Probably the most extreme example of an urban heat island (UHI) in the Southern Hemisphere is Jakarta. I first discussed it when analysing the temperature data of Indonesia for Post 31, but it is so dramatic that it needs further examination.
Jakarta is the largest city in the Southern Hemisphere with a population of over 33 million. Indonesia has a population of more than 270 million, but this is spread over an archipelago of islands that stretch over 5000 km. The result is that Indonesia has seen no warming over the last one hundred years while Jakarta has warmed by almost 3°C since 1880 (see Fig. 131.1 below).
Fig. 131.1: The change to the 5-year average temperatures of Jakarta (red curve) and Indonesia (blue curve) since 1920.
In Post 31 I examined the temperature trends for Indonesia. The mean temperature change since 1912 is shown in Fig. 131.2 below and it indicates that Indonesia outside of Jakarta has actually cooled slightly over the last one hundred years. The best fit for 1913-2012 indicates a temperature change of -0.08°C while the 5-year average suggests a small rise of about +0.16°C.
Fig. 131.2: The mean temperature change for Indonesia since 1912 relative to the 1961-1990 monthly averages. The best fit is applied to the monthly mean data from 1913 to 2012 and has a slight negative gradient of -0.08 ± 0.04 °C per century.
One of the oldest weather stations in Indonesia is Jakarta Observatorium (Berkeley Earth ID: 155660). It is located in the middle of Jakarta with almost continuous data stretching back as far as 1866, hence its significance as a case study of the urban heat island (UHI) effect. In contrast to the rest of Indonesia, Jakarta Observatorium shows significant and continuous warming since 1870 (see Fig. 131.3 below). The best fit for 1913-2012 indicates a temperature rise of more than 2.16°C in the one hundred years since 1913, while the 5-year average suggests a rise of over 2.35°C.
Fig. 131.3: The mean temperature change for Jakarta Observatorium since 1866 relative to its 1961-1990 monthly averages. The best fit is applied to the monthly mean data from 1913 to 2012 and has a positive gradient of +2.16 ± 0.08 °C per century.
It is important to note that while Jakarta Observatorium is the clearest example of a UHI in Indonesia, it is not the only one. Up until 1970 there was a second station in Jakarta (Berkeley Earth ID: 155660) which also exhibited over 1.8°C of warming from 1866 to 1970. But the city of Surabaya (Berkeley Earth ID: 155652) also appears to behave as a UHI. Its population is over twelve million making it the fifth largest city in the Southern Hemisphere. From 1949 to 2013 it appears to have exhibited warming of more than 1.7°C as well (or 2.75°C per century). Yet despite this the rest of Indonesia cooled.
Summary
The following temperature changes were observed from 1913 to 2012.
Indonesia: 0.16°C (trend -0.08°C).
Jakarta: 2.35°C (trend 2.16°C).
So Jakarta has warmed by at least 2°C more than the rest of Indonesia. It has also warmed while Indonesia has not. A classic UHI!
Hi, first at all, congratulations for your amazing job. Second, at the end of this six UHI articles, i end up with a big uncertanty you dont clear up at the end of each one, and is how much would be the impact of this UHI locations for the entire country mean temp change. In other words, and for the sake of the example, if this 2.16dC trend in Jakarta has an impact (and how much) on the -0.08dC Indonesia trend. Keep up this amazing work. Cheers.
ReplyDeleteThe impact of the UHI on the average for the entire country depends on the number of other temperature records. The more there are the less the impact. If there are more than 30 then the impact is small.
DeleteIn the case of Indonesia, as I showed in Post 31, there are less than 10 other temperature records before 1950 so the impact before 1950 is large. That is why the trend for Indonesia in Fig. 131.2 above does not include the Jakarta data.
The impact of the UHI on the average temperature for the entire country depends on the averaging method used as well. Ideally each station should be weighted by the area it represents, but an urban station should really only be weighted by the area of the urban region it represents. So as Jakarta is only 0.03% of the area of Indonesia (660 km^2 out of 1.9m km^2) its weighting should be small (0.03%). But as there are less than about 50 stations in Indonesia the typical weighting of each station will be 2%. So the weighting of UHI records could be over 60 times bigger than their true impact.
Delete