosmetic changes to the code.
    write_FVCOM_forcing.m
        * Fix a typo.
    read_fvcom_mesh.m [REMOVED]
        * Remove this duplicate function (leave the version in utilities
        instead).

swan_scripts:
    calc_tauw.m
        * Bring in the differences between this version of calc_tauw.m and the
        one in utilities/ (which has been removed).

utilities:
    calc_rad.m
        * New function to calculate theoretical shortwave irradiance based on
        latitude and time of day.
    read_fvcom_mesh.m
        * Bring in the changes from the (now removed) version of this script
        which was in fvcom_prepro.
    read_netCDF_FVCOM.m
        * Fixed a bug where if a 2D array was requested after a 3D array, the
        2D array would cause the function to crash (because it was using a 3D
        index for getVar).
        * Add the global verbose flag and tidy up the help a bit. Change some
        verbose statements to use fprintf instead of disp for better control
        over formatting.


20140131
--------

Update to include some fixes to the forcing scripts as well as new scripts to
better handle various river data sources.

README.md
    * Add note about the Tidal Model Driver toolbox being an optional
    dependency.

fvcom_prepro:
    get_EA_river_climatology.m
        * New function to use pre-calculated Environment Agency river
        temperature climatology data as input to the FVCOM river
        temperature time series.
    get_EHYPE_rivers.m
        * New function to apply E-HYPE river discharge data to a given model
        domain.
    get_FVCOM_rivers
        * Change the unique call to preserve the order by replacing 'first'
        with 'stable'. This requires a relatively modern MATLAB (post-2011b).
    get_HYCOM_forcing.m
        * Add sea surface height to the list of variables that can be
        downloaded and add the ability to specify particular variables to
        download.
    get_HYCOM_tsobc.m
        * Add support for sea surface height and make the interpolation to the
        boundary nodes less likely to fail when too few HYCOM points are found
        nearby.
    interp_HYCOM2FVCOM.m
        * Fix the identification of the time index in the HYCOM data (use
        hycom.time instead of Mobj.ts_times). Also ignore a field name of 'MT'
        if supplied in varlist.
    read_MetUM_forcing.m
        * Fix the way time is handled. Previously a time variable had to be
        specified in varlist. Now, each data variable's time is returned as an
        array within the MetUM.(variable) struct, giving
        MetUM.(variable).time and MetUM.(variable).data. This means if each
        data variable uses a different time sampling, that can be accounted for
        later (by interpolating to a common time series with interp3, for
        example). Currently the code extracts the first 6 hour's worth of data.
        The assumption there is that the Met Office do 4 runs per day, so
        6 hours of data from each run gives you a day's worth.
    read_sms_map.m
        * Minor cosmetic changes.
    read_sms_mesh.m
        * Minor cosmetic changes.
    write_FVCOM_elevtide.m
        * Simplify the verbose output stuff.
    write_FVCOM_forcing.m
        * Add support for writing all the variables needed for
        a HEATING_CALCULATED model run. This essentially makes
        write_FVCOM_heating redundant, but I'll leave it in as it's a bit
        simpler to understand what's going on there. Also update the way the
        net surface heat flux is calculated (sum long and short, subtract
        latent and sensible). Fix the way the wind variables are handled so
        that both the U10/V10 and uwind_speed/vwind_speed netCDF variables are
        written if only one of data.u10/data.v10 or data.uwnd/data.vwnd is
        given. Change the output of tri' to tri, as tri was being written the
        wrong way around (thanks to Rory O'Hara Murray for spotting that one).
    write_FVCOM_heating.m
        * Clarify the help a bit.
    write_FVCOM_restart.m
        * Make it more obvious when new data is being written in place of
        existing data.
    write_FVCOM_river.m
        * Tidy up the way the times are handled.
    write_FVCOM_river_nml.m
        * Fix the handling of the optional vertical distribution string
        argument.
    write_FVCOM_stations.m
        * Minor cosmetic changes.
    write_WRF_forcing.m
        * Add full suite of variables so that HEATING_CALCULATED can be used
        instead. Fix names of the fields for long and shortwave radiation to
        match the NCEP ones (from n{l,s}wrf to n{l,s}wrs i.e. change last
        letter from f to s).

utilities:
    grid2fvcom.m
        * Check for the presence of the input fields being requested in the
        input struct to avoid finding out that the last field in vars doesn't
        exist in data half way through a time consuming loop. Change the way
        the alternative coordinate arrays are used to accommodate subtleties in
        the parallel code in MATLAB. Also fix some problems which sometimes
        arose when interpolating using the Parallel Computing toolbox (they
        were not reproducible with the serial version, annoyingly).
    read_fvcom_mesh.m
        * Minor cosmetic changes.
    read_netCDF_FVCOM.m
        * Add support for missing Itime and Itime2 values in an output file,
        falling back on the time variable instead.


