Management models for aquatic systems can be used to determine which measures in the watershed or in the water body have been effective and/or which one should be used in future. The newly developed management models presented in the following for Lake Tegel and Schlachtensee are empirical and lake specific. The values for the unknown factors are estimated by an iterative process using optimisation routines and sensitivity analysis methods. The resulting models describe the water and phosphorus balance of each lake. The Lake Tegel water balance model calculates the unknown water inflow from the River Havel depending on the other main in- and outflows with very good validation results. The phosphorus models of both lakes quantify mixing of the upper and lower water body as well as sedimentation and release from the sediment as functions of measured variables. For Lake Tegel, management scenarios were run indicating effective management interventions. For Lake Schlachtensee, the phosphorus model captured the variations in the hypolimnion well but produced poorer results for the epilimnion because of unknown external phosphorus loads. For these the model indicated possible sources and magnitudes.

After installation of phosphorus elimination plants at the inflows of the eutrophic Lake Tegel and Schlachtensee, phosphorus (P) loading declined by a factor of 40 and 100, respectively. This resulted in a pronounced reduction of phosphorus concentrations in the lake water, followed by a decline of chlorophyll-a concentrations. However, for many years P release from sediments due to mineralization and desorption of sedimentary P continued. In Schlachtensee, the presence of nitrate above the sediment suppresses P release, because the Fe/P ratio is sufficient to provide enough aerobic sorption capacity. In Lake Tegel, some P release occurs even under aerobic conditions because of the low aerobic P sorption capacity of the sediments. There, nitrate could moderate the P release peaks which occur when the Fe-P cycle at the sediment water interface is disturbed by precipitation of iron sulfide after reduction of sulfate during times of high mineralization intensity. The potentially mobile P pool in the sediments of both lakes is rather small, indicating that the P release could subside after sufficient reduction of the external P load in Lake Tegel and the disruption of the internal P cycle in Schlachtensee.

Two major points of view prevail in modelling aquatic ecosystems. For practitioners, the key question in modelling aquatic ecosystems is how well models describe system reactions to changes – particularly trophic change and climate change – and in consequence, how useful models are for predicting system responses and for informing water-body management. For researchers, the modelling objective is to investigate functional interactions between ecosystem components, to understand complex causalities and to identify knowledge gaps to close with further lab and field work. Ideally, successful modelling should address both purposes. A need for more intensive communication between modellers of both groups, as well as other model users, was identified at a modellers’ workshop organised by the project “OLIGO” in autumn 2005. OLIGO (2003-2007) was a project of the Berlin Centre of Competence for Water (KWB), executed by researchers of the Federal Environmental Agency (UBA) and the Berliner Wasserbetriebe (BWB) and funded by Veolia Water. Following up on the recommendation of this workshop and in the context of finalising this project, UBA together with KWB organised a 2-day workshop in Berlin, 8-9 November 2007. The workshop was hosted by the Berliner Wasserbetriebe and brought together 60 modellers and model users from 6 countries. The workshop aimed to consolidate and document the state of the art in modelling water quality in reaction to nutrient load or climate change, to discuss how well current lake and reservoir models can support and inform decision-makers, including dealing with the uncertainty of model predictions, and to explore future perspectives and needs for the development of models.

