this is such fascinating stuff, yet in somewhat layman's language, so even I can enjoy it. Thank-you SO much for offering it to us......I love following the weather and the why's of it....
The enso forecast dashboard with interactive graphs is very nice. The plots, of course, are based on the traditional Nino 3.4, so have a warm bias. Please consider also adding the relative Nino 3.4 to your dashboard. Great work!
RONI would be great, but it would be a much bigger project to calculate as I'd have to use the gridded fields of every individual ensemble member (vs using pre-calculated Nino3.4 region anomalies).
Zeke, this is really helpful. How if at all should i understand the lower whiskers going back down in July/August when the bottom of the inter quartile just keeps rising?
I have a question about the recent letter from Howarth, Mann, et al. to Governor Hochul about the proposed changes to New York State's methane "accounting" under the state's CLCPA of 2019, which would have the effect of encouraging the use of methane gas for electrical generation and residential and commercial heating. I noticed that Dr. Hausfather did not sign the letter, and am wondering if this reflects a fundamental disagreement with these other climate scientists about the proper way to account for methane emissions (combustion and leaks) when comparing the warming potential of various energy sources.
Thank you for an interesting article. The unpredictability of ENSO events has always struck me as weird given the magnitude of the effect. Could you say a little bit about these forecasts. Are these global circulation models (is that the term?) working hour-by hour- to model the climate months ahead? Or are these forecasts looking at some specific indicators thought to be linked to El Niño events. Thanks. M
I don't think the forecasts work at all. II know many use GCMs, but there's a significant paradox in that ENSO researchers use toy models in trying to explain the behavior (using delay differential equations, recharge oscillators, etc) yet those are nowhere close to being full-fledged GCMs. In other words, they don't understand why they don't work, the full GCMs are too complex, so they opt for toy models out of desperation
The most mysterious El Nino, comes out of nowhere and wreaks havoc on global climate for the next year. How? They say it's caused by a sudden change in prevailing winds in the Pacific. But what causes that change? They don't know. Could it be that the change in prevailing winds is caused by something below the surface, so they are mistaking that causation? That's too obvious, so this paper gets dismissed "Switch Between El Nino and La Nina is Caused by Subsurface Ocean Waves Likely Driven by Lunar Tidal Forcing"
→Expert system← AIs, not communicate-with-humans "generative" AIs*, will be able to figure these things out.
I will trust expert systems to competently fold molecules, drive/fly vehicles, forecast weather, assess potential pharmaceuticals, map galaxies, forecast climate, design products and design the production lines for those products.
__________________
*Large language models (LLMs) that the public hears about and uses.
" drive/fly vehicles" -- I kind of know the diff between an expert system and what passes for AI these days. See my avionics paper "Expert system for redundancy and reconfiguration management" from 1992 (!). So, yes, in the past an expert system was essentially a set of rules programmed in to reason around a specific context or situation. In contrast, these days an LLM can reason about a specific situation by bringing in an entire universe of context, potentially finding a solution from an unrelated scientific discipline where the math happens to be the same. Much progress has been made since the early days.
Second, leveraging my past experience with Creationists, I subscribe here mostly to deal with climate deniers (and pontificate on everything else when those are not available, of course).
this is such fascinating stuff, yet in somewhat layman's language, so even I can enjoy it. Thank-you SO much for offering it to us......I love following the weather and the why's of it....
Barbara (David)
The enso forecast dashboard with interactive graphs is very nice. The plots, of course, are based on the traditional Nino 3.4, so have a warm bias. Please consider also adding the relative Nino 3.4 to your dashboard. Great work!
RONI would be great, but it would be a much bigger project to calculate as I'd have to use the gridded fields of every individual ensemble member (vs using pre-calculated Nino3.4 region anomalies).
Zeke, this is really helpful. How if at all should i understand the lower whiskers going back down in July/August when the bottom of the inter quartile just keeps rising?
Its two models (DWD and CanESM) that have lower Nino3.4 region anomalies after June that push the bottom of the whiskers down.
Aren't spring forecasts known to be unreliable?
I have a question about the recent letter from Howarth, Mann, et al. to Governor Hochul about the proposed changes to New York State's methane "accounting" under the state's CLCPA of 2019, which would have the effect of encouraging the use of methane gas for electrical generation and residential and commercial heating. I noticed that Dr. Hausfather did not sign the letter, and am wondering if this reflects a fundamental disagreement with these other climate scientists about the proper way to account for methane emissions (combustion and leaks) when comparing the warming potential of various energy sources.
Thanks, Zeke, always read your prognostications with interest (and, half the time at least, with cauterising alarm).
Does the ·intensity· of the El Niño directly correlate with the intensity of the wind shear in the Atlantic basin?
Enquiring minds with family in hurricane zones want to know!
Thank you for an interesting article. The unpredictability of ENSO events has always struck me as weird given the magnitude of the effect. Could you say a little bit about these forecasts. Are these global circulation models (is that the term?) working hour-by hour- to model the climate months ahead? Or are these forecasts looking at some specific indicators thought to be linked to El Niño events. Thanks. M
I don't think the forecasts work at all. II know many use GCMs, but there's a significant paradox in that ENSO researchers use toy models in trying to explain the behavior (using delay differential equations, recharge oscillators, etc) yet those are nowhere close to being full-fledged GCMs. In other words, they don't understand why they don't work, the full GCMs are too complex, so they opt for toy models out of desperation
GCMs are not good at El Nino predictions. There are specialized models for that. But I don't think they are reliable in March.
Zeke, thanks for that heads up that we'll get not just an early El Nino but a strong one later this year. It put War and Economic news in perspective.
Is it called el ninon ( Sp ending for big or monstrous) or el hombre ( grew up )
The most mysterious El Nino, comes out of nowhere and wreaks havoc on global climate for the next year. How? They say it's caused by a sudden change in prevailing winds in the Pacific. But what causes that change? They don't know. Could it be that the change in prevailing winds is caused by something below the surface, so they are mistaking that causation? That's too obvious, so this paper gets dismissed "Switch Between El Nino and La Nina is Caused by Subsurface Ocean Waves Likely Driven by Lunar Tidal Forcing"
ai-scientists-help-human-ones-answer-urgent-climate-questions in Bloomberg News today. AI will figure this all out, right?
→Expert system← AIs, not communicate-with-humans "generative" AIs*, will be able to figure these things out.
I will trust expert systems to competently fold molecules, drive/fly vehicles, forecast weather, assess potential pharmaceuticals, map galaxies, forecast climate, design products and design the production lines for those products.
__________________
*Large language models (LLMs) that the public hears about and uses.
" drive/fly vehicles" -- I kind of know the diff between an expert system and what passes for AI these days. See my avionics paper "Expert system for redundancy and reconfiguration management" from 1992 (!). So, yes, in the past an expert system was essentially a set of rules programmed in to reason around a specific context or situation. In contrast, these days an LLM can reason about a specific situation by bringing in an entire universe of context, potentially finding a solution from an unrelated scientific discipline where the math happens to be the same. Much progress has been made since the early days.
Whatever.
Chinese companies are working on cargo drones that can carry supplies to remote destinations (close to VTOL) that don't have airports.
You doing any climate simulations w/ or w/o AI, or just pontificating?
If you do wanna join in, visit geoenergymath dot com, OR github dot com/pukpr/GEM-LTE
First, I'm ·always· pontificating.
Second, leveraging my past experience with Creationists, I subscribe here mostly to deal with climate deniers (and pontificate on everything else when those are not available, of course).