Automated conservation assessment of the orchid family with deep learning

Using comparative extinction risk analysis to prioritize the IUCN Red List reassessments of amphibians

Assessing the extinction risk of species based on the International Union for Conservation of Nature (IUCN) Red List (RL) is key to guiding conservation policies and reducing biodiversity loss. This process is resource demanding, however, and requires continuous updating, which becomes increasingly difficult as new species are added to the RL. Automatic methods, such as comparative analyses used to predict species RL category, can be an efficient alternative to keep assessments up to date. Using amphibians as a study group, we predicted which species are more likely to change their RL category and thus should be prioritized for reassessment. We used species biological traits, environmental variables, and proxies of climate and land-use change as predictors of RL category. We produced an ensemble prediction of IUCN RL category for each species by combining 4 different model algorithms: cumulative link models, phylogenetic generalized least squares, random forests, and neural networks. By comparing RL categories with the ensemble prediction and accounting for uncertainty among model algorithms, we identified species that should be prioritized for future reassessment based on the mismatch between predicted and observed values. The most important predicting variables across models were species' range size and spatial configuration of the range, biological traits, climate change, and land-use change. We compared our proposed prioritization index and the predicted RL changes with independent IUCN RL reassessments and found high performance of both the prioritization and the predicted directionality of changes in RL categories. Ensemble modeling of RL category is a promising tool for prioritizing species for reassessment while accounting for models' uncertainty. This approach is broadly applicable to all taxa on the IUCN RL and to regional and national assessments and may improve allocation of the limited human and economic resources available to maintain an up-to-date IUCN RL.

Effectiveness of South Africa's network of protected areas: Unassessed vascular plants predicted to be threatened using deep neural networks are all located in protected areas

Globally, we are in the midst of a biodiversity crisis and megadiverse countries become key targets for conservation. South Africa, the only country in the world hosting three biodiversity hotspots within its borders, harbours a tremendous diversity of at-risk species deserving to be protected. However, the lengthy risk assessment process and the lack of required data to complete assessments is a serious limitation to conservation since several species may slide into extinction while awaiting risk assessment. Here, we employed a deep neural network model integrating species climatic and geographic features to predict the conservation status of 116 unassessed plant species. Our analysis involved in total of 1072 plant species and 96,938 occurrence points. The best-performing model exhibits high accuracy, reaching up to 83.6% at the binary classification and 56.8% at the detailed classification. Our best-performing model at the binary classification predicts that 32% (25 species) and 8% (3 species) of Data Deficient and Not-Evaluated species respectively, are likely threatened, amounting to a proportion of 24.1% of unassessed species facing a risk of extinction. Interestingly, all unassessed species predicted to be threatened are in protected areas, revealing the effectiveness of South Africa's network of protected areas in conservation, although these likely threatened species are more abundant outside protected areas. Considering the limitation in assessing only species with available data, there remains a possibility of a higher proportion of unassessed species being imperilled.

Inferring the extinction risk of marine fish to inform global conservation priorities

While extinction risk categorization is fundamental for building robust conservation planning for marine fishes, empirical data on occurrence and vulnerability to disturbances are still lacking for most marine teleost fish species, preventing the assessment of their International Union for the Conservation of Nature (IUCN) status. In this article, we predicted the IUCN status of marine fishes based on two machine learning algorithms, trained with available species occurrences, biological traits, taxonomy, and human uses. We found that extinction risk for marine fish species is higher than initially estimated by the IUCN, increasing from 2.5% to 12.7%. Species predicted as Threatened were mainly characterized by a small geographic range, a relatively large body size, and a low growth rate. Hotspots of predicted Threatened species peaked mainly in the South China Sea, the Philippine Sea, the Celebes Sea, the west coast Australia and North America. We also explored the consequences of including these predicted species' IUCN status in the prioritization of marine protected areas through conservation planning. We found a marked increase in prioritization ranks for subpolar and polar regions despite their low species richness. We suggest to integrate multifactorial ensemble learning to assess species extinction risk and offer a more complete view of endangered taxonomic groups to ultimately reach global conservation targets like the extending coverage of protected areas where species are the most vulnerable.

Extinction risk predictions for the world's flowering plants to support their conservation

More than 70% of all vascular plants lack conservation status assessments. We aimed to address this shortfall in knowledge of species extinction risk by using the World Checklist of Vascular Plants to generate the first comprehensive set of predictions for a large clade: angiosperms (flowering plants, c. 330 000 species). We used Bayesian Additive Regression Trees (BART) to predict the extinction risk of all angiosperms using predictors relating to range size, human footprint, climate, and evolutionary history and applied a novel approach to estimate uncertainty of individual species-level predictions. From our model predictions, we estimate 45.1% of angiosperm species are potentially threatened with a lower bound of 44.5% and upper bound of 45.7%. Our species-level predictions, with associated uncertainty estimates, do not replace full global, or regional Red List assessments, but can be used to prioritise predicted threatened species for full Red List assessment and fast-track predicted non-threatened species for Least Concern assessments. Our predictions and uncertainty estimates can also guide fieldwork, inform systematic conservation planning and support global plant conservation efforts and targets.

