Back to Search
Start Over
A Machine Learning Based‐Approach to Predict the Water Content of Mid‐Ocean Ridge Basalts.
- Source :
- Geochemistry, Geophysics, Geosystems: G3; Jul2023, Vol. 24 Issue 7, p1-16, 16p
- Publication Year :
- 2023
-
Abstract
- Water is critical in the evolution of the mantle due to its strong influence on the physicochemical properties of mantle rocks. Mid‐ocean ridge basalts (MORBs) are commonly used to study the compositional characteristics of the convecting upper mantle. However, there remains abundant samples in the global MORB data sets without direct measurements of water contents. The commonly observed good correlation between H2O and other incompatible trace components, such as Ce, has been applied to quantify water contents of MORBs. However, this approach assumes constant H2O/Ce in the target samples, which is not always true as the H2O/Ce ratios of MORBs could be rather heterogeneous even in some short ridge segments. Utilizing the present compositional data of global MORB glasses with measured water contents (n = 1,467), we construct a Random Forest Regression model based on machine learning, which can predict water concentrations of samples based on selected major and trace element data, without assuming a ratio between H2O and other trace elements. This model allows us to precisely recover water contents for MORBs with comparable accuracy with traditional analytical methods. The predicted results of MORB glasses from this model (n = 1,931) expand the water content database of global MORBs and indicate a broad existence of high‐H2O MORBs. This new approach allows us to investigate the water content of MORBs from some ridges lacking previous water content measurements (e.g., the Chile Ridge and the Pacific‐Antarctic Ridge) and infer changes in the water content of MORB sources through applying the model to transform fault samples. Plain Language Summary: There are trace amounts of water in the mantle, which can strongly influence the physicochemical properties of the mantle; therefore, exploring the distribution of water is significant for understanding the evolutionary processes of the mantle. Mid‐ocean ridge basalts (MORBs) are derived from the upper mantle, and MORB glasses usually preserve the compositional information (particularly water) of the upper mantle. Based on compiled MORB glass compositional data, we constructed a machine learning‐based regression model, which can precisely predict the water contents of samples by inputting their major and trace element data. Its application largely expands the global MORB geochemical data sets. The new model provides a reliable and low‐cost way to obtain water data from MORB samples and it also provides a new perspective to understand the temporal evolution of transform faults and constrain the water out‐flux through mid‐ocean ridges across the geological time. Key Points: A machine learning‐based model is established to predict water contents of mid‐ocean ridge basalt (MORB) glasses based on their major and trace element dataThe predicted results largely expand the water database of global MORBs with comparable uncertainty of normal instrumental analysis [ABSTRACT FROM AUTHOR]
- Subjects :
- MID-ocean ridges
MACHINE learning
BASALT
GEOLOGICAL time scales
REGOLITH
Subjects
Details
- Language :
- English
- ISSN :
- 15252027
- Volume :
- 24
- Issue :
- 7
- Database :
- Complementary Index
- Journal :
- Geochemistry, Geophysics, Geosystems: G3
- Publication Type :
- Academic Journal
- Accession number :
- 167371545
- Full Text :
- https://doi.org/10.1029/2023GC010984