Extract temperature and salinity forcing information from NOC Operational
Tide Surge model output and interpolate onto the FVCOM domain for use in
a reproduced FVCOM restart file.
function restart = get_POLCOMS_tsrestart_NOCL(Mobj, inputConf)
DESCRIPTION:
Interpolate temperature and salinity values onto the FVCOM nodes
at all sigma levels.
INPUT:
Mobj = MATLAB mesh structure which must contain:
- Mobj.siglayz - sigma layer depths for all model
nodes.
- Mobj.lon, Mobj.lat and/or Mobj.x, Mobj.y - node
coordinates.
inputConf = MATLAB structure which must contain:
- inputConf.polcoms_ts - location of NOC Operational
Model output containing 4D variables of temperature
(tem) and salinity (sal). They should have dimensions
(x, y, sigma, time).
- inputConf.polcoms_z - location of NOC Operational
Model output containing 4D variables of bathymetry
(XXX) and sigma layer thickness (XXX).
- inputConf.startDate - start date and time for FVCOM
model run
OUTPUT:
restart = MATLAB structure which contains three fields (called
temperature, salinity and ts_time). temperature and salinity
have sizes (Mobj.nVerts, sigma, time). The time dimension is
determined based on the input NetCDF file. The ts_time
variable is just the input file times in Modified Julian Day.
EXAMPLE USAGE
restart = get_POLCOMS_tsrestart_NOCL(Mobj, inputConf)
Author(s):
Pierre Cazenave (Plymouth Marine Laboratory)
Karen Amoudry (National Oceanography Centre, Liverpool)
PWC Revision history
2013-01-09 First version based on the FVCOM shelf model
get_POLCOMS_forcing.m script (i.e. not a function but a plain script).
KJA Revision history:
2014-01-15 First version, adapted from KJA's
'get_POLCOMS_tsobc_NOCL.m', which in turn was based on PWC's
'get_POLCOMS_tsobc.m'.
==========================================================================0001 function restart = get_POLCOMS_tsrestart_NOCL(Mobj, inputConf) 0002 % Extract temperature and salinity forcing information from NOC Operational 0003 % Tide Surge model output and interpolate onto the FVCOM domain for use in 0004 % a reproduced FVCOM restart file. 0005 % 0006 % function restart = get_POLCOMS_tsrestart_NOCL(Mobj, inputConf) 0007 % 0008 % DESCRIPTION: 0009 % Interpolate temperature and salinity values onto the FVCOM nodes 0010 % at all sigma levels. 0011 % 0012 % INPUT: 0013 % Mobj = MATLAB mesh structure which must contain: 0014 % - Mobj.siglayz - sigma layer depths for all model 0015 % nodes. 0016 % - Mobj.lon, Mobj.lat and/or Mobj.x, Mobj.y - node 0017 % coordinates. 0018 % inputConf = MATLAB structure which must contain: 0019 % - inputConf.polcoms_ts - location of NOC Operational 0020 % Model output containing 4D variables of temperature 0021 % (tem) and salinity (sal). They should have dimensions 0022 % (x, y, sigma, time). 0023 % - inputConf.polcoms_z - location of NOC Operational 0024 % Model output containing 4D variables of bathymetry 0025 % (XXX) and sigma layer thickness (XXX). 0026 % - inputConf.startDate - start date and time for FVCOM 0027 % model run 0028 % 0029 % OUTPUT: 0030 % restart = MATLAB structure which contains three fields (called 0031 % temperature, salinity and ts_time). temperature and salinity 0032 % have sizes (Mobj.nVerts, sigma, time). The time dimension is 0033 % determined based on the input NetCDF file. The ts_time 0034 % variable is just the input file times in Modified Julian Day. 0035 % 0036 % EXAMPLE USAGE 0037 % restart = get_POLCOMS_tsrestart_NOCL(Mobj, inputConf) 0038 % 0039 % Author(s): 0040 % Pierre Cazenave (Plymouth Marine Laboratory) 0041 % Karen Amoudry (National Oceanography Centre, Liverpool) 0042 % 0043 % PWC Revision history 0044 % 2013-01-09 First version based on the FVCOM shelf model 0045 % get_POLCOMS_forcing.m script (i.e. not a function but a plain script). 