Comprehensive conservation assessments reveal high extinction risks across Atlantic Forest trees

Biodiversity is declining globally, yet many biodiversity hotspots still lack comprehensive species conservation assessments. Using multiple International Union for Conservation of Nature (IUCN) Red List criteria to evaluate extinction risks and millions of herbarium and forest inventory records, we present automated conservation assessments for all tree species of the Atlantic Forest biodiversity hotspot, including ~1100 heretofore unassessed species. About 65% of all species and 82% of endemic species are classified as threatened. We rediscovered five species classified as Extinct on the IUCN Red List and identified 13 endemics as possibly extinct. Uncertainties in species information had little influence on the assessments, but using fewer Red List criteria severely underestimated threat levels. We suggest that the conservation status of tropical forests worldwide is worse than previously reported.

More than 17,000 tree species are at risk from rapid global change

Trees are pivotal to global biodiversity and nature's contributions to people, yet accelerating global changes threaten global tree diversity, making accurate species extinction risk assessments necessary. To identify species that require expert-based re-evaluation, we assess exposure to change in six anthropogenic threats over the last two decades for 32,090 tree species. We estimated that over half (54.2%) of the assessed species have been exposed to increasing threats. Only 8.7% of these species are considered threatened by the IUCN Red List, whereas they include more than half of the Data Deficient species (57.8%). These findings suggest a substantial underestimation of threats and associated extinction risk for tree species in current assessments. We also map hotspots of tree species exposed to rapidly changing threats around the world. Our data-driven approach can strengthen the efforts going into expert-based IUCN Red List assessments by facilitating prioritization among species for re-evaluation, allowing for more efficient conservation efforts.

Modelling the probability of meeting IUCN Red List criteria to support reassessments

Comparative extinction risk analysis-which predicts species extinction risk from correlation with traits or geographical characteristics-has gained research attention as a promising tool to support extinction risk assessment in the IUCN Red List of Threatened Species. However, its uptake has been very limited so far, possibly because existing models only predict a species' Red List category, without indicating which Red List criteria may be triggered. This prevents such approaches to be integrated into Red List assessments. We overcome this implementation gap by developing models that predict the probability of species meeting individual Red List criteria. Using data on the world's birds, we evaluated the predictive performance of our criterion-specific models and compared it with the typical criterion-blind modelling approach. We compiled data on biological traits (e.g. range size, clutch size) and external drivers (e.g. change in canopy cover) often associated with extinction risk. For each specific criterion, we modelled the relationship between extinction risk predictors and species' Red List category under that criterion using ordinal regression models. We found criterion-specific models were better at identifying threatened species compared to a criterion-blind model (higher sensitivity), but less good at identifying not threatened species (lower specificity). As expected, different covariates were important for predicting extinction risk under different criteria. Change in annual temperature was important for criteria related to population trends, while high forest dependency was important for criteria related to restricted area of occupancy or small population size. Our criteria-specific method can support Red List assessors by producing outputs that identify species likely to meet specific criteria, and which are the most important predictors. These species can then be prioritised for re-evaluation. We expect this new approach to increase the uptake of extinction risk models in Red List assessments, bridging a long-standing research-implementation gap.

Prioritizing the reassessment of data-deficient species on the IUCN Red List

Despite being central to the implementation of conservation policies, the usefulness of the International Union for Conservation of Nature (IUCN) Red List of Threatened Species is hampered by the 14% of species classified as data-deficient (DD) because information to evaluate these species' extinction risk was lacking when they were last assessed or because assessors did not appropriately account for uncertainty. Robust methods are needed to identify which DD species are more likely to be reclassified in one of the data-sufficient IUCN Red List categories. We devised a reproducible method to help red-list assessors prioritize reassessment of DD species and tested it with 6887 DD species of mammals, reptiles, amphibians, fishes, and Odonata (dragonflies and damselflies). For each DD species in these groups, we calculated its probability of being classified in a data-sufficient category if reassessed today from covariates measuring available knowledge (e.g., number of occurrence records or published articles available), knowledge proxies (e.g., remoteness of the range), and species characteristics (e.g., nocturnality); calculated change in such probability since last assessment from the increase in available knowledge (e.g., new occurrence records); and determined whether the species might qualify as threatened based on recent rate of habitat loss determined from global land-cover maps. We identified 1907 species with a probability of being reassessed in a data-sufficient category of >0.5; 624 species for which this probability increased by >0.25 since last assessment; and 77 species that could be reassessed as near threatened or threatened based on habitat loss. Combining these 3 elements, our results provided a list of species likely to be data-sufficient such that the comprehensiveness and representativeness of the IUCN Red List can be improved.

