SEARCH

SEARCH BY CITATION

Keywords:

  • Foraging;
  • habitat selection;
  • prey distributions;
  • shorebird consumers;
  • spatial scale

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

Food availability is a fundamental determinant of habitat selection in animals, including shorebirds foraging on benthic invertebrates. However, the combination of dynamic habitats, patchy distributions at multiple spatial scales, and highly variable densities over time can make prey less predictable on ocean-exposed sandy shores. This can, hypothetically, cause a mismatch between prey and consumer distributions in these high-energy environments. Here we test this prediction by examining the occurrence of actively foraging pied oystercatchers (Haematopus longirostris) in relation to physical habitat attributes and macrobenthic prey assemblages on a 34 km long, high-energy beach in Eastern Australia. We incorporate two spatial dimensions: (i) adjacent feeding and non-feeding patches separated by 200 m and (ii) landscape regions with and without foraging birds separated by 2–17 km. There was no support for prey-based or habitat-based habitat choice at the smaller dimension, with birds being essentially randomly distributed at the local scale. Conversely, at the broader landscape dimension, the distribution of oystercatchers was driven by the density of their prey, but not by attributes of the physical beach environment. This scale-dependence suggests that, on open-coast beaches, landscape effects modulate how mobile predators respond to variations in prey availability.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

Habitat selection is the primary mechanism for mobile animals to cope with changing conditions, such as spatial and temporal variability in resource availability (Railsback & Harvey 2002). Hence, resource-based habitat selection models, also known as resource-selection functions (RSFs), are often used to predict the distribution of animals, and to address one of the central challenges in conservation management – understanding how habitat change affects populations (Duchesne et al. 2010; McLoughlin et al. 2010). Spatial scale is a fundamental consideration in RSFs because it can influence the process of habitat selection (Boyce et al. 2003; Boyce 2006; McLoughlin et al. 2010). Animals often select foraging areas at small spatial scales, whereas home ranges are usually shaped at the landscape scale (Addicott et al. 1987; Klaassen et al. 2006; Ciarniello et al. 2007).

Birds have long been the focus of habitat-selection models (Cody 1981) and are an important biological link between habitats and ecosystems (Sekercioglu 2006). Shorebirds foraging on intertidal shores, in particular, are a widely used empirical system for field studies of RSFs and to test foraging theory (Goss-Custard et al. 1977; Ens & Goss-Custard 1984; Rutten et al. 2010). Although many factors, operating at a range of spatial scales, influence the choice of foraging sites by shorebirds (Kelsey & Hassall 1989; Finn et al. 2007, 2008), birds generally show a high fidelity to patches with consistent levels of food availability (Rutten et al. 2010).

At the landscape or regional scale (e.g. between estuaries and embayments), the densities of shorebirds are often correlated with the densities of their main prey (Colwell & Landrum 1993; Placyk & Harrington 2004; Spruzen et al. 2008). At smaller spatial scales, substrate penetrability (Grant 1984) and vegetation cover (Spruzen et al. 2008) can become important in determining habitat quality for birds. It has also been suggested that physical cues, such as sediment properties, are used by birds to select feeding sites because the density of harvestable prey within sediments may not be discernible from above (Kelsey & Hassall 1989; Finn et al. 2007).

The vast majority of studies on shorebirds focus on intertidal estuaries, bays and wetlands. These are heterogeneous landscapes where the association of birds with prey densities may be confounded by factors such as substrata properties, proximity to the water's edge, tidal emersion, vegetation cover, and other habitat features that change along the shoreline (Botton et al. 1994; Yasue 2006). Comparisons of RSFs across spatial scales can be further biased by spatial autocorrelation. This can arise along convoluted coastlines where distinct landscape elements occur regularly along the shoreline, interspersed by alternate habitat types. For example, feeding sites located within an embayment are likely to be much more similar than feeding sites between adjacent embayments (Colwell & Landrum 1993).

By contrast, exposed ocean beaches are essentially linear landscapes, and comparatively homogeneous in terms of substrata and vegetation cover (Schlacher et al. 2008c). This can make them, arguably, better study systems to gauge the role of food versus physical habitat attributes as drivers of bird occurrence. Like other intertidal shores, high-energy sandy shores are important habitats and foraging sites for shorebirds (Dugan et al. 2003, 2010; Hubbard & Dugan 2003; Weston & Elgar 2005; Peterson et al. 2006). Yet, what determines the distribution, movement, dispersal, diversity, and density of birds on open-coast beaches remains, surprisingly, poorly understood (Hubbard & Dugan 2003; Weston et al. 2009; Meager et al. 2012).

