The promotion of Nature-based solutions (NbS) through landscape planning is currently the focus of achieving climate resilience, not only through new urban development but also through retrofitting existing urban infrastructure. In this research, the opportunities for urban revitalization were investigated through a multi-leveled approach towards analysing the possibility for NbS interventions on a city scale, encompassing urban streetscape areas. Using the city of Novi Sad as a case study, five different locations were targeted for assessing the potential for retrofitting, based on previous conducted analysis, including citizen micro-scaled NbS allocation preferences, current climatic and soil conditions, and overall urban infrastructure. The design, area and type of chosen NbS tool, was accessed through computational modelling in order to measure positive outcomes of retrofitted micro-scaled locations in NBs. From the aspect of various layouts and design, the results revealed that the best performances of micro scaled streetscape NbS can be achieved when linking them to specific spatial context and project goals. The results of this study are important to critically analyse future NbS design choices and implement appropriate design elements from the standpoint of a city with a moderate continental climate, in developing resilience adaption plans, and for planning future infrastructure.

Constructed wetlands (CWs) are nature-based solutions increasingly applied for domestic wastewater treatment, yet their performance is sensitive to climatic conditions. This study presents an overview of key findings from the Gložan CW pilot site in Vojvodina, Serbia, assessed within the framework of the DALIA and SWIM Horizon Europe projects. Long-term treatment performance analysis demonstrated high and stable removal efficiency for organic matter, with average BOD₅ removal exceeding 90%, while phosphorus and suspended solids exhibited greater variability. Seasonal analysis indicated significantly lower BOD₅ removal during summer. Future climate projections were analysed using an ensemble of eight bias-corrected Euro-CORDEX regional climate models under the RCP8.5 scenario. Results indicate a progressive increase in thermal stress, with tropical days nearly doubling by 2071-2100, alongside a 49% reduction in frost days, a 21% increase in maximum one-day precipitation, and a modest decline in the Seljaninov Hydrothermal Coefficient. These changes present both opportunities - through extended frost-free operational seasons - and challenges, including increased hydraulic loading from extreme rainfall and moisture deficits during the growing season. Integrating climate adaptation into the long-term management of nature-based wastewater treatment systems is strongly recommended.

Reference evapotranspiration (ET0) is most commonly estimated using the FAO-56 Penman-Monteith (PM) equation. However, its application is often limited by the lack of required meteorological parameters. Due to their flexibility, ability to operate with limited input, and high accuracy in estimating ET0, machine learning models have become increasingly relevant in scientific research, offering a practical alternative under limited data conditions. In this study, artificial neural networks (ANNs) were applied to estimate daily ET0 using meteorological data from the Novi Sad station in Vojvodina (Serbia). The dataset consisted of eight meteorological variables relevant to evapotranspiration processes. Analysis showed that some variables had a stronger influence on ET0 prediction than others. To evaluate their combined effect, a series of ANN models with different input combinations were developed and tested. The random forests, gradient boosting and k-nearest neighbors models were used as a benchmark, and model performance was evaluated using R2, NSE, RMSE, and MAE. The highest accuracy was achieved when all variables were included, providing the model with maximum information. The best performance was obtained using a two-hidden-layer architecture with 32 and 16 neurons, resulting in R2 = 0.97, NSE = 97.07%, RMSE = 0.23 mm/day, and MAE = 0.21 mm/day. The results showed that a limited number of input variables can be used to estimate ET0 with high accuracy, achieving an R2 value of 0.95 using only three input variables. Therefore, the findings of this study may contribute to more accurate and cost-effective irrigation scheduling and water balance estimation, providing practical benefits for agricultural water management and farmers in Serbia.

