The evolution of Earth's climate over geological timescales is linked to surface erosion via weathering of silicate minerals and burial of organic carbon. However, methodological difficulties in reconstructing erosion rates through time and feedbacks among tectonics, climate, and erosion spurred an ongoing debate on mountain erosion sensitivity to tectonic and climate forcing. At the heart of this debate is the question of whether late Cenozoic climate cooling has increased global erosion rates or not. The Himalaya plays a prominent role in this debate as its erosion produces a large fraction of global sediments delivered to ocean basins. We report a 6-Myr-long record of Be10-derived erosion rates from the north-western Himalaya, which indicates that erosion rates in this region varied quasi-cyclically with a period of ∼1 Myr and increased gradually toward the present. We hypothesize that the observed pattern of erosion rates occurred in response to the tectonic growth of the Himalaya by punctuated basal and frontal accretion of rocks from the underthrusting Indian plate and concomitant changes in topography. In this scenario, basal accretion episodically changes rock-uplift patterns, which brings landscapes out of equilibrium and results in quasi-cyclic variations in erosion rates. We used numerical landscape evolution simulations to demonstrate that this hypothesis is physically plausible. We attribute the long-term increase in erosion rates to the erosional response of topography due to frequent basal accretion relative to frontal accretion. Because tectonic accretion processes are inherent to collisional orogenesis, they likely confound climatic interpretations of erosion rate histories.
