Late night 🧵 on how to think about extreme event attribution and whether the TX cold event was caused by climate change.

I think of extreme event attribution as resting on three pillars. 1/
Pillar 1: If you have observations of the climate over a long enough period, the data can be statistically analyzed to determine the likelihood that an observed extreme event occurring today could have occurred prior to human-induced warming. 2/
But even if the observations are good enough for that sort of analysis, they usually can’t tell you whether an observed trend was caused by global warming or by something else because correlation does prove causality. 

That brings us to ... 3/
Pillar 2: Our understanding of the physics of the phenomenon.  It should be obvious why, in a warmer world, we expect to get more frequent heat waves.  This physical understanding adds to our confidence that climate change is a factor in the occurrence of heat waves.  4/
Pillar 3: Climate models. We can run climate models with and without global warming and if we see that events like the one being studied occur in the warmer world but not in the colder one, then we have another piece of evidence. 5/
If you've got all three pillars (e.g., 2003 European heat wave), we can be very confident that climate change made the event worse.  6/
If you have zero pillars (e.g., tornadoes), you can’t confidently conclude anything — at least for now. The evidence for most types of extreme events is somewhere in between. 7/
For last week's TX cold event, we have maybe one pillar. There is a plausible mechanism concerning the equator-pole temperature gradient.

But we don't have observational evidence that this event is occurring more frequently today than prior to global warming. 8/
I also don't know of any computer simulations showing that these TX cold extremes become more frequent as the climate warms. [if someone's done the analysis, let me know] 9/
Given this, I am skeptical of the connection between the TX cold event and global warming. However, there are people I respect who disagree with me — that's science. As more work on this is done, consensus will emerge on way or the other. 10/10
I wonder what @MCHammer thinks about this. 11/11

• • •

Missing some Tweet in this thread? You can try to force a refresh

Keep Current with Andrew Dessler

Andrew Dessler Profile picture

Stay in touch and get notified when new unrolls are available from this author!

Read all threads

This Thread may be Removed Anytime!


Twitter may remove this content at anytime! Save it as PDF for later use!

Try unrolling a thread yourself!

how to unroll video
  1. Follow @ThreadReaderApp to mention us!

  2. From a Twitter thread mention us with a keyword "unroll"
@threadreaderapp unroll

Practice here first or read more on our help page!

More from @AndrewDessler

9 Jan
Free-market dudes don't understand the free market. As Milton Friedman said, corporations are in the business of making money. If having someone (Trump) on the platform costs it money, they have the right — nay, the responsibility — to kick him off to maximize profits.
Wow, another person who needs to read more Adam Smith. In a free market, transactions are voluntary. If Twitter doesn't want to transact with Trump, they don't have to. Nikki Haley can move to Cuba if she doesn't like free markets.
Read 6 tweets
8 Jan
Quick reminder: Elsevier is trash. Do not publish in their journals. I am going to avoid citing papers in those journals & I hope other researchers do this, too. That's the only thing that will stop them from bankrupting the academic community.

From @TAMU libraries:
In geosciences, there's no need to ever publish in Elsevier. The @theAGU @ametsoc and the @EuroGeosciences have all the journals we need. Elsevier journals are nothing special and can easily be avoided.
I know, I know. In other fields it's harder to avoid Elsevier. Obviously, everyone's situation is different. And the decision about which journal to publish in, like the decision to fly to a meeting, can be difficult.
Read 4 tweets
4 Jan
I'm a co-author on a new paper about committed warming! @mzelinka is a co-author, so you know it's good!…
Here is an explainer video. I'm using the fancy video set-up I use for teaching.

"Committed warming" is a standard calculation that asks how much warming you would get if you held the atmospheric composition fixed at today's values indefinitely.

This figure from the IPCC AR4 report shows about 0.6°C of committed warming in 2100 (the yellow line).
Read 11 tweets
23 Dec 20
There is no result that disqualifies climate economics more than this one. 8°C warming = loss of ~10% of GDP???? As a reminder, the last ice age was about 6°C cooler than today. Now imagine changes as big as those occurring over the next century or two. This result is absurd.
This plot shows GDP curves (growing at 2.5% and 2.46%) that differ by 4% in 2100. This is analogous to Lomborg's worlds with no global warming and a world that warms by 8°C in 2100! Image
8°C is getting close to temperatures during the Eocene, when there was no permanent ice on the planet.

NH land will warm more than the global average, so 8°C global avg. probably corresponds to something like 12°C (22°F) over NH land (and even more in the Arctic).
Read 5 tweets
27 Aug 20
If you hear a "reasonable middle bro" who "cares deeply about the environment" but tells you that "climate change is not making hurricanes worse," they're gas-lighting you.
Let's examine the peer-reviewed science on this issue. From chapter 6 of the IPCC Ocean and Cryosphere report.
See also this web page from NOAA:…
Read 17 tweets
19 Aug 20
REPOST: I've been looking for a good figure showing Extreme temperatures have gone up. After searching without success, I made this, adapted from @DrJamesEHansen's 2012 PNAS paper. Data are from Berkeley earth.
[previous post had an error in the plot; corrected here] Image
The left plot shows the 1951-1980 June-July-August seasonal average for northern hemisphere land (30°N-60°N). The blue area is the coolest is 33% of temperatures, the white area or the middle 33%, and the red or the highest 33%.
The right plot shows the 2011-2020 temperatures. Over the decade, 88% of the seasonal averages would have been in the top third in the 1951-1980 period. Only 1.2% would have been in the coolest third. And 14% are warmer than ANY temperatures in the 1951-1980 period.
Read 10 tweets

Did Thread Reader help you today?

Support us! We are indie developers!

This site is made by just two indie developers on a laptop doing marketing, support and development! Read more about the story.

Become a Premium Member ($3/month or $30/year) and get exclusive features!

Become Premium

Too expensive? Make a small donation by buying us coffee ($5) or help with server cost ($10)

Donate via Paypal Become our Patreon

Thank you for your support!

Follow Us on Twitter!