Abstract:
Intertidal algae are exposed to potentially severe environmental forces on a daily basis.
One of the most important of these is the pressure drag generated by crashing waves, the
magnitude of which is determined by the interaction of water velocity and an alga’s projected
area. This study addresses the effects of drag on the morphological and mechanical properties of
the common intertidal red alga Mastocarpus papillatus. Breaking force of the stipe is positively
correlated with drag, indicating morphological and/or mechanical differences between exposed
and protected individuals of this species. Stipe cross-sectional area and material strength are
both positively correlated with drag as well as with breaking force, suggesting that an individual
blade may be adjusting both of these parameters in response to its mechanical environment. A
single blade does not, however, appear to increase both of these factors simultaneously, as stipe
cross-sectional area and strength are negatively correlated. Thus, I find that M. papillatus
responds to increased drag by either increasing its stipe diameter or its tissue strength, but not
both. The range of variation in these two properties remains small when compared to the overall
natural variability within the population.
Introduction:
The effects of crashing waves almost certainly represent one of the most important
environmental variables affecting life on rocky shores. Of all the organisms that inhabit the
intertidal zone of rocky headlands, sessile macroalgae face particularly great challenges in
dealing with the effects of wave force. With no ability to "run and hide" during large wave
events and a generally low material strength relative to other biological materials (Denny et. al
1989), algal distribution and abundance often appear to be constrained by conditions of wave
force. The question of whether algal populations respond to this variation in wave intensity with
morphological or mechanical adjustments remains both open and intriguing.
Previous laboratory and field studies have demonstrated that some species of brown algae
(Ochrophyta, class Phaeophyceae) exhibit a considerable degree of mechanical plasticity in
response to differing exposure conditions (Charters et al. 1969, Armstrong 1987, McEacheron &
Thomas 1987, Kraemer & Chapman 1991b, Johnson & Koehl 1994, Milligan & DeWreede
2000). Within the red algae (Rhodophyta), differences in both biomass and planform area
between exposed and protected sites have been observed (Carrington 1990, Pratt & Johnson
2002). Variation in breaking force, cross-sectional area, and strength along a wave exposure
gradient, however, has never before been demonstrated for a species of intertidal red algae.
In this study, I assess the response of an intertidal red alga, Mastocarpus papillatus, to
varying conditions of wave exposure. In an attempt to uncover weak trends that may have not
been detectable in previous studies, I take a novel approach to the measure of "wave exposure.
Whereas prior work has focused on differences in algal properties between groups of individuals
-2-
from “exposed" and "protected" sites, I examine these same properties as they vary continuously
along a drag gradient.
Drag is the most important hydrodynamic force applied to intertidal organisms and as a
result, the primary means by which algae interact with their wave environment. The force of
drag pulls an intertidal alga in the direction of water flow and is dependent both upon water
velocity and the projected area of that alga into the flow. The previous studies cited above
largely examined groups of algae taken from “exposed" and "protected" sites for variation
between different wave exposure environments. For M. papillatus, however, this approach may
be confounded by the fact that the size or frontal area of this alga varies with wave exposure
(Carrington 1990). As such, a small alga from a wave-exposed site might experience similar
drag conditions to those of a large alga in a wave-protected site. Given that drag represents the
biologically important interaction between an organism and its wave environment, a test of how
the morphological and mechanical properties of an alga vary with drag likely represents a more
relevant test of an algal response to its wave environment.
Using the exposed / protected design described above, Emily Carrington (1990) found no
significant differences in stipe strength (breaking force divided by cross-sectional area at the
breaking point) between exposed and protected M. papillatus populations. In this study, I
examine variation in breaking force, cross-sectional area, and strength for M. papillatus along a
drag gradient in an attempt to detect relationships between these properties and the wave
