Spatial variability in river bed porosity determined by nuclear density gauging: A case study from a French gravel-bed river

Porosity is one of the key properties of fluvial sediments. It is defined as the ratio of pore volume to total volume. In river science, porosity is often assumed to be spatially constant, which might be a gross simplification of reality. Ignoring the spatial variations in porosity can cause errors in morphological, ecological, hydrological, hydrogeological and sedimentological applications. Although detailed information about spatial porosity variations can be obtained from porosity measurements at field sites, such information has never been collected where these variations might be important. In this study, field porosity measurements were carried out to quantify the magnitude of the spatial porosity variation for four different sedimentological environments of a braided river: a confluence, a tributary delta, a braid bar and a secondary channel. A nuclear density gauge was used for the measurement of porosity. The nuclear density gauge proved to be a time-saving and labour-saving technique that produces accurate porosity values with a root mean square error of 0.03. The four sedimentological environments showed significant differences in porosity, with mean porosity being lower for confluence and bar than for delta and secondary channel. Semi-variogram analysis showed the absence of any spatial correlation in porosity for distances beyond 4 m. This shows that distance cannot be used as a parameter for porosity extrapolation in a fluvial system unless the extrapolation distance is less than 4 m. At least eight measurements of porosity are required to obtain a reliable estimate of mean porosity in a sedimentary environment, i.e. with uncertainty <0.03. Although grain size characteristics were found to have a significant impact on porosity, the relationships between these parameters and porosity were not very strong in this study. The unique porosity dataset, presented in this article, provides a valuable source of information for researchers and river managers.

Automated summary

This study conducted porosity measurements in different sedimentological environments of a braided river, revealing significant differences in porosity among confluence, bar, delta, and secondary channel. There was no spatial correlation in porosity beyond 4 meters. At least eight porosity measurements are required for a reliable estimate of mean porosity.