Quantifying the global threat to native birds from predation by non-native birds on small islands

Although invasive non-native species can adversely affect biodiversity in many ways, predation of native species by non-native species on islands can be severely damaging. Results of numerous studies document non-native birds preying on birds on islands, but our understanding of the number and type of species affected has been limited by the lack of a global review of these impacts. I identified the non-native bird species that have been recorded preying on birds, the locations where this predation occurred, and the bird species affected. Because the impacts of non-native birds can be particularly severe on small islands, I then identified the islands <500 km² around the world that are occupied by predatory non-native birds. By taking into account their life-history traits and predation history, I also identified the near-threatened and threatened bird species on these islands that they may prey on. The results indicated that predation by non-native birds was primarily a concern for threatened bird conservation on small islands; almost all predation impacts (91%) on near-threatened and threatened birds were recorded on islands, and median island size was 106 km². I also found non-native bird predation was a poorly known and widespread potential threat to avian biodiversity; worldwide, 194 islands of <500 km² were occupied by predatory non-native birds, but information on their impacts was unavailable for most of these islands. On them, where the impacts of non-native species can be severe, non-native birds may be preying on approximately 6% of the world's near-threatened and threatened bird species. Four non-native bird species I identified have been successfully eradicated from islands. If they were eradicated from the small islands they occupy, 70% of the near-threatened and threatened bird species I identified would no longer be affected by nest predation by non-native birds on small islands.     

The Value of Primary, Secondary, and Plantation Forests for a Neotropical Herpetofauna

Abstract:  Plantation forests and second-growth forests are becoming dominant components of many tropical forest landscapes. Yet there is little information available concerning the consequences of different forestry options for biodiversity conservation in the tropics. We sampled the leaf-litter herpetofauna of primary, secondary, and Eucalyptus plantation forests in the Jari River area of northeastern Brazilian Amazonia. We used four complementary sampling techniques, combined samples from 2 consecutive years, and collected 1739 leaf-litter amphibians (23 species) and 1937 lizards (30 species). We analyzed the data for differences among forest types regarding patterns of alpha and beta diversity, species-abundance distributions, and community structure. Primary rainforest harbored significantly more species, but supported a similar abundance of amphibians and lizards compared with adjacent areas of second-growth forest or plantations. Plantation forests were dominated by wide-ranging habitat generalists. Secondary forest faunas contained a number of species characteristic of primary forest habitat. Amphibian communities in secondary forests and Eucalyptus plantations formed a nested subset of primary forest species, whereas the species composition of the lizard community in plantations was distinct, and was dominated by open-area species. Although plantation forests are relatively impoverished, naturally regenerating forests can help mitigate some negative effects of deforestation for herpetofauna. Nevertheless, secondary forest does not provide a substitute for primary forest, and in the absence of further evidence from older successional stands, we caution against the optimistic claim that natural forest regeneration in abandoned lands will provide refuge for the many species that are currently threatened by deforestation .     

Earthworm Invasion as the Driving Force Behind Plant Invasion and Community Change in Northeastern North American Forests

Abstract:  Identification of factors that drive changes in plant community structure and contribute to decline and endangerment of native plant species is essential to the development of appropriate management strategies. Introduced species are assumed to be driving causes of shifts in native plant communities, but unequivocal evidence supporting this view is frequently lacking. We measured native vegetation, non-native earthworm biomass, and leaf-litter volume in 15 forests in the presence and absence of 3 non-native plant species ( Microstegium vimineum, Alliaria petiolata, Berberis thunbergii ) to assess the general impact of non-native plant and earthworm invasions on native plant communities in northeastern United States. Non-native plant cover was positively correlated with total native plant cover and non-native earthworm biomass. Earthworm biomass was negatively associated with cover of native woody and most herbaceous plants and with litter volume. Graminoid cover was positively associated with non-native earthworm biomass and non-native plant cover. These earthworm-associated responses were detected at all sites despite differences in earthworm species and abundance, composition of the native plant community, identity of invasive plant species, and geographic region. These patterns suggest earthworm invasion, rather than non-native plant invasion, is the driving force behind changes in forest plant communities in northeastern North America, including declines in native plant species, and earthworm invasions appear to facilitate plant invasions in these forests. Thus, a focus on management of invasive plant species may be insufficient to protect northeastern forest understory species.     

