Assessing the effects of pseudo-absences on predictive distribution model performance

Modeling a hot, dry future: Substantial range reductions in suitable environment projected under climate change for a semiarid riparian predator guild

An understanding of species-environmental relationships is invaluable for effective conservation and management under anthropogenic climate change, especially for biodiversity hotspots such as riparian habitats. Species distribution models (SDMs) assess present species-environmental relationships which can project potential suitable environments through space and time. An understanding of environmental factors associated with distributions can guide conservation management strategies under a changing climate. We generated 260 ensemble SDMs for five species of Thamnophis gartersnakes (n = 347)-an important riparian predator guild-in a semiarid and biogeographically diverse region under impact from climate change (Arizona, United States). We modeled present species-environmental relationships and projected changes to suitable environment under 12 future climate scenarios per species, including the most and least optimistic greenhouse gas emission pathways, through 2100. We found that Thamnophis likely advanced northward since the turn of the 20th century and overwinter temperature and seasonal precipitation best explained present distributions. Future ranges of suitable environment for Thamnophis are projected to decrease by ca. -37.1% on average. We found that species already threatened with extinction or those with warm trailing-edge populations likely face the greatest loss of suitable environment, including near or complete loss of suitable environment. Future climate scenarios suggest an upward advance of suitable environment around montane areas for some low to mid-elevation species, which may create pressures to ascend. The most suitable environmental areas projected here can be used to identify potential safe zones to prioritize conservation refuges, including applicable critical habitat designations. By bounding the climate pathway extremes to, we reduce SDM uncertainties and provide valuable information to help conservation practitioners mitigate climate-induced threats to species. Implementing informed conservation actions is paramount for sustaining biodiversity in important aridland riparian systems as the climate warms and dries.

Free-ranging dogs are seriously threatening Himalayan environment: delineating the high-risk areas for curbing free-ranging dog infestation in the Trans-Himalayan region

It is becoming more widely recognised that free-ranging dogs, which have a nearly global distribution, threatening native wildlife. Their increasing population and spread to new areas is of growing concern for the long-term viability of wildlife species. Hence, it is imperative to understand the factors responsible for their infestation and map areas where native species are most vulnerable. Using the random forests algorithm, we modelled the free-ranging dog infestation in the Trans-Himalayan region to pinpoint the high-risk areas where free-ranging dogs are threatening the native wildlife species. We found that the likelihood of free-ranging dog occurrence is most in valley regions and up to 4000 m, often in proximity to roads. Our results also indicated that free-ranging dog prefers areas with wildlife near to protected areas. The predictor variables, such as potential evapotranspiration of the coldest quarter, distance to protected areas, elevation, distance to roads, and potential evapotranspiration of the driest quarter, significantly influence the distribution of the free-ranging dogs. We found that within the Ladakh region of the Trans-Himalayan area, the high-risk zones for free-ranging dogs are located in and around Hemis National Park, Karakoram Wildlife Sanctuary, and Changthang Wildlife Sanctuary. While, in the Lahaul and Spiti region the high-risk areas encompass Pin Valley National Park, Inderkilla National Park, Khirganga National Park, Kugti Wildlife Sanctuary, and several other protected areas. We identified the potentially high-risk areas for implementing strategies to mitigate the possible impact of free-ranging dogs on native wildlife of the Himalayas. Hence, the identified high priority areas can be used for implementing actions for controlling the population growth and further preventing the infestation of the free-ranging dogs into the new areas.

Comparing the Performance of CMCC-BioClimInd and WorldClim Datasets in Predicting Global Invasive Plant Distributions

Species distribution modeling (SDM) has been widely used to predict the distribution of invasive plant species based on bioclimatic variables. However, the specific selection of these variables may affect the performance of SDM. This investigation elucidates a new bioclimate variable dataset (i.e., CMCC-BioClimInd) for its use in SDM. The predictive performance of SDM that includes WorldClim and CMCC-BioClimInd was evaluated by AUC and omission rate and the explanatory power of both datasets was assessed by the jackknife method. Furthermore, the ODMAP protocol was used to record CMCC-BioClimInd to ensure reproducibility. The results indicated that CMCC-BioClimInd effectively simulates invasive plant species' distribution. Based on the contribution rate of CMCC-BioClimInd to the distribution of invasive plant species, it was inferred that the modified and simplified continentality and Kira warmth index from CMCC-BioClimInd had a strong explanatory power. Under the 35 bioclimatic variables of CMCC-BioClimInd, alien invasive plant species are mainly distributed in equatorial, tropical, and subtropical regions. We tested a new bioclimate variable dataset to simulate the distribution of invasive plant species worldwide. This method has great potential to improve the efficiency of species distribution modeling, thereby providing a new perspective for risk assessment and management of global invasive plant species.

