This paper reports an application of uncertainty visualisation of a regional scale (1:50 000) 3D geological geometry model to be involved in GIS-based 3D mineral potential assessment of the Xiangxibei lead-zinc mineral concentration area in northwestern Hunan District, China. Three-dimensional (3D) geological modelling is a process of interpretation that combines a set of input measurements in geometry. Today, technology has become a necessary part of GIS-based deep prospecting. However, issues of sparse data and imperfect understanding exist in the process so that there are several uncertainties in 3D geological modelling. And these uncertainties are inevitably transmitted into the post-processing applications, such as model-based mineral resource assessment. Thus, in this paper, first, a big-data-based method was used to estimate the uncertainty of a 3D geological model; second, a group of expectations of geological geometry uncertainty were calculated and integrated into ore-bearing stratoisohypse modelling, which is one of the major favourable parameters of assessment for Lead-Zinc (Pb-Zn) deep prospectivity mapping in northwestern Hunan; and finally, prospecting targets were improved.

本文报告了不确定性可视化在区域规模（1:50 000）3D地质几何模型中的应用，该模型将涉及基于GIS的湘西北地区湘西北铅锌矿集中区的3D矿产潜力评估。三维（3D）地质建模是一种解释过程，将几何学中的一组输入测量值组合在一起。如今，技术已成为基于GIS的深度勘探的必要组成部分。但是，在此过程中存在数据稀疏和理解不完善的问题，因此3D地质建模存在一些不确定性。这些不确定性不可避免地会传递到后处理应用程序中，例如基于模型的矿产资源评估。因此，在本文中，首先，基于大数据的方法用于估计3D地质模型的不确定性；其次，计算了一组对地质几何不确定性的期望值，并将其整合到含矿层状同位素分布模型中，这是湘西北铅锌（Pb-Zn）深远景测绘评估的主要有利参数之一；最后，改善了勘探目标。