We used 19 years of mark-recapture/resighting data collected on 11,020 birds from 1988-2006 at five colony sites in Massachusetts, New York, and Connecticut, USA, to examine temporal variation in the survival rates of adult Roseate Terns (Sterna dougallii) during periods of overall population increase (1988-2000) and decline (2000-2006). Roseate Terns nested at only one colony site in Buzzards Bay, Massachusetts at the start of this period, but two more sites in this area were recolonized as the study progressed. Adult survival rates varied temporally in different ways at the different colony sites, but for the five sites combined they did not differ between the period of overall population increase (0.835 ± 0.006 SE) and the period of population decline (0.835 ± 0.008 SE). As expected based on previous work, adult survival from 1991 to 1992 was lowered as a result of a severe hurricane in August 1991. An oil spill in Buzzards Bay in April 2003 did not appear to result in lower survival of the birds nesting at the sites in this area compared to those nesting at the other two study sites in New York and Connecticut. Temporal variation in other vital rates of this species (such as postfledging survival) needs to be examined to determine the likely cause(s) of the recent population decline.

Roseate terns are rare and restricted range breeders in NW Europe, with a population size of below 1,100 pairs and 95% of these within four colonies. The apparent survival and breeding dispersal rates of adults was estimated from live resighting data collected at three of the largest colonies between 1995 and 2007. These parameters were estimated using a multi-state model implemented in program MARK, with transition rates equating to breeding dispersal. The program U-CARE identified transience and trap-dependence in the data and so these effects were specified in the model. Time-averaged apparent survival rates were 0.735 yr-1 for birds during the interval following their first encounter and 0.855 yr-1 for “older” birds. Survival varied among years but not colonies. Breeding dispersal rates varied among colonies, being highest towards the largest and most productive colony, intermediate in the reverse direction and lowest between the two smaller colonies. Since the largest colony is not en route to either of the smaller colonies this observation cannot be explained by the order in which birds encounter sites during spring migration, so we propose that birds either follow the bulk of birds migrating north in spring or choose to breed there after prospecting during the staging period of the previous year. The differences in numbers breeding at each colony, however, mean that the numbers moving in different directions among colonies broadly balance out, such that net movements in terms of the number of individuals involved are relatively small.

Survival of tropical Roseate Terns (Sterna dougallii) from the western Indian Ocean was modeled using a capture-mark-recapture dataset of 102 breeding adults ringed and resighted on Aride Island, Seychelles, from 2002 to 2007. The effect of covariates reflecting oceanographic conditions and resighting effort was also examined during the modeling. A time-invariant survival rate was best supported by our data, with annual adult survival estimated at 0.807 ± 0.033 (SE). Resighting probability was found to be influenced by sea-surface temperature (SST) anomalies, as expressed by the mean Indian Ocean Dipole mode index recorded during the nonbreeding period. Years of higher SST in the western part of the Indian Ocean in September-April (8 months) corresponded to a lower probability of resighting adults at the colony during the breeding season (May-August), which may be related to a tendency for some adults to refrain from breeding in less favorable years. Comparing our results with temperate studies, Roseate Terns breeding on Aride were found to exhibit similar survival estimates. Consequently, this study does not support the hypothesis that tropical Roseate Terns may counterbalance their lower fecundity (clutch size and breeding success) compared to their temperate and northern-hemisphere counterparts (Europe, North America, Caribbean) by relatively higher survival rates.

Capture-recapture data from two disparate breeding populations of Roseate Terns (Sterna dougallii) captured together as non-breeding individuals from 2002 to 2007 in the southern Great Barrier Reef, Australia were analyzed for both survival rate and recapture rate. The average annual survival rate for the two populations in the study area was estimated at 0.850. There is strong evidence that the average annual survival rate for the birds from the Asian population (S. d. bangsi) (0.901) is higher than that of the other population of unknown breeding origin (0.819). There was large variability in survival in both populations among years, but the average survival rate of 0.85 is similar to estimates for the same species in North America. The Cormack-Jolly-Seber models used in program MARK to estimate survival rates also produced estimates of recapture probabilities and population sizes. These estimates of population size were 29,000 for S. d. bangsi and 8,300 for the ‘unknown’ group. The population estimate for S. d. bangsi is double the visual estimates of the numbers in the study area and much larger than the documented numbers in the likely breeding areas, suggesting that many breeding sites are currently unknown.