Responses of Tuatara (Sphenodon punctatus) to Removal of Introduced Pacific Rats from Islands

Abstract:  Invasive mammalian predators such as rats are now widespread on islands, but hypotheses about their effects have rarely been tested. Circumstantial evidence from New Zealand indicates that, when introduced to islands, Pacific rats ( Rattus exulans ) have negative effects on endemic plants, invertebrates, birds, and reptiles, including the tuatara ( Sphenodon punctatus ). We tested the effects of Pacific rats on tuatara by comparing the demographic structure and body condition of tuatara populations on three islands before and after removal of rats and on a fourth island where rats remained. In the presence of rats, juvenile tuatara constituted on average 0–5% of the sample tuatara populations. When Pacific rats were removed after at least 200 years' occupancy, the proportion of juvenile tuatara increased 3.5- to 17-fold and body condition of adult males and females also improved (sometimes dramatically). We predict that, unless Pacific rats are removed from Taranga Island, the tuatara population will collapse because of low population density and the lack of juvenile recruitment. Our results demonstrate that when invasive species exert subtle effects on recruitment and body condition, the effects on populations of long-lived endemic species may only become apparent long after the invasion.     

The Biogeography of Globally Threatened Seabirds and Island Conservation Opportunities

Seabirds are the most threatened group of marine animals; 29% of species are at some risk of extinction. Significant threats to seabirds occur on islands where they breed, but in many cases, effective island conservation can mitigate these threats. To guide island‐based seabird conservation actions, we identified all islands with extant or extirpated populations of the 98 globally threatened seabird species, as recognized on the International Union for Conservation of Nature Red List, and quantified the presence of threatening invasive species, protected areas, and human populations. We matched these results with island attributes to highlight feasible island conservation opportunities. We identified 1362 threatened breeding seabird populations on 968 islands. On 803 (83%) of these islands, we identified threatening invasive species (20%), incomplete protected area coverage (23%), or both (40%). Most islands with threatened seabirds are amenable to island‐wide conservation action because they are small (57% were <1 km²), uninhabited (74%), and occur in high‐ or middle‐income countries (96%). Collectively these attributes make islands with threatened seabirds a rare opportunity for effective conservation at scale. La Biogeografía de Aves Marinas Amenazadas Globalmente y las Oportunidades de Conservación en Islas     

Effects of Forest Roads on Macroinvertebrate Soil Fauna of the Southern Appalachian Mountains

Abstract: Many forested landscapes are fragmented by roads, but our understanding of the effects of these roads on the function and diversity of the surrounding forest is in its infancy. I investigated the effect of roads in otherwise continuous forests on the macroinvertebrate fauna of the soil. I took soil samples along transects leading away from the edges of unpaved roads in the Cherokee National Forest in the Southern Appalachian mountains of the United States. Roads significantly depressed both the abundance and the richness of the macroinvertebrate soil fauna. Roads also significantly reduced the depth of the leaf-litter layer. These effects persisted up to 100 m into the forest. Wider roads and roads with more open canopies tended to produce steeper declines in abundance, richness, and leaf-litter depth, but these effects were significant only for canopy cover and litter depth. The macroinvertebrate fauna of the leaf litter plays a pivotal role in the ability of the soil to process energy and nutrients. These macroinvertebrates also provide prey for vertebrate species such as salamanders and ground-foraging birds. The effect of roads on the surrounding forest is compounded by the sprawling nature of the road system in this and many other forests. My data suggest that even relatively narrow roads through forests can produce marked edge effects that may have negative consequences for the function and diversity of the forest ecosystem.     

A Natural Experiment on the Impact of Overabundant Deer on Forest Invertebrates

Abstract:  In large parts of North America and Europe, deer overabundance threatens forest plant diversity. Few researchers have examined its effects on invertebrate assemblages. In a natural experiment on Haida Gwaii (British Columbia, Canada), where Sitka black-tailed deer ( Odocoileus hemionus sitkensis ) were introduced, we compared islands with no deer, with deer for fewer than 20 years, and with deer for more than 50 years. We sampled invertebrates in three habitat categories: forest edge vegetation below the browse line, forest interior vegetation below the browse line, and forest interior litter. In forest edge vegetation, invertebrate abundance and species density decreased with increasing length of browsing history. In forest interior vegetation, decrease was significant only on islands with more than 50 years of browsing. Insect abundance in the vegetation decreased eightfold and species density sixfold on islands browsed for more than 50 years compared with islands without deer. Primary consumers were most affected. Invertebrates from the litter showed little or no variation related to browsing history. We attributed the difference between vegetation-dwelling and litter-dwelling invertebrates to differences in the effect of browsing on their habitat. In the layer below the browse line deer progressively removed the habitat. The extent of litter habitat was not affected, but its quality changed. We recommend more attention be given to the effect of overabundant ungulates on forest invertebrate conservation with a focus on edge and understory vegetation in addition to litter habitat.     