Pied oystercatchers, Haematopus longirostris, are a good candidate species to test models of habitat selection on high-energy sandy shores. They are a coastal bird species distributed around Australia, feeding on high-energy sandy shores where they prey mostly on bivalves, polychaetes and crustaceans (Marchant & Higgins 1993; Lauro & Nol 1995). The Eurasian oystercatcher, Haematopus ostralegus, is an important model species in studies of foraging behaviour, interference competition, and habitat selection (Goss-Custard et al. 1977; Ens & Goss-Custard 1984; Goss-Custard 1996; Rutten et al. 2010). Like their Eurasian congeners, Australian pied oystercatchers are also an interference-sensitive species occupying a large proportion of suitable habitat and are therefore a suitable model for habitat selection studies (Marchant & Higgins 1993; Folmer et al. 2010). Pied oystercatchers are also amenable to habitat selection studies because their foraging behaviour and location is easily determined from a distance.

Here we investigate habitat selection of pied oystercatchers in relation to prey density and physical habitat attributes at two spatial scales on a linear, high-energy, sandy shoreline. Specifically, we test two hypotheses: (i) there is a mismatch between the density of pied oystercatchers and the density of their prey on ocean shores; and (ii) spatial scale modulates habitat selection based on food resources, being less pronounced at small (e.g. adjacent patches 200 m apart) than at broader spatial scales (e.g. adjacent regions 2–17 km apart).

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

Study area

Field work was done on Main Beach of North Stradbroke Island (27.445°S, 153.537°E) in Southeast Queensland, Australia (Fig. 1) from 4 to 11 April 2010. Main Beach is a 34-km-long linear stretch of sandy beach that is fully exposed to the prevailing swell from the east to southeast (Schlacher & Thompson 2008). The study was done during the non-breeding season (Marchant & Higgins 1993) when birds did not show strong territorial behaviour; all observations and data pertain to adult birds.

image

Figure 1. Location of experimental sites in (a) Australia, (b) Moreton Bay, Southeast Queensland, (c) on Main Beach of North Stradbroke Island (black squares, pied oystercatcher feeding locations; hollow squares, non-feeding locations; triangles, southern non-feeding region). (d) Pied oystercatcher abundances from the five surveys of the study area between 28 March and 11 April (circles scaled by abundance from 1 for one oystercatcher to the largest symbols for six to nine individuals of oystercatchers). The rectangles in (d) show the two regions compared in Experiment 2.

Download figure to PowerPoint

We made extensive observations of oystercatcher foraging on sandy beaches prior to this study. The birds forage almost exclusively below the effluent line in the lower intertidal area and swash zone, preying on invertebrates buried in the sand (T. A. Schlacher, personal observations). Oystercatchers prey both on large clams and on smaller invertebrates. To capture the larger prey, the birds first visually locate the siphons of buried clams (Donax deltoides, shell length up to 60 mm) before dislodging the detected clams from the sand, prying the valves open, and consuming the soft parts. Oystercatchers also feed on smaller (<1 cm) prey items, indicated by frequently observed probing and swallowing of prey other than clams. This is consistent with the literature showing that oystercatchers have a catholic diet, consuming a very wide variety of prey items that include bivalves, gastropods, worms (polychaetes and oligochaetes), insects (adults and larvae), crabs, isopods and amphipods. All potential prey taxa sampled by us in this study, and used in the habitat selection models, have been observed to form part of the diet of oystercatchers (Marchant & Higgins 1993) and are therefore relevant to be included in the analysis. Thus, we treat all macrobenthic invertebrates as potential prey items for oystercatchers in this study. The macrobenthic communities of this beach are numerically dominated by crustaceans (several species of amphipods, isopods and crabs), polychaetes and molluscs (Schlacher & Thompson 2007; Schlacher et al. 2007, 2008a,b; Schlacher & Morrison 2008; Sheppard et al. 2009; Schlacher & Lucrezi 2010; Noriega et al. 2012; Schlacher & Thompson 2012).

Experimental approach

We conducted two complementary experiments to test whether macrobenthic prey and physical habitat attributes are significant predictors of shorebird occurrence, encompassing small (10s to 100s of metres: Experiment 1), and landscape scales (kilometres to 10s of kilometres: Experiment 2).

