Qualitative interpretation of physico-chemical and isotopic parameters in the Krka River (Croatia) assessed by multivariate statistical analysis

Abstract

Multivariate statistical analyses were applied on the measured physico-chemical (Cd, Pb, Cu, Zn, Mg, Ca, O2, alkalinity, temperature, pH, SAS, DOC and DIC) and isotopic parameters (δ13C and δ18O) to estimate and distinguish anthropogenic from natural influences to the water system of the Krka River. Analyses were conducted on the data collected during six years from twelve sampling sites. On the basis of orientation, positioning and grouping of parameters arranged by biplots, four main hypotheses were defined and finally statistically confirmed. Thereof, two main and distinct processes occurring in the Krka River could be highlighted: (i) upstream pollution, caused by the inflow of untreated waste-waters of city of Knin and (ii) downstream self-purification, caused by the sedimentation and/or co-precipitation of pollutants coupled by the inflow of clean subterranean water (groundwater recharge). Grouping of (i) hydrological and carbon cycle connected parameters, and (ii) anthropogenically influenced correlated parameters were proposed as a result of statistical analysis. Regarding the pH, it is shown that a stream section influenced by the subterranean inflow of Zrmanja River is statistically significantly different for all sampling campaigns during six years, being lower for about 0.5 pH unit.