Water quality research in Loosdrecht lakes: the salient features

Ecosystem research by the working group Water Quality Research Loosdrecht lakes (WQL) was carried out from 1979 to 1990. A coordinated research programme, involving several research institutes and laboratories in The Netherlands, was initiated in 1983, i.e. a year before the reduction of external phosphorus loading by stripping, became effective. The paper summarizes the main results, with emphasis on insight they provide into the lake ecosystems.     

Mathematical modelling as a tool for management in eutrophication control of shallow lakes

The eutrophication model Delwaq-Bloom-Switch is developed to be a functional tool for water management. Therefore it includes nutrients, algal biomass and composition as well as water transparency. A module describing the interaction between water and bottom gives the model the flexibility to deal with measures, such as a decrease of the external phosphorus loading and flushing with water differing in composition from the lake water. This paper focuses on the functional aspects of the model, the results of an application on Lake Veluwe, The Netherlands, and the implications for water management.With one set of coefficients DBS reproduces the most important characteristics of Lake Veluwe for a period of two years before measures (reduction of the external loading and flushing during the winter months) and eight years after the measures. The phosphorus concentration decreased and became growth limiting for algae instead of nitrogen and light. Both in measurements and modelling results, the algal composition changed from blue-green algae dominance to green algae and diatom dominance. Lake Veluwe had a relatively short transient phase after reduction of external loading, because high nitrate concentrations in the flushing water inhibited a long period with high phosphorus releases from the bottom.Model calculations were carried out to investigate the effects of fish stock management and optimization of flushing. Both measures are promising.     

Stochastic modelling of nutrient loading and lake ecosystem response in relation to submerged macrophytes and benthivorous fish

SUMMARY 1. The strong stabilising effect of increased submerged macrophytes (charophytes) and benthivorous fish reduction on the clear water state was shown for shallow Lake Veluwe and Lake Wolderwijd. 2. The first two links in the chain of relationships from external phosphorus (P) loading to in‐lake total‐P concentrations to chlorophyll a concentrations to water transparency, showed a significant correlation with the areal fraction of coverage with charophytes. Higher coverages lead to (i) lower ratios of the in‐lake total‐P concentration compared with the volume weighted average concentration in the inlet water, indicating a higher retention of P in the presence of charophytes (ii) lower chlorophyll a to total‐P ratios, indicating a positive effect of charophytes on top‐down control of algae, and (iii) higher water transparency because of lower algal turbidity. Transparency further improved as a result of benthivorous fish reduction and a significant positive correlation between non‐algal turbidity and benthivorous fish biomass. 3. A model was developed taking into account the inherent variability in precipitation and uncertainties in the empirical relationships determining phosphorus export from stream catchments and other sources and eutrophication variables in the receiving lakes. The model was used to compute (i) probability distributions for in‐lake total‐P, chlorophyll a and Secchi Disc transparency in relation to the coverage with charophytes and benthivorous fish biomass, and (ii) exceedence probabilities with respect to critical values for in‐lake total‐P and water transparency for several management scenarios. 4. The effects of an expected rise in external nutrient loading on the in‐lake total‐P and chlorophyll a concentrations and on water transparency can be compensated for by two proposed control measures: (i) extended treatment at a waste water treatment plant directly discharging into Lake Veluwe, and (ii) diverting the outlet of a stream draining a catchment with high fertilisation. The minimal internal charophyte coverage needed to sufficiently stabilise the clear water state and to meet with the objective of a summer mean water transparency of at least 1 m was estimated at well over 30% of the lake area, while the benthivorous fish stock should be maintained at the present level of c. 20 kg ha^?1.     