Deep learning in terrestrial conservation biology

Biodiversity is being lost at an unprecedented rate on Earth. As a first step to more effectively combat this process we need efficient methods to monitor biodiversity changes. Recent technological advance can provide powerful tools (e.g. camera traps, digital acoustic recorders, satellite imagery, social media records) that can speed up the collection of biological data. Nevertheless, the processing steps of the raw data served by these tools are still painstakingly slow. A new computer technology, deep learning based artificial intelligence, might, however, help. In this short and subjective review I oversee recent technological advances used in conservation biology, highlight problems of processing their data, shortly describe deep learning technology and show case studies of its use in conservation biology. Some of the limitations of the technology are also highlighted.

Why plant diversity and distribution matter

This article is the Editorial for the Special Collection ‘Global plant diversity and distribution’. See https://www.newphytologist.org/global-plant-diversity for more details.

Botanical Collection Patterns and Conservation Categories of the Most Traded Timber Species from the Ecuadorian Amazon: The Role of Protected Areas

The Ecuadorian Amazon is home to a rich biodiversity of woody plant species. Nonetheless, their conservation remains difficult, as some areas remain poorly explored and lack georeferenced records. Therefore, the current study aims predominantly to analyze the collection patterns of timber species in the Amazon lowlands of Ecuador and to evaluate the conservation coverage of these species in protected areas. Furthermore, we try to determine the conservation category of the species according to the criteria of the IUCN Red List. We identified that one third of the timber species in the study area was concentrated in three provinces due to historical botanical expeditions. However, a worrying 22.0% of the species had less than five records of presence, and 29.9% had less than ten records, indicating a possible underestimation of their presence. In addition, almost half of the species evaluated were unprotected, exposing them to deforestation risks and threats. To improve knowledge and conservation of forest biodiversity in the Ecuadorian Amazon, it is recommended to perform new botanical samplings in little-explored areas and digitize data in national herbaria. It is critical to implement automated assessments of the conservation status of species with insufficient data. In addition, it is suggested to use species distribution models to identify optimal areas for forest restoration initiatives. Effective communication of results and collaboration between scientists, governments, and local communities are key to the protection and sustainable management of forest biodiversity in the Amazon region.

Diversity and distribution of Orchidaceae in one of the world's most threatened plant hotspots (Madagascar)

INTRODUCTION

In recent decades, Madagascar has become one of the most important plant hotspots in the world. The country's remaining forests and vegetation are disappearing at an alarming rate, while dozens of new species are discovered each year. Amongst the plant families that have long been studied, Orchidaceae appear as one of the most charismatic, diverse and of high conservation concern. Based on a reviewed, comprehensive herbarium dataset, we have compiled a curated checklist of all orchid species occurring in Madagascar. Based on this complete dataset, we then compiled sampling effort, species diversity distribution and some general statistics on their ecology and IUCN conservation status.

METHODS

We compiled and standardised a global dataset using five public databases as the main data sources, supplemented by the most recent publications. The database contains ~ 10,000 geolocated records collected between 1816 and 2021. We used GIS software and rarefaction methods to examine sampling and diversity patterns.

RESULTS

According to our dataset, there are currently 913 orchid species collected in Madagascar, of which 759 orchid species (83.1%) are endemic. Doubling the sampling effort could lead to the discovery of around 100 more species, bringing the total estimated number of orchid species in Madagascar to between 986 and 1048. About one-third (297 species) of all orchid species are known only by type specimens (189 species) or have not been collected in Madagascar for more than 50 years (214 species). Although the raw data show that the Andasibe-Moramanga area would have the highest orchid species concentration, our analysis of the data, adjusted for bias, shows that the centres of orchid diversity in Madagascar are in the Tsaratanàna Strict Nature Reserve and the Ranomafana National Park. Life-form statistics show that 55.0% of orchid species are strict epiphytes. The main flowering period of orchids in Madagascar is between November and March. To date, 84% of the 226 Malagasy orchid species listed in the IUCN Red List are threatened with extinction (CR, EN or VU).

CONCLUSION

Despite geographically uneven coverage, the biodiversity of Malagasy orchids appears to be already well documented. We provide maps corrected for sampling bias that indicate priority areas for future surveys. Upcoming efforts should also focus on rediscovery and conservation of rare and/or threatened species and ensure that the protected area network is well aligned with the distribution of priority species for conservation. Finally, the conservation status of 75% of the orchid species found in Madagascar is not yet known and the inclusion of these species must be a top priority in the coming years.

