Correlation between biomass burning and air pollution in China: Spatial heterogeneity and corresponding factors

Abstract

Biomass burning (i.e., forest fires or open-field burning) is strongly correlated with regional air pollution. However, the correlation varied in different regions because of the complex interaction with other driving factors, leaving a substantial gap in understanding the mechanism of air pollution formation. This study aimed to clarify the spatial heterogeneity and corresponding factors of the biomass burning-air pollution correlation in China in recent decades (2001–2019). Multisource datasets, including satellite observations, in situ measurements, and annual statistics, were adopted for the evaluation through mathematical statistics and spatial-temporal analysis. The results showed that biomass burning mainly occurred in Northeast, Central and South China, which decreased slightly by −153.08 km 2 ·yr −1 during the study period. Under this background, air quality improved significantly, particularly after 2013. The strongest correlations occurred in Northeast China at the spatial scale or in spring at the temporal scale. Notably, the correlation was affected by several factors and their complex interactions. Specifically, the dominant factors of the correlations were gross domestic product (GDP) in Northeast China, atmospheric humidity in Central China, and population in South China. In addition, these factors exerted significant interactions, which were characterized by the two-factor enhancement of temperature-GDP in Northeast China, the nonlinear enhancement of temperature-DEM in Central China, and the nonlinear enhancement of temperature-population in South China. The results of this study were helpful for capturing the physical correlation between biomass burning and air pollution, which would further support relevant air pollution management and control. • Biomass burning impact on air pollution were evaluated in pixel-wise. • Strong correlations located in Northeast, Central and South China. • Dominant factors were GDP, humidity and population in three hotspots regions. • Interactions had a stronger effect on correlation than single factors.