An assessment of the importance of emergent and floating-leaved macrophytes to trophic status in the Loosdrecht lakes (The Netherlands)

The potential importance of the six major emergent and floating-leaved macrophyte species in recycling of sediment phosphorus in the Loosdrecht lakes was studied. Representative plant samples were collected at the time of maximum biomass, and analysed for biomass and carbon, nitrogen and phosphorus contents. Species cover was determined by aerial photography.Total cover in the seven lakes studied ranged between 2 and 26 percent. For the four main species, biomass per unit area increased with lake trophic status. Consistent differences in C, N and P contents per unit biomass were not observed. Although cover values were small, significant amounts of C, N and P were contained in the macrophytes when compared with maximum sestonic content.Potential P loads from macrophyte decay were calculated. In Lake Loosdrecht, the P load represented 15 percent of current external P inputs. The potential importance of macrophyte decay to P recycling in the other lakes is greater.Decay of macrophyte species at the end of the growing season appears to affect autumnal nutrient and chlorophyll a levels in the water column of some lakes. The re-establishment of submerged species following lake restoration may increase the importance of this pathway in the lakes.     

Effect of intensive catchment and in-lake restoration procedures on phosphorus concentrations in a eutrophic lake

Lake Okaro is a small, warm monomictic lake in central North Island, New Zealand, which progressed from oligotrophic to eutrophic through the 1960s. Trends in phosphorus (P) concentrations in the lake are linked to multiple restoration efforts over a 5-year period (2003–2008). The restoration procedures include a 2.3 ha constructed wetland established in February 2006 and riparian margin protection to reduce external loading, as well as an Alum application in December 2003 and sediment capping using modified zeolite in September 2008 to reduce internal loading. The annual average total phosphorus (TP) concentration in the lake decreased by 41% from 2004–2005 to 2007–2008. Two predictive models based on external P loading data generally underestimated the measured TP concentrations in the water column due to internal P loading. The relatively rapid response of TP concentrations after reduction of the internal loading using modified zeolite suggests that this technique can effect a rapid decrease in lake water TP concentrations though the trophic state of Lake Okaro showed high resilience to the reduced P loading. It is concluded that the combined effect of all restoration procedures resulted in a relatively rapid decrease in TP concentrations in Lake Okaro, which may be prolonged by continued external load reduction.     

A mathematical model of the phosphorus cycle in Lake Loosdrecht and simulation of additional measures

The phosphorus cycle in the ecosystem of the shallow, hypertrophic Loosdrecht lakes (The Netherlands) was simulated by means of the dynamic eutrophication model PCLOOS. The model comprises three algal groups, zooplankton, fish, detritus, zoobenthos, sediment detritus and some inorganic phosphorus fractions. All organic compartments are modelled in two elements, carbon and phosphorus. Within the model system, the phosphorus cycle is considered as completely closed. Carbon and phosphorus are described independently, so that the dynamics of the P/C ratios can be modelled. The model has been partly calibrated by a method based on Bayesian statistics combined with a Range Check procedure.Simulations were carried out for Lake Loosdrecht for the periods before and after the restoration measures in 1984, which reduced the external phosphorus loading to the lake from ca. 2 mgP m^–2 d^–1 to 1 mgP m^–2 d^–1. The model outcome was largely comparable withthe measured data. Total phosphorus has slowly decreased from an average 130 µgP l^–1 to ca. 80 µgP l^–1, but chlorophyll-a (ca. 150 µg 1^–1, summer-averaged) and seston concentrations (8–15 mgC 1^–1) hardly changed since the restoration measures. About two-thirds of the seston consisted of detritus, while the phytoplankton remained dominated by filamentous cyanobacteria. The P/C ratio of the seston decreased from ca. 1.0% to 0.7%, while the P/C ratios of zooplankton, zoobenthos and fish have remained constant and are much higher. The system showed a delayed response to the decreased phosphorus loading until a new equilibrium was reached in ca. five years. Major reasons for the observed resilience of the lake in responding to the load reduction are the high phosphorus assimilation efficiency of the cyanobacteria and the high internal recycling of phosphorus. A further reduction of nutrient loading, perhaps in combination with additional measures like biomanipulation, will be the most fruitful additional restoration measure.     

