A statistical evaluation of the morphological variability of shortening landforms on Mercury

Observations of Mercury from both the Mariner 10 and MESSENGER missions showed that Mercury has a global population of shortening landforms, with several thousands of individual structures identified to date. The accommodation of widespread tectonic shortening is widely regarded to be the result of global contraction—the long, sustained cooling of the interior that has caused the planet to shrink. Shortening landforms on Mercury have been traditionally categorized into three distinct categories: lobate scarps, wrinkle ridges, and high-relief ridges. Although the clearest examples of shortening landforms at the time were used to describe and define these categories qualitatively, later studies showed that shortening landforms on Mercury display morphological characteristics that do not make for a ready classification into one of these “traditional” groups. More recently, other studies have classified shortening landforms based on the terrain that those landforms reside in to avoid generalizing morphology. In this study, we quantitatively assess the shape of shortening landforms by measuring and compiling a suite of 12 morphological parameters for 100 such structures across the planet. These parameters were evaluated for their importance in defining categories using two multivariate statistical analyses, a Principal Component Analysis (PCA) and Linear Discriminant Analysis (LDA). These methods allow us to assess any correlation that the traditional categories, terrain types, or alternative classification schemes have with the variation observed across our set of measurements. Our results show that the morphologic characteristics of shortening landforms on Mercury are not accurately captured by traditionally recognized groups. Instead, shortening landforms fall along a morphological spectrum, where only a few ideal examples of lobate scarps or wrinkle ridges provide clear endmembers. Therefore, despite the frequent use of the terms “lobate scarps” and “wrinkle ridges” in works regarding planetary tectonics, we find that such terminology does not appropriately define the morphology of shortening landforms found on Mercury and may lead to the generalization, or misinterpretation of landforms described as accommodating shortening on Mercury's surface. Future studies should test if a distinction between the landforms is found in the underlying thrust fault systems.