The Effect of Human Disturbance on Foraging Behavior and Habitat Use in Piping Plover (Charadrius me/o&s)

Transcription

1 Estuaries Vol. 17, No. 3, p September 1994 The Effect of Human Disturbance on Foraging Behavior and Habitat Use in Piping Plover (Charadrius me/o&s) JOANNA BURGER Biological Sciences Rutgers Univmsity Piscataway, New Jersey l 059 ABSTRACT: Piping plovers breed in coastal areas where they experience intense competition with man. I studied habitat use (using transects) and foraging behavior (using focal animals) at three habitats on each of three nesting beaches over a 2-yr period ( ) in New Jersey, USA, to understand how plovers use space. Piping plovers forage along the tidal oceanfront, in the dunes, and in backbays, and their relative use of these habitats partially depends on the presence of people. Witbin each habitat the plovers select sites that contain fewer people than the habitat as a whole. The time devoted to vigilance (when tbey are not searching for food) is directly related to the number of people near them, and to the overall human use of that habitat. Thus, in habitats with few people the plovers can spend 90% of their foraging time actively searching for prey and feeding, whereas on beaches with many people they may spend less than 50% of their foraging time in direct feeding behaviors. A diversity of habitats allows the bids to move between habitats to minimize interactions with people and maximize the time devoted to foraging. The results suggest that it is critical to maintain high habitat diversity in coastal environments to help mitigate competition with people. Introduction Animals that live along temperate coasts are increasingly constrained by habitat destruction and human disturbance. Available habitat for breeding and foraging may be unsuitable because physical features have changed, human presence has increased, or predator numbers have increased because of human activities. In coastal habitats, native species are competing with people for undisturbed space. In competition with man, the birds sometimes shift to nearby, suboptimal habitats not preferred by people, populations may decline, or local extinctions may occur (Burger 1984; Anderson 1988). In this paper I examine differences in foraging behavior of piping plovers (Charadrius melodus) in different habitats (dunes, beach, backbay). Space partitioning can occur partially as a result of the presence of people. I predicted that plovers accommodate to the presence of people by using all available habitats for foraging. This paper follows from Burger (1991), where foraging plovers were examined only in one habitat (beach, ). Piping plovers nest on wide, sparsely-vegetated beaches or on inland lake shores or river banks (Wilcox 1959; Renaud 1979; Cairns 1982; Haig and Oring 1985). Reproductive success is depressed in areas with high human disturbances, populations have declined dramatically, and the species is listed as endangered in the United States and Canada (Dyer et al., 1987; Haig and Oring 1988). Piping plovers generally forage on intertidal beaches in coastal regions where their foraging behavior is af fected by the presence of people (Tull 1984; Johnson and Baldassarre 1988; Burger 1991). Yet their ability to reduce interactions with people by shifting among habitats has not been examined, and I do so in this paper. Study Areas and Methods Observations were made from April 15 to August 15 in 1988 and 1989 at Holgate, Brigantine, and Corson s Inlet, New Jersey (Fig. 1). Data were gathered from 0700 hours to 1800 hours 5-6 d a week at each of the three sites, requiring three field assistants each year. All assistants had made similar observations in previous years at the same site to establish that there were no significant interobserver variations. All study sites were on barrier beach islands, and all had exposed surf, dunes, and backbay habitats. The observers used binoculars and telescopes to remain far away from the plovers and to avoid affecting their behavior. Brigantine Beach, opposite Atlantic City, is a flat beach 2 km long and 200 m wide with a belt of dunes. On the back side is a bay not exposed to tidal waves, which is used extensively by fisherman during the day and night. Least terns (Sterna antillarum) nest on the flat sand in front of the dunes, and the colony is delineated with string. The dunes are not enclosed by string Estuarine Research Federation /94/ $01.50/O