The Danube River is one of Europe’s largest transboundary rivers, characterized by substantial spatial heterogeneity in environmental conditions, monitoring practices, and water management frameworks. Developing a harmonized approach for basin-wide surface-water quality assessment is therefore essential. This study presents the development and application of an adapted Water Quality Index (Danube WQI) for assessing and monitoring water quality along the Danube River, one of Europe’s largest and most complex transboundary systems. The Danube WQI is based on established WQI methodologies and integrates two objective weighting approaches—the Entropy Weight Method (EWM) and the CRITIC (Criteria Importance Through Inter-Criteria Correlation) method—to minimize subjectivity and improve the robustness of parameter weighting. Long-term water quality data from the TransNational Monitoring Network (TNMN) of the International Commission for the Protection of the Danube River (ICPDR) were used, covering 42 stations across nine countries (1996–2022). Nine parameters were selected: dissolved oxygen (DO), biochemical oxygen demand (BOD5), total nitrogen (TN), nitrate (NO3), ammonium (NH4), total phosphorus (TP), orthophosphate (PO4), electrical conductivity (EC), and pH. During the formation of sub-indices and rating curves, national water quality standards from the Danube countries were harmonized to ensure consistent parameter classification. Results indicate that the Danube River generally exhibits very good water quality, with most sections belonging to the first and second quality classes. Comparison with the Canadian Water Quality Index (CWQI) confirmed similar results but demonstrated higher seasonal sensitivity of the Danube WQI. Additionally, rankings obtained using the PROMETHEE II multicriteria method showed strong agreement with the Danube WQI classifications, further confirming the robustness of the proposed index. The proposed index provides a harmonized and transferable framework that can support integrated water management and policy evaluation across the Danube River Basin and within the EU Water Framework Directive context.

Waterlogging disasters are one of the most severe and widespread agricultural meteorological disasters. They affect about 15% of land surface globally, causing a significant reduction in crop growth and yields. This paper presents an objective methodology for assessing waterlogging risk, primarily in non-urban, predominantly agricultural areas. The waterlogging risk was assessed by evaluating vulnerability and hazard based on key environmental, anthropogenic, and climatic factors. The weights of factors affecting the waterlogging vulnerability were determined using the entropy weight method (EWM), assuring the objectivity of the overall evaluation results. The obtained waterlogging risk map was validated by comparing it with observed and detected waterlogged sites using Sentinel-2 imagery and Random Forest classification. The key novelties of this study are the use of the entropy weight method to objectively determine the relative importance of factors influencing waterlogging vulnerability, and a two-step validation process which includes field-based comparison and remote sensing validation. The presented methodology was demonstrated in the Vojvodina region, Serbia. The following waterlogging vulnerability factors were used: soil properties, geomorphology, surface depressions, average phreatic water table depth, and land cover. The EWM shows that surface depressions and soil properties have the most significant influence on waterlogging vulnerability. The highest waterlogging hazard classes occur in about 31% of the analyzed territory. The waterlogging hazard was estimated based on water balance for the non-vegetation season and maximum daily precipitation in spring, both modeled using the Generalize Extreme Value distribution function. The highest waterlogging hazard classes occur in about 31% of the analyzed territory. The final risk map shows that the high waterlogging risk occurs in about 11% of the territory. Those are mainly areas in the central, eastern, and southeastern parts of the Vojvodina region, usually along the main watercourses. High agreement between the detected waterlogged areas and the produced waterlogging risk map was achieved, validating the proposed methodology. The presented waterlogging risk assessment methodology is valuable for planning and policy-making for various water management and environmental activities. Although it is demonstrated in Vojvodina, by selecting the appropriate factors of vulnerability and hazard, it can be applied to any other region.

