Join us for an expert talk by Pedram Masoudi:

Sensory data fusion by kriging, application to porosity estimation by neutron and density well logs.

📅 Wednesday, 4 June 2025 | 🕚 11:25 AM – 11:35 AM CEST
📍 Digital Transformation Area Theatre
Authors: Pedram Masoudi, Jean Langanay and Hélène Binet, Geovariances

Discover how sensory data fusion using kriging can revolutionize porosity estimation from neutron and density well-logs.
Don’t miss this session led by our expert geomodeling consultant, Pedram Masoudi, showcasing an innovative approach to enhance subsurface modeling accuracy.

Full abstract
In the earth sciences, kriging is a frequent method for interpolating a random variable as a function of coordinates. In this article, kriging is used innovatively to estimate porosity as a function of neutron and density well-logs, replacing coordinates. So, kriging is introduced here as an estimator in the framework of data fusion. The dataset consists of core porosity and well logs of the Windalia Sand Member in two wells in the Barrow Island oilfield, Northwest Australia. In each well, variability of core porosity as a function of Euclidean distance (petrophysical well-logs as coordinates) was evaluated by variogam. Thereafter, the kriging method was applied to estimate the core porosity in the other well. The estimated values were then compared to the measured core porosity values and to a conventional empirical method of porosity estimation. The comparison confirmed the accuracy of the kriging-based method in estimating porosity in the sandy reservoir, with the estimation error of the proposed method just below the error of the empirical method. The proposed method has three advantages: (i) free of lithological parametrization, (ii) independent from core porosity for calibration, and (iii) applicable while only a limited number of data is available.