Andreas Kleeberg

Phosphorus release in Lake Großer Müggelsee and its implications for lake restoration

The shallow, hypertrophic L. Groser Muggelsee, Germany, needs to be converted into a meso- to eutrophic level by means of restoration of the catchment area. Despite efforts in phosphorus (P) loading reduction in the inflow, this polymictic lake still demonstrates an year-to-year fluctuating and strong internal P loading with increasing pore-water phosphate concentrations during summer due to formation of anoxic microlayers at the sediment surface. This P release is indirectly governed by external factors including the supply of dissolved oxygen and nitrate via inflow (high runoff), changes in land use (e.g., decreasing nitrogen fertilization) and climatic factors, and is facilitated by wind-induced rapid changes of stratification and mixing events. Important management implications are: first, declines in nitrate concentrations in the inflow may result in depletion of electron acceptors with respect to phosphate desorption and increase the P release. Second, the P release will continue for a long period and thus delay the recovery of the lake. In dredging the uppermost 1 m sediment layer, the P content of the sediment would decrease from 3.5 to between 0.5 and 1.0 mg P g dry weight–1 and result in reduced capability of internal P loading. Because of the high costs, a decision should be made about dredging or whether the limnological improvement of the lake can be achieved via dephosphorization between 11 and 16 years.

Redox sensitivity of iron in phosphorus binding does not impede lake restoration.

Iron salts have been regarded as unsuitable precipitants for sustainable sedimentary P retention, because Fe-bound P is released at low redox potential. The longevity of an Fe(3+) application (500 g m(-2)) in 1992 was studied in a dimictic lake. Release of Fe and P as well as their co-precipitation were observed dependent on artificial aeration in 2010 and only natural oxygenation in 2011. Sediment core stratigraphy by μX-ray fluorescence analysis revealed that Fe is relocating towards the surface, representing a dynamic P trap with a molar Fe:P ratio of 7. Even at this favourable ratio, P release cannot be suppressed. Settling fluxes of Fe, Mn and P, determined by a multi trap at two day resolution, during aeration and oxygenation, showed that released P can be efficiently precipitated independent of the nature of the oxygen supply. Thus, P release is not relevant for the P supply to the epilimnion, since at overturn most P is co-precipitated by the concurrently hypolimnetically accumulated Fe. To increase the availability of reactive (dithionite extractable) Fe for P binding, our Fe dosage calculation considers Fe in surplus. Beside external and internal P sources to be precipitated in a stoichiometric Fe:P ratio of 5, additional Fe equivalents of 25% for sedimentary organic carbon and to bind soluble sulfides are required. A long-term effect can be achieved only if the external P loading is sufficiently reduced, and Fe is added to ≥ 200 g m(-2).

Interactions between benthic phosphorus release and sulfur cycling in lake scharmützelsee (Germany)

Sulfur (S) conversions were determined during summer stratification in 1995/96 to assess the extent to which benthic release of phosphorus (P) is influenced by the S cycling in eutrophic, dimictic, sulfate-rich (61.33 ± 10.41 mg SO42- l-1) freshwater Lake Schamutzelsee. Hypolimnetic SO42- reduction (4.56 ± 0.73 g (S) m-2 d-1) forming ΣH2S (44.71 ± 17.57 mg ΣH2S m-2 d-1), leading to iron sulfide precipitation (5.62 ± 1.72 mg FeS m-2 d-1) and dissolved iron depletion in the hypolimnion has a major influence on benthic P mobilization and release. The most important inorganic S pool is the CRS (FeS2 + S° + H2S; 15.1 % total S), being 1.3 to 6.6 times higher than the AVS (FeS + H2S) in the uppermost 0 - 8 cm sediment. This diminishes the ability of the sediment to bind P (indicated by 14.6 % loosely bound P (NH4Cl-P) and an exhaustion of the redox-sensitive P (BD-P)), leading to interstitial water P concentrations up to 10.8 mg l-1 and P release rates of 2.64 ± 0.56 mg P m-2 d-1. As a consequence the P content of the lake increased fourfold within 58 days.