environment (represented by drag) that might not be apparent from Carrington's approach.
I found that with increasing drag, M. papillatus thalli show increases in all three of these
properties - breaking force, stipe cross-sectional area (the area over which a blade is most likely
to break), and tissue strength. Stipe cross-sectional area and strength, however, are negatively
-3.
correlated, suggesting that an individual may have a higher cross-sectional area or a higher tissue
strength in response to wave forces, but not both.
Materials and Methods:
Species and location.
Mastocarpus papillatus is perhaps the most common intertidal red alga along the central
California coast (Abbot & Hollenberg 1976). M. papillatus grows in semi-erect clumps with
multiple thalli emerging from a single holdfast. In this study, a single thallus was taken as the
sampling unit. Thallus location within a clump was not considered as a factor in analysis.
Although clumping has been shown to reduce drag in the related alga Chondrus crispus,
Carrington (1990) found only mild drag reducing interactions between groups of up to six
closely packed M. papillatus thalli. In the field, large clumps of more than approximately a
dozen densely packed individuals were not encountered in sampling.
Samples were collected from six sites along the rocky intertidal of Hopkins Marine
Station (HMS), Pacific Grove, California (36’37’ N, 121°54’ W). Sites were separated by a
minimum of 10 m and were all located at approximately 1.5 ft above MLLW. All sites were
located within 1 m* of locations at which dynamometers had previously been used to measure
wave-induced forces (Bell & Denny 1994, Helmuth & Denny 2003). This allowed water
velocity estimates to be made for each collection site.
-4-
Data collection:
Thalli were collected during springtime low tides from April through June 2003.
Samples were fully hydrated before all experimental procedures to control for the possible
effects of desiccation. Samples were selected that had a lack of obvious holes, tears, rotten areas,
and other imperfections that could affect the breaking force, and each sample had a low to
moderate degree of blade branching, as extensive branching patterns made accurate
measurements of projected area difficult.
Blades were isolated from their neighbors and grasped with either a small alligator clip
lined with rubber tubing or a small string with a slipknot. The clamp or string was then attached
to either a 500 g or 1000 g recording spring scale (resolution of -0.05 N) and pulled with a
steady force intended to separate the blade from the substratum within approximately 1 second.
A small slider on the scale recorded the maximum force at which the thallus broke.
All analyzed samples broke just above the holdfast at the lower part of the stipe. The
diameter of the stipe at the breaking point was measured for each sample with a dissecting scope
and ocular micrometer (resolution of -25 um). Cross-sectional area at the breaking point was
estimated as the area of a circle having this diameter. Each sample blade was then flattened
under a sheet of clear acrylic and digitally photographed. The planform area of each sample was
calculated using ImageJ, an image analysis program distributed by the Research Services Branch
of the NIH (http://rsb.info.nih.gov/ij)
Any sample that met one or more of the following criteria was discarded: (1) the holdfast
of the alga failed before the stipe tissue, (2) a break occurred at some location along the blade
rather than along the stipe, (3) the stipe was irregular, deformed, rotten, or damaged, (4) the
break occurred at the point of connection for two stipes, making measurements of cross-sectional
-5.
area inaccurate, or (5) no planform area measurements were possible due to the condition of the
sample or the degree of branching or bushiness. A summary of the number of samples discarded
for each of these reasons appears in Table 1. Additionally, approximately three times during the
course of data collection blades were encountered that were too small and had too high of a
break force to be removed from the substratum with my attachment procedures. These blades
could not be sampled.
Calculations.
The material strength of each blade was calculated by dividing the breaking force for that
sample by the cross-sectional area over which the break occurred.
Drag for each sample was modeled with the standard relationship
D=½ pU’SC
where p is the density of seawater (nominally 1025 kg/ m2), U is water velocity in m/s, S is
planform or projected area in m’ and Ca, a function of velocity, represents the coefficient of drag
for M. papillatus. The Ca used for this study was taken from Carrington (1990), who calculated