This paper presents a summary of the four preceding papers and integrates them with previously-published data to compare demographic parameters of Roseate Terns breeding in tropical and temperate regions. Tropical Roseate Terns appear to breed less frequently than temperate birds and raise far fewer chicks to fledging when they do breed. Nevertheless, the three estimates of adult survival rates of tropical populations that were presented in this symposium are lower than those for temperate populations (0.71-0.82 yr^-1 vs 0.83-0.90 yr^-1). Two estimates of juvenile survival (from fledging to age 1 or 2 years) of tropical birds are slightly higher than comparable estimates for a temperate population, but the differences are not sufficient to offset the lower productivity. Data on ages at first breeding are similar in the two regions. Given their irregular breeding and low productivity, the true survival rates of tropical Roseate Terns must be much higher than those estimated in this symposium. The adult survival rates of temperate Roseate Terns are unusually low compared to other seabirds with similar life-history characteristics. This is partially offset by consistently high productivity, but the reported demographic parameters cannot account for the observed long-term increases in numbers of two well-studied regional populations. Despite intensive study of Roseate Terns in several parts of their world range during the last 25 years, important features of the demography of both tropical and temperate populations remain poorly understood.

Parental care behaviors often differ in dimorphic seabirds, leading to sex-specific differences in foraging behaviors. However, few studies have examined sex-specific foraging behaviors in monomorphic seabirds. Using radio-telemetry, we studied Forster’s Terns (Sterna forsteri)-a monomorphic and socially monogamous seabird-breeding in the South San Francisco Bay, California. Space use did not differ between males and females. Instead, space use varied by breeding stage and colony affiliation. Forster’s Terns were located farthest from the nest during pre-breeding and post-breeding time periods, and closest to the nest during incubation and chick-rearing. Home-range size and core-use areas decreased as the breeding season progressed and were most concentrated in the post-breeding stage. The results of this and other studies indicate that terns, unlike other monomorphic seabirds studied, do not exhibit sex-specific differences in space use. .

Female American Avocets (Recurvirostra americana) were radio-marked (N = 15) and tracked in the South San Francisco Bay, California, to determine if space use varied by breeding stage. Visual observations were used to determine breeding stage (pre-incubation, incubation, brood-rearing, post-breeding) of marked avocets. Space use measurements (linear movements, home ranges, core areas, and average distance from nest) varied significantly among breeding stages. Space use was greatest for the post-breeding stage, followed by pre-incubation, incubation, and brood-rearing. Most avocet nests (93%) were located within their pre-incubation core area boundaries, whereas only 36% of nests were within post-breeding core areas. Distance between daily location and future nest sites decreased significantly as the number of days prior to incubation decreased, suggesting that avocets prospected future nest sites several weeks prior to nesting. These data indicate that breeding stage influences space use of female American Avocets and illustrates the importance of delineating breeding stages to better understand space use of avian species.

We compared the composition and overlap in the chick diets of the two most abundant auks in the Northwest Atlantic, the Common Murre (Uria aalge; hereafter murres) and the Atlantic Puffin (Fratercula arctica; hereafter puffins). Our objective was to assess how auks with different foraging strategies (dietary specialization among murres vs. generalization among puffins) and physiological constraints imposed by differences in body size and prey-loading meet the energy demands of reproduction. Murres provisioned chicks with female capelin (>100 mm): 98.8% by number in 2004 and 95.5% in 2005. The relative contribution of gravid capelin to murre chick diets decreased by an order of magnitude between years (41.1 and 4.9% by number in 2004 and 2005, respectively) and was related to the timing of capelin spawning. Puffins provisioned chicks with high proportions of forage fish in both years with the focal species changing from sandlance Ammodytes spp. in 2004 (97.6% by index of relative importance, IRI) to capelin (>100 mm) in 2005 (92.3% IRI). Murre fledglings were significantly lighter in 2005 (203.0 ± 4.6 g) compared to 2004 (215.0 ± 3.9 g) whereas puffins chicks exhibited no such differences. These results suggest that generalized feeding is a viable strategy for puffins but that specialized feeding by murres can lead to poor offspring condition when there is low overlap between peak food availability (i.e., capelin spawning) and chick-rearing.

