Moving towards better risk assessment for invertebrate conservation

Ecography, Год журнала: 2025, Номер unknown

Опубликована: Май 5, 2025

Global change threatens a vast number of species with severe population declines or even extinction. The threat status an organism is often designated based on geographic range, size, in either. However, invertebrates, which comprise the bulk animal diversity, are conspicuously absent from global frameworks that assess extinction risk. Many invertebrates hard to study, and it has been questioned whether current risk assessments appropriate for majority these organisms. As rare, we contend lack data organisms makes criteria apply. Using empirical evidence one largest terrestrial arthropod surveys date, consisting over 33 000 collected million hours survey effort, demonstrate estimates trends low sample sizes associated major uncertainty misclassification under defined by IUCN. We argue most ambitious monitoring efforts unlikely produce enough observations reliably estimate ranges more than fraction species, there likely be substantial assessing biodiversity using species‐level trends. In response, discuss need focus metrics can currently measure when conducting highlight modern statistical methods allow quantification could incorporate rare into conservation frameworks, suggest how might adapted meet needs biodiversity.

Язык: Английский

---

Deep learning with citizen science data enables estimation of species diversity and composition at continental extents

Ecology, Год журнала: 2023, Номер 104(12)

Опубликована: Окт. 2, 2023

Effective solutions to conserve biodiversity require accurate community- and species-level information at relevant, actionable scales across entire species' distributions. However, data methodological constraints have limited our ability provide such in robust ways. Herein we employ a Deep-Reasoning Network implementation of the Deep Multivariate Probit Model (DMVP-DRNets), an end-to-end deep neural network framework, exploit large observational environmental sets together estimate landscape-scale species diversity composition continental extents. We present results from novel year-round analysis North American avifauna using over nine million eBird checklists 72 covariates. highlight utility by identifying critical areas high for single group conservation concern, wood warblers, while capturing spatiotemporal variation associations interspecific interactions. In so doing, demonstrate type accurate, high-resolution on that learning approaches as DMVP-DRNets can is needed inform ecological research decision-making multiple scales.

Язык: Английский

---

Incorporating eco-evolutionary information into species distribution models provides comprehensive predictions of species range shifts under climate change

The Science of The Total Environment, Год журнала: 2023, Номер 912, С. 169501 - 169501

Опубликована: Дек. 23, 2023

Язык: Английский

---

Boosting biodiversity monitoring using smartphone-driven, rapidly accumulating community-sourced data

eLife, Год журнала: 2024, Номер 13

Опубликована: Март 26, 2024

Comprehensive biodiversity data is crucial for ecosystem protection. The Biome mobile app, launched in Japan, efficiently gathers species observations from the public using identification algorithms and gamification elements. app has amassed >6 million since 2019. Nonetheless, community-sourced may exhibit spatial taxonomic biases. Species distribution models (SDMs) estimate while accommodating such bias. Here, we investigated quality of its impact on SDM performance. accuracy exceeds 95% birds, reptiles, mammals, amphibians, but seed plants, molluscs, fishes scored below 90%. Our SDMs 132 terrestrial plants animals across Japan revealed that incorporating into traditional survey improved accuracy. For endangered species, required >2000 records accurate (Boyce index ≥ 0.9), blending two sources reduced this to around 300. uniform coverage urban-natural gradients by data, compared biased towards natural areas, explain improvement. Combining multiple better estimates distributions, aiding protected area designation service assessment. Establishing a platform accumulating will contribute conserving monitoring ecosystems.

Язык: Английский

---

Shrinkage-based Bayesian variable selection for species distribution modelling in complex environments: An application to urban biodiversity

Ecological Informatics, Год журнала: 2024, Номер 81, С. 102561 - 102561

Опубликована: Март 15, 2024

Robust, quantitative understanding of the diverse ecological needs species is needed to inform effective biodiversity conservation, now and in future, but lacking for most species. The advent "big data" ecology presents unprecedented opportunities fill this gap disentangle drivers biodiversity. Variable model selection sparse (small sample sizes species), high-dimensional (large pool candidate predictors) problems is, however, non-trivial. Here, we employ cross-validated Bayesian projection predictive variable shrinkage priors identify, from a list 70 biophysical predictor variables, minimal subset that best predicts habitat preferences distributions 103 amphibians, birds, butterflies, dragonflies, grasshoppers using city Zurich, Switzerland, as case study. We contrast performance inference models fit with full set predictors (exhaustive models) limited number obtained by compiling weakly informative (selective models). show exhaustive excel performance, albeit at cost greater complexity compared selective models. Results reveal importance access aquatic wide range taxa, relative other such urbanisation, vegetation environmental hazards. These results are complemented more nuanced insights into specific types (ponds, lakes, streams) (herb, shrub, canopy cover) distribution urban biodiversity, well different spatial scales which relevance. Our findings demonstrate potential shrinkage-based leverage big data modelling, contribute development concrete guidelines planning infrastructure design account conservation.

