Oligotrophication as a result of planktivorous fish removal with rotenone in the small,eutrophic, Lake Mosvatn,Norway

In September 1987 the shallow, eutrophic, Lake Mosvatn was treated with rotenone to eliminate planktivorous fish (mainly whitefish,Coregonus lavaretus, L.), and the effects were studied. The first summer after treatment the zooplankton community changed markedly from rotifer dominance and few grazers, to a community with few rotifers and many grazers. Accordingly there was a fivefold increase in the biomass ofDaphnia galeata. Adult females ofD. galeata approximately doubled in weight. The decrease in rotifer biomass was probably mainly due to a loss of food by competition with the daphnids. The phytoplankton community was also markedly affected. Prior to treatment Secchi depth was 1.7 m and Chl-a 23μg l⁻¹ in the summer. After treatment there was an increase in the proportion of small and gelatinous algae and the mean chlorophyll concentration fell to 7μg Chl-a l⁻¹. Secchi depth increased to>2.3 m (bottom-sight most of the season). After the treatment there were also fewer cyanobacterial blooms. This seems to be related to oligotrophication caused indirectly by increased grazing by the zooplankton. Total nutrient concentrations were affected. Prior to treatment the mean summer concentration of total phosphate was 44μg P l⁻¹. This decreased to 29μg P l⁻¹ in the first summer and 23μg P l⁻¹ the second summer after the treatment. Total nitrogen decreased from 0.68 mg N l⁻¹ before treatment to 0.32 mg N l⁻¹ the first summer after the treatment. The phosphate loading was not reduced, therefor it can be concluded that the fish removal provided a biomanipulation which caused the more oligotrophic conditions.     

Whole-lake food-web manipulation as a means to study community interactions in a small ecosystem

Whole-lake food-web manipulation was carried out in the hypertrophic Lake Zwemlust (The Netherlands), with the aim of studying the effects on the lake's trophic status and to gain an insight into complex interactions among lake communities. Before manipulation this small (1.5 ha) and shallow (1.5 m) lake was characterized byMicrocystis blooms in summer and high chlorophyll-a concentrations were common (ca. 250 μg 1⁻¹). In March 1987 the planktivorous and benthivorous fish species in the lake were completely removed (ca. 1000 kg ha⁻¹), a new simple fish community (pike and rudd) was introduced and artificial refuges were created. The effects of this manipulation on the light climate, nutrient concentrations, phytoplankton, zooplankton, fish, macrophytes, and macrofauna were monitored during 1987, 1988 and 1989. Community interactions were investigated in phytoplankton bioassays and zooplankton grazing experiments. After the manipulation, despite the still high P and N loads to the lake (ca. 2.2 g P m⁻² y⁻¹ andca. 5.3 g N m⁻² y⁻¹), the phytoplankton density was low (Chl-a<5μg l⁻¹), due to control by large-sized zooplankton in spring and N-limitation in summer and autumn. A marked increase in the abundance of macrophytes and filamentous green algae in 1988 and 1989, as well as N loss due to denitrification, contributed to the N limitation of the phytoplankton. Before manipulation no submerged macro-vegetation was present but in 1988, the second year after manipulation, about 50% of the lake bottom was covered by macrophytes increasing to 80% in 1989. This led to substantial accumulation of both N and P, namely 76% and 73% respectively of the total nutrients in the lake in particulate matter. Undesirable features of the increase in macrophytes were: 1) direct nuisance to swimmers; and, 2) the large scale development of snails, especiallyL. peregra, which may harbour the parasite causing ‘swimmers' itch’. But harvesting of only about 3% of the total macrophyte biomass from the swimmers' area, twice a year, reduced the nuisance for swimmers without adversely affecting the water clarity.     

