From the INSTITUTE OF ATMOSPHERIC PHYSICS, CHINESE ACADEMY OF SCIENCES
How much warmer has Hong Kong’s urban area become during the past 4 decades?
Characterizing the urban temperature trend using seasonal unit root analysis: Hong Kong from 1970 to 2015
Scientists from Macao Polytechnic Institute are pioneers in exploring urban temperature in Hong Kong using seasonal econometric models. In particular, the characterization of the urban temperature trend was investigated using a seasonal unit root analysis of monthly mean air temperature data over the period of January 1970 to December 2013.
“The seasonal unit root test makes it possible to determine the stochastic trend of monthly temperatures using an autoregressive model,” says Prof. Wai Ming To. “We found that Hong Kong’s urban mean air temperature has increased by 0.169°C per 10 years over the past four decades using monthly temperature data, or 0.174°C per 10 years using annual temperature data, and the trend is likely to persist.”
The increase in Hong Kong’s urban mean air temperature was higher than the increase in global mean air temperature [0.13°C (10 yr)-1 using data from 1956 to 2005, or 0.07°C (10 yr)-1 between 1906 and 2005 due to global warming]. The model of monthly temperature obtained from the seasonal unit root analysis was able to explain 95.9%of the variance in the measured monthly data–much higher than the variance explained by the ordinary least-squares model using annual mean air temperature data and other studies alike. The model accurately predicted monthly mean air temperatures between January 2014 and December 2015, with a root-mean-square percentage error of 4.2%.
By analyzing the monthly air temperatures recorded at an urban site and a rural site, it was found that the urban heat island effect led to the urban site being on average 0.865°C warmer than the rural site over the past two decades. Besides, it was shown that the increase in annual mean air temperature was significantly associated with the increase in population, gross domestic product, urban land use, and energy use.
###
The paper: http://link.springer.com/article/10.1007%2Fs00376-016-6113-z
Characterizing the urban temperature trend using seasonal unit root analysis: Hong Kong from 1970 to 2015
Abstract
This paper explores urban temperature in Hong Kong using long-term time series. In particular, the characterization of the urban temperature trend was investigated using the seasonal unit root analysis of monthly mean air temperature data over the period January 1970 to December 2013. The seasonal unit root test makes it possible to determine the stochastic trend of monthly temperatures using an autoregressive model. The test results showed that mean air temperature has increased by 0.169◦C (10 yr)−1 over the past four decades. The model of monthly temperature obtained from the seasonal unit root analysis was able to explain 95.9% of the variance in the measured monthly data — much higher than the variance explained by the ordinary least-squares model using annual mean air temperature data and other studies alike. The model accurately predicted monthly mean air temperatures between January 2014 and December 2015 with a root-mean-square percentage error of 4.2%. The correlation between the predicted and the measured monthly mean air temperatures was 0.989. By analyzing the monthly air temperatures recorded at an urban site and a rural site, it was found that the urban heat island effect led to the urban site being on average 0.865◦C warmer than the rural site over the past two decades. Besides, the results of correlation analysis showed that the increase in annual mean air temperature was significantly associated with the increase in population, gross domestic product, urban land use, and energy use, with the R2 values ranging from 0.37 to 0.43.
Discover more from Watts Up With That?
Subscribe to get the latest posts sent to your email.
Re UHI in Hong Kong accounts for most ‘warming’ since 1970.
The short form of the paper is that the city is warmer than the sticks. Wasn’t it always cooler in the country where grandma lives? She didn’t have, or for that matter need, a cooler. Trees just don’t get as hot as asphalt.
Today’s journal climatology is like a Christmas Tree, filled with ornaments, which have zero functional utility. UHI is one of them. (Others include upper atmosphere parameters, acidification, chaos, El Niño, hurricanes, land use, and aplenty more.) The reason for the irrelevance of the UHI phenomenon, of course, is that UHI measures the heat content where the heat capacity is negligible for anything that might predict global climate.
So I wondered, did the source article confuse UHI effects with global climate? The post didn’t answer the question, though the scare quotes around warming in the title seemed to promise something along those lines.
To & Wu’s abstract compares the Hong Kong temperature to the monthly mean air temperature data over [a 14 year] period. Temperature where? Global? They found a weak correlation of Hong Kong temperature with four parameters, the increase in population, gross domestic product, urban land use, and energy use. Parameters where? Just in Hong Kong? Seems unlikely, So I clicked on the link to the paper, only to find that it was for science for sale at $39.95. Bah, humbug. Is this blog just for the idle rich and academics? Seems improbable.
