Indices of sediment connectivity: opportunities, challenges and limitations

Indices of connectivity are critical means for moving from qualitative to (semi-)quantitative evaluations of material (e.g., water, sediment and nutrients) transfer across the building blocks of a terrestrial system. In geomorphology, compared to closely related disciplines like ecology and hydrology, the development of indices has only recently started and as such presents opportunities and challenges that merit attention. In this paper, we review existing indices of sediment connectivity and suggest potential avenues of development for meeting current basic and applied research needs. Specifically, we focus on terrestrial geomorphic systems dominated by processes that are driven by hydro-meteorological forcing, neglecting seismically triggered events, karstic systems and environments controlled by eolian processes. We begin by setting a conceptual framework that combines external forcings (drivers) and system (intrinsic) structural and functional properties relevant to sediment connectivity. This framework guides our review of response variables suitable for sediment connectivity indices. In particular, we consider three sample applications concerned with sediment connectivity in: (i) soil studies at the plot scale, (ii) bedload transport at the reach scale, and (iii) sediment budgets at the catchment scale. In relation to the set of response variables identified, we consider data availability and issues of data acquisition for use in indices of sediment connectivity. We classify currently available indices in raster based, object or network based, and indices based on effective catchment area. Virtually all existing indices address the degree of static, structural connectivity only, with limited attention for process-based, functional connectivity counterparts. Most recent developments in indices of sediment connectivity deal, to some extent, with different styles of anthropogenic and hydro-meteorological forcings and with the temporal variability of sediment connectivity, by incorporating additional variables and parameters in existing indices. We believe that, in order to use structural connectivity as explanatory or predictive tool, indices need to be interpretable in relation to geomorphic processes, material properties, and forcing styles and magnitude-frequency spectra. Improvements in this direction can be made through studies shaped to constrain structural-functional correlations across a range of hydro-meteorological scenarios, for example employing fieldbased techniques such as particle tracking and sediment provenance analysis, as well as numerical simulations. We further consider existing indices in relation to spatial and temporal scales. The latter have immediate implications on the distinction and application between indices and models of sediment connectivity. In this context, we suggest that sediment connectivity over millennial or longer time scales should be dealt with models, as opposed to indices