Increase in the extent of mass coral bleaching over the past half-century, based on an updated global database

The recurrence of mass coral bleaching and associated coral mortality in the past few decades have raised questions about the future of coral reef ecosystems. Although coral bleaching is well studied, our understanding of the spatial extent of bleaching events continues to be limited by geographical biases in data collection. To address this gap, we updated a previous observational database and spatially modelled the probability of past bleaching occurrence. First, an existing raw observational database was updated to cover the 1963-2017 period using searches of the academic and grey literature and outreach to coral reef monitoring organizations. Then, in order to provide spatially-explicit global coverage, we employed indicator kriging to spatially model the probability of bleaching occurrence each year from 1985 through 2017 at 0.05° x 0.05° lat-long resolution. The updated raw database has 37,774 observations, including 22,650 positive bleaching reports, three times that in the previous version. The spatial interpolation suggests that 71% of the world's coral reefs likely (>66% probability) experienced bleaching at least once during the 1985 and 2017 period. The mean probability of bleaching across all reefs globally was 29-45% in the most severe bleaching years of 1998, 2005, 2010 and 2016. Modelled bleaching probabilities were positively related with annual maximum Degree Heating Weeks (DHW), a measure of thermal stress, across all years (p<0.001), and in each global bleaching event (p<0.01). In addition, the annual maximum DHW of reef cells that very likely (>90% probability) experienced bleaching increased over time at three times the rate of all reef cells, suggesting a possible increase in reef thermal tolerance. The raw and spatially interpolated databases can be used by other researchers to enhance real-time predictions, calibrate models for future projections, and assess the change in coral reef response to thermal stress over time.

Quantifying the relative contributions of habitat modification and mammalian predators on landscape-scale declines of a threatened river specialist duck

Habitat modification and introduced mammalian predators are linked to global species extinctions and declines, but their relative influences can be uncertain, often making conservation management difficult. Using landscape-scale models, we quantified the relative impacts of habitat modification and mammalian predation on the range contraction of a threatened New Zealand riverine duck. We combined 38 years of whio (Hymenolaimus malacorhynchos) observations with national-scale environmental data to predict relative likelihood of occurrence (RLO) under two scenarios using bootstrapped boosted regression trees (BRT). Our models used training data from contemporary environments to predict the potential contemporary whio distribution across New Zealand riverscapes in the absence of introduced mammalian predators. Then, using estimates of environments prior to human arrival, we used the same models to hindcast potential pre-human whio distribution prior to widespread land clearance. Comparing RLO differences between potential pre-human, potential contemporary and observed contemporary distributions allowed us to assess the relative impacts of the two main drivers of decline; habitat modification and mammalian predation. Whio have undergone widespread catastrophic declines most likely linked to mammalian predation, with smaller declines due to habitat modification (range contractions of 95% and 37%, respectively). We also identified areas of potential contemporary habitat outside their current range that would be suitable for whio conservation if mammalian predator control could be implemented. Our approach presents a practical technique for estimating the relative importance of global change drivers in species declines and extinctions, as well as providing valuable information to improve conservation planning.

Global Warming and Long-Distance Spread of Invasive Discoglossus pictus (Amphibia, Alytidae): Conservation Implications for Protected Amphibians in the Iberian Peninsula

Discoglossus pictus is a North African amphibian that was introduced in southern France early the 20th century and has spread south and north along the Mediterranean coastal plains up to 170 km. In order to disentangle the conservation implications of the spread of D. pictus for sensitive native species, we examined the impact of long-term climate warming on the basis of niche overlap analysis, taking into account abiotic factors. The study area covered the distribution ranges of all genus Discoglossus species in northwestern Africa (659,784 km²), Sicily (27,711 km²), the Iberian Peninsula, and southern France (699,546 km²). Niche overlap was measured from species environmental spaces extracted via PCA, including climate and relief environmental variables. Current and future climatic suitability for each species was assessed in an ensemble-forecasting framework of species distribution models, built using contemporary species data and climate predictors and projected to 2070's climatic conditions. Our results show a strong climatic niche overlap between D. pictus and native and endemic species in the Iberian Peninsula. In this context, all species will experience an increase in climatic suitability over the next decades, with the only exception being Pelodytes punctatus, which could be negatively affected by synergies between global warming and cohabitation with D. pictus.

