Death Valley Temperatures, Part II: Thoughts from William T. Reid

From Dr. Roy Spencer’s Global Warming Blog

Roy W. Spencer, Ph. D.

NOTE: Since he has done extensive investigation into some implausibly hot temperatures reported in Death Valley, I asked Bill Reid to comment on my previous blog post where I maintain that the world record 134 deg. F highest recorded air temperature was likely biased warm by about 10 deg., and should not be accepted as a world record. What follows are Bill’s initial thoughts on the subject. Also, based upon his comments, I will likely update the charts found in my previous blog post with more realistic temperature lapse rate values in the early 20th Century when insufficient stations were available to determine accurate lapse rates.

by William T. Reid

A big thank you to Dr. Spencer for investigating the current (very dubious) world high-temperature record and for bringing attention to my Death Valley climate research. There are a handful of ways, both climatologically and meteorologically, to show that Greenland Ranch’s reported maximum of 134F on July 10, 1913, is likely not valid.

Dr. Spencer’s methodology here (comparing the Death Valley maximums to those the closest surrounding stations, with adjustments for station elevation) is indeed a devastating blow to the authenticity of the suspect observations. What it basically demonstrates is that the lower troposphere was not hot enough to support temperatures much above 125F in July, 1913. I have compared regional maximums for all of the hottest summertime events since 1911. In practically all instances (in which the Greenland Ranch and Death Valley reports appear reasonable), ALL of the maximums at the closest surrounding stations lend support to the maximums for Death Valley.

From July 7 to 14 of 1913, when eight consecutive afternoons had reported maximums of 127, 128, 129, 134, 129, 130, 131 and 127F in Death Valley, NONE of the maximums from the closest surrounding stations supported the Greenland Ranch maximums! The departures from average for maximums for the hottest five-day stretch were about +4 to +8 at the closest stations, while maximums at Greenland Ranch were nearly 15 degrees F above the average for July. (see table)

Annual maximums at Greenland Ranch from 1911 to 1960 ranged from 120F to 127F, except for the 134F in 1913. If the reported maximums at Greenland Ranch in July, 1913, were authentic, then the maximums at the closest surrounding stations in that month would have been much higher than reported. In addition, numerous regional heat waves have been hotter than the one during the first half of July, 1913. Why have Death Valley maximums failed to exceed 130F in the interim when three days in July 1913 purportedly reached 134, 130 and 131F?

In his “bias” chart, Dr. Spencer notes the “substantial warm biases in the temperatures reported at Greenland Ranch in the first 10-15 years.” And, he mentions that the observer(s) may have been relying to some extent on thermometers other than the official instrumentation. I do think that the observer was comparing “household thermometer” readings with the official equipment on occasion from spring to summer of 1913. Higher readings off of the poorly-exposed thermometers near the ranch house and under the veranda were probably (and inappropriately) entered onto the official climate form. But, I have not uncovered much evidence of this particular type of deviation from standard observational procedures outside of 1913.

I would contend that the generally higher “bias” numbers from the early years comparably are due primarily to changes at the closest area weather stations which promoted cooler maximums early on and warmer maximums later. For example, two of the closest stations to Greenland Ranch in 1913 were Independence and Lone Pine, in Owens Valley. In 1913, Owens River water was diverted to Los Angeles, and the Owens Valley gradually dried up. Summer maximums increased as Owens Lake evaporated, irrigation was not possible and farmland was abandoned, and desert-like conditions developed. (Roy’s note: The early years had very few stations within 100 miles of Death Valley, and the temperature lapse rates I computed from those few stations appear to be biased as a result. I will correct this in a future blog post, and will provide what should be better estimates of average July daily maximum Death Valley temperatures.)

Also, in the early decades of the 20th century, thermometer shelters were (almost invariably) sited above grass. This resulted in very conservative (i.e., coolish) maximums at desert stations. Low humidities promoted cooling due to evapotranspiration effects. In the early decades of the 20th century, desert weather stations were generally in towns, amidst shade trees and lawns. The resulting maximum temperature reports were very conservative. By mid-century and thereafter, the town weather stations were more likely to be at the airport or at a municipal utility site, fire station or equipment yard. Grass cover and shade trees were usually absent at these locales. Today, desert weather stations in towns and cities are (almost invariably!) above bare ground.

You can imagine the difference in maximums between desert stations above oft-irrigated grass and those above bare ground. (Roy’s note: In my experience, unless the vegetation area is rather large, and there is almost no wind, a weather station’s daily maximum temperature will still be largely determined by air flowing from the larger-scale desert surroundings. But note… this is different from, say a poorly sited thermometer next to a brick wall or heat pump where hot air from an isolated source can elevate the daily maximum temperature recorded).

The Greenland Ranch station was originally sited above a patch of alfalfa grass, immediately adjacent to forty acres of cultivated and irrigated land.

It is my belief that the new observer in 1913 (Oscar Denton) was rather disillusioned with the conservative maximums from the official station above grass and next to the evaporatively-cooled farmland. I think he felt compelled to fudge the maximums upwards in 1913. Photographs of the Greenland Ranch weather station show that it was above bare ground by about 1920 (see example photo at top of post).