Türkiye'de yer seviyesindeki partikül madde konsantrasyonlarının bölgesel tahmininde farklı uydulardan temin edilen aerosol optik derinliği verisine dayanan bir yöntem geliştirilmesi

Respirable particulate matter with aerodynamic diameters lower than 10 µm (PM10) is closely associated with adverse impacts on the atmospheric environment and human health. Estimation of daily ground-level PM10 concentrations based on satellite aerosol optical depth (AOD) products has been widely used over large regions with the help of statistical regression models. This study was the first attempt in Turkey to estimate daily PM10 concentrations based on the newly released Collection 6.1 (C6.1) Moderate Resolution Imaging Spectroradiometer (MODIS) DTB AOD products from Terra and Aqua satellites across Turkey. This study developed two different space-time random forest (STRF) models that outperformed most models with strong predictive power to evaluate the differences between the time windows representing the mean PM10 around the overpass times of the satellites with the synergetic use of space-time information over Turkey. This model uses spatiotemporal information along with MODIS AOD, meteorological, and land-related data as input to improve the overall accuracy of PM10 estimation for 2008-2019. Solar radiation and dust mass surface concentrations were the most effective predictors for Terra and Aqua overpass times, respectively. The STRF model performed moderately well, with a moderate cross-validation (CV) correlation coefficient (R) of 0.71 (0.67) and a low root-mean-square error over the national scale with AOD. The model better estimated PM10 concentrations at both regional (R up to 0.80) and individual site scales (R up to 0.90). Moreover, the STRF model developed in this study is a first step to construct a high-quality PM10 dataset across Turkey that is important for air pollution studies.