Distribution Of Rare Species Research Articles

Modeling the geographic distribution of pawpaw (Asimina triloba [L.] Dunal) in a portion of its northern range limits, western New York State, USA

Pawpaw (Asimina triloba [L.] Dunal) is a small deciduous tree in eastern North America notable for its edible fruit. Its northern range extends into New York State (NYS) where it is listed as threatened, with only approximately 20 sites known to contain pawpaw. This research models the geographic distribution of pawpaw in western NYS (land area = 27,263 km2) for four reasons: to understand factors shaping its distribution, to assess the degree to which it is protected, to locate potential introduction or reintroduction sites, and to locate potential sites with undiscovered pawpaw. This study uses an ensemble of small models (ESM) approach devised for modeling the distribution of rare species, which averages predictions from simple (i.e., bivariate) models weighted by model predictive performance. The ESM in this study, constructed from boosted regression tree (BRT) models, demonstrated a high ability to discern pawpaw presences from pseudo-absences. The area under the receiver operating characteristic curve (AUC) value was 0.986 when calculated using training data, and BRT models comprising the ESM exhibited AUC values calculated using cross-validation data as high as 0.957. The ESM suggests that pawpaw is found in the warmest (mean annual temperature > 9.0 °C,) and driest (May–September precipitation 110 cm) soils. Locations predicted to be suitable for pawpaw are overwhelmingly found on unprotected lands; forested areas in PAs with “high” or “highest” suitability for pawpaw only comprise 0.1–0.2% of the study area.

Exploring a legendary giant squid: an environmental DNA approach

The giant squid, Architeuthis dux, has been reported for centuries, but its distribution and ecology have remained famously mysterious. We investigated the A. dux distribution in the Sea of Japan using the environmental DNA (eDNA) method, which has recently been considered as a useful technique for evaluating the distributions of rare species. To develop the eDNA method to detect A. dux, we first created a species-specific polymerase chain reaction (PCR) primer–probe and tried to detect the eDNA in the western region of the Sea of Japan where A. dux has recently been observed. We successfully collected and amplified the A. dux eDNA in our study area using field water sampling and real-time PCR measurements. A. dux eDNA was detected in winter but not in summer, reflecting the historical record of A. dux observations in the region. The use of eDNA techniques could be a potential method for monitoring "invisible" and rare organisms, even in open ocean habitats.

Integrating species distribution and occupancy modeling to study hellbender (Cryptobranchus alleganiensis) occurrence based on eDNA surveys

Managers often rely on species surveys and distribution models to evaluate species occurrence and develop management and conservation plans. However, these tools are rarely used in concert. We used a three-step framework to evaluate the distribution of a declining and elusive freshwater amphibian species, the hellbender salamander (Cryptobranchus alleganiensis). We used the Maximum Entropy (MaxEnt) algorithm to develop a habitat suitability model to guide sampling, environmental DNA (eDNA) surveys to ground truth the habitat suitability model, and multi-level occupancy modeling to assess species presence, while accounting for eDNA detection errors. Our suitability model (AUC = 0.941, True Skill Statistic = 0.7, sensitivity and specificity = 0.86) identified the greatest amount of high and very highly suitable habitat in the Interior Plateau and Blue Ridge ecoregions of the study area. We used eDNA survey results (n = 284 sites) to evaluate model fit and detected the species at 65 sites. Detection probability (p) was 0.692 (95% CRI: 0.547, 0.818) at the site level and 0.674 (95% CRI: 0.621, 0.721) at the quantitative PCR level. Ecoregion was the primary covariate that explained occupancy, with greatest estimates in the Blue Ridge ecoregion 0.695 (95% CRI, 0.390, 0.925). Although the MaxEnt output was not significantly correlated with occupancy probability, we established a relationship between habitat quality and the number of eDNA detections. This study highlights the use of a multi-level framework to optimize sampling, assess model fit, account for imperfect detection, and evaluate the distribution of rare species that have limited occurrence data available.

