Why traditional climate accounting fails to capture the true risk of overshoot, and how a new framework reveals the dangerous “multipliers” altering Earth’s response.
Tom — thanks for the careful placement of the figure and for taking on something genuinely hard here. The ACS framing is doing something Dan Miller has been arguing in words on the Climate Brink threads: that the *effective* sensitivity we're living through is higher than the headline ECS number suggests, because unmasking, methane chemistry, contrails, and land-use feedbacks are all running in parallel with CO₂. Putting numerical multipliers on those is a useful step toward making that case concrete.
The Amazon Bowen-ratio threshold from Wunderling et al. is a particularly nice example — the kind of mechanism-specific, regionally-grounded multiplier that distinguishes ACS from generic ECS discussion. The point that a permanently shifted biome locks in a higher regional multiplier even after CO₂ comes down is exactly the asymmetry that simple recovery scenarios miss.
The overshoot/CDR section is where this framework earns its keep. The carbon-sink-weakening multiplier and albedo memory loss on the way down are the right way to think about why net-negative emissions don't simply reverse net-positive ones. That's a contribution worth developing further.
Nice to see some numbers to this, now we need to put some numbers to the fixes so we can most appropriately address the problem. I have always advocated for well hydrated riparian zones especially along intermittent water ways and irrigation in places like the amazon can replicate the hydrological and atmospheric qualities of the same size forest as biomes overproduce bioaerosols. Spreader levee dryland restoration and ocean nutrient balancing would also go a long way to decreasing these amplifying effects. Many thanks.
This is an outline of the issues and conversation going on, that begins to express the complexity of the situation in one dialogue. And points us in the right direction. Framing probable new issues, and identifying where we may have been deficient in previous calculations. Everyone concerned about climate, the environment, and biodiversity loss, should be reading this.
p.s. You would rather hear from Tom than me, I promise you.
I foresee the stillbirth of intellect in the Milky Way looming on the event horizon. There is time, I’m just saying I have seen the black dragon. It is here.
Tom, isn't your piece substantial enough for an academic paper? The term ACS is yours (I think) and I dare say you could add some detail (I wondered where your mu values came from and how you ended with the composite x1.8 ECS).
Thank you. It probably could be but would need a lot more work to calculate and justify the various multiplier values. The values used in the article are guesses based on gut field and discussions rather than empirical data.
Hi Tom--There is lots I agree with in your presentation, but I do have a problem with comments at the start. What is logarithmic is not the climate sensitivity but the change in radiative forcing in response to changes in the CO2 concentration. This change in forcing might then be said to lead to a warming through Stefan-Boltzmann response but that is not the sensitivity because there are then all the other feedbacks and effects that you mention--some, like water vapor have known relationships (Clausius-Clapeyron) but most others do not and you list lots of the factors. So, there are some like snow albedo that only apply when there is snow, and so on--and then there are lots of other factors, etc. I've not gone and read all the citations, etc., but I don't see why everything has to lead to a smooth curve nor that the sensitivity is independent of where the continents are, mountain ranges and heights, sea level, and more. I do agree that it is critical, and all this talk about the the highest IPCC scenario now being very unlikely really should have been accompanied by the trend of the climate sensitivity increasingly seeming to be toward the higher end of what was the original range of the climate sensitivity, so the end effect may be less emissions and so a lower concentration than the high end proposed a couple of decades ago, but a high climate sensitivity (and probably also more different GHGs--i.e., molecules in the atmosphere with three or more atoms) and so no way is the issue of less concern than it has been. Best, Mike MacCracken
The data does suggest that Earth System Sensitivity, the temperature rise when all the feedbacks have settled out, so after tens of thousands of years, is indeed logarithmic and bounded, at least for the last 66 million years, if not the last 485 million. Emily Judd’s paper “A 485-million-year history of Earth’s surface temperature” shows this to be the case.
The first graph in the article is Dean Rovang’s presentation of this data. In fact Judd herself says it is extraordinary that the relationship does indeed hold across states even though continents move about between them.
