Water table forecasting plays an important role in the management of groundwater resources in agricultural regions where there are drainage systems in river valleys. The results presented in this paper pertain to an area along the left bank of the Danube River, in the Province of Vojvodina, which is the northern part of Serbia. Two soft computing techniques were used in this research: an adaptive neurofuzzy inference system (ANFIS) and an artificial neural network (ANN) model for one-month water table forecasts at several wells located at different distances from the river. The results suggest that both these techniques represent useful tools for modeling hydrological processes in agriculture, with similar computing and memory capabilities, such that they constitute an exceptionally good numerical framework for generating high-quality models.

No abstract

In suburban Belgrade, there are some 200 local water supply systems which are not connected to either the Belgrade Water Supply System or to supply systems operated by municipal utilities. The small systems in Belgrade suburbs are either operated by local municipality (local government) or even by the group of local citizens who have neither technical capability nor financial resources to do it properly. Roughly 200,000 of Belgrade's inhabitants obtain their drinking water from these water supply systems. The water quality delivered by these local water supply systems is often compromised in terms of microbiological, physical and/or chemical compliance with drinking water standards in addition to the general lack of strategy on water safety plans and risk assessment. WHO Guidelines on water quality standards as well as the recommendations on safety plans and whole risk assessment are strictly respected in the main (central) Water Supply System in Belgrade. Most frequently, elevated concentrations of ammonia, nitrites, nitrates and iron lead to lack of chemical compliance, while elevated counts of aerobic mesophilic bacteria and the presence of bacteria indicative of faecal pollution tend to be behind microbiological lack of compliance with drinking water standards. In most cases, failure to meet drinking water standards can be attributed to groundwater pollution. No sewer system exists in these areas, and wastewater from septic tanks, in practice infiltration wells, is in direct contact with groundwater. Of a total of 72 laboratory-tested drinking water samples, 51.3% failed to meet physical and/or chemical standards, and 73.6% failed to meet microbiological standards. Groundwater pollution can only be prevented if wastewater disposal system is provided for all households and all suburban residential areas which obtain their water supply from local water supply systems. Some possible mitigation measures have been indicated. In the interim period, water must be disinfected continually, and the feasibility of ozonation or UV irradiation, in addition to chlorination, should be assessed.

The aim of this paper is to examine the water regime of chernozem under maize crops in the last half century (1966-2019) and to determine whether during that period and to what extent, there was an increase or decrease in maize irrigation requirements. The mathematical plant model FAO CROPWAT 8.0 was used for the calculation. The calculation was performed on the basis of monthly values of reference evapotranspiration (ETo) for the period 1966-2019, calculated by the modified Hargreaves method, daily values of precipitation from the meteorological station Surcin, data on the selected plant, which are in accordance with FAO56. The soil is chernozem on the Zemun les terrace. The analysis was performed by dividing the research period into three subperiods: the first twenty (1966-1985), the second twenty (1986-2005) and the last fourteen (2006-2019) years. It was found that the average values of potential evapotranspiration of maize were increasing, starting from the first (500mm) to the third (562mm) subperiod, while the average values of actual evapotranspiration, as well as the average amount of effective precipitation in the vegetation period of maize, decreased. Consequently, the average water deficit, i.e. maize irrigation requirements was increased by 56%, starting from the first (205mm) to the third (319mm) subperiod of the research. The increase in the water deficit also caused an increase in the projected reduction in maize yield related to its genetic capasity, which averaged 31% in the first subperiod and 47% in the third. Analysis of the results on a monthly and decadal level showed that maize irrigation requirements lasts from June to August, with a maximum in the second decade of July. In all three summer months, an increase in maize irrigation requirements was registered from the first to the third subperiod of the research, with the maximum increase during July. The conducted research, which generally gives an insight into the state of the water regime of Zemun chernozem in the last half century, shows that the conditions of maize production in the natural rain regime are deteriorating and that the irrigation requirement is increasing.

Normalized Difference Vegetation Index (NDVI) is an indicator of vegetation health and land cover changes, based on the reflectance of certain ranges in the electromagnetic spectrum. Land use, seasons and climate changes affect spatial variations in NDVI values. This study focuses on the basins of the rivers Tinja and Kozlica, located on the Eastern parts of the Maljen Mountain, and characterized by the dominant presence of grassy vegetation. Spatial and temporal changes in plant water supply are monitored using 10-meter Sentinel-2 imagery, and further processed on a monthly basis in QGIS for 2020-2021. For better elaboration of NDVI values basins of these two rivers were delineated into 305 sub-basins, on which further analysis was performed. NDVI data during both years range from < 0.1 - > 0.6. NDVI values change during different seasons, which is consistent with the increase and decrease of water stress during the studied period, which refers to changes in weather conditions during the growing season. In the summer months, the highest values exceed 0.6, and in some cases even 0.8. NDVI values in October and November decrease to 0.3 and 0.5, while in winter months NDVI values are <0.1. NDVI values are higher, and less variable, in sub-basins with woody, partially coniferous vegetation. This study contributes to increasing knowledge about the potential application of remote sensing as well as highresolution Sentinel-2 imagery for monitoring plant water supply because the assessment of drought impact on plant production requires the current monitoring of plant water regime. GIS tools enable the delineation of sub-catchments, which helps to better monitor the spatial variation of NDVI within natural landscape entities. NDVI and other indices are easy to calculate, and therefore, Sentinel-2 can play an important role in future drought early warning systemsand in determining conditions of the vegetation cover.