FAQ

Q: Why nctoolbox?

nctoolbox can access many common data model datasets

nc = ncdataset('localfile.nc') % NetCDF file
nc = ncdataset('http://geoport.whoi.edu/thredds/dodsC/coawst_2_2/fmrc/coawst_2_2_best.ncd') % DODS URL
nc = ncdataset('http://www.mbari.org/staff/brian/pub/m1_pco2.nc') % NetCDF URL

It's simple to use

nc = ncdataset('http://www.mbari.org/staff/brian/pub/m1_pco2.nc')
p = nc.data('pco2')
t = nc.data('time_d')
plot(t, p)
datetick('x')
set(gca, 'XLim', datenum(['2005-01-01'; '2008-01-01']))

Q: Why is nctoolbox READ-ONLY?

In short, Matlab already has built-in support for writing HDF, HDF5, and NetCDF files. Unfortunately, Matlab has no built-in support for OpenDAP and very complicated API's for reading common data model formats. For many researchers/scientists/students, this makes getting data into Matlab a big pain in the rear a serious challenge. nctoolbox greatly simplifies data access by providing a very simple and intuitive API. Here's a brief synopsis of the differences:

Code Comparison - Reading variable names from a local NetCDF file

Matlab's built in NetCDF API

ncid = netcdf.open('m1_pco2.nc', 'NC_NOWRITE');
[numdims, numvars, numglobalatts, unlimdimID] = netcdf.inq(ncid);
varnames = {};
for i = 0:(numvars - 1)
    [varnames{i + 1, 1}, xtype, dimids, numatts] = netcdf.inqVar(ncid,i);
end

nctoolbox

ds = ncdataset('m1_pco2.nc');
varnames = ds.variables;