Your point on the talk over scenarios is very insightful, linking the loss of SSP5-4.5 with ECS moving to 4.5 instead of 3 cancel each other out. And with the other forces I’ve highlighted, not linked to CO2, this makes it even more the case.
A new evapotranspiration (ET) correction method (PULSE) can recover up to 30% of the “missing” energy in eddy‑covariance tower measurements by using plant water‑use efficiency rather than assuming a fixed Bowen ratio. what this means from AI
If ET is under‑measured, then all downstream models — drought forecasts, convection initiation, cloud formation, and even carbon budgets — are biased toward a drier, hotter world.
Correcting ET upward by 20–30% would mean:
Lower modeled surface temperatures
Higher modeled humidity
Lower vapor pressure deficit
More frequent convective triggering
Greater moisture recycling
Higher carbon uptake (because WUE is tied to photosynthesis)
This is exactly the feedback loop you’ve been describing for riparian and spreader‑levee systems.
Tom separate to your analysis is the climate risks communication issue of thinking in terms of the higher "global average LAND surface warming" - that place where humans and terrestrial life lives.
Totally accept the scientific validity of GMST as a yardstick to measure, but as far as communicating the understandable impacts of that, surely telling the public and pollies what the avg land surface is going to hit soon is far more telling than the spinning wheels over rcp8.5 etc?
Tom, I'm with dan miller as well. I don't appreciate this other forum so will not comment there, I can't compete with the vanity. But what do you say think on the comments by Rees which to me reflect reality and the responses continue to promote the twisted belief that every failure every miss = a success?
There is a big push by the incumbents to say that renewables can’t be built without fossil fuels. That may have some truth now, but it won’t always be the case. Take mining which relies on diesel, but companies are developing electric diggers and so forth, in time these will replace the diesel fleet. They said it couldn’t be done for trucks, but 50% of new truck sales in December in China were electric.
All we can do is try, the alternative is highly undesirable.
"If there are no climate feedbacks, the equilibrium global temperature change in response to doubled CO2 climate forcing (4 W/m²) is
ΔTeq = 1.2°C (no feedbacks). (A1)
Water vapor and sea ice feedbacks are reasonably well understood and modeled. Together they produce a feedback gain g = gwv + gsi = ~0.5 and an equilibrium global warming
ΔTeq = 1.2°C/(1 – g) = 2.4°C
(water vapor and sea ice feedbacks, zero cloud feedback). (A2)
This doubling of the no-feedback response implies that the feedback (water vapor + sea ice)
supplies an equivalent forcing of 4 W/m²
Now consider a strong cloud feedback (gcl = 0.25): the gain is g = gwv + gsi + gcl = 0.75 and the equilibrium global warming is
Tom — thanks for the careful placement of the figure and for taking on something genuinely hard here. The ACS framing is doing something Dan Miller has been arguing in words on the Climate Brink threads: that the *effective* sensitivity we're living through is higher than the headline ECS number suggests, because unmasking, methane chemistry, contrails, and land-use feedbacks are all running in parallel with CO₂. Putting numerical multipliers on those is a useful step toward making that case concrete.
The Amazon Bowen-ratio threshold from Wunderling et al. is a particularly nice example — the kind of mechanism-specific, regionally-grounded multiplier that distinguishes ACS from generic ECS discussion. The point that a permanently shifted biome locks in a higher regional multiplier even after CO₂ comes down is exactly the asymmetry that simple recovery scenarios miss.
The overshoot/CDR section is where this framework earns its keep. The carbon-sink-weakening multiplier and albedo memory loss on the way down are the right way to think about why net-negative emissions don't simply reverse net-positive ones. That's a contribution worth developing further.
Dean
Thanks Dean, I appreciate that.
As you know I've been mulling this over for a while, so it was good to get it down on paper and out for comment.
Albedo is always ignored in mainstream media. Loss of albedo increases heat and that heat reverses carbon sinks.
Oh.. millions die too from the heat 🥵 now and tomorrow.
Let's help them brighten up and hydrate and green up.