One of the frequently used drought metrics in scientific research is the consecutive dry days (CDDs) because it effectively indicates short-term droughts important to ecosystems and agriculture. CDDs are expected to increase in many parts of the world in the future. In Serbia, both the frequency and severity of droughts have increased in recent decades, with most droughts being caused by a lack of precipitation during the warmer months of the year and an increase in evapotranspiration due to higher temperatures. In this study, the frequency and duration of extreme CDDs in the growing season in Serbia were analysed for the past (1950-2019) and the future (2020-2100) period. The Threshold Level Method over precipitation data series was used to analyse CDD events, where extreme CDDs are defined as at least 15 consecutive days without precipitation. In contrast to the original definition of the CDD as the maximum number of consecutive days with precipitation less than 1 mm, here we defined the threshold that is more suitable for agriculture because field crops can experience water stress after 15 days of no rainfall or irrigation. An approach for modelling the stochastic process of extreme CDDs based on the Zelenhasi & cacute;-Todorovi & cacute; (ZT) method was applied in this research. The ZT method was modified by selecting a different distribution function for modelling the durations of the longest CDD events, enabling a more reliable calculation of probabilities of occurrences. According to the results, future droughts in Serbia are likely to be more frequent and severe than those in the past. The duration of the longest CDDs in a growing season will be extended in the future, lasting up to 62 days with a 10-year return period and up to 94 days with a 100-year return period. Results indicate a worsening of drought conditions, especially in the eastern and northern parts of Serbia. The results can help decision-makers adapt agricultural strategies to climate change by providing information on the expected durations of extreme rainless periods in future growing seasons. Although the analysis was performed in Serbia, it can be applied to any other region.

Climate change has a potentially negative impact on the overall vitality of vegetation in both forested and agricultural areas. A comprehensive understanding of the interaction between climate and vegetation across various land cover types holds significant importance from multiple perspectives. This research examined the current state of vegetation trends and their interplay with climate parameters, specifically temperature and precipitation. Additionally, it aimed to provide insights into the anticipated changes in these climate parameters in the future, across the entire area of the Republic of Serbia. The vegetation was observed using the Normalized Difference Vegetation Index (NDVI) obtained from AVHRR/NOAA 11 satellite for the vegetation season (May–October) from 1981 to 2021, while the climate data records used the examination of the relationship between climate indicators and vegetation were monthly mean 2m temperature and precipitation obtained from the ERA5-Land (from April to October). The nonparametric Mann–Kendall test implemented with the Sen's slope estimator and the Pearson correlation coefficient (r) was utilized to identify trends (for the NDVI and climate variables) and the strength of the correlation, respectively. To obtain the information of temperature and precipitation change in future (from 2071 to 2100), the ensemble mean of the eight climate models, for vegetation period and summer season (June–July–August) from the EURO-CORDEX database was used. Results show relatively high NDVI values (> 0.5) over the entire area and the statistically significant (p < 0.005) positive NDVI trend increasing (up to 0.0006 year^-1 )from the north (mainly agriculture cover) to the south (forest cover). In agricultural areas, a positive statistically significant correlation (r = 0.4–0.6, p < 0.005) indicates that the quality of vegetation cover in rainfed agriculture is directly dependent on the amount of precipitation, which serves as the sole source of moisture input. In contrast, the situation differs in forested areas where the correlation between NDVI and precipitation is often statistically not significant (p > 0.005) indicating that forests, because of their characteristics, are less dependent on the amount of precipitation. Regarding temperature, in agricultural areas, there is a positive correlation with NDVI, although it does not reach statistical significance. Conversely, in forested areas, a significant positive correlation is observed between NDVI and temperature which even positively contributes to the development of forest vegetation. In future, the recorded decline in precipitation (a substantial 22.72

