Premise:  The resistance of macroalgae to hydrodynamic forces imposed by ambient water motion depends in part on the mechanical properties of their tissues.  In wave-swept habitats, tissues are stretched (strained) at different rates as hydrodynamic forces constantly change.  Kelp have tissues of different ages, and mechanical properties of kelp tissue change with age.  However, the effects of age on the strain-rate dependence of the mechanical behavior of kelp tissues is unknown. 

Methods:  Using the kelp Egregia menziesii, we measured how high strain rate (simulating wave impingement) vs. lower strain rate (simulating wave surge) affected mechanical properties of frond tissues of various ages.  

Results:  Stiffness of tissues of all ages increased with strain rate, whereas extensibility was unaffected.  Strength and toughness of old tissue were unaffected by strain rate but increased with strain rate for young tissue.  

Conclusions:  Young tissue is weaker than old tissue, and therefore the most susceptible to breakage from hydrodynamic forces.  The increased strength of young tissue at high strain rates can help the whole frond resist breaking when pulled rapidly during wave impingement.  Because breakage of young tissue can remove a frond’s meristem and negatively impact the survival of the whole kelp, strain-rate dependence of young tissue’s strength can enhance kelp’s survival.

Field site and collection – Fronds of Egregia menziesii were collected from two sites in northern California: 20 fronds were collected from Miwok Beach (38° 21’25” N, 123° 4’2” W) near Bodega, CA and 12 fronds were collected from McClures Beach (38° 11’3”N, 122° 58’2”W) in the Point Reyes National Seashore.  In a concurrent study, material properties of E. menziesii were compared between sites and found to be similar (Burnett and Koehl, 2019).  At each site, fronds were collected by walking along a transect parallel to the shoreline and haphazardly removing a frond from every third E. menziesii encountered.  Fronds were only collected if they were ≥ 40 cm long, unwounded (i.e., no damage from herbivores), and had their intercalary meristem (Burnett and Koehl, 2019).  Fronds were stored in a cooler between 11 and 15 ℃ with just the water that was trapped on the fronds and transported to UC Berkeley.  Material properties were measured within 12 hours.

Measuring material properties – We measured the material properties of rachis tissue by conducting tensile stress-strain tests (details in Koehl and Wainwright, 1985).  An Instron materials-testing machine (Model 5544, Norwood, MA, USA) was used to pull on a section of rachis at a defined strain rate until it broke, while simultaneously measuring the specimen extension and the force with which it resisted the deformation.  We define "strain rate" as the change in length per time of a sample divided by the sample’s original length.  We use engineering stress-strain definitions, which are common in studies of macroalgal mechanics, where “engineering stress”  is defined as the force acting on a sample divided by the sample’s original cross-sectional area, and “engineering strain” is the total change in the sample’s length divided by the sample’s original length.  In studies of macroalgal mechanics, engineering strain is often reported as extension ratio, which is the total length of the sample divided by the sample’s original length (i.e., engineering strain + 1), and we follow this convention here.  We also consider the true stress-strain definitions, which are appropriate for materials that are highly deformable, including macroalgal tissue:  “true stress” is the instantaneous force acting on a sample divided by the sample’s instantaneous cross-sectional area, and “true strain” is the instantaneous extension of the sample relative to the previous instantaneous length, which is calculated as the ln(1 + engineering strain) (Vincent 2012).  We calculate the instantaneous cross-sectional area of the sample by dividing the sample’s volume, which remains constant) by its instantaneous length.  A variety of material properties can be calculated from the resulting stress-strain curve, whether represented as engineering stress-strain or true stress-strain (Fig. 2).   

            We measured material properties of E. menziesii rachis tissue of different ages.  For each frond, we cut sections (10 cm in length) of the intact rachis at specific distances from the intercalary meristem, where distance from meristem indicates tissue age (Burnett and Koehl, 2019): 0 to 10 cm, 20 to 30 cm, 40 to 50 cm, and 60 to 70 cm.  For simplicity, we will hereafter refer to these rachis regions by the mean distance of each sample from the intercalary meristem (i.e., 5, 25, 45, and 65, respectively).  Some fronds were not long enough to provide tissue at each region.  The ends of each sample were wrapped with pieces of paper towel that were affixed to the specimen by cyanoacrylate glue to prevent slipping in the grips of the Instron.  Each sample was then secured in the grips of the Instron and strain was applied at a fixed rate (see below) until the sample failed.  The force with which the sample resisted the strain was measured at 10 Hz to the nearest 0.1 N, and the extension of the sample was based on the distance between the Instron’s clamps.  The cross-section of the sample next to the break location was then photographed, and its area (A) was measured using ImageJ software v1.52g (US National Institutes of Health, Bethesda, MD, USA).  