Achieving zero extinction for land plants

Despite the importance of plants for humans and the threats to their future, plant conservation receives far less support compared with vertebrate conservation. Plants are much cheaper and easier to conserve than are animals, but, although there are no technical reasons why any plant species should become extinct, inadequate funding and the shortage of skilled people has created barriers to their conservation. These barriers include the incomplete inventory, the low proportion of species with conservation status assessments, partial online data accessibility, varied data quality, and insufficient investment in both in and ex situ conservation. Machine learning, citizen science (CS), and new technologies could mitigate these problems, but we need to set national and global targets of zero plant extinction to attract greater support.

Review of Orchidaceae of the northern part of Kazakhstan

We present a review of Orchidaceae Juss. of the northern part of Kazakhstan, within the steppe, forest-steppe and semi-desert habitats of the country (Pavlodar, northern Kazakhstan, Kostanay, Akmola, Aktobe, West Kazakhstan, partially Karaganda and East Kazakhstan regions). The investigation is based on herbarium materials, literature data and field observations. We examined material from the following herbarium collections: LE, MW, TK, MHA, SVER, KUZ, ALTB, AA, NUR, KG, KSPI, NS, NSK, MOSP, ORIS, PPIU, totalling 288 herbarium specimens. The paper presents data in the form of revision, focusing on orchids of the northern part of Kazakhstan. It is accompanied by maps indicating localities, notes on habitat preferences, phenology and conservation status. A total of 25 species of 16 genera were recorded, of which eight are included in the Red Book of Kazakhstan (2014). According to our data, we propose to enlarge the number of protected orchids by adding the following nine species: Corallorhizatrifida, Epipactisatrorubens, Gymnadeniaconopsea, Hammarbyapaludosa, Herminiummonorchis, Liparisloeselii, Malaxismonophyllos, Neottiacamtschatea and Spiranthesaustralis. The most widespread species in the studied region are Dactylorhizaincarnata, D.umbrosa and Epipactispalustris. The rarest species (single locality only) are Epipactisatrorubens, E.helleborine, Epipogiumaphyllum, Hammarbyapaludosa and Herminiummonorchis.

Evidence-based guidelines for automated conservation assessments of plant species

Assessing species' extinction risk is vital to setting conservation priorities. However, assessment endeavors, such as those used to produce the IUCN Red List of Threatened Species, have significant gaps in taxonomic coverage. Automated assessment (AA) methods are gaining popularity to fill these gaps. Choices made in developing, using, and reporting results of AA methods could hinder their successful adoption or lead to poor allocation of conservation resources. We explored how choice of data cleaning type and level, taxonomic group, training sample, and automation method affect performance of threat status predictions for plant species. We used occurrences from the Global Biodiversity Information Facility (GBIF) to generate assessments for species in 3 taxonomic groups based on 6 different occurrence-based AA methods. We measured each method's performance and coverage following increasingly stringent occurrence cleaning. Automatically cleaned data from GBIF performed comparably to occurrence records cleaned manually by experts. However, all types of data cleaning limited the coverage of AAs. Overall, machine-learning-based methods performed well across taxa, even with minimal data cleaning. Results suggest a machine-learning-based method applied to minimally cleaned data offers the best compromise between performance and species coverage. However, optimal data cleaning, training sample, and automation methods depend on the study group, intended applications, and expertise.

Madagascar’s extraordinary biodiversity: Threats and opportunities

Madagascar’s unique biota is heavily affected by human activity and is under intense threat. Here, we review the current state of knowledge on the conservation status of Madagascar’s terrestrial and freshwater biodiversity by presenting data and analyses on documented and predicted species-level conservation statuses, the most prevalent and relevant threats, ex situ collections and programs, and the coverage and comprehensiveness of protected areas. The existing terrestrial protected area network in Madagascar covers 10.4% of its land area and includes at least part of the range of the majority of described native species of vertebrates with known distributions (97.1% of freshwater fishes, amphibians, reptiles, birds, and mammals combined) and plants (67.7%). The overall figures are higher for threatened species (97.7% of threatened vertebrates and 79.6% of threatened plants occurring within at least one protected area). International Union for Conservation of Nature (IUCN) Red List assessments and Bayesian neural network analyses for plants identify overexploitation of biological resources and unsustainable agriculture as the most prominent threats to biodiversity. We highlight five opportunities for action at multiple levels to ensure that conservation and ecological restoration objectives, programs, and activities take account of complex underlying and interacting factors and produce tangible benefits for the biodiversity and people of Madagascar.

