Authors: Juha-Matti Välimäki, Eliisa Lotsari & Tuure Takala
Publication type: A1
Bedload transport rate describes the amount of sediment moving within the active layer of a riverbed. Traditional mechanical sampling methods are labor-intensive, provide limited spatial and temporal resolution, and may introduce substantial uncertainty by disturbing local hydraulic conditions. Although computer vision–based image velocimetry methods have shown promise for quantifying bedload transport from video data, their use in ice-covered field conditions remains largely unexplored.This study aimed to (A) quantify bedload transport rates and improve process understanding by comparing traditional mechanical sampling, ADCP bottom tracking, and image-processing methods, and (B) characterize the pulsating behavior and spatial variability of bedload transport under ice-covered conditions. Bedload velocity was estimated using a background subtraction algorithm combined with Large-Scale Particle Image Velocimetry (LSPIV) applied to GoPro Hero 12 video recorded during a midwinter 2024 field campaign in a subarctic river. Reference bedload transport was measured using a Helley-Smith pressure-difference sampler, while reference velocity data were collected with an Acoustic Doppler Current Profiler (ADCP) and an Acoustic Doppler Velocimeter (ADV).The results indicate that image-based measurements are a promising tool for improving understanding of sediment transport processes in ice-covered rivers. Compared with pressure-difference samplers, the method provided more detailed temporal and spatial information and yielded consistent estimates of both bedload transport rate and bedform transport. In the companion paper (Part II), improved image velocimetry method is proposed and applied to the same dataset.