Structural and grazing responses of zooplankton community to biomanipulation of some Dutch water bodies

Structure and grazing activities of crustacean zooplankton were compared in five lakes undergoing manipulation with several unmanipulated eutrophic (shallow) and mesotrophic (deep) lakes in The Netherlands. The biomanipulated lakes had lesser number of species and their abundance, both of rotifers and crustaceans, and had much larger mean animal size (3–11 μg C ind.⁻¹) than in the unmanipulated eutrophic lakes (0.65 μG C ind.⁻¹). WhereasD. hyalina (=D. galeata) andD. cucullata generally co-occurred in the unmanipulated lakes, in the manipulated lakes bothD. hyalina and other large-bodied daphnids,D. magna,D. pulex (=D. pulicaria), were the important grazers. In the biomanipulated lakes an increase in the individual crustacean size and of zooplankton mass were reflected in a decrease in seston concentration, higher Secchi-disc depth and a marked decrease in the share in phytoplankton biovolume of cyanobacteria. Biomass relationship between seston (150 μm) and zooplankton indicated a Monod type relationship, with an initial part of the curve in which the zooplankton responds linearly to the seston increase up to aboutca. 2 mg C l⁻¹, followed by a saturation of zooplankton mass (0.39 mg C l⁻¹) at 3–4 mg C l⁻¹ seston, and an inhibitory effect on zooplankton mass at seston levels>4 mg C l⁻¹. This latter is related to predominance in the seston of cyanobacteria. In the biomanipulated lakes, the zooplankton grazing rates often exceeded 100% d⁻¹, during the spring, and food levels generally dropped to <0.5 mg C l⁻¹. The computed specific clearance rate (SCR) of zooplankton of 1.9 l mg⁻¹ Zoop C is well within the range of SCR values (1.7–2.2 l mg⁻¹ Zoop C) from deep and mesotrophic waters, but about an order of magnitude higher than in the eutrophic lakes, with the food levels 10-fold higher. For 25% d⁻¹ clearance of lake seston between 35 and 60 ind. l⁻¹ are needed in the biomanipulated lakes against 1200–1300 ind. l⁻¹ in eutrophic lakes. Similarly, about 10 to 15 times more crustacean grazers are required to eliminate the daily primary production in the eutrophic lakes than in the biomanipulated lakes. These numbers are inversely related to the differences in animal size. The corresponding biomass values of zooplankton needed to clear the daily primary production in the eutrophic waters were 0.1–0.2 mg C l⁻¹ in the biomanipulated lakes, but about 0.45 mg C l⁻¹ in the unmanipulated eutrophic waters. Only if the water was kept persistently clear by zooplankton was there a balanced seston budget between the inputvia primary production and elimination by zooplankton. Mostly, however, the input exceeded the assimilatory removal by zooplankton, such that the estimated seston loss could be attributed to sedimentation and mineralization.     

Fish manipulation as a lake restoration tool in shallow,eutrophic temperate lakes 1: cross-analysis of three Danish case-studies

The use of fish manipulation as a tool for lake restoration in eutrophic lakes has been investigated since 1986 in three shallow, eutrophic Danish lakes. The lakes differ with respect to nutrient loading and nutrient levels (130–1000 μg P l⁻¹, 1–6 mg N l⁻¹). A 50% removal of planktivorous fish in the less eutrophic cyanobacteria-diatom dominated Lake V?ng caused marked changes in lower trophic levels, phosphorus concentration and transparency. Only minor changes occurred after a 78% removal of planktivorous fish in eutrophic cyanobacteria dominated Frederiksborg Castle Lake. In the hypertrophic, green algae dominated Lake S?byg?rd a low recruitment of all fish species and a 16% removal of fish biomass created substantial changes in trophic structure, but no decrease in phosphorus concentration. The different response pattern is interpreted as (1) a difference in density and persistence of bloomforming cyanobacteria caused by between-lake variations in nutrient levels and probably also mixing- and flushing rates, (2) a difference in specific loss rates through sedimentation of the algal community prevaling after the fish manipulation, (3) a decreased impact of planktivorous fish with increasing mean depth and (4) a lake specific difference in ability to create a self-increasing reduction in the phosphorus level in the lake water. This in turn seems related to the phosphorus loading.     

