iiinova — where your digital self keeps evolving

chriswells95.bsky.social @chriswells95.bsky.social · 10/03/2026

(Though I should probably still mention I have a poster this afternoon in POS3.1 on our emulator work)

010

chriswells95.bsky.social @chriswells95.bsky.social · 10/03/2026

Might have to pivot this account to an International Conference Centre Kyoto fan page

110

Reposted by @chriswells95.bsky.social

Zarah Sultana MP @zarahsultana.bsky.social · 28/02/2026

An Omani mediator said last night a peace deal between the US and Iran was within reach. This morning, the US and Israel bombed Iran. No imminent threat. No legal basis. Another illegal war dressed up as “security” and “democracy”.

18445180

chriswells95.bsky.social @chriswells95.bsky.social · 18/02/2026

As promised, here's the representation of climate impacts in FRIDA:

100

chriswells95.bsky.social @chriswells95.bsky.social · 18/02/2026

The next step here is to use FRIDA-Clim to contribute to RCMIP3 (www.rcmip.org) which will compare SCM responses across a range of experiments. And as the FRIDA IAM develops, to ensure we are integrating the climate in in a reasonable way. Thanks to all involved! 6/6

010

chriswells95.bsky.social @chriswells95.bsky.social · 18/02/2026

This part of the model complements the representation of climate impacts (see this thread; compare the fig in 2/N above with 7/N there) to enable us to study the effects of climate impacts on the overall system. We calibrate it externally to the main IAM, to ensure an accurate climate response. 5/N

100

chriswells95.bsky.social @chriswells95.bsky.social · 18/02/2026

Crucially, we extend the representation in FaIR by including a process-based carbon cycle, which is integrated with the land use & agriculture side of the IAM - and the ocean representation - to track carbon through the system. This lets us look at emissions-driven experiments like flat10MIP. 4/N

100

chriswells95.bsky.social @chriswells95.bsky.social · 18/02/2026

but we reduce the number of species, as we need to be able to model the human drivers of the inputs within the IAM in order to close the human-Earth feedback loop. We model CO2, CH4, N2O, and SO2, and some minor effects, and correlate some species with them; we just have Montreal gases external. 3/N

110

chriswells95.bsky.social @chriswells95.bsky.social · 18/02/2026

We use a 3-layer Energy Balance Model to take the emissions and other forcings from the human modules in the IAM (plus solar & volcanic forcings) and estimate GMST. We also have a sea level rise module (documented here gmd.copernicus.org/articles/18/...). The model is heavily based on FaIR... 2/N

100

chriswells95.bsky.social @chriswells95.bsky.social · 12/02/2026

So what next? Analysis-wise: we need to see how these climate damages affect the trajectory of the human-Earth system, and interact with each other. Development-wise: we need to model the most important damages in more detail, and add any key ones missing. (lots) more info in the paper! 8/8

000

chriswells95.bsky.social @chriswells95.bsky.social · 12/02/2026

This results in a set of impact channels with varying levels of detail/certainty, with the whole IAM calibrated to historical data across all domains. Here we show all these, with the Climate Module (left) affecting the other modules (right) and the human -> climate feedback loop closed below. 7/N

110

chriswells95.bsky.social @chriswells95.bsky.social · 12/02/2026

1) condense a list from the IPCC AR6 WGII technical summary on climate impacts 2) implement the damages in there which we could find literature for at the global scale 3) use our historical calibration to estimate the functions for channels expected to be important but without sufficient data 6/N

100

chriswells95.bsky.social @chriswells95.bsky.social · 12/02/2026

can be explored in a coupled IAM like FRIDA, to the extent that the climate damages are represented. This is a challenge(!) because there are so many possible climate damages at different scales, with different levels of focus in the literature. So we followed a 3-step approach to this: 5/N

100

chriswells95.bsky.social @chriswells95.bsky.social · 12/02/2026

we have every reason to expect climate damages will affect the total system in complex ways- e.g. increased A/C demand from heatwaves will increase emissions, and climate damages to emissions-generating infrastructure will reduce emissions. These feedbacks are missed in linear frameworks, but... 4/N

