In this recent post I discussed a paper on regional temperature divergence issues related to irrigation from Dr. John Christy saying:
This press release below from Columbia University shown below suggests that irrigation cools the region undergoing irrigation. However, a study published three years ago of California’s central valley by Dr. John Christy suggests exactly the opposite. See this WUWT post from 2007, then read the Columbia story and decide for yourself.
A two-year study of San Joaquin Valley nights found that summer nighttime low temperatures in six counties of California’s Central Valley climbed about 5.5 degrees Fahrenheit (approximately 3.0 C) between 1910 and 2003. The study’s results will be published in the “Journal of Climate.”
Dr. Christy responds via email, and I’ve added graphs and links to enhance his presentation to us.
I would like to let the readers know (several of whom obviously did not read it) that the 2006 California paper (PDF here)was based on daily time series of 41 stations in the six south-central San Joaquin Valley and adjacent Sierra counties. 18 stations were in the valley, 23 in the Sierra (mainly the foothills).
The time series for each of these stations was divided into segments based on the examination of all NWS forms that NCDC was able to find – over 1500 pages. I examined each form and personally digitized its basic information (see attached.) In the example, you can see the results of reading 25 of these forms for Madera CA. I keyed in the
relevant information, then at the bottom of the spreadsheet summarized the results with a listing of the dates and reasons for establishing a segment breakpoint, in a sense treating each segment as a separate “station”.
See this PDF file showing metadata for Madera_045233
I sent the digitized metadata files for each station to the WRCC for safekeeping if for some reason I lose them.
I also hand-digitized several years of DAILY TMax, TMin too – for downtown Fresno, it was about 40 years worth … a very tedious project for a grandpa like me.
This method in essence provides a set of time series which can be debiased relative to each other through a method described in the paper that used mathematical game theory. By having this many stations (order 20 in Valley and Sierra separately) I have oversampled the region’s signal, thus allowing, as shown in the paper, several ways to test the resulting trends and issue statements of confidence.
We are very confident that TMin is warming rapidly in the Valley while the TMax trend is not significantly different from zero. The seasonal distribution of this result is consistent with irrigation. The Sierra time series show no such TMin trends – also consistent with
irrigation/development in the Valley but not Sierra. The fact trends in TMax and TMin in the Sierra are not rising is evidence against the notion that the lack of rise in TMax in the Valley is due to irrigation (though it could have some impact … but keep in mind that a number of these Valley stations have become urbanized in any case – perhaps a bigger daytime signal than irrigation.)
Since TMax is more highly coupled to the deep atmosphere (daytime mixing), TMax is therefore the better candidate to be used as a proxy for greenhouse warming since this type of warming is maximized in the troposphere according to model theory. The temperature results in this part of California (see also Christy and Hnilo 2010 on snowfall) do not provide evidence of greenhouse theory expectations.
Some have mentioned other factors that contribute to rising night temps, and these are discussed more fully in Christy et al. 2009 regarding East Africa temps (another tedious effort to digitize numerous stations here-to-fore never examined.) The key factor in all of these causes is the idea that the delicate, nocturnal boundary layer in which TMin is
measured can be easily disrupted by many factors – warmth from surface fluxes due to irrigation, IR forcing from aerosols, buildings which increase roughness (and thus mixing) etc. When the BL is disrupted, the warmer air above the inversion is mixed downward and is evidenced by a warmer TMin than would otherwise be the case.
I have not read the paper being discussed which suggests irrigation cools the temp stations. This would be a difficult thing to establish from typical datasets as these are “homogenized” in a way that smears the station trends from other stations of quite different micro-climate situations into each other. As noted above, I kept the Valley and
Sierra distinct and used many more stations (a station density of at least 10 times greater.)
John R. Christy
Professor, Atmospheric Science
Director, Earth System Science Center
Alabama State Climatologist