Physical and biological conditions on a steep intertidal gradient at Rottnest Island, Western Australia
Introduction
Vertical zonation of plants and animals in the intertidal zone is a well-known feature of seashores.
Early hypotheses emphasized the tide as a primary factor causing this zonation.
Connell's review (1972) suggests that physical factors set upper limits of species distributions, while biological interactions (competition, predation) affect lower limits.
Conditions for growth and survival are generally better lower on the shore:
Longer submersion times.
Greater algal production.
Less harsh physical conditions.
Aims of the study:
Quantitatively describe the zonation of the community on vertical coastal limestone shores.
Measure differences in physical and biological conditions on the intertidal gradient.
Measure algal production and utilization rates.
Study Area and Intertidal Animals
Rottnest Island (32° S, 155°30'E) has coastal limestone platforms at about low tide level, ending in 2 m high vertical rockfaces.
Previous studies described habitats and animals living on the vertical shore.
Study carried out in March and April 1977 at Strickland Bay and North Point.
Common animals, from highest to lowest:
Littorine snails: Littorina unifasciata, Nodilittorina rugosa
Pulmonate limpet: Siphonaria kurracheensis
Acmaeid limpet: Notoacmea onychitis
Nerite: Nerita atramentosa
Chiton: Clavarizona hirtosa
Acmaeid limpet: Patelloida alticostata
Species maintain zonation patterns with variations in species presence and densities.
Limpets and chitons return to home scars after grazing.
Nerites and littorines shift positions with tidal conditions.
Methods
Zonation of Animals
Ten areas were chosen based on similarity in rock texture, vertical profile, and wave exposure, each 50 cm wide.
Each area was divided into 10 cm high horizontal strips, numbered 1-18 from the platform upwards.
Numbers represent height in decimeters (dm) above the platform (at approximately the 0.40 m tide level).
Each individual of each species was measured in each level.
Length/weight regressions were used to calculate the dry flesh weight of each species per level.
Tidal Regime
The proportion of time levels were damp was estimated using electrode terminals at levels 2, 4, 6, 8, 10, and 12.
Electrodes were connected to a multi-channel chart recorder, monitoring each electrode for 3 minutes of each 30-minute interval.
Percentage of time each level was immersed/damp was calculated from 24h recordings at each of five areas, with two replicates per level.
Clod cards (Doty 1971) were used as a second measure of seawater presence.
Hemispherical plaster clods were filed to 7.5 g, glued to plastic cards, and left for 48 h.
Percentage weight loss measured the effect of seawater at different shore levels.
Standing Crop and Productivity of Algae
Standing crop of algae was measured as µg chlorophyll/cm\textsuperscript{2}.
Algal samples were obtained by scraping the algal-rock surface at the same levels as electrodes, down to pale rock (Underwood & Creese 1976; Stimson 1973).
Samples were funneled into test tubes, wrapped in foil, and stored in a cool, dark place until analyzed.
3 ml of 80% aqueous acetone (Kirk 1968) and 0.1 g calcium carbonate were added to reduce acidity (Underwood & Creese 1976).
Samples were placed in the dark for 3 h, then ground with mortar and pestle using acetone to a fine powder.
The suspension was made up to 10 ml with 80% acetone to rinse the test tube and mortar and pestle.
The suspension was centrifuged at 12,000 rev/min for 1 min and absorbances of the supernatants were read on a Bausch and Lomb spectrophotometer at 645 nm and 660 nm, corresponding to the peak absorbances of chlorophylls a and b respectively (Kirk 1968).
Kirk's (1968) nomogram converted spectrophotometer readings to µg of chlorophyll/sample.
Analyses on 24 samples from one site were measured at 664, 647 and 630 nm.
Kirk's (1968) nomogram values for the sum of chlorophylls a and b, and Jeffrey & Humphrey's (1975) equations for chlorophyll a, b and c summed were highly correlated (r = 0.966).