Cadmium effects and accumulation in cultures ofProrocentrum micans (dinophyta)

Effects and accumulation of cadmium were studied in unialgal 10-1 batch-culture experiments with the dinoflagellateProrocentrum micans Ehrenberg. Tests were made using sterile filtered North Sea water enriched with nitrate and phosphate only in order to avoid disturbances by complex formation. Cadmium was added to the cultures in amounts of 100 to 0.13μg l⁻¹. In one series it was added at the start of the experiments and in a second one after a growth period of 1 week. Addition of only 1.2μg Cd l⁻¹ reduces multiplication rates and maximum cell densities of the algae. Not until 0.4μg Cd⁻¹ does growth correspond to that of the controls. Cadmium concentrations were measured, after filtration, in the culture medium and in the biomass by means of flameless AAS. The cadmium content in algae increased from 2.7μg g⁻¹ (dry weight) in controls to 500μg g⁻¹ (dry weight) in media containing 100μg Cd l⁻¹. Uptake occurred rapidly during the first few days of the experiments, slowed down somewhat during exponential growth stage, and increased during decay of the cultures. Cadmium content of culture media remained nearly constant (Series 1) or decreased only slowly during experimental time (Series 2). The highest concentration factor was measured in the controls. It decreased with increasing metal concentration in the medium and increased with experimental time. Structural modifications of the cells were visible after Lugol fixation only, indicating brittleness of the cell walls.P. micans has shown to be extremely sensitive to cadmium and to accumulate this metal.     

Lake Muzahi,Rwanda: limnological features and phytoplankton production

Lake Muhazi, a small lake of Rwanda (East Africa) was studied from 1986 to 1990. A dramatic decrease of the catch of Oreochromis niloticus (350 T y⁻¹ in the fifties vs 30 T y⁻¹ in 1982) suggested a loss of productivity or overfishing. In the same period, other ecological changes occurred: the submerged macrophytes regressed and there was a decrease in Secchi depth (0.65 m in 1987 vs 1.5 m in the fifties). Compared to other lakes of the same area, the plankton production seemed low. The results of the present study characterize lake Muhazi as a shallow lake with a rather unstable diurnal stratification and with slight differences in mixing regime between its eastern, deepest part and its western, shallowest part. Secchi disk depth does not vary seasonally to a large extent. The water has a rather high mineral content (conductivity of about 500 μS cm⁻¹ at 25 °C) and low concentrations of dissolved N and P, except in the hypolimnion, where NH _inf4 ^sup+ -N can be high. Two species, Microcystis aeruginosa and Ceratium hirundinella, account for most of the phytoplankton biomass, which is about 50–80 mg chlorophyll a m⁻² in the euphotic zone, usually with little seasonal variation. Daily gross production estimates amount to about 6 to 9.5 g O₂ m⁻² d⁻¹ with a significant difference between the two parts of the lake. Data on C:N and C:P ratio in the phytoplankton suggest that some N deficiency might occur in the eastern part. Moreover, the Zm:Zc ratio could also lead to rather low net production rates (0.21–0.25 d⁻¹ for a mixed layer of 4 m) In conclusion, the primary production of lake Muhazi is medium for African lakes and the hypothesis that decreased planktonic production could account for a reduced fish production should be discarded. Whereas the present yield of the fishery is only 20 kg ha⁻¹ y⁻¹, the yield estimated from primary production ranges between 46 and 64 kg ha⁻¹ y⁻¹. This could be reached through proper management. Finally, some hypotheses are given to explain the ecological changes which occurred in the lake.     

