arXiv subjects

Nicholas J. Lutsko

Publications and source records attributed to Nicholas J. Lutsko.

2 recordsLinked to original sources

arXiv2026

The forgotten role of wave dynamics in modulating the low cloud response to warm pool warming

The Pattern Effect describes the dependence of top-of-atmosphere radiation anomalies on changes in the pattern of sea surface temperatures. The emerging consensus in the field explains the impact of Pacific warm pool temperature on radiation using Convective Quasi-Equilibrium Weak Temperature Gradient (QE-WTG) theory: warm pool warming leads to increase in free-tropospheric temperatures across the tropics, a strengthening of inversion, increased cloud cover in the East Pacific low cloud decks, and negative radiative anomalies. Here we call on overlooked past results and new simulations from the Energy Exascale Earth System model to show that Rossby waves dominate the low-cloud response over the subtropical East Pacific low cloud decks, leading to decrease cloud cover in the low cloud decks. While the global radiative response is negative and consistent with QE-WTG, it is dominated by the response of the deep tropics, rather than the subtropical low cloud decks.

physics.ao-ph↗

arXiv2024

Climate-Invariant Machine Learning

Projecting climate change is a generalization problem: we extrapolate the recent past using physical models across past, present, and future climates. Current climate models require representations of processes that occur at scales smaller than model grid size, which have been the main source of model projection uncertainty. Recent machine learning (ML) algorithms hold promise to improve such process representations, but tend to extrapolate poorly to climate regimes they were not trained on. To get the best of the physical and statistical worlds, we propose a new framework - termed "climate-invariant" ML - incorporating knowledge of climate processes into ML algorithms, and show that it can maintain high offline accuracy across a wide range of climate conditions and configurations in three distinct atmospheric models. Our results suggest that explicitly incorporating physical knowledge into data-driven models of Earth system processes can improve their consistency, data efficiency, and generalizability across climate regimes.

cs.LG↗