Explore the vast range of titles available.

Deforestation increases the odds of a species being threatened by extinction, and this effect is disproportionately strong in relatively intact landscapes, suggesting that efforts are needed to protect intact forest landscapes and prevent a new wave of extinctions. Pristine landscapes at highest risk of biodiversity loss Global efforts to reduce biodiversity loss must consider to what extent conservation efforts should focus on modified and fragmented landscapes where extinction threats are potentially greatest, versus landscapes that are largely intact. Matthew Betts and colleagues use International Union for Conservation of Nature (IUCN) data on species ranges and remote sensing data on forest cover to assess the effects of deforestation across intact and human-modified landscapes on the extinction risk of 19,432 vertebrate species worldwide. As expected, deforestation substantially increases the odds of a species being threatened with extinction. However, the strength of this relationship is disproportionately high in relatively intact landscapes. The authors estimate that under current rates of forest loss, 121–219 species will become threatened with extinction in Borneo, the central Amazon and the Congo Basin over the next 30 years. The findings suggest that large-scale efforts to reduce the degradation and loss of intact forest landscapes are needed to guard against an intensified wave of extinctions in the world's last wildernesses. Global biodiversity loss is a critical environmental crisis, yet the lack of spatial data on biodiversity threats has hindered conservation strategies 1 . Theory predicts that abrupt biodiversity declines are most likely to occur when habitat availability is reduced to very low levels in the landscape (10–30%) 2 , 3 , 4 . Alternatively, recent evidence indicates that biodiversity is best conserved by minimizing human intrusion into intact and relatively unfragmented landscapes 5 . Here we use recently available forest loss data 6 to test deforestation effects on International Union for Conservation of Nature Red List categories of extinction risk for 19,432 vertebrate species worldwide. As expected, deforestation substantially increased the odds of a species being listed as threatened, undergoing recent upgrading to a higher threat category and exhibiting declining populations. More importantly, we show that these risks were disproportionately high in relatively intact landscapes; even minimal deforestation has had severe consequences for vertebrate biodiversity. We found little support for the alternative hypothesis that forest loss is most detrimental in already fragmented landscapes. Spatial analysis revealed high-risk hot spots in Borneo, the central Amazon and the Congo Basin. In these regions, our model predicts that 121–219 species will become threatened under current rates of forest loss over the next 30 years. Given that only 17.9% of these high-risk areas are formally protected and only 8.9% have strict protection, new large-scale conservation efforts to protect intact forests 7 , 8 are necessary to slow deforestation rates and to avert a new wave of global extinctions.

Many of the world's vertebrates have experienced large population and geographic range declines due to anthropogenic threats that put them at risk of extinction. The largest vertebrates, defined as megafauna, are especially vulnerable. We analyzed how human activities are impacting the conservation status of megafauna within six classes: mammals, ray‐finned fish, cartilaginous fish, amphibians, birds, and reptiles. We identified a total of 362 extant megafauna species. We found that 70% of megafauna species with sufficient information are decreasing and 59% are threatened with extinction. Surprisingly, direct harvesting of megafauna for human consumption of meat or body parts is the largest individual threat to each of the classes examined, and a threat for 98% (159/162) of threatened species with threat data available. Therefore, minimizing the direct killing of the world's largest vertebrates is a priority conservation strategy that might save many of these iconic species and the functions and services they provide.

