I’ve just updated my HADCRU2 dataset and, for good measure, also the current edition of CRUTEM2. I’ve updated my scripts and, for those brave souls that want their very own copy of HADCRU2 or CRUTEM2 in R format, I’ve posted up scripts to make R tables organized like time series usually are, with each column being a gridcell time series. I’ll post up some utilities as well for locating a gridcell-column from lat-long coordinates and the inverse and some plotting layouts that I’ve been using recently.
The following downloads are semi-automatic. You have to manually download and unzip the data and locate the unzipped data in a directory called c:/temp. I think that the manual download and unzipping could be automated within R, but I don’t do it often enough to learn it yet.
My computer is a couple of years old and only has 128 MB of SDRAM. This necessitated that I do this in two parts; otherwise it crashed. However the datasets as assembled can be loaded and used easily. Machines being sold now have much more SDRAM and the division in two parts would not be necessary for such a machine. (I haven’t needed more horsepower for little datasets like proxy datsets, but it’s time for me to upgrade anyway.)
The Jones data was downloaded from http://www.cru.uea.ac.uk/ftpdata/hadcrut2.dat.gz on August 14, 2005. The August 2005 download produced an R-table of 1791 months from Jan 1856 to Mar 2005, with 2592 columns of 5×5 degreee grdicells, big hand: 36 lat bands N to S; little hand 72 long bands W to E (from dateline). Values of -9999 are changed to NA. Script
The Jones data was downloaded from http://www.cru.uea.ac.uk/ftpdata/crutem2.dat.gz on August 14, 2005. The August 2005 download produced an R-table of 1815 months from Jan 1851 to Mar 2005, with 2592 columns of 5×5 degreee grdicells, big hand: 36 lat bands N to S; little hand 72 long bands W to E (from dateline). Values of -9999 are changed to NA. Script
I know that there is quite a bit of interest in temperature datasets. Some of the insights to be gained are through examination of details. Many people who are working on them are using difficult languages like Fortran or time-consuming languages like Excel and I supsect that their productivity is much lower than it could be. One of the best decisions that I ever made was downloading R and getting familiar with it. You can download the R language from this link. While I’m now pretty good at R, I was able to get results right away and the payback time is pretty much instantaneous. R has a vast library of statistical and time series accessories, which provide remarkable power.
I’ve also gotten pretty good at tweaking special purpose scripts to make time series tables from ftp files of disparate formats. I’ll write a little manual some time of the tricks that I look for in downloading. I could also do a script pretty quickly (5 minutes to an hour) for some of the other data sets. so if someone who’s not used to R wants access to GISS or something else and posts up some details of access information, I’ll try to take a look at it.
I’m going to start making a file of utilities that I find handy so that others can do this as well. This script (not up yet) will have utilities as follows:
jones – calculates column from latitude and longitude coordinates;
jonesinv- calculates latitude and longitude center of gridbox from jones format column #
plot1 – plots autocorrelation function and time series plot for an individual gridcell
I’ve got lots of other utilities, which I’ll look at see what’s applicable to the temperature dataset.