Improved Machine Learning Climate Model Separates Effects of Sea Surface Temperature and CO₂
Researchers at Ai2 have developed an improved version of their AI-based climate emulator, ACE, that can accurately disentangle the separate effects of CO₂ concentrations and sea surface temperatures on global climate. Previous versions failed in extreme scenarios because their training data had CO₂ and SST varying together, preventing the model from learning their independent contributions. The advance allows the emulator to handle physically challenging scenarios—such as abrupt CO₂ quadrupling and perturbed SST conditions—that earlier versions could not simulate without producing unphysical results.
The Ai2 Climate Emulator (ACE), a machine-learning-based tool designed to replicate the behavior of physics-based climate models, has been significantly improved in a new study by Clark et al. Prior ACE versions were trained on datasets where sea surface temperature (SST) and CO₂ concentrations were correlated, causing the model to conflate their effects and break down under novel forcing scenarios such as AMIP SST perturbed by +4 K or an abrupt quadrupling of CO₂. To address this, the team introduced a new class of 'random-CO₂' reference simulations in which SST and CO₂ are prescribed to vary independently, allowing the model to learn their separate contributions. The updated model is trained on a balanced mix of AMIP, equilibrium-climate, and random-CO₂ data, and incorporates a total energy conservation constraint that improves physical interpretability. The result is a more data-efficient emulator that performs accurately across both the scenarios where previous versions excelled and the challenging new ones where they failed. The authors acknowledge key limitations: the model uses simplified or prescribed representations of the ocean, land, and sea ice; does not account for other known climate forcings; and is trained exclusively on physics-based model output, inheriting whatever biases those models carry relative to real-world observations. Each limitation is identified as an avenue for future research.
What's missing
No validation against observational data (as opposed to model output) is presented, leaving the magnitude of inherited model biases unquantified.
What different sources said
- arXiv physicsCenter
Disentangling the effects of sea surface temperature and CO$_2$ in global machine learned weather-climate emulators
Related
Gut Bacteria Enzyme Found to Break Down Heat-Processed Food Compounds, Producing Novel Biogenic Amines
Researchers have discovered that an enzyme in common gut bacteria can degrade N-epsilon-carboxymethyllysine (CML), a compound formed during thermal food processing, producing previously unknown biogenic amines. The enzyme, ornithine decarboxylase SpeC from enterobacteria, acts on CML and related modified lysine derivatives through a low-level 'underground' catalytic activity. This finding suggests a previously unrecognized communication axis between thermally processed dietary compounds and gut microbial physiology, with potential implications for host health.
Full-Length Gene Sequencing Reveals Two Distinct Bacterial Communities in Black-Legged Ticks Expanding Into Canada
Researchers used Oxford Nanopore full-length 16S rRNA gene sequencing to characterize the microbiome of Ixodes scapularis black-legged ticks collected in Nova Scotia, Canada, distinguishing between tick-adapted bacteria and environmentally acquired bacteria. The study comes as I. scapularis — the primary vector of Lyme disease — is rapidly expanding northward into Canada due to climate change. The findings suggest that environmentally derived bacteria in tick microbiomes are not mere contamination, which has implications for how tick microbiome data is collected and interpreted across surveillance studies.
Study Identifies Metabolic Link Between Cell Envelope Stress and Biofilm Formation in Bacteria
Researchers have discovered that the metabolite acetyl-CoA directly inhibits enzymes that degrade the bacterial signaling molecule c-di-GMP, connecting cell envelope biosynthesis stress to biofilm formation in Pseudomonas aeruginosa. The study found that sub-inhibitory concentrations of antibiotics targeting early peptidoglycan biosynthesis — but not other antibiotic classes — elevate c-di-GMP levels by reducing phosphodiesterase activity, with acetyl-CoA competing for the enzyme active site. Because the relevant enzyme domain is broadly conserved across bacterial species, this checkpoint mechanism may be widespread and could have implications for understanding antibiotic-induced biofilm responses.