We are all in this together.
Nice to see some numbers to this, now we need to put some numbers to the fixes so we can most appropriately address the problem. I have always advocated for well hydrated riparian zones especially along intermittent water ways and irrigation in places like the amazon can replicate the hydrological and atmospheric qualities of the same size forest as biomes overproduce bioaerosols. Spreader levee dryland restoration and ocean nutrient balancing would also go a long way to decreasing these amplifying effects. Many thanks.
Thank you Tom.
This is an outline of the issues and conversation going on, that begins to express the complexity of the situation in one dialogue. And points us in the right direction. Framing probable new issues, and identifying where we may have been deficient in previous calculations. Everyone concerned about climate, the environment, and biodiversity loss, should be reading this.
p.s. You would rather hear from Tom than me, I promise you.
I foresee the stillbirth of intellect in the Milky Way looming on the event horizon. There is time, I’m just saying I have seen the black dragon. It is here.
🫵👍
Tom, isn't your piece substantial enough for an academic paper? The term ACS is yours (I think) and I dare say you could add some detail (I wondered where your mu values came from and how you ended with the composite x1.8 ECS).
Philip
Thank you. It probably could be but would need a lot more work to calculate and justify the various multiplier values. The values used in the article are guesses based on gut field and discussions rather than empirical data.
Something to think about though.
Hi Tom--There is lots I agree with in your presentation, but I do have a problem with comments at the start. What is logarithmic is not the climate sensitivity but the change in radiative forcing in response to changes in the CO2 concentration. This change in forcing might then be said to lead to a warming through Stefan-Boltzmann response but that is not the sensitivity because there are then all the other feedbacks and effects that you mention--some, like water vapor have known relationships (Clausius-Clapeyron) but most others do not and you list lots of the factors. So, there are some like snow albedo that only apply when there is snow, and so on--and then there are lots of other factors, etc. I've not gone and read all the citations, etc., but I don't see why everything has to lead to a smooth curve nor that the sensitivity is independent of where the continents are, mountain ranges and heights, sea level, and more. I do agree that it is critical, and all this talk about the the highest IPCC scenario now being very unlikely really should have been accompanied by the trend of the climate sensitivity increasingly seeming to be toward the higher end of what was the original range of the climate sensitivity, so the end effect may be less emissions and so a lower concentration than the high end proposed a couple of decades ago, but a high climate sensitivity (and probably also more different GHGs--i.e., molecules in the atmosphere with three or more atoms) and so no way is the issue of less concern than it has been. Best, Mike MacCracken
Thank you Mike.
The data does suggest that Earth System Sensitivity, the temperature rise when all the feedbacks have settled out, so after tens of thousands of years, is indeed logarithmic and bounded, at least for the last 66 million years, if not the last 485 million. Emily Judd’s paper “A 485-million-year history of Earth’s surface temperature” shows this to be the case.
The first graph in the article is Dean Rovang’s presentation of this data. In fact Judd herself says it is extraordinary that the relationship does indeed hold across states even though continents move about between them.
Your point on the talk over scenarios is very insightful, linking the loss of SSP5-4.5 with ECS moving to 4.5 instead of 3 cancel each other out. And with the other forces I’ve highlighted, not linked to CO2, this makes it even more the case.
Cheers, Tom
Tom you may have seen this
https://phys.org/news/2026-05-evapotranspiration-method-recover-tower-energy.html
A new evapotranspiration (ET) correction method (PULSE) can recover up to 30% of the “missing” energy in eddy‑covariance tower measurements by using plant water‑use efficiency rather than assuming a fixed Bowen ratio. what this means from AI
If ET is under‑measured, then all downstream models — drought forecasts, convection initiation, cloud formation, and even carbon budgets — are biased toward a drier, hotter world.
Correcting ET upward by 20–30% would mean:
Lower modeled surface temperatures
Higher modeled humidity
Lower vapor pressure deficit
More frequent convective triggering
Greater moisture recycling
Higher carbon uptake (because WUE is tied to photosynthesis)
This is exactly the feedback loop you’ve been describing for riparian and spreader‑levee systems.