20130917
--------

Update to include the ability to use HYCOM and Met Office Unified Model
outputs.

fvcom_prepro:
   calc_sponge_radius.m
      * Fix typo.
   get_FVCOM_rivers.m
      * Use the full set of coastal coordinates when finding the appropriate
      river node.
   get_HYCOM_forcing.m
      * Major update to this function. Now it correctly downloads the relevant
      HYCOM data from the HYCOM OPeNDAP server. Does not work with the third-
      party OPeNDAP toolbox.
   get_HYCOM_tsobc.m
      * New function: interpolates HYCOM data to the boundary nodes for
      temperature and salinity boundary forcing.
   get_MetUM_pp.m
      * Update to build a list of the downloaded files from the BADC server
      Also add new model outputs to the list of files available for download
      (the new files are post-2011 but live in a different directory from the
      pre-2012 data). Also download files in parallel to speed up the process
      (this may be a bad idea so may be removed if that proves to be the
      case).
   get_NCEP_forcing.m
      * Fairly significant change in that this function can now download any
      of the Reanalysis-1, Reanalysis-2 and 20th Century Reanalysis data from
      the NCEP server. Where 4D data are requested, only the surface data is
      returned. I have not tested this with the third-part OPeNDAP toolbox.
   interp_HYCOM2FVCOM.m
      * Use to create spatially varying temperature, salinity, u and v data
      for use in an FVCOM restart file from HYCOM data.
   interp_POLCOMS2FVCOM.m
      * Add inteprolation of the u and v data in the POLCOMS netCDF files.
      Eventually this function will be adjusted to allow the selection of
      only specific variables to be interpolated (as is the case with
      interp_HYCOM2FVCOM).
   read_MetUM_forcing.m
      * New function: reads the specified variables from the netCDF files of
      the converted Met Office Unified Model PP files.
   read_sms_mesh.m
      * Fairly major restructuring to decrease the time it takes to read an
      SMS file by about 50%. This was mainly done to maintain my sanity.
   write_FVCOM_elevtide.m
      * Minor change to the verbose output following the rewritten
      mjulian2greg function.
   write_FVCOM_forcing.m
      * Add support for alternative name for the pressure variable. Also tweak
      the output of the net surface heat flux to only occur when we are
      writing out to a single file.
   write_FVCOM_heating.m
      * Add failsafe functionality when writing the variables to better handle
      NCEP and Met Office data.
   write_FVCOM_restart.m
      * Add support for writing out u and v data (on the element centres
      rather than node positions).
   write_FVCOM_river_nml.m
      * Add support for an optional string to be written out (e.g. 'uniform')
      for the vertical distribution of river discharge to bypass the automated
      string generation.
   write_WRF_forcing.m
      * New function: writes out regularly gridded data in wrf_grid format for
      direct use in FVCOM. The interpolation then occurs within FVCOM rather
      than here, in MATLAB. Thanks to Dmitry Aleynik for his MATLAB function
      which was used as the basis for this function.

utilities:
   grid2fvcom.m
      * Fix a bug in which the position arrays could be written out transposed
      relative to the data arrays (leading to forcing data being in the wrong
      place).
   grid_vert_interp.m
      * Remove any NaN values in the vertical profile to support HYCOM data
      (which have a set of fixed vertical levels).
   pp2nc.m
      * Incorporate changes to better support paths in Windows (in particular
      paths with spaces).
   pp2nc.tcl
      * Add (but leave commented out) functions to interpolate the Met Office
      data onto a geographical grid.
   pp2nc_subset.m
      * New function: wrapper function to use convsh to interpolates Met
      Office Unified Model PP files onto a geographical grid (i.e. lat/long).
   subset.tcl
      * New TCL script to interpolate the Met Office Unified Model PP files
      onto a geographical grid.