In Experiment 1 we used a matched ‘case-control’ design (McLoughlin et al. 2010). We compared macrobenthic prey density and physical beach attributes between locations where oystercatchers were actively feeding with matched non-feeding locations situated 200 m away (Fig. 1). Each matched pair of locations was termed a ‘site’, and there were nine sites distributed along 15 km of beach (Fig. 1). The position of a feeding location was determined by the presence of one or more actively foraging pied oystercatchers. All birds occurring on the beach were first mapped with a GPS by driving a vehicle along the entire length of the beach from the north to the south. Experimental feeding locations were then randomly selected, sampling two to four sites per day. Surveys and mapping of birds and site selection was done separately every day. Non-feeding locations at each site were located 200 m either north or south of the feeding locations. To minimize any confounding of spatial differences between locations with possible differences in tidal levels, selection of locations needed to be rapid and all samples were taken within 45 min in both locations. Thus, we first matched locations, and then measured physical beach attributes once faunal samples had been taken. All sampling was done within 2 h of low tide. Birds were not marked and the possibility exists that individuals were re-sampled on subsequent days.

In both the feeding and non-feeding locations, samples were taken from an area centred on the position where the oystercatchers were observed to be searching for food (i.e. probing the sediment) or were actively foraging (i.e. capturing, handling and swallowing food). This area extended for approximately 100 m alongshore and was 20–40 m wide. The upper limit of the sampling area was the highest reach of the swash and the seaward limit was the lowest swash recorded within 10 min of arriving at the site. Twenty sampling stations were distributed within the swash zone of each location: stations were spaced 4–6 m along the beach, while their across-shore position was determined by the up-reach of 20 consecutive swashes. This design ensured that sampling covered the part of the beach where oystercatchers normally forage (e.g. between the highest and lowest swash reach) and that samples were dispersed along the beach to avoid oversampling of small patches.

At each station a composite sample (five cores of 154 mm diameter) for macrofauna and one for sediment (five cores of 30 mm diameter) was collected. Sampling depth was 10 cm as the beaks of oystercatchers do not penetrate deeper (Zwarts & Wanink 1993). Macrobenthos samples were separated from the sediment on site through a sieve of 1-mm-mesh size and preserved in 75% ethanol. Physical habitat surveys and laboratory analyses of fauna and sediment samples followed standard methods for ecological beach surveys (Schlacher et al. 2008a).

In the second experiment, we tested which factors influence the distribution of pied oystercatchers at spatial scales larger than 200 m. To this end, prey assemblages and physical beach attributes were compared between two areas: a stretch of beach where oystercatchers were frequently observed to forage compared with a stretch of beach where oystercatchers were rarely observed or absent; we demarcated these areas based on the distribution of birds during the first 2 days of the field surveys. It is not unusual for highly mobile birds to shift their distribution slightly, resulting in some overlap at the border. However, what is important and relevant to the study is that there was a twofold difference in mean densities of pied oystercatchers between the northern feeding region (mean density of individuals per linear km of beach ± SE = 3.12 ± 0.42) and the southern non-feeding region (1.5 ± 0.41 ind.·km−1). Seven sites were sampled in the southern, 6.3-km-long, ‘few-oystercatcher’ region (Fig. 1). Sample collection and habitat surveys followed the same methods as in Experiment 1, except that only a single location (‘no bird feeding’) was available per site.

Data analyses

We tested whether the occurrence of oystercatchers at a location was influenced by prey availability, prey size, beach morphodynamics, or sediment characteristics separately for two spatial scales using data from Experiment 1 (fine scale) and Experiment 2 (landscape scale). The response variable in each analysis was coded as either 1 (oystercatchers present) or 0 (no oystercatchers) and was hence modelled using logistic regression (binomial link in function GLM of R version 2.12.1). This approach enabled us to develop predictive RSF models of the probability of birds occurring at locations separated by (i) 200 m (Experiment 1) and (ii) 2.5–17 km (Experiment 2), as a function of the availability and size of prey, and physical beach and sediment characteristics.

In each case, the initial (i.e. saturated) model included swash zone slope, mean grain size, shell grit (weight in samples) and the total density of macrofauna (i.e. all species pooled). These predictors were selected from the overall set of variables because they (i) had low collinearity with other predictors and (ii) they were the most likely to influence the distribution of pied oystercatchers based on the literature. The density values of individual macrofauna taxa (e.g. polychaetes, amphipods, bivalves and isopods), were collinear with other predictors (i.e. r values >0.4), and were therefore modelled in separate logistic regressions. Sizes (body length) of prey taxa were also modelled in separated logistic regressions that included only one faunal predictor. We only included taxa which were found at six or more locations and which occurred at each of these locations as five or more individuals. This selection was done to maintain adequate sampling precision and replication, and to avoid zero values for body size (i.e. the length of animal cannot equal zero). Taxa included in the models were the polychaetes Nephtys australiensis and Nephtys longipes, and the isopod Pseudolanna elegans.

In the logistic regressions for Experiment 1, a random effect was included for each site (paired locations, 9 levels) to incorporate spatial autocorrelation (McLoughlin et al. 2010) and models were fitted using mixed-model logistic regression (GLMM, lme4 library v. 0.999375-35 of the R package v. 2.11.1, R Development Core Team 2010). In all logistic regressions, nested log-likelihood ratios (G2) were used to test for the significance of fixed effects and to arrive at the final model (Faraway 2006).