The first biomanipulation conference: a synthesis

At the First Biomanipulation Conference held in Amsterdam (8–11 August 1989), studies presented considered mainly trophic interactions in lakes, enclosures and laboratory systems. Studies on the interactions between phytoplankton and zooplankton emphasized the edibility of the phytoplankton in relation to the zooplankton size structure and the trophic state. Most lake experiments involved 50–100% reduction in fish standing stock or alternatively heavy stocking with piscivorous fish. The most dramatic effects of biomanipulation were found in shallow, eutrophic lakes which exhibited radical changes in ecosystem structure because of changes in light climate and consequently, luxuriant development of macrophytes. There was still much controversy about the top-down effects in relation to trophic state, especially those concerning the role of fish and zooplankton in the development and succession of phytoplankton. Many experiments showed indirect effects within the food web, emphasizing the importance of feedbacks and the complexity of the food web rather than the simplicity of the food chain. The stabilizing effects of refugia for zooplankton and fish on the ecosystem were stressed. Shallow lakes responded generally more rapidly to biomanipulation and this was most successfully accomplished when TP concentration was<50μg l⁻¹, even though in a few cases at 10–20 fold higher TP concentrations (mostly PO₄-P lakes) the results achieved could be maintained for two or more summers. For a guaranteed success of the measures an almost complete removal of fish appeared to be indispensible; moreover in many cases removal of benthivorous fish appeared to be even more important than that of planktivorous fish.     

Phytoplankton ecology of Loch Achray (Scotland)

Loch Achray is a warm monomictic lake in the Scottish highlands with water temperature ranging from 4 °C to 18.7 °C. The oxygen distribution is mainly clinograde with saturation values ranging from 76 to 112%. The general pattern of nutrient variation is irregular particularly for phosphate. Dissolved silica (measured as silicate) showed a range of 24 μg at. Si l⁻¹; phosphate varied from 0.24 to 0.03 μg at. P l⁻¹ while nitrate ranged from 3.5 to 15.2 μg at. N l⁻¹. The pH remained below 7 for the entire year and the alkalinity was very low, ranging from 2.5 to 6 mg CaCO₃ 1⁻¹. The examination of net samples showed Loch Achray to possess a poor phytoplankton population, consisting mainly of desmids and a few diatoms. The quantitative investigation showed a different picture of variation. There was no clear phytoplankton increase till May when a simultaneous increase occurred involving Melosira, Asterionella, Tabellaria fenestrata and Cyclotella. In July the population was dominated by Anabaena, while Cyclotella and Staurastrum became dominant in August.     

Proposals for macrophyte restoration in eutrophic coastal lagoons

Based on the comparison of environmental requirements forRuppia cirrhosa andPotamogeton pectinatus growth, macrophyteversus phytoplankton biomass and production features, and differences in hydrological and nutrient balances between Tancada lagoon (where macrophytes form dense beds) and Enca?izada lagoon (with no macrophytes at all), several proposals for macrophyte restoration are presented. The highest photosynthetic efficiency ofR. cirrhosa takes place at high irradiance and it grows over a wide range of salinity.P. pectinatus is better adapted to lower light intensity and salinity thanRuppia. R. cirrhosa transplanted from Tancada to Enca?izada was successful in enclosures, where light availability increased (μ=0.013 cm⁻¹), but not in open waters where light extinction coefficient was 0.032 cm⁻¹. Phytoplankton biomass (0.11–2.15 g C m⁻²) is much lower than macrophyte biomass (16–200 g C m⁻²) in Tancada lagoon. However, phytoplankton production (165 g C m⁻² yr⁻¹ in Tancada, 480 g C m⁻² yr⁻¹ in Enca?izada) is the same order of magnitude as macrophyte production (244–467 g C m⁻² yr⁻¹). Turnover rates are 0.3–0.9 day⁻¹ for phytoplankton and 1.2–2.5 yr⁻¹ for macrophytes. Phytoplankton and inorganic particles are responsible for high turbidity of the water in Enca?izada lagoon. Phytoplankton blooms in Enca?izada lagoon are supported by high freshwater inflows from rice field drains from May to November. The Q_s (seawater discharge)/Q_F (Freshwater discharge) ratios are, respectively, 0.24 and 0.48, which denotes a higher seawater influence in Tancada than in Enca?izada lagoon. Decreasing freshwater inputs to Enca?izada lagoon both in May and November thus allowing greater inputs of sea water, is proposed as the most effective way to restore this eutrophic coastal lagoon. The objective being to reduce nutrient loadings to the lagoon and phytoplankton in order to favour macrophyte re-colonization.     

