While discussing Darwin data I was pointed to this file master.dat.com in the FOIA(leaked ClimateGate file) documents folders. I was amazed to find the 14MB file is in fact the Jones et al 1999 station data – another gem from FOIA.
So this is a considerable step forward in our uncovering of Jones et al station data – the previous most recent Jones station data I had was the much expanded Jones 1994 data. (note the 1996 update has few stations).
I was amazed to find that the 1999 station list (2664) has fewer stations than the 1994 update (2961 quoted in the paper but 3555 in the digital download from CRU). If I had been asked my opinion I would have bet that Jones steadily increased his stations list with each new version.
Notes re graphic added 2 Jan 2010. 1986, total from journal papers, 1994 ditto, 1995 total from digital file of monthly T data ex CRU website, 1999 total from digital file of monthly T data ex FOIA file ClimateGate, 2003 total from Jones & Moberg journal paper, 2006 total from Brohan et al journal paper, 2007 station list released by CRU on their website, UKMO09 – partial release of station monthly T data – presumably they have another ~2000+ to release. My use of “hide the decline” refs to the drop in station numbers from 2961 in 1994 to 2664 in 1999. Note this issue is NOTHING to do with the number of stations open and recording at any time. A good guide to that % can be arrived at by the Jones 1996 update file which listed only stations currently recording and that had 1226 stations. So at that time less than half of all stations used by Jones were still recording.
Now we only need to discover the Jones & Moberg 2003 monthly station T data – then if the UKMO station data keeps emerging, we are in sight of a sequence of how Jones et al stations changed over over 20 years.
The 1999 station data file relates to this paper.
Jones, P.D., M. New, D.E. Parker, S. Martin, and I.G. Rigor. 1999. Surface air temperature and its changes over the past 150 years. Reviews of Geophysics 37:173-199. (Free pdf available)
My use of the well known phrase from ClimateGate emails, “hide the decline” is to highlight the decrease in station numbers from the 1994 list to 1999. IMHO this is somewhat at odds with the following dscription from Jones et al 1999.
In section 2.1. Land Component on page 174
“Here we use the land station data set developed by Jones . All 2000+ station time series used have been assessed for homogeneity by subjective interstation comparisons performed on a local basis. Many stations were adjusted and some omitted because of anomalous warming trends and/or numerous nonclimatic jumps (complete details are given by Jones et al. [1985, 1986c]).”
My comments on the above Jones et al 1999 paragraph.
 I take the first sentence to mean that no new stations were examined or used – only those in Jones 1994.
 Surely all 1994 stations passed the Jones homogeneity checks as described in his 1985 & 1986c references, the DoE TR022 and TR027 books.
 Yet Jones et al 1999 above speaks of, “Many stations were adjusted and some omitted because of anomalous warming trends and/or numerous nonclimatic jumps..”
 The obvious question arises, why were stations omitted when all stations were from Jones 1994 which had already gone through the homogenization process ?
 Also why were many stations further adjusted when all stations were from Jones 1994 which had already gone through the homogenization process ?
 Why did not reviewers ask for some better precision than the use of 2000+ and some clear explanation of what was involved in the statement, “Many stations were adjusted and some omitted because of anomalous warming trends and/or numerous nonclimatic jumps..” ?
Post ClimateGate we know that Jones et al operated in a cozy bubble of permissive peer review populated by friendly allies.