Tracing silicon cycling in the Okavango Delta, a sub-tropical flood-pulse wetland using silicon isotopesTracing silicon cycling in the Okavango Delta, a sub-tropical flood-pulse wetland using silicon isotopes
Faculty of Sciences. Biology
Department of Biology - other
Geochimica et cosmochimica acta. - Oxford
142(2014), p. 132-148
University of Antwerp
Chemical weathering of silicate minerals releases elements into solution whereas the neoformation of secondary minerals works in the opposite direction, potentially confounding estimates of silicate weathering rates. Silicon isotopes (δ30Si) may be a useful tool to investigate these processes. Here, we present 82 δ30Si measurements from surface waters, pore waters, biogenic silica (BSi), clays, sand and vegetation from the Okavango Delta, Botswana, a freshwater sub-tropical, flood-pulse wetland. Hydrologically, the Okavango is dominated by evapotranspiration water losses to the atmosphere. It receives an annual pulse of water that inundates seasonal floodplains, while river baseflow is sufficient to maintain a permanent floodplain. δ30Si in dissolved silica (DSi) in surface waters along a 300 km transect at near-peak flood show a limited range (0.361.19), implying the Delta is well buffered by a balance of processes adding and removing DSi from the surface water. A key control on DSi concentrations is the uptake, production of BSi and recycling of Si by aquatic vegetation, although the net isotopic effect is necessarily small since all BSi re-dissolves on short timescales. In the sediments, BSi δ30Si (n = 30) ranges from −1.49 to +0.31 and during dissolution, residual BSi tends towards higher δ30Si. The data permit a field-based estimate of the fractionation associated with BSi dissolution, ε30BSi-DSi = −0.26, though it is unclear if this is an artefact of the process of dissolution. Clay δ30Si ranges from −0.97 to +0.10, (n = 15, mean = −0.31) and include the highest values yet published, which we speculate may be due to an equilibrium isotope effect during diagenetic transformation of BSi. Two key trends in surface water DSi δ30Si merit further examination: declining δ30Si in an area roughly corresponding to the permanent floodplains despite net DSi removal, and increasing δ30Si in the area corresponding to the seasonal floodplains. We infer that evaporative enrichment of surface waters creates two contrasting regimes. Chemical weathering of low δ30Si phases releases low δ30Si DSi in the relatively dilute waters of the permanent floodplains, whereas silicon removal via clay formation or vegetation uptake is the dominant process in the more enriched, seasonal floodplains.