Pollutant dispersal and stability in a severely polluted floodplain: A case study in the Litavka River, Czech Republic

Abstract

The fluvial system of the Litavka River in the Czech Republic has been severely polluted by polymetallic ore mining and smelting that occurred mainly between the 1780s and 1970s. To decipher the mechanisms of the pollution transport pathways, we analysed river valley sediments from river headwaters upstream from the ore district through mining and smelting areas to downstream sites. We sampled recently inundated areas as well as sites just outside 100-year (Q100) inundation. In the river valley, it was necessary to distinguish anthropogenic alluvium (AA) floodplain sediments produced due to ore mining and processing. AA had changed K/Rb signal ratios (measured by energy dispersive X-ray fluorescence, ED XRF); Pb and Zn pollution; and 206Pb/207Pb isotope ratios. The main part of the primary pollution of the river system was deposited in 19th century and during the first half of the 20th century, a few kilometres far downstream from the primary pollution sources (waste deposits of ore mining processing and the smelter). AA has a Zn/Pb ratio of ~1.5 and a local enrichment factor (LEF) of up to ~160 for Pb and up to ~130 for Zn. The floodplain further downstream has received diluted primary pollution with the same Zn/Pb ratio and a LEF of up to ~80 for Pb and up to ~50 for Zn. Much less severe pollution is recognised in fallout-polluted soils at the river valley edges outside Q100 (Zn/Pb ~0.9, LEF ~7 for Pb and LEF ~2 for Zn). The secondary pollution currently exported from the watershed and floodplain is substantially enriched in Zn (Zn/Pb ~1.5–6). That is obviously a consequence of the higher mobility of Zn in the temporary reservoirs in the mining and smelting area and floodplain. We demonstrate that simple element ratios obtained by convenient laboratory XRF spectrometry can considerably help in deciphering the complex structure of floodplain fill as well as pollution transport routes because that method allows us to adequately process a large number of samples to characterise the complexity of the pollutant distribution in floodplains.