Effects of planktivorous fish mass mortality on the plankton community of Lake Mendota,Wisconsin: implications for biomanipulation

Massive mortality of planktivorous fish had a dramatic impact on plankton community dynamics of Lake Mendota, Wisconsin, USA. After fish mortality, the largerDaphnia pulicaria replaced the smallerDaphnia galeata mendotae, resulting in greater grazing pressure on phytoplankton. This was accompanied by a much longer spring clear-water period and lower summer phytoplankton biomass compared to years before the fish mortality. Analysis of historical data (from the mid-1970's) showed that previous fluctuations in planktivorous abundance had similar effects onDaphnia abundance and species composition, and on spring phytoplankton biomass. However, the mid-1970's fish fluctuations had no detectable effect on summer phytoplankton. Concentrations of phosphorus were much higher in the 1970's (spring P 80–135 μg 1⁻¹) than in the 1980's (spring P 19–36 μg 1⁻¹) and it is possible that high P concentrations may reduce trophic cascade effects on summer phytoplankton communities. This suggests that the success of biomanipulation programs may be dependent on lake nutrient status.     

First attempt to apply whole-lake food-web manipulation on a large scale in The Netherlands

Lake Breukeleveen (180 ha, mean depth 1.45 m), a compartment of the eutrophic Loosdrecht lakes system, was selected to study the effects of whole-lake foodweb manipulation on a large scale. In Lake Loosdrecht (dominated by filamentous cyanobacteria), due to water management measures taken from 1970–1984 (sewerage systems, dephosphorization) the external P load has been reduced from 1.2 g m⁻² y⁻¹ to 0.35 g m⁻² y⁻¹. The water transparency (Secchi-depthca. 30 cm), however, has not improved. The aim of the food-web manipulation in Lake Breukeleveen was not only to improve the light climate of the lake, but also to study if the successfull effects observed in small lakes (a few ha) can be upscaled. In March 1989 the standing crop of planktivorous and bentivorous fish populations was reduced by intensive fishery, fromca. 150 kg ha⁻¹ toca. 57 kg ha⁻¹. The lake was made unaccessible to fish migrating from the other lakes and it was stocked with large-sized daphnids and 0⁺ pike. However, water transparency did not increase in the following summer and autumn 1989, which is in contrast with great improvement in the light conditions previously observed in smaller lakes. The main explanations for the negative outcome in Lake Breukeleveen are: 1) the rapid increase of the planktivorous fish biomass and carnivorous cladocerans, predating on the zooplankton community; 2) suppression of the large daphnids by the high concentrations of filamentous cyanobacteria; 3) high turbidity of the lake due to resuspension of bottom material induced by wind, unlike in smaller lakes, and thus inability of submerged macrophytes to develop and to stabilize the ecosystem.     

Total Mercury in Sediments,Macrophytes, and Fish from a Shallow Steppe Lake in Eastern Austria

