Complementing airborne laser bathymetry with UAV-based lidar for capturing alluvial landscapes

Abstract

In this paper we report on a flight experiment employing airborne laser bathymetry (ALB) and unmanned aerial vehicle (UAV) based laser scanning (ULS) for capturing very high resolution topography of shallow water areas and the surrounding littoral zone at the pre-alpine Pielach River in Austria. The aim of the research is to assess how information gained from non-bathymetric, ultra-high resolution ULS can support the ALB data. We focus first on the characterization of the water surface of a lowland river and provide validation results using the data of a topographic airborne laser scanning (ALS) sensor and a low flying ULS system. By repeat ULS survey of a the meandering river reach we are able to quantify short-term water level changes due to surface waves in high resolution. Based on a hydrodynamic-numerical (HN) model we assess the accuracy of the water surface derived from a water penetrating ALB sensor. In the second part of the paper we investigate the ability of ALB, ALS, and ULS to describe the complex topography and vegetation structure of the alluvial area. This is carried out by comparing the Digital Terrain Models (DTM) derived from different sensor configurations. Finally we demonstrate the potential of ULS for estimating single tree positions and stem diameters for detailed floodplain roughness characterization in HN simulations. The key findings are: (i) NIR scan data from ALS or ULS provide more precise water level height estimates (no bias, 1σ: 2 cm) compared to ALB (bias: 3 cm, 1σ: 4 cm), (ii) within the studied reach short-term water level dynamics irrelevant for ALB data acquisition considering a 60 cm footprint diameter, and (iii) stem diameters can be estimated based on ULS point clouds but not from ALS and ALB.