To test whether physical properties of the habitat across a suite of variables differed between areas, we compared key multivariate habitat metrics (e.g. physical extent, slope, substrate) between feeding and non-feeding locations (Experiment 1), and the northern and southern region (Experiment 2) using analysis of similarities (ANOSIM; Clarke 1993).

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

Physical habitat attributes

Habitats where oystercatchers fed closely resembled those without birds in terms of physical extent, slope, and substrate (Experiment 1: ANOSIM R = 0.04, P = 0.25; Experiment 2: ANOSIM R = 0.04, P = 0.31), at both spatial scales investigated (Fig. 2, Table 1). Beach morphodynamic state, measured by the dimensionless fall velocity Ω (Wright & Short 1984), was very similar in the northern feeding sites (Ω mean ± SE = 2.87 ± 0.11), the northern non-feeding sites (Ω = 2.91 ± 0.08), and the southern non-feeding region (Ω = 2.68 ± 0.10); median Ω values did not differ significantly (P > 0.05) in any spatial contrast.

Table 1. Comparison of physical habitat attributes between feeding and non-feeding locations of pied oystercatchers at two spatial scales. Wilcoxon signed-ranks tests compare medians between matched feeding and non-feeding locations at the small scale (Experiment 1), whereas Mann–Whitney U-tests compare medians between the northern feeding and southern non-feeding region at the landscape scale (Experiment 2)
physical attributesmall scale (Experiment 1)landscape scale (Experiment 2)
north – feeding (n = 9)north – non feeding (n = 9)south - non feeding (n = 7)
mean (SE)mean (SE)P (Wilcoxon)mean (SE)P (Mann–Whitney U)
beach-face width (m)72 (3.63)67 (2.76)0.16769 (5.17)0.593
beach-face slope (°)2.47 (0.11)2.71 (0.12)0.2032.70 (0.17)0.138
swash-zone height (m)1.60 (0.22)1.67 (0.10)0.7341.95 (0.15)0.314
swash-zone width (m)32 (3.88)26 (2.32)0.10832 (5.22)0.957
swash-zone slope (°)1.87 (0.14)1.90 (0.06)0.6521.97 (0.11)0.223
sediment grain size (μm)261 (7.60)257 (5.45)0.910273 (7.46)0.223
image

Figure 2. Comparison of beach profiles between feeding and non-feeding locations in Experiment 1 in the northern feeding region (cf. Fig. 1), and the southern, non-feeding region (Experiment 2). The arrow indicates the main part of the beach where oystercatchers forage during low tides. LWST = low-water spring tide.

Download figure to PowerPoint

Resource selection functions

In Experiment 1 (small scale), neither the overall density of macrofauna, the mean sediment grain size, the slope of the swash zone, nor the amount of shell grit influenced the probability of oystercatchers occurring at a location (Table 2, Fig. 3). Similarly, when analyzed by individual groups of prey taxa, oystercatcher occurrence was not influenced by the abundance of the isopod species Pseudolana elegans, polychaetes or amphipods (Table 2). The abundance of bivalves, was, however, likely to have influenced bird occurrence (P = 0.057, Table 2), with a trend of higher probability of oystercatchers to occur in patches with more bivalves (logit β ± SE = 0.84 ± 0.60). Similar to the weak or non-significant effects of prey density and habitat attributes, the size of prey items (i.e. body length of Nephtys australiensis, Nephtys longipes and Pseudolana elegans) did not influence the probability of bird occurrence (Table 2).

Table 2. Results of logistic regressions in the small-scale Experiment 1 (adjacent feeding and non-feeding locations) and the landscape scale Experiment 2 (northern feeding region and southern non-feeding region). All values are given as G2, the log-likelihood ratio test statistic, calculated from the difference in deviance of the full and reduced models. The sample size is given in subscript (feeding, non-feeding, northern feeding or southern non-feeding). Models for the density and size of prey taxa were compared to the intercept-only model
termssmall scale (Experiment 1)landscape scale (Experiment 2)
  1. *P < 0.05, #P = 0.057.

full model
macrofauna density1.8489,94.4379,7*
swash slope3 × 10−49,90.9509,7
shell grit2.5739,91.279,7
grain size0.8199,90.0259,7
density of prey taxa
bivalves3.6149,9#2.5019,9
amphipods0.7799,94.6779,9
polychaetes (Po)0.0669,90.6099,9
isopods (Is)0.0029,90.4089,9
size of prey taxa
Nephtys australiensis (Po)0.0119,90.3879,7
Nephtys longipes (Po)0.8749,90.2148,6
Pseudolanna elegans (Is)0.0028,80.1699,6
image

Figure 3. Comparison of physical habitat properties (a–i), and densities of benthic prey taxa (j–s) between oystercatcher feeding locations (n = 9) and non-feeding locations (n = 9) in Experiment 1 (left column), and between the northern feeding region (n = 9) and the southern non-feeding region (n = 7) in Experiment 2 (right column).

