It is difficult to vouch for the quality of these records. The lighthouses at Gabo Island, Wilson’s Promontory and Cape Ottway are isolated locations where there is no danger of urban heat island distortion. Other records, such as that from Maryborough, have been made at stations that once may have been located on the edge of a large town but are now engulfed by it. Still others, such as Ballarat Aerodrome appear from satellite images to be in open spaces, but still on the margins of a rural city.
It is also difficult to tell from the records or the site information provided by the BOM whether the records for each station are from a single location or a moveable location. Some—such as Ballarat Aerodrome—seem likely to represent a series cobbled together from different locations over the course of a century. Others, such as Gabo Island Lighthouse or Wilson’s Promontory Lighthouse could easily be from a single location.
The full list of stations, in the row-order of the graphic, is
- Ballarat Aerodrome
- Cape Ottway Lighthouse
- Deniliquin (NSW)
- Gabo Island Lighthouse
- Wilson’s Promontory Lighthouse
I used the following simple R script to graph the data from CSV files available here
# data from http://www.bom.gov.au/climate/data/weather-data.shtml
df - read.table(file, header=TRUE, sep=",", na.strings="null")
df - df[,3:4]
p-qplot(Year, Jan, data=df, main="Montly mean maximum temperature: Ballarat Aerodrome, Vic")
p + geom_point() + geom_smooth(method = "lm")