Regional collaboration to simultaneously mitigate PM2.5 and O3 pollution in Beijing-Tianjin-Hebei and the surrounding area: Multi-model synthesis from multiple data sources

Quantitative Estimation of the Impacts of Precursor Emissions on Surface O3 and PM2.5 Collaborative Pollution in Three Typical Regions of China via Multi-Task Learning

“…Most previous studies only explored the impact of the influencing factors on pollution [29] or O3 pollution [16,30], while there are few studies revealing the effect contributions of the influencing factors on PM2.5 and O3 collaborative pollution simu ously [31][32][33]. In addition, statistical models have also been used by few studies to r the influence and contribution of influencing factors to PM2.5 and O3 collaborative p tion, with the mainly considered influencing factors being meteorological [10] and pr sor [34][35][36] factors. In this work, the relative importance of precursor emissions, met logical factors, population density, NDVI, LULC, and other factors on PM2.5 and O laborative pollution in the NCP, the PRD, and the YRD were assessed comprehens In addition, few previous studies using statistical models are not able to estimate an veal the contribution of the same factor to both PM2.5 and O3 simultaneously, and was larger than that in 2010, but it decreased significantly from 2015 to 2020, and, therefore, values of the O 3 response to SO 2 were all positive in 2015-2020, but the spatial extent covered by positive values gradually decreased.…”

“…Most previous studies only explored the impact of the influencing factors on PM 2.5 pollution [29] or O 3 pollution [16,30], while there are few studies revealing the effects and contributions of the influencing factors on PM 2.5 and O 3 collaborative pollution simultaneously [31][32][33]. In addition, statistical models have also been used by few studies to reveal the influence and contribution of influencing factors to PM 2.5 and O 3 collaborative pollution, with the mainly considered influencing factors being meteorological [10] and precursor [34][35][36] factors. In this work, the relative importance of precursor emissions, meteorological factors, population density, NDVI, LULC, and other factors on PM 2.5 and O 3 collaborative pollution in the NCP, the PRD, and the YRD were assessed comprehensively.…”

“…In this work, the relative importance of precursor emissions, meteorological factors, population density, NDVI, LULC, and other factors on PM 2.5 and O 3 collaborative pollution in the NCP, the PRD, and the YRD were assessed comprehensively. In addition, few previous studies using statistical models are not able to estimate and reveal the contribution of the same factor to both PM 2.5 and O 3 simultaneously, and more than one model needs to be built to estimate and reveal the contribution of a particular factor to both PM 2.5 and O 3 [35,36]. In this work, daily gridded PM 2.5 and O 3 datasets and their relative influencing factors were utilized to establish the PM 2.5 and O 3 MTL estimation model.…”

“…Few previous studies that simultaneously reveal the influence and contribution of influencing factors to the synergistic pollution of PM 2.5 and O 3 mainly use atmospheric chemistry models, and the accuracy of the models needs to be improved [32,33]. In addition, the few studies that have used statistical modelling to reveal the influence and contribution of influencing factors to PM 2.5 and O 3 collaborative pollution have mainly considered meteorological [10] and precursor [34][35][36] factors. However, these statistical models are not able to estimate and reveal the contribution of the same factor to both PM 2.5 and O 3 simultaneously, and more than one model needs to be built to estimate and reveal the contribution of a particular factor to both PM 2.5 and O 3 [35,36].…”

Machine learning-based evaluation of dynamic thermal-tempering performance and thermal diversity for 107 Cambridge courtyards

“…The self-organising map (SOM) is an ANN algorithm for processing high-dimensional-low-sample-size (HDLS) datasets and detecting outliers. The SOM, also known as the Kohonen map (Kohonen, 1982), is one of the most widely applied unsupervised learning techniques in identifying spatial-temporal principles for environmental and ecological modelling results, such as synoptic climatology (Sheridan & Lee, 2011), air and water quality (Duan et al, 2022;Ejarque-Gonzalez & Butturini, 2014;Qu et al, 2021), land surface temperature (Hu & Weng, 2009), biodiversity (Shanmuganathan et al, 2006), smart metering (Mcloughlin et al, 2015), and perceptions towards climate change (Sugg, 2021).…”