2 J. Burger Albsny~ ; VT. f N.H.,- r..l_.. -~ ~MA~~AC~~USETTS NEW YORK I I EoSlO v \ IARYLANO Fig. 1. Map of New Jersey showing sites where the foraging behavior and effects of human disturbance on piping plovers were studied in New Jersey, Corson s Inlet is a state park at the end of a barrier beach by an inlet. The beach is less flat and has higher, less stable dunes than Brigantine. The least tern colony here also is delineated by signs and string. State park personnel keep people out of the colony and dunes whenever possible. Holgate, part of Edwin B. Forsythe National Wildlife Refuge, is on the northern barrier spit bordering Little Egg Inlet. It has a long, wide beach with stable dunes and a backbay beach bordering Barnegat Bay. As part of a protection plan for piping plovers, the nesting beach (with associated dunes and backbay) was closed to the public during the nesting season. This research was initiated to evaluate the effect of closure on foraging piping plovers. At each of the three study locations, we designated three habitats or census areas: beach, dunes, and backbay. The beach was the area bordering the Atlantic Ocean with direct wave action, the dunes were stable dunes, and the backbay was the area bordering bays where there were no direct tidal waves, but the tides rose and fell. Two types of data were gathered: foraging behavior and habitat use. Quantifying foraging behavior involved recording data on any foraging plovers encountered for a 2-min sample, and then moving on to another foraging plover (focal animal method). Habitat data were gathered by walking transects (300 m long by 20 m wide) through the three census areas (ocean, dunes, bay), and recording the number, ages, and activities of all plovers. The protocol was the same for all three locations in both years, and field assistants were trained for several days before their observations began. The daily schedule was to randomly select the first transect (ocean, dunes, bay), and then to alternate the transects throughout the day, running five to eight transects each day. Wherever foraging plovers were encountered, 2-min foraging samples were taken. Assistants remained at least 150 m away from the bird, and used binoculars and telescopes to census birds and record feeding data. Judging from the behavior of plovers when other people approached, we felt our presence was not influencing their behavior. There were pairs of nesting plovers per site, as well as some nonbreeding adults early in the season. Habitat data were taken by walking transects and recording how many plovers were present, what they were doing by age (foraging, resting, walking, running, flying, guarding chicks), the distance between family members and nonfamily members, people within 100 m of the plover, and the activities of the people. The transects were demarcated by stakes every 10 m to facilitate censusing and determinations of interbird distances. The transects were selected on the basis of high piping plover use in 1985 and 1986 (Burger 1991), and the dune transects included several nest sites. Foraging behavior data were taken whenever foraging plovers were encountered, whether within the transects or not. Foraging plovers were assigned to a census area (ocean, dunes, bay). Foraging behavior was recorded for 2 min because

3 Effect of Human Disturbance on Plovers 697 TABLE 1. Models (PROC GLM) for explaining variation in the feeding behavior and habitat use of piping plovers at three New Jersey study sites, for 1988 and Model F l;f lv Factors entering (F, p) Year Locatiot+ Habitar Month People within 100 m Seconds Feeding , (0.0001) 103 (0.0001) 12.6 (0.0001) 71.3 (0.0001) 12.8 (0.0002) a Type III. h Brigantine, Corson s Inlet, Holgate. c Dune, ocean, backbay. Plovers by Habitat ,282O (0.0001) 10.4 (0.0001) 40.0 (0.0001) 4.8 (0.0002) 4.9 (0.0001) birds could be easily followed for this period, were not always interrupted, and 2 min was long enough to include a variety of behaviors (see Burger 1991). Plovers that were interrupted and flew away were omitted from further analyses. Data recorded before the start of each 2-min sample included: location, census area, date, time, stage (pre-incubation, incubation, chick stage), sex (male, female, unknown), and interbird distance (distance to nearest neighbor). Stop watches were used to record the time spent in each activity. I defined an alert bird as one that looked up and about (often at