export of GISTEMP data
Hugo Ahlenius
I am a cartographer and GIS analyst, and I have need for preparing
some maps out of the NASA data - for a publication. The author of a
chapter has selected maps like:
http://tinyurl.com/37s78mz
...and it is my task to prepare something nice out of this. I could
work from the pdf-files from that link, but I also found ccc-gistemp,
and I am curious what I can do out of this.
What I need is to have some data exported that I can work with in the
software that I have handy - primarily ArcGIS and Global Mapper. I can
also add that I have some level of proficiency in Python (although a
bit basic).
Is it possible to export averages or annual/seasonal world wide data
using ccc-gistemp?
I ran ccc-gistemp, but I am not really sure what is happening, event
though I have skimmed through the documentation (please refer me to
any important points, if I can read up on this somewhere).
If I can get the data into some kind of grid/text/ascii-format I can
then import it into my GIS software. I can also do calculations - such
as surfaces, averages and calculate anomalies there, if needed!
Thanks!
Regards,
Hugo Ahlenius
http://nordpil.com
David Jones
> I am a cartographer and GIS analyst, and I have need for preparing
> some maps out of the NASA data - for a publication. The author of a
> chapter has selected maps like:
> http://tinyurl.com/37s78mz
Hi, and thanks for getting in touch.
> ...and it is my task to prepare something nice out of this. I could
> work from the pdf-files from that link, but I also found ccc-gistemp,
> and I am curious what I can do out of this.
>
> What I need is to have some data exported that I can work with in the
> software that I have handy - primarily ArcGIS and Global Mapper. I can
> also add that I have some level of proficiency in Python (although a
> bit basic).
>
> Is it possible to export averages or annual/seasonal world wide data
> using ccc-gistemp?
Yes. It may require some additional programming.
> I ran ccc-gistemp, but I am not really sure what is happening, event
> though I have skimmed through the documentation (please refer me to
> any important points, if I can read up on this somewhere).
Documentation is lacking. It would be useful to know what
documentation you tried to look for, or what documentation would be
most useful. Surely some of this e-mail I'm writing now will make its
way into documentation.
It's probably a good idea to read Hansen & Lebedeff 1987,
http://pubs.giss.nasa.gov/abstracts/1987/Hansen_Lebedeff.html , or at
least skim it. In particular, note how the grid is constructed.
> If I can get the data into some kind of grid/text/ascii-format I can
> then import it into my GIS software. I can also do calculations - such
> as surfaces, averages and calculate anomalies there, if needed!
I'm keen to see ccc-gistemp produce datasets that can be plugged into
GIS software, but I don't have or routinely use GIS software. That
suggests a useful colloboration.
Right now in terms of gridded datasets ccc-gistemp produces 2 files of
interest, but neither are quite what you want (hence, some programming
required).
result/SBBX1880.Ts.GHCN.CL.PA.1200
This is a Fortran binary dataset in the same format that GISTEMP emits
(see ftp://ftp.giss.nasa.gov/pub/gistemp/SBBX1880.Ts.GHCN.CL.PA.1200
). It consists of the land stations only (interpolated over the
ocean, but no Sea Surface Temperature records are included in this
dataset).
work/step3.v2 (or work/v2.step3.out in earlier versions, I very
recently renamed the intermediate files)
This is the same gridded data as above, but in ASCII format.
Specifically, a variant on the format used by GHCN v2 (hence the v2 in
the file name).
Specifically, an extract:
+64.8-175.5C1881-9999-9999-9999-9999-9999 -71 -101 37 107 289 198 -4
+64.8-175.5C1882 251 112 281 231 -20 29 49 -23 -23 169 48 -264
+64.8-175.5C1883 141 242 -179 -349 -60-9999-9999-9999-9999-9999-9999-9999
+64.8-175.5C1892-9999-9999-9999-9999-9999-9999-9999-9999 17 -111 58 56
+64.8-175.5C1893 -786 262 191 309 91 -77 -67 -78 -45 -118 -121 -262
(using python notation)
row[0:11] is the centre of the cell in ISO 6709 format, to the nearest
0.1 degree;
row[11] is the letter 'C';
row[12:16] is the calendar year
row[16:76] is the 12 monthly values, each is an integer formatted in 5
characters, -9999 denotes missing value.
(Thus the file is a map of time series, rather than a movie of maps)
The grid values are anomalies in units of 0.01C (the baseline period
is gridding_reference_period in parameters.py and is normally 1951 to
1980; but exercise caution when the cell has few or no years in that
period (see http://clearclimatecode.org/byrd/ )).
The grid used is unusual (see Hansen & Lebedeff 1987, and possibly
code/eqarea.py). It is not a regular lat/lon plate carree grid (as
seems to be popular with other datasets): each hemisphere is divided
into 4 bands by splitting at the latitudes that have sines of 0.4,
0.7, 0.9 (and 0 and 1). Thus the bands have areas in the ratio
4:3:2:1. Each band is divided into 10 equal area zonal strips, and
each strip is divided by longitude into a number of cells proportional
to the area of the band in which the strip is found: 160,120,80,40.
(internally this division takes place via 80 "boxes", but it's
equivalent). Thus each cell has equal area (and there are 8000 of
them). But cells do not, in general, have equal latitude or longitude
bounds.
Possibly your GIS software has a regridding function that you could use.
It seems to me that to be of use to you ccc-gistemp would need to:
- produce a gridded dataset that includes sea surface temperatures
(this is generated internally before computing zonal and global
anomalies, but not currently output);
- produce gridded datasets that could be consumed by GIS software;
- (and possibly) either regrid to a standard grid, or modify the
algorithm to use a standard grid rather than the Hansen & Lebedeff
grid.
Those all require effort. If you're working on a commercial project,
Climate Code Foundation would be happy to discuss rates, quotes, and
contracts for any of that.
Hope that helps,
David Jones
Climate Code Foundation