Machine learning (ML), a branch of artificial intelligence (AI), has been increasingly used in environmental engineering due to the ability to analyze complex nonlinear problems (such as ones connected with water quality management) through a data-driven approach. This study provides an overview of different ML algorithms applied for monitoring and predicting river water quality. Different parameters could be monitored or predicted, such as dissolved oxygen (DO), biological and chemical oxygen demand (BOD and COD), turbidity levels, the concentration of different ions (such as Mg2+ and Ca2+), heavy metal or other pollutant's concentration, pH, temperature, and many more. Although many algorithms have been investigated for the prediction of river water quality, there are several which are most commonly used in engineering practice. These models mostly include so-called supervised learning algorithms, such as artificial neural network (ANN), support vector machine (SVM), random forest (RF), decision tree (DT), and deep learning (DL). To further enhance prediction power, novel hybrid algorithms, could be used. However, the quality of prediction is not only dependent on the applied algorithm but also on the availability of previously mentioned water quality parameters, their selection, and the combination of input data used to train the ML model.

Changes in air pollution in the region of the city of Novi Sad due to the COVID-19 induced state of emergency were evaluated while using data from permanently operating air quality monitoring stations belonging to the national, regional, and local networks, as well as ad hoc deployed low-cost particulate matter (PM) sensors. The low-cost sensors were collocated with reference gravimetric pumps. The starting idea for this research was to determine if and to what extent a massive change of anthropogenic activities introduced by lockdown could be observed in main air pollutants levels. An analysis of the data showed that fine and coarse particulate matter, as well as SO₂ levels, did not change noticeably, compared to the pre-lockdown period. Isolated larger peaks in PM pollution were traced back to the Aralkum Desert episode. The reduced movement of vehicles and reduced industrial and construction activities during the lockdown in Novi Sad led to a reduction and a more uniform profile of the PM_2.5 levels during the period between morning and afternoon air pollution peak, approximately during typical working hours. Daily profiles of NO₂, NO, and NO_X during the state of emergency proved lower levels during most hours of the day, due to restrictions on vehicular movement. CO during the state of the emergency mainly exhibited a lower level during night. Pollutants having transportation-dominated source profiles exhibited a decrease in level, while pollutants with domestic heating source profiles mostly exhibited a constant level. Considering local sources in Novi Sad, slight to moderate air quality improvement was observed after the lockdown as compared with days before. Furthermore, PM low-cost sensors’ usefulness in air quality assessment was confirmed, as they increase spatial resolution, but it is necessary to calibrate them at the deployment location.

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.

In this study, we describe the development of seasonal winter and summer (heating and non-heating season) land use regression (LUR) models for PM_2.5 mass concentration for the city of Novi Sad, Serbia. The PM_2.5 data were obtained through an extensive seasonal measurement campaign conducted at 21 locations in urban, urban/industrial, industrial and background areas in the period from February 2020–July 2021. At each location, PM_2.5 samples were collected on quartz fibre filters for 10 days per season using a reference gravimetric pump. The developed heating season model had two predictors, the first can be associated with domestic heating over a larger area and the second with local traffic. These predictors contributed to the adjusted R² of 0.33 and 0.55, respectively. The developed non-heating season model had one predictor which can be associated with local traffic, which contributed to the adjusted R² of 0.40. Leave-one-out cross-validation determined RMSE/mean absolute error for the heating and non-heating season model were 4.04/4.80 μg/m³ and 2.80/3.17 μg/m³, respectively. For purposes of completeness, developed LUR models were also compared to a simple linear model which utilizes satellite aerosol optical depth data for PM_2.5 estimation, and showed superior performance. The developed LUR models can help with quantification of differences between seasonal levels of air pollution, and, consequently, air pollution exposure and association between seasonal long-term exposure and possible health risk implications.

The aim of the study was to analyse relationship between PM2.5 and PM10 concentrations, traffic density and meteorological factors during the week with regular traffic regime and first week of COVID-19 lockdown in Novi Sad, Serbia. During the study period, which included working days and weekends, traffic emission contributions to PMs were also determined. Obtained results have shown higher PM, pressure and emission contribution values, lower temperature, relative humidity values, and lower traffic counts for all vehicle categories during the COVID-19 week. A positive correlation was detected only during the first week, between PM2.5 and passenger vehicles and lightweight trucks, as well as between PM10 and all categories of vehicles. Background PM2.5 and PM10 concentrations were moderately correlated to total traffic during the first week as well. Very strong and moderate positive correlation was detected between PM2.5 and PM10 concentrations and temperature during COVID-19 week. PM concentrations increased during COVID-19 week, but total traffic decreased by 68% on workdays and 22% on weekends, proving the impact of lockdown measures on traffic regime and intensity. Since relationship between PM2.5 and PM10 with different vehicle categories was confirmed only during first week, and PM and emission contribution concentrations were higher during COVID- 19 week, a secondary emission source of PM was strongly indicated. Very strong and strong positive correlations of PM2.5 and PM10 with temperature during COVID-19 week have confirmed lower temperature impact on PM concentrations and, consequently, increased impact of heating, as an emission source, due to lockdown measures and people staying at their homes.

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.