The Times groups fires by the calendar year, obscuring an important pattern: fire season in Australia starts in June or July.¹ The Post’s chart nicely adjusts for this by zooming into fire season and avoiding the awkward two highlighted lines that mid-January updates to the Times’ version added.
Romain Vuillemot shifted the Times’ chart by six months to account for this. I was curious how smaller changes to the start date for the year would change the shape of the lines; try mousing over my version above to see them move.
The Times only includes fires in New South Wales² while the Post has all of Australia but only the last seven years of data. The Times backs up their selection of New South Wales with maps showing how close the fires got to cities. The Post seems to have only gone back to 2013 because there were significantly more fires across all of Australia in 2012 than in 2019.
I’m not sure if there’s a bright light between cherrypicking date ranges and focusing on impacted regions, but the Post’s arbitrary cutoff makes me a little uncomfortable. While every chart can’t include every data point, getting this right is especially important with unusual weather. Climate change denialists have spent the last decade complaining about a single graph.
¹ I wasn’t able to find an good reference for the start of fire season; this does make it a little harder to justify switching from the calendar year. And with only 2019 data, calendar years are easier to label. The trend is so dramatic that the chart doesn’t change much.
² Mapshaper makes this surprisingly easy! I’m not sure I got it completely right through; the Post has significantly more fires than I counted.