Effects of Dam‐Induced Landscape Fragmentation on Amazonian Ant–Plant Mutualistic Networks

Mutualistic networks are critical to biological diversity maintenance; however, their structures and functionality may be threatened by a swiftly changing world. In the Amazon, the increasing number of dams poses a large threat to biological diversity because they greatly alter and fragment the surrounding landscape. Tight coevolutionary interactions typical of tropical forests, such as the ant–myrmecophyte mutualism, where the myrmecophyte plants provide domatia nesting space to their symbiotic ants, may be jeopardized by the landscape changes caused by dams. We analyzed 31 ant–myrmecophyte mutualistic networks in undisturbed and disturbed sites surrounding Balbina, the largest Central Amazonian dam. We tested how ant–myrmecophyte networks differ among dam‐induced islands, lake edges, and undisturbed forests in terms of species richness, composition, structure, and robustness (number of species remaining in the network after partner extinctions). We also tested how landscape configuration in terms of area, isolation, shape, and neighborhood alters the structure of the ant–myrmecophyte networks on islands. Ant–myrmecophytic networks were highly compartmentalized in undisturbed forests, and the compartments had few strongly connected mutualistic partners. In contrast, networks at lake edges and on islands were not compartmentalized and were negatively affected by island area and isolation in terms of species richness, density, and composition. Habitat loss and fragmentation led to coextinction cascades that contributed to the elimination of entire ant–plant compartments. Furthermore, many myrmecophytic plants in disturbed sites lost their mutualistic ant partners or were colonized by opportunistic, nonspecialized ants. Robustness of ant–myrmecophyte networks on islands was lower than robustness near lake edges and in undisturbed forest and was particularly susceptible to the extinction of plants. Beyond the immediate habitat loss caused by the building of large dams in Amazonia, persistent edge effects and habitat fragmentation associated with dams had large negative effects on animal–plant mutualistic networks. Efectos de la Fragmentación del Paisaje Inducida por Presas sobre Redes Mutualistas Hormiga‐Planta Amazónicas     

Edge effects on trophic cascades in tropical rainforests

The cascading effects of biodiversity loss on ecosystem functioning of forests have become more apparent. However, how edge effects shape these processes has yet to be established. We assessed how edge effects alter arthropod populations and the strength of any resultant trophic cascades on herbivory rate in tropical forests of Brazil. We established 7 paired forest edge and interior sites. Each site had a vertebrate-exclosure, procedural (exclosure framework with open walls), and control plot (total 42 plots). Forest patches were surrounded by pasture. Understory arthropods and leaf damage were sampled every 4 weeks for 11 months. We used path analysis to determine the strength of trophic cascades in the interior and edge sites. In forest interior exclosures, abundance of predaceous and herbivorous arthropods increased by 326% and 180%, respectively, compared with control plots, and there were significant cascading effects on herbivory. Edge-dwelling invertebrates responded weakly to exclusion and there was no evidence of trophic cascade. Our results suggest that the vertebrate community at forest edges controls invertebrate densities to a lesser extent than it does in the interior. Edge areas can support vertebrate communities with a smaller contingent of insectivores. This allows arthropods to flourish and indirectly accounts for higher levels of plant damage at these sites. Increased herbivory rates may have important consequences for floristic community composition and primary productivity, as well as cascading effects on nutrient cycling. By interspersing natural forest patches with agroforests, instead of pasture, abiotic edge effects can be softened and prevented from penetrating deep into the forest. This would ensure a greater proportion of forest remains habitable for sensitive species and could help retain ecosystem functions in edge zones.     

Effects of Forest Fragmentation and Habitat Degradation on West African Leaf-Litter Frogs

Abstract:  Habitat degradation alters the dynamics and composition of anuran assemblages in tropical forests. The effects of forest fragmentation on the composition of anuran assemblages are so far poorly known. We studied the joint influence of forest fragmentation and degradation on leaf-litter frogs. We specifically asked whether the processes structuring leaf-litter anuran assemblages in fragmented forests are the same as those in continuous forests. We analyzed anuran assemblages with respect to habitat characteristics, including fragmentation and degradation parameters. In comparison with continuous forests, species richness and diversity were lower and assemblage composition was altered in forest fragments. These changes seemed to be mainly caused by habitat degradation rather than forest fragmentation. Availability of aquatic sites for breeding, vegetation structure (including those variables indicating degradation), and leaf-litter cover had the most influence on the presence of single species. The comparatively small impact of fragmentation on anurans might be due to the location of the study area; it still possessed large tracts of continuous forest. These forest blocks may stabilize the regional rainforest climate and thus weaken the effects of fragmentation .     

