The topographic implications deduced from relationships between different river concavity indices

Abstract

River longitudinal profiles are often smoothly concave up. Normalized Concavity Index (NCI) and concavity index ($\theta$) are usually used to quantify the degree of river concavity. These indices are calculated from distinct perspectives, and their interrelationships remain poorly understood. This study examines the relationships between NCI and $\theta$ in the Hulu River, a steady-state basin in the central Loess Plateau of China, using both all-channel and trunk-channel data. Landform evolution was modeled using Landlab to investigate changes in NCI and $\theta$ and their association. We find that (i) Due to the influence of tributary channels, NCI and $\theta$ are calculated using all channel data are larger than those for trunk channel data. $\theta$ calculated from all channels data may not effectively capture topographic implications, i.e., larger $\theta$ indicates more concave profiles. (ii) While NCI and $\theta$ derived from all-channel data show no significant correlation (P < 0.05), these concavity indices of trunk channels are significantly correlated (P < 0.05), indicating that tributary profiles may not be in a steady state even when trunk-channel profiles are in a steady state. (iii) Among model parameters, the drainage area exponent m in the stream power law strongly affects NCI and $\theta$, and their correlations. However, it is important to note that the sensitivity analysis conducted in this study was relatively simple, and further in-depth analysis is needed to fully confirm the extent of the influence of m. Our results highlight the distinct behaviors of NCI and $\theta$ due to difference in their calculation methods and provide guidance for selecting suitable metrics for assessing river concavity.