Quantitative 3D remote digital compositional and structural characterisation of outcrops

陆上盆地三维地下地理空间模型的建立是油气勘探的一个重要方面，它严重依赖于从地质露头和钻孔岩心分析中获得的信息。目前的分析技术产生的数据集非常有限，无法完全捕捉这些地质特征的内在3D性质，导致构造信息存在巨大差距，从而严重影响地质解释的准确性。一种新开发的野外便携式成像仪(MicroFTS)已经证明了能够远程准确识别关键沉积岩性的能力。该项目将展示MicroFTS以极低的成本显著增加样本-井点和盆地尺度的结构和成分信息的数量、覆盖范围和详细程度的能力。地质露头为地质学家提供了重要的主要数据来源，并在学术界和工业界广泛用于教学、培训和研究，以及开发概念和预测地质模型。准确描述岩石体积和油气盆地的沉积结构的岩性和矿物学特征对于预测油气资源的存在或量化油气资源的体积是至关重要的。地震反射数据可以成像大规模的储集层建筑，但垂直和水平分辨率通常限制在几十米。相反，岩心和电缆测井可以提供更高的分辨率，但井通常分布稀疏，只采样很小比例的岩石体积。然而，露头提供了从不到1厘米到几十公里的范围内对岩体及其几何、建筑和岩性非均质性的直接观测。传统的露头分析技术(如沉积测井、地表地质图和剖面)提供了关于地质系统的宝贵信息，为许多当代构造、地层和合成提供了便利。然而，野外数据通常是通过一维和二维纸质方法收集的，不能完全反映地质特征的内在3D性质。相关的精确度、精确度和不确定性很少被定义，而且往往很难提取关于沉积岩体的几何和空间非均质性的可靠定量信息，这些数据是基于计算机的三维地质统计油藏建模的关键数据。然而，可访问性问题极大地限制了使用传统地图绘制方法从这类网站获取的数据。地质暴露的范围也可能超过10平方公里，导致抽样严重不足和不具代表性的数据集。使用钻孔岩心和露头的定量、数字化特征来分析沉积结构的更全面的方法将提高盆地分析的准确性。尽管利用陆地激光雷达扫描(TLS)远程模拟地质界面几何形状的精度已经取得了重大进展，但岩石体积本身的定义受到可用照片反演的矿物学和岩性信息的范围和准确性的严重限制。虽然基于光谱反射率的遥感方法在解析岩石成分方面表现出了一定的能力，但由于可以探测到的矿物的范围和精度有限，以及观察结构和光照条件的影响，实际应用受到了严重限制。因此，迫切需要一种能够远程描述石油和天然气行业感兴趣的岩性的方法。发射光谱学有许多能力可以满足这一要求。

The construction of 3D sub-surface geospatial models of onshore basins is a vital aspect of hydrocarbon exploration and is critically dependant on the information derived from the analysis of geological outcrops and borehole cores. Current analysis techniques produce very limited datasets that do not fully capture the inherent 3D nature of these geological features which result in significant gaps in structural information with consequent severe effects on the accuracy of geological interpretation. A newly developed field portable imaging instrument (MicroFTS) has demonstrated the capability to remotely identify the key sedimentary lithologies accurately. This project will demonstrate the capability of the MicroFTS to significantly increase the volume, coverage, and level of detail, of structural and compositional information at sample-borehole-site and basin scales, at very low cost.Geological outcrops provide an important primary source of data for geologists and are extensively used within both academia and industry for teaching, training and research, and for the development of conceptual and predictive geological models. Accurate characterisation of lithology and mineralogy in rock volumes and the sedimentary architecture of a hydrocarbon basin is central to predicting the presence or quantifying the volumetrics of hydrocarbon resources. Seismic reflection data can image large-scale reservoir architectures but vertical and horizontal resolution is typically limited to tens of metres. Conversely, core and wireline logs can provide much higher resolution but wells are typically sparsely distributed, sampling only a very small percentage of the rock volume. Outcrops, however, offer direct observations of rock bodies and their geometries, architecture and lithological heterogeneities over scales ranging from less than 1 cm to several tens of kilometres. Conventional outcrop analysis techniques (e.g. sedimentary logs, surface geological maps and cross-sections provide invaluable information on geological systems, facilitating many contemporary structural and stratigraphical and syntheses.However, field data are typically collected by one- and two dimensional paper-based methods that do not fully capture the inherent 3D nature of geological features. Associated accuracy, precision and uncertainty are rarely defined, and it is often difficult to extract reliable quantitative information on the geometries and spatial heterogeneities of sedimentary rock bodies, data which are essential for 3D computer-based geostatistical reservoir modelling. However accessibility issues significantly limit the data acquired from such sites using traditional mapping methods. Geological exposures can also extend over 10s of square kilometres, resulting in a significantly under sampled and unrepresentative dataset. A more integrated approach to the analysis of the sedimentary architecture using a quantitative, digital- based characterisation of borehole cores and outcrops would enhance the accuracy of basin analysis.Although there have been significant advances in the accuracy of modelling the geometry of geological interfaces remotely using Terrestrial LiDAR Scanning (TLS), the definition of the rock volume itself has been restricted by severe limitations in the range and accuracy of the mineralogical and lithological information retrievable using photographs. While spectral reflectance based remote sensing methods have demonstrated some capabilities in resolving rock compositions the operational utility is severely restricted by the limited range and accuracy of minerals that can be detected and the effects of viewing configuration and illumination conditions. There is therefore an urgent requirement for a methodology that can remotely characterise the lithologies of interest to the oil and gas sector. Emission spectroscopy has a number of capabilities that can meet this requirement.