Search for: All records

Climatic warming and permafrost thaw are predicted to increase Arctic riverbank erosion, threatening communities and accelerating sediment, carbon and nutrient cycling between rivers and floodplains. Existing theory assumes that pore‐ice thaw sets riverbank erosion rates, but overpredicts observed erosion rates by orders of magnitude. Here, we developed a simple model that predicts more modest rates due to a sediment‐entrainment limitation and riverbank armoring by slump blocks. Results show that during times of thaw‐limited erosion, the river rapidly erodes permafrost and undercuts its banks, consistent with previous work. However, overhanging banks generate slump blocks that must thaw and erode by sediment entrainment. Sediment entrainment can limit bank and slump block erosion rates, producing seasonally averaged rates more consistent with observations. Importantly, entrainment‐limited riverbank erosion does not depend on water temperature, indicating that decadal erosion rates may be less sensitive to warming than predicted previously.

We use in situ measurements of suspended mud to assess the flocculation state of the lowermost freshwater reaches of the Mississippi River. The goal of the study was to assess the flocculation state of the mud in the absence of seawater, the spatial distribution of floc sizes within the river, and to look for seasonal differences between summer and winter. We also examine whether measured floc sizes can explain observed vertical distributions of mud concentration through a Rouse profile analysis. Data were collected at the same locations during summer and winter at similar discharges and suspended sediment concentrations. Measurements showed that the mud in both seasons was flocculated and that the floc size could reasonably be represented by a cross‐sectional averaged value as sizes varied little over the flow depth or laterally across the river at a given station. Depth‐averaged floc sizes ranged from 75 to 200 microns and increased slightly moving downriver as turbulence levels dropped. On average, flocs were 40 microns larger during summer than in winter, likely due to enhanced microbial activity associated with warmer water. Floc size appeared to explain vertical variations in mud concentration profiles when the bed was predominately composed of sand. Average mud settling velocities for these cases ranged from 0.1 to 0.5 mm/s. However, Rouse‐estimated settling velocities ranged from 1 to 3 mm/s at two stations during winter where the bed was composed of homogeneous mud. These values exceeded the size‐based estimates of settling velocity.

An inexpensive and compact underwater digital camera imaging system was developed to collect in situ high resolution images of flocculated suspended sediment at depths of up to 60 meters. The camera has a field of view of 3.7 × 2.8 mm and can resolve particles down to 5 . Depending on the degree of flocculation, the system is capable of accurately sizing particles to concentrations up to 500 mg/L. The system is fast enough to allow for profiling whereby size distributions of suspended particles and flocs can be provided at multiple verticals within the water column over a relatively short amount of time (approximately 15 min for a profile of 15 m). Using output from image processing routines, methods are introduced to estimate the mass suspended sediment concentration (SSC) from the images and to separate identified particles into sand and mud floc populations. The combination of these two methods allows for the size and concentration estimates of each fraction independently. The camera and image analysis methods are used in both the laboratory and the Mississippi River for development and testing. Output from both settings are presented in this study.