As a result of global climate change, the Vojvodina region in northern Serbia is witnessing more frequent extreme weather events. Furthermore, considering the existing trends and future climate change projections, a wide range of impacts are anticipated on the agricultural sector in Vojvodina. The Consecutive Dry Days (CDD) metric, commonly utilized in drought research, serves as an important indicator of drought severity by measuring the duration of dry spells. It is crucial to understand short-term droughts and their impact on agriculture and ecosystems. In this research, the occurrence and length of extreme CDDs during the growing season were studied for both historical (1950-2019) and future (2020-2100) periods in the Vojvodina region. Analyses of past and future occurrences of extreme CDD events were performed for 9 locations. This analysis utilized an ensemble of eight downscaled, bias-corrected regional climate models from the EURO-CORDEX project database, focusing on the RCP8.5 scenario to assess future CDD events. The analysis of CDD events was conducted using the Threshold Level Method on precipitation data, defining extreme CDDs as periods of at least 15 consecutive days without rain. The adapted threshold was chosen as it is more relevant for agriculture, considering that field crops may suffer from water stress after 15 days without rain or irrigation. The research examined various aspects of the stochastic process for CDDs, focusing on the distribution patterns of three key elements: distribution of the number of CDD events, distribution of the duration of CDDs, and distribution of the longest CDD events.To determine if extreme CDDs events act as independent and identically distributed random variables, run tests at a 5% significance level were conducted for all nine locations, utilizing both historical data and the chosen ensemble of eight regional climate models. These run tests confirmed the randomness hypothesis. Additionally, serial correlation coefficients for the series of extreme CDD events were computed, and a significance test at the 5% probability level indicated the independence of these CDDs, revealing no notable serial correlation within the data. The Poisson distribution was used to model the number of extreme CDD events, the exponential distribution function was used to model the distribution of the duration of CDDs, and the Gumbel distribution was selected to model the durations of the longest CDD events. The results indicate an increased likelihood of more frequent and severe droughts in the Vojvodina region in the future, compared to historical data. There is an expected rise in the probability of experiencing 3 to 6 dry periods in the growing season. Moreover, the lengths of the longest CDDs within a growing season are anticipated to extend, reaching up to 57 days for a 10-year return period and 83 days for a 100-year return period. This trend suggests a worsening in drought conditions, particularly in the eastern and northern areas of the Vojvodina region. These insights are valuable for predicting future agricultural drought scenarios, aiding decision-makers in adjusting agricultural practices to mitigate the adverse effects of climate change.

Extreme precipitation events, which are common natural hazards, are expected to increase in frequency due to global warming, leading to various types of floods, including pluvial floods. In this study, we investigated the probabilities of maximum 3-day precipitation amount (Rx3day) occurrences during spring in the Vojvodina region, covering both past (1971–2019) and future (2020–2100) periods. We utilized an ensemble of eight downscaled, bias-corrected regional climate models from the EURO-CORDEX project database, selecting the RCP8.5 scenario to examine future Rx3day amounts. The probabilities of occurrences of Rx3day were modeled using the GEV distribution, while the number of events where Rx3day in spring exceeds specific thresholds was modeled using the Poisson distribution. The results indicate that Rx3day with a ten-year return period during the spring months is expected to increase by 19% to 33%. Additionally, the probabilities of having more than one event where Rx3day exceeds thresholds are projected to rise by 105.6% to 200.0% in the future compared to the historical period. The analysis comparing the design values of Rx3day with future projections for the period 2020–2100 revealed that 51 drainage systems are likely to function without difficulties under future climate conditions. However, for the remaining 235 drainage systems, an increased risk of pluvial flooding was identified, as their design precipitation amounts are lower than the future projections. This study reveals that analyzing extreme rainfall events in the context of climate change yields crucial information that facilitates effective planning and policy making in water management, particularly flood protection.