Although chlorophyll c made up from 6 to 41% of the total chlorophyll a, b and c it averaged only 18.5%±2.22 (s.e.), and the total of chlorophylls a and b was not significantly different from the totals of chlorophylls a, b and c (r = 0.290, 22 d.f., p » 0.10).
Values for chlorophyll a and b were accepted as adequate estimates of the amount of chlorophyll on the shore.
The area of rockface scraped for the sample was calculated by cutting aluminium foil to the exact shape of the area. The foil shapes were weighed and divided by the weight of a known area of foil. The average sample area was 2.99 ± 0.35 (S.E.) cm\textsuperscript{2} (n = 124). Chlorophyll data are reported as µg/cm\textsuperscript{2}.
The relationship between chlorophyll content and algal dry weight was calculated from samples of rockface algae of different weights: µg chlorophyll = 13.42 + 3.22mg dry weight algae (n = 38; r^2 = 0.905; P < 0.001).
Normally the grazers harvest the algae growing there so that the vertical shores at Rottnest Island have almost no macroalgae and only small amounts of filamentous, encrusting, and microscopic algae occurring on and in the rock surface.
Following removal of grazers these small algae develop into a slick which we assume is an accumulation of algae normally eaten by the grazers as it is produced.
Algal productivity was estimated at various levels on the shore as the difference in algal standing crop before and after a 14-day period, in an area from which all animals had been cleared, and in another area as the difference in measurements of algal standing crop made on an area cleared of animals 28 days before and an adjacent area from which the animals were not removed. Both sets of data gave similar production rates per 14 days and were lumped in subsequent analyses.
Utilization of Food
Direct measurement of food consumption rates in the field was not attempted.
Food utilization was estimated indirectly by measuring faecal output, assuming egestion rate is proportional to ingestion rate.
Calow (1977) quotes maximum gross conversion efficiencies ranging from 32 to 63% for four molluscs early in their development.
The egestion estimate of ingestion is an underestimate affected by differences in gross conversion efficiencies between species and by their relative abundances at different shore levels.
Collections of each species of mollusc were separated into three to eight adjacent size classes with up to 100 individuals per class for the smaller size classes.
Faeces were collected over a 12 h period at room temperature from animals held in 10 cm deep seawater which was aerated only for hmpets.
Faeces were collected from the containers by decanting the excess water and filtering the remaining water and faeces through filter paper in a Biichner funnel.
The faeces were placed in bottles, refrigerated and later placed in pre-weighed crucibles, dried at 80°C for 12h, weighed and ashed at 500°C for 12 h and weighed again.
Total ash-free dry weight of faeces (Y) was compared with the total dry body weights of animals at the midpoint of the size classes (X).
All ln-ln regressions, except for S. kurracheensis had slopes indistinguishable from unity. Because faecal production is therefore allometrically related to body weight, the factor e^a, where a is the intercept of the hi-ln regression, coverted biomass on the censused areas to an estimate of egestion rate by multiplication. Egestion rate for the censused areas for S. kurracheensis was summed for all individuals from calculations from the regression equation.
Growth Rates
Notoacmea onychitis were individually marked and measured (without removal) at Strickland Bay in January 1977, and remeasured in June (159-177 days later).
The level on the beach was recorded for each individual in June.
Animals remain in the same location in home scars for several months.
Growth rates were calculated as change in length in mm/100 days (Y) and compared with initial lengths (X) for each level on the shore.
Results
Zonation
Significant overlap in the vertical distribution of the seven species of grazing molluscs on Rottnest Island's vertical shores.
Overlap within a census area was less than the overlap of averages.
Each species occupied a separate subset of shore levels.
The limpets Notoacmea and Siphonaria made up 75% of individuals but only 16% of biomass.
Clavaizona and Patelloida were 5% of individuals but 64% of biomass.
Average distributions showed no abrupt boundaries, but a sequence of species distributions with wide dispersion around modal and average heights.
Physical and Biological Correlates with Shore Level
Accurate height measurements are easily made on these shores due to horizontal platforms at the same tidal level and nearly vertical rockfaces.
Height-specific physical and biological characters were quantitatively estimated (Table 1).
