3D thermohydraulic model for Berlin's Urban Subsurface

Abstract

We present the data and methodology behind Model 2—a 3D thermohydraulic model of the Berlin subsurface that incorporates open boundary conditions to represent regional groundwater flow. This model covers the aquifer systems within the Northeast German Basin and simulates groundwater dynamics over an area of approximately 450 km². It was developed to examine the impact of regional flow on groundwater interactions and potential contamination risks for Berlin’s water supply, Model 2 differs from previous models by integrating cross-boundary flow and pressure influences.

The dataset, in the form of a FEFLOW file (.fem), depicts the 3D distribution of geological units in the subsurface up to a depth of 6 km. It includes thermal and hydraulic boundary conditions as well as the parameter ranges used in Frick et al., 2019. The model focuses particularly on the Rupelian clay layer, which partially separates fresh and saline aquifers. Model results indicate how regional flow affects inter-aquifer exchange, potentially amplifying the risk of saline intrusion into freshwater zones under anthropogenic stress. This dataset is particularly relevant for applications in groundwater management and urban water resource planning. Technical specifications of the model’s structure and parameterization are provided in the Technical Information section and Frick et. al., (2019).This dataset relates to several studies done in the past, focusing on the urban subsurface of Berlin: Frick et. al., (2018, 2019, 2020); Frick (2019), Haacke et al., (2018).

Model Area

in Model CRS (EPSG: 31468, DHDN / 3-Degree Gauss-Kruger zone 4):X: 4571450 to 4624450Y: 5800300 to 5843000

Methods

To develop the 3D Model 2 for groundwater dynamics in Berlin’s subsurface, we used the software FEFLOW© (DHI-WASY) to integrate geological and hydrogeological data across 21 geological layers, which represent both shallow freshwater and deep saline aquifers in the Northeast German Basin. The model construction relied on data from hydrogeological maps, well logs, and prior structural models, particularly the Haacke et al. (2018) model, which provided detailed information on regional aquitards and aquifers. 

To accommodate the open boundary conditions, we applied fixed hydraulic heads and temperatures at the model’s lateral boundaries. The temperatures were used from modeling results presented in Haacke et al., (2018). The hydraulic boundary conditions were derived by projecting the surface hydraulic heads downwards, therefore allowing cross-boundary flow. The upper boundary condition was set to reflect local measured hydraulic heads and surface water features, including lakes and rivers. The bottom of the model represents the depth limit of groundwater flow relevance in Berlin, approximately 6 km below the surface. Here, no flow is expected and hence no hydraulic boundary condition defined. 

Model discretization was performed in FEFLOW©, where we created a finite element mesh with an average horizontal resolution of 20 m × 20 m, and refined locally near production wells and surface water bodies. Vertically, the model was subdivided into 57 computational layers to ensure a better horizontal to vertical ratio of the mesh elements. Hydraulic properties were assigned based on lithology, using anisotropic values where vertical conductivity (κz) was set lower than horizontal conductivity (κx,y), in alignment with the aquifer and aquitard characteristics (see also Frick et al., (2019)).

Transient simulations were initialized by first calculating the model to a steady-state condition, representing pre-pumping groundwater dynamics. For each simulation, pumping wells were introduced as point sinks with variable rates, replicating the distribution and intensity of groundwater extraction in Berlin over a 100-year period. These scenarios allowed us to observe changes in inter-aquifer flow, particularly near regions of thinned Rupelian clay, where saline intrusion into freshwater zones was most probable.

Further methodological details on model parameterization and boundary conditions can be found in Frick et al., (2019).

Technical Information

We provide a single FEFLOW© file which incorporates all elevation distributions of the different geological layers (see table below for key), the hydraulic and thermal boundary conditions and the hydraulic and thermal properties of the model units. Pumping rates and wells were deleted due to the sensitivity of the data. 

3D structural modelling was primarily done in Petrel (© Schlumberger). Modelled unit thicknesses might slightly vary to the ones which can be derived from Frick et al., 2020, due to the fact that a structured mesh was used. Here, a minimum thickness of 0.1m was used in areas where any of the modelled units showed a thickness of 0m.

