Low Prey Abundance Research Articles

Ecology and conservation of a small insular bird population, the Réunion cuckoo-shrike Coracina newtoni

The Réunion cuckoo-shrike Coracina newtoni is a rare endemic forest bird surviving on Réunion island in the Indian Ocean. Its current population is estimated at 120–150 pairs, restricted within a 16-km 2 upland and montane forest area covering a mere 3% of the remaining forest cover of the island. This population has been stable in size and range for at least three decades but it is said to have been more widespread and abundant earlier. Its mountain range is cooler, with higher rainfall and cloud cover and lower prey abundance than the lowlands. Conversely, there are no apparent differences between its forest range and other existing forests, most of them now restricted within the same altitudinal level. It is hypothesized that the species was formerly a mainly lowland forest species which survives in a suboptimal habitat where poor breeding success or survival and lack of more suitable forests prevent it from spreading. Introduced predators, habitat degradation and poaching are additional limiting factors but are unlikely to explain the current status of the Réunion cuckoo-shrike and the difference from that of its Mauritius counterpart Coracina typica. Research guidelines and conservation priorities are suggested, including a feasibility study to establish a second population in the only lowland natural forest patch remaining in Réunion.

Secondary poisoning of stoats (Mustela erminea) at low mouse (Mus musculus) abundance in a New Zealand Nothofagus forest

Two different brodifacoum (Talon 20 P™) poisoning regimes effectively killed 100% of resident radio-tagged stoats (Mustela erminea) by secondary poisoning in a New Zealand Nothofagus forest when mice (Mus musculus) were scarce. Resident possums (Trichosurus vulpecula) and ship rats (Rattus rattus) were also killed. The relative importance of different prey species as sources of poison for stoats has not been clearly identified but availability of poisoned prey will determine the efficacy of secondary poisoning in years of low prey abundance. Tracking tunnels did not accurately measure the decline in the stoat population and were probably influenced by immigrant stoats that were kill-trapped and contained high levels of poison. This study corroborates the findings of several other similar studies that secondary poisoning using brodifacoum effectively kills stoats.

SEABIRD COMMUNITY STRUCTURE ALONG A PRODUCTIVITY GRADIENT: IMPORTANCE OF COMPETITION AND ENERGETIC CONSTRAINT

The oceanic eastern tropical Pacific supports a speciose seabird community that feeds in flocks and depends for food on schools of tunas and dolphins, which force prey to the surface. We analyzed data collected throughout an area of 40 3 10 6 km 2 of open ocean over 10 years (1979-1988) during 1136 at-sea days, from 1750 feeding flocks comprising 51 688 birds of 49 species. A PCA identified three distinct flock types based on species composition: ''Sooty Tern Flocks'' ( n 5 941 flocks) with a large number of Sooty Terns (Sterna fuscata), ''Juan-Wedge Flocks'' ( n 5 402 flocks) with a large number of Juan Fernandez Petrels (Pterodroma externa) and Wedge-tailed Shearwaters (Puffinus pacificus ), and ''Booby Flocks'' ( n 5 407 flocks) with a large number of Red-footed and Masked Boobies (Sula sula and S. dactylatra, respectively). Flock types exhibited largely disjunct distributions that mirrored patterns in thermocline depth and surface water chlo- rophyll content. Sooty Tern Flocks occurred in areas with deepest thermocline (77.6 m) and lowest chlorophyll (0.14 mg/m 3 ), Juan-Wedge Flocks in areas of intermediate ther- mocline (68.8 m) and chlorophyll (0.16 mg/m 3 ), and Booby Flocks in areas of shallowest thermocline (62.5 m) and highest chlorophyll (0.17 mg/m 3 ). These differences were sta- tistically significant. As thermocline depth and surface water chlorophyll are reliable in- dicators of surface water productivity, we conclude that Sooty Tern Flocks foraged in waters of low productivity, Juan-Wedge Flocks in waters of intermediate productivity, and Booby Flocks in waters of highest productivity. Differences in published cost-of-flight values support the hypothesis that energetic con- straint determines flock structure at the low-productivity end of the system. Sooty Terns have the lowest flight costs (4.8 W) and feed in waters of lowest productivity. Flight costs are progressively higher for those species feeding in more productive waters (9.9 W for Wedge-tailed Shearwaters, 19.0 W for Red-footed Boobies). Differences in body size sup- port the hypothesis that interference competition determines flock structure at the high- productivity end of the system. The largest species, Masked Boobies (1987 g) and Red- footed Boobies (1003 g), feed in areas of highest productivity; progressively smaller species, Juan Fernandez Petrels (430 g) and Wedge-tailed Shearwaters (388 g), and Sooty Terns (180 g), feed in areas of decreasing productivity. We hypothesize that this community is largely structured by two factors, competition and energetic constraint, that operate at opposite ends of a productivity gradient. As productivity decreases (low prey abundance), flight proficiency is increasingly important because only species with relatively low flight costs will be able to transit between prey patches. As productivity increases (high prey abundance), competitive ability is increasingly important because competitive dominants will exclude other species from feeding flocks. Our hypothesis is in accordance with em- pirical data and theoretical models designed to explain factors structuring communities for a wide variety of habitats and taxa.

