/NMR-K-prediction

Codes for estimating parameters in the Kozeny-Carmen relation between NMR measurements and hydraulic conductivity

Primary LanguageMATLABMIT LicenseMIT

Dec 2020: This repo won't work out of the box! Needs formatting/reorg

Predicting K from NMR data

This repo consists of code to find parameter values for models relating NMR measurements to hydraulic conductivity K. The models are given in Maurer and Knight, 2016.

Data

The data files are in the 'Aggregated_data' folder, and have columns: depth, T2ML, porosity, permeability, and optionally other measurements like Sum-of-Echoes (SOE).

Methods

Codes are provided for running grid search, bootstrap, MCMC, and direct solves for the parameters of the models.

Models

The codes are set up to handle the SDR model and various models using SOE. The Kozeny-Godefroy model can also be handled with some tweaking.

Licensing/Permissions/Acknowledgments

The data in the folder are the same as those published in Maurer and Knight 2016, see that paper for the original data sources and discussion. Original codes were developed by and are copyrighed by Jeremy Maurer, subsequently modified significantly by Alexander Kendrick. All codes are freely available for use under the MIT license.