100

chriswells95.bsky.social @chriswells95.bsky.social · 12/02/2026

with good reason. e.g. in ScenarioMIP, any representation of climate impacts is switched off in the IAMs because a) this would double count damages when running the ESM output in impact models (ISIMIP) and b) any representation in the IAMs is less detailed than those impact models anyway. But... 3/N

100

chriswells95.bsky.social @chriswells95.bsky.social · 12/02/2026

This work helps close the feedback loop in our initial attempt to represent the coupled human-Earth system in an IAM. But why are we doing this? Climate impacts have often been neglected in IAMs, either because they straddle the whole system or because they are deliberately excluded - often... 2/N

100

chriswells95.bsky.social @chriswells95.bsky.social · 12/02/2026

New paper in GMD presenting the climate impact functions in our new IAM FRIDAv2.1! We cover a broad range of impact channels at the global level, from Energy supply/demand to crops, mortality, labour productivity, etc. 1/N gmd.copernicus.org/articles/19/...

gmd.copernicus.org

The representation of climate impacts in the FRIDAv2.1 Integrated Assessment Model

Abstract. Feedbacks from the climate to other components of the coupled human-Earth system are expected to strongly influence the co-evolution of human society and its environment. Representing these ...

120

chriswells95.bsky.social @chriswells95.bsky.social · 02/02/2026

prior ECS distribution on the posteriors is not straightforward. iopscience.iop.org/article/10.1... Thanks to the co-authors and to ELR for having this in the Overshoot collection! I'll have a poster on this at CMIP26 in March! 8/8

010

chriswells95.bsky.social @chriswells95.bsky.social · 02/02/2026

2) All else being equal, we find higher warming and SLR. But that "all else equal" is important: we calibrate to ESM data, and we don't constrain these to observations. A next step will be to dig into a constrained ensemble using these, but this isn't straightforward, as the effect of the 7/N

110

chriswells95.bsky.social @chriswells95.bsky.social · 02/02/2026

What does this mean then? 1) to capture the full response to 4xCO2 we should have longer runs as standard, if possible. The CMIP7 protocol says 300 years but we find that is too short. We also aren't suggesting 900 years is the full response - the deep ocean timescale increases past this. 6/N

100

chriswells95.bsky.social @chriswells95.bsky.social · 02/02/2026

We look at policy-relevant metrics, finding that the year of crossing warming levels is brought forward, especially for higher ones (as lower ones occur too soon to be sensitive...). Peak warming in overshoot scenarios can be decently higher; 0.07C in SSP119 and 0.24C in SSP534-OS in median. 5/N

100

chriswells95.bsky.social @chriswells95.bsky.social · 02/02/2026

We force the emulators with SSP ERFs and as we might expect, the higher ECS leads to warmer temperatures, particularly around 2100-2200. The accumulation of this heat drives higher thermosteric sea level rise too. These differences don't arise when going from 150 to 300 years either. 4/N

100

chriswells95.bsky.social @chriswells95.bsky.social · 02/02/2026

for which we have 900+ years of 4xCO2 data for 15 models. We just calibrate our 3-layer Energy Balance Model to different time periods, and find that ECS increases in 14/15 models when calibrating on 900 years instead of 150. So what does this do to scenario projections? 3/N

100

chriswells95.bsky.social @chriswells95.bsky.social · 02/02/2026

expensive, so 150-year runs have had to do. This causes issues for equilibrium climate sensitivity - hence the ECS / EffCS difference - but the question is: how does this affect emulator projections? To study these questions, we use the LongRunMIP data, 2/N journals.ametsoc.org/view/journal...

journals.ametsoc.org

100

chriswells95.bsky.social @chriswells95.bsky.social · 02/02/2026

New paper in ERL in which we study the effect of a long-known issue in climate emulators: that we typically calibrate them on only 150 years of 4xCO2 ESM runs. This a problem as we know the climate isn't in equilibrium after this time, but running ESMs is 1/N iopscience.iop.org/article/10.1...

162

chriswells95.bsky.social @chriswells95.bsky.social · 30/01/2026

Was fortunate to see The Voice of Hind Rajab with a director Q&A the other day; so heartbreaking it feels odd to recommend viewing it but it was the most powerful reminder that these genocidal crimes cannot be forgotten. And gutting to realise her story is but one of tens of thousands.