Most measures were greatest low on the shore and least high on the shore, except for the standing crop of animals.
The relationship between percent time submersed or damp (Y) and height on the shore (X in dm) was best explained by the linear equation:
Y = 117.0 - 7.7X (n = 60; r^2 = 0.58; P < 0.001).
The fit was not good but the test for departures from linearity was not significant (F = 1.97; 4.54 d.f.).
The % weight loss by plaster clods (Y) decreased with increasing height on the shore (X in dm) according to a semi-logarithmic relationship:
Y = 134.5 - 27.5 \ln X (n = 76; r^2 = 0.73; P < 0.001).
Standing crop of chlorophyll (Y in µg/cm\textsuperscript{2}) showed a semi-logarithmic relationship with height (X in dm):
Y = 44.4 - 8.05 \ln X (n = 42; r^2 = 0.68; P < 0.001).
Multiple regression between growth rate (Y in mm/100 days), initial size (X1 in mm) and height on the shore (X2 in dm):
y = 3.95 - 0.176X1 - 0.131 X2 (n = 140; P1 = 0, P < 0.001; P2 = 0, P < 0.001; multiple r^2 = 0.02529).
Standard partial regression coefficients were -0.452 and -0.280 for initial size and height, respectively.
Initial length is more than 1.5 times as useful in predicting growth rate than height.
Biomass of herbivores was greatest at levels four, five, and six due to the presence of chitons.
Biomass decreased sharply toward lower shore levels, matching the large algal standing crop at the lowest level.
Biomass decreased more gradually upshore from level six.
All average measures of physical and biological characters showed significant Spearman rank correlations with height on the shore except the algal standing crop.
Four pairwise rank correlations were significantly positive: percent time submersed with algal standing crop, and growth rate with each of percent time immersed, percent weight loss of clods and animal biomass.
Algal Productivity and Utilization Rates
Algal production at the two sites was not statistically different.
Production decreased with increasing shore level (Table 2).
The combined data for both sites gave the relation between algal productivity (Y) and shore height (X) as:
y = 8.45 - 2.83 \ln X (n = 15; r^2 = 0.68; P < 0.01).
The total egestion rates of all herbivores at each shore level (Y) was related to the algal productivity (X) at the corresponding shore level according to:
\ln y = -0.025 + 0.733 Z (n = 8; r^2 = 0.54; P < 0.05).
Greater algal productivity correlated with greater egestion rate.
Discussion
Zonation
Zonation on Australian shores has been extensively described but lacked quantified estimates of animal abundance with relative tidal height, limiting comparisons.
This study provides replicated quantitative data on animal abundances on the vertical portion of coastal limestone shores.
The zonation pattern occurs over approximately 2 m due to the vertical profile and small tidal amplitude.
Physical conditions change rapidly over short distances.
A great range of conditions is within the normal range of mobile snails, less so for sedentary limpets and chitons.
Figure 1 shows that six of seven species could occupy levels four to eight, a distance of 50 cm.
Individual animals can regularly interact with individuals of several other species.
Physical and Biological Correlates with Shore Level
Relative tidal height is easily and accurately measurable and is an accurate index of other physical and biological characters.
Submersion time, water movement, algal production/standing crop, and growth rates show quantitatively predictable conditions (linear or semi-logarithmic).
Data are specific to locations, times of year, and variable.
Regression equations are convenient summaries of descriptive data.
The linear relation between time of submergence and height matches previously published relationships.
Growth, Algal Productivity, and Utilization Rates
Notoacmea onychitis grow fastest low on the shore where algal food is produced fastest.
High on the shore, there is either not enough food or time to graze.
Utilization of algal production (measured by egestion rate) was positively correlated to algal production.
Grazers use food as fast as it is produced except at low shore levels.
Food is limited relative to demand.
Experimental reductions in grazer density led to algal blooms.
Yearly growth rates of juvenile Patelloida alticostata were inversely related to adult limpet density.
Low shore levels are occupied by the two largest, patchily distributed grazers (Patelloida alticostata and Clavarizona hirtosa).