A suite of predictive quantitative models of phosphorus (P) dynamics in Lake Tegel and Schlachtensee has been developed. The results, specific to each lake, are set out below, together with general conclusions about management strategies, and some high priority areas for future research. Lake Tegel: 1. The inflow from the Havel to Lake Tegel has been estimated using both a discrete time step box model and a time integrated numerical model. There is good internal agreement between the 2 estimates of the Havel inflow as a mean fraction of the total inflows (~ 40 %) , as well as with the earlier work of Ripl (1993). The estimated residence times agree closely (~ 70 d). 2. There is considerable inter-annual and inter-seasonal variation in Havel inflows. The numerical model can be used to satisfactorily predict these as a function of the Havel discharge, OWA discharge, and water extraction (bank infiltration and recharge, r²=0.76). 3. Over the past 15 years Lake Tegel has been both a net source (1984-1992;2000-2002), and a net sink (1993 – 1999), for phosphorus. The Havel inflow is the most important component in the P budget of Lake Tegel. When the sediment is a source, the modelled internal P load is 2-4 fold of the OWA annual load. 4. The internal P load can be satisfactorily modelled (r²=0.72) as a function of the external P loads, the water works extractions, and the temperature and nitrate concentration in the hypolinmion. The sensitivity analyses indicate that temperature is the major controlling factor for the P release. The significance of nitrate has to be explored further, and identifying thresholds for parameters which trigger release remains to be done in years 2 and 3. 5.The sediment investigations indicate that the sediment P release is dominated by mineralisation, plus desorption at times of high mineralisation and FeS precipitation. 6. Sediment investigations indicate that artificial oxidation of the sediment surface will only impact on P release when the mineralisation is intense and sulphate reduction is prevented. 7. The internal store of mobilizable P in the sediments is small, the rate of mobilisation is high, and the water residence time is short; thus the internal P load will have no long term effects after the external load is reduced sufficiently (< 5 years, assuming an external load of zero). At present, the external P load is high enough to recharge the sediments. Schlachtensee: 1. The water balance of Schlachtensee can be modelled satisfactorily (r²=0.89) by considering precipitation, storm water discharges and a term to reflect groundwater flows, which yet needs to be validated. 2. Groundwater inflows, as unknown parameter, were determined from modelling by a constant groundwater inflow plus other variable components dependent on precipitation, the level of Schlachtensee, the extraction at Well Rehwiese and of the temperature; this still needs to be cross-checked with a more detailed analysis of groundwater data. 3. The long time development of the P concentration is dominated by the reduced external load from the OWA Beelitzhof. The modelled long term steady state is about 0.02 g P m3 (annual mean). Schlachtensee has been a sink for P since 1985. 4. Next to effects of the reduced external load, the P concentration in Schlachtensee is characterised by peaks occurring in autumn and winter. The cause is not conclusively identified, but is suspected to be due to loading from the steep shoreline, e.g. leaching P from fallen leaves or mobilisation of animal/human excreta deposited in the summer. 5. Modelling shows that in Schlachtensee the epilimnion exerts a dominant effect on the P dynamics. Although P accumulation occurs in the hypolimnion, this is only a small fraction of the total lake P content. P release is controlled mainly by temperature and redox conditions, as well as the hydrological regime. Whether or not thresholds for release can be identified from any of these remains to be investigated. 6.The sediment investigations indicate that the sediment P release is dominated by desorption due to FeS precipitation. 7. The internal store of mobilizable P in the sediments is small, the rate of mobilisation is moderate, and the water residence time is longer than Lake Tegel. Thus, though its contribution to the lake’s P pool is much smaller, the internal P load will continue to exert an effect for longer than in Lake Tegel after the removal of the external load. Assuming the external load to be zero, the mobilizable P-Pool will be released in about 5 years. Both lakes: Chlorophyll-a data is used to depict the reaction of phytoplankton biomass to reduced in-lake TP concentrations. Chlorophyll-a were recalculated without the phaeophytin correction, and investigations for TP thresholds that govern phytoplankton response were begun. TP thresholds in Lake Tegel appear to be higher (around 100 µg/L) than in Schlachtensee (around 30 µg/L). Further data evaluation, including analysis of monthly means and individual sampling dates, is needed. Management implications: 1. Lake Tegel and Schlachtensee have quite different behaviours and require different management strategies. The various models already developed provide a basis for exploring adapted management scenarios. An initial exploration has identified potentially effective strategies. 2. For Lake Tegel the results strongly point to the continuation of the current management strategy to limit the inflow of P rich Havel water into Lake Tegel, i.e. increasing the OWA discharge during summer, when the P concentration in the Havel, and the extraction by the Water Works, are at their highest. 3. As the P release from the sediment in Lake Tegel is mainly driven by the temperature above the lake bottom the stratification stability should be as high as possible.Therefore, operation of the aerators in a fashion to maintain the maximum possible stratification in summer is proving critically important. 4. The model results confirm that for Schlachtensee the P balance is no longer dominated by the inflows from the OWA Beelitzhof, thus any further efforts to reduce P loading will be more effective if concentrated on the other major external sources. 5. The dominant term in the P balance of Schlachtensee appears to be the autumn and winter deliveries, though the actual mode of delivery is still unclear. Identifying the source(s) is an important future research task. 6. Epilimnetic processes are dominant in Schlachtensee and thus no further measures are required to reduce the internal P loading from the sediments to the water column. Future Research Goals 1. Improving the P models for both lakes, for Lake Tegel particularly for the calculation of the internal loads and for Schlachtensee for calculating the external loads, 2. Developing the P models towards management models for both lakes by improving the calculation of the internal loads for Lake Tegel and the external loads for Schlachtensee, 3. Including model components for biological interactions and interfacing them within a transferable P process model to explain the process of trophic recovery, 5 4. Using the improved models for assessing the relative effects of external and internal measures aimed at modifying the P budget, e.g. seasonality of OWA output, aerator operation and seasonal changes in water residence time, 5. Analysing which responses of the lake components are continuous and which show thresholds, and identifying threshold values for the latter; in a second step including other lakes using literature and data provided by other partners, 6. Conducting specifically targeted field investigations to fill gaps, to validate the models and as supplement of monitoring by ILAT in order to uphold the long-term data series, as detailed in the proposal for continuation of the project; 7. Using the outcomes of 1 – 6 for optimised management scenarios for the two Berlin lakes. Together with the evaluation of literature and data from other lakes undergoing trophic recovery, general guidance on managing restoration and predictions of responses to reduced nutrient loading will be developed.