It’s been going up significantly over time—the trend-line goes from about 500k acres in 2000 to 1.5M acres in 2020, making me doubt a regression to the mean. Even excluding 2020, the trend-line goes from less than 500k acres in 2000 to about 1.1M acres in 2019. I’m expecting more years like this one in the future, although hopefully not quite as bad.
It’s worth noting that the R-squared value for a linear trend-line for 2000-2019 data has R2=0.07 so a constant prediction of 750k acres would only be marginally less accurate. (I think your excluding 2020 graph also excludes 2019 but the story doesn’t change much either way)
It looks like up until 2016 everything was fairly constant and since then 3 out of 4 years have been bad.
Looking at the acres of forest burned over the last twenty years:
It’s been going up significantly over time—the trend-line goes from about 500k acres in 2000 to 1.5M acres in 2020, making me doubt a regression to the mean. Even excluding 2020, the trend-line goes from less than 500k acres in 2000 to about 1.1M acres in 2019. I’m expecting more years like this one in the future, although hopefully not quite as bad.
(data from Wikipedia)
It’s worth noting that the R-squared value for a linear trend-line for 2000-2019 data has R2=0.07 so a constant prediction of 750k acres would only be marginally less accurate. (I think your excluding 2020 graph also excludes 2019 but the story doesn’t change much either way)
It looks like up until 2016 everything was fairly constant and since then 3 out of 4 years have been bad.
Hm, when I was making the excluding-2020 graph I was intending to include 2019 as well, but it might have been taken out accidentally.