Food resource utilization by juvenile Baltic cod Gadus morhua:a mechanism potentially influencing recruitment success at the demersal juvenile stage?

Pelagic and demersal juvenile Baltic cod Gadus morhua L. were collected during surveys in the Bornholm Basin (Baltic Sea) in autumn 1994. Stomach contents were examined for prey compo- sition in order to evaluate the potential importance of the pelagic and demersal habitats for recruitment success. Juvenile cod less than 40 mm fed exclusively on pelagic prey such as copepods and cladocer- ans. Between 40 and 50 mm the juveniles began to consume benthic prey such as mysids and amphipods; however, copepods were still the dominant food organisms. Between 50 and 70 mm the dominant prey items consumed were mysids and amphipods, with copepods comprising a minor com- ponent of the diet. Between 70 and 160 mm the juveniles' diet was composed exclusively of benthic prey with an increase in prey diversity. Mysids were slightly less important in the diet, while the abun- dance of polychaetes, decapods and fish increased with fish size. Amphipods remained almost constant in the diet of juveniles above 60 mm in length. These results indicate that a major change in food resource utilization started to occur at a fish length of 40 mm, with the main change taking place at approximately 50 mm, suggesting that juvenile Baltic cod make the transition to the benthic habitat at this length. This change in food preference is also reflected in the size of the prey items and the num- bers of prey consumed. Pelagic juveniles consumed much smaller prey than their demersal con- specifics. Neither prey size nor prey numbers consumed by pelagic and small demersal juveniles from this study differed from prey reported to be taken by juvenile cod of corresponding size in the Atlantic and the North Sea. However, prey organisms consumed by demersal juvenile Baltic cod larger than 90 mm were much smaller, and prey numbers much higher, than those reported to be consumed by juveniles in other areas. The apparent overlap in food resource utilization among the different size groups of demersal juveniles observed in this study suggests that in years with low prey abundance, or high abundance of pelagic juveniles, strong intra-specific competition for food resources may affect the recruitment success of Baltic cod.

Relationships beetween the regional distribution of carabid beetles ( Coleoptera, Carabidae) and the abundance of their potential prey

We studied regional variation in abundance of carabid beetles in relation to both abiotic factors and prey availability in 19 forest sites, with a focus on the relationship between the distributions of the carabid beetle Abax ater Villers and its prey, earthworms. No significant correlation was found between the abundance of Abax ater and any of the abiotic factors considered. On the other had, litter thickness was strongly associated with the composition and structure of the whole carabid community: an increase in litter thickness was accompanied by a decrease in both total carabid abundance and carabid species richnes, and by an increase in the proportion of forest species. The relationship between the regional distributions of abundance of carabid beetles and their potential prev was very different for soil-dwelling and litter-dwelling prey. There was a significant pattern of association between low soil prey abundance and low carabid abundance, which suggests that carabid abundance is partly controlled by prey numbers in the soil. On the other hand, there was a significant pattern of association between high prey density per volume of litter and low Abax ater abundance, thus showing a negative relationship between the abundance of this species and prey density in the litter. We propose two complementary hypotheses to explain this pattern: prey depletion by predation in the litter, and biased sampling of carabids due to differential mobility. The prey depletion hypothesis considers that soil acts as a refuge from predation, so that prey numbers in the soil control carabid abundance, while dominant carabids in turn deplete prey and control their abundance in the litter. This hypothesis suggests a complex relationship between the abundances of carabids and their prey.

