Accurate monitoring of water temperature (T), electrical conductivity (EC), and dissolved oxygen (DO) is essential for assessing river health, yet conventional methods are often limited by cost, coverage, and data latency. This study proposes an integrated framework that combines Sentinel-2 satellite imagery with machine learning (ML) models to enhance large-scale, real-time water quality monitoring. Five ML algorithms – deep neural networks, eXtreme gradient boosting, Kolmogorov–Arnold networks, long short-term memory (LSTM), and temporal Kolmogorov–Arnold networks – were evaluated using daily time-series data for the Danube River at Novi Sad from October 2012 to December 2023. Model interpretability was ensured through Shapley additive explanations and LIME. LSTM achieved the highest accuracy for temperature prediction (R2 = 0.97, RMSE = 1.45 °C, standard error = 1.31 °C), while XGBoost outperformed others for dissolved oxygen (R2 = 0.79). High accuracy was also observed for conductivity predictions (LSTM, R2 = 0.90). Seasonal variables and spectral bands (B8a, B4, B11) were found to be dominant predictors. The novelty of this work lies in the fusion of temporal deep learning, explainable AI, and multispectral satellite data for interpretable, scalable, and cost-effective water quality assessment. This approach supports timely water quality degradation and decision-making in river basin management.

Aquatic sediments serve as both a major sink and a potential secondary source of polybrominated diphenyl ethers (PBDEs), a class of persistent, bioaccumulative, and toxic pollutants. This study assessed ten PBDE congeners in Danube River sediments (Serbia), addressing their occurrence, spatial distribution, and potential sources. Σ10PBDE concentrations ranged from 0.52 µg/kg dry weight (War Island) to 31.21 µg/kg dry weight (Neštin), with BDE-209 as the predominant congener. Localized dominance of hepta- and penta-BDEs at Neštin and Šangaj suggested site-specific contamination patterns. Comparative analysis indicated levels comparable to or exceeding those in other industrialized regions, implicating anthropogenic activities as key contributors. Source apportionment using Principal Component Analysis identified three major pathways: (1) transport and debro-mination of higher-brominated congeners, (2) direct anthropogenic discharge, and (3) improper waste manage-ment. Hierarchical cluster analysis and Kohonen’s self-organizing maps pinpointed Neštin as a pollution hotspot. This study provides the most comprehensive dataset on PBDE contamination in Serbian Danube sediments, establishing a critical baseline for future monitoring and regulatory actions.

Air pollution poses a significant global health risk, with severe consequences for respiratory and cardiovascular diseases, particularly among vulnerable populations facing heightened exposure. The respiratory system is particularly susceptible to pollutants, resulting in various adverse health effects. Ambient air monitoring for benzo(a)pyrene (B(a)P) was conducted at four sites in Novi Sad throughout 2020, representing diverse environmental conditions. Sampling and analytical methods adhered to standardized protocols. B(a)P concentrations exceeded prescribed limits at most sites, with combustion processes and traffic emissions identified as primary pollution sources. Principal component analysis highlighted the dominance of combustion-related pollution, particularly during winter. Hierarchical cluster analysis revealed distinct pollution patterns across sites, with areas less burdened by traffic and combustion showing lower B(a)P levels. The study underscores the urgent need to mitigate local pollution sources, especially combustion and traffic emissions, to safeguard public health. The findings provide crucial baseline data for future monitoring efforts and highlight the importance of addressing air pollution for sustainable development and climate change mitigation.

The paper provides a comprehensive examination of resources available for the deployment of green hydrogen in Serbia. The assessment encompasses various aspects, including renewable energy potentials, technological advancements, and future projections. The evaluation considers factors such as solar and wind power capacities, which are pivotal for green hydrogen production. Additionally, the study delves into the policy landscape, addressing initiatives aimed at fostering the integration of green hydrogen into Serbia's energy matrix. The analysis combines quantitative data on energy production capacities with qualitative insights into the economic and environmental implications of green hydrogen utilization. While the nation boasts abundant renewable energy resources, challenges such as high production costs and infrastructure limitations hinder widespread adoption. However, with strategic initiatives and technological advancements, Serbia can overcome these hurdles and pave the way for a sustainable hydrogen economy. Assessing Serbia's green hydrogen potential, driven by over 24 095 MWp from solar and 10 750 MWp from wind, highlights the nation's capacity to harness renewable resources, with hydrogen production set to grow from 1915 tons in 2019 to 37 ,123 tons by 2040. The findings aim to contribute to the ongoing discourse on sustainable energy transitions and the role of green hydrogen in Serbia's evolving energy landscape.

Hydropower is a renewable energy source critical for balancing the electricity grid and integrating variable wind and solar energy. However, its clean energy credentials are increasingly scrutinised due to its potential greenhouse gas (GHG) emissions, particularly methane—a potent GHG. While some reservoirs act as carbon sinks, others are significant emission sources. Accurately quantifying and addressing these emissions is essential to ensure hydropower’s role as a low-carbon energy source and to mitigate the climate-finance risks associated with reservoir emissions.This paper critically analyzes existing methodologies for estimating hydropower-related GHG emissions, including direct field measurements, empirical and machine learning (ML) models, and simplified emission factors. These methods vary in their strengths, limitations, and uncertainties, with emissions being highly site-specific and influenced by climatic conditions, reservoir characteristics, water quality, and operational practices.The analysis highlights how ML and hybrid modeling approaches can improve accuracy, providing more dynamic and scalable predictions of GHG emissions. These advancements enable the identification of high-emission reservoirs and inform the development of targeted mitigation strategies.By advancing the understanding of hydropower emissions, this research supports the sustainable integration of hydropower into the energy mix, ensuring it displaces fossil fuel generation while maintaining its low-carbon status. Additionally, it provides actionable insights for policymakers to design strategies that promote low-carbon hydropower development, aligning with broader climate objectives.

Air pollution in Serbia is characterized by pronounced seasonal variability, with the heating period representing the most critical phase in terms of fine particulate matter (PM2.5). Understanding both short-term dynamics and the spatial distribution of PM2.5 in urban environments remains a key challenge. This study focuses on the assessment of PM2.5 variability in Novi Sad, combining high-resolution sensor measurements with spatial modeling perspectives. A field campaign was conducted at 21 locations across urban, industrial, mixed-use, and background environments during heating and non-heating seasons. The use of low-cost sensors enabled detailed insight into short-term fluctuations and daily patterns of PM2.5 concentrations. Wind and pollution roses were employed to explore dominant dispersion patterns and to better understand the influence of prevailing meteorological conditions on pollutant transport. To support data reliability, sensor measurements were compared with reference measurements, demonstrating a satisfactory level of agreement and confirming their suitability for further analysis and spatial interpretation. As a key outcome, high-resolution seasonal prediction maps of PM2.5 are presented as an extension of previously developed Land Use Regression (LUR) models for Novi Sad, representing the first implementation of this approach in Serbia. These maps reveal spatial distribution patterns, pollution hotspots, and potential exposure gradients across the urban area. In addition, a preliminary assessment of chronic health impacts based on PM2.5 exposure at selected monitoring locations will be discussed. Particular attention will be given to the interpretation of prediction maps and their role in understanding spatial exposure patterns in urban environments.