20130719
--------

This merges Geoff Cowles' GitHub repository with mine (at least that was the
theory). Hopefully nothing catastrophic has happened.

examples:
    create_files_tutorial.m
        * New directory into which scripts on how to use the tools in the
        toolbox can be placed.

fvcom_prepro:
    add_obc_nodes_list.m
        * Fix open boundary node ordering issues when using unique.
    add_river_nodes_list.m
        * Fix open boundary node ordering issues when using unique.
    calc_sponge_radius.m
        * Fix open boundary node ordering issues when using unique.
    extract_mesoscale.m
        * New script to extract POLCOMS meteorological forcing data.
    get_AMM.m
        * Fix typo.
    get_FVCOM_rivers.m
        * Add Karen to list of authors.
    get_MetUM_pp.m
        * Script to fetch Met Office (UK) operational model (Unified Model)
        forcing files.
    get_NAE2_forcing.m
        * Fix typos.
    get_NCEP_forcing.m
        * Add (optional) support for extracting only particular variables from
        the NCEP Reanalysis 2 OPeNDAP server. Also fix the version check.
    get_POLCOMS_netCDF.m
        * Use a variable rather than getting the name dynamically (makes it
        marginally easier to read).
    get_POLCOMS_tsobc.m
        * Fix some bugs in the way the open boundary node values were stored
        (the order in which they were stored did not match the order of the
        nodes in Casename_obc.dat). Also fix the order of the vertically
        interpolated values so that FVCOM starts at the surface instead of
        mirroring POLCOMS' approach (where the first value is the seabed).
    get_POLCOMS_tsobc_NOCL.m
        * Add NOC Liverpool specific temperature and salinity interpolation
        script.
    interp_POLCOMS2FVCOM.m
        * Fix the vertical ordering of the POLCOMS data. POLCOMS' scalar
        values (temperature, salinity etc.) are stored seabed to surface;
        its depths are stored surface to seabed; FVCOM stores everything
        surface to seabed.
    read_ERA_wind.m
        * Major changes to this function. It is now able to automatically load
        forcing data from a directory hierarchy for a given period.
    read_grid_mesh.m
        * New function to convert .grid format mesh files to mesh objects
        (Mobj).
    write_FVCOM_forcing.m
        * Remove support for pevpr (pevpr is in W/m^{2} from the NCEP
        Reanalysis 2 data; FVCOM wants evaporation in m/s). Update the help
        accordingly.
    write_FVCOM_heating.m
        * New function to write an FVCOM NetCDF heating file (for
        HEATING_CALCULATED).
    write_FVCOM_meanflow_ascii.m
        * Fix the way the open boundary nodes are obtained so their order
        matches the casename_obc.dat file.
    write_FVCOM_restart.m
        * Remove commented out warnings.

tests:
    test_get_POLCOMS_tsobc.m
        * Make minor process on getting this working (still not working).

utilities:
    get_BADC_data
        * Child function for get_MetUM_pp.m to do the actual downloading of
        the pp files.
    grid2fvcom.m
        * Add more elegant case statement rather than using string
        comparisons.
    pp2nc.m
        * Convert Met Office (UK) pp files to a more sensible NetCDF file.
    pp2nc.tcl
        * convsh script to convert from pp to NetCDF format (using with
        pp2nc.m).


20130521
--------

fvcom_prepro:
    get_FVCOM_rivers.m
        * Function now finds river nodes more intellgently. Also separated out
        the POLCOMS specific code to get_POLCOMS_river_positions.m and
        get_POLCOMS_river_discharge.m) to make the code more reusable if you
        have a different source of river data.
    get_NCEP_forcing.m
        * Expand the help a little.
    get_MetUM_forcing.m
        * New (incomplete) function. As yet unfinished function to download
        forcing data from the Met Office FTP server.
    get_POLCOMS_river_discharge.m
        * New function formerly part of get_FVCOM_rivers.m, but now extracts
        discharges from POLCOMS flow files only.
    get_POLCOMS_river_positions.m
        * New function formerly part of get_FVCOM_rivers.m, but now simply
        finds the coordinates of the rivers in a POLCOMS model grid.
    get_POLCOMS_tsobc.m
        * Open the MATLAB Parallel workers pool to make full use of the
        parallel capability in MATLAB (previously the parfor loop was running
        suboptimally).
    get_POLCOMS_tsobc_gcoms.m
        * New function to be used if the POLCOMS depths are stored in a second
        NetCDF file rather than with the temperature and salinity arrays.
    interp_POLCOMS2FVCOM.m
        * Parallelise the interpolation of the data onto the unstructured
        grid. This requires the Parallel Computing Toolbox (although if it is
        not available, the code still runs, albeit slower!).
    read_sigma.m
        * Updated to include better support for non-uniform sigma coordinates
        (e.g. geometric distributions). See sigma_gen.m, sigma_geo.m and
        sigma_tanh.m for more information.
    read_sms_map.m
        * New function to read an SMS .map file and import the coastline into
        MATLAB.
    mjul2str_noyear.m
        * Deleted this empty function.
    write_FVCOM_forcing.m
        * Fix the evaporation and precipitation arrays being used (should be
        in correct units now).

