Time Series Foundation Models and Deep Learning Architectures for Earthquake Temporal and Spatial Nowcasting

Advancing the capabilities of earthquake nowcasting, the real-time forecasting of seismic activities, remains crucial for reducing casualties. This multifaceted challenge has recently gained attention within the deep learning domain, facilitated by the availability of extensive earthquake datasets....

Full description

Saved in:
Bibliographic Details
Main Authors: Alireza Jafari, Geoffrey Fox, John B. Rundle, Andrea Donnellan, Lisa Grant Ludwig
Format: Article
Language:English
Published: MDPI AG 2024-11-01
Series:GeoHazards
Subjects:
Online Access:https://www.mdpi.com/2624-795X/5/4/59
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850242174081302528
author Alireza Jafari
Geoffrey Fox
John B. Rundle
Andrea Donnellan
Lisa Grant Ludwig
author_facet Alireza Jafari
Geoffrey Fox
John B. Rundle
Andrea Donnellan
Lisa Grant Ludwig
author_sort Alireza Jafari
collection DOAJ
description Advancing the capabilities of earthquake nowcasting, the real-time forecasting of seismic activities, remains crucial for reducing casualties. This multifaceted challenge has recently gained attention within the deep learning domain, facilitated by the availability of extensive earthquake datasets. Despite significant advancements, the existing literature on earthquake nowcasting lacks comprehensive evaluations of pre-trained foundation models and modern deep learning architectures; each focuses on a different aspect of data, such as spatial relationships, temporal patterns, and multi-scale dependencies. This paper addresses the mentioned gap by analyzing different architectures and introducing two innovative approaches called Multi Foundation Quake and GNNCoder. We formulate earthquake nowcasting as a time series forecasting problem for the next 14 days within 0.1-degree spatial bins in Southern California. Earthquake time series are generated using the logarithm energy released by quakes, spanning 1986 to 2024. Our comprehensive evaluations demonstrate that our introduced models outperform other custom architectures by effectively capturing temporal-spatial relationships inherent in seismic data. The performance of existing foundation models varies significantly based on the pre-training datasets, emphasizing the need for careful dataset selection. However, we introduce a novel method, Multi Foundation Quake, that achieves the best overall performance by combining a bespoke pattern with Foundation model results handled as auxiliary streams.
format Article
id doaj-art-523b0bfe3a664a3a98d7e6c2aa118ed4
institution OA Journals
issn 2624-795X
language English
publishDate 2024-11-01
publisher MDPI AG
record_format Article
series GeoHazards
spelling doaj-art-523b0bfe3a664a3a98d7e6c2aa118ed42025-08-20T02:00:22ZengMDPI AGGeoHazards2624-795X2024-11-01541247127410.3390/geohazards5040059Time Series Foundation Models and Deep Learning Architectures for Earthquake Temporal and Spatial NowcastingAlireza Jafari0Geoffrey Fox1John B. Rundle2Andrea Donnellan3Lisa Grant Ludwig4Computer Science Department, University of Virginia, Charlottesville, VA 22904, USAComputer Science Department, University of Virginia, Charlottesville, VA 22904, USAPhysics and Astronomy and Geology, University of California, Davis, CA 95616, USAJet Propulsion Laboratory, California Institute of Technology, Pasadena, CA 91109, USADepartment of Population Health & Disease Prevention, University of California, Irvine, CA 92697, USAAdvancing the capabilities of earthquake nowcasting, the real-time forecasting of seismic activities, remains crucial for reducing casualties. This multifaceted challenge has recently gained attention within the deep learning domain, facilitated by the availability of extensive earthquake datasets. Despite significant advancements, the existing literature on earthquake nowcasting lacks comprehensive evaluations of pre-trained foundation models and modern deep learning architectures; each focuses on a different aspect of data, such as spatial relationships, temporal patterns, and multi-scale dependencies. This paper addresses the mentioned gap by analyzing different architectures and introducing two innovative approaches called Multi Foundation Quake and GNNCoder. We formulate earthquake nowcasting as a time series forecasting problem for the next 14 days within 0.1-degree spatial bins in Southern California. Earthquake time series are generated using the logarithm energy released by quakes, spanning 1986 to 2024. Our comprehensive evaluations demonstrate that our introduced models outperform other custom architectures by effectively capturing temporal-spatial relationships inherent in seismic data. The performance of existing foundation models varies significantly based on the pre-training datasets, emphasizing the need for careful dataset selection. However, we introduce a novel method, Multi Foundation Quake, that achieves the best overall performance by combining a bespoke pattern with Foundation model results handled as auxiliary streams.https://www.mdpi.com/2624-795X/5/4/59deep learningearthquake nowcastingfoundation modelstransformerspre-trained modelsgraph neural networks
spellingShingle Alireza Jafari
Geoffrey Fox
John B. Rundle
Andrea Donnellan
Lisa Grant Ludwig
Time Series Foundation Models and Deep Learning Architectures for Earthquake Temporal and Spatial Nowcasting
GeoHazards
deep learning
earthquake nowcasting
foundation models
transformers
pre-trained models
graph neural networks
title Time Series Foundation Models and Deep Learning Architectures for Earthquake Temporal and Spatial Nowcasting
title_full Time Series Foundation Models and Deep Learning Architectures for Earthquake Temporal and Spatial Nowcasting
title_fullStr Time Series Foundation Models and Deep Learning Architectures for Earthquake Temporal and Spatial Nowcasting
title_full_unstemmed Time Series Foundation Models and Deep Learning Architectures for Earthquake Temporal and Spatial Nowcasting
title_short Time Series Foundation Models and Deep Learning Architectures for Earthquake Temporal and Spatial Nowcasting
title_sort time series foundation models and deep learning architectures for earthquake temporal and spatial nowcasting
topic deep learning
earthquake nowcasting
foundation models
transformers
pre-trained models
graph neural networks
url https://www.mdpi.com/2624-795X/5/4/59
work_keys_str_mv AT alirezajafari timeseriesfoundationmodelsanddeeplearningarchitecturesforearthquaketemporalandspatialnowcasting
AT geoffreyfox timeseriesfoundationmodelsanddeeplearningarchitecturesforearthquaketemporalandspatialnowcasting
AT johnbrundle timeseriesfoundationmodelsanddeeplearningarchitecturesforearthquaketemporalandspatialnowcasting
AT andreadonnellan timeseriesfoundationmodelsanddeeplearningarchitecturesforearthquaketemporalandspatialnowcasting
AT lisagrantludwig timeseriesfoundationmodelsanddeeplearningarchitecturesforearthquaketemporalandspatialnowcasting