people or dogs), rather than looking at the sand for prey or at its chicks. Thus, looking in search of prey was not considered being alert. Data were analyzed using SAS (Proc. GLM) regression procedures on log-transformed data to determine the factors that contribute to explaining the variation in the dependent variables (SAS 1985). This procedure selected the factor that con- tributed the most to the R2, and then selected the second variable that increased the R:! the most. Thus, variables that varied colinearly were not entered in the model (SAS 1985). Significant differences among groups were determined by Wilcoxon tests yielding a x2 statistic. Unless specified otherwise, means and one standard deviation are given in the text. Results FORAGING BEHAVIOR The best models for the time plovers spent feeding (out of the P-min samples) for all sites in both years accounted for 28% of the variation in terms of year, location, census area, month, and the presence of people (Table 1). There were significant location differences in almost all foraging parameters, numbers and activities of people, and numbers of birds foraging (Table 2). Based on the mean 2-min samples, piping plovers spent 106 s feeding at Holgate, 85 s at Corson s Inlet, and 73 s at Brigantine. Most of the other time was spent being alert. It should be noted that since these are P-min sample periods, a high value for one behavior means others must be low. Plovers were significantly more aggressive at Corson s Inlet compared to the other locations, and there were more birds feeding near the focal birds at Corson s Inlet than at the other sites (Table 2). Foraging plovers had more people nearby at Corson s Inlet than at Brigantine, and there were no people at Holgate. Most people were walking near the plovers, rather than fishing, sunbathing, or jogging. The models also indicated significant differences in feeding behavior as a function of habitat, and these are examined in Table 3. There were no significant habitat differences in time allocation for foraging piping plovers at Holgate (Table 3) ; the plovers devoted about 90% of their 2-min samples TABLE 2. Comparison of foraging piping plovers on the three study sites (1988 and 1989) in New Jersey. Each sample was 2-min observation of a Foraging Plover. Given are X 2 SD. Brigantine Holgate Carson s Inlet Wilcoxon x (p) Number of samples Seconds spent: Feeding Alert Conspecific aggression People within 100 m Number of people Fishing Sunbathing Walking Number of feeding birds f * (0.0001) 24.4 t I * (0.0001) t i (0.0001) t c (0.0001) * (Q.0001) f (0.0001) 2.12 * (0.0001) (

4 J. Burger TABLE 3. Habitat comparisons (ocean, dunes, bay) of feeding piping plover and presence of people at three study sites in New Jersey. Seconds ( k SD) devoted to feeding Brigantine Hdgate Carson s Inlet Ocean 76 f Dune ? Bay 70 k * Wilcoxon x2 (p) 48.3 (0.0001) NS People within 100 m (mean + SD) f f (0.01) Ocean 2.03 f k Dunes c f 0.4 Bay * Wilcoxon x2 (p) 19.9 (0.0001) 30.5 (0.0001) 11.8 (0.0006) to feeding and only 6% to being alert in all three habitats. There were almost no people at Holgate. At Corson s Inlet, piping plovers spent a significantly lower proportion of the time feeding in the bay compared to the ocean and dunes (Table 3). There were significantly fewer people in the dunes while the plovers were foraging compared to the ocean and bay transects. At Brigantine there were also significant habitat differences in allocation of time to foraging (Table 3). Plovers spent less time (of the 2-min samples) feeding in the dunes than in the bay or ocean. Like Corson s Inlet, there were differences in the number of people within 100 m. There were more people in the dunes than in the ocean or bay habitats. Overall, for both years, at Brigantine (r = 0.29, p < ) and Corson s Inlet (r = -0.14, p < ) the seconds devoted to foraging were indirectly related to the number of people (Fig. 2). That is, the more people there were, the less time the birds spent foraging. There were no significant correlations at Holgate in 1988, and few in HABITAT USE In the above section, the results show that there are differences in the time allocated to foraging as a function of location (Corson s Inlet, Holgate, Brigantine) and habitat (dunes, ocean, bay). However, these data do not indicate how many piping plovers use the space in relation to people. We used habitat transects to evaluate the presence of people, the