Rivers are the fluvial conveyor belts routing sediment across the landscape. While there are proper techniques for continuous estimates of the flux of suspended solids, constraining bedload flux is much more challenging, typically involving extensive measurement infrastructure or labor‐intensive manual measurements. Seismometers are potentially valuable alternatives to in‐stream devices, delivering continuous data with high temporal resolution on the average behavior of a reach. Two models exist to predict the seismic spectra generated by river turbulence and bedload flux. However, these models require estimating a large number of parameters and the spectra usually overlap significantly, which hinders straightforward inversion. We provide three functions contained in the R package “eseis” that allow generic modeling of hydraulic and bedload transport dynamics from seismic data using these models. The underlying Monte Carlo approach creates lookup tables of potential spectra, which are compared against the empirical spectra to identify the best fitting solutions. The method is validated against synthetic data sets and independently measured metrics from the Nahal Eshtemoa, Israel, a flash flood‐dominated ephemeral gravel bed river. Our approach reproduces the synthetic time series with average absolute deviations of 0.01–0.04 m (water depth, ranging between 0 and 1 m) and 0.00–0.04 kg/sm (bedload flux, ranging between 0 and 4 kg/sm). The example flash flood water depths and bedload fluxes are reproduced with respective average deviations of 0.10 m and 0.02 kg/sm. Our approach thus provides generic, testable, and reproducible routines for a quantitative description of key metrics, hard to collect by other techniques in a continuous and representative manner.

The monitoring of bedload flux under flash flood conditions has been successfully achieved since 1992 using slot samplers in the semiarid Nahal Eshtemoa. In the present study, a surrogate bedload monitoring techniquethe Japanese plate microphonehas been deployed and calibrated against data from the slot samplers. Since a slot sampler has a sensitivity threshold that becomes especially important when transport rates are low, different averaging periods should be considered for high and low fluxes. In order to overcome the deficiencies of time-based aggregation used hitherto, we have developed a new method involving mass aggregation and commensurably variable intervals, thereby enabling a more accurate analysis and optimizing the bedload sampler's capabilities. The data derived with this new method has then been utilized to calibrate the Japanese plate microphone. The Eshtemoa is an ephemeral gravel bed channel with a high proportion of fine gravel (< 0.02m); for these conditions, acoustic sensors have not been calibrated as yet. Two multiple linear regression models incorporating the effect of median bedload grain size on pulse rate have been established to predict bedload flux and cumulative transported bedload mass. The coefficients in these models are statistically significant. Good predictions are obtained for bedload flux (adj. r 2 = 0.83) and for cumulative bedload mass (adj. r 2 = 0.98) during flood recession. Overall, the multiple linear regression models, used in conjunction with the mass aggregation method of estimating bedload flux, suggest that field calibration of acoustic devices is feasible under these conditions for ca. 90% of the duration of bedload transport.

Rivers are key features of ecosystems, transferring water, dissolved, and particulate matter across the Earth's surface. Driven by the power of moving water, sediment helps rivers to shape landscapes and contribute to the evolution of river morphology (Leopold et al., 1964). Sediment in rivers is either carried as suspended load or as bedload (rolling, sliding, or saltating on the bed). Bedload contributes to channel changes, such as creating micro-and macroforms, narrowing, widening, shifting, aggrading, and degrading. It also affects riverbed and bank stability (Little & Mayer, 1976). From an engineering perspective, bedload transport and the channel changes that it causes can damage infrastructure and threaten near channel human activities (Badoux et al., 2014;Kondolf et al., 2002). Predictive models are needed to correctly constrain and understand the evolution of river morphology.The construction of predictive models of river morpho-dynamics requires high quality time-resolved quantitative data on bedload flux. In-stream monitoring has been developed to obtain these data, relying on devices such as basket samplers (i.e., portable traps, fixed basket), geophones, hydrophones, and underwater microphones (Ergenzinger & De Jong, 2003). However, these methods remain challenging. Basket samplers require manual maintenance and resolve only parts of an event (Vericat et al., 2006). In addition, they can also introduce a bias because they alter stream flow and transport patterns around them, thereby affecting local transport rates. Acoustic measurements of bedload can be achieved with geophones and hydrophones (Geay et al., 2017(Geay et al., , 2020Habersack et al., 2017;Rickenmann, 2017) calibrated by direct measurements. In addition, geophones require a stable cross section in the streambed. As a result, they are mostly used in small mountain streams. Because in-stream monitoring requires specific channel conditions (basket samplers, geophones, hydrophones), has low temporal resolution (basket samplers), or cannot be deployed during flood conditions (portable traps, e.g., Helley-Smith samplers or small boats carrying acoustic doppler

No abstract