Danomics + Claude: Evaluating Tops Completeness & Setting Up Zones

Your Claude (LLM) account can connect to your Danomics account. This is outlined in this article on how to set up a Danomics Connect for Claude. You will need to have completed this setup (or a similar one) to follow along with the tutorial below.

In this tutorial we will be using Claude evaluate our tops for completeness and based on those results set up our petrophysical zones.

Moving Forward

This tutorial continues on from using Claude + Danomics to Find and Set a Key well. If you haven't yet done so, take a look at that video.

When first starting a project one task you'll need to do is set up petrophysical zones. To do this properly you should analyze your tops coverage, which is normally a painful exercise done in spreadsheets and cross-sections.

Claude can make this easier for us. I'll start by giving it the following prompt:

In the new PRB.cpi can you evaluate the tops and determine if there are any formations that are sparsely picked?

Claude thinks for a while and then comes back with a formatted table of tops coverage and a recommendation. Here is a screenshot of the table:

Two things stand out.

Lower Parkman is the genuinely sparse one — 9 of 56 wells (16%). It's a sub-Parkman marker rather than a regional formation top, so this is almost certainly intentional — it was only picked where the subdivision is resolvable. Worth a decision: if you carry it as its own CPI zone it'll be null on 47 wells and your zone params for it will lean almost entirely on the key well. If you don't need it as a distinct interpretation interval, folding it into Parkman keeps the zone set clean. (For what it's worth, your Devon key well is not one of the 9 that has it.)

The deep section thins from ~Frontier down — Frontier 53, Mowry/Muddy 48, Dakota 47 — but this looks like well penetration, not missed picks. Three logged wells bottom out in the Carlile (49005287970000, 49005288830000, 49005289000000), each with only ~200–250 ft of log below their Carlile pick, so Frontier/Mowry are likely at or below TD. Another five reach only to Frontier (49005304140000, 49019206880000, 49019207330000, 49019207830000, 49045219920000). Since these are TD/log-limited, re-picking won't recover them.

This is really outstanding work that would have taken me over an hour to sort out myself, and I wouldn't have done it in this level of detail. I end by prompting it:

For this CPI can you please set up the zones, making sure to omit the Lower Parkman that you flagged as having sparse coverage.

Claude does as requested and sets up the stratigraphic zones.

Summary

Now Claude is saving us from the mundane grunt work that we need to do at the start of every project. With the analysis to find a key well and to evaluate which tops can be omitted we are now making significant progress faster than going it along.