During summer 2011, samples of sediment, macrophytes, and fish tissues from the shallow, slightly alkaline Lake Neusiedl, Austria, were evaluated for their total Hg content. This is the first report of Hg levels from this lake. Sediments displayed Hg contents between 0.025 and 0.113 μg g^?1 dw (dry weight), significantly correlating with the proportion of organic components pointing to a small anthropogenic impact on the lake's Hg content. Hg Levels in plants and fish were unexpectedly high: both investigated submerged plant species, Potamogeton pectinatus and Myriophyllum spicatum, showed mean values of 0.245±0.152 and 0.298±0.115 μg g^?1 dw, respectively. Biomagnification was evident when comparing muscle samples of the planktivorous fish species rudd Scardinus erythrophthalmus (n=10, mean=0.084 μg g^?1 ww (wet weight)) with the piscivorous perch Perca fluviatilis (n=21, mean=0.184 μg g^?1 ww) or pike‐perch Sander lucioperca (n=9, mean=0.205 μg g^?1 ww). Significantly lower values were found in the muscle of the piscivorous pike Esox lucius (n=25, mean=0.135 μg g^?1 ww), pointing to a specific Hg metabolism of this fish, presumably under the particular physicochemical properties of the lake. Hg Concentrations in fish could pose a risk to piscivorous birds in this protected wetland system.     

Trophic structure, species richness and biodiversity in Danish lakes: changes along a phosphorus gradient

1. Using data from 71, mainly shallow (an average mean depth of 3 m), Danish lakes with contrasting total phosphorus concentrations (summer mean 0.02–1.0 mg P L?l), we describe how species richness, biodiversity and trophic structure change along a total phosphorus (TP) gradient divided into five TP classes (class 1–5: <0.05, 0.05–0.1, 0.1–0.2, 0.2–0.4,> 0.4 mg P L?1).
2. With increasing TP, a significant decline was observed in the species richness of zooplankton and submerged macrophytes, while for fish, phytoplankton and floating‐leaved macrophytes, species richness was unimodally related to TP, all peaking at 0.1–0.4 mg P L?1. The Shannon–Wiener and the Hurlbert probability of inter‐specific encounter (PIE) diversity indices showed significant unimodal relationships to TP for zooplankton, phytoplankton and fish. Mean depth also contributed positively to the relationship for rotifers, phytoplankton and fish.
3. At low nutrient concentrations, piscivorous fish (particularly perch, Perca fluviatilis) were abundant and the biomass ratio of piscivores to plankti‐benthivorous cyprinids was high and the density of cyprinids low. Concurrently, the zooplankton was dominated by large‐bodied forms and the biomass ratio of zooplankton to phytoplankton and the calculated grazing pressure on phytoplankton were high. Phytoplankton biomass was low and submerged macrophyte abundance high.
4. With increasing TP, a major shift occurred in trophic structure. Catches of cyprinids in multiple mesh size gill nets increased 10‐fold from class 1 to class 5 and the weight ratio of piscivores to planktivores decreased from 0.6 in class 1 to 0.10–0.15 in classes 3–5. In addition, the mean body weight of dominant cyprinids (roach, Rutilus rutilus, and bream, Abramis brama) decreased two–threefold. Simultaneously, small cladocerans gradually became more important, and among copepods, a shift occurred from calanoid to cyclopoids. Mean body weight of cladocerans decreased from 5.1 μg in class 1 to 1.5 μg in class 5, and the biomass ratio of zooplankton to phytoplankton from 0.46 in class 1 to 0.08–0.15 in classes 3–5. Conversely, phytoplankton biomass and chlorophyll a increased 15‐fold from class 1 to 5 and submerged macrophytes disappeared from most lakes.
5. The suggestion that fish have a significant structuring role in eutrophic lakes is supported by data from three lakes in which major changes in the abundance of planktivorous fish occurred following fish kill or fish manipulation. In these lakes, studied for 8 years, a reduction in planktivores resulted in a major increase in cladoceran mean size and in the biomass ratio of zooplankton to phytoplankton, while chlorophyll a declined substantially. In comparison, no significant changes were observed in 33 ‘control’ lakes studied during the same period.     