the drag coefficient of M. papillatus as
Ca=0.156 U 0367
(2)
Carrington also noted that this standard model of drag accounted for 75.8% of variation in drag
for the 60 samples she examined. The addition of nine other morphological parameters,
-6-
including patterns of branching and papillae density, only explained an additional 9% of the
variation, and as such were not included in the model of drag for this study. This equation for Ca
takes into account the reconfiguration in flow (a decrease in projected area) of M. papillatus.
Relevant water velocities were calculated using the relationships between offshore
significant wave height and water velocity at each dynamometer site derived by Helmuth and
Denny (2003). A pressure transducer deployed offshore at HMS recorded significant wave
height data four times per day for a forty-day period just prior to the beginning of this study. The
average of the greatest one-third of these wave heights (129.5 cm) was used to calculate relevant
water velocities at each site. Relevant water velocity thus provides a measure of the average
stressful water velocities experienced by M. papillatus thalli at each site over the forty days prior
to the beginning of this study.
Systat 8.0 was used for plots and statistical analysis. For all linear regression
calculations, R* represents adjusted squared multiple R and probabilities are reported for the
outcome of a 2-tailed student’s t-test.
Results:
Of the 233 total samples collected, 121 were retained for data analysis. Departures from
normality were corrected in all data sets (cross-sectional area, breaking force, strength, and drag)
with log-transformations prior to linear regression calculations.
Linear regressions show significant positive relationships between drag and breaking
force (Fig. 1, R2 = 0.212, P «0.001), drag and cross-sectional area (Fig. 2, R“ = 0.048, P -
0.009), and drag and strength (Fig. 3, R° = 0.053, P = 0.007). Cross-sectional area and breaking
force (Fig. 4, R2 = 0.2531, P « 0.001) and strength and breaking force (Fig. 5, R° = 0.271, P«
-7-
0.001) are also positively correlated. Cross-sectional area and strength are inversely related, as
seen in Fig. 6 (R° = 0.209, P « 0.001).
Discussion:
Relationships along a drag gradient.
We can see in Fig. 1 that breaking force for M. papillatus increases with increasing drag,
implying that these algae are in some way responding to their wave environments through
modifications of their morphological or material properties. The question then becomes what
factors might explain this positive relationship. Figs. 2 and 4 together show that one possible
explanation for this effect lies in an increase in stipe cross-sectional area (from these
relationships, an increase in drag implies to an increase in cross-sectional area, and an increase in
cross-sectional area implies an increase in breaking force). Figs. 3 and 5 demonstrate a second
possible explanation. As drag increases, the material strength of the stipe tissue of M. papillatus
also increases, and as stipe strength increases breaking force increases as well.
The observed increases in stipe breaking force, cross-sectional area, and strength with
drag stand in contrast to the results of Pratt & Johnson (2002) and Carrington (2001). These
studies found no significant relationships between these three variables and wave exposure for
the related algae Mastocarpus stellatus and Chondrus crispus using the previously described
“exposed site" and "protected site" experimental design. Working with these same two species,
Johnson & Dudgeon (1992) report evidence that high wave exposure and freezing actually
weakens (lowers the strength of) the stipes of C. crispus and M. stellatus. Working with M.
-8-
papillatus, Carrington (1990) found no significant relationship between collection site and stipe
strength between one “exposed", one "protected", and one "intermediate exposure" site.
The increase in breaking force, cross-sectional area, and strength with increasing drag
shown here may be the result of either selective forces acting on the population as a whole (by
removing weak individuals) or plastic responses to increased drag forces within an individual
thallus. Further investigations will be needed to determine which of these two is actually
occurring for M. papillatus.
Strength and Cross-sectional area:
Figure 6 shows a relatively strong negative correlation between cross-sectional area and
strength. This result corroborates published work on the related red algae Chondrus crispus and
Mastocarpus stellatus (Dudgeon & Johnson 1992). It appears that for M. papillatus as well as
these other red algae, growth in stipe diameter outpaces the corresponding increase in stipe
