The mismatch between PM_2.5 exposure and related health burden during 2000-2023 in the Sichuan Basin, China

Sichuan Basin (SCB) is a critical region in China facing the dual pressures of air pollution and population aging. This study constructed high resolution (1 km) PM_2.5 datasets for SCB using advanced machine learning approaches - Super Resolution Generative Adversarial Networks (SRGAN) and Convolutional Neural Network - Long Short-Term Memory (CNN-LSTM). Evaluation results demonstrate good performance of the machine learning model (SRGAN: R² = 0.93, RMSE = 0.08; CNN-LSTM: R² = 0.93, RMSE = 9.79 µg/m³). Our analysis reveals distinct temporal patterns in PM_2.5 exposure and related public health burdens (PHB), closely tied to major policy interventions including the 11th Five-Year Plan (2006), Clean Air Action (2013), and COVID-19 lockdown (2020). While PM_2.5 exposure followed a "rise-slight fall-rapid fall-stable" pattern, PHB exhibited a mismatched dynamics as "rise-slight rise-fall-rebound" with the variation rates of +28.2, +7.1 %, -23.9 % and 27.9 %. PHB in SCB grew by 33.9 % from 2000 to 2023, with Chengdu contributing 49.8 % to the total burden. Population aging led to 62.2 % increase in PHB and substantially offsetting gains from pollution control measures. Before the COVID-19 breakout, stable decrease baseline mortality rate (BMR) helped alleviated PHB with -3.1 %, -4.8 % and -9.4 % in three stages. But during the post-COVID period, increase BMR begun to contribute to the rise in PHB. These findings underscore the urgent need for integrated policy approaches that simultaneously address air quality management and aging-related health vulnerabilities, particularly in urban centers where pollution and population density converge. Effective mitigation will require sustained, coordinated efforts across environmental and public health sectors to counteract the compounding effects of demographic change and environmental degradation. SYNOPSIS: Population aging and mortality rate variations caused the mismatch between PM_2.5 exposure and health burden during 2000-2023 in Sichuan Basin.