A meta-analysis of water quality and aquatic macrophyte responses in 18 lakes treated with lanthanum modified bentonite (Phoslock(®)).

Lanthanum (La) modified bentonite is being increasingly used as a geo-engineering tool for the control of phosphorus (P) release from lake bed sediments to overlying waters. However, little is known about its effectiveness in controlling P across a wide range of lake conditions or of its potential to promote rapid ecological recovery. We combined data from 18 treated lakes to examine the lake population responses in the 24 months following La-bentonite application (range of La-bentonite loads: 1.4-6.7 tonnes ha(-1)) in concentrations of surface water total phosphorus (TP; data available from 15 lakes), soluble reactive phosphorus (SRP; 14 lakes), and chlorophyll a (15 lakes), and in Secchi disk depths (15 lakes), aquatic macrophyte species numbers (6 lakes) and aquatic macrophyte maximum colonisation depths (4 lakes) across the treated lakes. Data availability varied across the lakes and variables, and in general monitoring was more frequent closer to the application dates. Median annual TP concentrations decreased significantly across the lakes, following the La-bentonite applications (from 0.08 mg L(-1) in the 24 months pre-application to 0.03 mg L(-1) in the 24 months post-application), particularly in autumn (0.08 mg L(-1) to 0.03 mg L(-1)) and winter (0.08 mg L(-1) to 0.02 mg L(-1)). Significant decreases in SRP concentrations over annual (0.019 mg L(-1) to 0.005 mg L(-1)), summer (0.018 mg L(-1) to 0.004 mg L(-1)), autumn (0.019 mg L(-1) to 0.005 mg L(-1)) and winter (0.033 mg L(-1) to 0.005 mg L(-1)) periods were also reported. P concentrations following La-bentonite application varied across the lakes and were correlated positively with dissolved organic carbon concentrations. Relatively weak, but significant responses were reported for summer chlorophyll a concentrations and Secchi disk depths following La-bentonite applications, the 75th percentile values decreasing from 119 µg L(-1) to 74 µg L(-1) and increasing from 398 cm to 506 cm, respectively. Aquatic macrophyte species numbers and maximum colonisation depths increased following La-bentonite application from a median of 5.5 species to 7.0 species and a median of 1.8 m to 2.5 m, respectively. The aquatic macrophyte responses varied significantly between lakes. La-bentonite application resulted in a general improvement in water quality leading to an improvement in the aquatic macrophyte community within 24 months. However, because, the responses were highly site-specific, we stress the need for comprehensive pre- and post-application assessments of processes driving ecological structure and function in candidate lakes to inform future use of this and similar products.

Lake responses following lanthanum-modified bentonite clay (Phoslock®) application: an analysis of water column lanthanum data from 16 case study lakes.

Phoslock(®) is a lanthanum (La) modified bentonite clay that is being increasingly used as a geo-engineering tool for the control of legacy phosphorus (P) release from lake bed sediments to overlying waters. This study investigates the potential for negative ecological impacts from elevated La concentrations associated with the use of Phoslock(®) across 16 case study lakes. Impact-recovery trajectories associated with total lanthanum (TLa) and filterable La (FLa) concentrations in surface and bottom waters were quantified over a period of up to 60 months following Phoslock(®) application. Both surface and bottom water TLa and FLa concentrations were <0.001 mg L(-1) in all lakes prior to the application of Phoslock(®). The effects of Phoslock(®) application were evident in the post-application maximum TLa and FLa concentrations reported for surface waters between 0.026 mg L(-1)-2.30 mg L(-1) and 0.002 mg L(-1) to 0.14 mg L(-1), respectively. Results of generalised additive modelling indicated that recovery trajectories for TLa and FLa in surface and bottom waters in lakes were represented by 2nd order decay relationships, with time, and that recovery reached an end-point between 3 and 12 months post-application. Recovery in bottom water was slower (11-12 months) than surface waters (3-8 months), most probably as a result of variation in physicochemical conditions of the receiving waters and associated effects on product settling rates and processes relating to the disturbance of bed sediments. CHEAQS PRO modelling was also undertaken on 11 of the treated lakes in order to predict concentrations of La(3+) ions and the potential for negative ecological impacts. This modelling indicated that the concentrations of La(3+) ions will be very low (<0.0004 mg L(-1)) in lakes of moderately low to high alkalinity (>0.8 mEq L(-1)), but higher (up to 0.12 mg L(-1)) in lakes characterised by very low alkalinity. The effects of elevated La(3+) concentrations following Phoslock(®) applications in lakes of very low alkalinity requires further evaluation. The implications for the use of Phoslock(®) in eutrophication management are discussed.

Assessing the mode of action of Phoslock® in the control of phosphorus release from the bed sediments in a shallow lake (Loch Flemington, UK).