On the positive side, the authors are courageous to admit in the abstract that the Coefficient of Determination, R-squared, was in the range of 37% to 43%. That is the ratio of the combined effect of the four parameters cited that accounts for the observed warming attributable to UHI. That quantifies that most of the statistical power in the Hong Kong temperature is not due to those parameters. An R-squared no better than reported may be valuable, as in invalidating someone’s conjecture. More likely, though, it reports a failed experiment, one that verges on junk because the regression model has so little predictive power.
You can download the PDF file of the paper from here for free. Just scroll down to the article.
http://159.226.119.58/aas/EN/0256-1530/current.shtml
I agree R-squared of 37 to 43% is not sufficient to explain anything.
Jeff Glassman: “To & Wu’s abstract compares the Hong Kong temperature to the monthly mean air temperature data over [a 14 year] period.”
The Abstract: “This paper explores urban temperature in Hong Kong using long-term time series. In particular, the characterization of the urban temperature trend was investigated using the seasonal unit root analysis of monthly mean air temperature data over the period January 1970 to December 2013”
WR: January 1970 to December 2013 = 44 years. Typo?
Wim Röst, 112-/16 @ur momisugly 4:15 pm
Thanks for the catch.
“Besides, the correlation between annual
mean air temperature and urbanized area was consistent with
the findings reported by He et al. (2013), who indicated that
the total urban area of a city was associated with the annual
mean air temperature recorded in the city’s meteorological
stations.
To identify the relationships between population, GDP,
urban land use, energy use, and annual mean air temperature,
multiple-linear regression was applied to these five time series
(Ning and Bradley, 2014). Population, GDP, urban land
use, and energy use were chosen as the independent variables,
while annual mean air temperature was selected as the dependent
variable. A stepwise procedure was used to identify significant
predictors of annual mean air temperature. Results
showed that energy use and GDP were able to explain 48.1%
of the variance in annual mean air temperature.”
Wai-Ming TO and Tat-Wai YU, I wonder if you are welcome to speak in the next climate conference.
GDP is Gross Domestic Product. “Hong Kong’s GDP increased from HKD 195.2
trillion to HKD 2246.4 trillion during the same period of time.”
Austin TX, too: https://politicsandprosperity.com/2016/09/03/agw-in-austin-ii/
I need more$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$ to continue my work
You’ll have to excuse the fact that my postings here are usually of fairly basic…..but still very important…..matters. In this case, any time I see temperatures or anomalies to three places after the decimal point I get very suspicious since I was taught long ago that one can’t have more places of decimals than the number of the lowest precision that you are working with. So, to say that one sees
“0.169°C per 10 years” does not seem correct to me given that thermometers are rarely readable past one place after the decimal point.
Ian M
One place that the rule you mention can safely be ignored is when calculating trends over time.
For example, if you have a thermometer that is accurate to 0.1C. Now take measurements 10 years apart, the one is 10.1C, the other is 10.2C. Both measurements are valid based on the accuracy of the thermometers.
It is quite legitimate to state that you have a trend of 0.1C per decade or 0.01C per year. Even though the trend per year is less than the accuracy of the thermometers.
From the mere mathematical point of view that is correct, but it is also practical significance we should be concerned about.
Has the calibration of your thermometer drifted in 10 years? The values 10.1 and 10.2 are single measurements, or averages?
Is a variation of 0.1 C at all significant for the system being investigated?
Mark
That’s the beauty of stats, you can manipulate data to tell any story you want.
It maybe –
“quite legitimate to state that you have a trend of 0.1C per decade or 0.01C per year. Even though the trend per year is less than the accuracy of the thermometers”
But it doesn’t show the true picture of what’s occurring
is it a gradual rise over 10 yrs,
a single step rise (where & when ?),
or have the readings been taken on a segment of a sine wave & the true trend is down ?? You just don’t know, there are not enough data points.
It maybe legitimate stats….but it’s junk science.
The use of a single regression line for the given data set is questionable. The period of analysis includes the 24-year interval 1975 to 1998, during which all global temperature time series shot up by around 0.8 oC. This is well-represented in the graph, but the data earlier than 1975 doesn’t show an upward trend at all, and neither does the data newer than 1998. Three different regression lines would be more appropriate to the data set, the pre-1975 and post-1998 lines being flat and the 1975 to 1998 line steeply ascending to the right.
Still waiting for Anthony to put up his UHI paper for peer review like he promised. How long ago was that? Like 3-4 years ago? I’ve lost track it since has been so long ago.