utilities:
    catstruct.m
        * MATLAB Central function to concatenate two structs.
    grid2fvcom.m
        * Parallelise the code. This requires the Parallel Computing Toolbox,
        but offers significant speed improvements if available.
    grid_vert_interp.m
        * Parallelise the vertical interpolation. As before, this requires the
        Parallel Computing Toolbox.
    sigma_gen.m
    sigma_geo.m
    sigma_tanh.m

tests:
    utilities:
        test_grid2fvcom.m
            * Unit test to ensure changes in grid2fvcom.m do not break its
            functionality (i.e. the modifications don't change the output
            values). This is obviously not valid if grid2fvcom.m is found to
            have a bug!
        test_get_POLCOMS_tsobc.m
            * Unit test to check get_POLCOMS_tsobc.m results don't change
            following changes to the script.
    data:
        * Directory in which data files are kept which are used by the unit
        test scripts.


20130403
--------

fvcom_prepro:
    add_river_nodes_list.m
        * Minor changes to code style.
    add_sponge_nodes_list.m
        * Fix bug which ignores constant sponge radius.
    add_stations_list.m
        * Add message about the outcome if no stations were within the domain.
    add_var_FVCOM_river.m
        * Turn warnings back on.
    example_FVCOM_tsobc.m
        * Add the global NetCDF history variable which includes a reference to
        the FVCOM toolbox.
    example_FVCOM_wind_ts.m
        * Turn warnings back on.
    example_FVCOM_wind_ts_speed.m
        * Turn warnings back on.
    example_init_lag.m
        * Turn warnings back on.
    get_FVCOM_rivers.m
        * Extract positions from the FVCOM-adapted POLCOMS rivers files. The
        POLCOMS positions are too coarse (12km) to be useful for FVCOM. As
        such, this function reads an ASCII file of positions and extracts the
        station's discharge data from the POLCOMS file.
    get_NCEP_forcing.m
        * Adds support for MATLAB's native OPeNDAP tools. Also fixes the
        calculation of net surface heat flux.
    get_POLCOMS_meanflow.m
        * Interpolate mean flow from POLCOMS output at the FVCOM open
        boundaries.
    get_POLCOMS_netCDF.m
        * More code outside the verbose section so it is executed even if
        verbose output is disabled.
    get_POLCOMS_rivers.m
        * Parse the POLCOMS rivers data file.
    get_POLCOMS_tsobc.m
        * Change the vertical interpolation to be scaled within the
        POLCOMS depth range for the current node. The net result is that the
        vertical profiles are squashed or stretched to fit within the FVCOM
        depths. This means the full profile structure is maintained in the
        resulting FVCOM boundary input despite the differing depths at the
        FVCOM boundary node.
    nodes2elems.m
        * Minor changes to coding style.
    read_ERA_wind.m
        * Turn warnings back on.
    read_NCEP_wind.m
        * Turn warnings back on.
    write_FVCOM_bedflag.m
        * Turn warnings back on.
    write_FVCOM_elevtide.m
        * Minor changes to coding style.
    write_FVCOM_forcing.m
        * Export of the surface forcing can now use a precomputed surface heat
        flux rather than having to interpolate the four components only to
        merge them (see write_FVCOM_forcing.m)
    write_FVCOM_meanflow.m
        * Preliminary support for mean flow at the open boundaries. At the
        moment, the mean flow is interpolated from POLCOMS daily mean flow
        values. The necessary tools to write out the FVCOM ASCII mean flow
        files have been created, but I have not yet managed to successfully
        run a model with mean flow. Consider it a work in progress.
    write_FVCOM_meanflow_ascii.m
        * Outputs the results of write_FVCOM_meanflow.m to ASCII for use in
        FVCOM.
    write_FVCOM_obs_TS.m
        * Add the global NetCDF history variable which includes a reference to
        the FVCOM toolbox.
    write_FVCOM_restart.m
        * The routine to replace variables within an FVCOM restart file now
        includes the ability to adjust the times for which the restart file is
        applicable. This is useful if you have a restart file for a model
        domain, but wish to use it for a different time.
    write_FVCOM_river.m
        * The existing river functions have been updated to reflect the
        changes needed in the latest versions of FVCOM (3.1.x). This has also
        been updated to use MATLAB's native NetCDF routines.
    write_FVCOM_river_nml.m
        * New function to write out the name list for the rivers in the model
        domain.
    write_FVCOM_spectide.m
        * Add the global NetCDF history variable which includes a reference to
        the FVCOM toolbox.
    write_FVCOM_sponge.m
        * Fix bug which ignores constant sponge radius.
    write_FVCOM_stations.m
        * Minor change to the error message for the incorrect number of
        arguments.
    write_FVCOM_tsobc.m
        * Add the global NetCDF history variable which includes a reference to
        the FVCOM toolbox. Also enable warnings.
    write_FVCOM_wind_ts_speed.m
        * Add the global NetCDF history variable which includes a reference to
        the FVCOM toolbox. Also enable warnings.
    write_FVCOM_z0.m
        * Add the global NetCDF history variable which includes a reference to
        the FVCOM toolbox. Also enable warnings.

