The unknown model parameters to be determined, which can be optimised using inverse modelling, consist of the following input data:

• Hydraulic conductivites:

Horizontal (KWER) and vertical (KWEV) K values can be optimised simultaneously in the 3D model. Horizontal K values can be optimised and the vertical K values recorded at the same time, or the vertical K values can be optimised and the horizontal K values recorded.

In these cases, it is not possible to influence the anisotropy factor between the horizontal and vertical K values. However, the horizontal K values can be optimised and the ratio between the horizontal and vertical K value can be defined by entering KFVH (Attributes Extras Adjust vertical/horizontal K value factor (KFVH)..., default setting: global KFVH = 0.1).

• Storage coefficients:

The storage coefficient (data type KSPE) in the horizontal model or the porosities (PORO) for calculating the storage coefficient in vertical or 3D models can be optimised (see explanations on the storage coefficient in the chapter "Transient calculation. ".

• Leakage coefficients:

The leakage coefficients for polygonal courses (data type LERA, per running metre) can be optimised.

There is usually prior knowledge of these parameters p, for example in the form of laboratory measurements, interpretations of field tests or subjective preliminary estimates. This prior knowledge serves as starting values for the iteration of the inverse problem. They must be defined in a separate file for an optimisation run. This file is referred to below as the parameter file. The structure of the parameter file is described in the chapter "Realization in SPRING.

Observation data and measured values