| Wood |
Wigley and Jones |
| Page 297 |
Page 313 |
| Page 298 |
Page 314 |
| Page 299 |
Page 315 |
| Page 300 |
Page 316 |
| Page 301 |
Page 317 |
| Page 302 |
Page 318 |
| Page 303 |
Page 319 |
| Page 304 |
|
| Page 305 |
Below this Table are
listed points 1 to 9 with page #'s from Wigley and Jones that they list
as "Errors in Wood's Analysis." |
| Page 306 |
I juxtapose these points
with the page location in Wood where you can read what he actually said |
| Page 307 |
Then in the right hand
column I comment. I will steadily build a separate page for each
of the W & J 9 points. |
| Page 308 |
Working through the 9
points where W & J say there are "Errors in Wood's Analysis",
I can not pin point any of the nine points |
| Page 309 |
where that is a
justifiable statement. I think Wood was very fair, faced
with the diehard agenda of Jones et al to include as many |
| Page 310 |
urban affected records into global trends as they could. |
| Page 311 |
|
| Page 312 |
| Wigley
and Jones points 1 to 9 |
What
Wood actually had to say |
Comments by
WS Hughes |
| 1, p313, claim that Wood
said of Jones et al 1986 , "urban warming was clearly identified at
only 38 stations out of 2666". W & J puff on that, "This is a
serious distortion of the facts." |
p298 Wood did not use "only". Reading on Wood seems to cover the issue fairly. On page 316 Wood draws attention to the fact that in the SH only 3 stations were excluded for urban bias, Sao Paulo, Madang and Whenuapai. | The UHI is the single
most pervasive non-climatic factor in global temperature data. We know
so much about the UHI that the most reasonable approach is to build a
series of wide area trends using various population categories. Then
the UHI influence will be obvious to see. I can not see where
Wood distorts any of W & J's facts. |
| 2, p313 W
& J make the point that they use many more stations for
intercomparisons then they record in their TR022 and TR027
documentation. |
Table II p 301 and p298
and 300. |
W & J should
record all the stations they use in intercomparisons. They can not
expect Wood or any of us to be mindreaders. I notice they still
do not document this information. |
| 3, p314 W & J struggling to muddy Wood's obvious point that "urban warming is correlated with population growth" | Table III p 303 and discussion 302. | W & J omit to mention that temperature records derive from stations established to provide weather data. The network was never designed to provide the precision to determine "climate" trends to tenths of a degree. Also never designed to track minute differences between urban and more rural settings. |
| 4, p314 W & J straining to make a lot of this point that you can not jump to the conclusion that because a station bears a city name, that it is central in the city. | Sorry, but I have not found where Wood does this. | I would say, sites
peripheral to a city are usually EXACTLY where the rate of urban growth
is highest. I also note W & J give no example of what Wood is supposed to have done here. |
| 5, p314 W & J Extension of point 4. W & J say that San Juan International Airport "did not survive our various tests" and that "..we interpreted the apparent trend as a step change of 0.8 degC around 1970 and corrected the record accordingly." | Wood discusses San Juan
Int AP mid page 302. He refers to the Duchon 1986 study of the UHI at
San Juan Int Airport which clearly proves the trend in San Juan is
urban warming and not a step change as Jones et al say. For
downloadable pdf of Duchon 1986 Click on, Print Version for Duchon, Claude E. |
There are listings for
San Juan in TR022 Appendix A and B which says it was "corrected"
and used in the gridding process. San Juan was compared to Santo
Domingo ~400 kms west and the years 1899 to 1969 were corrected
for what Jones et al saw as an upward step change at 1970. Looks like a
classic example of a proven Jones et al error. This Fig 4
from Duchon 1986 illustrates EXACTLY what Jones et al so
often did
wrong. They included the UHI affected trend from San Juan
with its conveniently more complete data and under-represented the
other less perfect data, which probably holds trends that more
accurately reflect regional temperature trends. Jones included 4
Puerto Rico stations that the GHCN classes as "Rural", Arecibo, Lajas,
and Corozal & Isabela Substations, which have data from 1951-1970.
If these are averaged and differenced with San Juan, the figures show
that San Juan warms at 0.56 degress over the 4 rurals for
1951-1970. One wonders how many bells would have had to ring for
Jones et al to wake up to their mistakes. See blog posts San Juan Puerto Rico, EXACTLY how UHI warming can get into global gridded T trends. and, True temperature trends for Puerto Rico hidden in fragmented data and universal UHI illustrations, GISS/NASA/NOAA graphics illustrate significant UHI truths |
| 6, p314 W & J point actually not numbered (good editing and reviewing there) but it has to be the para commencing "Wood suggests that the detection threshold used by Jones et al, in the last analysis stage may have been too high.........." | Wood discusses that Jones et al 1986 used an urban warming threshold that was too high, lower third of page 302 | The W & J reply
agrees that Wood got correct his number of 0.89 degrees for the
difference for the 38 NH stations and their rural pairs, rejected for
urban warming. A great result for Wood. How this makes Wood "in
error" is utterly beyond me. Then W & J go on to say that
they will keep their threshold secret !!! Not in the real world Tom and Phil. |
| 7, p314 & 315 W & J straining to make Wood seem misleading. Yet they have to admit to using ten urban stations where Kukla et al have proved urban warming. | Wood gives examples of Jones et al urban rural comparisons in Table II page 301 and discusses the Kukla et al 1986 paper with its 34 urban rural pairs on bottom page 303. | Tom and Phil., 10 cities is 10 too many. In fact 20% of the 297 USA cities used without correction, are greater than 50,000 in population. Using the out of date GHCN pop codes. Look at the cities used in the Southern Hemisphere, with populations. (much more to come on this point) |
| 8, p315 W & J saying that just because intercomparisons are not listed does not mean they have not been carried out.. | Wood discusses the sometimes lack of stations to compare on page 304 | I say, "document all your intercomparisons W & J", stop hiding data. Your intercomparisons for Australian cities, see point 9 below, show they can be over distances from hundreds of kms up to thousands of kms. Bizarre !! |
| 9, p315, W & J make
the astonishing claim that, "...the distance criticism is a red
herring." |
Wood notes top of
304 that some of Jones et al comparisons are between
station hundreds of miles apart. |
I have noticed this in
the Jones et al Australian data too and am
frankly staggered that W
& J would defend such an indefensible position.
Increase distance between intercomparisons in any particular region and
you decrease the validity of any correlation. Science 101. |
