Can Machine-Learning Algorithms Predict the Volcanic Source of Tephra? A Case Study from the South Aegean

We investigate the effectiveness of using machine-learning algorithms to identify the volcanic source of tephra deposits.
Can Machine-Learning Algorithms Predict the Volcanic Source of Tephra? A Case Study from the South Aegean
Like

Share this post

Choose a social network to share with, or copy the URL to share elsewhere

This is a representation of how your post may appear on social media. The actual post will vary between social networks

Read the paper

SpringerLink
SpringerLink SpringerLink

Application of machine-learning algorithms for tephrochronology: a case study of Plio-Quaternary volcanic fields in the South Aegean Active Volcanic Arc - Earth Science Informatics

We performed several machine-learning algorithms on a geochemical dataset including whole-rock (n = 1656) and glass (n = 1092) compositions of lavas and pyroclastics belonging to 8 volcanic fields along the South Aegean Active Volcanic Arc (SAAVA). We did not only test our trained model with the unknown distal tephras, but also controlled its performance using some known distal tephras (e.g., Nisyros-Kyra) from the easternmost part of the SAAVA. The different metrics and kappa values revealed that Naïve Bayes, Linear Discriminant Analysis, Artificial Neural Network, and Support Vector Machine (both probabilistic and non-probabilistic models) were the least performing algorithms; while the Random Forest and the gradient boosting algorithms (e.g., CatBoost, LightGBM) together with their average ensemble (Voting Classifier) were the best for the volcanic-source predictions of tephras. This also indicates that the latter algorithms give better results for the machine-learning applications on an imbalanced geochemical dataset, which was the main artifact in our training model. Despite the accurate prediction and training models especially for those having larger datasets (i.e., Santorini and Nisyros volcanoes), we here would like to express that the machine-learning can be as yet a time-saving tool (not an automatized decision-maker) in the tephrochronology studies providing a more efficient and rapid way of finding the possible volcanic sources for unknown tephras. In this regard, our freely-available Python codes would be easily implemented in further “tephra-hunting” studies in and around the SAAVA. However, there is a need for increasing the available geochemical (e.g., mineral chemistry) and also other interrelated datasets (e.g., geochronology) that should be as yet evaluated manually by the tephrochronologists to be able to improve the performances of machine-learning algorithms in the volcanic-source predictions.

Summary:
This scientific paper investigates the effectiveness of using machine-learning algorithms to identify the volcanic source of tephra deposits.  We used a database of geochemical data from eight volcanic fields in the South Aegean Active Volcanic Arc (SAAVA). We trained different machine-learning algorithms to recognize patterns in the chemical composition of volcanic rocks and tephra. We tested the algorithms' accuracy by asking them to predict the volcanic source of known tephra samples and then compared the results to predictions made using traditional methods.
The study found that certain algorithms, such as Random Forest and gradient boosting algorithms (like XGBoost and LightGBM), were the most accurate in predicting the source of the tephra. These algorithms were particularly effective in handling the imbalanced nature of the dataset, where some volcanic fields had significantly more data points than others. We also found that factors like the amount of data available for each volcanic field and the chemical similarity between different volcanic sources played a role in the algorithms' accuracy.


Why is it important?
Tephrochronology, the study of tephra deposits, is crucial for understanding past volcanic eruptions, their environmental impacts, and for dating geological and archaeological events. Identifying the source of a tephra deposit is often challenging, especially when dealing with deposits far from their source. This study explores the potential of machine learning as a valuable tool to aid tephrochronological studies, potentially saving time and improving the accuracy of volcanic source identification.

Additional Perspectives:
The study acknowledges that while machine learning shows promise in tephra studies, it should not replace traditional methods entirely. Instead, it should be seen as a complementary tool.  The accuracy of these machine-learning models relies heavily on the quality and quantity of data used for training. Therefore, future research should focus on expanding the geochemical database and incorporating other types of data, such as mineral chemistry and geochronological data, to improve the models' accuracy further.


Hashtags:
#volcanoes #geology #tephra #machinelearning #earthscience #SAAVA #archaeology #datingscience #volcaniceruptions #research #science

Please sign in or register for FREE

If you are a registered user on Research Communities by Springer Nature, please sign in

Follow the Topic

Geology
Physical Sciences > Earth and Environmental Sciences > Earth Sciences > Geology
Earth Sciences
Physical Sciences > Earth and Environmental Sciences > Earth Sciences
Machine Learning
Mathematics and Computing > Computer Science > Artificial Intelligence > Machine Learning

Related Collections

With collections, you can get published faster and increase your visibility.

Sustainable Development Goal 13: Climate Action (ESIN)

Sustainable Development Goal 13 (SDG 13) is to limit and adapt to climate change. It is one of 17 Sustainable Development Goals established by the United Nations General Assembly in 2015. The official mission statement of this goal is to "Take urgent action to combat climate change and its impacts". SDG 13 has five targets which are to be achieved by 2030. They cover a wide range of issues surrounding climate action. The first three targets are outcome targets: Strengthen resilience and adaptive capacity to climate-related disasters; integrate climate change measures into policies and planning; build knowledge and capacity to meet climate change. The remaining two targets are means of implementation targets: To implement the UN Framework Convention on Climate Change (UNFCCC), and to promote mechanisms to raise capacity for planning and management. Along with each target, there are indicators that provide a method to review the overall progress of each target. The UNFCCC is the primary international, intergovernmental forum for negotiating the global response to climate change. Papers in this collection address one or more of the 5 targets of SDG 13: 1. Strengthen resilience and adaptive capacity to climate-related disasters 2. Integrate climate change measures into policy and planning 3. Build knowledge and capacity to meet climate change 4. Implement the UN Framework Convention on Climate Change 5. Promote mechanisms to raise capacity for planning and management

Publishing Model: Hybrid

Deadline: Ongoing