The status of the Humboldt Penguin (Spheniscus humboldti) was investigated in Peru after the 1997-98 El Niño event, the strongest of the last century. Penguin numbers along the southern and central coast of Peru (97% of the total) did not differ significantly between 1999 and 2000; the average number was 4,425 individuals. In 1999, the proportion of juveniles (one-year-old birds) was only 0.2% compared with 7% in 2000, probably as a result of the 1997-98 El Niño. Penguins were found from La Foca Island (512’S) to Punta Coles (1742’S). However, the majority (78%) were clustered in five localities, Punta San Juan (36%), San Juanito Islet (11%), Hornillos Island (10%), Pachacamac Island (12%) and Tres Puertas (9%). The size and distribution of penguin colonies have changed over the last 15 years. Penguins have abandoned sites at Punta Corio, Sombrerillo and Morro Sama, and have decreased significantly in numbers in Punta San Fernando and Punta La Chira, where human disturbance has increased, mainly due to local fisheries activities. Penguins have increased at Punta San Juan, San Juanito Islet and San Gallan Island, all of which are partially protected. Half of the penguins were located in guano bird reserves, primarily at Punta San Juan. Guano bird reserves provide some protection against terrestrial predators and human disturbance; however, periodical guano extraction decreases their breeding success. Most penguin sites were found in inaccessible and marginal areas, which were vulnerable to occasional and unpredictable flooding from ocean swells. The methodology recommended by the Population and Habitat Viability Assessment workshop for a consistent census of penguins in Peru and Chile during the molting period was validated at the Punta San Juan Reserve. Continued monitoring of Humboldt Penguin numbers is recommended in order to more fully understand patterns of fluctuation and to be able to detect changes of conservation concern as early as possible. Collaborative efforts between local authorities and conservation biologists are needed to monitor and protect this vulnerable species.

In South Africa’s Western Cape, the main breeding season of Hartlaub’s Gull (Larus hartlaubii) is from January or February until May. At southern localities, there may be a second peak in breeding in August and September. Primary molt of adults mostly starts between August and December and lasts on average 115 days. In each of nine body measurements, averages were higher for males than for females, but values overlapped. In Western Cape during 1990-2001, the species bred at nine localities where breeding had not previously been reported. The maximum number recorded breeding at all localities in this region in any one year was about 5,500 pairs in 1995, less than the 10,000-11,000 pairs estimated for the region in the 1980s. Because the timing of the peak in breeding varies from year to year, it is considered that the 5,500 pairs represent an under-estimate, but it is likely that numbers did decrease after the 1980s. There are substantial inter-annual fluctuations in numbers of birds breeding, suggesting that in some years an appreciable proportion of the adults do not breed.

To boost productivity in the Piping Plovers (Charadrius melodus) breeding in the northern Great Plains, predator exclosure “cages” constructed of wire mesh fence were placed over 1,355 plover nests on alkali lake beaches in Alberta, Saskatchewan, North Dakota, and Montana during 1993-2002. Nesting plovers were killed, apparently by raptors, near cages at 68 (5%) of the nests. In contrast, no losses of adult plovers were detected at 420 nests that were not covered by cages. The predation was greatest (up to 48% of applications) when small (1-1.7 m) diameter cages with wire mesh tops were used at sites with low (mean, 4%) or moderate (15%) tree cover within two km. In areas with low tree cover, predation decreased to 0.7% of applications/year when large (3-4 m) diameter cages with soft netting tops replaced other designs. No predation was recorded in 393 applications of small cages at plover nests along the relatively treeless North Dakota-Montana border. Predator exclosure cages should be used cautiously for protecting eggs of endangered shorebirds. In some situations, enhanced productivity from use of the cages is outweighed by risks to adult birds.

Reproductive success of shorebirds can be improved by placement of predator exclosure fences along beaches or wire-mesh exclosure “cages” over nests. We predicted that these two types of exclosures used simultaneously might further improve reproductive success over that when cages alone are used. Field experiments were carried out on Piping Plovers (Charadrius melodus) on prairie alkali lakes in North Dakota and Montana. During 1996 and 1997, we compared success of nesting plover pairs provided with: (1) no protection, (2) cages that protected eggs in individual nests from both mammalian and avian predators, and (3) a combination of cages plus a temporary electric fence that excluded mammalian predators from the entire nesting beach where chicks were being reared. In 20 replicated trials, fledgling production rates were: no protection, 0.72 chicks/pair (95% CI: 0.29-1.15, N = 43 pairs); cage only, 1.73 (1.30-2.16, N = 46); fence plus cage, 2.06 (1.63-2.49, N = 50). Production by protected pairs was significantly greater than for unprotected pairs. However, no significant difference in production was detected between the two protection types. Temporary electric fences were relatively expensive to apply and added little to the effectiveness of cages, but may be appropriate in situations where cages cannot be used or where mammalian predation on chicks is a greater threat.

