One of the favourite appeals of Creationists is that science must be experimental and repeatable and as no one was there when rocks were laid down we cannot experiment on them and thus “historical sciences” like geology are not science but faith arguments from a set of beliefs. That is persuasive to some.
As a geologist I always groan when I read that the “scientific method” involves repeatable experiments.
Most of the time as a geologist you are study rocks in situ and observe , record and analyse and often do no experiments.
Here is one of my favourite Cross-sections of the Bowland shales. It is the top of the shales and then there are alternate shale and grit, followed by massive grit. You can observe the thickness and make-up of the strata and from Steno’s principle of superposition work out the order of deposition.
The next photo is a close up of Pendle Grits overlying Bowland Shales and for good measure cross-cut by some injectites of sand.
Wherever you go in the North of England and particularly in the Bowland area, you will find the same sequence with only minor variations. And so it follows my rough cross-section and, if you want more detail , the detailed cross-section in Little Mearley Clough on Pendle hill
I could have taken any of dozens of examples, but it shows that geology is, or rather was, primarily an observational and not an experimental science. That does not mean it is not repeatable as lots of geologists have looked at the same strata, to confirm or correct.
In this excellent article Paul Braterman deals with this issue and is critical of many popular definitions of the “scientific method” which litter our textbooks and misinform so many what science is.
I like Medawar”s comment that there is no one scientific method.
Now read what Paul says