Phoslock(®) is increasingly used worldwide to control sediment phosphorus (P) release and cyanobacterial blooms despite the fact that little is known about its mode of action in lake bed sediments. This study quantified the effects of Phoslock(®) on sediment elemental composition and P fractionation (one year pre- and post-application of 170 g Phoslock(®) m(-2)) in an attempt to address these knowledge gaps. Post-application, sediment La content was significantly higher in the top 10 cm of the sediment compared to pre-application conditions. Mass balance calculations indicated that the applied mass of La had the potential to bind 25% of potentially release-sensitive P (Pmobile; sum 'labile P', 'reductant-soluble P' and 'organic P' fraction) present in the top 4 cm or 10% of Pmobile present in the top 10 cm of the sediment. Assessing variation in sediment P partitioning indicated that the application caused a significant increase in the mass of P present in the more refractory 'apatite bound P' fraction between post-application month 4 and 7 compared to Pmobile. This suggests that Phoslock(®) controls sediment P release by increasing the mass of P permanently bound in the sediment. To address uncertainty in estimating product dose required to control sediment P release we conducted laboratory assays using intact sediment cores to which we added serial additions of Phoslock(®) under either aerobic or anaerobic conditions. The laboratory experiment indicated that the original dose to Loch Flemington was sufficient to control sediment P release under aerobic conditions but that significant P release will occur should prolonged anaerobic conditions persist. However, Phoslock(®) may be a viable option to control sediment P-release under anaerobic conditions which would require an estimated additional application of up to 510 g Phoslock(®) m(-2). A conceptual model is proposed for the use of P-capping agents in lake remediation projects which is likely to increase cost-effectiveness and reduce non-target effects by applying multiple smaller doses compared to a single high dose.

Comparison of phosphorus (P) removal properties of materials proposed for the control of sediment P release in UK lakes.

Of growing interest in the control of sediment phosphorus (P) release in lakes is the use of solid phase P-sorbing products (PSPs) including industrial by-products and naturally occurring or modified mineral complexes. However, there is a need to report on novel PSPs proposed by suppliers for use in lake remediation projects at the national scale. We comparatively assessed the elemental composition and P sorption properties of six industrial waste-products (waste-products from treatment of abandoned mine waters - 'red ochre' and 'black ochre'; waste products from building practices: 'gypsum', 'sander dust', 'mag dust' and 'vermiculite') and one commercially available modified lanthanum (La) bentonite product (Phoslock®), all of which have been proposed for use in remediation projects in UK lakes. P sorption was well described (r(2)>0.70) by Langmuir isotherms for all products with the exception of 'gypsum' for which no significant P sorption was reported. P sorption capacities ranged from 4 mgPg(-1) dry weight (d.w.) PSP ('black ochre') to 63 mgg(-1) d.w. ('sander dust'), with products composed mainly of calcium oxide (CaO) and silicon dioxide (SiO(2)) (i.e. 'vermiculite', 'mag dust' and 'sander dust' PSPs) having significantly higher sorption capacities than all other PSPs. Estimates of the equilibrium P concentration (EPC(0)) from Langmuir isotherms indicated that all PSPs, with the exception of 'gypsum', were capable of reducing soluble reactive P (SRP) concentrations to <5.1 µgL(-1). Further research and development required to underpin regulatory policy decisions regarding the use of PSPs is discussed. Dose estimates for each PSP required to treat a eutrophic shallow loch (40.6 ha) with persistent internal loading issues are reported.

Sediment amendment with Phoslock® in Clatto Reservoir (Dundee, UK): Investigating changes in sediment elemental composition and phosphorus fractionation.

Lanthanum-modified bentonite clay (Phoslock(®) is a lake remediation tool designed to strip dissolved phosphorus (P) from the water column and increase the sediment P-sorption capacity. This study investigated short term alterations in sediment elemental composition and sediment P-fractions based on sediment cores taken 2 days before and 28 days following the application of 24 t of Phoslock® to a 9 ha, man-made reservoir. Following the application, sediment lanthanum (La) content increased significantly (p < 0.05; n = 4) in the top 8 cm of the sediment, thereby theoretically increasing sediment P-binding capacity on the whole reservoir scale by 250 kg. Mass balance calculations were used to estimate the theoretical binding of release-sensitive P (P(mobile); sum of 'labile P', 'reductant-soluble P' and 'organic P' fraction) by La across the top 4 cm and 10 cm depth of sediment. The amended mass of La in the sediment had the potential to bind 42% of P(mobile) present in the top 4 cm or 17% of P(mobile) present in the top 10 cm. However, with the exception of a significant increase (p<0.05; n=4) in the 'residual P' fraction in the top 2 cm, sediment P-fractions, including P(mobile,) did not differ significantly following the Phoslock® application. Experimental P-adsorption studies indicated P-saturation values for Phoslock® of 21,670 mg P kg⁻¹ Phoslock®. Sequential extraction of P from saturated Phoslock® under laboratory conditions indicated that around 21% of P bound by Phoslock® was release-sensitive, while around 79% of bound P was unlikely to be released under reducing or common pH (5-9) conditions in shallow lakes. Applying Phoslock® is, therefore, likely to increase the P-sorption capacity of sediments under reducing conditions.