This thesis presents the methodology for determining die-off coefficient of faecal indicator bacteria (Escherichia coli and enterococci) in terms of seepage through the karst environment. The main problem in exploring karst environment, which this methodology strives to cope with, is lack of field measurements, poor data on karst rock formation, and fractures and channels within it, and groundwater level dynamics. The analysed karst catchment Bokanjac-Poličnik is situated in the hinterland of the city of Zadar and extends to 235.07 km2. In water supply wells within analysed catchment a frequent occurrence of faecal indicator bacteria was observed. The appearance of Escherichia coli and enterococci at the water supply wells of the karst aquifers Bokanjac-Poličnik is an indicator of the existence of faecal pollution that is very dangerous for the environment and people living in this area. During the period from 1/1/2012 to 19/10/2016 the EC/ENT monitoring was carried out in the supply wells. Besides concentration, in the implementation of the methodology the data of groundwater levels at the wells, the pumping rates and precipitation were used. The proposed methodology consists of three basic steps. Preliminary analyses as the initial step were use in concretisation of the certain assumptions related to the detection of increased concentrations of EC/ENT as well as in determination of the potential source of contamination. Time series analysis, Integral Pumping Test and quantification of flow (KarstMod model) were elementary parts of the preliminary analysis. Information from preliminary analysis were used in the development of the analytical model, the main part of the proposed methodology. The analytical model was constructed with the aim of resolving processes of sorption and die-off and determining the dominant factor in the process of natural removal of faecal indicator bacteria in terms of seepage through the karst environment. Within the model, two parts of the pollutant transport are integrated. The first part refers to percolation from septic tank to the saturated zone, and the second part refers to the horizontal flow and transport within the saturated zone (inverse procedure). Initial concentration at the septic tank C0 and final concentration in the water supply wells CB4,JEZ were the input dana in the model. The calculation yields pollutant concentration at the contact of vertical and horizontal part of the flow (CVAD and CSAT). The aim of the calculation was to obtain a set of optimization parameters-coefficients of the model that achieve a minimum difference in the concentrations ΔC = CVAD - CSAT. Reduction of pollutant concentration in the vertical section was associated with the process of natural die-off (ki) and sorption (k), while in the horizontal phase, due to the more pronounced rate of seepage, the effect of sorption was eliminated. The calculation was carried out for each of the suspicious sources of contamination and assumed seepage directions, with variation of the parameter-coefficient values in the transport model (percolation velocity vSAT, total die-off coefficient ktot = ki + k, hydraulic conductivity K and retardation coefficient R). Estimated die-off coefficient for both seepage phases were later used as an input in the numerical flow and transport model (MODFLOW). The numerical model gained insight into the progress of the pollution cloud, but also checked the accuracy of the applied methodology by comparing the measured and model pollutant concentration. The strategy used for creating the numerical flow and transport model was Single Continuum Porous Equivalent model.