Forest tree species distribution for Europe 2000-2020: mapping potential and realized distributions using spatiotemporal machine learning

This article describes a data-driven framework based on spatiotemporal machine learning to produce distribution maps for 16 tree species (Abies alba Mill., Castanea sativa Mill., Corylus avellana L., Fagus sylvatica L., Olea europaea L., Picea abies L. H. Karst., Pinus halepensis Mill., Pinus nigra J. F. Arnold, Pinus pinea L., Pinus sylvestris L., Prunus avium L., Quercus cerris L., Quercus ilex L., Quercus robur L., Quercus suber L. and Salix caprea L.) at high spatial resolution (30 m). Tree occurrence data for a total of three million of points was used to train different algorithms: random forest, gradient-boosted trees, generalized linear models, k-nearest neighbors, CART and an artificial neural network. A stack of 305 coarse and high resolution covariates representing spectral reflectance, different biophysical conditions and biotic competition was used as predictors for realized distributions, while potential distribution was modelled with environmental predictors only. Logloss and computing time were used to select the three best algorithms to tune and train an ensemble model based on stacking with a logistic regressor as a meta-learner. An ensemble model was trained for each species: probability and model uncertainty maps of realized distribution were produced for each species using a time window of 4 years for a total of six distribution maps per species, while for potential distributions only one map per species was produced. Results of spatial cross validation show that the ensemble model consistently outperformed or performed as good as the best individual model in both potential and realized distribution tasks, with potential distribution models achieving higher predictive performances (TSS = 0.898, R2 logloss = 0.857) than realized distribution ones on average (TSS = 0.874, R2 logloss = 0.839). Ensemble models for Q. suber achieved the best performances in both potential (TSS = 0.968, R2 logloss = 0.952) and realized (TSS = 0.959, R2 logloss = 0.949) distribution, while P. sylvestris (TSS = 0.731, 0.785, R2 logloss = 0.585, 0.670, respectively, for potential and realized distribution) and P. nigra (TSS = 0.658, 0.686, R2 logloss = 0.623, 0.664) achieved the worst. Importance of predictor variables differed across species and models, with the green band for summer and the Normalized Difference Vegetation Index (NDVI) for fall for realized distribution and the diffuse irradiation and precipitation of the driest quarter (BIO17) being the most frequent and important for potential distribution. On average, fine-resolution models outperformed coarse resolution models (250 m) for realized distribution (TSS = +6.5%, R2 logloss = +7.5%). The framework shows how combining continuous and consistent Earth Observation time series data with state of the art machine learning can be used to derive dynamic distribution maps. The produced predictions can be used to quantify temporal trends of potential forest degradation and species composition change.

An integrated high-resolution mapping shows congruent biodiversity patterns of Fagales and Pinales

The documentation of biodiversity distribution through species range identification is crucial for macroecology, biogeography, conservation, and restoration. However, for plants, species range maps remain scarce and often inaccurate. We present a novel approach to map species ranges at a global scale, integrating polygon mapping and species distribution modelling (SDM). We develop a polygon mapping algorithm by considering distances and nestedness of occurrences. We further apply an SDM approach considering multiple modelling algorithms, complexity levels, and pseudo-absence selections to map the species at a high spatial resolution and intersect it with the generated polygons. We use this approach to construct range maps for all 1957 species of Fagales and Pinales with data compilated from multiple sources. We construct high-resolution global species richness maps of these important plant clades, and document diversity hotspots for both clades in southern and south-western China, Central America, and Borneo. We validate the approach with two representative genera, Quercus and Pinus, using previously published coarser range maps, and find good agreement. By efficiently producing high-resolution range maps, our mapping approach offers a new tool in the field of macroecology for studying global species distribution patterns and supporting ongoing conservation efforts.

