I've been thinking about this after working with geological maps and modelling.
When we look at a geological map, we tend to treat the boundaries and structures as if they are established facts. But in many areas, only a relatively small part of the geology is actually exposed or directly measured. We might have some outcrops, structural measurements, drill holes, geophysics, remote sensing and DEM data. Then we connect the dots and build a geological interpretation.
That made me wonder:
How much of the geology in our maps and 3D models is actually observed, and how much is interpretation?
For example, if a fault is mapped between two well-exposed areas, but the fault itself isn't exposed in the ground between them, where does observation end and interpretation begin?
The same problem becomes even more interesting with 3D geological modelling. A model can look extremely detailed and convincing even when much of the subsurface is poorly constrained.
I'm not saying geological models are unreliable. We obviously have to interpret incomplete data. That's part of geology.
But perhaps we should be more explicit about the difference between:
- what we actually observed
- what we can reasonably infer
- what is simply the model's best interpretation given the available data
I'm curious how other geologists think about this.
Where do you personally draw the line between geological observation and geological interpretation?
And have you ever had a situation where new field evidence forced you to completely rethink a geological model you were previously confident about?