Predicting the emplacement of Cordilleran porphyry copper systems using a spatio-temporal machine learning model

Diaz-Rodriguez, Julian; Müller, R. Dietmar; Chandra, Rohitash

doi:10.1016/j.oregeorev.2021.104300

Cited by 9 publications

(4 citation statements)

References 87 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Reconstructing plate motions is a fundamental step towards understanding what drives geodynamic motions, the nature of the platemantle system and how this system could change over geological time 174 . Indeed, a primary motivation for developing plate-focused, 169 , and (parts c-e) the along-strike variations on inputs to the subduction zone and how these could relate to porphyry copper formation 30 . Kinematic parameters based on ref.…”

Section: Plate-mantle Dynamicsmentioning

confidence: 99%

“…Such integration has already advanced understanding of the internal dynamics of the Earth [16][17][18][19][20][21][22] , long-term biogeochemical cycles 23,24 and biodiversity through time [25][26][27] . The accessibility of these models has also facilitated advances in applied areas, for example, linking mineral systems and their prospectivity to tectonic parameters [28][29][30] and unravelling tectonic controls on natural hazards 31,32 . Likewise, images and animations produced from plate reconstruction tools have promoted Earth science education and outreach 33,34 .…”

mentioning

confidence: 99%

“…A, An illustration of the reconstructed subduction history along the Andean margin as a plate kinematics and data mining example. Plots illustrate (parts a and b) the potential link between overall subduction flux and zircon age frequency, with a time lag169 , and (parts c-e) the along-strike variations on inputs to the subduction zone and how these could relate to porphyry copper formation30 . Kinematic parameters based on ref 10.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Deconstructing plate tectonic reconstructions

Seton

Williams²,

Domeier³

et al. 2023

Nat Rev Earth Environ

View full text Add to dashboard Cite

The evolving mosaic of tectonic plates across the surface of the Earth sets boundary conditions for the evolution of biotic and abiotic processes and helps shape the dynamics of its interior. Reconstructing plate tectonics back through time allows scientists from a range of disciplines (such as palaeobiology, palaeoclimate, geodynamics and seismology) to investigate Earth evolution through these spatiotemporal dimensions. However, the variety and complexity of plate reconstructions can lead to some of their limitations being overlooked. In this Technical Review, we discuss the domain-specific knowledge underpinning modern quantitative plate reconstructions and convey a set of principles on how to use (but not abuse) the software or results. Open-source plate tectonic reconstruction software, like GPlates, has led to a major shift in working practices, handing non-specialists the tools to develop and integrate reconstructions based on their own datasets and expertise. However, there is no 'one-size-fits-all' and users need to understand what data and underlying assumptions go into making different, sometimes competing reconstruction models. It is therefore essential to consider the many ways reconstructions simplify reality when interpreting them to avoid circular reasoning. Although many aspects of deep-time reconstructions remain unresolved, future work on intercomparisons between models and uncertainty quantification is an essential pathway towards next-generation plate reconstructions. Key points• Plate tectonic reconstructions have evolved from simple, rigid reconstructions to ones that incorporate the time-dependent evolution of plates and their boundaries, deformation and/or the history of subduction from seismic tomography.• Reconstructions can be powerfully predictive for a wide range of disciplines beyond tectonics, including palaeobiology, palaeoclimate, geodynamics and seismology.• With the advent of community-driven, open-source software and tools, plate models have become accessible and practicable to the wider geosciences community.• With this accessibility comes a responsibility for specialists and non-specialists alike to understand how these plate models are built, their weaknesses and pitfalls and how they can be used effectively to ensure correct inferences are made.• When reconstructing the tectonic plates of the Earth, there is no 'one-size-fits-all'. Different data types and techniques are more applicable for different time periods, resolutions and purposes, for example, palaeomagnetics for pre-Pangea time periods.• Important areas of ongoing research include the quantification of uncertainty, incorporation of machine learning techniques and linking reconstructions to physics-based deep Earth models and surface (and/or biogeochemical) models.

show abstract

Section: Plate-mantle Dynamicsmentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 1 more Smart Citation

Deconstructing plate tectonic reconstructions

Seton

Williams²,

Domeier³

et al. 2023

Nat Rev Earth Environ

View full text Add to dashboard Cite

show abstract

“…The extraction of spatio‐temporal data from plate reconstructions using GPlately presents exciting new opportunities in machine learning. Training neural networks on a range of plate kinematic parameters can be used to predict the formation of certain types of mineral deposits such as porphyry copper (Julian Diaz‐Rodriguez et al, 2021). In addition, integrating recycled oceanic lithosphere at subduction zones through deep geological time using GPlately can bring context to intraplate volcanism (Mather et al, 2020), among many other future applications.…”

Section: Future Outlookmentioning

confidence: 99%

Deep time spatio‐temporal data analysis using pyGPlates with PlateTectonicTools and GPlately

Mather

Müller

Zahirovic

et al. 2023

Geoscience Data Journal

Self Cite

View full text Add to dashboard Cite

PyGPlates is an open‐source Python library to visualize and edit plate tectonic reconstructions created using GPlates. The Python API affords a greater level of flexibility than GPlates to interrogate plate reconstructions and integrate with other Python workflows. GPlately was created to accelerate spatio‐temporal data analysis leveraging pyGPlates and PlateTectonicTools within a simplified Python interface. This object‐oriented package enables the reconstruction of data through deep geologic time (points, lines, polygons and rasters), the interrogation of plate kinematic information (plate velocities, rates of subduction and seafloor spreading), the rapid comparison between multiple plate motion models, and the plotting of reconstructed output data on maps. All tools are designed to be parallel‐safe to accelerate spatio‐temporal analysis over multiple CPU processors.

show abstract