Predicting population trends of birds worldwide with big data and machine learning

Zhang, Xuan; Campomizzi, Andrew J.; Lebrun‐Southcott, Zoé M.

doi:10.1111/ibi.13045

Cited by 2 publications

(2 citation statements)

References 95 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several studies have demonstrated proof-of-concept for predictive modelling approaches for monitoring or conservation prioritization for plant and animal groups, but at continental or global scales (Bastin et al, 2019; Parsons, Pelletier, Wieringa, Duckett, & Carstens, 2022; Pelletier et al, 2018; Wieringa, 2022; Zhang, Campomizzi, & Lebrun-Southcott, 2022; Zizka et al, 2022). Because decisions about monitoring, research, and protection are often made at smaller spatial scales, before recommending adoption of these tools for conservation practice, it is crucial to determine if they still provide useful predictions at those scales (Meyer & Pebesma, 2021; Wyborn & Evans, 2021), as we have done here.…”

Section: Discussionmentioning

confidence: 99%

Global conservation prioritization approach provides credible results at a regional scale

Roswell,

Espíndola

2023

Preprint

View full text Add to dashboard Cite

Aim: Conservationists and managers must direct resources and enact measures to protect species, despite uncertainty about their present statuses. One approach to covering the data gap is borrowing information from data-rich species or populations to guide decisions about data-poor ones, via machine learning. Recent efforts demonstrated proof-of-concept at the global scale, leaving unclear whether similar approaches are feasible at the local and regional scales at which conservation actions most typically occur. To address this gap, we tested a global-scale predictive approach at a regional scale, using two groups of taxa. Location: Maryland, USA. Taxa: Vascular land plants and lepidopterans. Methods: Using publicly available occurrence and biogeographic data, we trained random forest classifiers to predict the state-level conservation status of species in each of the two focal taxa. We assessed model performance with cross-validation, and explored trends in the predictions. Results: Our models had strong discriminatory ability, accurately predicting status for species with existing status assessments. They predict that the northwestern part of Maryland, USA, which overlaps the Appalachian Mountains, harbors a higher concentration of unassessed, but threatened plants and lepidopterans. Our predictions track known biogeographic patterns, and unassessed species predicted as most likely threatened in Maryland were often recognized as also needing conservation in nearby jurisdictions, providing external validation to our results. Main Conclusions: We demonstrate that a modelling approach developed for global analysis can be downscaled and credible when applied at a regional scale that is smaller than typical species ranges. We identified select unassessed plant and lepidopteran species, and the western, montane region of Maryland as priority targets for additional monitoring, assessment, and conservation. By rapidly aggregating disparate data and integrating information across taxa, models like those we used can complement traditional assessment tools and assist in prioritization for formal assessments, as well as protection.

show abstract

Section: Discussionmentioning

confidence: 99%

Global conservation prioritization approach provides credible results at a regional scale

Roswell,

Espíndola

2023

Preprint

View full text Add to dashboard Cite

show abstract

“…The steepest declines were observed in forest-specialized invertivores, a group known to be particularly sensitive to the effects of deforestation and fragmentation ( 14 , 29 ), and matching findings from other Neotropical forests ( 14 , 28 ). A recent global analysis of avian population declines found that severe population fragmentation was the best predictor for declines in nonmigratory birds in South America ( 30 ). In this system, for example, the Rufous Mourner ( Rhytipterna holerythra ) and the Northern Nightingale-wren ( Microcerculus philomela ) are sensitive to habitat fragmentation, effects that have driven local extirpations throughout Costa Rica ( 15 , 31 ); these same species were characterized by rapidly declining population sizes in this study.…”

Section: Discussionmentioning

confidence: 99%

Diversified farms bolster forest-bird populations despite ongoing declines in tropical forests

Hendershot,

Echeverri,

Frishkoff

et al. 2023

Proc. Natl. Acad. Sci. U.S.A.

View full text Add to dashboard Cite

While some agricultural landscapes can support wildlife in the short term, it is uncertain how well they can truly sustain wildlife populations. To compare population trends in different production systems, we sampled birds along 48 transects in mature forests, diversified farms, and intensive farms across Costa Rica from 2000 to 2017. To assess how land use influenced population trends in the 349 resident and 80 migratory species with sufficient data, we developed population models. We found, first, that 23% of species were stable in all three land use types, with the rest almost evenly split between increasing and decreasing populations. Second, in forest habitats, a slightly higher fraction was declining: 62% of the 164 species undergoing long-term population changes; nearly half of these declines occurred in forest-affiliated invertivores. Third, in diversified farms, 49% of the 230 species with population changes were declining, with 60% of these declines occurring in agriculture-affiliated species. In contrast, 51% of the species with population changes on diversified farms showed increases, primarily in forest-affiliated invertivores and frugivores. In intensive farms, 153 species showed population changes, also with similar proportions of species increasing (50%) and decreasing (50%). Declines were concentrated in agriculture-affiliated invertivores and forest-affiliated frugivores; increases occurred in many large, omnivorous species. Our findings paint a complex picture but clearly indicate that diversified farming helps sustain populations of diverse, forest-affiliated species. Despite not fully offsetting losses in forest habitats, diversified farming practices help sustain wildlife in a critical time, before possible transformation to nature-positive policies and practices.

show abstract

Predicting population trends of birds worldwide with big data and machine learning

Cited by 2 publications

References 95 publications

Global conservation prioritization approach provides credible results at a regional scale

Global conservation prioritization approach provides credible results at a regional scale

Diversified farms bolster forest-bird populations despite ongoing declines in tropical forests

Contact Info

Product

Resources

About