Historical Records of Cloudiness and Sunshine in Australia

P. A. Jones Macquarie University, New South Wales, Australia

Search for other papers by P. A. Jones in
Current site
Google Scholar
PubMed
Close
and
A. Henderson-Sellers Macquarie University, New South Wales, Australia

Search for other papers by A. Henderson-Sellers in
Current site
Google Scholar
PubMed
Close
Full access

Abstract

Historical records of mean monthly cloud amount over Australia have been studied to determine whether there is any long-term trend. Of 318 stations with more than 30 years of data, 252 show an increase and 66 a decrease. The cloud amount shows a rise of 5% between 1910 and 1989, when averaged over all stations. The trend is not uniform, however, with a slight fall in cloud between 1910 and 1930 and with most of the rise between 1930 and 1980. Sunshine records were used to check the cloud record for systematic errors. Monthly average cloud and sunshine fractions are correlated with coefficient r=−0.87 and with best-fit slope −1.00. The sum of cloud and sunshine fractions is around 1.2, whereas it may be expected that the sun should be 1.0 if the cloud and sunshine fractions are complementary. The sunshine and cloud variations are in close agreement for the period 1950 to 1989. The subset of stations that have sunshine records shows no overall change in cloudiness or sunshine over this period, with 31 stations showing an increase in cloud and 28 a decrease. An independent dataset of 41 stations, mostly airports, shows no significant trend over the period from 1940 to 1988, with 24 stations showing a decrease in cloud and only 17 showing an increase over this period. It is suggested that there is an overall long-term increase in total cloud amount over Australia, but that it does not occur uniformly for all stations, so that some groups of stations show no increase. However, the overall trend must remain tentative until the reason for the differences between the datasets is clarified.

Abstract

Historical records of mean monthly cloud amount over Australia have been studied to determine whether there is any long-term trend. Of 318 stations with more than 30 years of data, 252 show an increase and 66 a decrease. The cloud amount shows a rise of 5% between 1910 and 1989, when averaged over all stations. The trend is not uniform, however, with a slight fall in cloud between 1910 and 1930 and with most of the rise between 1930 and 1980. Sunshine records were used to check the cloud record for systematic errors. Monthly average cloud and sunshine fractions are correlated with coefficient r=−0.87 and with best-fit slope −1.00. The sum of cloud and sunshine fractions is around 1.2, whereas it may be expected that the sun should be 1.0 if the cloud and sunshine fractions are complementary. The sunshine and cloud variations are in close agreement for the period 1950 to 1989. The subset of stations that have sunshine records shows no overall change in cloudiness or sunshine over this period, with 31 stations showing an increase in cloud and 28 a decrease. An independent dataset of 41 stations, mostly airports, shows no significant trend over the period from 1940 to 1988, with 24 stations showing a decrease in cloud and only 17 showing an increase over this period. It is suggested that there is an overall long-term increase in total cloud amount over Australia, but that it does not occur uniformly for all stations, so that some groups of stations show no increase. However, the overall trend must remain tentative until the reason for the differences between the datasets is clarified.

Save