netCDF Operators [NCO].ppt

netCDF Operators [NCO] Introduction and History Suite of Command Line Operators Designed to operate on netCDF/HDF files Each is a stand alone executable Very efficient for specific tasks Available for various computer architectures: Solaris, Irix, AIX, Linux, Windows Appending vs. Concatenation Appending is the merging of files: file1 = T,U,V file2 = PSI,CHI file3 = T,U,V,PSI,CHI Concatenation is the combination of variables along a record dimension: file 1 = T(0:12,:,:) file 2 = T(13:24,:,:) concatenated file = T(0-24,:,:) Hyperslabs A hyperslab is a subset of data. Missing Values NCO identifies missing data by the _FillValue attribute. [v 3.9.2 8/2007] No arithmetic operations on these values. No longer recognizes missing_value Best to create netCDF with both _FillValue and missing_value ncea: grid point averaging Performs gridpoint averages across an arbitrary number of input files Each file is weighted evenly. Each variable in the output file will be the same size as the same variable in any input file. Coordinate variables are not averaged. ncea -v T,U jan_1.nc jan_2.nc ave.nc ncra: record averaging Averages record variables across an arbitrary number of input files The record dimension is retained as a degenerate (size 1) dimension. Weights each record in the input files equally ncra 12.nc 01.nc 02.nc DJF.nc ncecat: ensemble concatenator Concatenates an arbitrary number of input files into a single output file. Wild characters allowed. Each input file is stored consecutively as a single record in the output file. Input files are glued together by the creation of a record dimension. ncecat case-1.nc case-2.nc total.nc ncecat case*nc TOTAL.nc ncrcat: record concatenator Concatenates record variables across an arbitrary number of input files. Unix wild characters allowed Final record dimension is the sum of the lengths of the input files. Input files may vary in length, but EACH must have an UNLIMITED record