The model layers are as follows:

ID
Name

1
Colmation Layer  

2
Holocene Weichsel  

3
Holocene Weichsel  

4
Saale  

5
Saale  

6
Holstein  

7
Holstein  

8
Elster  

9
Elster  

10
Miocene  

11
Miocene  

12
Cottbus  

13
Cottbus  

14
Rupelian  

15
Rupelian  

16
Rupelian  

17
Pre-Rupelian  

18
Pre-Rupelian  

19
Pre-Rupelian  

20
Pre-Rupelian  

21
Upper Cretaceous  

22
Upper Cretaceous  

23
Upper Cretaceous  

24
Lower Cretaceous  

25
Lower Cretaceous  

26
Jurassic  

27
Jurassic  

28
Jurassic  

29
Jurassic  

30
Keuper  

31
Keuper  

32
Keuper  

33
Muschelkalk  

34
Muschelkalk  

35
Muschelkalk  

36
Upper Buntsandstein  

37
Upper Buntsandstein  

38
Middle Buntsandstein  

39
Middle Buntsandstein  

40
Lower Buntsandstein  

41
Lower Buntsandstein  

42
Lower Buntsandstein  

43
Zechstein  

44
Zechstein  

45
Zechstein  

46
Zechstein  

47
Zechstein  

48
Rotliegend  

49
Rotliegend  

50
Permocarboniferous  

51
Permocarboniferous  

52
Permocarboniferous  

53
Permocarboniferous  

54
Basement  

55
Basement  

56
Basement  

57
Basement  

Table 1. Layers of the 3D Model

Related works

Frick, M., Bott [Sippel], J., Scheck-Wenderoth, M., Cacace, M., Haacke, N., Schneider, M., 2020. 3D geological model of Berlin - Germany. https://doi.org/10.5880/GFZ.4.5.2020.005

Frick, M., Scheck-Wenderoth, M., Cacace, M., Schneider, M., 2019. Boundary condition control on inter-aquifer flow in the subsurface of Berlin (Germany) – new insights from 3-D numerical modelling, in: Advances in Geosciences. Presented at the European Geosciences Union General Assembly 2019, EGU Division Energy, Resources & Environment (ERE) - EGU General Assembly 2019, Vienna, Austria, 7–12 April 2019, Copernicus GmbH, pp. 9–18. https://doi.org/10.5194/adgeo-49-9-2019

Frick, M., Scheck-Wenderoth, M., Schneider, M., Cacace, M., 2018. Surface to groundwater interactions beneath the city of Berlin - Results from 3D models. Geofluids In Press.

Haacke, N., Frick, M., Scheck-Wenderoth, M., Schneider, M., Cacace, M., 2018. 3-D Simulations of Groundwater Utilization in an Urban Catchment of Berlin, Germany. Advances in Geosciences 45, 177–184. https://doi.org/10.5194/adgeo-45-177-2018

Frick, M., 2019. Towards a more sustainable utilization of the urban geological subsurface: Insights from 3D thermohydraulic models (PhD Thesis). FU Berlin, Berlin.

References

Frick, M., Bott [Sippel], J., Scheck-Wenderoth, M., Cacace, M., Haacke, N., Schneider, M., 2020. 3D geological model of Berlin - Germany. https://doi.org/10.5880/GFZ.4.5.2020.005

Frick, M., Scheck-Wenderoth, M., Cacace, M., Schneider, M., 2019. Boundary condition control on inter-aquifer flow in the subsurface of Berlin (Germany) – new insights from 3-D numerical modelling, in: Advances in Geosciences. Presented at the European Geosciences Union General Assembly 2019, EGU Division Energy, Resources & Environment (ERE) - EGU General Assembly 2019, Vienna, Austria, 7–12 April 2019, Copernicus GmbH, pp. 9–18. https://doi.org/10.5194/adgeo-49-9-2019

Frick, M., Scheck-Wenderoth, M., Schneider, M., Cacace, M., 2018. Surface to groundwater interactions beneath the city of Berlin - Results from 3D models. Geofluids In Press.

Haacke, N., Frick, M., Scheck-Wenderoth, M., Schneider, M., Cacace, M., 2018. 3-D Simulations of Groundwater Utilization in an Urban Catchment of Berlin, Germany. Advances in Geosciences 45, 177–184. https://doi.org/10.5194/adgeo-45-177-2018

Frick, M., 2019. Towards a more sustainable utilization of the urban geological subsurface: Insights from 3D thermohydraulic models (PhD Thesis). FU Berlin, Berlin.

Update

Because it has been requested, we added a FEFLOW file that uses the hydraulic module only. The name is: Berlin_3D_H.fem