Presence-absence surveys are frequently used to monitor populations of rare and elusive species. Such data may also be used as a proxy for breeding activity, but links between presence-absence data and higher-order processes must be validated to determine their reliability. The Marbled Murrelet ( Brachyramphus marmoratus ) is a threatened seabird that nests in older-aged forests along the Pacific Coast. Its nests are exceptionally difficult to find, so we tested whether presence-absence surveys can help identify nesting sites. Between 2018 and 2022 we located 17 trees containing active murrelet nests in the Oregon Coast Range (USA) and 38 trees that purportedly contained no active nests (26 in occupied murrelet stands and 12 in unoccupied stands). Observers surveyed within 200 m of focal trees using standard presence-absence surveys, and we modeled the effects of site status (active nest or control) and distance from the focal tree on probability of recording murrelets. We never detected murrelets in unoccupied control sites. We found some evidence that the probability of recording presence was higher at active nesting sites (0.81, 95% CI: 0.71, 0.88) than at occupied control sites (0.71, 95% CI: 0.64, 0.78) although a null model had similar support. The probability of recording murrelet breeding behaviors in nesting and occupied control sites was 0.20 (95% CI: 0.14, 0.27) regardless of distance to a known active nest. These results suggest that presence-absence surveys may be useful for identifying plausible murrelet nesting habitat, but they are ineffective for identifying active nesting sites. Moreover, we estimated that 20 repeated surveys at a point in space are required to reasonably conclude there are no active nesting sites within 200 m. These findings serve as an important reminder of the limitations that can come with relying on presence-absence data alone to identify breeding sites.

1. Recent advances in occupancy estimation that adjust for imperfect detection have provided substantial improvements over traditional approaches and are receiving considerable use in applied ecology. To estimate and adjust for detectability, occupancy modelling requires multiple surveys at a site and requires the assumption of 'closure' between surveys, i.e. no changes in occupancy between surveys. Violations of this assumption could bias parameter estimates; however, little work has assessed model sensitivity to violations of this assumption or how commonly such violations occur in nature. 2. We apply a modelling procedure that can test for closure to two avian point-count data sets in Montana and New Hampshire, USA, that exemplify time-scales at which closure is often assumed. These data sets illustrate different sampling designs that allow testing for closure but are currently rarely employed in field investigations. Using a simulation study, we then evaluate the sensitivity of parameter estimates to changes in site occupancy and evaluate a power analysis developed for sampling designs that is aimed at limiting the likelihood of closure. 3. Application of our approach to point-count data indicates that habitats may frequently be open to changes in site occupancy at time-scales typical of many occupancy investigations, with 71% and 100% of species investigated in Montana and New Hampshire respectively, showing violation of closure across time periods of 3 weeks and 8 days respectively. 4. Simulations suggest that models assuming closure are sensitive to changes in occupancy. Power analyses further suggest that the modelling procedure we apply can effectively test for closure. 5. Synthesis and applications. Our demonstration that sites may be open to changes in site occupancy over time-scales typical of many occupancy investigations, combined with the sensitivity of models to violations of the closure assumption, highlights the importance of properly addressing the closure assumption in both sampling designs and analysis. Furthermore, inappropriately applying closed models could have negative consequences when monitoring rare or declining species for conservation and management decisions, because violations of closure typically lead to overestimates of the probability of occurrence.

A topic of recurring interest in ecological research is the degree to which vegetation structure influences the distribution and abundance of species. Here we test the applicability of remote sensing, particularly novel use of waveform lidar measurements, for quantifying the habitat heterogeneity of a contiguous northern hardwoods forest in the northeastern United States. We apply these results to predict the breeding habitat quality, an indicator of reproductive output of a well‐studied Neotropical migrant songbird, the Black‐throated Blue Warbler (Dendroica caerulescens). We found that using canopy vertical structure metrics provided unique information for models of habitat quality and spatial patterns of prevalence. An ensemble decision tree modeling approach (random forests) consistently identified lidar metrics describing the vertical distribution and complexity of canopy elements as important predictors of habitat use over multiple years. Although other aspects of habitat were important, including the seasonality of vegetation cover, the canopy structure variables provided unique and complementary information that systematically improved model predictions. We conclude that canopy structure metrics derived from waveform lidar, which will be available on future satellite missions, can advance multiple aspects of biodiversity research, and additional studies should be extended to other organisms and regions.

