Atmospheric particulate matter (PM) has major threats to global health, especially in urban regions around the world. Dhaka, Narayanganj and Gazipur of Bangladesh are positioned as top ranking polluted metropolitan cities in the world. This study assessed the performance of the application of hybrid models, that is, Autoregressive Integrated Moving Average (ARIMA)-Artificial Neural Network (ANN), ARIMA-Support Vector Machine (SVM) and Principle Component Regression (PCR) along with Decision Tree (DT) and CatBoost deep learning model to predict the ambient PM2.5 concentrations. The data from January 2013 to May 2019 with 2342 observations were utilized in this study. Eighty percent of the data was used as training and the rest of the dataset was employed as testing. The performance of the models was evaluated by R2, RMSE and MAE value. Among the models, CatBoost performed best for predicting PM2.5 for all the stations. The RMSE values during the test period were 12.39 µg m−3, 13.06 µg m−3 and 12.97 µg m−3 for Dhaka, Narayanganj and Gazipur, respectively. Nonetheless, the ARIMA-ANN and DT methods also provided acceptable results. The study suggests adopting deep learning models for predicting atmospheric PM2.5 in Bangladesh.

Drought is one of the most significant repercussions of climate change. Worldwide droughts affect food security and ecological productivity. Bangladesh has faced a series of droughts over the past few decades, with significant economic and environmental consequences. The north-western region of Bangladesh is the most affected by drought because of its geographical location and semi-arid climate. With the increasing frequency and severity of droughts, rapid and reliable drought information is essential for agro-ecological production and food security. Using the Standardized Precipitation Index (SPI) and three models (Auto Regressive Moving Average (ARMA), PROPHET, and ARMA-Generalized Autoregressive Conditional Heteroskedasticity (ARMA-GARCH)), we assessed the trends of drought in the five meteorological stations (Bogra, Dinajpur, Ishwardi, Rajshahi, and Rangpur) in the north-western region of Bangladesh for the period 1980–2019. Results show that the SPI trends were significant for Dinajpur and Ishwardi stations but insignificant for the other three stations (Bogra, Rajshahi, and Rangpur). Among the three models, the hybrid model (ARMA-GARCH) outperformed the individual models (ARMA and PROPHET), which suggests that the ARMA-GARCH model could be utilized to predict droughts as it showed higher accuracy than that of individual models. This study provides empirical evidence of (i) the intensification of drier climates in the north-western region of Bangladesh over the 40 years, which has practical implications for introducing climate adaptive practices in agriculture and other livelihood sectors, and (ii) the better performance of a hybrid model compared to individual models in predicting drought, which is of great significance for government decision-making.