Download figure to PowerPoint

By contrast, in Experiment 2, which focused at the landscape scale, actively foraging oystercatchers were significantly more likely to occur in the region where benthic prey items were more numerous (Table 2, logit β ± SE = 0.050 ± 0.028, Fig. 3). The final model predicted that oystercatchers had only a 7% chance of occurring in a patch with low macrobenthos density (11 ind. m−2), compared with an 81% chance of occurring in a patch with high macrobenthos density (54 ind. m−2). As was the case for small-scale models, sediment grain size, swash slope and the amount of shell grit did not influence the overall likelihood of oystercatcher occurrence (Table 2). Bird occurrence was positively associated with amphipod abundance (logit β = 0.12 ± 0.079) but not with polychaetes or isopods (Table 2). Prey size, measured for N. australiensis, N. longipes and P. elegans, did not influence bird occurrence (Table 2).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

Habitat selection by shorebirds has long been the subject of intensive research but the spatial dependence of habitat selection by shorebirds has not been tested on open-coast beaches. Here we tested the predictive hypothesis that the spatial distribution of pied oystercatchers does not closely match that of invertebrate prey on a high-energy beach. This prediction was supported at small scales (i.e. adjacent patches separated by 200 m): we found no evidence for a strong influence of prey density on oystercatcher distributions. However, when we expanded the analysis to the landscape scale that included regions separated by 2–17 km, the opposite was true: the distribution of oystercatchers corresponded to that of the overall density of their invertebrate prey. This suggests that there are similarities to sheltered shores, where resource selection by shorebirds can be more strongly expressed at broader spatial scales (Colwell & Landrum 1993; Folmer et al. 2010).

Small scale: habitat selection between matched feeding and non-feeding locations

There were very few differences in the measured biotic or abiotic variables between feeding and non-feeding locations, suggesting that pied oystercatchers had a random distribution at this spatial scale. Furthermore, locations over the full range of prey densities in our study were occupied by oystercatchers (Fig. 3). The most likely explanation for these results is that both matched feeding and non-feeding locations occurred within a larger foraging area, suggesting that effective feeding patches for oystercatchers are larger than 200 m. Macrofauna on sandy beaches can be aggregated in alongshore elliptical patches that range from several metres up to several kilometres in size (Defeo & McLachlan 2005). Macrofauna prey densities in our study were also highly similar between matched pairs (Fig. 3), suggesting that they occurred within larger homogeneous patches.

It is also possible that oystercatchers were simply unable to perceive the profitability of feeding patches at this small spatial scale, i.e. 100s of metres, because many of their prey were either too small or buried in the sediments (sensu Santos et al. 2009). The density of pied oystercatchers is also associated with the density of bivalves on sheltered shores in Australia (Spruzen et al. 2008). One of the few conspicuous prey species of pied oystercatchers in this area were bivalves, Donax deltoides (Schlacher et al. 2008b; Sheppard et al. 2009). The siphons of this species regularly become visible in the swash zone when the clams feed. The presence of clam siphons on the beach surface initiates a feeding responses in pied oystercatchers up to around 10 m (T. A. Schlacher, personal observations). This could explain the trend towards more bivalves in feeding than non-feeding locations in our small-scale experiment.

Traditionally, mismatches between the distribution of shorebirds and their prey have also been explained by social interactions (Folmer et al. 2010; Rutten et al. 2010). Oystercatchers are known to be sensitive to interference because of kleptoparasitism (Rutten et al. 2010). However, pied oystercatchers are only strongly territorial at the height of the breeding season (Marchant & Higgins 1993) and we observed no antagonistic interactions during our study, which was done during the non-breeding season. Furthermore, feeding locations in our study were separated by at least a few 1000 m (Fig. 1), making interference behaviour highly unlikely. We therefore do not think that social interactions explained the mismatch between oystercatchers and prey density at the small spatial scale recorded here.

