Yuanxin Liu, Fandong Meng, Zheng Lin, Peng Fu, Yanan Cao, Weiping Wang, Jie Zhou. Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies. 2022.

The 2019 novel coronavirus disease (COVID-19) has become a severe public health and social problem worldwide. A limitation of the existing literature is that multiple environmental variables have not been frequently elaborated, which is why the overall effect of the environment on COVID-19 has not been conclusive. In this study, we used generalized additive model (GAM) to detect the relationship between meteorological and air pollution variables and COVID-19 in four urban agglomerations in China and made comparisons among the urban agglomerations. The four urban agglomerations are Beijing-Tianjin-Hebei (BTH), middle reaches of the Yangtze River (MYR), Yangtze River Delta (YRD), and the Pearl River Delta (PRD). The daily rates of average precipitation, temperature, relative humidity, sunshine duration, and atmospheric pressure were selected as meteorological variables. The PM_2.5, PM₁₀, sulfur dioxide (SO₂), nitrogen dioxide (NO₂), ozone (O₃), and carbon monoxide (CO) contents were selected as air pollution variables. The results indicated that meteorological and air pollution variables tended to be significantly correlated. Moreover, the nature of the relationship between severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and meteorological and air pollution variables (i.e., linear or nonlinear) varied with urban agglomerations. Among the variance explained by GAMs, BTH had the highest value (75.4%), while MYR had the lowest value (35.2%). The values of the YRD and PRD were between the above two, namely 45.6% and 62.2%, respectively. The findings showed that the association between SARS-CoV-2 and meteorological and air pollution variables varied in regions, making it difficult to obtain a relationship that is applicable to every region. Moreover, this study enriches our understanding of SARS-CoV-2. It is required to create awareness within the government that anti-COVID-19 measures should be adapted to the local meteorological and air pollution conditions.

Electronic proceedings of IJCAI 2021

Unraveling the evolution of "landscape pattern (LP) - ecosystem service (ES) - human well-being (HW)" (LEH) cascade process is crucial for optimizing LPs, mitigating ES degradation, and enhancing HW. This study focuses on Qinghai Province, China, employing social network analysis and Geodetector to explore the evolution and driving mechanism of LEH cascade. The results demonstrated that, ES synergies increased and trade-offs decreased in Qinghai between 2000 and 2020. The LEH cascade network grew more cohesive and interconnected, with system boundaries as interactions strengthened key factors -including the number of patches, soil conservation, and residential savings balance-emerged as crucial nodes for sustaining network's functionality. Single-factor detection showed that drivers of HW were best explained by soil organic carbon content (q = 0.42). For ESs, population density (q = 0.34) was the key factor, while LP was primarily influenced by factors like topographic roughness, albeit with weaker explanatory power. Factor interactions provided greater explanatory power for LP, ESs, and HW than any single factor. Both dual-factor and nonlinear enhancements significantly improved the model's explanatory power. The population density-fractional vegetation cover pair showed the strongest synergistic effect on ESs (q = 0.94), indicating nonlinear enhancement, while the temperature-precipitation pair was most explanatory for HW(q = 0.88). The study offers a valuable reference for the systematic understanding of the LEH cascade and provide a scientific foundation for the sustainable management of ecosystems and the enhancement of HW.