GISS warms rural stations in Florida

GISS adjustments insert warming into rural stations

While working on the Florida 5 degree grid cell page, strange high warming trends were apparent in GISS (NASA's Goddard Institute for Space Studies in New York) graphics at their web site for (at least) two rural stations, Arcadia and Everglades.

Comparisons are shown below between the US HCN data (from the V2 GHCN adjusted data files) and the more recent GISS homogeneity adjusted data using satellite night lighting.

Arcadia Comparison

It is immediately apparent that a huge warming adjustment of circa one degree Celsius has been inserted into Arcadia over last century.

Everglades Comparison

A similar but slightly less severe adjustment is seen in Everglades which has data from 1926.

Seeking enlightenment from Dr Jim Hansen of GISS I was told that;

yes the adjustments above looked questionable;
this happened because some of the rural areas in Florida did not come out black enough in the satellite night lights images which leads to the computer treating them as some category of Urban;
then data "smoothing" leads to insertion of the above warming trends;
it's really not kosher to go in on a case by case basis;
our approach is to set up an objective scheme and then let the computer and the satellite make the decision - its not necessarily the best way, but its objective, even though, as our caveat states, it is going to give "bad" results at some individual stations.

The caveat referred to above is from a paper in press at Journal of Geophysical Research-Atmospheres, titled, "A closer look at United States and global surface temperature change", by J. Hansen, R. Ruedy, M. Sato of NASA Goddard Institute for Space Studies NY, M. Imhoff, W. Lawrence of NASA Goddard Space Flight Center MD and D. Easterling, T. Peterson and T. Karl of NOAA National Climatic Data Center NC.

I am sure interested readers can obtain a copy of this paper from GISS.

Comments:

I can not accept that it is "not kosher" to go in and look at case by case studies of components of large area data studies.
If the results can not stand up to case by case testing then it is highly likely that the entire process is flawed and it is better to look for a better process that will stand up to detailed investigation at a station by station level.

I remain puzzled that if the satellite wrongly classifies a rural station such as Arcadia as urban, that this could lead to a warming adjusment ?? Would not logic dictate that the computing process should subtract some warming trend from Arcadia leading to a version that cools (wrongly) even more ?

Can people with local knowledge tell me if there is urban development near these stations or is the satellite picking up spurious night light for some other reason.

My advice to the authors above would be to get on to the Editor at JGR fast and withdraw the paper from publication, hoping like hell it has not gone to the printers.

I have not done any other checking of rural trends in USA stations using GISS data in the last few months, who knows what else may turn up.
As featured on my page ghcn.htm I have pointed out non-climatic cold anomalies in early 20C & late 19C USA rural data from the V2 GHCN adjusted data files.

What is the significance of the new GISS rural adjustments for the global warming debate

Global warming as promoted by the IPCC is defined by the Jones et al datsets and I am progessively demonstrating on these pages that those data are badly contaminated by UHI warming at a station by station and grid cell level.

Many sceptics out here think that UHI contamination is involved in the difference between surface and satellite temperature trends, an issue strenously debated far from the IPCC.

In 1997-98 the American GHCN team from NOAA NCDC published findings from rural data supporting the Jones et al global trends. Many of us are still sceptical of this conclusion but we do not have the computing power to check what they have done..

Now we have "son of GHCN" in press with NASA programs chiming in with NOAA / NCDC and no doubt they will also come up with global trends that agree with the Jones et al studies (that we know contain UHI contamination).

Purely by chance I find absolutely ridiculous non-climatic warming adjustments to two Florida rural stations. Does anyone seriously think I just accidently stumbled across the only non-climatic warming corrections in USA rural data ?
Of course not, there will be many others to be found.
Will anyone put money on there being compensating large non-climatic cooling corrections ?

Posted 19, July, 2001
You saw it first here.

Back to Review comments on climate papers by key "IPCC Supportive" scientists main page

Back to National Met. Institutes page
Back to Front Page