Identifying Highly Diverse Areas of Rare Plant Species as a Basis for Assessing Representativeness and Improving the Network of Protected Areas

Assessing the representativeness of protected areas is necessary to optimize their national and regional networks, as well as to improve the protection of rare and endemic species. In the Republic of Bashkortostan, an analysis of the distribution of rare vascular plant species in need of protection has been carried out on the basis of synthesizing all sources of high-precision georeferenced data, i.e., databases of herbarium specimens and geobotanical releves. The result of these works was the creation of a graphic database: the Rare and Endangered Species of Vascular Plants of the Republic of Bashkortostan GIS map. This GIS map includes three main vector layers: a point layer containing information about 4932 rare species localities, a Protected Areas Boundaries layer, and a grid layer with cells 6' in latitude by 10' in longitude (10.8 × 10.2 km). Based on these layers and standard procedures of the QGIS 3.4 program, an algorithm has been developed to analyze the patterns of distribution of rare species throughout the region, as well as to identify areas with a high diversity of these species and their covering by the existing network of protected areas. Using the example of two grid map cells with a high diversity of rare species, approaches to identifying and delineating further sites of protected areas have been considered using space images and a digital elevation model. To estimate the protection level of the unique combinations of rare species, the distribution of all sets of these species within the existing protected areas is assessed. This makes it possible to justify the creation of two new protected areas. This approach does not exclude, but significantly reduces the volume of field investigations and can significantly reduce the time and financial costs of creating new protected areas to optimize the protection of rare and endangered plant species.

Формування та трансформація штучних комплексних пам’яток природи другої половини XX століття Українського Полісся

Protected man-made parks of Ukrainian Polissya, created in the period of postmodernism, are conservation oriented, historical, cognitive and educational objects. Landscape design is a synthesis of traditions and new transformation trends. On the territories of protected man-made parks, natural existence and distribution of rare species are maintained, endangered and vulnerable species are protected, hydrological functions are improved, and valuable landscapes are preserved. Concentrated in Rivne region, five parks, which have the status of complex natural monuments of local significance, were created in Ukrainian Polissya in the 1960s. According to the results of the analysis of formation and development history, the transformation of complex natural monuments (CNMs) of Ukrainian Polissya for a period of over 60 years is shown. CNMs have a simple planning structure of the following types: cross-shaped ("Vysotsky Arboretum"), star-shaped ("Bilsky Arboretum", "Sarny Arboretum"), loop-shaped ("Rokytne Arboretum"), and combined ("Tryputnyansky Park"). Of the three parks with regular planning, the biggest change can be observed in the complex natural monument "Bilsky Arboretum". This is due to the growth of hedges, which are not currently formed, and to the violation of the boundaries of meadows, where self-seeding plants appear. The dendroflora of five man-made parks is represented by 157 species, which belong to 19 orders, 32 families, and 81 genera, with 111 of them being included to the IUCN Red List. In two parks during their existence, there was an increase in the number of species by 11-50%, with the number of rare tree species amounting to 50-71%. In three parks under the influence of anthropogenic and ecological factors, there was a decrease in dendrodiversity, with 55-81% of species of woody plants remaining; аmong them 71-75% are rare tree species. Protected man-made areas of Ukrainian Polissya are valuable objects that need to be further developed and protected, whereas their collections of woody plants replenished and maintained.

Adaptive niche‐based sampling to improve ability to find rare and elusive species: Simulations and field tests

Abstract Sampling efficiency is crucial to overcome the data crisis in biodiversity and to understand what drives the distribution of rare species. Adaptive niche‐based sampling (ANBS) is an iterative sampling strategy that relies on the predictions of species distribution models (SDMs). By predicting highly suitable areas to guide prospection, ANBS could improve the efficiency of sampling effort in terms of finding new locations for rare species. Its iterative quality could potentially mitigate the effect of small and initially biased samples on SDMs. In this study, we compared ANBS with random sampling by assessing the gain in terms of new locations found per unit of effort. The comparison was based on both simulations and two field surveys of mountain birds. We found an increasing probability of contacting the species through iterations if the focal species showed specialization in the environmental gradients used for modelling. We also identified a gain when using pseudo‐absences during first iterations, and a general tendency of ANBS to increase the omission rate in the spatial prediction of the species' niche or habitat. Overall, ANBS is an effective and flexible strategy that can contribute to a better understanding of distribution drivers in rare species.