The state of the environment of the Loosdrecht lakes

The Loosdrecht lakes are a system of shallow, interconnected, peat lakes in the centre of The Netherlands. The main environmental functions of the Loosdrecht lakes are nature and recreation. From the point of view of the Dutch policy, a Specific Environmental Quality (Bijzondere Milieukwaliteit) should be set for these lakes.The most serious environmental problem of the area is eutrophication. The Loosdrecht lakes have, by increasing external phosphorus loading, changed, from clear lakes with few macrophytes, followed by a period of abundant characean growth, to turbid lakes dominated by cyanobacteria and detrital matter. Eutrophication was counteracted by use of sewerage systems and dephosporization of the supply water. The resultant decrease in external phosphorus loading did not result in a decrease of turbidity by suspended particles.The eutrophication of the lake ecosystems was described as a series of phases. One of those phases, the status around 1940, has been used as an ecological reference system.By means of a graphical presentation technique, the so-called AMOEBE-approach, the state of the environment of the Loosdrecht lakes has been visualized. Thirty-two ecological parameters, including both biotic and abiotic factors, have been selected and quantified. Concrete target values for these parameters have been derived from historical reports and from Lake Western Loenderveen, located close to the Loosdrecht lakes, but less eutrophic.The general conclusion is that the state of the environment of the Loosdrecht lakes is far from what is required with respect to a Specific Environmental Quality, as many of the selected parameters, like water transparency, total phosphorus, mineral nitrogen, cyanobacteria, bream, pike, macrophytes, birds and otter, deviate by over an order of magnitude from their desired levels.     

Water quality research in the Loosdrecht lakes: proposals for the follow-up restoration measures

Reduction of the external phosphorus load in 1984 did not lead to recovery of Loosdrecht lakes. Therefore the Provincial Water Authorities of the Province of Utrecht proposed, after consulting the research group Water Quality Research Loosdrecht lakes (WQL), measures to reduce the present external phosphorus loading from 0.35 to 0.1 g Pm^–2y^–1. Mathematical modelling suggested that such a reduction would benefit the water quality of the lakes. The results of the WQL investigation did not lead to additional measures (dredging, biomanipulation, chemomanipulation).     

Lake restoration by reducing external phosphorus loading: the influence of sediment phosphorus release

SUMMARY. 1. This review considers the factors which determine the recovery of eutrophic lakes following a reduction in the external phosphorus loading.
2 The mean phosphorus content of a lake should decrease roughly in proportion to the reduction in phosphorus input. Where the lake phosphorus concentration does not decrease as predicted, then the release of phosphorus from the sediment is implicated.
3. The current understanding of the processes by which sediment phosphorus is mobilized and transported into the photic zone is described. The extent to which phosphorus release can maintain lake phosphorus concentrations following the reduction in external loading is influenced by: lake morphometry, flushing rate, sediment type, trophic state and history of enrichment.
4. A reduction in the phytoplankton biomass of a lake is dependent upon the size of the decrease in lake phosphorus concentration and the degree to which phosphorus limits primary production. The importance of phosphorus in limiting phytoplankton production tends to decrease with increasing lake trophic status.
5. Improvements in the condition of highly eutrophic lakes require very large reductions in external phosphorus loading, whereas in mildly enriched lakes moderate changes in the supply of phosphorus have noticeable effects on phytoplankton biomass.     

The spatio-temporal variations of sedimentary phosphorus in Taihu Lake and the implications for internal loading change and recent eutrophication

Internal phosphorus loading is particularly concerned for the shallow lakes due to the frequent sediment disturbance, which may play a vital role in changing nutrient level in overlying water. A historical perspective on internal phosphorus loading may contribute to understanding its contribution to recent eutrophication. In this work, a study on the changes in internal phosphorus loading and release potential in Taihu Lake, a shallow eutrophic lake in China, was performed based on the analysis of spatio-temporal variations of sedimentary total phosphorus and three operationally defined fractions (NaOH-P, HCl-P and OP). The influencing factors for changing internal loading were discussed. The results showed that internal phosphorus loading was elevated compared to pre-eutrophication periods and the increase has occurred since approximately the late 1970s to early 1980s. Changes in internal phosphorus loading were primarily attributed to the NaOH-P and OP fractions, relating to anthropogenic inputs and enhanced productivity, respectively. The internal phosphorus release potential may be enhanced by up to 22% currently relative to the pre-eutrophication period; however, it should play a secondary role to external input in enhancing nutrient levels and sustaining the eutrophication in Taihu Lake.     

