Palaeoenvironmental changes and anthropogenic impact recorded in floodplain sediments: A case study from the lower Morava River Basin (Czech Republic)

Abstract

The Morava River catchment in the Czech Republic serves as sedimentary archive reflecting various natural and man-made processes, and was studied by several authors from the point of view of flood frequency, anthropogenic contamination, or alluvial history. However, more consistent and detailed information on the effect of human activities, namely in terms of land use and river regulation, is missing. The aim of this study is to reconstruct the palaeoenvironmental and anthropogenic processes affecting the floodplain sediments in this area, focusing on the last few hundreds of years. Alluvial sediments from river banks and drilled cores were collected in the Strážnické pomoraví area from sites both inside and outside dams constructed to avoid flooding. Magnetic parameters, reflecting the composition and grain-size distribution of iron oxides, which serve as fingerprints of lithogenic vs. pedogenic vs. anthropogenic origin, are complemented by the radiocarbon and dendrochronological dating. In addition, deposition age was estimated using the 137Cs activity and persistent organic pollutant content. Our results reveal continuous increase of ferrimagnetic minerals (magnetite) input to the floodplain, suggesting increased soil erosion in the catchment. Significantly different pattern was observed inside and outside the flood dykes. The erosion accelerated since the 1950s due to incorrect land use and introduction of modern agriculture techniques. Finally, the industrial pollution significantly contributes to the magnetic enhancement of the topmost 50 cm of the floodplain sequences. Although the results represent local case study, they have more general validity in the sense that records of paleoenvironmental changes and human activities in floodplain sediments may not be well preserved in all the strata, and also their spatial distribution of individual markers may show significant variability.