FEFLOW | Boundary Condition Constraints

MODELING TASK DESCRIPTION

Boundary conditions (BC) are applied to groundwater models to fix the hydraulic head, concentration values, fluxes or water transfers between surface waters and aquifers. In some cases, however, the user might want to activate boundary conditions only under certain conditions.

For example, if the the concentration of inflowing water is defined through a mass-concentration BC, outflowing water should be able to leave the domain with whatever concentration value has been calculated during the simulation. To do so, the mass-concentration BC needs to be switched off for outflowing water. Another example for the application of constraints is river infiltration to an aquifer with variable groundwater levels by means of a fluid-transfer BC. When groundwater levels drop below the river bed, this BC needs to be deactivated to prevent artificial infiltration to the system from the river. 

SOLUTION

Boundary condition constraints (BCC) are available to limit boundary conditions in FEFLOW. Such limitations can be applied to any type of boundary condition, as they can be added manually from the context menu of the each BC in the Data Panel. 

In the first example mentioned above, the user could assign a fixed concentration boundary of 0 mg/l on either side of the model to define the inflow of freshwater in the domain - Fig. 1 (upper sketch). If a contamination source inside the model domain adds contaminant mass, the boundary condition on the outflow side would artificially fix the water concentration on the outflow boundary to 0 mg/l, which is conceptually wrong. Combining the BC with a minimum mass flow constraint of 0 g/d, the model activates the BC for positive flows (inflows) only and switches it off otherwise - Fig. 1 (lower sketch). This means that water leaving the domain on the outflow side can have any concentration that has been calculated, eventually removing mass from the model. 
 

Fig. 1 – Effect of applying only boundary conditions for mass flow simulations or a combination of boundary conditions and boundary condition constraints.  

Please note the different sign conditions for boundary conditions and boundary condition constraints - Fig. 2. Boundary condition constraints are assigned positive ranges (e.g. min flow constraint of 0) to specify the boundary conditions for any inflows, and negative for outflows.

Fig. 2 – Opposite algebraic sign convention for boundary conditions and boundary condition constraints. 

Another example for the application of boundary conditions constraints is is the deactivation of fluid-transfer BC for low groundwater heads. The infiltration of river water is calculated based on the current difference between the river water level and the actual hydraulic head in the aquifer Δh = h_ref - h_gw. While this is a solid approach for groundwater levels higher than the river bottom, it leads to overestimated infiltration for groundwater levels lower than the river bed. In reality, the unsaturated area below the river limits the amount of leakage from the river and therefore, the Fluid-transfer BC needs to be switched off.

As a solution, the Fluid-transfer BC can be accompanied with a minimum head constraint h_min [m] defining the lowest possible river water level, i.e., the river bed elevation. If such min. head constraint is applied, the BC will be deactivated as soon as the water level recedes below the minimum head constraint, effectively limiting the the maximum infiltration proportional to the difference (h_ref - h_min)- Fig. 3.  

Fig. 3 – Scheme of boundary condition constraint applied to a fluid-transfer boundary condition for groundwater heads below the river bed.

FURTHER INFORMATION & USEFUL LINKS

Manuals and Guidelines
FEFLOW 10.0 Documentation - Boundary Conditions Constraints
FEFLOW 10.0 Documentation - Fluid-Flow Boundary Conditions