Tissue-Specific Multielement Distribution in Freshwater Fish from Vrana Lake, Croatia: Insights from Advanced Multivariate Classification Techniques

This study examined the distribution of 24 metal(loid)s in muscle, liver, gills, and gonads of dominant fish species from Vrana Lake, Croatia. Element concentrations were determined by inductively coupled plasma mass spectrometry (ICP-MS). The effects of tissue type, species, fish size, and season were evaluated using classical statistics, principal component analysis (PCA), and the Neural Gas (NG) algorithm. Tissue type was the primary determinant of elemental composition. Concentrations of most elements were highest in liver and gills, intermediate in gonads, and lowest and most uniform in muscle. Species-related differences were generally limited, while size-related relationships were tissue- and species-dependent, with generally weak or negative trends in muscle and more variable responses in metabolically active organs. PCA revealed clear tissue-based structuring but comparatively limited separation among species. NG complemented these findings by providing additional resolution of within-tissue heterogeneity and low-frequency multielement profiles. Overall, our findings provide a baseline reference dataset for the biota of Vrana Lake, indicate tissue physiology as the main factor influencing elemental variability, and indicate that NG may serve as a useful complementary approach for exploring complex multielement biomonitoring data.