Quantifying fluvial (dis)connectivity in an agricultural catchment using a geomorphic approach and sediment source tracing

PURPOSE: Catchments subject to land clearance, soil tillage and grazing suffer legacy effects from altered sediment and hydrological regimes and from changes in sediment connectivity between hillslopes and channels. Sediment dynamics are routinely investigated using the fallout radionuclides (FRNs) caesium-137 (¹³⁷Cs) and excess lead-210 (²¹⁰Pbₑₓ), which provide information regarding sediment sources and fluvial processes if source types are differentiated with confidence. Downstream transport, mixing and dilution of FRN-labelled fine sediment can obliterate the tracer signal from sources with low initial concentrations, so geomorphic evidence for downstream changes in sediment source types, mixing and fluvial (dis)connectivity should be used with tracers to ascertain the degree of sediment source variation and dominant fluvial processes. MATERIALS AND METHODS: We coupled sediment source tracing with a geomorphic assessment of downstream hydrological, morphological and sedimentological change to quantify key components of fluvial (dis)connectivity in Coolbaggie Creek, a major source of sediment to the Macquarie River, southeastern Australia. RESULTS AND DISCUSSION: Cs and ²¹⁰Pbₑₓ discriminated between the <63-μm fraction of topsoils from agricultural and forested land (mean ± standard error of ¹³⁷Cs is 5.7 ± 1.1 Bq kg⁻¹ and of ²¹⁰Pbₑₓ is 76.2 ± 22.4 Bq kg⁻¹) and subsoils from channel banks and gullies (¹³⁷Cs = 1.5 ± 0.5 Bq kg⁻¹; ²¹⁰Pbₑₓ = 3.6 ± 2.4 Bq kg⁻¹). In-channel sediment was a mix of these two source types (¹³⁷Cs = 0.9 ± 0.3 Bq kg⁻¹; ²¹⁰Pbₑₓ = 18.8 ± 6.8 Bq kg⁻¹). FRN concentrations also declined significantly downstream. Mixing model results show topsoils account for 39 % (95 % confidence interval (CI) = 21–88 %) of fine in-channel sediment in the upper reaches, but this declines to 2 % (CI = 0–19 %) after 25 km, and thereafter, subsoils account for an estimated 100 % of fine sediment in the system. This rapid decrease in topsoil contributions coincides with a downstream change in hydrogeomorphic character, including large increases in channel cross-sectional area (∼20 to >200 m²) and unit stream power (5.4 to 1,382 W m⁻²), and evidence of greater subsoil and reworked sediment contributions from bank and gully erosion. Limited topsoil supply to the trunk stream suggests low catchment erosion rates and reduced connectivity between the catchment and the river. Enlargement and entrenchment of the trunk stream in the lower reaches have resulted in lateral channel–floodplain disconnection, while a sediment slug impedes longitudinal coarse sediment transfer. CONCLUSIONS: Hydrogeomorphic change and sediment source variations downstream explain the short-term sediment dynamics in this agricultural catchment, which has broader implications for understanding sediment transport processes and fluvial (dis)connectivity when interpreting sediment source and tracer data.