International Journal of Remote Sensing

Tariq, Muhammad Hamza; Hussain, Musawar; Khaliq, Muhammad Mahad; Saeed, Talha; Bergin, Michael Howard; Khokhar, Muhammad Fahim

doi: 10.1080/01431161.2025.2546157pmid: N/A

The proposed study explores the prediction of ground truth PM2.5 ug/m3 concentration using satellite imagery via a combined deep learning approach followed by a machine learning model. We employ a convolutional neural network (CNN) based on a modified version of the state-of-the-art VGG16 model for the sake of extracting deep features from satellite images. The output from the CNN are then concatenated with additional meteorological features including temperature and relative humidity, and the seasonal factors Month, Day, Year, and then fed into a Random Forest regression model, which is responsible for the prediction of PM2.5 ug/m3 concentration. The proposed methodology is further tested on different sites of Rawalpindi and Islamabad, Pakistan, including Road sites and Non-Road sites (residential). Various experiments were performed in order to test the robustness of the presented machine learning pipeline, in the first experiment, the model is trained and tested on a shuffled dataset of all the 30 sites, achieving a minimum Mean Absolute Error (MAE) of 11.27 µg/m3, a Root Mean Square Error (RMSE) of 16.07 µg/m3, and Pearson Correlation of 0.85 between the actual and the predicted PM2.5 ug/m3 concentration. The second experiment involves training separate models for each site type to evaluate their performance on unseen data, achieving a minimum RMSE of 16.08 ug/m3, MAEs of 12.64 ug/m3, and Pearson Correlations of 0.84, respectively. The results demonstrate the effectiveness of the proposed methodology in accurately predicting PM2.5 ug/m3 concentrations and highlight the potential for model improvement by targeting site-specific characteristics.

Behera, Sibu; Kannemadugu, Hareef Baba Shaeb; K.G, Divya; Taori, Alok

doi: 10.1080/01431161.2025.2559424pmid: N/A

This study investigates the spatial and temporal variations in tropospheric columnar NO₂ concentrations over thermal power plant (TPP) locations surrounding Delhi, utilizing data from the TROPOMI instrument on board the Sentinel-5P satellite. The data analysis analyses yearly, seasonal and monthly variability in NO₂ and influence by meteorological parameters from 2019 to 2024. Findings reveal significantly higher NO₂ concentrations during the pre-monsoon (3.1 × 10− 5 ± 2.1 × 10−6 mol m− 2) and winter (3.0 × 10− 5 ± 3.6 × 10−6 mol m− 2), with the lowest levels recorded during the monsoon (1.8 × 10− 5 ±1.43 × 10−6 mol m− 2), followed by the post-monsoon (2.1 × 10− 5 ±2.98 × 10−6 mol m− 2). Among the monitored TPPs, the Jajjar and Indira Gandhi plants consistently exhibited the highest annual NO₂ levels, likely due to their proximity to Delhi and additional emissions from urban sources. A significant decline in NO₂ concentrations was observed in 2020 during the COVID-19 lockdown, reflecting a temporary reduction in anthropogenic activities. However, levels rebounded post-2022, with a slight decrease in 2024, except during the Jajjar and Indira TPPs. Meteorological analysis indicates that wintertime inversions and low wind speeds contribute to pollution build-up, while high-magnitude monsoon winds enhance dispersion and can transport pollutants away from urban centres. Forward trajectory modelling using HYSPLIT reveals frequent air mass transport from NCTPP and Panipat TPPs towards Delhi. The other TPPs, such as Suaratgarh, Talwandi and Harduaganj, contribute significantly lower number of transport trajectories towards Delhi than others. This study illustrates the importance of satellite-based monitoring in evaluating NO₂ concentrations from TPPs and emphasizes the necessity for season-specific policy changes and strong emission regulations. This study approach is also valuable for monitoring NO₂ concentrations over other TPPs in India and the Globe and assessing their influence on the pollutant levels of the nearby cities.

Mondal, Sandipan; Kamaruzzaman, Yeny Nadira; Ray, Aratrika; Biswas, Ipsita; Ghosh, Arpita; Pradhan, Alakesh; Lee, Ming-An; Hsieh, Hung-Yen

doi: 10.1080/01431161.2025.2545639pmid: N/A

The impact of increasing global temperatures, and changing ocean conditions affect the distribution of marine organisms as well as on the marine ecosystem. This study investigates the implications of predicted climate change on the distribution of Scomberomorus commerson in the Taiwan Strait (TS). Ensemble modelling (using generalized additive model, gradient boosting machine and random forest) and two shared socio-economic pathways (SSPs: SSP 2.6–1.2 to1.8°C increase in temperature expected by 2100, and SSP 8.5–3.8 to 5°C increase in temperature expected by 2100) scenarios were used in this study. Our results indicated that the current range of S. commerson in the TS is between 22°N–25°N and 117.5°E–119°E. Both SSPs demonstrated that over 2045–2050, a notable increase in habitat range area will occur in the extent of the distribution of S. commerson. However, after this period, more habitat loss will occur for this species under SSP 8.5 than under SSP 2.6 (60%–80% vs. 2%–10%). Given climate change, policymakers should establish sustainable and management strategies including the identification and protection of climate refugia, the implementation of seasonal closures in susceptible southern habitats, and the enhancement of regional collaboration to promote sustainable fisheries in the TS and to ensure the well-being of the communities dependent on them. These measures are essential not only for the sustainability of S. commerson but also for other small pelagic fisheries and gillnet-dependent groups in the Taiwan Strait, which are particularly vulnerable to climate-induced habitat degradation.

Pathakoti, Mahesh; Sundara, Rajan Krishnan; Asuri, Kanchana Lakshmi; Dangeti, Mahalakshmi Venkata; Thadishetty, Santhoshi; Peethani, Sujatha; Shaik, Ibrahim; Mamidala, Taran; Pushpanathan, Raja; Goru, Srinivasa Rao; Mandal, Swastika; Remala, Naga Anjani; Sitiraju, Srinivasa Rao;