Acts of aggression by Roseate Tern (Sterna dougallii) adults on conspecific chicks were examined on Aride Island, Seychelles, from 13 June-10 July 1998. Data were used to examine two hypotheses to explain adult aggression towards chicks: (1) the adoption avoidance hypothesis, and (2) the feeding unfamiliar chicks hypothesis. Throughout the entire nestling development, adults pecked mostly unattended chicks (96%). The intensity of pecking was higher on b-chicks than on a-chicks and 72% of the pecking acts were on chicks that were on their own nests. During each pecking act, young chicks (0-6 days) were pecked by a greater number of adults than were old chicks (>6 days). When pecked, young chicks moved significantly greater distances away from nests than did old chicks (1.41 ± 1.92 m vs. 0.18 ± 0.42 m). Of the 23 chicks killed by adults, 11 were a-chicks, representing 8.5% of the a–chicks in the study area. Parents sometimes rescued a-chicks, but did not attempt to rescue b-chicks from adult attacks. Chicks that died had moved significantly greater distances away from nests than chicks that survived (2.8 ± 2.5 m vs. 0.3 ± 0.67 m). This study presents evidence that both hypotheses may be important in explaining intraspecific adult aggression by Roseate Terns on chicks. In particular, the feeding unfamiliar chicks hypothesis may be invoked to explain the significant negative correlation between daily intensity of adult aggression and the amount of food delivered to chicks. The adoption avoidance hypothesis may explain why b-chicks were disproportionately involved in pecking in relation to a-chicks. Adult aggression is an additive factor of chick mortality, operating mostly on days of food shortage.

The White-rumped Sandpiper (Calidris fuscicollis) breeds in Alaska and Canada, and during the nonbreeding season migrates to South America. The reported diet of this sandpiper is principally invertebrates; seeds are a very rare item. This species was studied at Guaminí Pond, Buenos Aires Province, Argentina, in December 1997. The stomach contents of 23 adults were investigated. The analysis showed that during the sampling period the only food eaten by the White-rumped Sandpiper were seeds. Them were represented by the following families: Caryophylaceae, Chenopodiaceae, Fabaceae, Malvaceae, Polygonaceae, Ulmaceae, Cyperaceae and Poaceae. The seeds ranged between 0.83 and 2.25 mm in length. Stones and fragments of algae were found in all samples, the algae possibly ingested accidentally during feeding. Our findings indicate that the White-rumped Sandpiper forages on different items depending on the area, season and availability, and may be an opportunistic feeder.

Over the past 70 years, three counts have been conducted on the Rockhopper Penguin (Eudyptes chrysocome) population on the Falkland Islands during the 1932/33, 1995/96 and 2000/01 breeding seasons. The results indicated a population decrease of more than 90% during this period, from more than three million breeding pairs in 1932/33 to less than 300,000 breeding pairs in the mid-1990s. However, a re-evaluation of these data revealed that the original population was substantially overestimated and the 1930s numbers were probably closer to 1.5 million breeding pairs. Modifications to the mid-1990s data produce a revised population estimate of about 263,000, rather than 297,000 breeding pairs. Based on these revised values, the overall decrease in the Rockhopper Penguin at the Falkland Islands between 1932 and 1995 still exceeded 80%, at a rate of ca. 2.75% per annum. In the most recent census, the population was estimated to be 272,000 breeding pairs, suggesting a stable population since the mid 1990s. These re-calculations of historical Rockhopper Penguin population trends in the Falkland Islands have important implications for the assessment of the global population size and long-term trends of this species.

We studied the habitat requirements and nest site characteristics of Imperial Cormorants (Phalacrocorax atriceps) and Rock Shags (P. magellanicus) in northern Golfo San Jorge, Argentina. Colonies of both species were only located on islands. We characterized nesting habitat at 15 and 31 islands with and without Imperial Cormorant colonies, and 19 and 27 islands with and without Rock Shag colonies, respectively. The probability of finding nesting cormorants and shags was independent of island size. Both cormorants and shags nested on only 18% of islands or islets closer than one km to the mainland, but nested in between 35% and 52% of islands located further offshore. All Imperial Cormorant and Rock Shag colonies were located on bare, rocky substrate. All colonies of both species were located more than 50 m from shrubs. The mean slope of the substrate where colonies were located was significantly greater for Rock Shags than for Imperial Cormorants (73° vs. 3°). Orientation of cliffs and steep slopes where Rock Shags nested was mostly to the east. Over 86% of Imperial Cormorant colonies were located on the side of the island facing the open sea. Both height and depth of the nest were significantly larger for the Rock Shag, while internal diameter was significantly larger for Imperial Cormorant nests. The distribution patterns of cormorants and shags at the northern sector of Golfo San Jorge appears to be largely determined by the presence of islands, with availability of open rocky substrate and low shrub vegetation cover.

