M.G. Chapman

Observations in ecology : you can't make progress on processes without understanding the patterns

Coastal marine ecology is, quite properly, increasingly focussed on experimental tests of hypotheses about processes. These are, however, done to explain observations and patterns. It is therefore appropriate to be able to publish quantitative observations to provide the context and basis for studying mechanisms and processes. Ecologists are concerned about very different types of observations. Some areas of study are still totally dependent on observational, descriptive evidence; some depend on mensurative tests of hypotheses about patterns. Tests of hypotheses about patterns are also needed to validate casual or qualitative observations. Guide-lines for what constitutes appropriate or publishable ecological descriptions are discussed here. These recognize the experimental, hypothesis-testing nature of many descriptive studies and consider the relevance of sound logic and experimental design in the planning, collection and interpretation of observations.

Variability at different spatial scales between a subtidal assemblage exposed to the discharge of sewage and two control assemblages

Abstract Environmental disturbances can alter the variability of assemblages of organisms in impacted sites compared to control sites. It has therefore been proposed that increased variability might be an important feature of stressed populations. Increased variability may be due to changes in the population structure of individual species or changes in the suite of species. In this study, spatial variances of shallow subtidal assemblages of organisms inhabiting vertical cliff-faces were compared among two control locations and one location that had for many years been exposed to the discharge of sewage. These assemblages covered nearly all available space on the substratum and consisted primarily of encrusting and foliose macro-algae and numerous filterfeeding animals, such as ascidians, sponges and bryozoans. Mean differences in abundances between these locations were investigated using Beyond BACI designs. In addition, these locations were used to examine the model that assemblages are more variable in disturbed than undisturbed environments and to try to distinguish differences in variability due to differences in the population structure of individual species from that due to changes in species composition. The assemblages were sampled at two spatial scales at each of three depths in each location. There were significant differences between the polluted location and one or other of the control locations in the mean abundances of some organisms, the variances of certain species (or recognisable types) at each spatial scale and for multivariate measures of species composition. There was, however, no evidence to support the prediction of increased variability in the apparently polluted location compared to the control locations. Importantly, for many measures of abundance and variability, the control locations were as different from each other as they were from the polluted location, suggesting that the latter fell within the range of natural subtidal assemblages separated by these spatial scales. These findings emphasise the need to include more than one control location in any study of a potential environmental impact, so that any effects of that impact can be distinguished from the range of natural variability.

Power, precaution, Type II error and sampling design in assessment of environmental impacts

Abstract Increasingly, environmental managers attempt to incorporate precautionary principles into decision making. In any quantitative analysis of impacts, precaution is closely related to the power of the analysis to detect an impact. Designs of sampling to detect impacts are, however, complex because of natural spatial and temporal variability and the intrinsic nature of the statistical interactions which define impacts. Here, pulse and press responses and impacts that affect time courses (temporal variance) were modelled to determine the influences of increasing temporal replication—sampling more times in each of several longer periods before and again after an impact. Increasing the number of control or reference locations and number of replicate sample units at each time and place of sampling investigated the influence of spatial replication on power. From numerous scenarios of impacts, with or without natural spatial and temporal interactions (i.e. not caused by an impact), general recommendations are possible. Detecting press impacts requires maximal numbers of control locations. Shorter-term pulse impacts are best detected when the number of periods sampled is maximized. Impacts causing changes in temporal variance are most likely to be detected by sampling with the greatest possible number of periods or times within periods. To allow precautionary decision making, the type of predicted impact should be specified with its magnitude and duration. Only then can sampling be designed to be powerful, thereby allowing precautionary concepts to be invoked.

Assessment of some controls in experimental transplants of intertidal gastropods

Abstract Some factors associated with transplant experiments using a small high-shore gastropod, Littorina unifasciata Gray have been investigated. These included: (1) the effects of marking the test animals and the associated disturbance; (2) the pattern of dispersion of the test animals at the start of the experimental period; (3) the past history of the test animals, moving animals into a new patch of the same habitat or zone as opposed to a different habitat or zone; and (4) differing densities of conspecifics in the experimental areas. In addition, spatial and temporal variability in the results were examined. Disturbance, translocation into a new area and density of conspecifics had little effect on the subsequent movement of the test animals, and animals from different areas at the same level on the shore behaved similarly when moved between areas. The area in which the experiment was done, rather than the characteristics of the sample of snails used in the experiment, determined the pattern of movement. The pattern of dispersion of the test animals at the start had a significant effect on subsequent movement, because animals which were placed in clumps tended to move in the same direction. Of more importance, however, the data showed considerable spatial and temporal variability, in both the directions and distances displaced by the animals. It is extremely important to investigate the variability in the results obtained in such experiments, before detailed transplant experiments are done. In the case of L. unifasciata replicate experiments in a number of areas, on a number of occasions, and for differing periods are clearly needed before conclusions can be drawn about patterns of movement, and their importance in the establishment and maintenance of this snails patterns of distribution and abundance.