Understanding the mechanisms enabling coevolution in complex mutualistic networks remains a central challenge in evolutionary biology. We show for the first time, to our knowledge, that a tropical plant species has the capacity to discriminate among floral visitors, investing in reproduction differentially across the pollinator community. After we standardized pollen quality in 223 aviary experiments, successful pollination of Heliconia tortuosa (measured as pollen tube abundance) occurred frequently when plants were visited by long-distance traplining hummingbird species with specialized bills ([Formula] pollen tubes = 1.21 ± 0.12 SE) but was reduced 5.7 times when visited by straight-billed territorial birds ([Formula] pollen tubes = 0.20 ± 0.074 SE) or insects. Our subsequent experiments revealed that plants use the nectar extraction capacity of tropical hummingbirds, a positive function of bill length, as a cue to turn on reproductively. Furthermore, we show that hummingbirds with long bills and high nectar extraction efficiency engaged in daily movements at broad spatial scales (∼1 km), but that territorial species moved only short distances (<100 m). Such pollinator recognition may therefore affect mate selection and maximize receipt of high-quality pollen from multiple parents. Although a diffuse pollinator network is implied, because all six species of hummingbirds carry pollen of H. tortuosa , only two species with specialized bills contribute meaningfully to its reproduction. We hypothesize that this pollinator filtering behavior constitutes a crucial mechanism facilitating coevolution in multispecies plant–pollinator networks. However, pollinator recognition also greatly reduces the number of realized pollinators, thereby rendering mutualistic networks more vulnerable to environmental change. Significance By using structural characteristics, such as long tubular flowers, plants are known to achieve selective visitation by certain pollinator species. These morphological traits typically arise over evolutionary timescales. We show for the first time, to our knowledge, that at least one plant has also evolved the capacity to recognize pollinator species immediately after visitation, thereby increasing the likelihood that a flower visitor has delivered high-quality pollen. This novel responsiveness by the plant leads to functional specialization in an apparently generalized tropical plant–pollinator network. Such specialized linkages likely facilitate coevolution but also, render pollination mutualisms more vulnerable to environmental change.

Protected areas are a key tool in the conservation of global biodiversity and carbon stores. We conducted a global test of the degree to which more than 18,000 terrestrial protected areas (totalling 5,293,217 km 2 ) reduce deforestation in relation to unprotected areas. We also derived indices that quantify how well countries’ forests are protected, both in terms of forested area protected and effectiveness of protected areas at reducing deforestation, in relation to vertebrate species richness, aboveground forest carbon biomass and background deforestation rates. Overall, protected areas did not eliminate deforestation, but reduced deforestation rates by 41%. Protected area deforestation rates were lowest in small reserves with low background deforestation rates. Critically, we found that after adjusting for effectiveness, only 6.5%—rather than 15.7%—of the world’s forests are protected, well below the Aichi Convention on Biological Diversity’s 2020 Target of 17%. We propose that global targets for protected areas should include quantitative goals for effectiveness in addition to spatial extent. A global analysis of deforestation rates in more than 18,000 terrestrial protected areas shows that, once protected area effectiveness is taken into account, only 6.5%—rather than 15.7%—of the world’s forests are protected, well below the Aichi Target of 17%.

To maximize fitness, organisms must assess and select suitable habitat. Early research studying birds suggested that organisms consider primarily vegetation structural cues in their habitat choices. We show that experimental exposure to singing in the post-breeding period provides a social cue that is used for habitat selection the following year by a migrant songbird, the black-throated blue warbler (Dendroica caerulescens). Our experimental social cues coerced individuals to adopt territories in areas of very poor habitat quality where individuals typically do not occur. This indicates that social information can override typical associations with vegetation structure. We demonstrate that a strong settlement response was elicited because post-breeding song at a site is highly correlated with reproductive success. These results constitute a previously undocumented, but highly parsimonious mechanism for the inadvertent transfer of reproductive (public) information from successful breeders to dispersers. We hypothesize that post-breeding song is a pervasive and reliable cue for species that communicate vocally, inhabit temporally autocorrelated environments, produce young asynchronously and/or abandon territories after reproductive failure.