Landscape scale: habitat selection between feeding and non-feeding region

Social interactions were also unlikely to have influenced the broader scale distribution of oystercatchers, because few oystercatchers were observed in the southern region (Fig. 1). The most parsimonious explanation for oystercatchers preferring the northern region is that they chose to forage within an area that had higher prey density, as has been found in numerous studies on sheltered shores (Colwell & Landrum 1993; Piersma et al. 1993; Ribeiro et al. 2004; Spruzen et al. 2008). A recent study suggests that in interference-sensitive shorebirds, such as pied oystercatchers, a large proportion of suitable habitat becomes occupied (Folmer et al. 2010). The fact that a sizeable proportion of potential beach habitat in our study (the southern non-feeding region) was not occupied by pied oystercatchers further suggests that the availability of food resources was likely to have been important at the landscape scale.

Whether decisions on habitat choice in our study were based on prey density per se, prior knowledge of habitat quality or another unmeasured factor is unknown (sensu Santos et al. 2009). Distance from high-tide roosting sites is unlikely, as it is not an important predictor of oystercatcher densities elsewhere (Folmer et al. 2010) and the distance between high-shore roosting sites and the feeding areas was very similar in each of the regions included in our study (i.e. beach width; Table 1). The higher density of oystercatchers in an area with higher prey density may have been facilitated by the greater temporal stability of prey distributions generally exhibited by invertebrates at larger spatial scales (Morrisey et al. 1992; Santos et al. 2009).

Influence of the physical environment on habitat choice

At both the small and landscape spatial scale, beach morphodynamics was similar irrespective of whether they were preferred feeding locations of oystercatchers or not. Key metrics for sandy beach habitats (e.g. slope and width of the swash zone where oystercatchers fed) were relatively invariable along the shore (Table 1). Similarly, sediment characteristics had little influence on the distribution of oystercatchers because there were no systematic differences in physical substrate properties between feeding and non-feeding locations at either the small or the landscape scale (Table 1, Fig. 3). Hence, unlike the influence of coarse or compacted substrata on bird distribution recorded elsewhere (Grant 1984), the largely unconsolidated, medium-grained sands across all of our study locations were unlikely to have limited the ability of oystercatchers to probe for buried prey.

Conclusions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

Shorebirds on ocean-exposed sandy beaches were not distributed randomly in relation to the density of prey, despite foraging on mobile and spatially highly variable benthic prey in a dynamic, high-energy environment. However, the association between the distribution of oystercatchers and their prey was only exhibited at the landscape scale (i.e. 2–17 km) examined, where it resembled, conceptually, results obtained on more sheltered shores (Colwell & Landrum 1993; Placyk & Harrington 2004; Spruzen et al. 2008). This suggests that the proximate behavioural mechanisms driving habitat choice in oystercatchers can be similar between sheltered and high-energy shores, but that the predictability and size of prey patches modulates this behavioural response in a spatial context.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References