We documented the movement and distribution patterns of wintering Surf Scoters (Melanitta perspicillata) and White-winged Scoters (Melanitta fusca) in relation to herring spawn events in the Strait of Georgia, British Columbia. Radio-telemetry and surveys were conducted in Baynes Sound, an important wintering area where scoters feed primarily on clams. In early March, herring spawn events in areas adjacent to Baynes Sound provide a short-term pulse of abundant and easily accessible food, which could affect habitat use by wintering scoters from Baynes Sound. Radio-marked Surf Scoters and White-winged Scoters exhibited limited movements during winter, in contrast to the spring herring spawn season, when both scoter species moved greater distances to access herring eggs. Most individuals were located near spawning locations at least once during the spawning season, and the majority of telemetry locations were close to spawning sites, with Surf Scoters showing a higher association with spawn for both metrics. A marked decrease (66-98%) in the abundance of both scoter species in Baynes Sound was observed coincident with spawn initiation in adjacent sites. We conclude that scoters altered their movement and habitat use patterns in spring to take advantage of herring roe, an energy-rich food source. This dramatic change in behaviour suggests that herring spawn may be of particular importance to these species.

Migration chronology was quantified for 15 waterfowl species on 58 playa wetlands in the Southern High Plains of Texas from February 2004 through April 2006. Abundance of each species was estimated on playas once every two weeks during the nonbreeding season (16 August to 30 April); presence of ice was also recorded. Dabbling ducks were most common (N = 250,668) and most tended to exhibit either a bimodal migration pattern (lower abundance in winter than during fall and spring passage) or a unimodal pattern (one defined peak). Abundance of the most common dabbling ducks was skewed toward late winter and spring. Most species of diving ducks (N = 15,128) tended to exhibit irregular migration patterns. Canada Geese (both Branta canadensis and B. hutchinsii, N = 15,347) had an abundance pattern that gradually increased, peaking in midwinter, and then decreased, which is typical for a terminal wintering area. Ice was most common on playas during the first half of December, which coincided with the lowest winter abundance in dabbling ducks. Data from this study will support management efforts focused on playa wetlands, including the development of population goals and habitat objectives that span the entire non-breeding season.

Nests of many over-water nesting birds are frequently subjected to wave action from repeated windstorms. Colonial grebes often lose over 50% of nests in any particular breeding season to wave action, depending on factors such as storm activity and habitat quality. We examined nest structure survival in a colony of Western Grebes (Aechmophorus occidentalis) in relation to wave action due to wind. Initial nests were 66% successful in hatching at Lake Christina, Minnesota, in 2005 compared to only 48% in 2006. In 2005, initial nests that survived and hatched young were closer to the central tuber-mat of the colony and closer to individual bulrush island centers than were nests lost during storms. In 2006, a year of high storm activity, nests located near the colony edge most prone to wave attack were less likely to last the season and produce young than were nests in central or more protected areas of the colony. Buffer zones of vegetation surrounding the nest were the single most important factor in attenuating waves. Buffer zone stem density was less important than the amount of buffer between nest and wave attack, which may be a major factor colony founders select for when evaluating nest placement sites within a colony.

