UHI has influenced China climate record, according to new research by Nicola Scafettaa & Shenghui Ouyang.
Scafettaa is a well known climate sceptic, and he's a theoretical physicist. Hardly an expert. The paper is published in Global and Planetary Change, which is an Elsevier journal. I'm no fan of Elsevier, it epitomises everything wrong with academic publishing: take publicly-funded research, publish it behind a paywall, and then shake-down the institutions responsible for the research by selling back access to the research they produced for as much as you can possibly get. The only industry in the world similar to this is the pharmaceutical industry - they too shake-down governments for everything they can get, which is why some countries literally speed 1,000x the amount for the same pill that another country pays.
However, this paper produces a compelling case that can't be ignored. China is the perfect location to look for a bias in the temperature record due to the urban heat island (UHI) effect. In 1950 88% of the population lived in rural areas, by 2015 that was down to 45% (55% living in urban environments). In the period 1950-2010 mean surface temperatures in China rose by 0.8 °C. The hypothesis presented seems robust, it goes like this. UHI is a localised phenomenon where large urban landscape objects like concrete and asphalt absorb heat during the day, and release it slowly at night. A bias in the temperature record can be introduced if a rural environment is converted into an urban environment. Because the effects of UHI happen at night, Tmax would be affected less and provide a better indicator than Tmin for changes in surface temperature due to climatic forces rather than non-climatic forces like UHI. If this hypothesis is correct it would also mean a narrowing of the range between Tmin and Tmax in the climate data. The fingerprint of this difference would also be visible against the computer climate models, or to put it crudely the data would over-represent the cities as hotplates.
The results are pretty damning: "Somehow the combined might and supercomputers at NOAA, NASA, Hadley and the Bureau of Met experts all missed this."
![[Image: uhi-china-scafetta-2019-fig6.jpg]](https://s3.amazonaws.com/jo.nova/guest/eu/scafetta/uhi-china-scafetta-2019-fig6.jpg)
Fig. 6. Map of the divergence (ΔTMin − ΔTMax) between the warmings registered by the minimum and maximum temperature records (CRU TS4) between 1945 and 1954 and 2005–2014. The cyan dots indicate the 200 most populated cities in China according to the Free World City Database. (White regions over India and the ocean indicate missing data).
![[Image: cUNtUAD.jpg]](https://i.imgur.com/cUNtUAD.jpg)
Fig. 15. CRU TS4.01 Tmax (red) and Tmin (blue) records for each month over China smoothed with an acsplines filter: See Table 6. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Compare to what was predicted in the climate models:
It's difficult to put into words just what a monumental clusterfuck this represents. It means that we can confirm there was a systemic problem in the climate record data for China. Whoopsie. UHI should not be introducing any bias, indeed virtually all climate scientists will tell you that - even the really sceptical ones. Why not? Because all the urbanised data is carefully corrected to remove the UHI so we can separate that warming from the atmospheric warming. And it's not like they didn't have fair warning, Soon et al. 2015 & 2019 said the Chinese climate data looked like it suffered from urbanisation bias. The paper's results are thoroughly robust - they've taken into account that Tmin was projected to rise 0.2°C more than Tmax by the climate models (see figure above). Between 1950-2010 the increase in mean surface temperature in China "should be 0.46 ± 0.13°C, which is 42% less than the reported 0.79 ± 0.10°C".
Can we generalise these results? Of course not! Indeed we mus'n't. However if you wanted to look for other areas where the climate data might be biased by UHI, this paper has provided one way in which to detect it.
Scafettaa is a well known climate sceptic, and he's a theoretical physicist. Hardly an expert. The paper is published in Global and Planetary Change, which is an Elsevier journal. I'm no fan of Elsevier, it epitomises everything wrong with academic publishing: take publicly-funded research, publish it behind a paywall, and then shake-down the institutions responsible for the research by selling back access to the research they produced for as much as you can possibly get. The only industry in the world similar to this is the pharmaceutical industry - they too shake-down governments for everything they can get, which is why some countries literally speed 1,000x the amount for the same pill that another country pays.
However, this paper produces a compelling case that can't be ignored. China is the perfect location to look for a bias in the temperature record due to the urban heat island (UHI) effect. In 1950 88% of the population lived in rural areas, by 2015 that was down to 45% (55% living in urban environments). In the period 1950-2010 mean surface temperatures in China rose by 0.8 °C. The hypothesis presented seems robust, it goes like this. UHI is a localised phenomenon where large urban landscape objects like concrete and asphalt absorb heat during the day, and release it slowly at night. A bias in the temperature record can be introduced if a rural environment is converted into an urban environment. Because the effects of UHI happen at night, Tmax would be affected less and provide a better indicator than Tmin for changes in surface temperature due to climatic forces rather than non-climatic forces like UHI. If this hypothesis is correct it would also mean a narrowing of the range between Tmin and Tmax in the climate data. The fingerprint of this difference would also be visible against the computer climate models, or to put it crudely the data would over-represent the cities as hotplates.
The results are pretty damning: "Somehow the combined might and supercomputers at NOAA, NASA, Hadley and the Bureau of Met experts all missed this."
Fig. 6. Map of the divergence (ΔTMin − ΔTMax) between the warmings registered by the minimum and maximum temperature records (CRU TS4) between 1945 and 1954 and 2005–2014. The cyan dots indicate the 200 most populated cities in China according to the Free World City Database. (White regions over India and the ocean indicate missing data).
Fig. 15. CRU TS4.01 Tmax (red) and Tmin (blue) records for each month over China smoothed with an acsplines filter: See Table 6. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Compare to what was predicted in the climate models:
Spoiler:
![[Image: JRxOzL5.jpg]](https://i.imgur.com/JRxOzL5.jpg)
It's difficult to put into words just what a monumental clusterfuck this represents. It means that we can confirm there was a systemic problem in the climate record data for China. Whoopsie. UHI should not be introducing any bias, indeed virtually all climate scientists will tell you that - even the really sceptical ones. Why not? Because all the urbanised data is carefully corrected to remove the UHI so we can separate that warming from the atmospheric warming. And it's not like they didn't have fair warning, Soon et al. 2015 & 2019 said the Chinese climate data looked like it suffered from urbanisation bias. The paper's results are thoroughly robust - they've taken into account that Tmin was projected to rise 0.2°C more than Tmax by the climate models (see figure above). Between 1950-2010 the increase in mean surface temperature in China "should be 0.46 ± 0.13°C, which is 42% less than the reported 0.79 ± 0.10°C".
Can we generalise these results? Of course not! Indeed we mus'n't. However if you wanted to look for other areas where the climate data might be biased by UHI, this paper has provided one way in which to detect it.
