Global Quantification of Glacial versus Fluvial Erosion Rates: Limits to Timescale Dependence

Geologists have long debated the erosive power of glaciation. At one extreme, glaciers and ice sheets have been viewed as non-erosive protective blankets, while at the other extreme, they are considered among Earth’s most potent erosive forces. As such, the question of whether glaciers or rivers are, on average, more effective erosive agents remains contentious. This problem is further complicated by the “Sadler effect,” which describes how erosion and deposition rates appear to decrease as the timescale of observation increases, posing a considerable challenge for directly comparing glacial and fluvial erosion rates. The Sadlerian dilemma is particularly relevant to discussions of Cenozoic paleoclimate, where intensified erosion due to enhanced glaciation has been argued to be both a cause and consequence of Cenozoic cooling. Here, we present a comprehensive global compilation of glacial and fluvial erosion rates supplemented by simple numerical experiments. Our analysis shows that globally averaged glacial erosion rates surpass fluvial erosion rates by an order of magnitude (0.51 mm/year vs. 0.067 mm/year), a difference that cannot be attributed to the Sadler effect. We further demonstrate that the Sadler effect arises from three distinct biases: a measurement thickness bias (primarily related to the average penetration depth of cosmogenic rays into rocks), an erosion-redeposition bias (reflecting the bidirectional nature of certain geological processes whereby previous progress may be undone), and a non-observation bias (resulting from unobserved erosionally or depositionally quiescent intervals). Taken together, these findings support the notion that erosion rates increased with Cenozoic cooling and glaciation, highlighting the global importance of glacial erosion across diverse timescales.