What does nature know that engineers do not? Lessons from natural rock arches for rock mass classification and design

Summary

Abstract

Natural rock arches have persisted over geomorphological timescales, encoding in their continued existence a form of knowledge about long-term rock mass behaviour that engineering practice has yet to fully articulate. This paper critically examines whether empirical tools developed for engineering contexts can explain what nature has already demonstrated. Applying the Q-system maximum unsupported span chart and RMR stand-up time diagram to a database of 100 arches from the Natural Arch and Bridge Society (NABS), we find that estimated Q values would require correction by up to two orders of magnitude to satisfy unsupported span criteria, while RMR-based predictions would demand ratings of 85-100 for rock masses that are observably fractured, blocky, and weathered. At the same time, many natural arches behave as slender beams supported by rock pillars, a failure mechanism that classification systems were never designed to recognize. The results suggest that engineering practice is simultaneously overconservative in its support recommendations and underconservative in its understanding of the actual failure mechanisms at play. The question this paper ultimately poses is a simple but uncomfortable one: what does nature know that engineers do not? In the end, all that we see or seem is but a rock bridge within a rock bridge, and therein lies the challenge. The stability of natural arches (i.e., natural rock bridges) is sustained by mechanisms that current design methods were never built to see, let alone quantify. The objective of this paper is to open a discussion on what it would take for rock engineering practice to finally account for what has been holding these structures standing all along.