Hydro‐Geomorphic Metrics for High Resolution Fluvial Landscape Analysis. Issue 3 (21st March 2022)

• Record Type: Journal Article
• Authors: Bernard, Thomas G.; Davy, Philippe; Lague, Dimitri
• Is Part Of: Journal of geophysical research. Volume 127:Issue 3(2022)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2022-03-21
• Subjects: Geomorphology -- Periodicals; 551.3
• DOI: 10.1029/2021JF006535
• Languages: English
• ISSNs: 2169-9003

Abstract

Abstract: Topographic metrics are designed to quantify scale‐relevant relationships between geometric properties of landscapes to reveal the processes shaping them. They have long been derived from topographic flow routing algorithms, initially developed for coarse Digital Elevation Models (DEMs), whose resolution (≥30 m) and poor precision did not resolve correctly flow patterns and channel flow width. Since high resolution and precision DEMs make the description of meter‐scale flow patterns possible, new methods are required to analyze high resolution landforms structures such as hillslope‐channel connections, channel width or floodplains. Here, we investigate the potential of 2D hydraulic simulations based on the shallow water equations to replace the classical slope versus drainage area analysis, to analyze river morphology and to identify floodplains. We apply the Floodos model to the 1 m resolution DEM of the Elder Creek catchment, California, from which we derive three hydro‐geomorphic metrics accounting for the river geometry: a specific drainage area extended to channels, an effective flow width and the hydraulic slope. We analyze the Elder Creek catchment through what we call the hydraulic slope‐area diagram allowing a better identification of hillslope‐channel connections than the slope‐area approach. The effective flow width is analyzed along the drainage network and is characterized by a power‐law relationship consistent with previous observations. We derive metrics based on a multi‐runoff approach to automatically identify floodplains and evaluate along‐stream variations in hydraulic geometry. The hydro‐geomorphic metrics offer a geomorphic analysis suitable for high resolution DEMs and opens up new perspectives in fluvial landscape analysis. Plain Language Summary: Landform analysis is essential to identify and understand the processes shaping landscapes. It is traditionally based on topographic metrics following surface flow paths designed to reveal surface processes and to characterize the spatial organization of landscapes through scale‐relevant relationships. Current methods to compute these metrics are not able to resolve both high resolution flow patterns and width. New methods are required to analyze fine scale landforms structures such as hillslope‐channel connections, channel width or floodplains. We investigate the potential of 2D hydraulic simulations, which consists in simulating water depth and velocity on the topographic surface, to analyze such landform structures. We applied the Floodos model to a meter‐scale resolution topographic data from the Elder Creek catchment in California. We demonstrate that 2D hydraulic simulation allows to better identify hillslope‐channel connections than traditional approach and to easily estimate river flow width through the use of derived hydro‐geomorphic metrics. The analysis of these metrics at various input discharges enables to support description of floodplains and along‐stream variations in hydraulic geometry allowing a richer description of landscape organization. The use of 2D hydraulic simulation opens up new perspectives to analyze high resolution landform structures crucial in the understanding of processes controlling landscape evolution. Key Points: We use a 2D hydraulic model on a 1 m Digital Elevation Model and derive hydro‐geomorphic metrics of slope, drainage area and river width Analyses of scaling laws based on these metrics provide a better identification of hillslopes and channels than a traditional approach A multi‐runoff approach allows to characterize the hydraulic geometry along the river network and to identify floodplains