In the context of global climate change, waterlogging is a severe agricultural meteorological disaster, causing substantial crop yield losses. It results from heavy rainfall, floods, and phreatic rise, leading to excess water accumulation on the surface or seeping through porous soil. In Vojvodina, waterlogging primarily occurs due to significant precipitation in the non-vegetation period or heavy spring rainfall. This study assessed waterlogging hazard in Vojvodina using soil water balance during the non-vegetation period and maximum daily precipitation in spring (March to May). Data from 1971 to 2021 from eight principal meteorological stations were used. Thiessen polygons defined proximal regions around these stations, assigning meteorological parameters across each polygon. Monthly soil water balance was calculated by considering the difference between precipitation and evapotranspiration, using the FAO-56 Penman-Monteith method and taking into account soil water reserves. Soil water reserves were computed based on solum depth and available water, which varies from 10 mm to 150 mm across the area. Next, time series of soil water balance and maximum daily precipitation in early spring were modeled using the Generalized Extreme Value (GEV) distribution function, a common choice in hydrological analyses. The goodness-of-fit for the GEV distribution was tested using the Anderson-Darling test (significance level 0.05). The GEV distributions were fitted to empirical data using the R package 'fitdistrplus', and critical values for the Anderson-Darling test were calculated via Monte Carlo bootstrap simulations using the 'nsRFA' R package. In the Vojvodina region, drainage systems are generally designed based on the relevant amount of excess water with a return period of 10 years. Therefore, for this study, the amount of excess water in the soil water balance during the non-vegetation period, as well as the maximum daily spring precipitation, both with a ten-year return period, were calculated for all analyzed meteorological stations. The waterlogging hazard map was developed by combining the waterlogging hazard assessments based on the amount of excess water during the non-vegetation period with those based on the maximum daily precipitation in spring. It is estimated that the areas facing the highest waterlogging hazards, which constitute about 30% of Vojvodina's territory, are primarily located in the central, eastern, and southeastern parts of the region. The methodology presented for assessing waterlogging hazard provides a clearer understanding of its spatial distribution, enabling the implementation of measures to improve the planning, management, and maintenance of drainage systems, thereby enhancing prevention and mitigation of its negative impacts. This research was conducted within the the European Union’s Horizon Europe innovation action programme under grant agreement No 101094070, project DALIA (Danube Region Water Lighthouse Action) and COST Actions CA22162 – FutureMed and CA22122 RethinkBlue.

In the last decade, small unmanned aerial vehicles (UAVs/drones) have become increasingly popular in the airborne observation of large areas for many purposes, such as the monitoring of agricultural areas, the tracking of wild animals in their natural habitats, and the counting of livestock. Coupled with deep learning, they allow for automatic image processing and recognition. The aim of this work was to detect and count the deer population in northwestern Serbia from such images using deep neural networks, a tedious process that otherwise requires a lot of time and effort. In this paper, we present and compare the performance of several state-of-the-art network architectures, trained on a manually annotated set of images, and use it to predict the presence of objects in the rest of the dataset. We implemented three versions of the You Only Look Once (YOLO) architecture and a Single Shot Multibox Detector (SSD) to detect deer in a dense forest environment and measured their performance based on mean average precision (mAP), precision, recall, and F1 score. Moreover, we also evaluated the models based on their real-time performance. The results showed that the selected models were able to detect deer with a mean average precision of up to 70.45% and a confidence score of up to a 99%. The highest precision was achieved by the fourth version of YOLO with 86%, as well as the highest recall value of 75%. Its compressed version achieved slightly lower results, with 83% mAP in its best case, but it demonstrated four times better real-time performance. The counting function was applied on the best-performing models, providing us with the exact distribution of deer over all images. Yolov4 obtained an error of 8.3% in counting, while Yolov4-tiny mistook 12 deer, which accounted for an error of 7.1%.

Abstract Extreme hydrological events, such as floods and droughts, are becoming more frequent as a result of climate change, leading to negative impacts on various economic sectors. The Pannonian-Carpathian Basin is particularly affected by the increasing frequency of hazardous hydrological events. Agricultural production, which is a highly significant economic sector in the region, is particularly vulnerable to these unfavourable climatic conditions. Changes in precipitation patterns and soil moisture levels can lead to reduced crop yields, while floods can pollute water sources and erode fertile soil. Mapping of Inland Excess Water (IEW), also known as ponding water or waterlogged areas, is crucial for informed decision-making, damage compensation, risk management, and future prevention planning. Remote sensing technology and machine learning have been demonstrated to be valuable tools for the mapping of IEW. The 2014 floods in Southeastern and Central Europe serve as a reminder of the importance of effective flood risk management. This study used a Geographical Information System (GIS) and a Semi-automated Classification Processing (SCP) tool to process high-resolution RapidEye satellite images from the 2014 floods in the Srem region of Serbia. The Spectral Angle Mapping (SAM) classification model was used to produce a map of IEW. The SAM model achieved an overall accuracy of 92.68 %. The study found that IEW affected approximately 2.90 % or 99.59 km² of the territory in Srem. The obtained maps can be used by responsible water management agencies to prevent and control excessive inland water.