Assessment of the long-term effectiveness of sediment dredging to reduce benthic phosphorus release in shallow Lake Müggelsee (Germany)

Two series of laboratory experiments mimicking dredging of the uppermost phosphorus (P) rich sediment layers of the shallow eutrophic L. Müggelsee were carried out to study the extent of P release from deeper sediment layers, and changes in P mobility by means of fractional P composition of the in 10 cm steps ‘dredged’ sediment cores. In the first run over 38 days, the aerated controls reached 55%, and the ‘dredged’ cores from 18.1% (−10 cm) down-core to 0.4% (−50 cm) of the non-aerated control (50.82 mg P m−2 d−1 = 100%). In the second run over one year, the fractional P composition in the revealed sediment layers changed slightly. The water-soluble P (H2O-P) increased for the respective ‘dredged’ horizon by between 1.5 and 5.6% TP. The redox-sensitive P (BD-P) increased in each horizon from the intact core to the situation following ‘dredging’, as well as with depth in each horizon from 4.9% TP (−10 cm) to 11.4% TP (−40 cm). The organic bound P portion (NaOH-NRP) decreased least (1.7% TP) in the uppermost layer and most (15.6% TP) in the deepest horizon exposed to water after ‘dredging’. Based on the changes in P pools following dredging at the future sediment–water interface, it is predictable that dredging without reduction of the external loading may give only temporary improvement followed by a slow return to the present situation.

The Quantification of Sulfate Reduction in Sulfate-Rich Freshwater Lakes - A Means for Predicting the Eutrophication Process of Acidic Mining Lakes?

A comparison of neutral freshwater lakes and acidic coal mining lakes with respect to both, in-lake alkalinity generation and P mobilization, has been made to predict the extent of the possibility of P remobilisation in acidic mining lakes creating eutrophication. It is hypothesized that the maturing process of an acid mining lake is comparable to the recent history of the increasing productivity observed in SO42--rich freshwater lakes. This hypothesis is based on the observation that (1) with rising pH over time the atomic S:Fe ratio in the acidic waters is increasing because only a fraction of the SO42--S but nearly all Fe is usually buried in the sediment; (2) the potential of their sediments to immobilize P is at present linked to its continuous accumulation at the sediment together with the stock of its binding partners, mainly Fe(III) compounds; (3) the input of organic matter stimulating the SO42- reduction and the formation of insoluble complexes of sulfide with ferrous Fe will enhance not only the generation of alkalinity, but will also increase the mobility and release of P.

Phosphorus sedimentation in seasonal anoxic Lake Scharmützel, NE Germany

The seasonal course of phosphorus (P) fractions of sinking particulate matter has been studied at the deepest location of dimictic eutrophic Lake Scharmützel (29.5 m) by paired sediment traps at 9 and 27 m water depth from May 1996 to December 1997. Relatively large depositional fluxes and considerably variable P fluxes, mainly carried by allochthonous particles, diatoms in spring, and iron during overturns transport almost 60% of the average water column P pool to the sediment surface. The contribution of resuspension and sediment focusing (24–34%) is relatively small. A sequential chemical P extraction of the matter entrapped revealed that ,loosely adsorbed` P contributed to 5–14%, organic bound P to 55–68%, and Ca-associated P to 3–6% of the annual P flux, depending on season and depth. The redox-sensitive (iron bound) P ranged from only 12% of the annual P flux during anoxic sulfidic conditions to 28% during oxic conditions. On an annual basis, 16–18% of sedimenting P was recycled within the water column, and 71–75% of total primary P flux was recycled at the sediment surface. Ultimately, 10–23% of P became incorporated into recent sediments indicating the long residence time for P and a high internal resupply of P for primary production.

Quantification of phosphorus entrainment in a lowland river by in situ and laboratory resuspension experiments

Abstract.Hydrodynamics at the sediment water interface initiate and control the exchange of particles and associated phosphorus (P) between river sediment and the water column. Currently, no general analytical theory for cohesive sediment resuspension is available, so these properties must be inferred from direct measurements. In a moderately slow-flowing (annual mean velocity 0.1–0.3 m s-1) stretch of the lowland River Spree, Germany, an in situ experiment (INS) and a laboratory experiment (LAB) with surface sediment (0–3 cm) were conducted concurrently in May 2005 using an erosion chamber. In both approaches, the entrainment of particles and particulate P (PP) increased significantly as shear velocities were incrementally increased from 0.57 to 1.67 cm s-1. In repetitive INS runs, particle and PP entrainment rates at the lowest shear velocity applied differed by a factor of 5 and 11 (0.25 – 1.18 g m-2 h-1 and 4.1 – 45.9 mg m-2 h-1), respectively, due to in situ river bed heterogeneities, specifically the presence or absence of a high-porosity fluff layer. These rates were on average 48 and 3 times higher, respectively, than those of LAB runs (0.01– 0.02 g m-2 h-1 and 3.1– 14 mg m-2 h-1, respectively), suggesting the temporary availability of local, in situ, fluffy surface layers on the river bed that are not preserved during sediment preparation for LAB runs. Such a transient storage feature has not been described previously in the literature for river sediments. The entrainment of this easily resuspendable and P rich material at a low flow-generated shear velocity can lead to batch-wise P burdens in the water column and consequently to a displacement of eutrophication potential. Those events would be underestimated in LAB experiments devoid of this fluffy layer.