湖泊富营养化治理的生态工程

1996年对长春南湖的富营养化实施了生治理工作,调查结果表明,通过收获水生高等植物和鱼产品带出湖体的P量分别为149．6和189．9kg,通过蚌体生长固定的P量为153.4kg,三者合计492.9kg,与湖体会年P输入量大体持平,生态工程运转后,水质明显好转,湖水中的总P浓度逐年下降,浮游植物个体密度减小,种类数增加,生态工程是城市湖泊富营养化治理较为理想的方法。     

Wind-induced increases in trophic state characteristics of a large (27 km 2), shallow (1.5 m mean depth) Florida lake

Nutrient and chlorophyll concentrations in Lake Newnan (27 km², mean depth 1.5 m), Florida showed dramatic increases from 1991 to 1998. Historical data showed Lake Newnan never had sufficient aquatic macrophyte abundance for a shift in alternate stable states to account for increases in trophic state characteristics. External phosphorus and nitrogen loads from incoming streams were monitored from August 1997 to July 1998 to determine if external supplies of nutrients were responsible for increases in lake nutrient and chlorophyll concentrations. During the study period, external nutrient loading rates were not correlated to lake nutrient concentrations. Phosphorus and nitrogen models based on the external loading estimates predicted the lake total phosphorus and total nitrogen concentrations to be 370% and 680% less, respectively, than the observed lake total phosphorus and total nitrogen mean concentrations. Consequently, phosphorus and nitrogen exports were 280% and 540% greater, respectively, than stream input loading. Data during the study period revealed strong inverse relations between lake stage and total phosphorus (r=–0.78), total nitrogen (r=–0.71), and chlorophyll (r=–0.90) concentrations. Long-term data (1965–1998) also revealed inverse correlations (r=–0.48 to –0.52) between lake stage and total phosphorus, total nitrogen, and chlorophyll concentrations. Applying fundamental wave theory and using a bathymetric map, it is probable that as much as 70% of the lake bottom sediment could be subjected to resuspension 50% of the time when the lake stage falls below 19.9 m mean sea level (msl). Above a lake stage of 19.90 m msl, less than 20% of the lake bottom sediment can potentially be resuspended 50% of the time. A percent frequency distribution from 1991 to 1998 showed that over 30% of the lake stages fell below 19.9 m msl. However, from 1967 to 1990, only 8% of the lake stage values fell below 19.9 m msl. Increases in total phosphorus, total nitrogen and chlorophyll concentrations in Lake Newnan were likely caused by an increased probability of internal loading due to decreased lake levels, and not to external loading of phosphorus and nitrogen.     

A first step in modelling plankton growth in the Loosdrecht Lakes

In this paper a dynamic model is presented to simulate plankton growth in the Loosdrecht Lakes. The present, first step version of the model, contains only one algal super-species and incorporates fairly standard formulations for the effects of light irradiance, the extinction coefficient of the water, temperature and the phosphate levels on algal growth. Grazing by zooplankton and detritus dynamics (non-living particulate organic material) are considered essential parts of the carbon cycle. The total phosphorus concentration is used as a timevariable forcing function, which transforms the problem of predicting future water quality levels into the problem of predicting future total phosphorus levels. This requires a separate estimation procedure. However, both types of estimation should be and can be integrated into one single water quality model.The Loosdrecht Lakes seem P-limited in summer. Re-adjustment of certain constants is necessary to simulate Lake Breukeleveen, which differs from the Loosdrecht Lakes in certain characteristics. It is hypothesized that some form of phosphorus cell-quota dynamics may solve this problem. Furthermore, it is discussed as to what extent the hydrological interactions between the Loosdrecht Lakes and Lake Breukeleveen may have caused the differences observed and how this matter could be evaluated through simulation.     