Evolution of Nesting Height in an Endangered Hawaiian Forest Bird in Response to a Non‐Native Predator

Abstract: The majority of bird extinctions since 1800 have occurred on islands, and non‐native predators have been the greatest threat to the persistence of island birds. Island endemic species often lack life‐history traits and behaviors that reduce the probability of predation and they can become evolutionarily trapped if they are unable to adapt, but few studies have examined the ability of island species to respond to novel predators. The greatest threat to the persistence of the Oahu Elepaio (Chasiempis ibidis), an endangered Hawaiian forest bird, is nest predation by non‐native black rats (Rattus rattus). I examined whether Oahu Elepaio nest placement has changed at the individual and population levels in response to rat predation by measuring nest height and determining whether each nest produced offspring from 1996 to 2011. Average height of Oahu Elepaio nests increased 50% over this 16‐year period, from 7.9 m (SE 1.7) to 12.0 m (SE 1.1). There was no net change in height of sequential nests made by individual birds, which means individual elepaios have not learned to place nests higher. Nests ≤3 m off the ground produced offspring less often, and the proportion of such nests declined over time, which suggests that nest‐building behavior has evolved through natural selection by predation. Nest success increased over time, which may increase the probability of long‐term persistence of the species. Rat control may facilitate the evolution of nesting height by slowing the rate of population decline and providing time for this adaptive response to spread through the population.     

Resilience of Southwestern Amazon Forests to Anthropogenic Edge Effects

Abstract:  Anthropogenic edge effects can compromise the conservation value of mature tropical forests. To date most edge-effect research in Amazonia has concentrated on forests in relatively seasonal locations or with poor soils in the east of the basin. We present the first evaluation from the relatively richer soils of far western Amazonia on the extent to which mature forest biomass, diversity, and composition are affected by edges. In a southwestern Amazonian landscape we surveyed woody plant diversity, species composition, and biomass in 88 × 0.1 ha samples of unflooded forest that spanned a wide range in soil properties and included samples as close as 50 m and as distant as >10 km from anthropogenic edges. We applied Mantel tests, multiple regression on distance matrices, and other multivariate techniques to identify anthropogenic effects before and after accounting for soil factors and spatial autocorrelation. The distance to the nearest edge, access point, and the geographical center of the nearest community ("anthropogenic-distance effects") all had no detectable effect on tree biomass or species diversity. Anthropogenic-distance effects on tree species composition were also below the limits of detection and were negligible in comparison with natural environmental and spatial factors. Analysis of the data set's capacity to detect anthropogenic effects confirmed that the forests were not severely affected by edges, although because our study had few plots within 100 m of forest edges, our confidence in patterns in the immediate vicinity of edges is limited. It therefore appears that the conservation value of most "edge" forests in this region has not yet been compromised substantially. We caution that because this is one case study it should not be overinterpreted, but one explanation for our findings may be that western Amazonian tree species are naturally faster growing and more disturbance adapted than those farther east.     

Determinants of population persistence and abundance of terrestrial and arboreal vertebrates stranded in tropical forest land-bridge islands

Megadams are among the key modern drivers of habitat and biodiversity loss in emerging economies. The Balbina Hydroelectric Dam of Central Brazilian Amazonia inundated 312,900 ha of primary forests and created approximately 3500 variable-sized islands that still harbor vertebrate populations after nearly 3 decades after isolation. We estimated the species richness, abundance, biomass, composition, and group size of medium- to large-bodied forest vertebrates in response to patch, landscape, and habitat-quality metrics across 37 islands and 3 continuous forest sites throughout the Balbina archipelago. We conducted 1168 km of diurnal censuses and had 12,420 camera-trapping days along 81 transects with 207 camera stations. We determined the number of individuals (or groups) detected per 10 km walked and the number of independent photographs per 10 camera-trapping days, respectively, for each species. We recorded 34 species, and patch area was the most significant predictor of vertebrate population relative abundance and aggregate biomass. The maximum group size of several group-living species was consistently larger on large islands and in continuous patches than on small islands. Most vertebrate populations were extirpated after inundation. Remaining populations are unlikely to survive further ecological disruptions. If all vertebrate species were once widely distributed before inundation, we estimated that approximately 75% of all individual vertebrates were lost from all 3546 islands and 7.4% of the animals in all persisting insular populations are highly likely to be extirpated. Our results demonstrate that population abundance estimates should be factored into predictions of community disassembly on small islands to robustly predict biodiversity outcomes. Given the rapidly escalating hydropower infrastructure projects in developing counties, we suggest that faunal abundance and biomass estimates be considered in environmental impact assessments and large strictly protected reserves be established to minimize detrimental effects of dams on biodiversity. Conserving large tracts of continuous forests represents the most critical conservation measure to ensure that animal populations can persist at natural densities in Amazonian forests.     

