Comparison of Ordinary Kriging (Ok) and Inverse Distance Weighting (Idw) Methods for the Estimation of a Modified Palaeoplacer Gold Deposit: A Case Study of the Teberebie Gold Deposit, Sw Ghana

Abstract

The study described in this paper involves the application of a conventional resource estimation method, inverse distance weighting (IDW), and univariate geostatistical technique, ordinary kriging (OK) to the gold grades data from the modified palaeoplacer Teberebie gold deposit, in Ghana. The deposit consists of 4 layered well-defined orebodies referred to as A reef, CDE reef, F24 reef and G reef at the mine environment. Simple, reliable, and adequately accurate resource/reserve estimation are essential to mining operations. Data used for the research were collected by diamond and reverse circulation (RC) drilling. A total of 19353 one-meter composite samples, consisting of 18962 RC chip samples from 695 RC drill holes, and 391 diamond drill core samples from 11 DD holes. Samples were analysed by atomic absorption spectrometry (AAS) for gold (Au). Descriptive statistical treatment was conducted on grade values for the reefs. To analyse for spatial structure of Au mineralisation, experimental downhole, and several horizontal directional semi-variograms were computed, and models fitted. Ore reserves were estimated by OK and IDW methods, and results of the various reefs compared. Regression analysis of estimated results indicate that, the inverse distance square (ID2) model produced estimates that compared well with the OK model in all the ore zones. It is therefore, appropriate to use ID2 as an alternative estimation method to the OK method for purposes of mine planning and grade control.

A Grade Control Method without Sampling Preparation Based on Spectrometric Techniques and the Use of Electronic Detonators in an Open Pit Tungsten Deposit

“…Several methods can be used to infer the grades of the whole bench from the borehole samples. While the nearest-neighbor interpolation method is commonly used, Amadu et al [18] discovered that the inverse distance weighting method with an exponent of 2 is more effective than the nearest neighbour method. Additionally, vasylchuk and deutch [19] recommend using kriging for this work.…”