Mesocosm experiments on nutrient and fish effects on shallow lake food webs in a Mediterranean climate

1. Nutrient and fish manipulations in mesocosms were carried out on food‐web interactions in a Mediterranean shallow lake in south‐east Spain. Nutrients controlled biomass of phytoplankton and periphyton, while zooplankton, regulated by planktivorous fish, influenced the relative percentages of the dominant phytoplankton species. 2. Phytoplankton species diversity decreased with increasing nutrient concentration and planktivorous fish density. Cyanobacteria grew well in both turbid and clear‐water states. 3. Planktivorous fish increased concentrations of soluble reactive phosphorus (SRP). Larger zooplankters (mostly Ceriodaphnia and copepods) were significantly reduced when fish were present, whereas rotifers increased, after fish removal of cyclopoid predators and other filter feeders (cladocerans, nauplii). The greatest biomass and diversity of zooplankton was found at intermediate nutrient levels, in mesocosms without fish and in the presence of macrophytes. 4. Water level decrease improved underwater light conditions and favoured macrophyte persistence. Submerged macrophytes (Chara spp.) outcompeted algae up to an experimental nutrient loading equivalent to added concentrations of 0.06 mg L^?1 PO₄‐P and 0.6 mg L^?1 NO₃‐N, above which an exponential increase in periphyton biomass and algal turbidity caused characean biomass to decline. 5. Declining water levels during summer favoured plant‐associated rotifer species and chroococcal cyanobacteria. High densities of chroococcal cyanobacteria were related to intermediate nutrient enrichment and the presence of small zooplankton taxa, while filamentous cyanobacteria were relatively more abundant in fishless mesocosms, in which Crustacea were more abundant, and favoured by dim underwater light. 6. Benthic macroinvertebrates increased significantly at intermediate nutrient levels but there was no relationship with planktivorous fish density. 7. The thresholds of nutrient loading and in‐lake P required to avoid a turbid state and maintain submerged macrophytes were lower than those reported from temperate shallow lakes. Mediterranean shallow lakes may remain turbid with little control of zooplankton on algal biomass, as observed in tropical and subtropical lakes. Nutrient loading control and macrophyte conservation appear to be especially important in these systems to maintain high water quality.     

Reconstructing the past density of planktivorous fish and trophic structure from sedimentary zooplankton fossils: a surface sediment calibration data set from shallow lakes

1. As quantitative information on historical changes in fish community structure is difficult to obtain directly from fish remains in lake sediments, transfer function for planktivorous fish abundance has been developed based on zooplankton remains in surface sediment (upper 1 cm). The transfer function was derived using weighted average regression and calibration against contemporary data on planktivorous fish catch per unit effort (PF-CPUE) in multiple mesh size gill nets. Zooplankton remains were chosen because zooplankton community structure in lakes is highly sensitive to changes in fish predation pressure. The calibration data set consisted of thirty lakes differing in PF-CPUE (range 18–369 fish net^–1), epilimnion total phosphorus (range 0.025–1.28 mg P l^–1) and submerged macrophyte coverage (0–57%). 2. Correlation of log-transformed PF-CPUE, total phosphorus and submerged macrophyte coverage v the percentage abundance in the sediment of the dominant cladocerans and rotifers revealed that the typical pelagic species correlated most highly to PF-CPUE, while the littoral species correlated most highly to submerged macrophyte coverage. Consequently, only pelagic species were taken into consideration when establishing the fish transfer function. 3. Canonical correspondence analysis (CCA) revealed that the pelagic zooplankton assemblage was highly significantly related to PF-CPUE (axis 1), whereas the influence of total phosphorus and submerged macrophyte coverage was insignificant. Predicted PF-CPUE based on weighted average regression without (WA) and with (WA(tol)) downweighting of zooplankton species tolerance correlated significantly with the observed values (r² = 0.64 and 0.60 and RMSE = 0.54 and 0.56, respectively). A marginally better relationship (r² = 0.67) was obtained using WA maximum likelihood estimated optima and tolerance. 4. It is now possible, quantitatively, to reconstruct the historical development in planktivorous fish abundance based on zooplankton fossil records. As good relationships exist between contemporary PF-CPUE data and indicators such as the zooplankton/phytoplankton biomass ratio, Secchi depth and the maximum depth distribution of submerged macrophytes, it is now also possible to derive information on past changes in lake water quality and trophic structure. It will probably prove possible further to improve the transfer function by including other invertebrate remains, e.g. chironomids, Chaoborus, snails, etc., and its scope could be widened by including deeper lakes, more oligotrophic lakes, more acidic lakes and lakes with extensive submerged macrophyte coverage (in the latter case to enable use of the information in the fossil record on plant-associated cladocerans).     

