Assessing the short-term stability of spatial patterns of macrobenthos in a dynamic estuarine system

Recent studies of macrofauna on sandflats have emphasized postsettlement dispersal, particularly associated with sediment bedload transport. This study investigated the short-term (3 weeks) stability in spatial patterns of eight common, potentially mobile, near-surface macrofaunal species inhabiting a 4,000-m2 area of intertidal sandflat. Over this period, wind conditions reworked the sediment to a depth of 3 cm, thus generating sediment and animal transport. Movement and changes in spatial arrangement of juvenile M’acomona Miana and Paphies australis demonstrated the need to consider postsettlement dispersal in survivorship and experimental studies and to conduct such studies at a relevant scale. However, the majority of the species exhibited stability of spatial pattern, even when undergoing movement and changes in mean density. This finding suggests that biological interactions and organismhabitat relationships are important processes even in habitats where the sediment and organisms are mobile. Estuarine and marine intertidal sandflats often form a dynamic environment for surfaceand near-surface-dwelling organisms and there is increasing evidence for high fluxes of infaunal organisms. Even small waves can generate extensive sediment bedload movement, and recent studies have shown benthic macrofaunal species move with the sediment (Emerson 1991; Emerson and Grant 199 1; Comrnito et al. 199%). Other studies have found a variety of benthic phyla entering and being transported in the water column (Armonies unpubl. rep. Martel and Chia 1991; Cummings et al. 1995). Yet an assumption often important to generalizing from experimental or behavioral research is that spatial pattern and variability are constant over time scales of days to months and spatial scales representing the natural ambit of the organisms involved. Ecologists working in dynamic softsediment habitats need to establish the temporal persistence of spatial patterns relative to potential physical forcing. If this is not done there is no way to be sure that the smallscale processes under study are not swamped by larger-scale phenomena (Raffaelli et al. 1994). The few studies that have discussed the stability of spatial patterns of macrofauna (McArdle and Blackwell 1989; Krager and Woodin 1993) have not studied potentially mobile, near-surface species or juvenile stages. It is also important to evaluate, explicitly, the form of spatial pattern rather than just the intensity. Intensity is assessed by methods that sort distributions of density estimates into aggregated, uniform, or random. Only the form of spatial pattern utilizes information contained in the spatial location of individuals. Therefore, two distributions may have a similar intensity of pattern (e.g. aggregated) yet represent different spatial arrangements (e.g. small patches versus gradients) (Thrush 1991). We investigated the stability of spatial arrangements (i.e. form of spatial pattern) in the abundance of common nearAcknowledgments We thank Pierre Legendre for suggesting we investigate the spatial structure of differences in density. We also thank John Commito, Paul Dayton, Pierre Legendre, and an anonymous reviewer for commenting on an earlier draft of the manuscript. surface macrofauna on an extensive sandflat in an estuary, where wave-generated bedload and suspended-sediment transport are common (Dolphin et al. 1995). A site of 4,000 m2 was surveyed on two dates, 3 weeks apart. We tested the following predictions. First, if hydrodynamic conditions were suffi.cient to generate transport of organisms, the spatial arrangemlznts of those species undergoing transport would not be persistent. Postsettlement movement could occur on two scales: movement could take place within the site, resulting in changes in the spatial arrangement of density but not of mean site density; and emigration from or immigration into the site could occur, resulting in changes in mean density and, possibly, spatial arrangement. Second, small juvenile sta.ges and epibenthic animals (e.g. juvenile bivalves, cumaceans) were more likely to be transported than larger or fast-burrowing animals (e.g. polychaetes, isopods). Third, lack of persistence in spatial arrangements would result in lack of persistence of interand intraspecific correlations. Materials and methods The study was conducted on an extensive (5 km2) intertidal sandflat near Wiroa Island (37’01 ‘S, 174’49’E) within Manukau Harbor, New Zealand. Manukau Harbor covers 368 km2 and has a tidal range of 3.4 m on spring tides and 2.1 m on neap tides. The intertidal sandflat near Wiroa Island is exposed to the prevailing southwesterly wind with a fetch of 4 km at midtide and 17 km at high tide. The study site was chosen for its overall homogeneity in physical appearance, tidal height (ca. midtide), wave exposure, and nearsurface-sediment composition (predominantly fine sands). Light to moderate (5-10 m s-l) west and southwest winds commonly blow over the sandflat, producing waves lo-30 cm high and reworking sediments to a depth of l-3 cm (Cummings et al. 1995; Dolphin et al. 1995). Sand ripples l-2 cm high and 2-3 cm in wavelength were common during this study. Maximum water velocities around 40 cm s-l (60 cm above the bottom) and 36 cm s-l (10 cm above the bottom) recorded over a neap tidal cycle at this site (Cummings et al. 1995; pers. obs.) exceed the Miller threshold for