In Australian coalmines there are a number of coal quality parameters that are measured by analysing samples from boreholes. The data points are used to generate spatial models of these parameters, typically using inverse distance as an interpolation method. Numerous additional boreholes are drilled which have geophysical logs measured downhole but from which no coal quality samples are taken. Full suites of coal quality analyses are expensive and time consuming to drill and sample. If geophysical proxies for coal quality could be used, this would greatly improve the accuracy of the resource estimation without additional cost.This paper examines two of the most important coal quality parameters - ash and volatile matter - and assesses the results of a number of geostatistical interpolation methods. The aim is to improve the accuracy of the interpolated models. Auxiliary variables from geophysical logs are also trialled to improve the estimation.A multivariate analysis of ash, volatile matter and several auxiliary variables of a selected coal seam was undertaken in a spatial context, using variography. This was used to construct a model of co-regionalisation that relates all of the variables as manifestations of an underlying geological theme.The traditional inverse distance interpolation approach was compared to ordinary kriging, universal kriging, co-kriging, regression kriging and kriging with external drift approaches. The relative merits of the six interpolation methods were compared by removing all points, one at a time and estimating at the thus known location. The mean error and the root mean square error were evaluated as a measure of bias and precision.This paper demonstrates that there is significant opportunity to improve the estimations of coal quality when using kriging with external drift. For example, the results show that when using the depth of a sample as an external drift variable there is a significant improvement in the accuracy of estimation for volatile matter. Furthermore, when using borehole wireline density logs as an auxiliary variable for ash content, there is improvement in the estimation of the in situ ash. The underpinning reasons for this are discussed in the context of the underlying geology and the physical properties of the coal. These findings hold promise for utilising cheaper proxies for coal quality parameters to dramatically improve data density and the quality of estimations.CITATION:Jeuken, R, Xu, C and Dowd, P, 2017. Improving coal quality estimations using geostatistics and auxiliary variables, in Proceedings Tenth International Mining Geology Conference 2017, pp 157-168 (The Australasian Institute of Mining and Metallurgy: Melbourne).