Experimental analyses of the influences of topography of the substratum on movements and density of an intertidal snail, Littorina unifasciata

Abstract Topographic complexity of the substratum influences the movements of the intertidal snail Littorina unifasciata. To determine the relationships between movements and local density and dispersion, an objective quantitative measure of topography was evaluated for use in the field. Heights of 100 vertical pins placed a few cm apart in a grid were measured to determine a topographic index (TI), defined as the circular variance of vectors normal to the triangular planes formed by every triangle produced by three adjacent pins in the grid. Snails moved further on less complex surfaces (those with smaller TI). Their movements were more directional on simpler surfacesand, in most such sites, movements tended to be towards the sea. On more complex surfaces, movements were generally random, or, if directional, in any direction. There was a negative correlation between the distance moved and the topographic index. Snails on simpler surfaces also tended to move in the same direction on successive days; those on surfaces with a larger TI moved in different directions on successive days. There was no relationship between TI and mean density, nor with an index of dispersion. Small-scale spatial variance was negatively correlated with TI but long-term temporal variance of numbers in 30 populations was positively correlated with TI. The lack of relationship between density and TI, coupled with greater movements on less complex surfaces suggested that turnover of individuals in local populations should be negatively correlated with TI (i.e., more rapid turnover where surfaces were simple). This hypothesis was corroborated in short-term experimental manipulations. Emigration and immigration were both greater where surfaces were simple (i.e., surfaces with small values of TI). Implications of this for the population dynamics of L. unifasciata are discussed.

Inconsistency and variation in the development of rocky intertidal algal assemblages

Abstract Experimentally-cleared patches were used to test hypotheses about the relative importance of broad-scale biogeographic processes and small-scale historical processes in the development of low-shore algal assemblages on wave-exposed rocky coasts of New South Wales (Australia). In addition, the applicability of generalizing from patterns of recruitment and development at one time was tested by providing similar cleared patches in the algal assemblage at three-monthly intervals and quantifying early development of the algal assemblage in these clearings fifteen times over a period of four years. The early stages of development of these assemblages differed significantly from shore to shore and time to time and there was no evidence for common broad-scale patterns of recruitment, nor for any biogeographic trend. Nevertheless, these assemblages developed from a limited common pool of species, some of which recruited fairly regularly on most shores and some of which only arrived sporadically on some shores. Despite different starting points, assemblages converged towards the surrounding assemblages on most shores, so that within less than 1–2 years, cleared areas resembled the surrounding assemblages. Although the specific changes leading to convergence differed from shore to shore, there was a general pattern of early colonizers, such as ephemeral algae and sessile animals being gradually replaced by larger perennial algae. These results demonstrate no simple seasonal nor clear-cut biogeographical patterns in the development of algal assemblages on these shores and indicated the relative importance of local influences. Results are discussed with respect to the need to do experiments at numerous places in order to examine responses of assemblages to, or recovery from, environmental disturbances.

Dispersal of the intertidal snail, nodilittorina pyramidalis, in response to the topographic complexity of the substratum

The littorinid gastropod, Nodilittorina pyramidalis (Quoy & Gaimard), is commonly found at the top of rocky shores in New South Wales, Australia. On sandstone shores around Sydney, the snails are usually restricted to patches of topographically complex rock during low tide and are mainly found within the pits and crevices in these patches. Differences in densities and sizestructures of populations on adjacent patches suggest little intermixing of these populations. Experiments showed that adult Nodilittorina had limited dispersal which tended to retain them within these topographically complex patches. This was a response to the topographic complexity of the habitat, but other factors were necessary to retain populations within these patches for more than a few days. Small snails were more responsive to the complexity of the habitat than large snails. In addition, Nodilittorina actively avoided topographically simple areas on these shores and rapidly returned to complex patches after experimental transplantation. These patterns of behaviour differed from those shown by the co-existing littorinid, Littorina unifasciata Gray, although both species are found in the same areas of the shore and feed in the same way on the same source of food.

Poor design of behavioural experiments gets poor results: examples from intertidal habitats.