Response of a eutrophic, shallow subtropical lake to reduced nutrient loading

1. Lake Apopka (FL, U.S.A.) was subjected to decades of high nutrient loading from farms developed in the 1940s on converted riparian wetlands. Consequences included perennially high densities of cyanobacteria, low water transparency, elimination of submerged vegetation, modified fish community, and deposition of nutrient‐rich, flocculent sediments. 2. Initial steps were taken to reduce phosphorus (P) loading. Through strengthened regulation and purchase of farms for restoration, external P loading was reduced on average from 0.56 to 0.25 g P m^?2 year^?1 (55%) starting in 1993. The P loading target for the lake is 0.13 g P m^?2 year^?1. 3. For the first 6 years of P loading reduction the annual sedimentation coefficient (σ) averaged 13% less than the prior long‐term value (0.97 versus 1.11 year^?1). The sedimentation coefficient, σ, was lower in the last 3 years of the study, but this period included extreme low‐water conditions and may not be representative. Annual σ was negative (net P flux to the water column) only 1 year. 4. Wind velocity explained 43% of the variation in σ during the period before reductions in total phosphorus (TP) concentration of lake water, but this proportion dropped to 6% after TP reductions. 5. Annual mean TP concentrations differed considerably from values predicted from external loading and hydraulic retention time using the Vollenweider–Organization for Economic Co‐operation and Development relationship. Reductions in lake water TP concentration fit model predictions better when multiyear (3‐year) mean values were used. 6. Evidence available to date indicates that this shallow, eutrophic lake responded to the decrease in external P loading. Neither recycling of sediment P nor wind‐driven resuspension of sediments prevented improvements in water quality. Reductions in TP concentration were evident about two TP‐resident times (2 × 0.9 year) after programmes began to reduce P loading. Improvements in concentrations of chlorophyll a and total suspended solids as well as in Secchi transparency lagged changes in lake‐water TP concentration but reached similar magnitudes during the study.     

Rehabilitation and management of a moderately deep-stratifying reservoir by the use of nutrient reduction and food-web management

As expected from its short renewal time, the hypertrophic Lingese Reservoir responded rapidly but incompletely to external nutrient reduction in 1993. Although there was a sudden decline in lake phosphorus and chlorophyll concentrations, the ‘turbid regime’ revealed resistance in this weakly stratified reservoir of intermediate depth. Draining the reservoir in autumn 1995 provided an unprecedented opportunity for sediment treatment—the phosphorus-rich surface layer was inverted and covered with deeper nutrient-poor layers—and removal of the dense cyprinid fish stock which had not responded to sewage diversion, at least in the short term. Commencing with refilling in 1999, a new fish stock was built up from 2000 by only stocking predators (fingerlings of pike, pike-perch and larger specimens of rainbow-trout in the first years) in combination with catch restrictions. Concomitantly, with the appearance of daphnids in 1999, a ‘clear water regime’ was established and lake water phosphorus concentrations decreased at unchanged external loading. Reduced zooplanktivory as well as reduced fish-mediated phosphorus release from the sediments were driving mechanisms behind successful reduction of internal loading and achievement of a ‘clear water regime’. Hence, phosphorus concentrations were revealed to be a response variable not only to input management but also to food-web management. As the development of cyprinid dominance was prevented in the long-term, there is ample evidence that the fish community responded to the applied management measures as expected featuring successful food-web management. Overall, the biological structure was revealed to be of major importance for lake phosphorus availability and turbidity as in shallow lakes, without, however, the establishment of macrophyte dominance.     

Predicting concentration of total phosphorus and chlorophyll a in a lake with short hydraulic residence time

The relationship between total phosphorus and chlorophyll a concentration was determined for Skinner Lake, Indiana over an annual cycle in 1978–79. Total nitrogen:total phosphorus ratios in the epilimnion ranged from 19 to 220 suggesting a phosphorus-dependent algal yield in the epilimnion. Approximately 90% of annual TP loading reached the lake via streamflow, and 93% of this entered during snowmelt and spring-overturn periods. At that time incoming water flushed the lake 2.4 times. Atmospheric loading accounted for 1.4% of annual TP load. Internal hypolimnetic TP loading occurred during summer stratification. Mean [chl a] for the ice-free period was 15.15 mg m^–3, within the range expected for eutrophic lakes.The 1978–79 data were used in conjuction with the Vollenweider & Kerekes (1980) model to produce a model specific for the Skinner Lake system. The model predicted mean epilimnetic total phosphorus and chlorophyll a concentrations from mean total phosphorus concentration in inlet streams and from lake water residence time during the period of spring overturn and summer stratification. The Skinner-specific model was tested in 1982 and it closely predicted observed mean epilimnetic [TP] and [chl a] during the ice-free period. This study shows that variability in lake models which average data over an annual period can be reduced by considering lake-specific seasonal variation in hydrology and external TP loading.     