A comparison in species distribution model performance of succulents using key species and subsets of environmental predictors

Identifying the environmental drivers of the global distribution of succulent plants using the Crassulacean acid metabolism pathway of photosynthesis has previously been investigated through ensemble‐modeling of species delimiting the realized niche of the natural succulent biome. An alternative approach, which may provide further insight into the fundamental niche of succulent plants in the absence of dispersal limitation, is to model the distribution of selected species that are globally widespread and have become naturalized far beyond their native habitats. This could be of interest, for example, in defining areas that may be suitable for cultivation of alternative crops resilient to future climate change. We therefore explored the performance of climate‐only species distribution models (SDMs) in predicting the drivers and distribution of two widespread CAM plants, Opuntia ficus‐indica and Euphorbia tirucalli. Using two different algorithms and five predictor sets, we created distribution models for these exemplar species and produced an updated map of global inter‐annual rainfall predictability. No single predictor set produced markedly more accurate models, with the basic bioclim‐only predictor set marginally out‐performing combinations with additional predictors. Minimum temperature of the coldest month was the single most important variable in determining spatial distribution, but additional predictors such as precipitation and inter‐annual precipitation variability were also important in explaining the differences in spatial predictions between SDMs. When compared against previous projections, an a posteriori approach correctly does not predict distributions in areas of ecophysiological tolerance yet known absence (e.g., due to biotic competition). An updated map of inter‐annual rainfall predictability has successfully identified regions known to be depauperate in succulent plants. High model performance metrics suggest that the majority of potentially suitable regions for these species are predicted by these models with a limited number of climate predictors, and there is no benefit in expanding model complexity and increasing the potential for overfitting.

Assessing the Invasion Risk of Humulus scandens Using Ensemble Species Distribution Modeling and Habitat Connectivity Analysis

Given the rapid spread of invasive alien plant species in Europe and limited information regarding their distribution and dispersion patterns, we analyzed the invasive risk of Humulus scandens, a species with an increased invasive potential. We collected occurrence records from Romania within an EU funded project and literature data, in order to perform an ensemble distribution model. Environmental variables varied from downscaled topoclimatic continuous entries to categorical ones, such as soil class, texture, or land use. Results showed potential core areas of the species within the study region. By inverting the probability output of the models, we have created a resistance surface which helped us model its dispersion patterns. Further, we assessed the probability of invasion for each resulted corridor using the species dispersion ecology and created an invasion risk map. H. scandens is highly influenced by milder climates and areas with constant flooding events, thus we found that the Tisa basin and its tributaries can be under a high invasion risk, spreading through the entire catchment, in Central, Western, and Northern Romania, towards the Eastern Carpathians. The Danube acted as a dispersion corridor for major river systems in southern Romania, but the dispersion capability of the species dropped in steppe areas with higher aridity and limited water course network. This approach is useful for creating adequate action plans in relation to invasive alien plant species, and should urgently be regarded, as results show a potentially large distribution of H. scandens across entire water catchment areas, with devastating effects on natural ecosystems.

Phylogenetic relationships and evolutionary patterns of the genus Psammolestes Bergroth, 1911 (Hemiptera: Reduviidae: Triatominae)

BACKGROUND

The evolutionary history of biodiversity in South America has been poorly studied in the seasonal dry tropical forest (SDTF). Species diversification in this ecosystem may have a twofold explanation. First, intermittent connections in the middle and late Pleistocene promoted species dispersal and/or genetic connectivity between lineages isolated in disjunct patches of forest. Second, allopatric speciation proceeded immediately after the formation and colonization of the SDTF in the Neogene. Here we studied the diversification of Psammolestes, a genus endemic of the SDTF and naturally infected with Trypanosoma cruzi (agent of Chagas disease), using a combination of phylogenetic, population genetics and niche model methods, and evaluated the reliability of the three morphospecies currently recognized.

RESULTS

Our multilocus analyses recovered P. coreodes and P. tertius in a monophyletic clade sister to P. arthuri. Species delimitation tests recovered these lineages as different species despite the shared genetic variation observed between P. coreodes and P. tertius in five genes. Also, genetic variation of the genus clustered in three groups that were consistent with the three morphospecies. Our demographic model predicted a scenario of divergence in absence of gene flow, suggesting that mixed haplotypes may be the result of shared ancestral variation since the divergence of the subtropical-temperate species P. coreodes and P. tertius. In contrast, the tropical species P. arthuri was highly differentiated from the other two in all tests of genetic structure, and consistently, the Monmonier's algorithm identified a clear geographical barrier that separates this species from P. coreodes and P. tertius.

CONCLUSIONS

We found three genetically structured lineages within Psammolestes that diverged in absence of gene flow in the late Miocene. This result supports a scenario of species formation driven by geographical isolation rather than by divergence in the face of gene flow associated with climatic oscillations in the Pleistocene. Also, we identified the Amazon basin as a climatic barrier that separates tropical from subtropical-temperate species, thus promoting allopatric speciation after long range dispersion. Finally, each species of Psammolestes occupies different climatic niches suggesting that niche conservatism is not crucial for species differentiation. These findings influence the current vector surveillance programs of Chagas disease in the region.

