In another thread
East Anglia Refusal Letter
Steve posted a reply from the University of East Anglia,
at which CRU is located, and that reply indicated that
the data are somewhere in GHCN, so Jones is not required
to be more specific.

JerryB

Thanks very very much for all the trouble! I’ll look carefully at the refs.

One comment: if CRU & Jones et al don’t give the origins of their data, this is ?analogous to an historian writing something but not giving the documents on which it’s based…(?)

page linking PDF files describing GHCN

“>
page linking PDF files describing GHCN

I am concerned that a brief reply may mislead, but an adequate reply may
take me so long to compose that I won’t finish it. So, let me try
something, but be cautious about inferring too much from it.

A large amount of historical temperature data has been gathered into
something called GHCN (Global Historical Climatolgy Network). Various
sources were used for various lengths of time. How accurate the numbers
are is not clear. The exact locations often not clear. Multiple
sets of numbers occur for some locations, but when are different numbers
due to different sources, and when are they due to actually being for
different locations, is not clear.

GHCN monthly data

page linking PDF files describing GHCN

In any case, different people use various subsets of GHCN data to concoct
global, and hemispheric, temperature averages, usually in 5 by 5 latitude/
longitude grid cells. Some, such as GISS in the US, make it clear which
“locations” they use, and publish their adjustments location by location,
month by month. One may agree, or disagree, with their adjustments, but
you can see what the adjustments are.

For GISS see
for station data.

GISS station data processing is briefly discussed in the paper
“GISS analysis of surface temperature change”

linka

The paper “A closer look at United States and global surface temperature change”
linkb
discusses the inclusion of some USHCN adjustments (not including
estimates of missing data, and not including urban heat island
adjustments) into the GISS station data files.

On the other hand, CRU (Climate Research Unit) in the UK publishes grid
cell, and hemispheric, and global, average estimates, but does not
publish which stations they use, in what years, or with what adjustments.

For CRU see
but no station data
and at is an older
version, also with no station data.

Brohan et al

linkc briefly discusses recent changes of CRU versions.

A GRL paper, abstract at
linkd
compares some aspects of GISS, CRU, and NCDC(GHCN) grid cell concoctions,
and may, if not explain, suggest some reasons for some differences.

Now, back to the topic of this thread: one of the issues of historical
temperature adjustments concerns UHI, or UHIE (Urban Heat Island (Effect)).
One of the papers that claimed to support very little need for UHI
adjustment was Jones et al 1990, Jones being Phil Jones of CRU. It seems
that he has avoided publishing the names of the stations, and the
temperature numbers, used in that study, thereby preventing anyone from
checking his work.

JerryB,

Thanks very much for the clarification.

Do I understand correctly:

1. Over long periods, the provenance of the ‘raw’ data _may_ change. 2. In any case, unknown ‘adjustments’ are made. 3. These last are the basis for policy discussion/recommendation. 4. Questioners are, ipso facto, ‘fringe’ scientists. ??

Thanks for the clarification.

Do I understand correctly: Unknown ‘adjustments’ are made to ‘observed’ data, whose provenance _may_ have changed over long periods. And these ‘adjusted’ data are the basis of policy discussion/recommendation? And those who question this are treated as ‘fringe’ scientists?

Sudha,

There is no indication that anyone is altering records of
“raw” or “original” temperature data.

When someone attempts to find “trends” in temperature data,
they often meet certain kinds of problems, sometimes due to
changes that occurred over many years. For example, the
times of day at which temperature observations were made may
have changed, the location of the thermometer may have
changed enough to give different readings than it would have
if the location remained the same. Also, estimating missing
observations can be regarded as an adjustment.

Some adjustments are clearly stated, as in the USHCN (United
States Historical Climatology Network), which lists raw
monthly averages, and separately lists results of various
adjustments to those monthly averages, month by month, year
by year. One may, or may not, be confident in their
adjustments, but they are clearly distinct from the raw
numbers.

Others may not be clear about the adjustments that they may
make, which would not much matter if their results were not
being used for major policy decisions.