Machine learning for weather and climate are worlds apart

Adaptive bias correction for improved subseasonal forecasting

Subseasonal forecasting-predicting temperature and precipitation 2 to 6 weeks ahead-is critical for effective water allocation, wildfire management, and drought and flood mitigation. Recent international research efforts have advanced the subseasonal capabilities of operational dynamical models, yet temperature and precipitation prediction skills remain poor, partly due to stubborn errors in representing atmospheric dynamics and physics inside dynamical models. Here, to counter these errors, we introduce an adaptive bias correction (ABC) method that combines state-of-the-art dynamical forecasts with observations using machine learning. We show that, when applied to the leading subseasonal model from the European Centre for Medium-Range Weather Forecasts (ECMWF), ABC improves temperature forecasting skill by 60-90% (over baseline skills of 0.18-0.25) and precipitation forecasting skill by 40-69% (over baseline skills of 0.11-0.15) in the contiguous U.S. We couple these performance improvements with a practical workflow to explain ABC skill gains and identify higher-skill windows of opportunity based on specific climate conditions.

A framework for benchmarking uncertainty in deep regression

We propose a framework for the assessment of uncertainty quantification in deep regression. The framework is based on regression problems where the regression function is a linear combination of nonlinear functions. Basically, any level of complexity can be realized through the choice of the nonlinear functions and the dimensionality of their domain. Results of an uncertainty quantification for deep regression are compared against those obtained by a statistical reference method. The reference method utilizes knowledge about the underlying nonlinear functions and is based on Bayesian linear regression using a prior reference. The flexibility, together with the availability of a reference solution, makes the framework suitable for defining benchmark sets for uncertainty quantification. Reliability of uncertainty quantification is assessed in terms of coverage probabilities, and accuracy through the size of calculated uncertainties. We illustrate the proposed framework by applying it to current approaches for uncertainty quantification in deep regression. In addition, results for three real-world regression tasks are presented.

Sensitivity of Air Pollution Exposure and Disease Burden to Emission Changes in China Using Machine Learning Emulation

Machine learning models can emulate chemical transport models, reducing computational costs and enabling more experimentation. We developed emulators to predict annual-mean fine particulate matter (PM_2.5) and ozone (O₃) concentrations and their associated chronic health impacts from changes in five major emission sectors (residential, industrial, land transport, agriculture, and power generation) in China. The emulators predicted 99.9% of the variance in PM_2.5 and O₃ concentrations. We used these emulators to estimate how emission reductions can attain air quality targets. In 2015, we estimate that PM_2.5 exposure was 47.4 μg m^-3 and O₃ exposure was 43.8 ppb, associated with 2,189,700 (95% uncertainty interval, 95UI: 1,948,000-2,427,300) premature deaths per year, primarily from PM_2.5 exposure (98%). PM_2.5 exposure and the associated disease burden were most sensitive to industry and residential emissions. We explore the sensitivity of exposure and health to different combinations of emission reductions. The National Air Quality Target (35 μg m^-3) for PM_2.5 concentrations can be attained nationally with emission reductions of 72% in industrial, 57% in residential, 36% in land transport, 35% in agricultural, and 33% in power generation emissions. We show that complete removal of emissions from these five sectors does not enable the attainment of the WHO Annual Guideline (5 μg m^-3) due to remaining air pollution from other sources. Our work provides the first assessment of how air pollution exposure and disease burden in China varies as emissions change across these five sectors and highlights the value of emulators in air quality research.