Lecture 6b - Inverting Resistivity Soundings

Flashcards about inverse modeling, error metrics, and resistivity sounding, covering key concepts and processes.

What is inverse modeling?

Estimating the model from the data, starting with data to arrive at a best fit model.

What are the characteristics of most inversion problems?

Usually nonlinear and solved with an iterative numerical technique.

What are the inputs required for inverse modeling?

Measured data and acquisition parameters, initial estimate of subsurface model.

Why are initial estimates of subsurface models needed in inverse modeling?

To help the numerical computer algorithms converge on an appropriate solution since most inverse models are highly nonlinear.

What is the first component within an iteration of inverse modeling?

Calculating the surface response from model parameters.

What is said when the misfit is sufficiently small and the model can be accepted?

The algorithm has converged.

What are the characteristics observed in the example sounding data?

High resistivity, decreased resistivity, starts to increase again.

What was considered to be the apparent resistivity of the first and shallowest layer?

About 600 ohm meters.

What do the model parameters represent for resistivity sounding?

The layers thicknesses and resistivities.

What should interpretations align with in the absence of external information?

Soil, geology, or hydrological conditions.

What could a submeter thickness of relatively low resistivity reflect?

Soil.

What could a relatively low resistivity beneath the top layer represent?

A very clay-rich layer or an increasing influence of water.

What is the most common way to quantify misfit and derive an optimum?

Least Squares.

What it is calculated within the square brackets in the equation to calculate e?

The difference between observed and modeled apparent resistivities.

How to normalize the error metric to account for the numbers of measurements involved?

Normalizing by dividing by the number of measurements included.

What are common sources of error in resistivity inversion?

Random error from the measurement instrument, poor electrode contact with the ground.

What does the model assume about resistivity?

Lateral continuity i.e. resistivity only varies vertically.

What is the idea behind equivalence analysis?

Running the inversion repeatedly, making small perturbations each time to the data.

For resistivity sounding, what do the model parameters represent?

Thicknesses and the resistivities of the layers.

What is data inversion?

Iterative numerical modeling process, repeatedly runs a Ford model, automatically adjusting parameters until the model output best fits the measured data.