Control of Feral Cats for Nature Conservation. II. Population Reduction by Poisoning

A feral cat population was substantially reduced by poisoning at a semi-arid site in Western Australia. The control programme was designed to protect two species of endangered native mammals that had recently been reintroduced to the site. Feral cats were poisoned with carcasses of laboratory mice, each impregnated with 4.5 mg of sodium monofluoroacetate (1080). Baits were placed at 100-m intervals along the track system each night for four consecutive nights. Kill rates were assessed by monitoring survival of radio- collared cats and by spotlight counts of cats before and after baiting. All radio-collared cats were killed and there was a 74% reduction in spotlight counts of cats after baiting. Bait removal varied with the abundance of rabbits, the primary prey item for cats in this area. Effectiveness of control operations against feral cats is maximised by baiting at times of low prey abundance. Monitoring the changing abundance of the primary prey species provides important information for timing control operations against feral cats.

Breeding Season Irruptions of Rough-legged Hawks (<i>Buteo Lagopus</i>) on Insular Newfoundland

Rough-legged hawks (Buteo lagopus Pontoppidan), a species typically associated with more northerly regions, occurred in unusually high numbers on insular Newfoundland in the summers of 1988 and 1993, and some pairs nested in noncoastal areas on the island in 1993. We suggest that the island is an alternative breeding ground for the species during years of low prey abundance on the Labrador Peninsula. Our observations concur with hypotheses that rough-legged hawks nomadically search for nesting territories and exploit areas near the limits of their breeding range during such years. Forest clear-cuts were used for foraging by pairs nesting in interior Newfoundland during 1993, which suggests that forest harvesting may have opened new alternative habitats for rough-legged hawks.

A risk assessment for lead in birds

Ecotoxicologists and ecologists have examined the effects of pollutants on individuals and populations largely in terms of one or only a few effects. Yet the recent trend toward a holistic approach to ecological risk assessment suggests that a rigorous paradigm should be applied to toxicants, from hazard identification to risk characterization. Recent discussions have recognized that an up-front problem formulation phase is more critical in ecological risk assessment than it is for human health risk assessment. In this article a modified environmental health risk assessment paradigm is used to examine the risk of lead to birds. This risk analysis is largely conceptual, based on laboratory and field data, and incorporates information currently available. The model expands the hazard identification phase to create a target identification phase that includes the identification of receptors, endpoints, relationships, spatial and temporal scales, and indicators. The target identification phase is unique to the particular hazard, species, population, or community being examined. Lead can cause mortality, or can indirectly affect populations through effects on the food base, avian behavior, reproductive success, and recruitment. Lead can (1) decrease the abundance and availability of prey, (2) bioaccumulate in prey causing increased lead toxicosis in predators, or (3) increase prey availability by interfering with its hiding or escape behavior. Moreover, lower abundance of prey can lead to starvation or nutrient deficiencies, which amplify the absorption and retention of lead. Lead also causes decreases in clutch and egg size, mortality of embryos and nestlings, depression of growth, and deficits in behavior that affect survival. Lead decreases migratory behavior, and increases vulnerability to cold stress, hunters, and other predators. Research needs for evaluating the risk of lead in birds include obtaining data on (1) metal dynamics within various tissues as a function of dose and time since initial exposure, (2) low-level effects on embryos, (3) effects on chicks following fledging and in the period prior to recruitment, (4) effects on adult foraging skills and reproductive behavior, and (5) the relationship between effects from exposure in the laboratory and those from exposure in the wild. This latter point is extremely important, particularly if wild birds have other means of ridding the body of lead not available or less apparent to laboratory birds.