Habitat associations and distributions of two endemic crayfishes, Cambarus (Erebicambarus) maculatus Hobbs & Pflieger, 1988 and Faxonius (Billecambarus) harrisonii (Faxon, 1884) (Decapoda: Astacoidea: Cambaridae), in the Meramec River drainage, Missouri, USA

Abstract Understanding the habitat associations and distributions of rare species is important to inform management and policy decisions. Cambarus (Erebicambarus) maculatus Hobbs & Pflieger, 1988, the freckled crayfish, and Faxonius (Billecambarus) harrisonii (Faxon, 1884), the belted crayfish, are two of Missouri’s endemic crayfish species. Both species are listed as Vulnerable (S3) on Missouri’s Species and Communities of Conservation Concern Checklist due to their limited range within the Meramec River drainage (MRD) and the impact of anthropogenic activities therein. Their distributional overlap offers an opportunity for multi-species research to address gaps in information required for conservation. We sampled 140 sites throughout the MRD during the summers of 2017 and 2018 for crayfishes and associated habitat variables, which we related to crayfish presence in an occupancy modeling framework. We found that C. maculatus occupancy was associated with larger stream size, boulder substrate, dolomite lithology, aquatic vegetation beds, dissolved oxygen, and pool mesohabitat. Faxonius harrisonii occupancy increased with boulder substrate, aquatic vegetation beds, the presence of C. maculatus, and decreased in third-order streams. We also expanded the known range for both species within the MRD. Range estimates (watershed area) for C. maculatus and F. harrisonii were 4,347 km2 and 3,690 km2, respectively. This study demonstrates the importance of targeted rather than opportunistic sampling for species distribution.

Predicting the distribution of a rare chipmunk (Neotamias quadrivittatus oscuraensis): comparing MaxEnt and occupancy models.

Species distribution models (SDMs) use presence records to determine the relationship between species occurrence and various environmental variables to create predictive maps describing the species’ distribution. The Oscura Mountains Colorado chipmunk (Neotamias quadrivittatus oscuraensis) occurs in central New Mexico and is of conservation concern due to its relict distribution and threats to habitat. We previously created an occupancy model for this taxon, but were concerned that the model may not have adequately captured the ecological factors influencing the chipmunk’s distribution because of the data hungry nature of occupancy modeling. MaxEnt is another SDM method that is particularly effective at testing large numbers of variables and handling small sample sizes. Our goal was to create a MaxEnt model for the Oscura Mountains Colorado chipmunk and to compare it with our previous occupancy model for this taxon, either to strengthen our original assessment of the relevant ecological factors or identify additional factors that were not captured by our occupancy model. We created MaxEnt models using occurrence records from baited camera traps and opportunistic surveys. We adjusted model complexity using a novel method for tuning both the regularization multiplier and feature class parameters while also performing variable selection. We compared the distribution maps and variables selected by MaxEnt to the results of our occupancy model for this taxon. The MaxEnt and occupancy models selected similar environmental variables and the overall spatial pattern of occurrence was similar for each model. Likelihood of occurrence was positively related to elevation, piñon woodland vegetation type, and topographic variables associated with escarpments. The overall similarities between the MaxEnt and occupancy models increased our confidence of the ecological factors influencing the distribution of the chipmunk. We conclude that MaxEnt offers advantages for predicting the distribution of rare species, which can help inform conservation actions.

Diversity patterns of reef fish along the Brazilian tropical coast

Causal mechanisms for broad-scale reef fish diversity patterns are poorly understood and current knowledge is limited to trends of species richness. This work compared the effects of ecological drivers on components of fish diversity across reefs spanning over 2.000 km of the tropical Brazilian coastline. A quarter of communities’ diversity is accountable to common and dominant species, while remaining species are rare. Low-latitude sites were more diverse in rare species. Communities along the coast share common and dominant species, which display high densities across all reefs, but differ in rare species that show abundance peaks in particular reef morphotypes. The disproportionate distribution of rare species reveals a higher vulnerability of these communities to impacts and stochastic density fluctuations. Uneven conservation efforts directed to these morphotypes pose a threat to the maintenance of a paramount component of the reef fish diversity represented by rare species.

Predictor species: Improving assessments of rare species occurrence by modeling environmental co-responses.

Designing an effective conservation strategy requires understanding where rare species are located. Because rare species can be difficult to find, ecologists often identify other species called conservation surrogates that can help inform the distribution of rare species. Species distribution models typically rely on environmental data when predicting the occurrence of species, neglecting the effect of species' co‐occurrences and biotic interactions. Here, we present a new approach that uses Bayesian networks to improve predictions by modeling environmental co‐responses among species. For species from a European peat bog community, our approach consistently performs better than single‐species models and better than conventional multi‐species approaches that include the presence of nontarget species as additional independent variables in regression models. Our approach performs particularly well with rare species and when calibration data are limited. Furthermore, we identify a group of “predictor species” that are relatively common, insensitive to the presence of other species, and can be used to improve occurrence predictions of rare species. Predictor species are distinct from other categories of conservation surrogates such as umbrella or indicator species, which motivates focused data collection of predictor species to enhance conservation practices.

