A novel approach to the classification of terrestrial drainage networks based on deep learning and preliminary results on solar system bodies

Carlo Donadio,Michele Delli Veneri,Alessia Riccardo,Giuseppe Riccio,Massimo Brescia,Giuseppe Angora

doi:10.1038/s41598-021-85254-x

Abstract

Several approaches were proposed to describe the geomorphology of drainage networks and the abiotic/biotic factors determining their morphology. There is an intrinsic complexity of the explicit qualification of the morphological variations in response to various types of control factors and the difficulty of expressing the cause-effect links. Traditional methods of drainage network classification are based on the manual extraction of key characteristics, then applied as pattern recognition schemes. These approaches, however, have low predictive and uniform ability. We present a different approach, based on the data-driven supervised learning by images, extended also to extraterrestrial cases. With deep learning models, the extraction and classification phase is integrated within a more objective, analytical, and automatic framework. Despite the initial difficulties, due to the small number of training images available, and the similarity between the different shapes of the drainage samples, we obtained successful results, concluding that deep learning is a valid way for data exploration in geomorphology and related fields.

Highlights

Several approaches were proposed to describe the geomorphology of drainage networks and the abiotic/biotic factors determining their morphology
The hydrographic networks of the Earth represent morphologies of the continental landscape characterized by variable dimensions from hundreds of meters to thousands of kilometers and complex geometry, derived by the mutual interaction of numerous physical, biotic and anthropic factors evolving in space and time due to a non-linear physics[1,2]
This favored the possibility of carrying out a comparative analysis of drainage patterns on different bodies of the Solar System, classifying them directly from the images, fully combining the experience acquired on Earth and the effectiveness of recent data-driven methodologies

Summary

Introduction

Several approaches were proposed to describe the geomorphology of drainage networks and the abiotic/biotic factors determining their morphology. In very different climatic and geological-structural contexts, characterized by active tectonics or volcanism, similar patterns are often observed, due to the morphological convergence and morpho-selection phenomena These considerations apply to geometries present in environments very different from terrestrial ones, such as in regions of some Solar System planets and satellites, e.g., Mars and T itan[6,7,8,9]. Unlike the Earth, in the other two cases the lack of telluric phenomena in the recent past, together with the absence of movements of the tectonic plates, avoided the succession of deviations and anomalies of the river paths, which instead have controlled the modeling of many drainage patterns on the Earth This means that on the surface of both Mars and Titan, mainly mechanical erosion processes acted. This favored the possibility of carrying out a comparative analysis of drainage patterns on different bodies of the Solar System, classifying them directly from the images, fully combining the experience acquired on Earth and the effectiveness of recent data-driven methodologies

Objectives

Findings

Discussion

Conclusion