Wet Channel Network Extraction based on LiDAR Data

Abstract:

The temporal dynamics of stream network is vitally important for understanding hydrologic processes including groundwater interactions and hydrograph recessions. However, observations are limited on flowing channel heads, which are usually located in headwater catchments and under canopy. Near infrared LiDAR data provides an opportunity to map the flowing channel network owing to the fine spatial resolution, canopy penetration, and strong absorption of the light energy by the water surface. A systematic method is developed herein to map flowing channel networks based on the signal intensity of ground LiDAR return, which is lower on water surfaces than on dry surfaces. Based on the selected sample sites where the wetness conditions are known, the signal intensities of ground returns are extracted from the LiDAR point data. The frequency distributions of wet surface and dry surface returns are constructed. With the aid of LiDAR-based ground elevation, the signal intensity thresholds are identified for mapping flowing channels. The developed method is applied to Lake Tahoe area based on eight LiDAR snapshots during recession periods in five watersheds. A power-law relationship between streamflow and flowing channel length during the recession period is derived based on the result.