Abstract/Summary

Global climate change is associated with significant changes to short-term weather extremes as well as long-term weather characteristics in different regions. Whilst the magnitude of climate changes are extremely uncertain, it is likely that summers will become warmer and drier, and there will be an increase in the intensity of rainfall events. These intense rainfall events will lead to an increased number of flooding events that remain for longer periods of time, and the occasional inundation of land that has rarely been flooded in the past. There is the possibility that increased flooding intensity and frequency will influence the soil properties, which in turn may affect the behaviour and mobilisation of potentially toxic elements (PTEs) in floodplain soils. It likely that many floodplains downstream of urban catchments, particularly those catchments with a history of industrial development, may harbour a legacy of contaminants that have been deposited with floodplain sediments. To investigate the impact of fluvial flooding on PTE mobility in floodplain soils, I used the Loddon Meadow floodplain site; situated adjacent to the River Loddon in the Southeast of England, as a model floodplain, typical of a lowland floodplain downstream of an urban catchment. Preliminary work characterised the floodplain topography using geospatial techniques and compared elevation with the spatial distribution of soil PTEs concentrations. The topography of the floodplain was found to influence the deposition of some PTEs (e.g. Cr, Cu, Ni and Zn), providing strong evidence that the source of these PTEs to this floodplain site originated from point source or diffuse pollution upstream in the urban catchment. The novel combination of geospatial mapping of elevation and geochemical analyses could be adopted as a method for determining the source of PTEs to other study sites. Analysing soil pore water chemistry provides the more useful measurement of the mobile fraction of PTEs, rather than the total concentration bound to the solid fraction. There are a number of methods for extracting pore water from soil samples; we compared an example of an in-situ method (RhizonTM sampler) with an example of an ex-situ method (centrifugation). There were no significant differences found in the pore water chemistry, despite the centrifugation exerting a pressure on the soil sample orders of magnitude higher than the RhizonTM sampler. We found, however, that in terms of useability through a range of soil moistures and consistency of sample volume extracted, the centrifuge was the preferred method for this particular study. We highlight examples where the opposite conclusion might be reached. Laboratory mesocosm studies have reported increased PTE mobilisation with artificial flooding events. However, it can be difficult to extrapolate these finding due to the controlled conditions of the laboratory set-up (e.g. room temperatures are often higher than found in the field). We found that there was a need for on-site experiments that consider the effects of flooding using real-time field observations. We therefore took a field-based approach; extracting soil pore waters, by centrifugation, from the Loddon Meadow floodplain pre-flood, during a flood and post-flood. We found that the flooding event did not influence the mobility of all of the PTEs in the same way. However, we found concentrations of Cd, Cu and Cr significantly decreased post-flood compared to pre-flood. The dominant process identified to explain this decrease was precipitation with sulphides, which occurred during the flood and subsequently resulted in the significant decrease in concentrations post-flood. A slight increase in pH may have aided adsorption processes onto organic matter and clay minerals. We also found a decrease in dissolved organic matter in solution and this would have reduced the capability of the pore water to complex PTEs in solution. It is possible that the decreased concentrations found were a result of dilution, due to the increased water volume from the river and ground water. When analysed, the river and ground water had considerably lower concentrations of PTEs than the soil pore waters. The impact of a flooding event on PTEs mobility is the combination of multiple processes. So, while we observed some processes increasing the concentrations of PTEs; for example, the reductive dissolution of Mn oxides, predominantly in the lower elevation areas of the floodplain. The overall net effect of the flooding event was a decrease in PTE concentrations, because processes like sulphide precipitation were dominant. There were no significant increases in PTEs mobility due to the flooding event and as such, no evidence to support the idea that floodplains become a source of PTEs. This is contrary to the evidence from laboratory studies, that found there is mobilisation of PTEs due to flooding. This study highlights the importance of understanding the dominant processes that drive the mobility of individual PTEs on specific floodplains, so that site-specific predictions can be made on the impact of future flooding on the mobilisation of legacy contaminants. Further field-based monitoring; collecting data pre-flood, during the flood and post-flood, from varying soil types and composition (e.g. clay, sand, silt, peat and loam) is required to support future modelling exercises. This would improve our capability to predict the impact of increased intensity and duration of flooding on soil porewater chemistry and PTE mobility.