An overview of remote monitoring methods in biodiversity conservation

Conservation of biodiversity is critical for the coexistence of humans and the sustenance of other living organisms within the ecosystem. Identification and prioritization of specific regions to be conserved are impossible without proper information about the sites. Advanced monitoring agencies like the Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES) had accredited that the sum total of species that are now threatened with extinction is higher than ever before in the past and are progressing toward extinct at an alarming rate. Besides this, the conceptualized global responses to these crises are still inadequate and entail drastic changes. Therefore, more sophisticated monitoring and conservation techniques are required which can simultaneously cover a larger surface area within a stipulated time frame and gather a large pool of data. Hence, this study is an overview of remote monitoring methods in biodiversity conservation via a survey of evidence-based reviews and related studies, wherein the description of the application of some technology for biodiversity conservation and monitoring is highlighted. Finally, the paper also describes various transformative smart technologies like artificial intelligence (AI) and/or machine learning algorithms for enhanced working efficiency of currently available techniques that will aid remote monitoring methods in biodiversity conservation.

The likely extinction of hundreds of palm species threatens their contributions to people and ecosystems

Protecting nature's contributions to people requires accelerating extinction risk assessment and better integrating evolutionary, functional and used diversity with conservation planning. Here, we report machine learning extinction risk predictions for 1,381 palm species (Arecaceae), a plant family of high socio-economic and ecological importance. We integrate these predictions with published assessments for 508 species (covering 75% of all palm species) and we identify top-priority regions for palm conservation on the basis of their proportion of threatened evolutionarily distinct, functionally distinct and used species. Finally, we explore palm use resilience to identify non-threatened species that could potentially serve as substitutes for threatened used species by providing similar products. We estimate that over a thousand palms (56%) are probably threatened, including 185 species with documented uses. Some regions (New Guinea, Vanuatu and Vietnam) emerge as top ten priorities for conservation only after incorporating machine learning extinction risk predictions. Potential substitutes are identified for 91% of the threatened used species and regional use resilience increases with total palm richness. However, 16 threatened used species lack potential substitutes and 30 regions lack substitutes for at least one of their threatened used palm species. Overall, we show that hundreds of species of this keystone family face extinction, some of them probably irreplaceable, at least locally. This highlights the need for urgent actions to avoid major repercussions on palm-associated ecosystem processes and human livelihoods in the coming decades.

More than half of data deficient species predicted to be threatened by extinction

The IUCN Red List of Threatened Species is essential for practical and theoretical efforts to protect biodiversity. However, species classified as “Data Deficient” (DD) regularly mislead practitioners due to their uncertain extinction risk. Here we present machine learning-derived probabilities of being threatened by extinction for 7699 DD species, comprising 17% of the entire IUCN spatial datasets. Our predictions suggest that DD species as a group may in fact be more threatened than data-sufficient species. We found that 85% of DD amphibians are likely to be threatened by extinction, as well as more than half of DD species in many other taxonomic groups, such as mammals and reptiles. Consequently, our predictions indicate that, amongst others, the conservation relevance of biodiversity hotspots in South America may be boosted by up to 20% if DD species were acknowledged. The predicted probabilities for DD species are highly variable across taxa and regions, implying current Red List-derived indices and priorities may be biased.

Data Deficient species are more likely to be at extinction risk than previously thought across multiple taxonomic groups.

Automated assessment reveals that the extinction risk of reptiles is widely underestimated across space and phylogeny

The Red List of Threatened Species, published by the International Union for Conservation of Nature (IUCN), is a crucial tool for conservation decision-making. However, despite substantial effort, numerous species remain unassessed or have insufficient data available to be assigned a Red List extinction risk category. Moreover, the Red Listing process is subject to various sources of uncertainty and bias. The development of robust automated assessment methods could serve as an efficient and highly useful tool to accelerate the assessment process and offer provisional assessments. Here, we aimed to (1) present a machine learning–based automated extinction risk assessment method that can be used on less known species; (2) offer provisional assessments for all reptiles—the only major tetrapod group without a comprehensive Red List assessment; and (3) evaluate potential effects of human decision biases on the outcome of assessments. We use the method presented here to assess 4,369 reptile species that are currently unassessed or classified as Data Deficient by the IUCN. The models used in our predictions were 90% accurate in classifying species as threatened/nonthreatened, and 84% accurate in predicting specific extinction risk categories. Unassessed and Data Deficient reptiles were considerably more likely to be threatened than assessed species, adding to mounting evidence that these species warrant more conservation attention. The overall proportion of threatened species greatly increased when we included our provisional assessments. Assessor identities strongly affected prediction outcomes, suggesting that assessor effects need to be carefully considered in extinction risk assessments. Regions and taxa we identified as likely to be more threatened should be given increased attention in new assessments and conservation planning. Lastly, the method we present here can be easily implemented to help bridge the assessment gap for other less known taxa.