Aim Climate changes are anticipated to have pervasive negative effects on biodiversity and are expected to necessitate widespread range shifts or contractions. Such projections are based upon the assumptions that (1) species respond primarily to broad-scale climatic regimes, or (2) that variation in climate at fine spatial scales is less relevant at coarse spatial scales. However, in montane forest landscapes, high degrees of microclimate variability could influence occupancy dynamics and distributions of forest species. Using high-resolution bird survey and under-canopy air temperature data, we tested the hypothesis that the high vagility of most forest bird species combined with the heterogeneous thermal regime of mountain landscapes would enable them to adjust initial settlement decisions to track their thermal niches. Location Western Cascade Mountains, Oregon, USA. Methods We used dynamic occupancy models to test the degree to which microclimate affects the distribution patterns of forest birds in a heterogeneous mountain environment. In all models we statistically accounted for vegetation structure, vegetation composition and potential biases due to imperfect detection of birds. We generated spatial predictions of forest bird distributions in relation to microclimate and vegetation structure. Results Fine-scale temperature metrics were strong predictors of bird distributions; effects of temperature on within-season occupancy dynamics were as large or larger (1–1.7 times) than vegetation effects. Most species (86.7%) exhibited apparent within-season occupancy dynamics. However, species were almost as likely to be warm associated (i.e., apparent settlement at warmer sites and/or vacancy at cooler sites; 53.3% of species) as cool associated (i.e., apparent settlement at cooler sites and/or vacancy at warmer sites; 46.7% of species), suggesting that microclimate preferences are species specific. Main conclusions High-resolution temperature data increase the quality of predictions about avian distribution dynamics and should be included in efforts to project future distributions. We hypothesize that microclimate-associated distribution patterns may reflect species' potential for behavioural buffering from climate change in montane forest environments.

The biodiversity crisis necessitates spatially extensive methods to monitor multiple taxonomic groups for evidence of change in response to evolving environmental conditions. Programs that combine passive acoustic monitoring and machine learning are increasingly used to meet this need. These methods require large, annotated datasets, which are time‐consuming and expensive to produce, creating potential barriers to adoption in data‐ and funding‐poor regions. Recently released pre‐trained avian acoustic classification models provide opportunities to reduce the need for manual labelling and accelerate the development of new acoustic classification algorithms through transfer learning. Transfer learning is a strategy for developing algorithms under data scarcity that uses pre‐trained models from related tasks to adapt to new tasks. Our primary objective was to develop a transfer learning strategy using the feature embeddings of a pre‐trained avian classification model to train custom acoustic classification models in data‐scarce contexts. We used three annotated avian acoustic datasets to test whether transfer learning and soundscape simulation‐based data augmentation could substantially reduce the annotated training data necessary to develop performant custom acoustic classifiers. We also conducted a sensitivity analysis for hyperparameter choice and model architecture. We then assessed the generalizability of our strategy to increasingly novel non‐avian classification tasks. With as few as two training examples per class, our soundscape simulation data augmentation approach consistently yielded new classifiers with improved performance relative to the pre‐trained classification model and transfer learning classifiers trained with other augmentation approaches. Performance increases were evident for three avian test datasets, including single‐class and multi‐label contexts. We observed that the relative performance among our data augmentation approaches varied for the avian datasets and nearly converged for one dataset when we included more training examples. We demonstrate an efficient approach to developing new acoustic classifiers leveraging open‐source sound repositories and pre‐trained networks to reduce manual labelling. With very few examples, our soundscape simulation approach to data augmentation yielded classifiers with performance equivalent to those trained with many more examples, showing it is possible to reduce manual labelling while still achieving high‐performance classifiers and, in turn, expanding the potential for passive acoustic monitoring to address rising biodiversity monitoring needs.