This work was financially supported by funds from three sources: Redlands Shire Council (RSC), the Department of Environment and Resource Management (DERM, formerly Environmental Protection Agency, EPA), and the Queensland Wader Study Group (QWSG). We especially thank Daniel Carter (RSC), David Rissik (EPA) and Dave Milton (QWSG) for championing our shorebird work on sandy shores. The hundreds of core samples could not have been collected without the valiant efforts of Mathew Nielsen, Margunn Kalvatn, Rudi de Jaeger and Jeffrey Ward – ‘thanks mates’.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusions
  8. Acknowledgements
  9. References
  • Addicott J.F., Aho J.M., Antolin M.F., Padilla D.K., Richardson J.S., Soluk D.A. (1987) Ecological neighborhoods: scaling environmental patterns. Oikos, 49, 340346.
  • Botton M.L., Loveland R.E., Jocobsen T.R. (1994) Site selection by migratory shorebirds in Delaware Bay, and its relationship to beach characteristics and abundance of horseshoe crab (Limulus polyphemus) eggs. The Auk, 111, 605616.
  • Boyce M.S. (2006) Scale for resource selection functions. Diversity and Distributions, 12, 269276.
  • Boyce M.S., Mao J.S., Merrill E.H., Fortin D., Turner M.G., Fryxell J., Turchin P. (2003) Scale and heterogeneity in habitat selection by elk in Yellowstone National Park. Ecoscience, 10, 421431.
  • Ciarniello L.M., Boyce M.S., Seip D.R., Heard D.C. (2007) Grizzly bear habitat selection is scale dependent. Ecological Applications, 17, 14241440.
  • Clarke K.R. (1993) Non-parametric multivariate analyses of changes in community structure. Australian Journal of Ecology, 18, 117143.
  • Cody M.L. (1981) Habitat selection in birds: the roles of vegetation structure, competitors, and productivity. BioScience, 31, 107113.
  • Colwell M.A., Landrum S.L. (1993) Nonrandom shorebird distribution and fine-scale variation in prey abundance. Condor, 95, 94103.
  • Defeo O., McLachlan A. (2005) Patterns, processes and regulatory mechanisms in sandy beach macrofauna: a multi-scale analysis. Marine Ecology Progress Series, 295, 120.
  • Duchesne T., Fortin D., Courbin N. (2010) Mixed conditional logistic regression for habitat selection studies. Journal of Animal Ecology, 79, 548555.
  • Dugan J.E., Hubbard D.M., McCrary M.D., Pierson M.O. (2003) The response of macrofauna communities and shorebirds to macrophyte wrack subsidies on exposed sandy beaches of southern California. Estuarine, Coastal and Shelf Science, 58(Suppl S), 2540.
  • Dugan J.E., Defeo O., Jaramillo E., Jones A.R., Lastra M., Nel R., Peterson C.H., Scapini F., Schlacher T., Schoeman D.S. (2010) Give beach ecosystems their day in the sun. Science, 329, 1146.
  • Ens B.J., Goss-Custard J.D. (1984) Interference among oystercatchers, Haematopus ostralegus, feeding on mussels, Mytilus edulis, on the Exe Estuary. Journal of Animal Ecology, 53, 217231.
  • Faraway J.J. (2006). Extending the Linear Model with R: Generalized Linear, Mixed Effects and Nonparametric Regression Models. IX. Chapman & Hall/CRC, Boca Raton: 301.
  • Finn P.G., Catterall C.P., Driscoll P.V. (2007) Determinants of preferred intertidal feeding habitat for Eastern Curlew: a study at two spatial scales. Austral Ecology, 32, 131144.
  • Finn P.G., Catterall C.P., Driscoll P.V. (2008) Prey versus substrate as determinants of habitat choice in a feeding shorebird. Estuarine Coastal and Shelf Science, 80, 381390.
  • Folmer E.O., Olff H., Piersma T. (2010) How well do food distributions predict spatial distributions of shorebirds with different degrees of self-organization? Journal of Animal Ecology, 79, 747756.
  • Goss-Custard J. (1996). The Oystercatcher: from Individuals to Populations. Oxford University Press, Oxford: 442.
  • Goss-Custard J.D., Jones R.E., Newbery P.E. (1977) The ecology of the wash. I. Distribution and diet of wading birds (Charadrii). Journal of Applied Ecology, 14, 681700.
  • Grant J. (1984) Sediment microtopography and shorebird foraging. Marine Ecology Progress Series, 19, 293296.
  • Hubbard D.M., Dugan J.E. (2003) Shorebird use of an exposed sandy beach in southern California. Estuarine Coastal and Shelf Science, 58, 4154.
  • Kelsey M.G., Hassall M. (1989) Patch selection by dunlin on a heterogeneous mudflat. Ornis Scandinavica, 20, 250254.
  • Klaassen R.H.G., Nolet B.A., de Fouw J. (2006) Intake rate at differently scaled heterogeneous food distributions explained by the ability of tactile-foraging mallard to concentrate foraging effort within profitable areas. Oikos, 112, 322331.
  • Lauro B., Nol E. (1995) Feeding behaviour, prey selection, and bill size of pied oystercatchers in Australia. The Wilson Bulletin, 107, 629640.
  • Marchant S., Higgins P.J. (1993). The Handbook of Australian, New Zealand and Antarctic Birds. Volume 2: Raptors to Lapwings. Oxford University Press, Melbourne: 984.
  • McLoughlin P.D., Morris D.W., Fortin D., Vander Wal E., Contasti A.L. (2010) Considering ecological dynamics in resource selection functions. Journal of Animal Ecology, 79, 412.
  • Meager J.J., Schlacher T.A., Nielsen T. (2012) Humans alter habitat selection of birds on ocean-exposed sandy beaches. Diversity & Distributions, 18, 294306.