Many patterns of distribution and abundance of intertidal animals are explained by processes of movements of animals, selecting particular habitats or levels on the shore, or interacting with other species. Movements of intertidal animals have therefore been studied over many years. During this long history, much intertidal ecology has changed in focus from broad-scale to small-scale patterns and processes, although there has been recent refocus on a combination of many scales. Simultaneously, there has been an increase in the incidence of field experiments and growing recognition that behaviour is more flexible than originally thought. This review examines changes in the ways that experiments on movements on intertidal animals have been and are being done, taking into account these changes in emphasis. Although some progress has been made, there is still a long way to go. The idea is still prevalent that behaviour is simple, rather invariant and that the animals respond to broad-scale cues that have traditionally been of interest to many investigators. This means that many experiments are still designed to minimise (or ignore) natural variation in behaviour rather than to measure it and that any associated disturbances are considered irrelevant and therefore not evaluated. Understanding the role that behaviour has in establishing and maintaining many of the patterns observed on intertidal shores is crucial to our understanding of the ecology of these habitats. Better experiments, designed logically with appropriate controls to evaluate realistic processes and to measure how behaviour varies among places and from time to time can only improve this understanding.

Influences of tidal conditions, temperature and desiccation on patterns of aggregation of the high-shore periwinkle, Littorina unifasciata, in New South Wales, Australia

The small high-shore littorinid, Littorina unifasciata, is the most abundant gastropod on many shores in New South Wales, Australia. It is usually found in dense clusters while emersed during low tide. Previous work has shown that the degree of aggregation varies spatially among shores, between different heights on the shore, between habitats with different aspect and among replicate sites in the same general area. Factors that might influence the degree of aggregation were not identified. In this study, aggregation at different stages of the tide and under different weather conditions was examined to test the models that aggregation and dispersal in L. unifasciata occur in response to the drying and wetting of the substratum by the tides or weather. Aggregation was found to be greater when the substratum was dry, irrespective of the state of the tide. It was proposed that environmental conditions would be harsher during neap than spring tides and when the time of low tide was nearer mid-day and, hence, snails would be more aggregated at these times. Aggregation was therefore examined with respect to the timing and duration of low tide. Although aggregation could be correlated with the time of low tide in some mid-shore sites, there was no such correlation higher on the shore. In addition, the time of low tide accounted for very little of the variability in aggregation from time to time. Finally, the influence of high temperatures and desiccation on the tendency to aggregate and the effects of aggregation on the temperatures and water reserves of the snails were examined experimentally in the field. Aggregated snails had larger water reserves and greater temperatures than did solitary animals, but neither the temperature of the snails, nor their water reserves, influenced their tendency to aggregate.

Experiments on effects of sampling biota under intertidal and shallow subtidal boulders

Abstract Diverse assemblages of sessile and mobile fauna live under intertidal and subtidal boulders. When these biota are sampled, it is necessary temporarily to overturn the boulders. Many mobile species respond to overturning by moving onto the other side of the boulder or moving off the boulder into the surrounding area. Overturning without replacing a boulder has been shown to affect abundances and diversity of its sessile and mobile organisms. Overturning when sampling has, however, not previously been examined as a potential source of disturbance, because replacement of the boulders has been assumed not to disturb the organisms. These experiments were designed to test the hypothesis that sampling is a measurable disturbance to the mobile fauna and groups of sessile biota living under boulders. The sessile organisms were mainly encrusting algae, encrusting tubeworms and bryozoans, with smaller amounts of foliose algae, other colonial animals, bivalves and cnidarians. The mobile fauna were primarily large (> 5 mm long) chitons, gastropods and echinoderms. Assemblages on boulders that had been previously sampled were compared to those on undisturbed boulders. Effects of sampling on temporal changes in abundances were also compared to similar changes on sets of undisturbed boulders. There were no or very few differences in assemblages, mean densities of selected species or the proportions of boulders occupied by selected species among undisturbed boulders and boulders subjected to different frequencies of prior sampling when there were 3–4 month intervals between samples. There were, however, short-term responses to disturbance associated with sampling. These were measured as differences in densities and the proportion of boulders occupied by some species of molluscs and echinoderms and differences in the molluscan assemblage between undisturbed and disturbed, i.e. previously sampled, boulders. These differences disappeared during the following month. These data suggest that organisms on boulders are disturbed by careful sampling, but that they rapidly recover during the following few weeks. Thus, sampling is possible without altering the abundances of the target species if there is long enough between samples for the densities under the disturbed boulders to recover.