Many species of waterbirds nest over-water in shallow lakes and marshes in the Northern Plains region of North America. High winds are common in this region and the resulting waves can cause significant loss of nesting platforms, yet the behavior of waves in these systems has not been well-studied. We used observational and experimental techniques to examine how Hardstem Bulrush (Schoenoplectus acutus Muhl.) impacts wave attenuation through its clonal growth characteristics. The predictable growth patterns influence nest placement and nesting success in colonial nesting grebes and other over-water nesting waterbirds. We constructed an artificial nesting colony to experimentally test how location with respect to wave-exposure affects nest structural integrity in grebes. Additional experiments with boat wakes confirmed the importance of emergent vegetation in attenuating waves caused by watercraft. Nests gain more protection from being placed on the lee side of a stand of vegetation than by having denser vegetation right at the nest. Over-water nests that are more protected from wave action by vegetation are at least three times more likely to hatch eggs than more exposed nests.

We investigated the use of crayfish gastroliths and exoskeleton pieces for quantifying crayfish abundance and size-structure in diets of White Ibis (Eudocimus albus) chicks in the Everglades. We then quantified crayfish and fish abundance from various small hard parts and intact fish heads in 23 boluses, taken from two nesting colonies in Water Conservation Area 3 (WCA 3) and Arthur R. Marshall Loxahatchee National Wildlife Refuge (Lox) in the Florida Everglades during May 2006. We determined that using gastroliths to estimate crayfish numbers in boluses had significant drawbacks; only a small fraction of field-caught crayfish bear gastroliths and the two crayfish species in the Everglades differ in percentage bearing gastroliths. In contrast, counts of crayfish rostrums and chelae pairs gave simple and similar estimates of crayfish in the boluses. The two colonies had strikingly different diets in May 2006; New Colony 3 (Lox) boluses were dominated by crayfish while birds from the Alley North (WCA 3) boluses were fish-dominated and had few crayfish. Using measurements of the crayfish rostrums we determined the size-structure of crayfish found in the diets of the New Colony 3 birds, and determined that the crayfish in the diet were relatively large (mean = 19 mm carapace length) when compared to the available crayfish in the marsh. These crayfish were also large relative to previous reports of crayfish found in White Ibis diets in the Everglades.

Diet of Great Cormorant (Phalacrocorax carbo sinensis) nestlings was studied during four consecutive breeding seasons (1999-2002) at three Greek colonies (Axios Delta, and Lakes Kerkini and Mikri Prespa) in order to assess variation and commercial value of prey. A variety of fish taxa were found in nestlings’ regurgitates in each area and season, but only one or two dominated by numbers or biomass. Black Goby (Gobius jozo), Round Sardinella (Sardinella aurita) and Twaite Shad (Alosa fallax) were the most important prey in the Axios Delta; Bleak (Alburnus alburnus), Giebel (Carassius auratus gibelio) and Roach (Rutilus rutilus) at Lake Kerkini; Chalcalburnus belvica and Giebel at Lake Mikri Prespa. Nestling diet varied both seasonally (but only at Lake Kerkini significantly so) and annually (significantly in the Axios Delta and at Lake Mikri Prespa). Temporal changes can be attributed to changes in prey availability and abundance and confirm this bird’s opportunistic behavior. Between-colonies, differences in diet were significant, probably due to differences in habitat and prey species diversity and composition. The low consumption of valuable fish prey by Great Cormorant nestlings (<10%, numbers and biomass) suggests minimal competition with human interests.

Diet of the Great Cormorant (Phalacrocorax carbo) was studied using regurgitated pellets, individual fish bones and fish remains collected from below roosting trees at Želivka and Slapy Reservoirs, Czech Republic, during summer, a warm winter and a cold winter. Both reservoirs have the same trophic status and similar fish assemblages. Using diagnostic bones (os pharyngeum, dentale, praeoperculare) and our own linear regression equations relating dimensions of the diagnostic bone and fish total length (L_T), a total of 2,055 fish of 18 species and four families were identified in the diet of Great Cormorants and their size was reconstructed. Both fish total length and weight differed significantly between seasons being, on average, 12.0 cm and 30 g during summer, 18.3 cm and 109 g during a warm winter and 22.8 cm and 157 g during the cold winter. The average weight of fish taken by Great Cormorants significantly increased with decreasing air and water temperature. The contribution of the dominant “large growing”, torpedo-shaped fish species in the diet of Great Cormorants dramatically increased from summer to the cold winter. In contrast, the contribution of dominant “small growing”, torpedo-shaped species, or humped body-shaped species, showed completely the opposite tendency. Great Cormorants seem to consume all fish of appropriate size that they are able to catch in summer and select for larger fish in winter. Thus, the winter elevation of foraging efficiency described for Great Cormorants in the literature is due to capturing larger fish not to capturing more fish.

