Extending the limits of divide migration criteria: A downscale approximation

Over recent decades, innovative quantitative methods have been developed to study the internal dynamics of river basins. Most of these approaches assume stable or minimally drainage divide mobility (DDM), based on the premise that the rate of river network evolution is significantly faster—by several orders of magnitude—than the mobility of drainage divides. However, recent studies challenge this assumption, showing that the Gilbert metrics can effectively measure DDM and reveal scenarios where drainage divides shift more rapidly than river network adjustments. This dynamic complicates the interpretation of river profiles. In particular, χ-values near headwaters have been identified as indicators of DDM, with low χ-value catchments tending to migrate towards those with higher χ-values. Yet, this interpretation hinges on the premise that mean rock uplift rates, erodibility, and base level heights are consistent across the system. Over geologic timescales, such uniformity is improbable across entire mountain ranges, thereby limiting the universal applicability of this tool. Therefore, thoughtful evaluation of potential DDM is crucial for understanding landscape evolution and also how drainage divide contains information on past climatic and tectonic forcings itself. In this study, we focus on an endorheic watershed in the hyperarid core of the Atacama Desert, northern Chile. The hyperarid conditions present in the Atacama Desert provides a unique opportunity to capture geomorphic signatures as far back as the Paleogene evolutionary stages. We propose a downscaling analysis of DDM applied to three local-scale watershed divides. By integrating geomorphological mapping, morphometric indices, and existing geochronological data, we aim to evaluate DDM across representative sections at a local scale. Furthermore, these results will be validated using forward landscape evolution modeling in Landlab 2.0. Our findings are expected to provide quantitative estimates by comparing widely recognized metrics. We also emphasize the utility of χ-values and Gilbert metrics in sub-catchments to decipher local-scale landscape changes within a basin characterized by multiple evolutionary stages. This downscaling approach refines our capacity to interpret and forecast landscape evolution with exceptional spatial precision, even within the context of an extreme climatic environment, offering profound insights into the complex interactions among tectonic, climatic, and geomorphic processes.