Utilising pykrige and matplotlib for Spatial Visualisation of Geological Variations

<p>When working with geological and petrophysical data, we often want to understand how that data changes over our field or region. One of the ways we can do this is to grid our actual measurement values and extrapolate what those values may be in other areas that have yet to be explored using boreholes.</p> <p>One particular method for carrying out this extrapolation is kriging, a geostatistical procedure named after South African mining engineer Danie G. Krige. The idea behind kriging lies in its estimation technique: it uses spatial correlation between observed data to predict values at unmeasured locations.</p> <p>By gauging how variables change over a distance, this method establishes a statistical relationship that can be used to predict values across an area, transforming scattered data points into a coherent spatial map.</p> <p>Within this tutorial, we will look at a Python library called&nbsp;<a href="https://github.com/GeoStat-Framework/PyKrige" rel="noopener ugc nofollow" target="_blank"><strong>pykrige</strong></a><strong>.&nbsp;</strong>This library has been designed for 2D and 3D kriging calculations and is easy to use with well log data.</p> <h1>Importing Libraries &amp; Data</h1> <p>First, we need to import the libraries we are going to need. For this article, we will require the following libraries:</p> <p><a href="https://towardsdatascience.com/utilising-pykrige-and-matplotlib-for-spatial-visualisation-of-geological-variations-a288b186bfd6">Read More</a></p>