presence of plovers, and the activities and habitat shifts of the plovers. Using all data combined, 28% of the variability in the number of all plovers present was explained by year, location, habitat, month, and the number of people within 100 m (Table 1). People within 100 m were significant only for Brigantine (F = 3.4, p < , R2 = 0.26) and Corson s Inlet (F = 4.6, p < , R2 = 0.21) because there was no variation in the number of people at Holgate. At Holgate there were significantly more adults 120, l3 iooz 6 P g 80 - DUNEsl OCEAN/BAY DUNES m OCEAN BAY I 5 BAY m ; 60-1 DUNES Q 401. I. I.,.,.I. OCEAN NUMBER OF PEOPLE WITHIN I OOM TRANSECT El BRIGANTINE. HOLGATE CORSON S Fig. 2. Mean seconds devoted to foraging by piping plovers as a function of the number of people in particular habitats (dunes, ocean, bay) at three study sites in New Jersey. feeding on the bay and ocean than in the dunes, but there were significantly more young feeding in the dunes and ocean than in the backbay (Table 4). Both adults and young rested in the dunes and ocean rather than the bay. Family members remained closer together in the bay, and were at intermediate distances on the other habitats, whereas nonfamily members were closer together on the dunes (often roosting at high tide). At Corson s Inlet there were also significantly more adult plovers feeding on the bay and ocean than the dunes, but there were no differences in where young fed (Table 4). Young rested on the dunes, whereas there were no significant differences in where adults rested. There were no significant differences in interbird distance for either family or nonfamily members as a function of census area. However, there were significantly fewer people in the dune transects than in the ocean or bay transects (Table 4). At Brigantine, more adults and young fed in the ocean transect, intermediate members fed in the bay, and the least number fed on the dunes (Table 4). Most adults and young rested in the dunes. Family members were closer together in the bay transect compared to the other transects. At Brigantine there were more people in the ocean transect, intermediate numbers in the bay transect, and lower numbers in the dunes. Discussion FORAGING BEHAVIOR AND THE EFFECT OF COMPETITION WITH PEOPLE Previously I have shown (Burger 1991) that at several beaches in New Jersey (including Brigantine, Holgate, and Corson s Inlet) breeding piping plovers remain in the vicinity of their territory during the breeding season, often keep visual or vocal contact with their incubating mates, and time devoted to feeding depends on the location,

5 Effect of Human Disturbance on Plovers 699 TABLE 4. Feeding and resting behavior of piping plover in three different habitats, at Holgage, Brigantine and Corson s Inlet, New Jersey. Given are X k SD (1988 and 1989). Holgate Number of adults Backbay DUlV3 OCGSI Wilcoxon,y (p) Feeding 1.3 k k k (0.005) Resting 0.03 k t (0.0007) Number of young Feeding 0.07 * (0.0001) Resting 0.04 k k (NS) Distance to family members (m) When feeding 0.85 k k (NS) Distance to nonfamily plover (m) When feeding 2.06 i t (0.0001) People within 100 m 0.18 t zk (0.002) Brigantine Number of adults Feeding k (0.0001) Resting c AZ (0.0001) Number of young Feeding c t k (0.0001) Resting c (0.0001) Distance to family members (m) While feeding 1.01 k k (0.0001) Distance to nonfamily plover (m) While feeding 2.3 k f f (0.0001) People within 100 m 6.9?z k (0.0001) Corson s Inlet Number of adults Feeding 0.54 i f f (0.0001) Resting 0.04 f * k 8.5 NS Number of young Feeding 0.09 f f k 0.6 NS Resting i k (0.01) Distance to family member While feeding 0.28 k k (NS) Distance to nonfamily plover While feeding t t (NS) People within 100 m 4.6 -c f (0.0001) time of year, time of day, and presence of people. However, that study did not critically examine foraging behavior in different habitats available to the plovers. Foraging plovers can move among habitats (dunes, ocean, bay), balancing several abiotic and biotic factors such as tides and inclement weather, with the presence of conspecifics and people. One objective of this research was to document the use of these different habitats by foraging plover. Habitat entered all location foraging models as an important contributor to variation in the time devoted to feeding (looking for prey and eating). Although there were no habitat differences at Hol- gate, there were differences at the other sites with plovers devoting more time to feeding in the dunes and ocean at Corson s Inlet, and more time in the ocean and bay at Brigantine (Table 3). These differences seem partially attributable to the presence of people. Without people at Holgate, the plovers devote similar amounts of time to feeding in each habitat, and spend more time feeding than do plover at Brigantine or Corson s Inlet. At Corson s Inlet, people are forbidden to go in the dunes by the presence of fences, signs, and park personnel. Thus, the plovers can feed in the dunes relatively undisturbed. The greatest number of people at Corson s Inlet were in the bay transect