Adult Humboldt Penguins (Spheniscus humboldti) are sexually monomorphic in feather coloration, therefore sexing by direct observation is difficult. To determine if subtle morphometric differences exist, we compared gender as determined by DNA analysis with logistic regression leave-one-out analysis of bill length, bill depth and flipper length measurements collected from free-ranging Humboldt Penguins inhabiting a coastal island in Chile. The final chosen model used bill length and bill depth, with no interactions, and correctly predicted the sex of 87.7% of the 756 individuals. Of the 368 females, 13.5% were misclassified as males, while 11.1% of the 388 males were misclassified as females. Including flipper length improved the predictive value of the model slightly. The analysis was not useful in determining the sex of a captive population of Humboldt Penguins of known sex because of excessive beak growth seen in captive penguins.

Common Eider (Somateria mollissima dresseri) breeding success and gull-eider interactions were studied at Stratton Island, Maine in 2004 and 2005. Eiders suffered little nest predation, and most egg losses to gulls were either facilitated by researcher intrusions or confined to newly initiated, unattended nests. Despite high nest success (>80%) in both study years, predation watches indicated that few, if any, ducklings survived to fledging as a result of extreme harassment and predation by Great Black-backed Gulls (Larus marinus). Gull attacks were opportunistic, involved one to 36 gulls, and often resulted in complete crèche destruction. Herring Gulls (L. argentatus) also took occasional young and eggs. Although Stratton Island is managed as a tern restoration site, and gull control measures to enhance tern productivity include nest destruction and shooting of tern predators, gulls continued to congregate around crèching areas and to prey on ducklings. We suggest that additional gull control measures, particularly at a nearby gull colony, may enhance duckling survival. We also recommend monitoring of other eider colonies in the region to better assess duckling survival and recruitment rates.

Wetlands are critical foraging areas for many waterbird species, and their relatively high productivity often determines the coexistence of several species using food resources for breeding and/or refueling during migration. Between 18 October and 20 December 2000, we gathered information on foraging habitat use and feeding techniques of sixteen waterbirds during the austral spring at Malaspina inlet, Chubut, Argentina. We quantified the utilization by waterbirds of ten foraging habitat types and seven feeding techniques on seven fixed survey stations along the coastline, representing the different coastal habitats of the Malaspina inlet. The waterbird assemblage was structured in three main guilds: shallow water frequenters (three cormorants, two terns, a grebe, and two steamer-ducks), hard habitat peckers (two gulls, an egret, and an oystercatcher) and soft habitat frequenters (two oystercatchers, a sandpiper, and a duck). These differed greatly in their use of habitats and feeding techniques, suggesting clear differences in the use of food resources. Most of the groups defined used almost exclusively one feeding technique or habitat type. This fact and the low resource utilization breath values suggest the existence of a relatively high degree of specialization within this waterbird assemblage.

Humans may modify winter habitat of the imperiled Piping Plover (Charadrius melodus), yet published accounts of the species’ winter ecology are rare. We studied Piping Plovers at Oregon Inlet, North Carolina from December 2005 to March 2006. Plovers used a 20.1 km² area (100% minimum convex polygon home range) containing narrow barrier islands with ocean and sound-side beaches, and small shoals, dredged-material islands, and marsh islands in shallow-water sounds. Plover activity was concentrated in twelve areas totaling 2.2 km2 (95% fixed kernel home range). When plovers were on ocean beaches, they spent less time foraging (18%) than when on Sound Island beaches (88%) and islands (83%, P = 0.003). Sound island use increased and beach use decreased as the tide dropped (Logistic regression, P < 0.001). Plover use of dredged-material islands implied that habitat managers can create or restore attractive foraging sites where habitat may be declining or limiting. Wintering habitat management should aim to provide foraging opportunities during most of the day and across a range of tide conditions and ensure that foraging habitat is close to roost sites.

