Controls of climate, topography, vegetation and lithology on drainage density extracted from high resolution topography

Abstract:

Drainage density is a useful topographic metric that varies as a function of geomorphic processes and that serves to quantify links with topography, climate, vegetation, and lithology. Here we analyze 101 sub-basins across thirteen states in the USA using high-resolution digital terrain models (DTMs) in combination with data on the spatial variation of precipitation, soil, geology, and land cover. We test the following hypotheses: (1) Drainage density carries strong, codependent signatures of rainfall variability, soil type, and topographic relief; (2) Drainage density reflects the extent of landscape dissection on the sub-catchment scale and the subsequent processes of vegetation recovery and gullying.

We employ a dimensionless drainage density (Ddd) metric defined as the ratio of likely channelized pixels in a basin to its total number of pixels, and map this metric across meter-resolution lidar DTMs using GeoNet [Passalacqua et al., 2010]. We assess the resolution-dependent scaling of Ddd and observe that it is a much weaker scaling function of DTM resolution than the dimensional formulation of drainage density (Dg), which is classically defined as the ratio of total channel length to total basin area.

In order to characterize the correlation structure of drainage density with climatic parameters such as mean annual precipitation (MAP), we use a Gaussian mixture model and identify two sub-groups of landscapes that display different correlations. We observe that Ddd and MAP are negatively correlated in arid and semi-arid environments and positively correlated in humid environments. The transition occurs at a MAP around 900-1000mm/yr and coincides with the maximum observed values of soil thickness and available water content. Landscape relief has a negative correlation with Ddd in arid environments while the correlation is positive in humid climates. We discuss the implication of our results for understanding eco-geomorphic processes and for modeling landscape evolution.

References:

Passalacqua, P., Do Trung, T., Foufoula‐Georgiou, E., Sapiro, G., & Dietrich, W. E. (2010). A geometric framework for channel network extraction from lidar: Nonlinear diffusion and geodesic paths. Journal of Geophysical Research: Earth Surface (2003–2012), 115(F1).