Why biomanipulation can be effective in peaty lakes

The effects of fish stock reduction (biomanipulation) was studied in an 85 ha shallow peaty turbid lake. The lake cleared in a 4-week period in April–May 2004, which demonstrated that biomanipulation can be effective in peaty lakes. We demonstrated that it is possible to reduce the fish stock to <25 kg ha⁻¹ benthivorous fish and <15 kg ha⁻¹ planktivorous fish, sufficiently low to switch the lake from a turbid to a clear state. Knowledge of lake morphology, fish stock, fish behaviour, and a variety of fishing methods was necessary to achieve this goal. It is expected that continuation of fisheries to remove young of the year planktivorous species is needed for several years, until macrophytes provide sufficient cover for zooplankton and can compete with phytoplankton. Cladocerans developed strongly after fish removal. The clearing of the lake coincided with a sudden decrease of filamentous cyanobacteria and suspended detritus, and a strong increase of Bosmina. We assume that Bosmina was able to reduce filamentous prokaryotes and detritus. After the disappearance of the cyanobacteria, Bosmina disappeared too. After the clearing of the lake Daphnia dominated in zooplankton and apparently was able to keep phytoplankton levels low. In our case, wind resuspension did not prevent biomanipulation from being successful. No correlation between windspeed and turbidity was found, neither in an 85 ha nor in a 230 ha shallow peaty lake. Regression analysis showed that on average 50% of the amount of suspended detritus can be explained by resuspension by fish and 50% by phytoplankton decomposition. The main goal of this biomanipulation experiment, clear water and increased submerged plant cover in a shallow peaty lake, was reached.     

The influence of plant-associated filter feeders on phytoplankton biomass: a mesocosm study

Low phytoplankton biomass usually occurs in the presence of submerged macrophytes, possibly because submerged macrophytes enhance top-down control of phytoplankton by offering a refuge for efficient grazers like Daphnia against fish predation. However, other field studies also suggest that submerged macrophytes suppress phytoplankton in the absence of Daphnia. In order to investigate these mechanisms further, we conducted an outdoor mesocosm experiment to study the effect of submerged macrophytes (Elodea nuttallii) on phytoplankton and zooplankton biomass. The experiment combined four nutrient addition levels (0, 10, 100, and 1000 μg P l⁻¹; N/P ratio: 16) with three macrophyte levels (no macrophytes, artificial macrophytes, and real macrophytes). We inoculated the tanks with species-rich inocula of phytoplankton and zooplankton but excluded fish or macro-invertebrates. Probably due to the lack of predators in the mesocosms, potential grazing rates of pelagic zooplankton (estimated from zooplankton biomass) did not differ between the macrophyte treatment combinations. Compared to the treatment combinations without macrophytes, lower phytoplankton biomass occurred in the treatment combinations with real macrophytes at all the nutrient addition levels and in those with artificial macrophytes at all the nutrient levels except the highest. Significantly, higher abundances of plant-associated filter feeders (Simocephalus vetulus and Ceriodaphnia spp.) occurred in the treatment combinations with real and artificial macrophytes. The estimated potential grazing rate of these plant-associated filter feeders indicated that these filter feeders could be responsible for the lower phytoplankton biomass in the presence of real and artificial macrophytes. Our results suggest that the plant-associated filter feeders may be significant grazers in vegetated shallow lakes.