Aphyllophoroid funga (Basidiomycota) of Finland: range extensions and records of nationally new and rare species

Knowledge of the Finnish aphyllophoroid funga has increased substantially in recent years. In this article, we present two species new to Finland: <em>Spiculogloea subminuta</em> Hauerslev and <em>Typhula suecica</em> I. Olariaga, G. Corriol, I. Salcedo & K. Hansen, and document <em>Sistotrema luteoviride</em> Kotir. & K.-H. Larss. for the third time globally. We also contribute 50 new records of 33 nationally rare species (with a maximum of ten previous records in Finland) and list 52 regionally new species, found for the first time in a certain subzone of the boreal vegetation zone in Finland. Each record is enclosed and contains notes on the substrate. Furthermore, the ecology of the nationally new species and the distribution of rare species are discussed.

Risk assessment and conservation strategies for rare lichen species and communities threatened by sea-level rise in the Mid-Atlantic Coastal Plain

The Mid-Atlantic Coastal Plain (MACP) is an ecoregion with high biodiversity that is under imminent threat from sea level rise and habitat loss. Previous studies have shown i) the area to be high in lichen biodiversity, including endemic species, ii) the most lichen species-rich sites are the most imperiled by sea-level rise, and iii) common, widespread lichens have lost significant suitable habitat and are projected to lose more due to sea-level rise. Despite this, the spatial distributions of rare species and the compositional dynamics of lichen communities has not previously been examined. Here, we evaluate the community composition of 599 lichen species in 215 sites across the MACP. We tested whether community similarity is correlated with spatial proximity or environmental conditions. We found that sites with similar environmental characteristics had similar species composition, after controlling for geographic distance. Inland swamps had the highest per-site average number of unique species (1.3) and the two coastal ecosystems, maritime forests and pocosin (a unique form of peatland vegetation), each had an average of nearly one unique species per site. Across the entire MACP, 42% of species were found at five or fewer sites and these were mostly at sites under high risk of being affected by sea level rise. While habitats and species throughout the MACP have been projected to be negatively affected by sea level rise, our results suggest that rare lichens face particularly acute threats. Lack of suitable habitat inland at higher elevations may necessitate intensive mitigation, including facilitated transplants of both lichens and their host trees.

Contribution of rare and common species to subterranean species richness patterns.

AimCommon species contribute more to species richness patterns (SRPs) than rare species in most studies. Our aim was to test this hypothesis using a novel model system, species living exclusively in subterranean habitats. They consist of mainly rare species (small ranges), only a few of them being common (large ranges), and challenge whether rare species are less important for the development of SRPs in this environment. We separately analyzed aquatic and terrestrial species.LocationWestern Balkans in southeastern Europe.MethodsWe assembled two datasets comprising 431 beetle and 145 amphipod species, representing the model groups of subterranean terrestrial and aquatic diversity, respectively. We assessed the importance of rare and common species using the stepwise reconstruction of SRPs and subsequent correlation analyses, corrected also for the cumulative information content of the subsets based on species prevalence. We applied generalized linear regression models to evaluate the importance of rare and common species in forming SRPs. Additionally, we analyzed the contribution of rare and common species in species‐rich cells.ResultsPatterns of subterranean aquatic and terrestrial species richness overlapped only weakly, with aquatic species having larger ranges than terrestrial ones. Our analyses supported higher importance of common species for forming overall SRPs in both beetles and amphipods. However, in stepwise analysis corrected for information content, results were ambiguous. Common species presented a higher proportion of species than rare species in species‐rich cells.Main ConclusionWe have shown that even in habitats with the domination of rare species, it is still common species that drive SRPs. This may be due to an even spatial distribution of rare species or spatial mismatch in hotspots of rare and common species. SRPs of aquatic and terrestrial subterranean organisms overlap very little, so the conservation approaches need to be habitat specific.