We surveyed a heronry for 16 years to study nest site selection in relation to bird species presence and vegetation structure. Black-crowned Night Heron (Nycticorax nycticorax) was the most abundant species, while Little Egret (Egretta garzetta) and Grey Heron (Ardea cinerea) increased only in the last years. In a first phase, Black-crowned Night Herons selected tall Black Locusts (Robinia pseudoacacia) for nesting. Later they moved into the understory, selecting Hawthorn (Crataegus monogyna) and Black Elderberry (Sambucus nigra), while the higher canopy was used by Grey Herons. The result was a stratification of nests of different species. Considering the succession in use of different forest strata by the medium-sized heron species and years of Grey Heron arrival, it is likely that the forest structure more than competition between species influenced nest site selection in this heronry.

Although the collection of air-dried blood samples on filter paper has been used for field work as a routine, it has required the use of special and expensive types of paper. In the present manuscript the use of common filter paper to collect and store air-dried bird blood samples is validated for its use in DNA based sexing techniques with a very simple method avoiding DNA purification for PCR. This method not only simplifies the laboratory work but also does not add any additional cost for DNA-based sexing techniques. The method was tested and successfully employed for sex determination by molecular techniques on blood samples taken on common filter paper during different field trips from four seabird species.

The Crested Ibis (Nipponia nippon) is one of the most endangered avian species in the world. In 1981, seven ibis were discovered in the Qinling Mountain, central China. The ibis population responded positively to 20 years of intensive conservation growing to more than 500 birds. Population growth of the Crested Ibis exhibited a nonlinear response to increased population size along an altitudinal gradient (600-1,400 m). Prior to 1993, the ibis population was constrained to high mountains with elevations greater than 1,000 m and exhibited density dependent growth in spite of small population size (less than 20 birds). Limited availability of habitats at high elevations might result in competition for food and other resources among the ibis. The ibis returned to traditional nesting habitats at low elevations (<1,000 m) and has grown exponentially since 1992. Fledging success of the ibis did not differ between low- and high-elevation habitats. Rapid population growth of the ibis in low-elevation habitats was mainly due to increased breeding population size and reduced anthropogenic disturbances. Differences in population growth pattern along an altitudinal gradient indicate that high-elevation habitats are suboptimal. Likewise, the recovery of other endangered or threatened species may be constrained by suboptimal habitats. Therefore, caution is needed when interpreting habitat requirements of endangered species based only on presently occupied habitats.

Habitat selection in Black-tailed Gulls (Larus crassirostris) on Hongdo Island, Korea, was studied during the breeding period in 2002-2003. To compare topographical advantages and disadvantages on breeding, we examined two habitats on the island: rocky-cliffs (lower and edge) and grassy (upper and inside) for differences in breeding biology and feeding frequency. In rocky-cliff habitat, Black-tailed Gulls had higher clutch size, faster laying and hatching date, and higher hatching and fledging success. Topographically, rocky-cliff habitat had two advantages-it was closer to sea and difficult to access. Proximity to the sea allowed higher feeding frequency whereas difficult access restricted predators. Conversely, grassy habitat was farther from the sea and allowed easier access to predators, but a lot of grass covered nests to protect eggs and chicks. These disadvantages caused low feeding frequency and higher hatching failure. Egging by fisherman was also a disadvantage of grassy habitat because of easy access from landing places. Therefore, our results suggested that rocky-cliff habitat was more profitable habitat than grassy habitat and this profitability was related to topographical difference between habitats.