Язык: Английский

---

Boosting biodiversity monitoring using smartphone-driven, rapidly accumulating community-sourced data

eLife, Год журнала: 2024, Номер 13

Опубликована: Июнь 20, 2024

Comprehensive biodiversity data is crucial for ecosystem protection. The Biome mobile app, launched in Japan, efficiently gathers species observations from the public using identification algorithms and gamification elements. app has amassed >6 million since 2019. Nonetheless, community-sourced may exhibit spatial taxonomic biases. Species distribution models (SDMs) estimate while accommodating such bias. Here, we investigated quality of its impact on SDM performance. accuracy exceeds 95% birds, reptiles, mammals, amphibians, but seed plants, molluscs, fishes scored below 90%. Our SDMs 132 terrestrial plants animals across Japan revealed that incorporating into traditional survey improved accuracy. For endangered species, required >2000 records accurate (Boyce index ≥ 0.9), blending two sources reduced this to around 300. uniform coverage urban-natural gradients by data, compared biased towards natural areas, explain improvement. Combining multiple better estimates distributions, aiding protected area designation service assessment. Establishing a platform accumulating will contribute conserving monitoring ecosystems.

Язык: Английский

---

Surface and subsurface oceanographic features drive forage fish distributions and aggregations: Implications for prey availability to top predators in the US Northeast Shelf ecosystem

Ecology and Evolution, Год журнала: 2023, Номер 13(7)

Опубликована: Июль 1, 2023

Forage fishes are a critical food web link in marine ecosystems, aggregating hierarchical patch structure over multiple spatial and temporal scales. Surface-level forage fish aggregations (FFAs) represent concentrated source of prey available to surface- shallow-foraging predators. Existing survey analysis methods often imperfect for studying at scales appropriate foraging predators, making it difficult quantify predator-prey interactions. In many cases, general distributions species known; however, these may not surface-level availability Likewise, we lack an understanding the oceanographic drivers patterns aggregation or community patterns. Specifically, applied Bayesian joint distribution models bottom trawl data assess species- community-level across US Northeast Continental Shelf (NES) ecosystem. Aerial digital surveys gathered on surface FFAs two project sites within NES, which used spatially explicit model estimate abundance size FFAs. We examine aggregations. Our results suggest that, regions high richness consistent with FFA abundance. Bathymetric depth drove both patterns, while subsurface features, such as mixed layer depth, primarily influenced behavior sea temperature, sub-mesoscale eddies, fronts diversity. combination, help predators novel application aerial data.

Язык: Английский

---

BiodiversityR: Package for Community Ecology and Suitability Analysis

Опубликована: Июль 28, 2007

Язык: Английский

---

Boosting biodiversity monitoring using smartphone-driven, rapidly accumulating community-sourced data

Опубликована: Май 17, 2024

Ecosystem services, which derive in part from biological diversity, are a fundamental support for human society. However, activities causing harm to biodiversity, ultimately endangering these critical ecosystem services. Halting nature loss and mitigating impacts necessitates comprehensive biodiversity distribution data, requirement implementing the Kunming-Montreal Global Biodiversity Framework. To efficiently collect species observations public, we launched ‘ Biome ’ mobile application Japan. By employing identification algorithms gamification elements, app has gathered >6M since its launch 2019. community-sourced data often exhibit spatial taxonomic biases. Species models (SDMs) enable infer while accommodating such bias. We investigated data’s quality how incorporating influences performance of SDMs. accuracy exceeds 95% birds, reptiles, mammals, amphibians, but seed plants, molluscs, fishes scored below 90%. The distributions 132 terrestrial plants animals across Japan were modeled, their was improved by our into traditional survey data. For endangered species, required >2,000 records build accurate (Boyce index ≥ 0.9), though only ca.300 when two sources blended. unique may explain this improvement: covers urban-natural gradients uniformly, is biased towards natural areas. Combining multiple offers insights Japan, aiding protected area designation service assessment. Providing platform accumulate improving processing protocol will contribute not conserving ecosystems also detecting changes testing ecological theories.

Язык: Английский

---

Climate refugia along Lake Superior’s shores: disjunct arctic–alpine plants rely on cool shoreline temperatures but are restricted to highly exposed habitat under climate warming

Journal of Plant Ecology, Год журнала: 2024, Номер 17(4)

Опубликована: Июнь 6, 2024

Abstract Climate refugia can serve as a remnant habitat or stepping stones for species dispersal under climate warming. The largest freshwater lake by surface area, Lake Superior, USA and Canada, serves model system understanding cooling-mediated local refugia, its cool water temperatures wave action have maintained shoreline habitats suitable southern disjunct populations of arctic–alpine plants since deglaciation. Here, we seek to explain spatial patterns environmental drivers plant along Superior’s shores, assess future risk moderate (+3.5 °C) warmest (+5.7 warming scenarios. First, examined how the interactive effects summer wind affected onshore temperatures, resulting in areas cooler refugia. Second, developed an ecological niche presence (pooling 1253 occurrences from 58 species) lake’s shoreline. Third, fit distribution models 20 most common predicted identify hotspots. Finally, used two scenarios predict changes Bedrock type, elevation above water, inland distance, July land temperature MODIS/Terra satellite near-shore depth were best predictors occurrences. Overall, 2236 km (51%) at least one current conditions, but this was reduced 20% 7% with (894 km) (313 change projections.

Язык: Английский

---