Through‐Water Close Range Digital Photogrammetry in Flume and Field Environments

Measurement of the structure of gravel‐bed river surfaces is crucial for understanding both bed roughness and the sediment entrainment process. This paper describes the use of close range digital photogrammetry to measure and monitor change occurring in submerged river gravel‐beds in both flume and field environments. High‐resolution digital elevation models (DEMs) were obtained and two‐media (through air and water) techniques were used to correct for the e.ects of refraction at the air/water interface. Although suitable refractive models have been developed, the use of proprietary software to generate DEMs automatically introduces the problem of how to re‐establish collinearity. A simple refraction correction algorithm based upon analytical geometry was developed and is described. This algorithm was designed for use after initial DEM acquisition and allows any photogrammetric software package to be used for data acquisition. Application of this algorithm led to improvements in DEM accuracy by reducing the systematic, depth‐dependent bias caused by refraction Research carried out in a flume environment allowed the algorithm to be tested by measuring a flooded and drained bed surface. Non‐systematic differences between the “dry” and “wet” DEMs arose from reductions in stereomatching success in the two‐media case. This effect was thought to be due to light attenuation and the introduction of residual parallax. Results suggest that close range digital photogrammetry can be used to extract high quality DEMs of submerged topography in both flume and field fluvial environments, which represents a particularly exciting development for fluvial geomorphologists