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.

With more than 1.2 million fans expected in the country for the World Cup, Qatar has set up a tech hub that uses artificial intelligence to keep an eye on the spectators, predict crowd swells and even control stadium temperature. More than 100 technicians will be working around the clock at the Aspire Command and Control Center, closely monitoring images flashing across their screens via 200,000 integrated units, from 22,000 security cameras spread across all eight World Cup stadiums. It is from here that they can operate entry gates, ensure there is running water and keep the air conditioners humming smoothly. Facial recognition technology will enable the crew to zoom in on each of the 80,000 seats at Lusail Stadium, which is set to host 10 matches, including the final. Experts from cybersecurity to anti-terrorism to transport will be stationed at the centre, along with Qatari and FIFA officials.