Has anyone validated a log-to-core workflow that ties XRD mineralogy and sonic-derived dynamic moduli to triaxial UCS/static E–ν for Wolfcamp A in the Delaware? We’re seeing quartz 38–42%, NMR phi 9–11%, and BI from logs overestimates fracability versus core, so I’m hunting for a paper, spreadsheet, or example dataset to tighten the mapping for stage spacing and cluster count.
If your “BI from logs overestimates fracability,” fit a net-stress dynamic→static E–ν/UCS transform using Vp/Vs plus quartz/clay and NMR phi (PLS works fine), then sanity-check it with DFIT closure in your A benches. For a starting point on mineralogy brittleness, Rickman et al. SPE-115258-MS is solid: A Practical Use of Shale Petrophysics for Stimulation Design Optimization: All Shale Plays Are Not Clones of the Barnett Shale | SPE Annual Technical Conference and Exhibition | OnePetro, but add a porosity/TOC term or it’ll still sweet-talk you.
And i’d first fix the lab‑vs‑log apples-to-oranges: shift sonic dynamic E–ν to pseudo‑static with a confining‑pressure/frequency correction and re‑reference core UCS/E to in‑situ effective stress; in Wolfcamp A we saw the BI bias drop once T2LM (bound fluid) and clay type were added as ductility regressors. Do you have dipole Vp/Vs and the triaxial Pc/σ3 schedule to run that ‘log‑to‑core’ reconciliation?
What’s worked for us is to anchor the log-to-core mapping with a continuous scratch-test profile on Wolfcamp A core (≈1 ft spacing), then regress Vp/Vs and XRD to the scratch UCS/E rather than sparse triaxials… It’s cheap and fast (about a half-day per well), and it corrected our BI overestimate where “quartz 38–42%” was masking clay-fabric effects. Do you have slabbed core to run a scratch line, or want the template we used?
Don’t trust BI here — first apply a low‑frequency correction (squirt‑flow/Gassmann) using your NMR T2 and “quartz 38–42%,” which cut our Wolfcamp A E_dyn→E_static gap by about 25%. Then regress UCS/E to λρ–μρ from the dipole plus clay%, and sanity‑check against DFIT closure, not just triaxial. @kevin_m193 scratch profiles help, but DFIT‑anchored transforms gave tighter stage‑spacing/cluster picks for us — I can share the λρ–μρ workbook if you want.
I’d anchor it with a continuous drilling MSE‑derived UCS along Wolfcamp A tied to your triaxial set, then back‑regress Vp/Vs and XRD to that UCS curve… With quartz about 40% and “NMR phi 9–11%,” MSE‑UCS has stayed within about 10–15% of core for us and tightened stage spacing/cluster count picks. Template here (SPE‑176884, MSE→UCS workflow): https://onepetro.org/SPEAPOG/proceedings/15APOG/All-15APOG/SPE-176884-MS/180169; you’ll need high‑res torque/ROP.