Thanks Theodore, I hadn’t spotted that. I’ll read the paper.
Quite rational considerations to me, leaving the detailed math to you. I liked it as a concept would love to hear Hansen's feedback on it.
I was going to make comment with my favourite points but there were too many Tom.
The bowen ratio and amazon ref were particularly haunting. Thanks.
Tom separate to your analysis is the climate risks communication issue of thinking in terms of the higher "global average LAND surface warming" - that place where humans and terrestrial life lives.
Totally accept the scientific validity of GMST as a yardstick to measure, but as far as communicating the understandable impacts of that, surely telling the public and pollies what the avg land surface is going to hit soon is far more telling than the spinning wheels over rcp8.5 etc?
Thanks Sean, that’s a very good point.
Tom, I'm with dan miller as well. I don't appreciate this other forum so will not comment there, I can't compete with the vanity. But what do you say think on the comments by Rees which to me reflect reality and the responses continue to promote the twisted belief that every failure every miss = a success?
https://www.theclimatebrink.com/p/on-the-death-of-rcp85/comment/261069042
There is a big push by the incumbents to say that renewables can’t be built without fossil fuels. That may have some truth now, but it won’t always be the case. Take mining which relies on diesel, but companies are developing electric diggers and so forth, in time these will replace the diesel fleet. They said it couldn’t be done for trucks, but 50% of new truck sales in December in China were electric.
All we can do is try, the alternative is highly undesirable.
Great equation!
Great "Let's tell the truth and say the way it REALLY is".
👍
Am always mindful of the tropic zones. They are getting the worst now and tomorrow.
ACS = ECSbaseline × ∑( Forcingi × μi ) yet another brilliant finding
Question1 is it ok to copy large part of ACS = ECSbaseline × ∑( Forcingi × μi ) and add more items?
Please do.
The Tom Harris Anthropogenic Climate Sensitivity
ACS = 1.8 x ECS = 5.4ºC
is consistent to Hansen ForestTrees.06August2025.pdf https://substack.com/@jimehansen
"If there are no climate feedbacks, the equilibrium global temperature change in response to doubled CO2 climate forcing (4 W/m²) is
ΔTeq = 1.2°C (no feedbacks). (A1)
Water vapor and sea ice feedbacks are reasonably well understood and modeled. Together they produce a feedback gain g = gwv + gsi = ~0.5 and an equilibrium global warming
ΔTeq = 1.2°C/(1 – g) = 2.4°C
(water vapor and sea ice feedbacks, zero cloud feedback). (A2)
This doubling of the no-feedback response implies that the feedback (water vapor + sea ice)
supplies an equivalent forcing of 4 W/m²
Now consider a strong cloud feedback (gcl = 0.25): the gain is g = gwv + gsi + gcl = 0.75 and the equilibrium global warming is
ΔTeq = 1.2°C/(1 – g) = 4.8°C (water vapor, sea ice, + strong cloud feedback). (A3)
" // end of Hansen gain feedback concept //
- my modified Tom Harris values follows herafter:
ACS = 1.2°C (no feedbacks and CO2 doubling). (TA1)
ACS = 1.2°C
* 1.45 CH4 μ≈1.45
* 1.8 Natural aerosol decline (μ≈1.2–1.8) includes forest burning for palm oil
* 1.3 The Bowen Ratio Shift (μ≈1.1–1.3): Deforestation
* 1.2 Aviation Contrails (μ≈1.2): High-altitude contrails act as artificial cirrus clouds
* 0.75 SOx Sulphate Aerosols (μ≈0.6–0.9)
* 0.9 Agricultural Irrigation (μ≈0.9)
* 2.3 regional Black Carbon on Snow (μ≈1.5–3.0)
* 0.65 Stratospheric ozone depletion (μ≈0.5-0.8)
ACS = 4.9°C (TA2)
https://www.columbia.edu/~jeh1/mailings/2025/ForestTrees.06August2025.pdf