Skip to Main Content
Book Chapter

Lava lake thermal pattern classification using self-organizing maps and relationships to eruption processes at Kīlauea Volcano, Hawai‘i

By
A.M. Burzynski
A.M. Burzynski
Department of Earth and Atmospheric Science, University of Northern Colorado, Greeley, Colorado 80639, USA
Search for other works by this author on:
S.W. Anderson
S.W. Anderson
Department of Earth and Atmospheric Science, University of Northern Colorado, Greeley, Colorado 80639, USA
Search for other works by this author on:
K. Morrison
K. Morrison
School of Mathematical Sciences, University of Northern Colorado, Greeley, Colorado 80639, USA
Search for other works by this author on:
M.R. Patrick
M.R. Patrick
U.S. Geological Survey Hawaiian Volcano Observatory, Hawai‘i National Park, Hawai‘i 96718, USA
Search for other works by this author on:
T. Orr
T. Orr
U.S. Geological Survey Hawaiian Volcano Observatory, Hawai‘i National Park, Hawai‘i 96718, USA
Search for other works by this author on:
W. Thelen
W. Thelen
U.S. Geological Survey Hawaiian Volcano Observatory, Hawai‘i National Park, Hawai‘i 96718, USA
Search for other works by this author on:
Publication history
08 December 201719 September 2018

ABSTRACT

Kīlauea Volcano’s active summit lava lake posed hazards to downwind residents and over 1.6 million Hawai‘i Volcanoes National Park visitors each year during 2008–2018. The lava lake surface was dynamic; crustal plates separated by incandescent cracks moved across the lake as magma circulated below. We hypothesize that these dynamic thermal patterns were related to changes in other volcanic processes, such that sequences of thermal images may provide information about eruption parameters that are sometimes difficult to measure. The ability to learn about concurrent gas emissions and seismic activity from a remote thermal time-lapse camera would be beneficial when conditions are too hazardous for field measurements. We applied a machine learning algorithm called self-organizing maps (SOM) to thermal infrared time-lapse images of the lava lake collected hourly over 23 April–21 October 2013 (n = 4354). The SOM algorithm can take thousands of seemingly different images, each representing the spatial distribution of relative temperature across the lava lake surface, and group them into clusters based on their similarities. We then related the resulting clusters to sulfur dioxide emissions and seismic tremor activity to characterize ties between the SOM classification and different emplacement conditions. The SOM classification results are highly sensitive to the normalization method applied to the input images. The standard ­pixel-by-pixel normalization method yields a cluster of images defined by the highest observed SO2 emission levels, elevated surface temperatures, and a high proportion of cracks between crustal plates. When lava lake surface patterns are isolated by minimizing the effect of temperature variation between images, relationships with seismic tremor activity emerge, revealing an “intense spatter” cluster, characterized by unstable, broken-up crustal plate patterns on the lava lake surface. This proof of concept study provides a basis for extending the SOM classification method to hazard forecasting and real-time volcanic monitoring applications, as well as comparative studies at other lava lakes.

You do not currently have access to this article.

Figures & Tables

Contents

GSA Special Papers

Field Volcanology: A Tribute to the Distinguished Career of Don Swanson

Geological Society of America
Volume
538
ISBN electronic:
9780813795386

GeoRef

References

Related

Citing Books via

Close Modal
This Feature Is Available To Subscribers Only

Sign In or Create an Account

Close Modal
Close Modal