Atmospheric stability sets maximum moist heat and convection in the midlatitudes

Funing Li, Talia Tamarin-Brodsky

Extreme near-surface moist heat and severe convective storms are among the leading causes of weather-related damages worldwide. Here, we show that episodes of extreme moist heat and severe convection frequently co-occur across midlatitude land regions, and develop a theoretical framework that links their maximum potential intensities to preexisting low-level energy inversions. By accounting for the stored-energy nature of midlatitude severe convection, where moist heat and atmospheric instability accumulate before convection initiates, our work advances the understanding of convective constraints on extreme heat events. The theory identifies low-level inversions as a critical factor shaping compound extreme heat and convective weather risks, and offers a pathway for improving the modeling and future projection of these events.

Li, Funing, and Talia Tamarin-Brodsky. 2025. “Atmospheric Stability Sets Maximum Moist Heat and Convection in the Midlatitudes.” arXiv:2501.05351. Preprint, arXiv, August 6. https://doi.org/10.48550/arXiv.2501.05351.

Dependence of convective precipitation extremes on near-surface relative humidity

Robert J.van der Drift, Paul A. O'Gorman

Precipitation extremes produced by convection have been found to intensify with near-surface temperatures at a Clausius-Clapeyron rate of 6 to 7% K−1 in simulations of radiative-convective equilibrium (RCE). However, these idealized simulations are typically performed over an ocean surface with a high near-surface relative humidity (RH) that stays roughly constant with warming. Over land, near-surface RH is lower than over ocean and is projected to decrease by global climate models. Here, we investigate the dependence of precipitation extremes on near-surface RH in convection-resolving simulations of RCE. We reduce near-surface RH by increasing surface evaporative resistance. Simultaneously, we hold free-tropospheric temperatures fixed by increasing surface temperatures. This “top-down” approach produces an RCE state with a deeper, drier boundary layer, which weakens convective precipitation extremes in three distinct ways. First, the lifted condensation level is higher, leading to a small thermodynamic weakening of precipitation extremes. Second, the higher lifted condensation level also reduces positive buoyancy in the lower troposphere, leading to a dynamic weakening of precipitation extremes. Third, precipitation re-evaporates more readily when falling through a deeper, drier boundary layer, leading to a substantial decrease in precipitation efficiency. These three effects all follow from changes in near-surface relative humidity and are physically distinct from the mechanism that underpins the Clausius-Clapeyron scaling rate. Overall, our results suggest that changes in relative humidity must be taken into account separately from specific humidity when seeking to understand and predict changes in convective precipitation extremes over land.

Drift, Robert J. van der, and Paul A. O’Gorman. 2025. Dependence of Convective Precipitation Extremes on Near-Surface Relative Humidity. Journal of Climate. August 4. https://doi.org/10.1175/JCLI-D-24-0738.1.

CERA: A Framework for Improved Generalization of Machine Learning Models to Changed Climates

Shuchang Liu, Paul A. O'Gorman

Robust generalization under climate change remains a major challenge for machine learning applications in climate science. Most existing approaches struggle to extrapolate beyond the climate they were trained on, leading to a strong dependence on training data from model simulations of warm climates. Use of climate-invariant inputs improves generalization but requires challenging manual feature engineering. Here, we present CERA (Climate-invariant Encoding through Representation Alignment), a machine learning framework consisting of an autoencoder with explicit latent-space alignment, followed by a predictor for downstream process estimation. We test CERA on the problem of parameterizing moist-physics processes. Without training on labeled data from a +4K climate, CERA leverages labeled control-climate data and unlabeled warmer-climate inputs to improve generalization to the warmer climate, outperforming both raw-input and physically informed baselines in predicting key moisture and energy tendencies. It captures not only the vertical and meridional structures of the moisture tendencies, but also shifts in the intensity distribution of precipitation including extremes. Ablation experiments show that latent alignment improves both accuracy and the robustness across random seeds used in training. While some reduced skill remains in the boundary layer, the framework offers a data-driven alternative to manual feature engineering of climate invariant inputs. Beyond parameterizations used in hybrid ML-physics systems, the approach holds promise for other climate applications such as statistical downscaling.