utilities:
    centroid.m
        * Fix problem with checking the input arguments.
    find_boundary_elements.m
        * New utility to find the element numbers which run along the open
        boundary (needed for the mean flow).
    fix_inside_boundary.m
        * A new function to reorganise open boundary elements to have an edge
        approximately normal to the open boundary has been added
        (fix_inside_boundary.m). The routine is somewhat brute force, and can
        result in invalid unstructured grids. As such, any changes should be
        reviewed in SMS prior to being used in a model run.
    grid2fvcom.m
        * Improve the documentation.
    write_SMS_2dm.m
        * Export unstructured grid data from MATLAB to SMS format.
    write_SMS_cst.m
        * Export open boundary from MATLAB to SMS format.


20130204
--------

fvcom-prepro:
    add_obc_nodes_graphic.m:
        * Minor fix to make the defined function name match the file name.
    add_obc_nodes_list.m:
        * Add optional new argument to plot a figure of the open boundary
        nodes.
    add_sponge_nodes.m:
        * Minor code cleanup.
    add_sponge_nodes_list.m:
        * New function: For a given list of nodes, apply a given sponge
        coefficient over a specified sponge radius (both given as arguments
        to the function. Follows the same syntax as add_sponge_nodes.m.
    add_stations_list.m:
        * New function: Add a set of stations at which FVCOM will output time
        series. Requires station coordinates and names, and a threshold
        distance beyond which the station is skipped.
    calc_sponge_radius.m:
        * New function: Create sponge layer with variable width radius based on
        the proximity of the closest element or 100km (whichever is smaller).
    estimate_ts.m:
        * Add arguments to the function so that the estimated current velocity
        and tidal range aren't hard-coded in the function.
        * Also add a crude conversion from lat/long distances to UTM (metres)
        using a great circle appoximation.
    example_FVCOM_tsobc.m:
        * Convert the method of writing NetCDF files from third-party toolbox
        to the MATLAB native version.
        * Add arguments to set dynamic file names (rather than hard-coded).
    example_FVCOM_wind_ts_speed.m:
        * Minor code change to calculate the number of elements in an array.
    get_AMM.m:
        * New function: Extract boundary forcing information from NOC
        Operational Tide Surge Model output.
    get_HYCOM_forcing.m:
        * New function: INCOMPLETE! Function to extract boundary forcing
        information from HYCOM model output through OPeNDAP.
    get_NAE2_forcing.m:
        * New function: Get the required parameters from NAE2 MetOffice model
        data to force FVCOM at the surface.
    get_NCEP_forcing.m:
        * New function: Get the required parameters from NCEP through OPeNDAP.
        Requires the air_sea and OPeNDAP toolboxes.
    get_POLCOMS_tsobc.m:
        * New function: Extract temperature and salinity data from the PML
        POLCOMS NetCDF files and interpolate them to the FVCOM open boundary
        nodes.
    get_POLPRED_spectide.m:
        * New function: Extract POLPRED harmonic amplitude and phases for the
        nearest point in the POLPRED grid to the open boundary nodes in the
        FVCOM grid.
    grid2fvcom.m:
        * New function: Interpolate regularly gridded surface forcing data onto
        a given FVCOM grid.
    read_ERA_wind.m:
        * New function: Extract wind data from ERA Interim NetCDF files.
    read_NCEP_wind.m:
        * New function: Read in pairs of NCEP wind vector files and output
        arrays of the required velocity components.
    read_sigma.m:
        * New function: Extract sigma layer and level information from a given
        sigma.dat file.
    read_sms_mesh.m:
        * Add the ability to extract open boundaries from the defined
        nodestrings in the SMS grid file (.2dm).
