Call via: SPRING Calculation Inverse calibration… A distinction is made according to the method of calculation:

Steady state flow…

Transient flow…

Mass transport (density dependent)…

Heat transport computation…

 

After creating the groundwater model and input of all data follows the calibration. SPRING supports the Calibration with the gradient method or the method of the inverse modelling.

The inverse modelling is an iterative process in which the system will be explored step by step and this will be judged over and over again. The user must constantly review the sense of the calculated values. After correcting false measurements or other incompatibilities of the input data remain the following model parameters as uncertain variables:

 

Permeabilities (K-values)

Leakage coefficients (LERA, LEKN)

Boundary in- and outflow (VORF and LERA)

Storage coefficients (transient calculation, KSPE)

 

The unknown model parameters p are normally determined by measured potentials on groundwater wells. For transient calculations hydrographs of potentials should also be present for some measuring points. In addition, e.g. discharge measurements give information about the magnitude of leakage rates, and the volume of boundary in- and outflow may be known. These are measurements or observations data (m).

In the calculation of the direct problem the potential heads and the quantities (measures m) are calculated from estimated, but unsecured model parameters p. In the inverse method, however, the inverse problem is to be solved, in which the unsecured model parameters p are determined from the backed measures m.

 

Method	Input data	Calculated data
direct	Model parameters p	measured data m
inverse	Measured data m	Model parameters p

 

 

The model parameters are in the inverse problem the dependent variables which should be determined by the independent observations. The aim of the calibration is to estimate the unknown model parameters so that a measured state is reproduced by the numerical model in the best way possible.

In the formulation of the inverse problem is to be taken into account that the hydrogeological concept might be faulty. Through a statistical formulation of the inverse problem, errors in the subsequent interpretation of the measured data can be compensated.

 

Model parameters as unknown sizes