A Machine Learning Approach for the Non-Destructive Estimation of Leaf Area in Medicinal Orchid Dendrobium nobile L

In this study, leaf area prediction models of Dendrobium nobile, were developed through machine learning (ML) techniques including multiple linear regression (MLR), support vector regression (SVR), gradient boosting regression (GBR), and artificial neural networks (ANNs). The best model was tested using the coefficient of determination (R2), mean absolute errors (MAEs), and root mean square errors (RMSEs) and statistically confirmed through average rank (AR). Leaf images were captured through a smartphone and ImageJ was used to calculate the length (L), width (W), and leaf area (LA). Three orders of L, W, and their combinations were taken for model building. Multicollinearity status was checked using Variance Inflation Factor (VIF) and Tolerance (T). A total of 80% of the dataset and the remaining 20% were used for training and validation, respectively. KFold (K = 10) cross-validation checked the model overfit. GBR (R2, MAE and RMSE values ranged at 0.96, (0.82–0.91) and (1.10–1.11) cm2) in the testing phase was the best among the ML models. AR statistically confirms the outperformance of GBR, securing first rank and a frequency of 80% among the top ten ML models. Thus, GBR is the best model imparting its future utilization to estimate leaf area in D. nobile.

Improving biodiversity protection through artificial intelligence

Over a million species face extinction, urging the need for conservation policies that maximize the protection of biodiversity to sustain its manifold contributions to people. Here we present a novel framework for spatial conservation prioritization based on reinforcement learning that consistently outperforms available state-of-the-art software using simulated and empirical data. Our methodology, CAPTAIN (Conservation Area Prioritization Through Artificial INtelligence), quantifies the trade-off between the costs and benefits of area and biodiversity protection, allowing the exploration of multiple biodiversity metrics. Under a limited budget, our model protects substantially more species from extinction than areas selected randomly or naively (such as based on species richness). CAPTAIN achieves substantially better solutions with empirical data than alternative software, meeting conservation targets more reliably and generating more interpretable prioritization maps. Regular biodiversity monitoring, even with a degree of inaccuracy characteristic of citizen science surveys, substantially improves biodiversity outcomes. Artificial intelligence holds great promise for improving the conservation and sustainable use of biological and ecosystem values in a rapidly changing and resourcelimited world.

Improving biodiversity protection through artificial intelligence

Over a million species face extinction, urging the need for conservation policies that maximize the protection of biodiversity to sustain its manifold contributions to people. Here we present a novel framework for spatial conservation prioritization based on reinforcement learning that consistently outperforms available state-of-the-art software using simulated and empirical data. Our methodology, CAPTAIN (Conservation Area Prioritization Through Artificial INtelligence), quantifies the trade-off between the costs and benefits of area and biodiversity protection, allowing the exploration of multiple biodiversity metrics. Under a limited budget, our model protects substantially more species from extinction than areas selected randomly or naively (such as based on species richness). CAPTAIN achieves substantially better solutions with empirical data than alternative software, meeting conservation targets more reliably and generating more interpretable prioritization maps. Regular biodiversity monitoring, even with a degree of inaccuracy characteristic of citizen science surveys, substantially improves biodiversity outcomes. Artificial intelligence holds great promise for improving the conservation and sustainable use of biological and ecosystem values in a rapidly changing and resourcelimited world.

Global Estimation and Mapping of the Conservation Status of Tree Species Using Artificial Intelligence

Trees are fundamental for Earth's biodiversity as primary producers and ecosystem engineers and are responsible for many of nature's contributions to people. Yet, many tree species at present are threatened with extinction by human activities. Accurate identification of threatened tree species is necessary to quantify the current biodiversity crisis and to prioritize conservation efforts. However, the most comprehensive dataset of tree species extinction risk-the Red List of the International Union for the Conservation of Nature (IUCN RL)-lacks assessments for a substantial number of known tree species. The RL is based on a time-consuming expert-based assessment process, which hampers the inclusion of less-known species and the continued updating of extinction risk assessments. In this study, we used a computational pipeline to approximate RL extinction risk assessments for more than 21,000 tree species (leading to an overall assessment of 89% of all known tree species) using a supervised learning approach trained based on available IUCN RL assessments. We harvested the occurrence data for tree species worldwide from online databases, which we used with other publicly available data to design features characterizing the species' geographic range, biome and climatic affinities, and exposure to human footprint. We trained deep neural network models to predict their conservation status, based on these features. We estimated 43% of the assessed tree species to be threatened with extinction and found taxonomic and geographic heterogeneities in the distribution of threatened species. The results are consistent with the recent estimates by the Global Tree Assessment initiative, indicating that our approach provides robust and time-efficient approximations of species' IUCN RL extinction risk assessments.

