Simulating Bank Filtrate Dynamics in Berlin - Decision Support Under Climate and Water Use Changes
Baldwin, Dwight; Haacke, Nasrin; Sprenger, Christoph; Wicke, Daniel; Monninkhoff, Bertram; Gnirss, Regina
- Publisher
- Zenodo
- Language
- en
Abstract
Effective decision-making in urban water management requires integrating outputs from specialized models. Berlin’s drinking water supply relies on induced bank filtration and managed aquifer recharge from the Spree and Havel rivers. However, river inflows into Berlin are declining -e.g., in summer 2019, the Spree’s inflow was half that of an average dry summer year- and are expected to decrease further over the next decade due to the ending of coal sump water discharge into the Spree. Long-term impacts from climate change are anticipated to exacerbate this trend. Additionally, an analysis of streamflow data and bank filtrate rate-corrected groundwater extraction has identified regions where maximum monthly extractions from drinking water wells already exceed the lowest monthly river flows in Berlin. This imbalance, combined with increasing water demand driven by population growth, leads to a higher proportion of treated wastewater in Berlin’s streams. As a result, risks to drinking water quality intensify, and the complexity and costs of water and wastewater treatment escalate. Furthermore, higher extraction levels are associated with increased bank filtrate fractions, amplifying system stress and emphasizing the need for sustainable water management practices. In collaboration with the Belin Waterworks (Berliner Wasserbetriebe), we applied a well-calibrated FEFLOW© model of the Berlin-Friedrichshagen waterworks to simulate bank filtrate rates under various recharge and groundwater extraction scenarios. The model was run under three historical well configurations (2010, 2015, and 2019) and then well pumping rates were adjusted in the same relative configuration under three groundwater recharge scenarios. A review of prior investigations revealed groups of well galleries exhibiting similar changes in bank filtrate fractions in response to extraction levels; our results complement these former investigations. Bank filtrate behavior across well galleries was found to depend on several factors, including well depth, distance to the riverbanks, the presence of opposing riverbanks, and regional groundwater heads. Relating bank filtrate change groups to site characteristics and bank filtrate fractions in other Berlin develops a city-wide understanding of changes in bank filtrate. Future FEFLOW© modeling scenarios, including commissioning and decommissioning of well galleries, and implementing managed aquifer recharge will be essential to address remaining uncertainties. Outputs from this modeling effort contribute to regional dynamic water balance modeling for Berlin’s semi-closed water cycle in order to support sustainable water management decision-making amid evolving climatic and regulatory challenges.
Full text
climate-impetus.eu This project has received funding from the European Union’s Horizon 2020 research and innovation programme under grant agreement No 101037084. Funded in the EU Horizon 2020 Green Deal call •Lower groundwater recharge increases the bank filtrate share, while higher recharge reduces it. •Numerical instabilities at extreme recharge–extraction setups highlight model limitations and the need for cautious interpretation. •Next steps should focus on scaling the analysis to all Berlin galleries and exploring the role of managed aquifer recharge (MAR) as a stabilising measure. •The study identified six characteristic types of bank filtrate contribution, which are influenced by factors such as: → Distance to the river bank → Surrounding surface water availability → Well depth →Regional groundwater levels Simulating Bank Filtrate Dynamics in Berlin Decision Support Under Climate and Water Use Changes CHALLENGE & CONTEXT •A calibrated groundwater flow model (FEFLOW), provided by Berliner Wasserbetriebe (BWB), was used to simulate groundwater dynamics within the model area. •A series of scenarios were developed by varying: → Groundwater recharge: 107 mm (baseline), 87 mm (−19%), 70 mm (−35%), and 54 mm (−50%, extreme case) → Pumping rates: increased by up to 30% relative to 2019 abstraction values •Pathlines were simulated to delineate the catchment areas of different well galleries under each scenario. •This approach allows assessing how changes in recharge and abstraction impact the spatial extent and composition of well catchments. Fig.1: (a) Map of Berlin showing the city outline, surface water bodies (blue), and the model domain (grey) used for simulating bank filtration processes. (b) Simulated flow pathlines within the model domain delineate the catchment areas of different well galleries (colored). The example shown corresponds to a total extraction of 60 Mio m³/yr, based on 2019 abstraction data. The bar chart indicates the share of total extraction per gallery. Fig. 3: Change in bank filtrate fraction across scenarios with differing groundwater recharge rates. SUMMARY & CONCLUSION METHOD & RESULTS Dwight Baldwin1, Nasrin Haacke1, Christoph Sprenger1, Daniel Wicke1, Bertram Monninkhoff2, Regina Gnirss2 1. Kompetenzzentrum Wasser Berlin gGmbH 2. Berliner Wasserbetriebe Fig. 2: Relationship between total extraction and bank filtrate share, based on a combination of literature data and model results. The distribution allows for the identification of six distinct classes, each representing a different range of bank filtrate contribution. 107 mm / year 87 mm / year 70 mm / year Bank filtrate share [%] Type 1 SW > 90% Type 3 SW = 50-90% 3,000 < Q < 50,000 m3/d SW [%] = -58.3 + 13.5 × LN(Q) Type 2 SW = 50-100% 2,000 < Q < 30,000 m3/d SW [%] = -79.3 + 17.4 × LN(Q) Type 4 SW = 0-60% 500 < Q < 50,000 m3/d SW [%] = -70.34 + 12 × LN(Q) Type 5 SW = 0-40 % 4,000 < Q < 60,000 m3/d SW [%] = -77.4 + 10.4 × LN(Q) Type 6 SW = 0-20 % 20,000 < Q < 70,000 m3/d SW [%] = -151.8 + 15.4 × LN(Q) •Six distinct bank filtrate types were identified based on the relationship between extraction rates and filtrate share. •Higher groundwater recharge scenarios led to a lower share of bank filtrate, while reduced recharge increased the relative contribution of bank filtrate to total extraction. •Model results for the 70 mm recharge scenario were implausible, likely due to numerical instability under extreme combinations of low recharge and high extraction. GOAL •Quantify the share of bank filtrate in a Berlin catchment across different well galleries. •Assess how changes in groundwater recharge and water use scenarios affect the share of bank filtrate. •Support adaptive management to ensure a resilient drinking water supply. • Berlin’s drinking water relies on bank filtration and managed aquifer recharge, covering ~70% of supply in a semi-closed water cycle. •River inflows may decline due to climate change and land use changes (e.g. lignite phase-out) •Longer dry spells and more intense rainfall events reduce natural groundwater recharge, stressing water availability. •The resulting impact on the share of bank filtrate remains uncertain. ab