Liu, Shuchang, and Paul A. O’Gorman. 2025. “CERA: A Framework for Improved Generalization of Machine Learning Models to Changed Climates.” arXiv:2509.00010. Preprint, arXiv, August 15. https://doi.org/10.48550/arXiv.2509.00010.

Boosting Ensembles for Statistics of Tails at Conditionally Optimal Advance Split Times

Justin Finkel, Paul A. O'Gorman

Climate science needs more efficient ways to study high-impact, low-probability extreme events, which are rare by definition and costly to simulate in large numbers. Rare event sampling (RES) and ensemble boosting offer a novel strategy to extract more information from those occasional simulated events: small perturbations in advance can turn a moderate event into a severe one, which otherwise might not come for many more simulation-years. But how to choose this ``advance split time'' (AST) remains a challenge for sudden, transient events like precipitation. In this work, we formulate a concrete optimization problem for the AST and instantiate it on an idealized but physically informative model system: a quasigeostrophic turbulent channel flow advecting a passive tracer, which captures key elements of midlatitude storm track dynamics. Three major questions guide our investigation: (1) Can RES methods, in particular \emph{ensemble boosting} and \emph{trying-early adaptive multilevel splitting}, accurately sample extreme events of return periods longer than the simulation time? (2) What is the optimal AST, and how does it depend on the definition of the extreme event, in particular the target location? (3) Can the AST be optimized ``online'' while running RES?
Our answers are tentatively positive. (1) RES can meaningfully improve tail estimation, using (2) an optimal AST of 1-3 eddy turnover timescales, which varies weakly but detectably with target location. (3) A certain functional that we call the \emph{thresholded entropy} successfully picks out near-optimal ASTs, eliminating the need for arbitrary thresholds that have thus far hindered RES methods. Our work clarifies aspects of the optimization landscape and can, in our view, guide future research efforts on optimizing and sampling transient extreme events more efficiently in general chaotic systems.

Finkel, Justin, and Paul A. O’Gorman. 2025. “Boosting Ensembles for Statistics of Tails at Conditionally Optimal Advance Split Times.” arXiv:2507.22310. Preprint, arXiv, July 30. https://doi.org/10.48550/arXiv.2507.22310.

Rare event sampling for moving targets: extremes of temperature and daily precipitation in a general circulation model

Justin Finkel, Paul A. O'Gorman

Extreme weather events epitomize high cost: to society through their physical impacts, and to computer servers that are used to simulate them to provide information to mitigate those impacts. It costs hundreds of years to sample a few once-per-century events with straightforward model integration, but that cost can be much reduced with rare event sampling, which nudges ensembles of simulations to convert moderate events to severe ones, e.g., by steering a cyclone directly through a region of interest. With proper statistical accounting, rare event algorithms can provide quantitative climate risk assessment at reduced cost. But this can only work if ensemble members diverge fast enough. Sudden, transient events characteristic of Earth's midlatitude storm track regions, such as heavy precipitation and heat extremes, pose a particular challenge because they come and go faster than an ensemble can explore the possibilities. Here we extend standard rare event algorithms to handle this challenging case in an idealized atmospheric general circulation model, achieving 5-10 times sped-up estimation of long return periods, such as 100-150 years from only 20 years of simulation for extremes of daily precipitation and surface temperature. The algorithm, called TEAMS (``trying-early adaptive multilevel splitting''), was developed previously in Finkel and O'Gorman (2024) using a toy chaotic system, and relies on a key parameter -- the advance split time -- which may be estimated based on simple diagnostics of ensemble dispersion rates. The results are promising for accelerated risk assessment across a wide range of physical hazards using more realistic and complex models with acute computational constraints..

Finkel, Justin, and Paul A. O’Gorman. 2025. “Rare Event Sampling for Moving Targets: Extremes of Temperature and Daily Precipitation in a General Circulation Model.” arXiv:2508.13120. Preprint, arXiv, August 18. https://doi.org/10.48550/arXiv.2508.13120.