strength. The biological explanation for this phenomenon remains unknown, although suggested
mechanisms in other algae include tissue differentiation in thick stipes (Koehl & Wainwright
1977), tissue weakening with age, endophytic infections, grazing, scarring, or general damage
due to repeated stress over time (Dudgeon & Johnson 1992).
The relationship shown in Figure 6 may also suggest that perhaps that the stipe tissue of
M. papillatus is not homogenous with regards to strength. The majority of the stipe’s strength
may be found, for example, in the outer ring of the stipe where the smaller, denser
photosynthetic cells of the plant are concentrated (Gabrielson et. al 2000). As stipe radius, r,
increases, the cross-sectional area increases as ar’ while the circumference of the stipe increases
-9-
as 2ir, and a concentration of strength in the circumference of the stipe could lead to the inverse
relationship seen in Fig. 6. As of the moment, however, this hypothesis is simply speculation.
The biological mechanisms by which tissue strength may vary in red algae are also
unknown, although many have been suggested. Previous studies suggest that the types and
amounts of structural materials, the proportion of these materials, and the orientation of fibers in
thalli may play a role in brown algal tissue strength variation (Wainwright et al. 1976, Koehl &
Wainwright 1977, Babb 1985). Modification of cell wall composition, including an
augmentation of structural compounds, in specific response to wave forces has also been
previously documented for brown algae (Kraemer and Chapman 1991b). Kraemer and Chapman
(1991a) also noted an increase in incorporation of radioactively tagged carbon into cell wall
tissue of Egregia under the application of a continuous force, which may play a role in strength
variation.
Given the very low R“ values calculated for the relationship between drag and cross-
sectional area (R° = 0.048) and that between drag and strength (R° = 0.053), interpretations of
these data must be somewhat guarded. Although both relationships are statistically significant
with P-values of 0.009 and 0.007 respectively, the natural variation within this M. papillatus
population is clearly very great compared to the magnitude of the adjustments of either cross¬
sectional area or strength with increasing drag. The lack of evidence for either of these
relationships in other red algae suggest that investigations specifically relating drag to
morphological and mechanical parameters, not just “exposed" and "protected" environments,
would be needed to detect these relatively weak signals in other species of red algae.
These high levels of natural variation should not, however, be taken to imply that the
increases in cross-sectional area and strength shown here are biologically irrelevant.
-10-
Examination of the untransformed data indicates that, on average, strength varies by 40-50% and
cross-sectional area by 25-35% over just the small range of drags sampled in this study. Thus,
despite the high degree of variation in these graphs and the lack of predictive power they imply,
differences in strength and cross-sectional area across a drag gradient could still be of biological
importance for individuals of this species.
Comparison with data from Carrington (1990):
Many of the results of this study corroborate the previous work of Emily Carrington
(1990), also at Hopkins Marine Station. A brief comparison between these data sets is outlined
in Table 2. The difference in percentage of holdfast failures might be explained by the different
times of year our studies were conducted (I sampled during springtime while Carrington
collected her data during the winter). The effects of seasonal changes on the properties of M.
papillatus have yet to be examined. Carrington and I also report significantly different mean
cross-sectional areas (student’s t-test with unequal variances, P = 0.005). Since 1 find that cross-
sectional area varies with wave environment, this difference may be due to a higher average drag
for Carrington’s samples. Carrington’s study did not include drag values, however, so this
hypothesis cannot be tested. No significant difference was detected between our values for mean
stipe strength (student's t-test with unequal variances, P2 0.10).
Carrington also found no effect of collection site on stipe strength, despite the fact that
she collected in three sites she considered protected, exposed, and intermediate. There are
several possible explanations for this apparent discrepancy between her findings and mine.
Possibly the wave force variation between these sites was not as great as Carrington had
expected (recall that she did not have access to accurate water velocity measurements at the time
