Earth Surface Dynamics (Jun 2020)

Parameterization of river incision models requires accounting for environmental heterogeneity: insights from the tropical Andes

  • B. Campforts,
  • B. Campforts,
  • B. Campforts,
  • V. Vanacker,
  • F. Herman,
  • M. Vanmaercke,
  • W. Schwanghart,
  • G. E. Tenorio,
  • G. E. Tenorio,
  • P. Willems,
  • G. Govers

DOI
https://doi.org/10.5194/esurf-8-447-2020
Journal volume & issue
Vol. 8
pp. 447 – 470

Abstract

Read online

Landscape evolution models can be used to assess the impact of rainfall variability on bedrock river incision over millennial timescales. However, isolating the role of rainfall variability remains difficult in natural environments, in part because environmental controls on river incision such as lithological heterogeneity are poorly constrained. In this study, we explore spatial differences in the rate of bedrock river incision in the Ecuadorian Andes using three different stream power models. A pronounced rainfall gradient due to orographic precipitation and high lithological heterogeneity enable us to explore the relative roles of these controls. First, we use an area-based stream power model to scrutinize the role of lithological heterogeneity in river incision rates. We show that lithological heterogeneity is key to predicting the spatial patterns of incision rates. Accounting for lithological heterogeneity reveals a nonlinear relationship between river steepness, a proxy for river incision, and denudation rates derived from cosmogenic radionuclide (CRNs). Second, we explore this nonlinearity using runoff-based and stochastic-threshold stream power models, combined with a hydrological dataset, to calculate spatial and temporal runoff variability. Statistical modeling suggests that the nonlinear relationship between river steepness and denudation rates can be attributed to a spatial runoff gradient and incision thresholds. Our findings have two main implications for the overall interpretation of CRN-derived denudation rates and the use of river incision models: (i) applying sophisticated stream power models to explain denudation rates at the landscape scale is only relevant when accounting for the confounding role of environmental factors such as lithology, and (ii) spatial patterns in runoff due to orographic precipitation in combination with incision thresholds explain part of the nonlinearity between river steepness and CRN-derived denudation rates. Our methodology can be used as a framework to study the coupling between river incision, lithological heterogeneity and climate at regional to continental scales.