Although there have been several studies of the foraging behavior of the Imperial Cormorant (Phalacrocorax atriceps) and the Rock Shag (Phalacrocorax magellanicus), none document the feeding performance and the use of feeding areas by these birds breeding in mixed or adjacent colonies. We studied birds nesting sympatrically in two colonies at Malaspina Inlet, Patagonia, Argentina. Both colonies were located on islands separated by 2.2 km: Vernacci Oeste, inside Malaspina Inlet and Vernacci Este at the mouth of the inlet. Rock Shags from both islands and Imperial Cormorants from Vernacci Oeste fed inside the inlet, and showed similar dive duration, surface interval and foraging range. Imperial Cormorants from Vernacci Este foraged outside the inlet and showed longer dive duration, surface interval and foraging range. Imperial Cormorants made longer foraging trips than Rock Shags (4.9 ± 1.9 vs. 1.9 ± 0.7 h) regardless of its colony. There was no overlap between the foraging areas used by Imperial Cormorants from the two colonies. Despite Rock Shags from both islands feeding inside the inlet, there was a little overlap in their foraging locations (3-22%). Rock Shags and Imperial Cormorants breeding in the same colony showed an overlap in their foraging areas, being lower at Vernacci Este than at Vernacci Oeste. Although both species are usually described as having different diving capacities, we found that their diving behavior was similar when feeding in areas of similar environmental conditions.

Nocturnal behavior of 113 American Avocets (Recurvirostra americana) was observed in playa wetlands on the Southern High Plains of Texas, USA, during April and May 2000. Nocturnal time activity budgets consisted primarily of foraging (62%) and resting (20%). Nocturnal time activity budgets were compared to diurnal time activity budgets. Differences in behavior existed between diurnal and nocturnal periods. Greater percentages of time were devoted to foraging and aggression during the night, but less time was spent resting during the night. Time spent in locomotion, maintenance, and alert behavior was comparable among diurnal and nocturnal periods.

Double-crested Cormorant (Phalacrocorax auritus) numbers are increasing throughout eastern North America. We compared variation for five portions of mtDNA to determine if genetic differences existed among portions of the breeding range that would need to be considered when formulating management programs. Sequences for four mtDNA regions were identical across sample locations; frequencies of two haplotypes of the mitochondrial Control Region were similar across sampling locations. There is no evidence of restricted gene flow among breeding areas, or between subspecies with different migratory patterns.

Food intake (fish biomass), during the winters of 1997-98 and 1998-99, by Great Crested Grebes (Podiceps cristatus) on a 47 km² area of the western basin of Lake Como, North Italy, was estimated by analyzing recognizable remains of the fish prey in the stomach contents of birds found dead. Using the size of otoliths and pharyngeal bones, we showed that grebes mainly ate the younger age-classes (2-yr. old and younger) of Bleak (Alburnus alburnus), 80% of all fish prey. Monthly rates of Bleak consumption by wintering grebes were estimated by multiplying daily food intake, extrapolated from the relationship between food biomass intake and water temperature in the different winter months, with the proportion of Bleak in the diet and with the number of wintering grebes present each month. Great Crested Grebes were most numerous in the study area in January-February and there were no significant differences in numbers between the two winters. Size of Bleak eaten increased from November to February and the biomass of Bleak consumed was highest at the end of winter, when water temperature was lowest. Overall in the western basin, grebes consumed between 640 kg (conservative estimate) and 1,000 kg (maximum estimate) of Bleak per winter, which corresponded to 10% to 20% of the annual commercial Bleak harvest in that area. We suggest that grebe predation on Bleak might increase the mortality rates in the younger age-classes of the fish, and thus contribute to a decrease in the stock, but that economic impact of grebe predation on gross annual income from the local fishery is negligible.