Dynamic Forecast of Desert Locust Presence Using Machine Learning with a Multivariate Time Lag Sliding Window Technique

Desert locust plagues can easily cause a regional food crisis and thus affect social stability. Preventive control of the disaster highlights the early detection of hopper gregarization before they form devastating swarms. However, the response of hopper band emergence to environmental fluctuation exhibits a time lag. To realize the dynamic forecast of band occurrence with optimal temporal predictors, we proposed an SVM-based model with a temporal sliding window technique by coupling multisource time-series imagery with historical locust ground survey observations from between 2000–2020. The sliding window method was based on a lagging variable importance ranking used to analyze the temporal organization of environmental indicators in band-forming sequences and eventually facilitate the early prediction of band emergence. Statistical results show that hopper bands are more likely to occur within 41–64 days after increased rainfall; soil moisture dynamics increasing by approximately 0.05 m³/m³ then decreasing may enhance the chance of observing bands after 73–80 days. While sparse vegetation areas with NDVI increasing from 0.18 to 0.25 tend to witness bands after 17–40 days. The forecast model combining the optimal time lags of these dynamic indicators with other static indicators allows for a 16-day extended outlook of band presence in Somalia, Ethiopia, and Kenya. Monthly predictions from February to December 2020 display an overall accuracy of 77.46%, with an average ROC-AUC of 0.767 and a mean F-score close to 0.772. The multivariate forecast framework based on the lagging effect can realize the early warning of band presence in different spatiotemporal scenarios, supporting early decisions and response strategies for desert locust preventive management.

Response of an Afro-Palearctic bird migrant to glaciation cycles

We combine tracks of a long-distance migratory bird with high–temporal resolution climate data to reconstruct habitat availability month by month for the past 120,000 y. The seasonal changes of suitable habitat in the past imply that continued seasonal migration was necessary during the glacial maxima. Genomic-based estimates of effective population size indicate that more generally migratory lifestyles can be beneficially adapted to various climatic conditions. Our results provide a major step forward in understanding how migratory species will fare in the future and have important implications for how we understand the role of migration in the distribution of species and potentially speciation.

Migration allows animals to exploit spatially separated and seasonally available resources at a continental to global scale. However, responding to global climatic changes might prove challenging, especially for long-distance intercontinental migrants. During glacial periods, when conditions became too harsh for breeding in the north, avian migrants have been hypothesized to retract their distribution to reside within small refugial areas. Here, we present data showing that an Afro-Palearctic migrant continued seasonal migration, largely within Africa, during previous glacial–interglacial cycles with no obvious impact on population size. Using individual migratory track data to hindcast monthly bioclimatic habitat availability maps through the last 120,000 y, we show altered seasonal use of suitable areas through time. Independently derived effective population sizes indicate a growing population through the last 40,000 y. We conclude that the migratory lifestyle enabled adaptation to shifting climate conditions. This indicates that populations of resource-tracking, long-distance migratory species could expand successfully during warming periods in the past, which could also be the case under future climate scenarios.

Evaluating the Effects of Climate Change on Spatial Aggregation of Giant Pandas and Sympatric Species in a Mountainous Landscape

Understanding how climate change alters the spatial aggregation of sympatric species is important for biodiversity conservation. Previous studies usually focused on spatial shifting of species but paid little attention to changes in interspecific competitions under climate change. In this study, we evaluated the potential effects of climate change on the spatial aggregation of giant pandas (Ailuropoda melanoleuca) and three sympatric competitive species (i.e., black bears (Ursus thibetanus), golden takins (Budorcas taxicolor), and wild boars (Sus scrofa)) in the Qinling Mountains, China. We employed an ensemble species distribution modeling (SDM) approach to map the current spatial distributions of giant pandas and sympatric animals and projected them to future climate scenarios in 2050s and 2070s. We then examined the range overlapping and niche similarities of these species under different climate change scenarios. The results showed that the distribution areas of giant pandas and sympatric species would decrease remarkably under future climate changes. The shifting directions of the overlapping between giant pandas and sympatric species vary under different climate change scenarios. In conclusion, future climate change greatly shapes the spatial overlapping pattern of giant pandas and sympatric species in the Qinling Mountains, while interspecific competition would be intensified under both mild and worst-case climate change scenarios.