-11-
of her study). Additionally, cross-sectional area clearly has an impact on measurements of
strength (Fig. 6), and Carrington’s failure to control for stipe diameter in her strength
comparisons may have confounded her results. Her experimental design, utilizing “exposed,
"intermediate," and "protected" sites, may also not have had the resolution necessary to detect
these relatively weak relationships. Additional variation between my results and Carrington's
could also be explained by the greater resolution of my data set: I measured breaking force to the
nearest 0.05 N and cross-sectional diameter to the nearest 25 um. Carrington measured these
two quantities to the nearest 1 N and 50 um, respectively.
Carrington also found no significant relationship between her untransformed data relating
cross-sectional area and thallus planform area (P2 0.05). In my log-transformed data set, I also
observe a non-significant relationship between these two quantities (R° = 0.023, P = 0.054).
Carrington used this conclusion to extrapolate to a model of the maximum planform area of a
blade at a given water velocity, assuming that this planform area is the only parameter that might
vary systematically with water velocity. While cross-sectional area may not vary with frontal
area, I find here that both stipe cross-sectional area and strength vary positively with drag (Figs.
2 and 3) as does breaking force (Fig. 1)
These results indicate that Carrington’s model for calculating maximum thallus size for
M. papillatus in a given wave environment may need to be reconsidered. Combining the linear
regression equation of breaking force and drag (Fig. 1) with Carrington's standard model for
drag, the maximum thallus area for a given water velocity is given by:
A=0.125 U-1633
(3)
-12-
where A is thallus area in m’ and U is water velocity in m/s. This relationship can be seen
graphically in Fig. 7. I predict the maximum, wave-limited thallus area of M. papillatus to be
approximately twice as high as Carrington (1990)
Conclusion:
Mastocarpus papillatus shows statistically significant increases in breaking force, cross¬
sectional area and strength with an increase in drag, although there is substantial variation about
this trend. This is the first time that relationships of this type have been reported for any alga of
the order Rhodophyta, which may be explained by my examination of these three relationships
along a continuous drag gradient instead of between sites of qualitatively different wave
exposure
Attention should be paid to the possible biological explanations behind these
relationships, especially to the relationship between strength and drag. Possible chemical or
physical variation within M. papillatus tissue itself, particularly in the cell wall composition,
might help explain this relationship and should be examined. The inverse relationship between
strength and cross-sectional area remains unexplained as well, and investigations should be
directed towards examining the question of whether the stipe tissue of M. papillatus might in fact
be heterogeneous with regards to strength. Future investigators might be advised to attempt to
control for age of the individuals.
Acknowledgements:
1 would like to thank the entire Denny lab for all of their help and advice in every stage of
my project design, implementation and analysis, especially Chris Harley, Patrick Martone, Luke
Hunt, Luke Miller, and my advisor Mark Denny. Special thanks also go to Jim Watanabe for his
help with my statistical analysis and all of my classmates for their interest and support
throughout the quarter.
Citations:
Abbott, I. A., Hollenberg, G. J., 1976. Marine algae of California. Stanford University Press,
Stanford, California, 827 pp.
Armstrong, S.L., 1987. Mechanical properties of the tissues of the brown alga Helophyllum
sessile (C. Ag.) Setchell: variability with habitat. J. Exp. Mar. Biol. Ecol. Vol. 114, pp. 143-
151.
Babb, I., 1985. The biomechanics of Maine coast kelps: their distribution, morphology, and
mechanical properties. M.Sc. thesis, University of maine, Orono, Maine, 126 pp.
Bell, E.C., Denny, M.W., 1994. Quantifying "wave exposure": a simple device for recording
maximum velocity and results of its use at several field sites. J. Exp. Mar. Biol. Ecol. Vol.
181, pp. 9-29.
Carrington, E., 1990. Drag and dislodgment of an intertidal macroalga: consequences of
morphological variation in Mastocarpus papillatus Kützing. J. Exp. Mar. Biol. Ecol., Vol.
139, pp. 185-200.
Carrington, E., Grace, S.P., Chopin, T., 2001. Life history phases and the biomechanical
properties of the red alga Chondrus crispus (Rhodophyta). J. Phycol., Vol. 37, no. 5, pp.
699-704.