Demography of a high-density coyote population

The demography of a lightly exploited population of coyotes (Canis latrans) in Webb County, Texas, which has had historically high densities, was described from 1976 to 1986. The population fluctuated at high abundance, except for a decline in 1981–1982 following an epizootic of canine parvovirus. Sex ratios were even. Population natality was low, as only 39% of females had viable fetuses. Natality of yearlings (1.5–2.0 years) was lower and more variable than among adults (≥2.5 years). Juvenile (0.8 year) females rarely ovulated. A comparison among 33 radiotelemetered females in 1985 indicated that only territorial adults produced viable fetuses. The proportion of juveniles in spring averaged 0.34. Indices of juvenile survival from birth to the following spring varied from 0.09 to 0.73. Annual survival rates of adults (0.64–0.73) were high and constant. The annual population growth rate (spring to spring) was correlated positively with prey abundance in the preceding winter and negatively with coyote abundance at the beginning of each annual period. Although failure to ovulate and juvenile mortality were identified as key factors in annual variation in population loss, variability in natality among yearlings was identified as the demographic variable associated with relative abundance of coyotes and prey. Body condition of coyotes, as indexed by body mass and internal fat, was not poorer during years of low prey abundance. Interactions between social organization and food availability were implicated in regulation of the lightly exploited high-density population.

Coyote – kit fox interactions as revealed by telemetry

Red foxes (Vulpes vulpes) avoid areas used extensively by coyotes (Canis latrans), apparently to reduce encounters with coyotes and possible injury or death. It is conceivable that kit foxes (Vulpes macrotis) behave in a similar manner. We used telemetry to examine the spacing patterns and interactions among 24 coyotes and 38 San Joaquin kit foxes (V. m. mutica) in California during a period of declining and low prey abundance (1989–1991). Coyotes showed either territorial (n = 13) or transient (n = 11) spacing patterns, whereas all kit foxes were territorial. Spatially and temporally, coyote home ranges overlapped >30% of each fox home range. Both species used the overlap areas more than expected (P < 0.001), possibly because prey such as kangaroo rats were more abundant in these areas. Fifty-five percent of the areas used extensively by foxes during their nocturnal activities were overlapped by coyote nocturnal-use areas. Concurrently monitored coyotes and foxes that occupied overlapping home ranges (n = 25 combinations) were neither farther apart nor closer together than expected (P = 0.09–0.87), even when the coyote was in the fox's home range. Kit foxes did not avoid coyotes and may be able to coexist with them by exploiting certain prey species better than coyotes and maintaining numerous (≥20) dens throughout their home ranges to facilitate escape. Apparent differences in coyote – red fox and coyote – kit fox relations may be partially attributed to differences in resource selection and predator-avoidance strategies among fox species.

Foraging Behaviour of Antarctic Fur Seals During Periods of Contrasting Prey Abundance

1. Foraging behaviour of Antarctic fur seals rearing pups at Bird Island, South Georgia, was assessed using at-sea activity patterns measured by electronic time-depth recorders. Information was obtained for a total of 75 individuals and 191 foraging trips to sea over five reproductive seasons from 1988/89 to 1992/93; this included one season (1990/91) of low prey abundance. A method was developed to divide the diving record up into logical units or bouts which differed from past methods used for defining bouts of behaviour. 2. Foraging trips were significantly longer in 1990/91 than in the other years. There were significant differences between years in the proportion of time spent foraging when at sea and in the distribution of foraging through the day and night. These differences probably represent behavioural responses to changes in prey distribution and abundance and were reflected in the frequency of occurrence of different types of foraging behaviour. 3. Four types of foraging bout were recognized using a cluster analysis. Type I (short) bouts were of short duration (17 min) and occurred mainly during daytime and at dusk. They probably represented exploratory behaviour. Type II (long) bouts occurred mainly at night and were of long duration (80 min). They increased in frequency in 1990/91 when food was scarce and 61-73% of time spent foraging was in these bouts. Type III (shallow) bouts occurred mainly at night, were of short duration (12 min) and represented feeding close to the surface, possibly in association with other, surface-feeding krill predators. Shallow bouts accounted for 8-14% of time spent foraging. Type IV (deep) bouts were of medium duration (19 min) and represented feeding at greater depth (40-50 m) than other bout types. They were most abundant around dawn. 4. Mean dive duration during bouts exceeded the theoretical aerobic dive limit on > 30% of occasions for short, long and deep bouts. There were positive correlations between mean dive duration and surface interval duration for most of these bout types in most years. This suggested that long dives incurred a cost in terms of the amount of time spent at the surface between dives. 5. The study demonstrated that female fur seals invest a significantly greater effort in foraging during periods of low prey abundance by both increasing the time spent foraging and by increasing activity during foraging. This could represent a 30-50% increase in the costs of foraging during years of low food abundance.

