Dataset: Sediment resuspension summary

We resuspended the surface 5 cm of natural muddy sediment cores in the lab and compared temporal changes in sediment compaction to changes in surface and subsurface cohesion over 30 days post resuspension. Sediment-water interface (SWI) height and acoustic sound speed through sediment, which depends on bulk density, provided continuous and nondestructive metrics of compaction, and sediment porosity and grain size were measured destructively to characterize sediment physical structure. We determined surface cohesion by measuring both eroded mass and turbidity resulting from increasing shear stress. Subsurface cohesion was determined from the force required for sediments to fail in tension. We compared surface and subsurface exopolymeric substance (EPS) concentrations to surface and subsurface cohesion measurements. We differentiated between water-soluble (colloidal) and sediment-bound EPS as we expected bound EPS to contribute more to sediment-organic matrix development and thus cohesion because they are directly bound to sediment grains rather than dissolved in porewater.

These data include the summary of data collected on cores processed over time points 0 days (no resuspension), then 1, 2, 3, 7, 14, and 30 days post resuspension. There were 5 replicate cores per time point, but not all data were collected on each replicate core. Detailed data on erosion measurements, as well as repeated non-destructive measurements of sediment-water interface height and sound speed on cores processed on day 30 are provided in separate datasets.

Cores were extruded and sliced at 1 cm depth intervals. We calculated water content from sediment mass differences before and after drying at 65° C for 24 h as mass of water divided by mass of dry sediment (Eq. 4.7 from Jackson and Richardson 2007). We measured grain size every cm in the top 5 cm and at 8 and 10 cm for undisturbed cores and cores 3 and 30 days after resuspension. We determined grain size distribution with a Malvern Mastersizer 3000 particle analyzer (Malvern Panalytical, Malvern, UK) and data were analyzed with Gradistat (Kenneth Pye Associates, Ltd., Berkshire, UK) and classified according to Folk and Ward (1957).

We performed acoustic measurements following methods from Dorgan et al. (2020). Within a seawater tank, a 400 kHz three-cycle sinusoidal tone burst was transmitted horizontally through sediment cores to a receiver at 3 depths below the sediment surface (2.5, 5, 10 cm) (see Fig. 1 in Dorgan et al., 2020). To account for sound speed differences due to temporal variability in temperature and salinity, sound speed through sediment was normalized by the sound speed in seawater to obtain sound speed ratio (SSR). Each day, we also performed acoustic measurements on cores filled with seawater and with no core present. Sound speed in seawater and the lag time between the transmitted and received signals (time of flight) through sediment and seawater cores were used to calculate sound speed in sediment (νp):

    ν_p=c_w/(1-(c_w * ∆t/d_s )

where cw is sound speed in water, Δt is the difference in time of flight between seawater core (tw) and sediment core (ts), and ds is the inner diameter of the core (Jackson and Richardson, 2007; Dorgan et al., 2020). SSR was then calculated by dividing νp by cw, where a higher SSR indicates more compact sediment.

To determine if differences in erodibility and tensile strength were driven by variability in surface and subsurface EPS, we analyzed the subcore used for water content and grain size measurements for EPS carbohydrate concentrations. Following methods of de Brouwer and Stal (2001), we lyophilized frozen sediment and extracted colloidal carbohydrates with purified water (E-Pure) for 1 h at 30 °C. We then extracted bound carbohydrates with 0.1 M Na2EDTA for 16 h at room temperature. We measured both carbohydrate fractions with the sulfuric acid-UV assay (Albalasmeh et al., 2013), which is based on the phenol-sulfuric acid assay (DuBois et al., 1956). 900 µL 96 % sulfuric acid was added to 300 µL carbohydrate solution to dehydrate dissolved carbohydrates into furfural derivatives, which absorb UV light. This solution was vortexed for 30 s, allowed to return to room temperature for approximately 5 min, then UV absorbance at 315 nm was measured using a SpectraMax M5 microplate reader (Molecular Devices). We determined carbohydrate concentration from UV absorbance of a glucose reference.

To determine subsurface cohesion changes over time, we measured tensile force (N) using a custom probe modified from a fracture toughness probe developed by Johnson et al. (2012). A helical probe is rotated and translated into the sediment like a corkscrew, then pulled upward, breaking off a plug of sediment. Force, measured with an in-line force sensor (Futek LS-200 2-lb), increases to a peak force, then drops when the plug breaks free of the sediment below. Forces from friction with the surrounding sediment and the weight of the sediment plug are removed by repeating the corkscrew motion and subtracting the force profile from the second upward pull. The peak force in the plot of net force as a function of upward distance corresponds to the tensile strength of the sediment, a metric of cohesion. Fracture toughness can be calculated from this peak force (Johnson et al. 2012), but due to some concerns about the effect of sediment depth on these calculations (Dorgan, unpublished data), only force is presented here. These force measurements are comparable across the same depth in different cores, with higher force indicating greater cohesion.

Most instruments are custom built. Acoustics measurements were done following Dorgan et al. 2020, JASA. Other measurements are described in Clemo et al., submitted, Limnology and Oceanography.