            We used the measurements of force (F) and change in sample length ΔL, along with the sample’s cross-sectional area (A) and original length (L_o), to construct an engineering stress-strain curve and true stress-strain curve for each sample (Fig. 2).  We then calculated four pairs of material properties for each sample, with one set based on engineering stress-strain curves and the other set based on the true stress-strain curves:  (1) Tensile strength, σ_max is the stress (F/A) at which the sample broke; (2) extensibility, λ_max, is the strain at which the sample broke; (3) elastic modulus, E, is the slope of the initial linear portion of the stress-strain curve between strains 0.0 and 0.1; (4) yield stress, σ_yield, is the stress at which the sample exhibited plastic deformation after it was extended beyond the range of its elastic behavior.  Yield stress was identified by fitting a cubic spline regression model to the stress-strain curve and finding the stress at which the rate of change in stress per change in strain was lowest (i.e., where the sample suddenly started deforming without a proportional increase in stress).  We also calculated (5) work per volume to fracture, W/V, which is the mechanical work to fracture the sample normalized by the sample’s volume (A L_o), as the total area under the stress-strain curve (i.e., it is independent of whether the curve is presented as engineering vs. true stress-strain).

Strain rates –  Material properties were measured at one of two strain rates that were chosen to represent strain rates that may be encountered by E. menziesii exposed to breaking waves:  a high strain rate that could occur during wave impingement and a lower strain rate that could occur during wave surge.  Each frond was randomly assigned a strain rate and all samples from that frond were measured at the same strain rate.  In situ measurements of hydrodynamic forces on whole E. menziesii showed that loads during the surge portion of waves can increase at rates of approximately 10 N/s (Gaylord et al., 2008).  From preliminary stress-strain measurements of E. menziesii rachis tissue, we found that a loading rate of 10 N/s corresponds to a strain rate of approximately 3.33 x 10^-3/s.  In situ measurements of water accelerations during complete wave cycles showed that water accelerates on the order of 100 m/s² during wave impingement and on the order of 1 m/s² during wave surge (Gaylord 1999).  We estimate that the water accelerations during impingement that are 100-fold higher than during wave surge could produce loading and strain rates in E. menziesii that are 100-times greater than the rates experienced during wave surge.  Therefore, to simulate strain that could occur during wave impingement, we used a strain rate of 3.33 x 10^-1/s during stress-strain tests on E. menziesii rachis tissue.  It is important to note that these two strain rates, 3.33 x 10^-3/s and 3.33 x10^-1/s, are not the only strain rates that may occur during wave surge and wave impingement, respectively, because waves are highly variable across space and time (O’Donnell and Denny, 2008), and many factors influence the specific strain rates that occur in different regions of the kelp’s tissues (e.g., the kelp’s position in the water, the length of the kelp) (Koehl, 1986, 1999).  Instead, these two strain rates represent a range of strain rates that could occur as a wave passes over the kelp, but which has not been examined for its effects on the kelp’s material properties.

This dataset contains the following data:

• Date collected: (m/dd/yyyy)
• Site: Miwok Beach, McClures Beach
• Specimen label: Fronds 1 to 32
• Distance from meristem: distance of the measured rachis tissue from the frond's intercalary meristem (units = cm)
• Strain rate: units = s^-1
• Initial sample length: Starting distance between the Instron clamps (units = mm)
• Cross-sectional area of sample at breakage location (units = m²)
• Maximum extension ratio of sample (units = dimensionless)
• Maximum engineering stress of sample (units = MPa)
• Engineering modulus (units = MPa)
• Engineering yield stress (units = MPa)
• Maximum true strain (units = dimensionless)
• Maximum true stress (units = MPa)
• True modulus (units = MPa)
• True yield stress (units = MPa)