Hunting Ranges and Habitat Use and Selection of Urban-Breeding Merlins

We studied hunting ranges and habitat requirements of 27 radio-tagged Merlins (Falco columbarius) breeding in Saskatoon, Saskatchewan from May to July, 1987-1990. Mean hunting range sizes of resident (hatched in the city) and immigrant (hatched outside the city) males were 6.3 _ 1.3 km2 (2.2-13.7 km2) and 33.7 _ 12.1 km2 (12.5-64.3 km2), respectively. Mean hunting range sizes of resident and immigrant females were 6.6 _ 3.4 km2 (2.5-13.4 km2) and 8.6 + 1.6 km2 (0.6-17.5 km2), respectively. Spatial overlap in hunting ranges between neighboring Merlins ranged from 0 to 77.3%. Most immigrants frequently left the city to hunt and had less urban habitats in their ranges than did residents. Merlins that hunted exclusively within the city used habitats in relation to their availability. Resident and immigrant merlins that hunted both within and outside the city avoided hunting in agricultural habitats, which had relatively low prey abundance.

Transtrophic interactions in cassava

Cassava plants exude phloem saps at the base of the petioles of the youngest leaves. The effect of these nutrient-rich droplets on the interaction between the predatory miteTyphlodromalus limonicus and the herbivorous miteMononychellus tanajoa was investigated in a semi-field setting. These two organisms were chosen as a model system due to their strong association with cassava. The hypothesis that exudate production can be considered in terms of extrinsic defense was tested experimentally. Because non-producing clones could not be found exudate production was mimicked (or not) by applying honey droplets on the petioles of plants from which exuding plant parts were aborted. The experiments indicated that: (1) The presence of exudate on otherwise clean plants does not prevent extinction of the predator population, but the rate of population decrease is consistently lower than when no exudate is present. (2) When a second food source enables the predators to reproduce, higher population densities are always attained when the sugar source is also present. (3) Higher predator numbers invariably coincide with lower herbivore abundance. (4) Lower prey abundance does not lead to a reduction in egg production when honey is present. (5) Presence of honey leads to enhanced juvenile/adult survival or reduced emigration and thus to a higher number of female predators.

Ctenophora in the Arctic: the abundance, distribution and predatory impact of the cydippid ctenophore Mertensia ovum (Fabricius) in the Barents Sea

The Ctenophora Mertensia ovum and Beroe cucumis, collected using both conventional sampling gear and scuba divers, were studied in the Barents Sea east of Bjornoya and North Norway in spring 1987 and summer 1988. Among the gelatinous zooplankton, Mertensia ovum was the most consistently abundant copepod predator. Feeding experiments were conducted to evaluate the predation rate of M. ovum in various trophic regimes. This ctenophore can take prey varying in size from small copepods to amphipods and krill, but gut-content analyses from field-collected specimens as well as experimental results showed that the main food source for adults was large-sized copepods (e.g. Calanus finmarchicus, C. glacialis, C. hyperboreus, Metridia longa). The robust tentacle arrray of M. ovum makes this species effective as a predator on large prey. The high potential predation rate of this ctenophore relative to its estimated metabolic cost of only 1.7% of the body energy content d?1 suggests that M. ovum may be able to maintain a positive energy balance even in conditions of low prey abundance. It is suggested that a single exploitation of a zooplankton patch may provide energy for survival for a very long time. The potential impact of M. ovum on Barents Sea copepod populations is estimated on the basis of the minimal observed average daily ration in experiments and from field data on gut contents. Using abundances of copepods for the area, and the actual predator biomass collected, it was estimated that an average of 0.7% of the copepod fauna per day could fall prey to this predator.