0046 % 0047 % KJA Revision history: 0048 % 2014-01-15 First version, adapted from KJA's 0049 % 'get_POLCOMS_tsobc_NOCL.m', which in turn was based on PWC's 0050 % 'get_POLCOMS_tsobc.m'. 0051 % 0052 %========================================================================== 0053 0054 subname = 'get_POLCOMS_tsrestart_NOCL'; 0055 0056 global ftbverbose; 0057 if ftbverbose 0058 fprintf('\n') 0059 fprintf(['begin : ' subname '\n']) 0060 end 0061 %% 0062 % Which variables do we want from the POLCOMS file? 0063 varlist = {'lon', 'lat', 'tem', 'sal', 'time'}; 0064 0065 % Create the POLCOMS filename based on the year and month of interest 0066 polcoms_ts = [inputConf.polcoms_ts,num2str(inputConf.startDate(1)),... 0067 '-',num2str(inputConf.startDate(2),'%02i'),'.nc']; 0068 0069 % Get the results from the POLCOMS file 0070 nc = netcdf.open(polcoms_ts, 'NOWRITE'); 0071 0072 for var=1:numel(varlist) 0073 0074 getVar = varlist{var}; 0075 varid_pc = netcdf.inqVarID(nc, getVar); 0076 0077 data = netcdf.getVar(nc, varid_pc, 'single'); 0078 0079 pc.(getVar).data = double(data); 0080 0081 % Try to get some units (important for the calculation of MJD). 0082 try 0083 units = netcdf.getAtt(nc,varid_pc,'units'); 0084 catch 0085 units = []; 0086 end 0087 pc.(getVar).units = units; 0088 end 0089 0090 netcdf.close(nc) 0091 0092 % Get rid of data outside the time we're interested in 0093 % Convert FVCOM timestep into AMM/S12 output (seconds since 20071101:000000) 0094 AMM_time = etime(inputConf.startDate,[2007,11,01,0,0,0]); 0095 keep_time = (pc.time.data == AMM_time+(12*60*60)); % Allowance for POLCOMS data being at noon, not midnight 0096 pc.tem.data = pc.tem.data(:,:,:,keep_time); 0097 pc.sal.data = pc.sal.data(:,:,:,keep_time); 0098 0099 % Import the POLCOMS sigma info 0100 pc = get_POLCOMS_sigma(pc,inputConf); 0101 0102 % Data format: 0103 % pc.tem.data and pc.sal.data are x, y, sigma 0104 [~, ~, nz] = size(pc.tem.data); 0105 0106 % Make rectangular arrays for the nearest point lookup. 0107 [plon, plat] = meshgrid(pc.lon.data, pc.lat.data); 0108 0109 % Number of sigma layers. 0110 fz = size(Mobj.siglayz, 2); 0111 0112 if ftbverbose 0113 tic 0114 end 0115 0116 % Flip the vertical layer dimension to make the POLCOMS data go from 0117 % surface to seabed to match its depth data and to match how FVCOM 0118 % works. 0119 pctemp3 = flipdim(pc.tem.data, 3); 0120 pcsalt3 = flipdim(pc.sal.data, 3); 0121 0122 % Preallocate the FVCOM results arrays 0123 fvtemp = nan(Mobj.nVerts, fz); % FVCOM interpolated temperatures 0124 fvsal = nan(Mobj.nVerts, fz); % FVCOM interpolated salinities 0125 0126 % Preallocate the intermediate results arrays. 0127 itempz = nan(Mobj.nVerts, nz); 0128 isalz = nan(Mobj.nVerts, nz); 0129 idepthz = nan(Mobj.nVerts,nz); 0130 0131 for j = 1:nz 0132 % Transpose the POLCOMS data to be (x,y) rather than (y,x) 0133 pctemp2 = pctemp3(:, :, j)'; 0134 pcsalt2 = pcsalt3(:, :, j)'; 0135 pcdepth2 = squeeze(pc.depth.data(:, :, j))'; 0136 0137 % Reshape the arrays to allow the sort to work properly later 0138 tlon=reshape(plon,(size(plon,1)*size(plon,2)),1); 0139 tlat=reshape(plat,(size(plat,1)*size(plat,2)),1); 0140 pctemp2 = reshape(pctemp2,(size(pctemp2,1)*size(pctemp2,2)),1); 0141 pcsalt2 = reshape(pcsalt2,(size(pcsalt2,1)*size(pcsalt2,2)),1); 0142 pcdepth2 = reshape(pcdepth2,(size(pcdepth2,1)*size(pcdepth2,2)),1); 0143 0144 % Find the points which aren't NaNs 0145 keeptemp = find(~isnan(pctemp2)); 0146 keepsalt = find(~isnan(pcsalt2)); 0147 keepdepth = find(~isnan(pcdepth2)); 0148 0149 keep = intersect(keeptemp,keepsalt); 0150 keep = intersect(keep, keepdepth); 0151 0152 % Interpolate the POLCOMS results by (1) turning the non-NaN POLCOMS 0153 % results into a Tri object and (2) extracting the values of that Tri 0154 % Object at the FVCOM node locations. 