Effects of Forest Fragmentation on a Dung Beetle Community in French Guiana

Abstract:  Fragmentation is the most common disturbance induced by humans in tropical forests. Some insect groups are particularly suitable for studying the effects of fragmentation on animal communities because they are taxonomically and ecologically homogenous. We investigated the effects of forest fragmentation on a dung beetle species community in the forest archipelago created in 1994–1995 by the dam of Petit Saut, French Guiana. We set and baited an equal number of pitfall traps for dung beetles on three mainland sites and seven island sites. The sites ranged from 1.1 to 38 ha. In 250 trap days, we captured 50 species in 19 genera. Diversity indices were high (2.18–4.06). The lowest diversity was on the small islands and one mainland site. Species richness and abundance were positively related to fragment area but not to distance from mainland or distance to the larger island. The islands had lower species richness and population than mainland forest, but rarefied species richness was relatively invariant across sites. There was a marked change in species composition with decreasing fragment that was not caused by the presence of a common fauna of disturbed-area species on islands. Small islands differed from larger islands, which did not differ significantly from mainland sites. Partial correlation analyses suggested that species richness and abundance of dung beetle species were positively related to the number of species of nonflying mammals and the density index of howler monkeys ( Alouatta seniculus ), two parameters positively related to fragment area.     

Nesting Success of a Neotropical Migrant in a Multiple-Use, Forested Landscape

Abstract: We studied the nesting success of an individually marked population of Kentucky Warblers ( Oporornis formosus ), a species that nests in disturbed and undisturbed forests, in a heterogeneous, managed forest site in the Shawnee National Forest in southern Illinois from 1992 to 1995. We examined the effects of forest stand type (clearcuts of various ages, tree plantations, and older forest) and distance from habitat edges on rates of nest predation and brood parasitism by Brown-headed Cowbirds (   Molothrus ater ). Brood parasitism levels gradually decreased from 60% to 3% ( nests) over a distance of 2 km from an agricultural edge proximal to a known cowbird foraging site (a pig feedlot), but they did not vary with distance from any other kinds of edges or with forest stand type. Rates of nest predation ( nests) did not vary with distance from any edges, but they were significantly lower in older forest than within even-aged clearcuts, a tree plantation, and in successional vegetation adjacent to a residential facility. These results suggest that, even in fragmented landscapes with high overall levels of parasitism and nest predation, management practices within and immediately adjacent to forest tracts can affect the nesting success of some species, but not necessarily as a simple function of distance from edge. For the Kentucky Warbler, our results suggest that a management strategy that avoids even-age silviculture and leaves core stands of older forest far from cowbird feeding areas can increase nesting success to levels similar to those measured in more forested landscapes.     

Comparison of Outcomes of Permanently Closed and Periodically Harvested Coral Reef Reserves

Abstract: In many areas of the developing world, the establishment of permanent marine reserves is inhibited by cultural norms or socioeconomic pressures. Community conserved areas that are periodically harvested are increasingly being implemented as fisheries management tools, but few researchers have empirically compared them with permanently closed reserves. We used a hierarchal control‐impact experimental design to compare the abundance and biomass of reef fishes, invertebrates, and substrate composition in periodically harvested and permanent reserves and in openly fished (control sites) of the South Pacific island country of Vanuatu. Fished species had significantly higher biomass in periodically harvested reserves than in adjacent openly fished areas. We did not detect differences in substratum composition between permanent reserves and openly fished areas or between permanent reserves and periodically harvested reserves. Giant clams (tridacnids) and top shells (Trochus niloticus) were vulnerable to periodic harvest, and we suggest that for adequate management of these species, periodically harvested community conservation areas be used in conjunction with other management strategies. Periodic harvest within reserves is an example of adaptive and flexible management that may meet conservation goals and that is suited to the social, economic, and cultural contexts of many coastal communities in the developing world.     

