Published data were used to analyse the long-term annual temperature trends at thirteen locations located in Pakistan, India and Sri Lanka. Similarities in the trends for individual stations enabled a comprehensive trend to be established for the whole region.
The data used were all from the website of KNMI (the Royal Netherlands Meteorological Institute). Only raw data were used (ie as originally recorded, without any adjustments). Data were only used from stations located in rural areas or in small towns so as to avoid the possible influence of urbanisation on the temperatures.
The periods of the data ranged from 47 years to 137 years of complete annual data. The earliest record, at Port Blair, commenced in 1868.
The plots shown below are of the annual mean temperatures at each of the thirteen locations. On each plot a polynomial “best fit” trendline has been added. It should be noted that the trendlines near each extremity (ie near the start and end of the records) are quite sensitive to individual data points, whereas trendlines within the main body of the record are much more stable and reliable.
COMPARISON OF TRENDLINES
Annual Mean Temperature Trends
[ The basis of these trendlines can be seen from the analyses for the thirteen individual stations – see next section. To avoid congestion and a condensed scale the trendlines are also shown on two separate plots. ]
Annual Mean Temperature Residual Trends
[ The individual trend-lines have been adjusted by constant amounts to be approximately zero at trend troughs around 1950 (for all 13 stations) and around 1920 (for 7 selected stations. ]
ANNUAL MEAN TEMPERATURE TRENDS FOR INDIVIDUAL STATIONS
Fort Sandeman (Pakistan) 1931-2015
Dal Bandin (Pakistan) 1931-2015
Panjgur (Pakistan) 1931-2015
Dwarka (India) 1901-2015
Mukteshwar (India) 1897-2015
Daltonganj (India) 1893-2015
Cherrapunji (India) 1903-2015
Minicoy (India) 1931-2015
Pamban (India) 1891-2015
Puttalam (Sri Lanka) 1869-2015
Hambantota (Sri Lanka) 1921-2015
Port Blair (India) 1868-2015
DISCUSSION AND COMMENTS
- The data analysed was limited to that which were available on the KNMI website.
- The annual mean temperature trend-lines for 12 the 13 stations are somewhat similar. Port Blair is the exception.
- Port Blair has a well-defined cyclical pattern but with a High around 1890 and a Low around 1990.
- Well-defined cyclical patterns in the trend-lines are apparent at the 12 stations, but the magnitude of the trend-lines vary considerably.
- A Low in the trendline may have occured around 1920, but a High around 2010 is much more evident. There are no common Highs or Lows between 1920 and 2010.
- The ~2010 “High” was generally about 0.5-1.0 deg C higher that that in ~1920.
- The consistency of the annual mean temperature trend-lines of 12 of the 13 widely separated stations provides support for the good quality of the data and the conclusions drawn from the analysis.
- Port Blair is an exception with a significantly different trend-line with the Low in ~1990 about 1.0 deg C lower than the High in ~1890.
- Except for Port Blair there is evidence of temperature rises of around 0.5-1.0 deg C since 1920, but that a cyclical peak occurred around 2010.
This article was written by Brian Gunter of Narooma, NSW. In his previous life Brian was an engineering hydrologist involved over many years in the analysis of rainfall and river flow data for the planning of water resources projects in Australia, Asia and Africa. In recent years he has been one of the Marine Rescue NSW (previously Royal Volunteer Coastal Patrol) volunteer weather observers who operate the Narooma station for the Bureau of Meteorology.