  • Morrisey D.J., Howitt L., Underwood A.J., Stark J.S. (1992) Spatial variation in soft-sediment benthos. Marine Ecology Progress Series, 81, 197204.
  • Noriega R., Schlacher T.A., Smeuninx B. (2012) Reductions in ghost crab populations reflect urbanization of beaches and dunes. Journal of Coastal Research, 28, 123131.
  • Peterson C.H., Bishop M.J., Johnson G.A., D'Anna L.M., Manning L.M. (2006) Exploiting beach filling as an unaffordable experiment: benthic intertidal impacts propagating upwards to shorebirds. Journal of Experimental Marine Biology and Ecology, 338, 205221.
  • Piersma T., Hoekstra R., Dekinga A., Koolhaas A., Wolf P., Battley P., Wiersma P. (1993) Scale and intensity of intertidal habitat use by knots Calidris canutus in the Western Wadden Sea in relation to food, friends and foes. Netherlands Journal of Sea Research, 31, 331357.
  • Placyk J.S., Harrington B.A. (2004) Prey abundance and habitat use by migratory shorebirds at coastal stopover sites in Connecticut. Journal of Field Ornithology, 75, 223231.
  • Railsback S.F., Harvey B.C. (2002) Analysis of habitat-selection rules using an individual-based model. Ecology, 83, 18171830.
  • Ribeiro P.D., Iribarne O.O., Navarro D., Jaureguy L. (2004) Environmental heterogeneity, spatial segregation of prey, and the utilization of southwest Atlantic mudflats by migratory shorebirds. Ibis, 146, 672682.
  • Rutten A.L., Oosterbeek K., Verhulst S., Dingemanse N.J., Ens B.J. (2010) Experimental evidence for interference competition in oystercatchers, Haematopus ostralegus. II. Free-living birds. Behavioral Ecology, 21, 12611270.
  • Santos C.D., Saraiva S., Palmeirim J.M., Granadeiro J.P. (2009) How do waders perceive buried prey with patchy distributions? The role of prey density and size of patch Journal of Experimental Marine Biology and Ecology, 372, 4348.
  • Schlacher T.A., Lucrezi S. (2010) Experimental evidence that vehicle traffic changes burrow architecture and reduces population density of ghost crabs on sandy beaches. Vie et Milieu, 60, 313320.
  • Schlacher T.A., Morrison J.M. (2008) Beach disturbance caused by off-road vehicles (ORVs) on sandy shores: relationship with traffic volumes and a new method to quantify impacts using image-based data acquisition and analysis. Marine Pollution Bulletin, 56, 16461649.
  • Schlacher T.A., Thompson L.M.C. (2007) Exposure of fauna to off-road vehicle (ORV) traffic on sandy beaches. Coastal Management, 35, 567583.
  • Schlacher T.A., Thompson L.M.C. (2008) Physical impacts caused by off-road vehicles to sandy beaches: spatial quantification of car tracks on an Australian barrier island. Journal of Coastal Research, 24, 234242.
  • Schlacher T.A., Thompson L. (2012) Beach recreation impacts benthic invertebrates on ocean-exposed sandy shores. Biological Conservation, 147, 123132.
  • Schlacher T.A., Thompson L., Price S. (2007) Vehicles versus conservation of invertebrates on sandy beaches: mortalities inflicted by off-road vehicles on ghost crabs. Marine Ecology, 28, 354367.
  • Schlacher T.A., Richardson D., McLean I. (2008a) Impacts of off-road vehicles (ORVs) on macrobenthic assemblages on sandy beaches. Environmental Management, 41, 878892.
  • Schlacher T.A., Thompson L.M.C., Walker S.J. (2008b) Mortalities caused by off-road vehicles (ORVs) to a key member of sandy beach assemblages, the surf clam Donax deltoides. Hydrobiologia, 610, 345350.
  • Schlacher T.A., Schoeman D.S., Dugan J., Lastra M., Jones A., Scapini F., McLachlan A. (2008c) Sandy beach ecosystems: key features, sampling issues, management challenges and climate change impacts. Marine Ecology, 29, 7090.
  • Sekercioglu C. (2006) Increasing awareness of avian ecological function. Trends in Ecology & Evolution, 21, 464471.
  • Sheppard N., Pitt K.A., Schlacher T.A. (2009) Sub-lethal effects of off-road vehicles (ORVs) on surf clams on sandy beaches. Journal of Experimental Marine Biology and Ecology, 380, 113118.
  • Spruzen F.L., Richardson A.M.M., Woehler E.J. (2008) Influence of environmental and prey variables on low tide shorebird habitat use within the Robbins Passage wetlands, Northwest Tasmania. Estuarine Coastal and Shelf Science, 78, 122134.
  • Weston M.A., Elgar M.A. (2005) Disturbance to brood-rearing Hooded Plover Thinornis rubricollis: responses and consequences. Bird Conservation International, 15, 193209.
  • Weston M.A., Ehmke G.C., Maguire G.S. (2009) Manage one beach or two? Movements and space-use of the threatened hooded plover (Thinornis rubricollis) in south-eastern Australia Wildlife Research, 36, 289298.
  • Wright L.D., Short A.D. (1984) Morphodynamics variability of surf zones and beaches: a synthesis. Marine Geology, 56, 93118.
  • Yasue M. (2006) Environmental factors and spatial scale influence shorebirds' responses to human disturbance. Biological Conservation, 128, 4754.
  • Zwarts L., Wanink J.H. (1993) How the food supply harvestable by waders in the Wadden Sea depends on the variation in energy density, body weight, biomass, burying depth and behaviour of tidal-flat invertebrates. Netherlands Journal of Sea Research, 31, 441476.