Agriculture is one of the largest consumers of water and the importance of its quality need to be usable, because the consequences of applying water of unsuitable quality are permanent and far-reaching. Assessment of groundwater usability should be performed according to available parameters. For the needs of classifications, water parameters were analyzed all cations, anions, total dissolved salt and electrical conductivity. According to all the classifications, the analyzed groundwater can be a good source of water for irrigation in terms of its quality, but with control and appropriate measures.

Conserving clean and safe freshwater is a global challenge, with nitrogen (N) and phosphorus (P) as frequent limiting factors affecting water quality due to eutrophication. This paper provides a critical overview of the spatiotemporal variability in both nutrient concentrations and their total mass ratio (TN:TP) in the canal network of the Hydro system Danube–Tisza–Danube at 21 measuring locations monitored by the Environmental Protection Agency of the Republic of Serbia over a length of almost 1000 km, collected once a month during the last decade. A spatiotemporal variation in nutrient concentrations in the tested surface water samples was confirmed by correlations and cluster analyses. The highest TN concentrations were found in winter and early spring (non-vegetation season), and the highest TP concentrations in the middle of the year (vegetation season). The TN:TP mass ratio as an indicator of the eutrophication pointed out N and P co-limitation (TN:TP 8–24) in 64% of samples, N limitation (TN:TP < 8) was detected in 27% and P limitation (TN:TP > 24) in the remaining 9% of water samples. Such observations indicate slow-flowing, lowland water courses exposed to the effects of non-point and point contamination sources as nutrient runoff from the surrounding farmlands and/or urban and industrial zones, but further investigation is needed for clarification. These results are an important starting point for reducing N and P runoff loads and controlling source pollution to improve water quality and underpin recovery from eutrophication in the studied watershed.

In this paper, the potential impact of the waters of the Banat watercourses of Zlatica, Brzava and Kikindski kanal as potential sources for irrigation is assessed on the basis of monthly water samples from the measuring stations of Markovićevo, Vrbica and Novo Miloševo, for the vegetation period from the year 2007 to 2019. The aim of the research is to get hydrochemical assesment of irrigation water quality from the basic chemical aspect in order to examine the possibility of using these watercourses for irrigation purposes. Most of the samples according to different quality parameters belong to the class of water that is of suitable quality, which indicates the fact that the examined watercourses are good as sources for irrigation.

Summary In this study, wet periods were analyzed using the Standardized Precipitation Index (SPI) on timescales of 3, 6 and 12 months (SPI3, SPI6 and SPI12) at eight stations in Vojvodina Province during the period 1971-2019. The obtained results show that there were very wet years in the observed period, so that regular maintenance of drainage channels is of great importance. Also, in order to illustrate the wet periods for the area of Vojvodina, maps for a timescale of 6 months (SPI6) were made for the vegetation season (April-September) of 2010. By analyzing the maps, it can be concluded that there were four categories of humidity conditions occurring in the observed period – from normal humidity conditions through moderate to very humid and extremely humid conditions.

Wind erosion is a widespread phenomenon causing serious soil degradation. It is estimated that about 28% of the global land area suffers from this process. Global climate changes are expected to accelerate land degradation and significantly affect the intensity of wind erosion. Shelterbelts are linear multi-row planting strips of vegetation (trees or shrubs) established for numerous environmental purposes. Shelterbelts are a specific type of agroforestry system which could reduce soil degradation (soil erosion). Shelterbelts mitigate greenhouse gas through trees storing carbon (C) in their above- and below-ground biomass, wherefore they are highlighted as one of the potential ways to mitigate climate change. The purpose of this chapter is to present wind erosion as a land degradation problem, especially in line with climate changes and the present concept of vegetation establishment in the form of shelterbelts for long-term multi-functional provision of ecosystem services, in particular carbon sequestration.