Headwater riparian forest-floor invertebrate communities associated with alternative forest management practices.

Headwater streams and their riparian zones are a common, yet poorly understood, component of Pacific Northwest, USA, landscapes. We describe the ecological importance of headwater stream riparian zones as habitat for forest-floor invertebrate communities and assess how alternative management strategies for riparian zones may impact these communities. We compared community composition of forest-floor invertebrates at increasing distances along trans-riparian (stream edge to upslope) transects in mature forests, clearcuts, and riparian buffers of approximately 30-m width with upslope clearcuts. Invertebrates were collected using pitfall traps in five replicate blocks of three treatments each in the Willamette National Forest, Oregon, USA. We measured microclimate and microhabitat variables at pitfall locations. Despite strong elevation and block effects on community composition, community analyses revealed a distinct "riparian" invertebrate community within 1 m of the stream edge in mature forest treatments, which was strongly related to cool, humid microclimate conditions. Invertebrate community composition in buffer treatments was far more similar to that of mature forests than to clearcuts; a pattern mirrored by microclimate. These results suggest that, within our study sites, forest-floor invertebrate distributions are strongly associated with microclimate and that riparian buffers of approximately 30-m width do provide habitat for many riparian and forest species. Riparian reserves may serve as effective forest refugia and/or dispersal corridors for invertebrates and other taxa, and their incorporation into watershed management plans likely will contribute to meeting persistence and connectivity objectives.     

Control of invasive rats on islands and priorities for future action

Invasive rats are one of the world's most successful animal groups that cause native species extinctions and ecosystem change, particularly on islands. On large islands, rat eradication is often impossible and population control, defined as the local limitation of rat abundance, is now routinely performed on many of the world's islands as an alternative management tool. However, a synthesis of the motivations, techniques, costs, and outcomes of such rat‐control projects is lacking. We reviewed the literature, searched relevant websites, and conducted a survey via a questionnaire to synthesize the available information on rat‐control projects in island natural areas worldwide to improve rat management and native species conservation. Data were collected from 136 projects conducted over the last 40 years; most were located in Australasia (46%) and the tropical Pacific (25%) in forest ecosystems (65%) and coastal strands (22%). Most of the projects targeted Rattus rattus and most (82%) were aimed at protecting birds and endangered ecosystems. Poisoning (35%) and a combination of trapping and poisoning (42%) were the most common methods. Poisoning allows for treatment of larger areas, and poison projects generally last longer than trapping projects. Second‐generation anticoagulants (mainly brodifacoum and bromadiolone) were used most often. The median annual cost for rat‐control projects was US$17,262 or US$227/ha. Median project duration was 4 years. For 58% of the projects, rat population reduction was reported, and 51% of projects showed evidence of positive effects on biodiversity. Our data were from few countries, revealing the need to expand rat‐control distribution especially in some biodiversity hotspots. Improvement in control methods is needed as is regular monitoring to assess short‐ and long‐term effectiveness of rat‐control.     

Threats to Avifauna on Oceanic Islands

Abstract:  Results of the study by Blackburn et al. (2004 a ) of avifauna on oceanic islands suggest that distance from the mainland and time since European colonization have major influences on species extinctions and that island area is a significant but secondary contributing factor. After augmenting the data of the study on geographical properties for some of the islands they examined, we used a causal analysis approach with structural equation modeling to reexamine their conclusions. In our model geographical properties of islands, such as island area and isolation, were considered constraints on biological factors, such as the number of introduced mammalian predators and existing number of avifauna, that can directly or indirectly influence extinction. Of the variables we tested, island area had the greatest total influence on the threat of extinction due to its direct and indirect effects on the size of island avifauna. Larger islands had both a greater number of threatened bird species and more avifauna, increasing the number of species that could become threatened with extinction. Island isolation also had a significant, positive, and direct effect on threats to island avifauna because islands farther from the mainland had fewer current extant avifauna. Time since European colonization had a significant negative, but relatively weaker, influence on threats compared with the traditional biogeographic factors of island area and distance to the mainland. We also tested the hypothesis that the amount of threat is proportionally lower on islands that have had more extinctions (i.e., there is a "filter effect"). Because the proportion of bird extinctions potentially explained only 2.3% of the variation in the proportion of threatened species on islands, our results did not support this hypothesis. Causal modeling provided a powerful tool for examining threat of extinction patterns of known and hypothesized pathways of influence.