Ink Blot Rorschach Test: What do you see?

Graph of global temperatures as portrayed by Wikipedia.

The reconstructions used, in order from oldest to most recent publication are:

1. (dark blue 1000-1991): P.D. Jones, K.R. Briffa, T.P. Barnett, and S.F.B. Tett (1998). , The Holocene, 8: 455-471.
2. (blue 1000-1980): M.E. Mann, R.S. Bradley, and M.K. Hughes (1999). , Geophysical Research Letters, 26(6): 759-762.
3. (light blue 1000-1965): Crowley and Lowery (2000). , Ambio, 29: 51-54. Modified as published in Crowley (2000). , Science, 289: 270-277.
4. (lightest blue 1402-1960): K.R. Briffa, T.J. Osborn, F.H. Schweingruber, I.C. Harris, P.D. Jones, S.G. Shiyatov, S.G. and E.A. Vaganov (2001). , J. Geophys. Res., 106: 2929-2941.
5. (light green 831-1992): J. Esper, E.R. Cook, and F.H. Schweingruber (2002). , Science, 295(5563): 2250-2253.
6. (yellow 200-1980): M.E. Mann and P.D. Jones (2003). , Geophysical Research Letters, 30(15): 1820. doi:10.1029/2003GL017814.
7. (orange 200-1995): P.D. Jones and M.E. Mann (2004). , Reviews of Geophysics, 42: RG2002. doi:10.1029/2003RG000143
8. (red-orange 1500-1980): S. Huang (2004). , Geophys. Res Lett., 31: L13205.doi:10.1029/2004GL019781
9. (red 1-1979): A. Moberg, D.M. Sonechkin, K. Holmgren, N.M. Datsenko and W. Karlén (2005). , Nature, 443: 613-617. doi:10.1038/nature03265
10. (dark red 1600-1990): J.H. Oerlemans (2005). , Science, 308: 675-677.doi:10.1126/science.1107046

(black 1856-2004): Instrumental data was jointly compiled by the Climatic Research Unit and the UK Meteorological Office Hadley Centre. Global Annual Average data set TaveGL2v [2] was used.

Notice that the references are chock full of names under investigation in association with the Climategate scandal.

The most important feature of the Wikipedia graph is the upward spike on the right side, to be interpreted as rapid, out of control global warming.

Temperatures in the distant past were estimated by studying the signs left by nature. These are called "proxies" because they are not actually thermometers. We would expect thermometers to be the very best means to measure temperature.

Modern temperature records in the graph are heavily dependent on thermometer measurements made at stations all across many continents. They should be the most accurate, but they have been subjected to recent urbanization: Buildings, pavement and more. Here is a picture of a station that has been making measurements for a century but only recently surrounded by buildings and roads:


Here is the temperature recorded at this site since 1900. Of course the air conditioner, buildings and other hot surfaces were not present when this sensor was first installed. Notice the sharply rising trend.


The above graph shows temperatures rising by more than three degrees. But how much is due to global climate and how much from local urbanization? Unfortunately, how much change to the local terrain, and when are not documented. The accuracy of such sites are given a rank, as follows.


The above temperature station rates a 5, "worst". It is no exception. I photographed a station near my house in the country, but the station only qualifies as 4, "poor", surrounded by buildings and pavement, with errors that could exceed 2 degrees C according to the chart above. The majority of stations in North America have been compromised by urbanization effects to a comparable extent. In the below map, the desired color of sensors is blue. Unfortunately, blue is an all but nonexistent Waldo:

The country as a whole remains undeveloped. Yet virtually all sensors are located next to developed structures that get hot in sunlight. Returning to the original graph, how much can we trust the right side of the graph, the only time period which uses modern temperature sensors? The global warming graph at beginning of this article is attempting to show tenths of a degree increase. But urbanization can introduce up to five degrees. We don't have records of precisely how much the change is or when it happened, so the sensor record is fairly useless. It cooks the data and burns a big tall spike on the right side of the graph.

And we thought the temperature record based on thermometers would be the gold standard. It is next to useless for this purpose. Does that mean that farther in the past proxies are more accurate, say within tenths of a degree to match the accuracy implied in the graph? Not a chance! We have even more difficulty verifying their accuracy due to many confusing and contaminating factors. The entire graph is in serious doubt now.

All those errors assume people did the best job humanly possible to collect and analyze their results. But they did not, so the errors are likely much greater. A large percentage of the people who made the graph at the top of the page are under investigation for the climategate scandal and they are global warming activists, factors which seriously questions the science. At this point it is appropriate to throw all conclusions out the window and wait decades for new data to be collected. Just don't let it hit the sensor next to the building on its way out.

Putting it all together, the wikipedia graph says very little factually. About the only thing you can see here is whatever you expect to see. You either see warming or sure signs of misleading "noise", completely opposite conclusions depending on what is in your mind. Like an ink blot, nothing in the drawing is real, it's all about how you interpret it. The warming interpretation is incorrect.