Evaluating the fidelity of detrital thermochronology as a recorder of catchment-scale exhumation

Detrital low-temperature thermochronology has become a widely used tool to infer source-area exhumation histories and catchment-averaged erosion rates. However, a fundamental question remains insufficiently explored: to what extent do detrital thermochronological age distributions faithfully record the full spectrum of bedrock exhumation events within a drainage basin? In particular, it is unclear which exhumation phases are robustly captured, which are selectively amplified, and which may be systematically filtered by geomorphic and fluvial processes. Here we address this problem using the Lhasa river catchment in southern Tibet as a natural laboratory, where abundant bedrock low-temperature thermochronological data coexist with multiple detrital samples from tributaries and the trunk stream. Our approach treats detrital signals as the outcome of a transfer process from bedrock exhumation to river sediments, rather than as a direct proxy. We first compile and statistically characterize bedrock-derived exhumation phases using age-elevation relationships and cooling-path constraints, which serve as physically grounded prior information. Detrital age distributions are then binned consistently with these bedrock exhumation phases, allowing a direct comparison of their relative importance. To quantify potential biases, we develop a bedrock-detrital transfer framework that compares the expected contribution of each exhumation phase, parameterized by its spatial extent and inferred exhumation rate, with its observed detrital fraction. This enables us to identify amplified versus suppressed exhumation signals. We further evaluate the role of geomorphic filtering by integrating catchment-scale metrics, including channel steepness index, hypsometric integrals, and relief, as proxies for erosion and sediment transport efficiency. Finally, we apply a Bayesian forward-inverse framework to estimate catchment-averaged exhumation rates from detrital thermochronological data, partially calibrated by bedrock constraints, and assess under which geomorphic conditions these estimates converge with bedrock-derived exhumation rates. Our results aim to provide a quantitative framework for interpreting detrital thermochronology and for assessing when, and why, detrital records succeed, or fail to capture source-area exhumation histories.