Large-Scale Tectonic Forcing of the African Landscape Revealed by Drainage Analysis.

Inverse modeling of an inventory of 1500 longitudinal river profiles across Africa permits recovery of a Neogene uplift rate history, which agrees closely with independent geological constraints. Forward simulations using the Badlands model, benchmarked against the Landlab model, are forced using this uplift history in order to test inherent limitations of existing inverse models (e.g. static catchments and drainage divides, constant precipitation rate). First, given the assumption of constant precipitation, naturalistic forward models closely reproduce observed topography, drainage network patterns, river profiles and sedimentary flux patterns. Secondly, spatial variation in precipitation based on latitudinally-controlled patterns is applied to forward models. The results of these models are in agreement with those run using a constant precipitation assumption. Thirdly, we apply variable temporal periodicities in combination with the spatially varying pattern of precipitation. Predicted elevation, drainage network planforms, and sedimentary flux histories are relatively insensitive to these severe precipitational tests through space and time. Inversion of river profiles produced by these forward simulations (with no inherent 'knowledge' of modern observed drainage patterns), reproduces the input uplift history. This result holds even if the constant precipitation assumption is applied for inverse modeling of drainage networks produced in Badlands under variable precipitation. Thus the precipitational forcing is shown to have a secondary effect on the recoverability of uplift histories from drainage networks.