Deep Species Distribution Modeling From Sentinel-2 Image Time-Series: A Global Scale Analysis on the Orchid Family

Species distribution models (SDMs) are widely used numerical tools that rely on correlations between geolocated presences (and possibly absences) and environmental predictors to model the ecological preferences of species. Recently, SDMs exploiting deep learning and remote sensing images have emerged and have demonstrated high predictive performance. In particular, it has been shown that one of the key advantages of these models (called deep-SDMs) is their ability to capture the spatial structure of the landscape, unlike prior models. In this paper, we examine whether the temporal dimension of remote sensing images can also be exploited by deep-SDMs. Indeed, satellites such as Sentinel-2 are now providing data with a high temporal revisit, and it is likely that the resulting time-series of images contain relevant information about the seasonal variations of the environment and vegetation. To confirm this hypothesis, we built a substantial and original dataset (called DeepOrchidSeries) aimed at modeling the distribution of orchids on a global scale based on Sentinel-2 image time series. It includes around 1 million occurrences of orchids worldwide, each being paired with a 12-month-long time series of high-resolution images (640 x 640 m RGB+IR patches centered on the geolocated observations). This ambitious dataset enabled us to train several deep-SDMs based on convolutional neural networks (CNNs) whose input was extended to include the temporal dimension. To quantify the contribution of the temporal dimension, we designed a novel interpretability methodology based on temporal permutation tests, temporal sampling, and temporal averaging. We show that the predictive performance of the model is greatly increased by the seasonality information contained in the temporal series. In particular, occurrence-poor species and diversity-rich regions are the ones that benefit the most from this improvement, revealing the importance of habitat's temporal dynamics to characterize species distribution.

Nectar Chemistry or Flower Morphology-What Is More Important for the Reproductive Success of Generalist Orchid Epipactis palustris in Natural and Anthropogenic Populations?

The aim of this study was to determine the level of reproductive success (RS) in natural and anthropogenic populations of generalist orchid Epipactis palustris and its dependence on flower structure and nectar composition, i.e., amino acids and sugars. We found that both pollinaria removal and female reproductive success were high and similar in all populations, despite differences in flower traits and nectar chemistry. Flower structures were weakly correlated with parameters of RS. Nectar traits were more important in shaping RS; although, we noted differentiated selection on nectar components in distinct populations. Individuals in natural populations produced nectar with a larger amount of sugars and amino acids. The sucrose to (fructose and glucose) ratio in natural populations was close to 1, while in anthropogenic ones, a clear domination of fructose and glucose was noted. Our results indicate that the flower traits and nectar composition of E. palustris reflect its generalist character and meet the requirements of a wide range of pollinators, differing according to body sizes, mouth apparatus, and dietary needs. Simultaneously, differentiation of nectar chemistry suggests a variation of pollinator assemblages in particular populations or domination of their some groups. To our knowledge, a comparison of nectar chemistry between natural and anthropogenic populations of orchids is reported for the first time in this paper.

The adaptive challenge of extreme conditions shapes evolutionary diversity of plant assemblages at continental scales

The tropical conservatism hypothesis (TCH) posits that the latitudinal gradient in biological diversity arises because most extant clades of animals and plants originated when tropical environments were more widespread and because the colonization of colder and more seasonal temperate environments is limited by the phylogenetically conserved environmental tolerances of these tropical clades. Recent studies have claimed support of the TCH, indicating that temperate plant diversity stems from a few more recently derived lineages that are nested within tropical clades, with the colonization of the temperate zone being associated with key adaptations to survive colder temperatures and regular freezing. Drought, however, is an additional physiological stress that could shape diversity gradients. Here, we evaluate patterns of evolutionary diversity in plant assemblages spanning the full extent of climatic gradients in North and South America. We find that in both hemispheres, extratropical dry biomes house the lowest evolutionary diversity, while tropical moist forests and many temperate mixed forests harbor the highest. Together, our results support a more nuanced view of the TCH, with environments that are radically different from the ancestral niche of angiosperms having limited, phylogenetically clustered diversity relative to environments that show lower levels of deviation from this niche. Thus, we argue that ongoing expansion of arid environments is likely to entail higher loss of evolutionary diversity not just in the wet tropics but in many extratropical moist regions as well.

Orchids of Mongolia: Taxonomy, Species Richness and Conservation Status

Orchidaceae is a diverse, globally important plant family with high conservation assessment and prioritization requirements. The checklist of Orchidaceae in Mongolia was updated based on herbarium materials, literature, and field observations. Mongolian orchids were revised as comprising 26 taxa belonging to 14 genera with major updates were conducted on Herminium and Epipactis. In particular, H. alaschanicum, previously noted in the Alashan Gobi region, was added to the flora of Mongolia based on literature and type specimens. Epipactis helleborine and E. palustris were excluded from the Mongolian flora owing to the absence of herbarium specimens and wild collection from Mongolia. Assessment of all orchid species at the national level resulted in 1, 4, 7, 11, and 2 species as critically endangered (CR), endangered (EN), vulnerable (VU), near threatened (NT), and data deficient (DD), respectively, according to IUCN criteria. Species richness and conservation gap analyses of 970 georeferenced orchid records based on 0.5° × 0.5° grid cells across 16 phytogeographical regions of Mongolia, showed that four phytogeographical regions, Khangai, Khuvgul, Khentii and Mongolian Dauria, have a high number of orchids. Regrettably, most orchid-rich locations in Mongolia are not fully within protected areas, highlighting the need for protection management updates. Based on herbarium collections, we prepared grid distribution maps of the 26 taxa using 40 × 40 km2 grids. Photographs of 18 taxa taken during fieldwork were included, providing valuable information on species morphology and typical habitat.

