Bogus Siberian Warming

Northern Siberia, Taymir Peninsular Region

This review reveals shoddy scientific procedures used by the IPCC to conjure up glossy warming claims for the 1976-1999 period when in fact temperatures in that region have cooled slightly over 60 years.

Figure 1 shows the region being examined within a rectangle and the very eye catching IPCC presentation of warming trends 1976 to 1999 by 5 degree grid cells from Figure 2.9 in the IPCC Third Annual Report (TAR) draft.

Data Sources: The writer has examined CRU 1991, Jones 1994, V2 GHCN and GISS temperature records for this region.

The station by station data presented below is only from GISS for the following reasons;

Jones 1994 data is not as long term as GHCN and GISS and have little data past 1991,
GHCN records are generally not updated past 1989-1990,
several GISS records extend through to 1999, they often extend back to a few earlier years, gaps are often filled and monthly data errors in Jones 1994 are often corrected in GISS.
there are no significant trend differences between the datasets apparent as yet,
it helps keep this paper simple.

Figure 2.9 from the TAR shows a block of four maps and the first impression for any reader is that the eye is drawn to the cluster of large red dots over Siberia on the 1976-1999 map. In the TAR text there is no qualification for this cluster of high warming anomalies but Figure 2.10 goes on to present the four seasonal 1976-1999 global anomalies, where the winter & spring anomaly also shows the eye-catching patch of Siberian high magnitude warming anomalies leading the way in "global warming".

Only a handfull of people would be familiar with temperature records from that region, so most readers of the TAR would carry away an impression that there is significant Siberian warming underway. The warming is in fact a partial recovery from a cold period in the early 1960's and vanishes rapidly if you choose a different starting year near 1976. The TAR mentions circulation perturbations that might make 1976 an interesting start point but that is no excuse for the portrayal of a region with little temperature trend as a major focus of global warming.

The data show that temperatures in that region have not yet recovered to levels seen in the 1940's.

Note also that the start point for the 1946-1975 anomaly map in the TAR Fig 2.9 does not correspond to the peak of the 1940's warming in the Jones 1994 global data. Those data peak in 1944, so if that anomaly map was 1944-1975 there would be more blue and less red. On the general subject of bias, why use red dots for zero values, can't they afford software that makes the zero's black ?

Figure 2 shows trends from five stations across the region and the salient facts are obvious, (a) that there is no warming trend 1940's to 1999 and (b) the data are very noisy and (c) warming or cooling can be measured over various periods depending on the start / finish years.

Figure 2

Figure 3 shows a map of with stations named, grid cells numbered 1 to 9 and grid cells with no stations shaded grey.

Figure 4 shows an enlargement of the subject region in Figure 1 with anomalies numbered 1 to 9 corresponding to numbers on the grid cell map Figure 3.

Non Climatic Warming Two stations, Verhojansk Figure 5 and Olenek Figure 6 show obvious non climatic warming.
Olenek can be seen to be cooler than neighbours before 1967 and warmer than neighbours after.
Verhojansk warms steadily compared to neighbours.
Two examples of non climatic warming not detected by statistical or other homogeneity testing.

Comments by Grid Cells 1 to 9 as on the maps Fig 3 and 4.

[1]    77.5 N, 102.5 E    The only station Mys Celjuskin or GMO IMEKF cools slightly from the 1940's, similar to Hatanga.
[2]    72.5 N, 102.5 E    The only station Hatanga cools slightly from the 1940's, see Fig 2.
[3]    67.5 N, 102.5 E    The only station, Essej is similar to neighbours, see Fig 2   For some reason the IPCC have no red dot anomaly here, see Fig 4. Maybe Essej is short on data, extending only to 1993.
[4]    67.5 N, 107.5 E     No stations in this grid cell. So the anomaly is interpolated using the 8 neighbours; procedure yet to be published I am told.   No problems with reasonable interpolations.
[5]    67.5 N, 112.5 E     In this grid cell we have Olenek with the non climatic warming jump, see Figure 6. So the anomaly red circle number 5 in Figure 4 is larger than it should be.   Selagoncy only has data to 1990 so may not contribute here.
[6]    67.5 N, 117.5 E     No stations in this grid cell, yet we see a LARGER red circle anomaly than for neighbouring grid cells !!   No interpolation can look like this.
[7]    67.5 N, 122.5 E    Dzardzan cools slighlty since the 1940's see Fig 2, while Zhigansk stops at 1990. Figure 4 shows a smaller red circle anomaly here.
[8]    67.5 N, 127.5 E     No stations here but red circle anomaly is on a par with number 9, the size of which is dictated by Verhojansk,   see Fig 5.   No valid interploation process I am aware of should produce a result like this.   This interpolation has four smaller neighbours that should have pulled the size down.
[9]   67.5 N, 132.5 E       Only Verhojansk in this cell with the non climatic warming see Fig 5. So this red circle anomaly is larger than it should be.

Summary

[1]    The IPCC are using non-climatic warming trends as evidence of climate change.
[2]    Interpolations of data to grid cells devoid of stations ends up with higher warming trends than the parent data ???
[3]    Can anyone offer any explanations.

You read it first here

Posted 12, September, 2000

Back to Main Page
.