Bringing statistics to storylines: rare event sampling for sudden, transient extreme events

Justin Finkel, Paul A. O'Gorman

A leading goal for climate science and weather risk management is to accurately model both the physics and statistics of extreme events. These two goals are fundamentally at odds: the higher a computational model's resolution, the more expensive are the ensembles needed to capture accurate statistics in the tail of the distribution. Here, we focus on events that are localized in space and time, such as heavy precipitation events, which can start suddenly and decay rapidly. We advance a method for sampling such events more efficiently than straightforward climate model simulation. Our method combines elements of two recent approaches: adaptive multilevel splitting (AMS), a rare event algorithm that generates rigorous statistics at reduced cost, but that does not work well for sudden, transient extreme events; and "ensemble boosting" which generates physically plausible storylines of these events but not their statistics. We modify AMS by splitting trajectories well in advance of the event's onset following the approach of ensemble boosting, and this is shown to be critical for amplifying and diversifying simulated events in tests with the Lorenz-96 model. Early splitting requires a rejection step that reduces efficiency, but nevertheless we demonstrate improved sampling of extreme local events by a factor of order 10 relative to direct sampling in Lorenz-96. Our work makes progress on the challenge posed by fast dynamical timescales for rare event sampling, and it draws connections with existing methods in reliability engineering which, we believe, can be further exploited for weather risk assessment.

Finkel, Justin, and Paul A. O’Gorman. “Bringing Statistics to Storylines: Rare Event Sampling for Sudden, Transient Extreme Events.” arXiv, February 2, 2024. https://doi.org/10.48550/arXiv.2402.01823.

Climate Change Contributions to Increasing Compound Flooding Risk in New York City

Ali Sarhadi , Raphaël Rousseau-Rizzi , Kyle Mandli , Jeffrey Neal , Michael P. Wiper , Monika Feldmann , and Kerry Emanuel

Efforts to meaningfully quantify the changes in coastal compound surge- and rainfall-driven flooding hazard associated with tropical cyclones (TCs) and extratropical cyclones (ETCs) in a warming climate have increased in recent years. Despite substantial progress, however, obtaining actionable details such as the spatially and temporally varying distribution and proximal causes of changing flooding hazard in cities remains a persistent challenge. Here, for the first time, physics-based hydrodynamic flood models driven by rainfall and storm surge simultaneously are used to estimate the magnitude and frequency of compound flooding events. We apply this to the particular case of New York City. We find that sea level rise (SLR) alone will increase the TC and ETC compound flooding hazard more significantly than changes in storm climatology as the climate warms. We also project that the probability of destructive Sandy-like compound flooding will increase by up to 5 times by the end of the century. Our results have strong implications for climate change adaptation in coastal communities.

Sarhadi, Ali, Raphaël Rousseau-Rizzi, Kyle Mandli, Jeffrey Neal, Michael P. Wiper, Monika Feldmann, and Kerry Emanuel. "Climate change contributions to increasing compound flooding risk in New York City." Bulletin of the American Meteorological Society 105, no. 2 (2024): E337-E356. DOI: https://doi.org/10.1175/BAMS-D-23-0177.1

Thunderstorm straight line winds intensify with climate change.

Andreas F. Prein

Straight line winds (SLWs), or non-tornadic thunderstorm winds, are causing widespread damage in many regions around the world. These powerful gusts are associated with strong downdraughts in thunderstorms, rear inflow jets and mesovortices. Despite their significance, our understanding of climate change effects on SLWs remains limited. Here, focusing on the central USA, a global hot spot for SLWs, I use observations, high-resolution modelling and theoretical considerations to show that SLWs have intensified over the past 40 years. Theoretical considerations suggest that SLWs should intensify at a rate of ~7.5% °C−1, yet the observed rates show a more pronounced increase of ~13% °C−1. The simulation results indicate a 4.8 ± 1.2-fold increase in the geographical extent affected by SLWs during the study period. These findings underscore the importance of incorporating intensifying SLWs into climate change adaptation planning to ensure the development of resilient future infrastructure.

Prein, A.F. Thunderstorm straight line winds intensify with climate change. Nat. Clim. Chang. (2023). https://doi.org/10.1038/s41558-023-01852-9