1NCES(1)                     General Commands Manual                    NCES(1)
2
3
4

NAME

6       nces - netCDF Ensemble Statistics
7

SYNTAX

9       nces  [-3]  [-4]  [-5]  [-6]  [-7]  [-A] [--bfr sz_byt] [-C] [-c] [--cb
10       y1,y2,m1,m2,tpd]  [--cnk_byt  sz_byt]  [--cnk_csh  sz_byt]   [--cnk_dmn
11       nm,sz_lmn]   [--cnk_map   map]   [--cnk_min   sz_byt]  [--cnk_plc  plc]
12       [--cnk_scl sz_lmn] [-D dbg_lvl] [-d dim,[  min][,[  max]]]  [--dbl|flt]
13       [-F]   [--fl_fmt=fmt]   [-G   gpe_dsc]  [-g  grp[,...]]   [-h]  [--hdf]
14       [--hdr_pad sz_byt] [-L  dfl_lvl]  [-l  path]  [--msa]  [-N]  [-n  loop]
15       [--no_cll_msr]  [--no_frm_trm]  [--no_tmp_fl]  [--nsm_sfx grp_sfx] [-O]
16       [-p path] [--ppc var1[, var2[,...]]= prc]] [-R]  [-r]  [--ram_all]  [-t
17       thr_nbr]  [--uio] [--unn] [-v var[,...]]  [-X box] [-x] [-y op_typ] in‐
18       put-files output-file
19

DESCRIPTION

21       nces performs gridpoint averages of variables across an arbitrary  num‐
22       ber  (an  ensemble)  of  input files, with each file receiving an equal
23       weight in the average.  Each variable in the output-file  will  be  the
24       same  size  as  the same variable in any one of the in the input-files,
25       and all input-files must be the same size.  Whereas ncra only  performs
26       averages  over  the  record  dimension  (e.g.,  time), and weights each
27       record in the record dimension evenly, nces averages entire files,  and
28       weights  each file evenly.  All dimensions, including the record dimen‐
29       sion, are treated identically and preserved in the output-file.
30
31       The file is the logical unit of organization for the  results  of  many
32       scientific  studies.   Often one wishes to generate a file which is the
33       gridpoint average of many separate files.  This may be to  reduce  sta‐
34       tistical  noise  by  combining the results of a large number of experi‐
35       ments, or it may simply be a step in a procedure whose goal is to  com‐
36       pute  anomalies  from  a  mean state.  In any case, when one desires to
37       generate a file whose properties are the mean of all the  input  files,
38       then nces is the operator to use.  nces assumes coordinate variable are
39       properties common to all of the experiments and  so  does  not  average
40       them  across  files.  Instead, nces copies the values of the coordinate
41       variables from the first input file to the output file.
42

EXAMPLES

44       Consider a model experiment which generated five  realizations  of  one
45       year of data, say 1985.  You can imagine that the experimenter slightly
46       perturbs the initial conditions of the problem before  generating  each
47       new  solution.  Assume each file contains all twelve months (a seasonal
48       cycle) of data and we want to produce a single file containing the  en‐
49       semble average (mean) seasonal cycle.  Here the numeric filename suffix
50       denotes the experiment number (not the month):
51              nces 85_01.nc 85_02.nc 85_03.nc 85_04.nc 85_05.nc 85.nc
52              nces 85_0[1-5].nc 85.nc
53              nces -n 5,2,1 85_01.nc 85.nc
54       These three commands  produce  identical  answers.   The  output  file,
55       85.nc,  is the same size as the inputs files.  It contains 12 months of
56       data (which might or might not be stored in the record  dimension,  de‐
57       pending  on  the input files), but each value in the output file is the
58       average of the five values in the input files.
59
60       In the previous example, the user could have obtained the ensemble  av‐
61       erage  values in a particular spatio-temporal region by adding a hyper‐
62       slab argument to the command, e.g.,
63              nces -d time,0,2 -d lat,-23.5,23.5 85_??.nc 85.nc
64       In this case the output file would contain only three slices of data in
65       the  time  dimension.   These three slices are the average of the first
66       three slices from the input files.  Additionally, only data inside  the
67       tropics is included.
68
69

AUTHOR

71       NCO  manual pages written by Charlie Zender and originally formatted by
72       Brian Mays.
73
74

REPORTING BUGS

76       Report bugs to <http://sf.net/bugs/?group_id=3331>.
77
78

80       Copyright © 1995-present Charlie Zender
81       This is free software; see the source for copying conditions.  There is
82       NO  warranty;  not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR
83       PURPOSE.
84
85

SEE ALSO

87       The full documentation for NCO is maintained as a Texinfo manual called
88       the  NCO Users Guide.  Because NCO is mathematical in nature, the docu‐
89       mentation includes TeX-intensive portions not  viewable  on  character-
90       based  displays.  Hence the only complete and authoritative versions of
91       the NCO Users Guide are the PDF (recommended), DVI, and Postscript ver‐
92       sions  at <http://nco.sf.net/nco.pdf>, <http://nco.sf.net/nco.dvi>, and
93       <http://nco.sf.net/nco.ps>, respectively.  HTML and  XML  versions  are
94       available          at          <http://nco.sf.net/nco.html>         and
95       <http://nco.sf.net/nco.xml>, respectively.
96
97       If the info and NCO programs are properly installed at your  site,  the
98       command
99
100              info nco
101
102       should  give  you access to the complete manual, except for the TeX-in‐
103       tensive portions.
104
105       ncap(1), ncap2(1), ncatted(1), ncbo(1), ncclimo(1), nces(1), ncecat(1),
106       ncflint(1),  ncks(1), nco(1), ncpdq(1), ncra(1), ncrcat(1), ncremap(1),
107       ncrename(1), ncwa(1)
108
109

HOMEPAGE

111       The NCO homepage at <http://nco.sf.net> contains more information.
112
113
114
115                                                                       NCES(1)
Impressum