000

chriswells95.bsky.social @chriswells95.bsky.social · 15/12/2025

New paper today in GMD describing the Sea Level Rise module in the process-based FRIDA IAM, to examine the role of SLR impacts and adaptation in future scenarios. Also given as a standalone SLR model named FRISIA! Huge shoutout to Lennart Ramme for leading this gmd.copernicus.org/articles/18/...

021

chriswells95.bsky.social @chriswells95.bsky.social · 05/12/2025

Hmm, looking forward to it already 😬

000

chriswells95.bsky.social @chriswells95.bsky.social · 04/12/2025

Update: I read exactly zero pages while away, instead devoting my time to eating an industrial amount of houmus… winter reading it is

120

chriswells95.bsky.social @chriswells95.bsky.social · 08/11/2025

And what of the technologies used in the IAMs to allow for this theoretical overshoot? Luckily the authors have managed to publish a sequel asking just this question! I’ll be diving in this month while visiting family in Iraq.

000

chriswells95.bsky.social @chriswells95.bsky.social · 08/11/2025

A key point made is that the arguments for overshoot won’t go away when crossing a given threshold; the goalposts will be moved indefinitely. The question then becomes: how does climate politics change when we fully enter the “overshoot era”?

100

chriswells95.bsky.social @chriswells95.bsky.social · 08/11/2025

as this seems to not be occurring by itself. Related to this is the insight (again building on other work) that profit, not price, drives private investment in energy systems, including some incredible anecdotes of fossil companies using cheap solar power to fuel oil rigs!

110

chriswells95.bsky.social @chriswells95.bsky.social · 08/11/2025

(Building on other recent work they argue for recognising an energy addition rather than transition, in which renewables add capacity onto existing fossil stocks without displacing them). Ultimately they call for a “moment of reckoning” to enforce a climate Minsky bubble-popping moment,

110

chriswells95.bsky.social @chriswells95.bsky.social · 08/11/2025

This leads to their take on the energy transition, which they argue is being indefinitely postponed due to the build up of stranded assets, and the political inertia this implies.

100

chriswells95.bsky.social @chriswells95.bsky.social · 08/11/2025

They are clear they aren’t blaming scenarios researchers for the lack of climate action, as the material driver is ultimately the fossil economy.

120

chriswells95.bsky.social @chriswells95.bsky.social · 08/11/2025

Parts of this narrative are familiar but articulated brilliantly here. They argue this reflects dominant social interests and their influence on science more broadly, and while the specific history isn’t as detailed as I’d like here, it becomes hard to argue with the broader point.

110

chriswells95.bsky.social @chriswells95.bsky.social · 08/11/2025

The next section is provocative - arguing that historical IAMs, far from being neutral explorers of future scenarios, baked in an “overshoot ideology” by negating the possibility of more radical interventions to reduce emissions, and over-relying on negative emissions technologies.

100

chriswells95.bsky.social @chriswells95.bsky.social · 08/11/2025

Also featuring a damning opinion of the politics of the Paris Agreement, which they see as recognising the harm of 2C+ but baking in the impossibility of meeting this (via voluntary contributions; as always begs the question of what else to have done…).

100

chriswells95.bsky.social @chriswells95.bsky.social · 08/11/2025

It starts with a pretty bleak - and hard to argue with - assessment of the last couple years of global retreat from serious talk on climate, especially in finance…

110

chriswells95.bsky.social @chriswells95.bsky.social · 08/11/2025

I read this over the summer - quite the read! Thoroughly recommend for anyone working in climate especially on scenarios or overshoot. It’s a pessimistic read which unfortunately is hard to disagree with; not sure what to make of all of it but very thought-provoking.

1113

chriswells95.bsky.social @chriswells95.bsky.social · 06/11/2025

More detailed documentation hopefully coming soon... including the representation of climate and climate impacts

100

chriswells95.bsky.social @chriswells95.bsky.social · 06/11/2025

Newly published: an overview of FRIDAv2.1, a new IAM to model the coupled human-Earth system (including climate impacts!) gmd.copernicus.org/articles/18/...

151