Coyote space use in relation to prey abundance

Food abundance is an important factor determining space use in many species, but its effect on carnivore home range and territory size has rarely been investigated. We explored the relationship between food abundance for the coyote (Canis latrans) and space use in two study areas in the northern Great Basin, where the primary prey, the black-tailed jackrabbit (Lepus californicus), fluctuates dramatically in abundance. At one site, home ranges and territories were significantly larger during a time of prey-scarcity than when prey was abundant. Coyotes on the second site had similar-size home ranges and territories at low and high prey abundance, but a higher proportion and probably a higher number of individuals were transients during the prey-scarcity period. We propose mortality rates of coyotes as an important factor mediating adjustments in space use to food abundance, and suggest two mechanisms by which mortality might interact with food abundance. Higher mortality rates may simply permit more rapid adjustment of home range size to changing food conditions. Alternatively, higher mortality may selectively eliminate transients, thus reducing the impact of intruders in limiting the size of the remaining territories.

Ctenophora in the Arctic: the abundance, distribution and predatory impact of the cydippid ctenophore Mertensia ovum (Fabricius) in the Barents Sea

The Ctenophora Mertensia ovum and Beroe cucumis, collected using both conventional sampling gear and scuba divers, were studied in the Barents Sea east of Bjørnøya and North Norway in spring 1987 and summer 1988. Among the gelatinous zooplankton, Mertensia ovum was the most consistently abundant copepod predator. Feeding experiments were conducted to evaluate the predation rate of M. ovum in various trophic regimes. This ctenophore can take prey varying in size from small copepods to amphipods and krill, but gut-content analyses from field-collected specimens as well as experimental results showed that the main food source for adults was large-sized copepods (e.g. Calanus finmarchicus, C. glacialis, C. hyperboreus, Metridia longa). The robust tentacle arrray of M. ovum makes this species effective as a predator on large prey. The high potential predation rate of this ctenophore relative to its estimated metabolic cost of only 1.7% of the body energy content d?1 suggests that M. ovum may be able to maintain a positive energy balance even in conditions of low prey abundance. It is suggested that a single exploitation of a zooplankton patch may provide energy for survival for a very long time. The potential impact of M. ovum on Barents Sea copepod populations is estimated on the basis of the minimal observed average daily ration in experiments and from field data on gut contents. Using abundances of copepods for the area, and the actual predator biomass collected, it was estimated that an average of 0.7% of the copepod fauna per day could fall prey to this predator.

The influence of increasing hunger on the hunting behaviour of Southern Kalahari leopards

The influence of hunger on the hunting behaviour of the leopard in the southern Kalahari Desert is examined, based on ground tracking. It is suggested that increasing hunger affects the various hunting actions of the leopards involved. The leopards are unable to increase the rate of hunting per km travelled, as hunger increases, because of a limited prey base which is already utilized fully. However, by moving increasingly longer distances, they do increase the actual number of hunting actions per 24-h period. The conservation implication is that, based on the energy balance, some arid areas with excessively low prey abundance may be unsuitable for permanent occupation by free-ranging leopards.

Food selection and the timing of breeding of the Blue‐and‐White Swallow Notiochelidon cyanoleuca in Venezuela

The breeding biology of the Blue‐and‐White Swallow in Mérida, Venezuela, was studied in relation to prey availability. Breeding occurred at the end of the dry season and beginning of the wet season. The diet consisted mainly of Hymenoptera but Ephemeroptera, Coleoptera and Diptera were also important prey. Wet or cloudy weather was associated with low feeding rates at the nest and with low prey abundance. The adults spent more time gliding and used more aerial feeding stations in cloudy weather. Food abundance was highest in the wet season. Breeding thus occurs at a time when prey availability is increasing but the wet season, with its high probability of bad weather occurring, tends to be avoided.

Feeding Behavior, Prey Consumption, and Growth of Juvenile Red Hake

Abstract Juvenile red hake Urophycis chuss were collected while they were inhabiting sea scallops Placopecten magellanicus, and the effects of appetite and food availability on prey consumption and growth were examined under laboratory conditions. Fish took increasing amounts of live amphipods Gammarus annulatus with increasing periods of food deprivation up to 40–50 hours. The largest portion of a meal was consumed within the first 3 minutes following prey introduction. With longer periods of food deprivation, red hake increased the time their chemosensitive pelvic fins were extended, indicative of a rise in feeding motivation. The average amount of sand shrimp Crangon septemspinosa consumed by fish was 7.4% of body weight/day. Under conditions of high prey abundance, fish grew at rates equal to field growth rates, but under conditions of low prey abundance, growth rate was inhibited. Prey availability can influence the growth rate, which might lead to size differentials within a cohort. Received Novembe...