Wetland use was studied in the Pied-billed Grebe (Podilymbus podiceps) and Horned Grebe (Podiceps auritus) in southwestern Manitoba during April to August of 1997. Of the 180 wetlands surveyed, 74 had no grebe use, six were sporadically used by either grebe species, 19 were consistently used by Horned Grebe, nine switched from Horned Grebe to Pied-billed Grebe use, none switched from Pied-billed Grebe to Horned Grebe use, 68 were consistently used by Pied-billed Grebe, one was regularly used by both species but at different times, and three were concurrently used by both species throughout the breeding season. Wetlands used by grebes were larger, deeper, and had more vegetated area than wetlands not used by grebes, wetlands where occupancy changed were larger, deeper, and had more vegetated area than wetlands consistently used by one species, and wetlands that were concurrently used by both species were larger, deeper, and had more vegetated area than wetlands used by only one species at a time. Five logistic regression models were used to model the probability of observing a grebe brood as a function of the habitat characteristics of the wetland, grebe species, and their interaction. The three best models all contained a parameter describing the habitat characteristics of the wetlands. In two of the three best models, the probability of observing a grebe brood increased with increased wetland size, depth, and vegetated area. These results indicate that habitat characteristics influence breeding quality of a wetland and that grebes may be competing for high quality wetlands in this study area.

We studied the basic biology of the Red Fox (Vulpes vulpes) and its interactions with the waterbird colonies in the Ebro Delta natural park, Spain. Dens (earths) were found in areas characterized by sandy dune vegetation, a few buildings, areas where human entry was prohibited, and areas where hunting of waterfowl was banned. Regression associated the presence of dens to the abundance of dunes, halophytic vegetation and prohibited entry. Birds constituted 96% of the prey found at dens, mainly waterfowl (68%), gulls (16%) and rails (10%), but this differed markedly between areas. During the breeding season, foxes obtained their food from a reduced number of sources in the area surrounding their dens, and impacted the nearest breeding birds. We propose changes in the management of foxes in the natural park to safeguard internationally threatened bird species and to prevent them from decreasing, or even disappearing, as a result of fox predation.

Urban areas are increasingly occupying the margins of estuarine wetlands. Consequently, it becomes important to understand how the presence of these urban areas influences the use of adjacent intertidal habitats by waterbirds. Wintering waterbirds were regularly counted on 28 mudflat sectors along the northern margin of the river Tagus estuary, Portugal, which includes large urbanized extensions. Sectors were characterized using 16 variables related to its physical characteristics and human influence and related to bird density and diversity using multiple linear regression. Species richness was positively influenced by wider mudflats and a margin covered with saltmarsh. The individual response of seven of the most abundant bird species to these variables differed between species, but the abundance of half of these species was significantly influenced by variables potentially related to the presence of urban areas.

We examined the relationship between number of wetlands and occurrence of five waterbird and one waterfowl species in the Prairie Pothole Region of North Dakota, USA, from 1980-2000. Data from 13 Breeding Bird Survey routes provided an index to regional density and distribution of Pied-billed Grebe (Podilymbus podiceps), Black Tern (Chlidonias niger), American Bittern (Botaurus lentiginosus), Northern Pintail (Anas acuta), Sora (Porzana carolina), and American Coot (Fulica americana), while 69 segments from annual Waterfowl Breeding Ground Population and Habitat Surveys provided an index to regional wetland availability. Numbers of wetlands and birds varied among years, and density and distribution of all six species showed a strong positive correlation with number of wetlands. Correlations were weaker when the number of wetlands was lagged one year, suggesting that waterbird distributions shift in response to water availability rather than respond locally. Spatial and temporal variation of waterbird habitat and numbers should be considered in monitoring and management of waterbirds in the Prairie Pothole Region.

Wading bird nesting colonies were surveyed in the Upper St. Johns River Basin, east central Florida, USA in 1993-1995 and 1998-2000 using aerial survey methods. A total of 62 colony locations were found over six years, with a maximum of 35 sites active in each of two years. Borrow pits and managed impoundments were the most important nesting locations based on size and persistence. Most of these sites were in or adjacent to the Upper St. Johns River Basin Project, a wetland restoration project. Higher numbers of nests were counted during nesting seasons preceded by above average rainfall than during seasons characterized by drought. Cattle Egrets (Bubulcus ibis) were the most common nesting species in all years, however, the proportion of the total nests that were Cattle Egrets decreased over the study period. Wood Storks (Mycteria americana), a federally endangered species, nested in increasing numbers within three borrow pits adjacent to the Upper St. Johns River Basin Project. This study reveals the importance of borrow pits, most of which are on private land where sites are unprotected, to wading bird nesting in east central Florida.