Charters, A.C., Neushul, M., Barilotti, C., 1969. The functional morphology of Eisenia arborea.
Proc. Int. Seaweed Symp. Vol. 6, pp. 89-105.
Denny, M.W., Brown, V., Carrington, E., Kraemer, G., Miller, A., 1989. Fracture mechanics
and the survival of wave-swept macroalgae. J. Exp. Mar. Biol. Ecol., Vol. 127, pp. 211-228.
-15.
Dudgeon, S.R., Johnson, A.S., 1992. Thick vs. thin: thallus morphology and tissue mechanics
influence differential drag and dislodgement of two co-dominant seaweeds. J. Exp. Mar.
Biol. Ecol., Vol 165, pp. 23-43.
Gabrielson, P.W., Widdowson, T.B., Lindstrom, S.C., Hawkes, M.W., Scagel, R.F., 2000. Keys
to the Benthic Marine Algae and Seagrasses of British Columbia, Southeast Alaska,
Washington and Oregon. Phycological Contribution Number 5. Department of Botany,
University of British Columbia. iv + 189 pp.
Helmuth, B., Denny, M.W., 2003. Predicting wave exposure in the rocky intertidal zone: Do
bigger waves always lead to larger forces?. Limnology and Oceanography. Vol 48 (3), pp.
1338-1345.
Johnson, A.S., 2001. Drag, drafting, and mechanical interactions in canopies of the red alga
Chondrus crispus. Biol. Bull. Vol 201, pp. 126-135.
Johnson, A.S., Koehl, M.A.R., 1994. Maintenance of dynamic strain similarity and
environmental stress factor in different flow habitats: thallus allometry and material
properties of a giant kelp. J. Exp. Mar. Biol. Ecol. Vol. 195, pp. 381-410.
Koehl, M.A.R., Wainwright, S.A., 1977. Mechanical adaptations of a giant kelp. Limnol.
Oceanogr., Vol. 22, pp. 1067-1071.
Kraemer, G.P., Chapman, D.J., 1991a. Effects of tensile force and nutrient availability on
carbon uptake and cell wall synthesis in blades of juvenile Egregia menziesii (Turn.) Aresch.
(Phaeophyta). J. Exp. Mar. Biol. Ecol. Vol. 149, pp. 267-277.
Kraemer, G.P., Chapman, D.J., 1991b. Biomechanics and alginic acid composition during
hydrodynamic adaptation by Egregia menziesii (Phaeophyta) juveniles. J. Phycol. Vol. 27,
pp. 47-53.
-16-
McEacheron, J.C.T., Thomas, M.L.H., 1987. Attachment strength of Ascophyllum nodosum (L.)
0
LeJolis and exposure to wave action. Bot. Mar. Vol. 30, pp. 217-222.
Milligan, K.L.D., DeWreede, R.E., 2000. Variations in holdfast attachment mechanics with
development stage, substratum-type, season, and wave exposure for the intertidal kelp
species Hedophyllum sessile (C. Agardh) Setchell. J. Exp. Mar. Biol. Ecol. Vol. 254, 189.
209.
Wainwright, S.A., Biggs, W.D., Currey, J.D., Gosline, J.M., 1976. Mechanical design in
organisms. Edward Arnold, London, UK, 423 pp.
Table 1: Samples removed from analysis
Problem
Holdfast break
Blade break
Stipe Irregularities
Double Stipe
No area data (accurate area measurement impossible due to condition of
sample at time of measurement or extreme degree of size and branching
Total
Samples
112
Table 2: Comparison of data from Carrington (1990) and Kitzes (2003)
Carrington (1990)—
Kitzes (2003)
Calculations
% holdfast failures
36% (n=240)
12% (n = 8
Mean stipe strength
6.73 MN/m
6.27 MN/m
(n=73, SD -2.83)
(n= 121, SD =0.233)
Mean stipe cross¬
0.71 mm
0.60 mm
(n=125, SD=0.34)
(n= 121, SD =0.27)
sectional are
Figure Legend:
Fig. 1: Logarithmic plot of breaking force as a function of drag. Dashed lines represent 95%
confidence intervals. (n = 121, R° = 0.212, P £0.001)
Fig. 2: Logarithmic plot of cross-sectional area as a function of drag. Dashed lines represent
95% confidence intervals. (n = 121, R2 = 0.048, P = 0.009)
Fig. 3: Logarithmic plot of strength as a function of drag. Dashed lines represent 95%
confidence intervals. (n = 121, R° = 0.053, P = 0.007)
Fig. 4: Logarithmic plot of breaking force as a function of cross-sectional area. Dashed lines
represent 95% confidence intervals. (n = 121, R° = 0.2531, P £ 0.001)
Fig. 5: Logarithmic plot of breaking force as a function of strength. Dashed lines represent 95%
confidence intervals. (n = 121, R° = 0.271, P £0.001)
Fig. 6: Logarithmic plot of cross-sectional area as a function of strength. Dashed lines represent
95% confidence intervals. (n = 121, R° = 0.209, P « 0.001)
Fig. 7: Prediction of maximum thallus size as limited by water velocity: A = 0.125 U 1.053
0
Fig. 1
1.0
0.5
0.
og

oo

8
g
00
O0
-1.5 -1.
-0.5
0.0
Log Drag (N)
0.5
0
Fig. 2

6.0
OO
000 9 000


Se
o
8
ooc
900

o
O
oo
-1.5
-0.5
0.0
-1.0
Log Drag (N)
0.5
0
Fig. 3

O
1.0
80

o
8
o% 00
S

ge

g
O
0
000
8
0.0
-1.
-1.0
-0.5
0.0
Log Drag (N)
0.5
Fig. 4

1.0
0.5
0.0
-O.5
-7.0



o

O
g8
ga
o


-6.5
-6.0
Log Cross-sectional Area (m2)
-5.5
0
Fig. 5
1.0
0.5
0.0
-0.5
0.0
o008


06
gde o

O
O
0
L
0.5
1.0
Log Strength (MN / m2)
1.5
5
Fig. 6
-5.5
+p; -6.0
8 -6.5
-7.0
0.0

oo
8
.



a
oo
0.5
1.0
Log Strength (MN 7 m2
1.5
Fig.7
100
90
80
70
60
50
40
30
20
10
kaaaa-



9 11 13 15 17 19 21 23 25
Water Velocity (m/s)