Reversing extinction trends: new uses of (old) herbarium specimens to accelerate conservation action on threatened species

Although often not collected specifically for the purposes of conservation, herbarium specimens offer sufficient information to reconstruct parameters that are needed to designate a species as 'at-risk' of extinction. While such designations should prompt quick and efficient legal action towards species recovery, such action often lags far behind and is mired in bureaucratic procedure. The increase in online digitization of natural history collections has now led to a surge in the number new studies on the uses of machine learning. These repositories of species occurrences are now equipped with advances that allow for the identification of rare species. The increase in attention devoted to estimating the scope and severity of the threats that lead to the decline of such species will increase our ability to mitigate these threats and reverse the declines, overcoming a current barrier to the recovery of many threatened plant species. Thus far, collected specimens have been used to fill gaps in systematics, range extent, and past genetic diversity. We find that they also offer material with which it is possible to foster species recovery, ecosystem restoration, and de-extinction, and these elements should be used in conjunction with machine learning and citizen science initiatives to mobilize as large a force as possible to counter current extinction trends.

Extinction Risk Assessment of the Greek Endemic Flora

Human-induced biodiversity decline has been on the rise for the past 250 years, due to various causes. What is equally troubling, is that we are unaware which plants are threatened and where they occur. Thus, we are far from reaching Aichi Biodiversity Target 2, i.e., assessing the extinction risk of most species. To that end, based on an extensive occurrence dataset, we performed an extinction risk assessment according to the IUCN Criteria A and B for all the endemic plant taxa occurring in Greece, one of the most biodiverse countries in Europe, in a phylogenetically-informed framework and identified the areas needing conservation prioritization. Several of the Greek endemics are threatened with extinction and fourteen endemics need to be prioritized, as they are evolutionary distinct and globally endangered. Mt. Gramos is identified as the most important conservation hotspot in Greece. However, a significant portion of the identified conservation hotspots is not included in any designated Greek protected area, meaning that the Greek protected areas network might need to be at least partially redesigned. In the Anthropocene era, where climate and land-use change are projected to alter biodiversity patterns and may force many species to extinction, our assessment provides the baseline for future conservation research, ecosystem services maintenance, and might prove crucial for the timely, systematic and effective aversion of plant extinctions in Greece.

Orchid Extinction over the Last 150 Years in the Czech Republic

Understanding temporal changes in the distribution and abundance of various species is one of the key goals of conservation biology. During recent decades, the abundance and distribution of many species of plants and animals have declined dramatically, mainly because of habitat loss and fragmentation. The purpose of this study is to analyze the rate of extinction of orchids at various sites in different 20-year time intervals over the last 150 years, determined according to changes in society. Using the dataset of the orchid records of the Nature Conservation Agency of the Czech Republic, we determined the disappearance rate of orchids from sites using a grid of 1 × 1 km. We found that the vast majority of orchids disappeared from many of their historical localities in all time intervals analyzed. The number of sites suitable for Czech orchids declined by 8–92%, depending on the species. The most threatened orchid species in the Czech Republic are Spiranthes spiralis, Anacamptis palustris, Epipogium aphyllum and Goodyera repens. This all seems to be closely related with changes in agricultural practices in the open as well as in forest habitats. Preserving suitable orchid habitats seems to be the key for keeping Czech orchid flora alive.

Assessing extinction risk from geographic distribution data in Neotropical freshwater fishes

Abstract The IUCN Red List (RL) provides high-quality conservation assessments for individual species, yet the rate and scale of environmental deterioration globally challenges the conservation community to develop expedited methods for risk assessment. Here we compare threat assessments for 3,001 species of Neotropical freshwater fishes (NFF) in the IUCN–RL using readily accessible data types as proxies for extinction risk: geographic range, elevation, and species publication date. Furthermore, using geographic and taxonomic data alone, we generated preliminary conservation assessments for 2,334 NFF species currently awaiting IUCN assessment, identifying an additional 671 NFF species as potentially threatened. This number of potentially threatened species represents an increase of 59% over the number of species currently assigned to threat categories by the IUCN–RL. These results substantially expand the number of threatened NFF species from 422 currently on the IUCN RL to 1,093 species as threatened or potentially threatened, representing about 18% of all NFF species. Extinction risk is greater in species with smaller geographic ranges, which inhabit upland rivers, and which were described more recently. We propose the Central and Southern Andes, and Eastern Guiana Shield as priorities in the upcoming IUCN RL assessment of NFF species conservation risk.