How to regrid/interpolate netCDF data (2.5x2.5 degree) to higher resolution (e.g. 0.25x0.25 degree)?

I am trying to use the script below by Hongxiao Jin in her Master thesis to regrid my netCDF NCEP Reanalysis II data to higher resolution at 0.25x0.25 degree. However, there are some errors in the script occurred. I installed mexcdf package to analyze netCDF data. Some errors occurred include: 1) Undefined function 'netcdf' for input arguments of type 'char'; 2) The expression to the left of the equals sign is not a valid target for an assignment in line 6. Thanks for your help.

 %Extracting Air relative humidity Example 2004 
  nc = netcdf('rhum.mon.mean.nc', 'nowrite'); %matlab function
  description = nc.description(:);    % Global attribute.
  variables = var(nc);    % Get variable data.
  Rhum2004=squeeze(variables{6}(301:312,1,:,:));%variables{6} 360X17X73X144 
  Rhum2004(find(Rhum2004==32766|Rhum2004==-32767)=nan; 
  nc = close(nc); 
  Rhum2004=permute(Rhum2004,[3 1 2]);%144X12X73 0->360
  Rhum2004=[Rhum2004(73:144,:,:);Rhum2004(1:72,:,:)];%-180->180 
  Rhum2004=permute(Rhum2004,[2 3 1]);%12X73X144
  Rhum2004=Rhum2004.*0.01+302.65; 
  [OX,OY]=meshgrid(-180:2.5:177.5,90:-2.5:-90); 
  [NX,NY]=meshgrid(-179.875:0.25:179.875,89.875:-0.25:-89.875); 
  RH2004=nan(12,720,1440); 
  for i=1:12 
  RH2004(i,:,:) = single(interp2(OX,OY,squeeze(Rhum2004(i,:,:)),NX,NY,'bilinear')); 
  end
  save RH2004 RH2004

[Ashish Uthama]

If you have a newer version of MATLAB, give NCDISP, NCINFO and NCREAD a try.

[tchu]

I am using Matlab R2013a with mexcdf package. If I change netcdf to ncread, remove some lines including description, variables, permute; and change the order of lon and lat values in meshgrid function, then it works. I am checking the result to see whether it is correct and compatible to the original data (before interpolation). Thanks for your suggestion.

[Ashish Uthama]

If that works, you could post that as an answer and "accept" it.

Answer 1

[tchu]

%Regrid Air relative humidity Example 2004 
rhum = ncread('rhum.mon.mean.nc','rhum');
Rhum2004 = squeeze(rhum(:,:,:,301:312));
size(Rhum2004)
Rhum2004(find(Rhum2004==32766|Rhum2004==-32767))=nan; 
Rhum2004=[Rhum2004(73:144,:,:);Rhum2004(1:72,:,:)];%-180->180 
[OY,OX]=meshgrid(90:-2.5:-90, -180:2.5:177.5); 
[NY,NX]=meshgrid(89.875:-0.25:-89.875,-179.875:0.25:179.875); 
RH2004=nan(1440,720,12); 
for i=1:12 
RH2004(:,:,i) = ...
single(interp2(OY,OX,Rhum2004(:,:,i),NY,NX,'bilinear')); 
end
save RH2004 RH2004