This is a nice way to try and show the point about records being broken. But would you mind explaining how you deal with more weather stations being bought online over time? I guess that doesn't make any difference as long as the grid size is consistent over that time? (The only difference is that measurements for the grid will be more accurate).
And, following on from that, can you therefore also go further back in time then 1980? I'm asking because someone showed me a blog post where they were looking at tmax from 1900's onwards and claiming that there's no great records being set for this July 2023 and i highly suspect their methods are totally flawed. (They're basically just looking at tmax for stations which existed from 1900 onwards.)
I'm using the gridded Berkeley Earth data set here, so I'm letting them deal with it. It might be an issue if you go back in time far enough, but since the 1980s we've had good coverage in the northern hemisphere mid-latitudes. Certainly we can go back further, but then you run into the time series not being very long and getting lots of records due to that — which makes it hard to pick out a global warming signal.
"fraction" in the plots is the (approx.) area fraction. so a fraction of 0.1 means that 10% of the region is experiencing a record temperature during a year.
yes, that's a valid hypothesis. and it can be tested by looking at other latitudes and they all show the same increase in temperature records (those plots are not in the post). it's happening basically everywhere. so we can falsify this hypothesis.
You would like my latest. It compliments yours.
https://open.substack.com/pub/richardcrim/p/the-crisis-report-39?r=a66g&utm_campaign=post&utm_medium=web
This is a nice way to try and show the point about records being broken. But would you mind explaining how you deal with more weather stations being bought online over time? I guess that doesn't make any difference as long as the grid size is consistent over that time? (The only difference is that measurements for the grid will be more accurate).
And, following on from that, can you therefore also go further back in time then 1980? I'm asking because someone showed me a blog post where they were looking at tmax from 1900's onwards and claiming that there's no great records being set for this July 2023 and i highly suspect their methods are totally flawed. (They're basically just looking at tmax for stations which existed from 1900 onwards.)
I'm using the gridded Berkeley Earth data set here, so I'm letting them deal with it. It might be an issue if you go back in time far enough, but since the 1980s we've had good coverage in the northern hemisphere mid-latitudes. Certainly we can go back further, but then you run into the time series not being very long and getting lots of records due to that — which makes it hard to pick out a global warming signal.
"fraction" in the plots is the (approx.) area fraction. so a fraction of 0.1 means that 10% of the region is experiencing a record temperature during a year.
the warming of the climate system is global, so shifts in the jet stream or anything like can't explain the observations.
yes, that's a valid hypothesis. and it can be tested by looking at other latitudes and they all show the same increase in temperature records (those plots are not in the post). it's happening basically everywhere. so we can falsify this hypothesis.