Related Experiment Video
Updated: Jun 17, 2026
07:58
Data Processing Methods for 3D Seismic Imaging of Subsurface Volcanoes: Applications to the Tarim Flood Basalt
Published on: August 7, 2017
Geochemical characterization of oceanic basalts using Artificial Neural Network.
1National Institute of Oceanography (Council of Scientific & Industrial Research) Dona Paula Goa 403004, India. pranab@nio.org
Geochemical Transactions
|December 24, 2009
Summary
Artificial Neural Networks successfully classified Central Indian Ocean Basin basalts. This method distinguishes ocean floor basalts (OFB) into normal (N-MORB), enriched (E-MORB), and ocean island basalts (OIB), overcoming limitations of traditional diagrams.
Area of Science:
- Geochemistry
- Petrology
- Machine Learning
Background:
- Geochemical diagrams often misclassify ocean floor basalts (OFB) as mid-oceanic ridge basalts (MORB).
- A specific method is needed to differentiate OFB into normal (N-MORB), enriched (E-MORB), and ocean island basalts (OIB).
Purpose of the Study:
- To apply Artificial Neural Network (ANN) techniques for classifying Central Indian Ocean Basin (CIOB) basalts.
- To identify geochemical signatures distinguishing N-MORB, E-MORB, and OIB within the CIOB.
Main Methods:
- Utilized a supervised Learning Vector Quantisation (LVQ) approach, a type of ANN.
- Trained and tested the network using diverse N-MORB, E-MORB, and OIB datasets.
Main Results:
- The LVQ method successfully identified geochemical characteristics within CIOB basalts.
- CIOB basalts were geochemically delineated as N-MORB, exhibiting moderate enrichment in rare earth and incompatible elements, indicative of E-MORB and OIB influences.
Conclusions:
- The ANN-LVQ method provides a successful approach for classifying OFB.
- This technique enhances the understanding of geochemical variations in basalts from different tectonic settings, despite challenges in deciphering magmatic processes.