The objective of this study is to assess the possibility of using unmanned aerial vehicle (UAV) multispectral imagery for rapid monitoring, water stress detection and yield prediction under different sowing periods and irrigation treatments of common bean (Phaseolus vulgaris, L). The study used a two-factorial split-plot design, divided into subplots. There were three sowing periods (plots; I—mid April, II—end of May/beginning of June, III—third decade of June/beginning of July) and three levels of irrigation (subplots; full irrigation (F)—providing 100% of crop evapotranspiration (ETc), deficit irrigation (R)—providing 80% of ETc, and deficit irrigation (S) providing—60% of ETc). Canopy cover (CC), leaf area index (LAI), transpiration (T) and soil moisture (Sm) were monitored in all treatments during the growth period. A multispectral camera was mounted on a drone on seven occasions during two years of research which provided raw multispectral images. The NDVI (Normalized Difference Vegetation Index), MCARI1 (Modified Chlorophyll Absorption in Reflectance Index), NDRE (Normalized Difference Red Edge), GNDVI (Green Normalized Difference Vegetation Index) and Optimized Soil Adjusted Vegetation Index (OSAVI) were computed from the images. The results indicated that NDVI, MCARI1 and GNDVI derived from the UAV are sensitive to water stress in S treatments, while mild water stress among the R treatments could not be detected. The NDVI and MCARI1 of the II-S treatment predicted yields better (r2 = 0.65, y = 4.01 tha−1; r2 = 0.70, y = 4.28 tha−1) than of III-S (r2 = 0.012, y = 3.54 tha−1; r2 = 0.020, y = 3.7 tha−1). The use of NDVI and MCARI will be able to predict common bean yields under deficit irrigation conditions. However, remote sensing methods did not reveal pest invasion, so good yield predictions require observations in the field. Generally, a low-flying UAV proved to be useful for monitoring crop status and predicting yield and water stress in different irrigation regimes and sowing period.

Drainage systems in Serbia are mainly designed to evacuate excess water generated in the winter-spring period, which occurs as a result of snow accumulation during the long and wet winter and its sudden melting with the parallel appearance of spring rains. Dimensioning of the drainage system is done in such a way as to satisfy the needs of draining the design excess water, which is usually calculated using the water balance. Applying statistical analysis based on distributions of probability, the results of the future occurrence of excess water can be predicted. The paper tests the distribution that best corresponds to the empirical distribution of excess water obtained by applying the water balance. The Kolmogorov-Smirnov, Anderson-Darling, and χ2 tests were used to test a number of theoretical distributions, and basis on those tests Generalized Extreme Value (GEV) distribution was selected, which is often used in hydrological analyzes. The probabilities of excess water on drainage systems for the return period of 5, 10, 50, and 100 years were obtained. The results of the calculations can be used in the reconstruction of existing drainage systems, since most of them were designed more than 50 years ago, or in the planning and design of new drainage systems.

The Obedska Bara Special Nature Reserve is one of the oldest protected areas in the world, also enlisted as an Important Bird Area, Ramsar and UNESCO world heritage site. False indigo bush (Amorpha fruticosa L.) represents an invasive alien species which is significantly deteriorating the biodiversity of the Obedska Bara Special Nature Reserve in Serbia. Mapping of A. fruticosa, using an unmanned aerial vehicle and GIS tools, showed that in meadows, pastures, ponds and wetland areas, A. fruticosa covered 85 ha or 11% of the area. However, coverage was uneven, and the most overgrown areas were some meadows (up to 35%), while flooded areas and areas where human impact is significant, as on pastures, were not so affected (1–3%). The most susceptible parts were forest edges. Active management practices, such as cattle grazing and topsoil tarping, and occasional moving, gave positive effects in A. fruticosa, but also other invasive terrestrial plant species spreading control in the reserve. This has also been confirmed by the UAV survey and mapping, which has proven to be an effective method for A. fruticosa monitoring over large areas.