Effects of benthic filamentous algae on the sediment-water interface in an acidic mining lake

Natural alkalinity generation by microbial sulphate reduction in acidic lakes is usually inhibited by the low pH and a low primary production which results in a lack of suitable organic carbon sources. In some acidic mining lakes mass developments of filamentous benthic algae occur. The effects of this periphyton layer on the biogeochemistry of the sediment–water interface were investigated by in situ microsensor measurements and laboratory incubations in Mining Lake Grünewalder Lauch (Germany). Microsensor measurements showed that the oxic–anoxic boundary was located in the periphyton layer and was moving up and down depending on light triggered photosynthesis. The sediment itself was permanently anoxic. The diurnal redox shift and the maintenance of neutral conditions in the periphyton layer lead to an effective precipitation of iron and phosphorus. Under the periphyton layer very high sulphate reduction rates up to 265 nmol cm−3 d−1 were measured in the sediment. These are the highest rates reported for mining lakes so far. The microbial activity was high enough to keep the pH in the surface sediment neutral and contributed to natural alkalinity production.

Sedimentary Sulphur:Iron Ratio Indicates Vivianite Occurrence: A Study from Two Contrasting Freshwater Systems

An increasing number of studies constrain the importance of iron for the long-term retention of phosphorus (P) under anoxic conditions, i.e. the formation of reduced iron phosphate minerals such as vivianite (Fe3(PO4)2⋅8H2O). Much remains unknown about vivianite formation, the factors controlling its occurrence, and its relevance for P burial during early sediment diagenesis. To study the occurrence of vivianite and to assess its relevance for P binding, surface sediments of two hydrologically contrasting waters were analysed by heavy-liquid separation and subsequent powder X-ray diffraction. In Lake Arendsee, vivianite was present in deeper sediment horizons and not in the uppermost layers with a sharp transition between vivianite and non-vivianite bearing layers. In contrast, in lowland river Lower Havel vivianite was present in the upper sediment layers and not in deeper horizons with a gradual transition between non-vivianite and vivianite bearing layers. In both waters, vivianite occurrence was accompanied by the presence of pyrite (FeS2). Vivianite formation was favoured by an elevated iron availability through a lower degree of sulphidisation and was present at a molar ratio of total sulphur to reactive iron smaller than 1.1, only. A longer lasting burden of sediments by organic matter, i.e. due to eutrophication, favours the release of sulphides, and the formation of insoluble iron sulphides leading to a lack of available iron and to less or no vivianite formation. This weakening in sedimentary P retention, representing a negative feedback mechanism (P release) in terms of water quality, could be partly compensated by harmless Fe amendments.

Surficial sediment composition as a record of environmental changes in the catchment of shallow Lake Petersdorf, Brandenburg, Germany

A chemical characterization of the surficial sediment (0–20 cm) of type ‘dy’ (org-Cpart/TNpart> 10) of the anthropogenically polluted shallow Lake Petersdorf is presented. Eighty samples were analyzed for a set of parameters, i.e. dry weight, loss on ignition (LOI), total inorganic carbon (TIC), N, S, P, Ca, Si, As, Fe, Al, Mn, Zn, Cd, and Pb. LOI, TIC, N, S, Ca, P, and Zn tend to accumulate at a water depth >2 m (70.6% of the lake surface) in contrast to Mn and Fe which are more widely distributed, and Cd and Pb which are accumulated in the lake part close to a road. The enrichment factors of certain elements, e.g. Al (23.9%), Si (31.5%), P (12.2%), and Ca (68.7%) from 20 cm sediment depth up to the surface, are attributed to incisive changes in the immediate catchment. Erosion, fertilization, amelioration, and separation of the peatland north of the lake by a dam within the last 60 yrs resulted in the change of Lake Petersdorf from a dystrophic to a eutrophic stage. This enhanced the mineralization of its meso-humic (LOI/TNpart=20.6) sediment.