        * Note: the SMS grid name must be a single word (no spaces). If the
        script fails to load your grid, chances are its name is "Default
        Coverage". If you change it to omit spaces in either the .2dm file or
        the SMS project, this script should work.
    replace_FVCOM_restart_vars.m:
        * New function: For a given FVCOM restart file, replace select variables
        with values from a PML POLCOMS model NetCDF output file. POLCOMS data
        are interpolated first on the FVCOM vertical grid, then the vertically
        interpolated data are interpolated onto the horizontal FVCOM grid.
    set_elevtide.m:
        * New function: Write out timeseries of surface elevations to a NetCDF
        file.
        * Requires the Tide Model Driver MATLAB Toolbox from Oregon State
        University at http://polaris.esr.org/ptm_index.html.
    set_spectide.m:
        * Replace hard-coded variables with arguments to function call.
        * Add support for adding equilibrium amplitudes and beta love numbers.
    write_FVCOM_elevtide.m:
        * New function to accompany set_elevtide.m to output a time series of
        surface elevations.
    write_FVCOM_forcing.m:
        * New function: Write forcing data (u and v winds, heat flux etc.) out
        to NetCDF file(s) depending on specified FVCOM version.
    write_FVCOM_obs_TS.m:
        * Converted to use MATLAB native NetCDF routines.
    write_FVCOM_spectide.m:
        * Replace use of third-party NetCDF library with MATLAB native tools.
        * Add support for writing out equilibrium amplitudes and beta love
        numbers.
    write_FVCOM_sponge.m:
        * Add support for variable width sponge layers (see
        calc_sponge_radius.m).
    write_FVCOM_stations.m:
        * New function: Output list of stations at which FVCOM will output 1D
        time series.
    write_FVCOM_tsobc.m:
        * Create a new function to output temperature and salinity at the open
        boundaries (either spatially uniform or varying (e.g. interpolated from
        POLCOMS using get_POLCOMS_tsobc.m). Based on the example_FVCOM_tsobc.m
        file.
    write_FVCOM_wind_ts_speed.m:
        * Create a new function to output spatially uniform but temporally
        varying wind field. Based on example_FVCOM_wind_ts_speed.m.
    write_FVCOM_z0.m:
        * Add support for MATLAB native NetCDF routines.

fvcom-postproc:
    example_surface_plot.m:
        * Now includes a lot more examples of different types of surface plots
        (e.g. vectors at specific layers).

utilities:
    centroid.m:
        * New function: Calculate the centroid of a given polygon.
    connectivity.m:
        * New function: From Mesh2D toolbox. Read unstructured grid
        connectivity (useful to find grid boundary, for example).
    deg2utm.m:
        * New function: Convert from lat/long to UTM (but automatically
        determine relevant UTM zone). See also wgs2utm (where UTM zone can be
        forced).
    do_residual.m:
        * New function: Calculate the residual vector for a given 2D time
        series.
    do_residual_plot.m:
        * New function: Use the output of do_residual.m to plot residual
        vectors.
    do_transect_plot.m:
        * New function: Pick and plot a transect through model output.
    greg2mjulian.m:
        * Format help to match better with other utilities.
    get_NCEP_year.m:
        * Extract the year from an NCEP Reanalysis file name.
    deg2utm.m:
        * Function to convert from lat/long to UTM.
        * From http://www.mathworks.com/matlabcentral/fileexchange/10915
    show_sigma.m:
        * Made the reading in of the sigma.dat file more resilient (notably to
        comments and blank lines).
    utm2deg.m:
        * Function to convert from UTM to lat/long.
        * From http://www.mathworks.com/matlabcentral/fileexchange/10914
    wgs2utm.m:
        * New function: Convert from lat/long to UTM whilst being able to force
        the UTM zone (allowing for coordinates which spread over several UTM
        zones). See also deg2utm.