Combining impact sensor field and laboratory flume measurements with other techniques for studying fluvial bedload transport in steep mountain streams

• Impact sensors can provide continuous data on the intensity of bedload transport.
• Impact sensors provide high-resolution data on start and end of bedload transport.
• The total mass of all separately moving particles > 11.3 mm can be determined.
• Combined with other methods bedload in supply-limited systems can be quantified.
• Low bedload transport rates in the supply-limited systems Erdalen and Bødalen

The timing and rate of fluvial bedload transport are of central importance within sediment budget studies and in many applications in river science and engineering. During the years 2010, 2011 and 2012 detailed field measurements with portable impact sensors as a non-invasive technique for indirectly determining fluvial bedload transport intensity were conducted in two instrumented and supply-limited drainage basin systems (Erdalen and Bødalen) in the fjord landscape in western Norway. Additional field measurements with portable impact sensors were carried out in 2010 and 2011 in selected transport-limited fluvial systems in the Coast Mountains of western Canada. The collected impact sensor field data were calibrated with laboratory flume experiments. The data from the impact sensor field measurements in western Norway and the flume experiments were combined with field data from continuous discharge monitoring, repeated surveys of channel morphometry and sediment texture, particle tracer measurements, Helley–Smith samplings, underwater video filming and biofilm analyses. The combination of methods and techniques applied provides insights into the temporal variability and intensity of fluvial bedload transport in the selected mountain streams: (i) in the transport-limited systems with generally high bedload transport rates during high discharge and with bedload material moving in clusters over the impact sensor plates, impact sensor data (based on a 1 s measuring interval) provide the opportunity to detect the start and end of bedload transport, thus to identify discharge thresholds for sediment entrainment, and to roughly estimate the intensity and relative intensity of change of bedload transport during the measuring period; (ii) in the supply-limited systems with low bedload transport rates and bedload components moving separately (as single particles) over the impact sensor plates, impact sensor data (with a 1 s measuring interval) allow the detection of the start and end of transport of bedload components > 11.3 mm, thus the identification of discharge thresholds for possible entrainment of particles, the quantification of the number of particles > 11.3 mm moving over the impact sensor plates during the measuring period, the rough estimation of grain sizes of the particles moving separately over the impact sensor plates, and the calculation of the total mass of the bedload material > 11.3 mm moving over the impact sensor plates during the measuring period; (iii) when combined with other methods and techniques (Helley–Smith sampling, particle tracer measurements, biofilm analyses, underwater video filming) which provide information on the active bedload transport channel width, on discharge thresholds for possible entrainment of particles of different grain sizes, and on transport rates of bedload material < 11.3 mm, total rates of fluvial bedload transport, covering all given grain sizes of the bedload material, can be calculated for the supply-limited mountain streams with generally low bedload transport. The higher measured annual bedload yield in Bødalen (13.6 t km− 2 yr− 1) compared to Erdalen (2.6 t km− 2 yr− 1) reflects a higher level of slope–channel coupling in Bødalen than in Erdalen.