Barcoding Magmas: applying machine learning in zircon geochemistry to increase provenance accuracy

Project Description

Provenance studies analyse minerals in sedimentary successions to ascertain the source of that material in order to reconstruct past environments & climate as well as assessing the suitability of the sediment for resource exploration or carbon capture/storage (CCS). Traditionally this is done by collecting large suites of zircons from sediments, dating them using U-Pb geochronology & matching the acquired ages to geological terranes. This gives a very broad picture of the source of the sediment but lacks accuracy. Zircons grow as accessory minerals in magmatic bodies in orogenic belts. During their growth, zircons geochemically record the conditions in the magma chamber much like tree rings record the passing of the seasons. Each individual zircon records part of the geochemical history of that magma chamber. We propose to use machine learning to analyse the growth patterns of large populations of zircons from a single suite of magmatic bodies (the Donegal Granites) to create a unique geochemical barcode representing continuous timeline of conditions within their corresponding magma chambers. Zircons from sediment believed to derive from the orogenic belt containing our magmatic systems (the Carboniferous sandstones of NW Ireland) will also be geochemically analysed. Using bespoke pattern matching software these zircons will be compared to new barcodes of the nearby granite bodies to ascertain their exact source & reconstruct the regional paleoenvironmental configuration.