0155 tritemp = TriScatteredInterp(tlon(keep),tlat(keep),pctemp2(keep),'natural'); 0156 itemp = tritemp(Mobj.lon,Mobj.lat); 0157 0158 trisalt = TriScatteredInterp(tlon(keep),tlat(keep),pcsalt2(keep),'natural'); 0159 isalt = trisalt(Mobj.lon,Mobj.lat); 0160 0161 tridepth = TriScatteredInterp(tlon(keep),tlat(keep),pcdepth2(keep),'natural'); 0162 idepth = tridepth(Mobj.lon,Mobj.lat); 0163 0164 % Put the results in this intermediate array. 0165 itempz(:, j) = itemp; 0166 isalz(:, j) = isalt; 0167 idepthz(:,j) = idepth; 0168 end 0169 0170 % Now we've interpolated in space, we can interpolate the z-values 0171 % to the sigma depths. 0172 % for zi = 1:fz 0173 for pp = 1:Mobj.nVerts 0174 % Get the FVCOM depths at this node 0175 tfz = Mobj.siglayz(pp, :); 0176 % Now get the interpolated POLCOMS depth at this node 0177 tpz = idepthz(pp, :); 0178 0179 % To ensure we get the full vertical expression of the vertical 0180 % profiles, we need to normalise the POLCOMS and FVCOM 0181 % depths to the same range. This is because in instances where 0182 % FVCOM depths are shallower (e.g. in coastal regions), if we 0183 % don't normalise the depths, we end up truncating the vertical 0184 % profile. This approach ensures we always use the full 0185 % vertical profile, but we're potentially squeezing it into a 0186 % smaller depth. 0187 A = max(tpz); 0188 B = min(tpz); 0189 C = max(tfz); 0190 D = min(tfz); 0191 norm_tpz = (((D - C) * (tpz - A)) / (B - A)) + C; 0192 0193 % Get the temperature and salinity values for this node and 0194 % interpolate down the water column (from POLCOMS to FVCOM). 0195 % Change to 'pchip' to match PWC parent code. 0196 if ~isnan(tpz) 0197 fvtemp(pp, :) = interp1(norm_tpz, itempz(pp, :), tfz, 'pchip', 'extrap'); 0198 fvsal(pp, :) = interp1(norm_tpz, isalz(pp, :), tfz, 'pchip', 'extrap'); 0199 else 0200 warning('Should never see this... ') % because we test for NaNs when fetching the values. 0201 warning('FVCOM boundary node at %f, %f is outside the POLCOMS domain. Skipping.', Mobj.lon(pp), Mobj.lat(pp)) 0202 continue 0203 end 0204 end 0205 % end 0206 0207 if ftbverbose 0208 toc 0209 end 0210 0211 % Convert NOC model output temperatures from Kelvin to Celsius 0212 fvtemp = fvtemp - 273.15; 0213 0214 % Timeshift to match the expected FVCOM input times. The temperature and 0215 % salinity values are a daily average (midnight to midnight). They are 0216 % given a timestamp of the middle time in this period (noon). Each day's 0217 % value applies to the whole day (e.g. from midnight to midnight). 0218 % Therefore, we can shift the time to midnight at the start of the relevant 0219 % day and apply the value to the whole day. This makes FVCOM (and me) 0220 % happy. 0221 restart.ts_times = datenum(2007, 11, 0, 0, 0, 0) + ... % Convert NOC model reference time to Matlab datenum 0222 ((pc.time.data(keep_time)+(12*3600)) / 3600 / 24); % Add NOC model output time and 12 hour offset 0223 restart.ts_times = datevec(restart.ts_times); % Convert times to datevec format 0224 0225 % Final output variable names must match FVCOM restart file variable names 0226 restart.temp = fvtemp; 0227 restart.salinity = fvsal; 0228 0229 if ftbverbose 0230 fprintf(['end : ' subname '\n']) 0231 end