Towards an Improved Nodule Resource Estimation and Classification Using Hard and Soft Data

The deep-sea ocean floor offers a great potential for mineral resources. The evaluation of these resources is both time consuming and costly and require as any resource evaluation data; geodata. The collection of hard data, physical samples, is of course preferred because the associated uncertainty is low. However, the collection of soft data or image derived data can offer a valuable supplement. In this paper image derived nodule uncertain abundance data from an area in the Pacific is utilized. Implementing these uncertain data improves the estimation when the quality of the estimation is measured using the classification indicators like the kriging standard deviation, the relative prediction error, the slope of regression and the weight of the mean. It is concluded that uncertain data is better than no data. INTRODUCTION The deep-sea offers vast mineral resources (Cathles (2011), Hannington et al. (2011), Hannington (2013), Singer (2014)). A mineral resource can according to increasing geological confidence be classified as an inferred, an indicated or a measured resource. The classification is performed by a competent- or qualified person and is dependent on the amount, quality and characteristic of available geodata. The competent person must have at least five years of experience relevant to both the style of mineralisation, the type of deposit and the performed task. Typical examples of a “task” is for example resource estimation and classification or planning- and execution of exploration activities and presentations of the results. The competent person must also be a member or a fellow of a professional organisation and thereby be bound by reporting code (-s) that defines minimum standards, recommendations and guidelines for public reporting.