GEOGRAPHIC DISTRIBUTION OF ENDEMIC TYPES OF THE KREMENETS MOUNTAINS FLORA

The article is devoted to the study of rare and endangered species of flora of the Kremenets mountains. A list of 19 rare species has been compiled, ecological-cenotic features and current status of their populations have been analyzed. The dynamics of the geographical distribution of rare species over the last 150-200 years has been determined, and recommendations for improving their conservation have been developed.A considerable group consists of endemic species, including: Lesser Poland - Volyn-Podolia: Rosa czackiana Bess., Carlina onopordifolia Bess.ex Szaf., Kulcz.et Pawl .; Volyn-Podilskyi: Dianthus pseudoserotinus Blocki, Euphorbia volhynica Bess. ex Racib., Senecio besserianus Minder., Anthyllis schiwereckii (DC.) Blocki., Symphytum besseri Zaverucha, Galium exoletum Klok., Salvia dumetorum Andrz., Ranunculus zapalowiczii Pacz .; Podillian: Spiraea pikoviensis Bess., Salvia cremenecensis Bess., Lapulla semicinkta (Stew.) M.Pop., Aconitum besserianum Andrz., Arum besserianum Schott; North-Podolia: Minuartia aucta Klok., Jurinea pachysperma Klok .; Kremenets Mountains: Betula klokovii Zaverucha, Myosotis ludomilae Zaverucha.Based on studies of the dynamic trends in the geographical distribution of endemic flora of the Kremenets Mountains over the last 150-200 years, it has been found that eleven species, which were listed in the 19th - 1st half of the 20th century, have tendencies to reduce their range , Salvia nutans, Carlina onopordifolia Myosotis ludomilae, Aconitum besserianum, Ranunculus zapalowiczii, Senecio besserianus, Lapulla semicinkta); 3 species (Dianthus pseudoserotinus, Euphorbia volhynica, Minuartia aucta) have progressive tendencies to expand their range. Populations of Spiraea pikoviensis, Salvia dumetorum, Rosa czackiana, Galium exoletum, Jurinea pachysperma are characterized by stable positions in the flora of the Kremenets mountains.

SIMULATION OF SPATIAL DISTRIBUTION OF RARE SPECIES IN PROTECTED NATURAL AREAS (ON THE EXAMPLE OF THE CENTRAL FOREST RESERVE, RUSSIA)

An important task of modern ecology is the modeling of the spatial distribution of organisms. Of particular relevance is the modeling of the distribution of rare species in protected natural areas. The paper discusses the main stages and presents the results of modeling the habitat suitability of the Central Forest State Nature Biosphere Reserve (Russia, Tver region) for the red vole. This species is a rare and relict for the center of the Caspian-Baltic watershed. In the modelling the maximum entropy method (MaxEnt) was used. The basis for building the model was the field sampling of the authors, performed in July-August 2010-2014. Each year, trapping were carried out at 745 sampling points each of which has geographical referencing in the WGS 84. A total of 12238 trap-days were worked out and 141 red vole individuals were caught. In the process of modeling, the contribution to the species distribution of certain integral characteristics of habitats (relief, vegetation type and three vegetation spectral indices) was determined. A map model of habitat suitability for the red vole on the territory of the Reserve was built, which allowed to describe the spatial structure of the population groups of the species. The total areas of the most favorable sites for the habitat of the red vole on the reserve's territory were calculated.

Using owl pellets to infer habitat associations and clarify the regional distribution of a cryptic shrew

AbstractAccurate knowledge of the distribution of rare species remains a great challenge in conservation biology because these species are prone to extinction and the number of species becoming rare is increasing. However, the distribution of rare species is usually difficult to delimit because of their low abundances and/or small distribution ranges. These difficulties are even greater when they are sympatric with morphologically similar species that can lead to identification errors. Here, we use barn owl (Tyto alba) pellets to infer, indirectly, potential habitat associations of the lesser white‐toothed shrew (Crocidura suaveolens) in the Gulf of Cádiz, south‐western Iberia and to clarify the distribution of the species in this region. In the Gulf of Cádiz, C. suaveolens is a rare species that lives in sympatry with its congener, the greater white‐toothed shrew (Crocidura russula), a much more abundant and extremely similar species. Through several indirect lines of evidence we inferred a possible association of C. suaveolens with tidal marsh habitat. This was a very surprising finding since C. suaveolens is a habitat generalist throughout its global distribution range. The species presented a highly restricted distribution in the Gulf of Cádiz occurring only in four tidal marshes. The causes of this limited distribution remain poorly understood to date, but our results are consistent with the hypothesis that C. suaveolens was displaced to tidal marshes through competitive exclusion by C. russula, although the geomorphological history of the tidal marshes of the Gulf of Cádiz may also explain its regional distribution and the phylogeographical patterns recently described in the region.

