Effects of differential rainfall on the dynamics of landscape evolution

The field of landscape evolution modeling focuses on the evolution of the structure of landscapes over geologic time. The elevations of rivers, valleys, hills, and mountains make up the structure of the landscape. Generally, these elevations change due rock uplift as well as incision. Rivers are the major driving force for incision in the landscape, and the amount that a river incises is greatly dependent on the amount of discharge flowing through it. Discharge is dependent on the structure of the drainage network in the landscape. Therefore, it is important to understand drainage networks in order to understand how landscapes evolve. Dendritic drainage networks are the most general form that exists in the environment. Some researchers hypothesize that the dendritic form is the optimal form of a drainage network. In order for this to be true, we postulate that landscapes seeded initially with non-dendritic drainage networks must be able to convert to dendritic drainage networks through reorganization. So far, some physically scaled models have been able to capture drainage network reorganization; however, no numerical models have been capable of doing so. In this study, we demonstrate the limitations of current landscape evolution models reorganize drainage networks. Through 1D analysis, we analyze the general behavior of the stream power law, and using a 2D numerical model, we conclude the following: (1) The most general landscape evolution model tends towards a static steady state, where elevations no longer change. (2) Landscape evolution models are very sensitive to their initial conditions. (3) Because landscapes tend toward a static steady state, and they are sensitive to their initial conditions, we demonstrate that initial conditions seeded with non-dendritic signals tend towards a static steady state that contain non-dendritic drainage networks signals. (4) Randomized perturbations in the spatial distribution of rainfall are not enough to reorganize the drainage network. (5) Systematic perturbations in precipitation are necessary to reorganize the drainage network. We conclude that drainage networks described by our model can indeed be reorganized using differential spatial rainfall patters; however, the perturbations applied in this study were generally deliberate, extreme, and had no physical bases. For future studies, we propose to understand the coupling between precipitation and the landscape’s topography in order to study how landscape reorganize themselves