Thanks Matt! I read something like this in the Wallace textbook, but there aren't any examples of this (like, an EOF picking out something variance maximising but physically-bogus). Maybe that's what the paper you referenced talks about
Climate people, help! I'm late to the EOF game, and am trying to build up some *intuition* for what an EOF actually measures. Are there are good examples online(/colab notebooks?!) which go through this? Much obliged.
So, how do we rescue millions of historical weather observations that are not currently available to science, such as those in this table? Surely, ML can do this by now... 🧵
"Stationarity" is the belief that past climate is an accurate predictor of future conditions. Sounds silly, right? But we've baked this false sense of stability into all our human systems. From building codes to water systems to infrastructure, it's all designed for a planet that no longer exists.
ALT: a man is standing on the side of the road holding a shovel and a sign that says this would be us .
Not sure what I am achieving by writing this, but sometimes screaming into the void can be cathartic.
Being an academic can be the best job in the world, but sometimes it can be really tough - especially when you have issues with anxiety/self-esteem. To all others out there who are struggling, you're not alone!
That was my thinking. Trends in this region are strongly affected by land use change
We're seeing news of 1 in a 1000 year rainfall events, and of rivers reaching record height from Helene For a while, something has worried me about estimates of flood and precipitation return levels in the South and Southeast US, and (for once) it's not climate change. 1/
We also revisit the arguments of Held & Soden (2006) and show that "P=Mq" is a poor model for changes in cloud-base mass flux with warming! This is because the HS06 paper neglects changes in "precipitation efficiency". For this reason, and others, we argue that the HS06 constraint is not robust.
This 4%/K weakening in temperature coordinates is an emergent result across models, which we explain using simple moist physics. TLDR; the lapse rate weakens under warming (panel b), which is what drives the mass flux decline.