6 J. Burger because it is close to the parking lot and does not require a 2-km walk. The situation at Brigantine differs in that no one patrols the dunes, and there are no string fences. People often picnic on the dunes, accounting for higher vigilance rates and lower foraging rates in the dunes compared to the ocean or bay transects. During the breeding season, foraging shorebirds must simultaneously attend to their incubating or brooding mate, chicks when they have them, competitive conspecifics, predators, and people as well as the availability of prey. Thus, even when they are foraging some time must be devoted to vigilance. Presumably piping plovers devote similar time and energy to vigilance for predators regardless of the habitat. Yet the results of this study indicate that time devoted to vigilance varies by habitat (dunes, ocean, bay). That the variations are due to differences among the habitats is countered by the lack of differences at Holgate where there were no people. I suggest these differences are due to the presence of people: with an increase in the number of people the birds increase vigilance time proportionate to the frequency of disturbances. HABITAT USE In coastal areas, piping plovers usually are distributed longitudinally along the coast, with a nesting pair every 100 m or so (Wilcox 1959). In some cases nests may be closer, particularly if there are wide expanses of low-lying foredunes (Burger 1987). Piping plovers nest in the higher areas of the beach to avoid tidal flooding, and normally move to the tidal surf to forage within their territory (Wilcox 1959; Cairns 1977; Johnson and Baldassarre 1988). Barrier beaches and coastal islands often have a variety of habitats that include tidal bays, marshes, dunes, and open beach. These areas are usually in close proximity, or are near enough for plovers to move among them. By using several habitats, plovers can find suitable foraging sites during all tidal stages. Before initiation of this study it was unclear to what degree piping plovers fed in different habitats. The results clearly indicate that adults and young forage in all three habitats (dunes, ocean, backbay), and their relative use depends on local tidal conditions and the presence of people. Even in the absence of people (Holgate), plovers fed in all three habitats, but only half as many plovers fed in the dunes as the ocean or bay. The variability in use of these habitats at Holgate indicates the importance of protecting the entire beach-dune ecosystem and that plovers are flexible in their habitat use. This flexibility is not solely attributable to the presence of people since it exists in the absence of IO NUMBER OF PEOPLE WITHIN I OOM OF FORAGING PLOVER Fig. 3. Mean seconds devoted to foraging by piping plovers as a function of the number of people within 100 m of foraging plover in 1985 through 1989 at five study sites in New Jersey (Brigantine, Holgate, Corson s Inlet, Little Beach, and Whale Beach). Data from this paper and Burger (1991). people. Yet the flexibility allows the plovers to adapt to the presence of people. FORAGING BEHAVIOR, HABITAT USE AND THE EFFECTS OF PEOPLE This study also examined foraging behavior and habitat use as it relates to the presence and activities of people. If piping plover have developed mechanisms of minimizing direct competition with people for space then they should be found with fewer people nearby than occurs in similar-size transects in the same habitat. This was uniformly the case for foraging piping plovers in 1988 and The mean numbers of people around foraging plovers were always lower than in similar-size transects nearby. Similarly, the time devoted to foraging should be inversely related to the number of people in transects (bay, dunes, ocean) if people are affecting habitat use, and this