Elsevier

Computers & Geosciences

Volume 56, July 2013, Pages 76-82
Computers & Geosciences

Relationships between palaeogeography and opal occurrence in Australia: A data-mining approach

https://doi.org/10.1016/j.cageo.2013.02.002Get rights and content

Abstract

Age-coded multi-layered geological datasets are becoming increasingly prevalent with the surge in open-access geodata, yet there are few methodologies for extracting geological information and knowledge from these data. We present a novel methodology, based on the open-source GPlates software in which age-coded digital palaeogeographic maps are used to “data-mine” spatio-temporal patterns related to the occurrence of Australian opal. Our aim is to test the concept that only a particular sequence of depositional/erosional environments may lead to conditions suitable for the formation of gem quality sedimentary opal. Time-varying geographic environment properties are extracted from a digital palaeogeographic dataset of the eastern Australian Great Artesian Basin (GAB) at 1036 opal localities. We obtain a total of 52 independent ordinal sequences sampling 19 time slices from the Early Cretaceous to the present-day. We find that 95% of the known opal deposits are tied to only 27 sequences all comprising fluvial and shallow marine depositional sequences followed by a prolonged phase of erosion. We then map the total area of the GAB that matches these 27 opal-specific sequences, resulting in an opal-prospective region of only about 10% of the total area of the basin. The key patterns underlying this association involve only a small number of key environmental transitions. We demonstrate that these key associations are generally absent at arbitrary locations in the basin. This new methodology allows for the simplification of a complex time-varying geological dataset into a single map view, enabling straightforward application for opal exploration and for future co-assessment with other datasets/geological criteria. This approach may help unravel the poorly understood opal formation process using an empirical spatio-temporal data-mining methodology and readily available datasets to aid hypothesis testing.

Highlights

► Palaeogeography was found to be a strong indicator of Australian precious opal. ► A data-mining methodology was used to quantitatively extract key time sequences. ► Only 11% of the entire Great Artesian Basin undergoes favourable palaeogeography. ► Important environmental transitions were extracted using data-mining approaches. ► The methodology was used to construct a map indicating favourable regions.

Introduction

Exploration practices and data analysis and modelling related to mineral and hydrocarbon systems are making increasing use of integrated geological datasets for understanding resource formation processes and thus improving exploration decision making. This is made possible by the rapid increase in the interconnected digital storage of data and the rapidly increasing power of computing systems. An important factor for many studies is the consideration of the palaeogeographic/depositional environment through time as inferred from the geological record, since the configuration of a particular region as observed today is nearly always substantially different from the geological past. This, together with an understanding of the physical processes involved, is particularly important for identifying niche geological conditions that result in a phenomena such as ore deposits, since only a consideration of the relevant geological history leads to an appropriate contextualisation of present-day observations. Thus investigating the nature and relationships involved in time-varying spatial data is a promising area for developing and applying machine learning to geological data. In this paper we develop a quantitative approach for utilising the palaeo-environmental history over a large portion of Australia to investigate some key factors associated with opal formation, as a step towards establishing more quantitative exploration criteria. A data-mining approach is taken here to cope with the large datasets involved, and to handle some degree of noise present in the datasets utilised, especially considering that the digital palaeogeographic maps this approach is based on involves interpretations and interpolations of sparse data points. The study is made possible through use of the GPlates plate tectonic geographic information system (Qin et al., 2012) and Gplates data-mining functionality (Landgrebe and Mueller, 2008), in which time-varying data are explicitly modelled, allowing for direct extraction of spatio-temporal associations.

Opal is a form of hydrated silica (SiO2·n H2O) found predominantly throughout the Great Artesian Basin in Australia. Although the Australian opal fields are responsible for over 90% of the world's opal supply, there has been a decline over the past 20 years in the number of high quality gemstones produced by the fields (Smallwood, 1997). This is accounted for, to some extent, by the absence of formal exploration models for Australian opal, which has led to an over-reliance on mining old opal fields discovered in the early 20th century (Barnes and Townsend, 1990). Making use of known locations of opal, we construct a method allowing us to search for other locations that possess similar palaeogeographic sequences throughout the entire Great Artesian Basin, and help establish important time-varying palaeogeographic/depositional/facies conditions favourable for opal formation.

Section snippets

Eastern Australian opals and palaeogeography

Opal in the Great Artesian Basin is found within fractures and primary and secondary pore spaces in the upper 30 m throughout the highly weathered Cretaceous sedimentary sequence of the Eromanga and Surat basins (Barnes and Townsend, 1990). The location of opal deposits is shown in Fig. 1, and includes primary opal mining regions from which a total of 1036 mining localities were identified for this study. The locations were taken from maps published by the state geological surveys (Carr, 1979,

Formalising the palaeogeographic time-series analysis

The quantitative methodology followed in this paper utilises the palaeogeographic dataset, referenced with known opal localities to derive a data-driven model of target palaeogeographic sequences. The digital representation consists of spatial zones defined by polygons, with associated geographical attributes. We define the unique set of 16 different palaeo-environment types ω=[ω1,ω2,,ω16], where the ith type is referred to as ωi (the environment types are listed in Table 1). In this study it

Event sequence pattern-matching

Eq. (2) provides the means to represent the nature of the palaeogeographic evolution at a particular location without requiring absolute ages. This is important for opal exploration, since direct radiometric dating of Australian opal has not been possible due to its low uranium content (Gaillou et al., 2008). Consequently, it has not been possible to constrain the age of Australian opal further than the age of its host rock, which is predominantly Cretaceous, with a single location (Mintabie)

Computing methodology

The plate-tectonic Palaeogeographic Information System GPlates was used as the basis for this study (Boyden et al., 2011). GPlates provides the ability to assess multiple spatial datasets that include time-varying properties, allowing for visualisations and analyses coherent both in space and time. Thus factors such as a varying geographic environment, or spatial relationships between different datasets undergoing plate motions or crustal deformation, can be assessed. GPlates has been augmented

Results and validation

The result of the pattern matching methodology is shown in Fig. 5, with the Cretaceous sedimentary rocks highlighted, and filled black regions depicting areas matching the trained palaeogeographic patterns (targeted regions). Two general observations can be made. The first observation is that there is a strong correlation between the region encompassing Cretaceous sedimentary rock and the targeted regions. This is an expected result since this region has a unique palaeographic characteristic

Conclusions

This paper presents a quantitative methodology for utilising age-coded geological data to develop models/targeting criteria for opal exploration. Such data-driven approaches are becoming increasingly viable as the growth in the storage of digital geological data continues to accelerate, allowing for repeatable and less-subjective analyses. A palaeogeographic dataset consisting of age-coded geological environments in Australia was used to assess the palaeogeographic evolution at known opal

Acknowledgements

This research was funded by ARC Grants FL0992245 and DP0987604. We would also like to thank the reviewers for their constructive feedback.

References (23)

  • N.F. Exon et al.

    The Cretaceous of the Eromanga and Surat Basins

    BMR Journal of Australian Geology and Geophysics

    (1976)
  • Cited by (0)

    View full text