I

O Observed field mesurement values m Computed model values

Continuous curve through computed values Continuous curve through model values w r ** «

H—I—|—|—|—|—|—)—|—|—|—| ■■(■ H

Measurement Location along the Surface (b) Curve matching computed data and field data

Measurement Location along the Surface (b) Curve matching computed data and field data

(c) Flow chart for interpretation using a model

FIGURE 6.16 Generalized process of inversion: (a) computed and field measurement values plotted as a function of distance along the surface of the ground, (b) the model used to compute the values for the computed curve, and (c) a flowchart of the process of inversion.

(c) Flow chart for interpretation using a model

FIGURE 6.16 Generalized process of inversion: (a) computed and field measurement values plotted as a function of distance along the surface of the ground, (b) the model used to compute the values for the computed curve, and (c) a flowchart of the process of inversion.

through the observed field data points. There are some basic rules for deciding on the type of model to be used for the inversion:

• The model must be appropriate for the situation at hand. You would not use the numerical model that represents a sphere to determine the parameters of a pipe.

• The model must be relatively easy to compute. It is generally inappropriate to use a model when the forward model takes an enormous amount of computer memory and time to compute. Several iterations to enable convergence should only take a short period of time and not require a supercomputer, or should not require a system with massive parallel processors. Of course, there is the rare situation where it might be appropriate to utilize a very complicated model.

0 0

Post a comment