was the case (Fig. 2). Plovers devoted more time to feeding when there were fewer people in the transect. These data, however, are from only 2 yr, and the initial models indicated that year is an important variable. The mean time devoted to foraging and the number of people within 100 m of foraging piping plovers are negatively correlated (Fig. 3)) using foraging data from 1985 to 1987 (Burger 1987, 1991) as well as from this study. This dataset involves 5 yr of data from Corson s Inlet, 4 yr from Brigantine, 3 yr from Holgate, and 1 yr each from Little Beach and Whale Beach. The great variation in the number of people on the beach partially reflects weather differences (1988 was often wet and cloudy so there were fewer people than usual) and medical waste scares (1989). I suggest that piping plovers have adapted to the coastal environment and to the space competition

7 Effect of Human Disturbance on Plovers 701 with people by diversifying their habitat use, sometimes moving among habitats as a function of the presence of people as well as tide. They select those parts of the habitat with the fewest people. This allows the plovers to coexist by temporally and spatially avoiding people. These results suggest that piping plovers that have a diversity of habitats available nearby can more easily cope with space competition from people than those nesting in places with only one or two of these habitats. Much of the beach-dune complex in coastal New Jersey is narrow with eroded dunes. Plovers nesting in such areas have fewer habitats available to avoid competition with people. This could lead to lowered reproductive success, lower population levels, or abandonments. Eventually, these narrow beaches may no longer have breeding piping plovers. Indeed, in New Jersey, almost half of the successful piping plover pairs nest at Holgate, Brigantine, and Corson s Inlet, which are all sites that have complex habitats. These complex habitats may allow the plovers to exist with the presence of people, whereas they have been extirpated from many of the narrow beaches. ACKNOWLEDGMENTS I especially thank M. Gochfeld, K. Staine, and C. Salina for valuable discussions throughout this study, and R. Trout of the Rutgers Statistics Department for advice and help with analysis. Several people helped with the field work and I thank them: D. J. Gochfeld, D. Grembowicz, J. Lucitte, C. Miller, K. Staine, and V. Turner. I thank D. L. Beall and H. Laskowski of the E. B. Forsythe National Wildlife Refuge for permission to work at Holgate. This research was partially funded by the Endangered and Nongame Species Program of the New Jersey Department of Fish, Game, and Wildlife, and I thank J. Frier-Murza, L. Niles, and D. Jenkins for their continued help and advice. LITERATURE CITED ANDERSON, D. W Dose-response relationship between disturbance and brown pelican breeding success. Wild& Society Bulletin 16~ BURGER, J Colony stability in least terns. Condor 86: BURGER, J Foraging behavior and habitat use, Report to New Jersey Department of Environmental Protection. Trenton, New Jersey. BURGER, J Foraging behavior and the effect of human disturbance on the piping plover (Charadrius melodus). Journal of Coastal Research 7: CAIRNS, W. E Breeding biology and behavior of the piping plover in southern Nova Scotia. M.Sc. Thesis. Dalhousie University, Halifax, Nova Scotia. CAIRNS, W. E Biology and behavior of breeding piping plovers. Wilson Bulletin DYER, R. W., A. HECHT, C. RAITHEL, K TERWILLINGER, AND S. MELVIN Atlantic Coast piping plover recovery plan April Draft report to United States Fish and Wildlife Service. Newton Corner, Massachusetts. HAIG, S. M. AND L. W. ORING Distribution and status of the piping plover throughout the annual cycle. Journal offield Ornithology 56: HAIG, S. M. AND L. W. ORING Mate, site, and territory fidelity in piping plovers. Auk JOHNSON, C. M. AND G. A. BALDASSARRE Aspects of the wintering ecology of piping plovers in coastal Alabama. Wilson Bulletin 100: RENAUD, W. E The piping plover in Saskatchewan: A status report. Blue Jay 37: SAS User s Guide: Statistics. SAS Institute Inc. Gary, North Carolina. TULL, E. C A study of the nesting piping plover of Kouchibouguac National Park Contract Report PKB Kouchibouguac National Park, New Brunswick, Canada. WILCOX, L A twenty year banding study of the piping plover. Auk 76: Received for consideration, December 17, 1993 Accepted jw publication, May 19, I994