Does Station Siting Have a Significant Impact on CONUS Temperatures?

I frequently see challenges to the US temperature record that claim that many stations are affected by poor station sighting, and that biases introduced by poor station siting may account for a significant fraction of observed warming in the US. The claim is that a significant number of stations located near asphalt or at airports or too close to buildings can cause thermometers to record artificially high temperatures, adding spurious warming trends to CONUS temperature trends.

With the USCRN project, NOAA developed a classification system based on exposure characteristics affecting the siting of stations, and this classification system was retroactively applied to the stations in the old USHCN network by Anthony Watts and surfacestations.org. In this classification system, ratings of 1 and 2 indicate "good siting" and ratings of 3, 4, and 5 indicate "poor siting." These classifications can actually be used to test whether station siting has any impact on CONUS temperature trends, and that's what was done in Menne et al 2010.[1] At the time of publication, 525 stations had been classified, and of them 71 were rated good (1 or 2), while 454 were rated poor. However, the sites with good exposure ratings were well-distributed across CONUS. The study compared both raw and homogenized data from stations with good siting and poor siting to see to what extent the stations differed, and to what extent homogenization corrected differences. It turns out that poor siting had only a small effect on Tmax and Tmin, and with homogenization those differences were corrected in favor of the well-sited stations.


The effectiveness is more clearly seen by comparing trends in Poor-Good sites. The adjusted Tmax and Tmin plots of Poor-Good stations show very little trend remaining.

At the time of publication, NOAA had also just started using the USCRN network, which is an all rural network of 114 well-sited stations. So Menne et al plotted both Tmax and Tmin for USCHN v2 with "all stations" stations with "good siting" and stations with "poor siting" along with about 5 years of data from the USCRN pristine dataset. The results look like this.
If the plots of good vs poor siting were not labeled, you wouldn't be able to tell which was which. The differences between them are minimal, and the trend for the poorly-sited stations is not higher than teh trend for all stations. This seems conclusive evidence that station siting is having very little effect on CONUS temperature trends. The authors conclude, "the adjusted USHCN temperatures are extremely well aligned with recent measurements from instruments whose exposure characteristics meet the highest standards for climate monitoring. In summary, we find no evidence that the CONUS average temperature trends are inflated due to poor station siting."
We can take this a step further. The USHCN network ha been defunct since 2014; it was replaced by nClimDiv, a homogenized dataset that includes 10,000 stations in the GHCN-daily network, and the USHCN stations are a subset of GHCN-daily. And now in 2025, the all-rural, well-sited USCRN network has 20 years under its belt. If poor siting is causing spurious warming to CONUS trends, this would show up in nClimDiv while being absent from USHCN. So if we compare the two, we can get either definitive confirmation or falsification of the claim that poorly sited stations are adding spurious trends to CONUS temperatures. As you can see above, nClimDiv is not warming more rapidly than USCRN. If anything, USCRN is warming more rapidly.
In fact, if we compare trends for Jan 2005 to May 2025 in USCRN, nClimDiv, the ERA5 reanalysis and UAH satellite TLT data for CONUS, USCRN is warming marginally more rapidly than all the others. Of the four, only nClimDiv could be affected by poor station siting, yet its trend ranks third, after both USCRN and ERA5 in terms of warming rate. There is simply no case to be made that poor station siting is having a significant effect on CONUS temperature trends.


References:

[1] Menne, M. J., C. N. Williams Jr., and M. A. Palecki (2010), On the reliability of the U.S. surface temperature record, J. Geophys. Res., 115, D11108, doi:10.1029/2009JD013094.
https://www.ncei.noaa.gov/pub/data/ushcn/papers/menne-etal2010.pdf


Comments

Post a Comment

Popular posts from this blog

Data Tampering by Shewchuk and Heller

Image
If you follow climate discussions on X, you're bound to see John Shewchuk and/or Tony Heller show graphs that reportedly show that NOAA is tampering with temperature data to fabricate global warming with spurious warming trends. I've gone over many of the reasons why this is nonsense before in posts about bias correction and so-called  ghost stations . I think it's good to show what's actually going on with the graphs they present as "proof" of data manipulation, though. I think it can be easily demonstrated here that it's actually Shewchuk and Heller that are tampering with data. Shewchuk (Top) and Correct (Below) Above are two graphs. The top graph shows what John Shewchuk claims shows that NOAA is manipulating data. It shows USHCN "raw" and "altered" Tmax data for 1900 to 2023. The bottom graph above is the correct plot of NOAA's published data from the current and correct dataset (nClimDiv) with a 5-year running mean to match S...
Read more

Was There a "Mike's Nature Trick" to "Hide the Decline?" Part 1 - Misreading CRU Emails

Image
In November 2009, someone likely stole[1] and publicly released over 1,000 emails from Climate Research Unit (CRU) at the University of East Anglia. Following this, people combed through all these emails  looking for evidence of frauds and coverups on the part of climate scientists. They found very little damaging information, but not everything they found was was good. There were, as best I can tell, four categories of content that were discovered: The normal kinds of communication you'd expect to find in correspondence between scientists Words and phrases that were twisted out of context and redefined to support an imagined conspiracy that had been imposed on the emails. Evidence of mean-spirited attacks on contrarian scientists and heavy-handed tactics intended to punish journals that published contrarian studies.  A single email describing the process of creating a graph for a WMO statement that used the phrase "hide the decline."  Here I plan to tackle the "hid...
Read more

Debunking the Latest CO2 "Saturation" Paper

Image
Update (12/29/2024):  The paper I discuss below was retracted a couple days ago, citing failures in the peer-review process. "Subsequent to acceptance of this paper, the rigor and quality of the peer-review process for this paper was investigated and confirmed to fall beneath the high standards expected by Applications in Engineering Science . After review by additional expert referees, the Editor-in-Chief has lost confidence in the validity of the paper and has decided to retract." A paper published earlier this year is the latest in the long history of attempts to show that CO2 is already "saturated" in the atmosphere, and therefore increased CO2 cannot cause any more warming. The latest in a series of these kinds of papers from Kubicki et al 2024[1] attempts to make this point by modeling the atmosphere in a couple different experiments. The result of their experiments suggested to them that we should question whether "additionally emitted CO2 in the atmosp...
Read more