Automated neural network identification of cirques

ABSTRACT

Morphological characteristics of cirques have been studied for decades; however, no repeatable set of metrics has been derived that can consistently identify them. Perhaps more importantly, there is no consensus definition of the form that distinguishes cirques and clusters of cirques from non-cirques. In our approach, we use Shuttle Radar Topography Mission (SRTM) digital elevation models (DEMs) in a Convolutional Neural Network (CNN) framework to identify cirques in 20 mountain ranges globally. We extracted bounding boxes of cirques in 19 of these study areas and used them to develop a training set for a cirque identification model. The trained model was applied to the Sierra Nevada California to assess whether this algorithmic approach derived from a global dataset could produce consistent results in complex terrain with mutually interacting cirque forms. Using commonalities revealed using this approach, we find that there is a basic, recognizable and morphometrically quantifiable cirque form. This approach can be used to automate the identification of cirque locations and to guide the quantification of cirque form independent of the subjective definitions of individual workers. The approach can also be used to understand cirque form under different environmental conditions, including similar forms on Mars.

KEYWORDS:

• Cirque
• morphometry
• convolutional neural network
• digital elevation model
• fuzzy landforms

Acknowledgments

This work was conducted while LAS was on sabbatical leave from the University of New Mexico. The authors would like to thank the Department of Earth and Planetary Sciences at UNM for computing support. LAS and TN would also like to acknowledge the support of TN under NASA 80NSSC19K1676. We would also like to acknowledge Dr. Ian Evans and an anonymous reviewer who provided insightful reviews that materially improved the manuscript.

Disclosure of potential conflicts of interest

The authors declare that they have no competing interests or financial, commercial, legal, or professional relationships with other organizations, or with the people working with them, that could influence our research.

Dedication to Antony Orme

Many years ago, I was an undergraduate student who, after a month-long backpack in California’s Sierra Nevada, had become totally enthralled by glacial landforms. Wanting to know more I enrolled in Tony’s Geomorphology class at UCLA in 1974. That experience and subsequent classes, along with Tony’s willingness to mentor me, changed my career path. I remember the long conversations with him about the geomorphic characteristics of cirques and the processes that produced them. Much of the content of those discussions became embodied in my master’s thesis. Tony was especially interested in how geomorphologists defined cirques and other landforms. This paper attempts to distill those conversations and more recent work into the essence of what we currently believe are key issues with delimiting cirques and defining cirque morphology. In this paper, we attempt to update our understanding of cirques using techniques and tools that Tony and I could barely imagine back in the 1970s. As Tony noted in his acceptance of the Melvin G. Marcus Distinguished Career Award, “ … change was coming – in the guise of plate tectonics, revitalized concepts of climate change, dating techniques, process studies, quantitative analysis, remote sensing, and computer technology. Such changes were essential to the refurbishment of geomorphology, as we know it today. Future changes are inevitable as fresh generations of scholars” (Orme, 2002).

I close this dedication with the words of another great scientist. These words are as appropriate now for the work presented herein as they were when Tony introduced Lord Kelvin’s writings to me early in my career.

“In physical science, a first essential step in the direction of learning any subject is to find principles of numerical reckoning and practicable methods for measuring some quality connected with it. I often say that when you can measure what you are speaking about and express it in numbers you know something about it; but when you cannot measure it, when you cannot express it in numbers, your knowledge is of a meagre and unsatisfactory kind: it may be the beginning of knowledge, but you have scarcely, in your thoughts, advanced to the stage of science, whatever the matter may be.” Thomson (1891).

Louis A. Scuderi

Additional information

Funding

This work was supported by the National Aeronautics and Space Administration [80NSSC19K1676].