Spatial distributions of common and rare species in a widespread group of stream insects

The phenomenon of rarity is understudied in freshwater systems compared with terrestrial systems. We examined whether spatial distributions of rare species of black flies (Diptera: Simuliidae) differed from those of common species. We used mean pairwise distance between stream sites as a proxy for dispersal ability to describe spatial distributions. From a master database of 1445 North American stream-site collections, in which all immature black flies were identified to species, we created one database of 12 ecoregion-level data sets (476 sites) and a second database of 5 larger area-level data sets (504 sites) to address the following questions: 1) Does the spatial distribution (clumped, random, or overdispersed) differ between common and rare species? 2) Does the mean distance among sites occupied by rare species differ from that for common species? 3) Is species frequency (species occurrences/sites in a data set) related to distance among occupied sites? 4) Does the classification of an individual species as common, rare, or vagrant vary with location? We found that 1) spatial distribution patterns were generally the same for common and rare species, 2) mean distances among stream sites typically did not differ between common and rare species, 3) species frequency was rarely related to distance among occupied sites, and 4) species classified as common, rare, or vagrant in one ecoregion data set were often classified differently in other data sets, indicating the importance of location in determining the status of a species. The results suggest that rare species have dispersal abilities similar to those of common species.

Identifying hotspots for rare species under climate change scenarios: improving saproxylic beetle conservation in Italy

Our aim is to model rare species (with few occurrences) but modelling the distribution of species with few occurrence data and many predictor variables leads to model overfitting. Thus, we use the recently developed ensemble of small models, which showed high predictive accuracy in modelling the distribution of rare species to estimate the current and future distribution of 56 rare (and endangered) saproxylic beetle species. Thus, we stacked predictions from individual species distribution models to derive rare species richness. We used current and five future general circulation models and three representative concentration pathways to test whether the distribution of hotspots for rare species shifts due to climate change under different future scenarios. Moreover, we verified the representativeness of existing protected area systems under future climate conditions in Italy. Specifically, we identified potential hotspots for rare species richness through a cumulative relative frequency distribution function. The current surface covered by hotspots is 50.4% of the study area corresponding to 151,223 km2 (mainly from central to northern Italy). Currently, only 35,124 km2 of rare saproxylic hotspots are covered by protected areas (PAs) and they will decrease by about 2–72% in 2070 depending on the future scenarios considered. Our results confirmed that the shift of the distribution of hotspots for rare species might occur due to climate change under different future scenarios and that the existing PAs system would be inadequate for assuring the conservation of rare saproxylic beetles in Italy under current and future climate conditions.

A Statistical Comparison between Less and Common Applied Models to Estimate Geographical Distribution of Endangered Species (Felis margarita) in Central Iran

Species distribution in space is important in habitat conservation and biodiversity protection, so gaining knowledge about species range would be worthwhile to rescue endangered species and plan conservation policy. This study evaluates and compares the performance of an array of Species Distribution Models (SDMs), namely RF, SVM, MaxEnt, GLMNET, and MARS, in predicting rare sand cat distribution across a large unprotected sand dune area in central Iran. Due to absence of reliable data and difficulties in recording the species itself, the SDMs were challenged by limited data including 55 absence (background) and 40 presence points as well as nine climatic and geological parameters that influence on species distribution, including humidity, maximum, minimum and mean temperature, precipitation, amount of sunshine, ground water level, aspect, and DEM. Moreover, each model was replicated 20 times and the statistics including TSS, AUC, COR and Deviance were computed. Then, based on computed statistics, the model performances were evaluated by TUKEY and ANOVA. Finally, ensemble map was obtained by weighted approach using AUC. The results of this study showed that complex machine learning methods, like SVM, RF, and MaxEnt are more outperformed to predict the distribution of rare species.

К морфологии, таксономии и распространению редких видов рода Sellaphora Mereschkovsky (Bacillariophyta) в России

К морфологии, таксономии и распространению редких видов рода Sellaphora Mereschkovsky (Bacillariophyta) в России