Application of SWITCH,a model for sediment-water exchange of nutrients,to Lake Veluwe in The Netherlands

The formulations of SWITCH, a model for prediction of nutrient fluxes across the sediment-water interface, are presented. Results of the application to data on the sediment of Lake Veluwe are presented and discussed.SWITCH calculates the thicknesses of the aerobic and denitrifying layers on the basis of a step-wise steady state approach. The concentrations of detritus, ammonium, nitrate and phosphate in the sediments and the pore water are simulated dynamically using mass balance equations.Analysis of the data for Lake Veluwe show large spatial heterogeneity. This presents a major drawback for the calibration of SWITCH, which focused on the silty part of the lake. The results show that the model simulates realistically and consistently layer thicknesses, concentrations and mass fluxes connected with the transport and conversion processes. The model appears to have potential for describing both seasonal patterns and developments on the long term.SWITCH calculates strongly increased phosphate return fluxes, following total reduction of the top sediments. An important hypothesis in the model is that phosphate precipitated in reduced sediment layers is transferred to the oxidized layer and dissolves instantaneously. This results in a decrease of the phosphorus content of the sediment, but also maintains high release rates of phosphorus after the reduction of the external phosphorus loading of Lake Veluwe. Model results and mass balance studies for the overlying water indicate that the removal of phosphorus to deeper sediment layers is underestimated or that dilution of the sediments occurs as the result of sedimentation.     

Coupling of phytoplankton and detritus in a shallow,eutrophic lake (Lake Loosdrecht,The Netherlands)

An oscillating steady state is described of phytoplankton, dominated by Prochlorothrix hollandica and Oscillatoria limnetica, and sestonic detritus in shallow, eutrophic Lake Loosdrecht (The Netherlands). A steady-state model for the coupling of the phytoplankton and detritus is discussed in relation to field and experimental data on phytoplankton growth and decomposition. According to model predictions, the phytoplankton to detritus ratio decreases hyperbolically at increasing phytoplankton growth rate and is independent of a lake's trophic state. The seston in L. Loosdrecht contains more detritus than phytoplankton as will apply to many other lakes. The model provides a basis for estimating the loss rate of the detritus, including decomposition, sedimentation and hydraulic loss. In a shallow lake like L. Loosdrecht detritus will continue to influence the water quality for years.     

Modelling phosphorus fluxes in the hypertrophic Loosdrecht Lakes

A dynamic, deterministic model is presented to simulate the phosphorus cycle and plankton growth in the shallow, hypertrophic Loosdrecht Lakes (The Netherlands) before and after restoration measures. Besides inorganic phosphorus (SRP) in both the surface water and the interstitial water, the model comprises three algal groups, zooplankton, fish, detritus, zoobenthos and upper sediment (all modelled both in carbon and in phosphorus). Within the model system, the phosphorus cycle is completely closed. Carbon and phosphorus are described independently, so that the dynamics of the P/C ratios can be modelled. Sediment processes are described in a simplified form.Simulated values are largely within the range of observed ones. The detrital fraction of the seston (=phytoplankton+detritus) varies from 50–60% in summer to about 90% in winter. SRP in the surface water is very low during most of the year. Sensitivity for external phosphorus input is larger for algal and detrital P than for algal and detrital C and chlorophyll-a. So the P/C ratio of the seston decreases following restoration measures, as is observed in the lakes, while the much higher P/C ratios of zooplankton and fish remain constant. Phosphorus mobilisation from the sediment decreases with decreasing external input. Adaptation of the model system to the reduced loading takes place within about two years.Sources of uncertainty in the model include the limited knowledge on selective grazing as well as on mortality and mineralisation processes.