Ensemble ecological niche modeling of West Nile virus probability in Florida

Ecological Niche Modeling is a process by which spatiotemporal, climatic, and environmental data are analyzed to predict the distribution of an organism. Using this process, an ensemble ecological niche model for West Nile virus habitat prediction in the state of Florida was developed. This model was created through the weighted averaging of three separate machine learning models—boosted regression tree, random forest, and maximum entropy—developed for this study using sentinel chicken surveillance and remote sensing data. Variable importance differed among the models. The highest variable permutation value included mean dewpoint temperature for the boosted regression tree model, mean temperature for the random forest model, and wetlands focal statistics for the maximum entropy mode. Model validation resulted in area under the receiver curve predictive values ranging from good [0.8728 (95% CI 0.8422–0.8986)] for the maximum entropy model to excellent [0.9996 (95% CI 0.9988–1.0000)] for random forest model, with the ensemble model predictive value also in the excellent range [0.9939 (95% CI 0.9800–0.9979]. This model should allow mosquito control districts to optimize West Nile virus surveillance, improving detection and allowing for a faster, targeted response to reduce West Nile virus transmission potential.

Biogeography Meets Niche Modeling: Inferring the Role of Deep Time Climate Change When Data Is Limited

Geographic range shifts are one major organism response to climate change, especially if the rate of climate change is higher than that of species adaptation. Ecological niche models (ENM) and biogeographic inferences are often used in estimating the effects of climatic oscillations on species range dynamics. ENMs can be used to track climatic suitable areas over time, but have often been limited to shallow timescales; biogeographic inference can reach greater evolutionary depth, but often lacks spatial resolution. Here, we present a simple approach that treats them as independent and complementary sources of evidence, which, when used in partnership, can be employed to reconstruct geographic range shifts over deep evolutionary timescales. For testing this, we chose two extreme African disjunctions: Camptoloma (Scrophulariaceae) and Canarina (Campanulaceae), each comprising of three species disjunctly distributed in Macaronesia and eastern/southern Africa. Using inferred ancestral ranges in tandem with preindustrial and paleoclimate ENM hindcastings, we show that the disjunct pattern was the result of fragmentation and extinction events linked to Neogene aridification cycles. Our results highlight the importance of considering temporal resolution when building ENMs for rare endemics with small population sizes and restricted climatic tolerances such as Camptoloma, for which models built on averaged monthly variables were more informative than those based on annual bioclimatic variables. Additionally, we show that biogeographic information can be used as truncation threshold criteria for building ENMs in the distant past. Our approach is suitable when there is sparse sampling on species occurrences and associated patterns of genetic variation, such as in the case of ancient endemics with widely disjunct distributions as a result of climate change.

Integrating demography and distribution modeling for the iconic Leontopodium alpinum Colm. in the Romanian Carpathians

Both climate change and human exploitation are major threats to plant life in mountain environments. One species that may be particularly sensitive to both of these stressors is the iconic alpine flower edelweiss (Leontopodium alpinum Colm.). Its populations have declined across Europe due to over-collection for its highly prized flowers. Edelweiss is still subject to harvesting across the Romanian Carpathians, but no study has measured to what extent populations are vulnerable to anthropogenic change.Here, we estimated the effects of climate and human disturbance on the fitness of edelweiss. We combined demographic measurements with predictions of future range distribution under climate change to assess the viability of populations across Romania.We found that per capita and per-area seed number and seed mass were similarly promoted by both favorable environmental conditions, represented by rugged landscapes with relatively cold winters and wet summers, and reduced exposure to harvesting, represented by the distance of plants from hiking trails. Modeling these responses under future climate scenarios suggested a slight increase in per-area fitness. However, we found plant ranges contracted by between 14% and 35% by 2050, with plants pushed into high elevation sites.Synthesis. Both total seed number and seed mass are expected to decline across Romania despite individual edelweiss fitness benefiting from a warmer and wetter climate. More generally, our approach of coupling species distribution models with demographic measurements may better inform conservation strategies of ways to protect alpine life in a changing world.

Pneumonia Detection Using an Improved Algorithm Based on Faster R-CNN

Pneumonia remains a threat to human health; the coronavirus disease 2019 (COVID-19) that began at the end of 2019 had a major impact on the world. It is still raging in many countries and has caused great losses to people's lives and property. In this paper, we present a method based on DeepConv-DilatedNet of identifying and localizing pneumonia in chest X-ray (CXR) images. Two-stage detector Faster R-CNN is adopted as the structure of a network. Feature Pyramid Network (FPN) is integrated into the residual neural network of a dilated bottleneck so that the deep features are expanded to preserve the deep feature and position information of the object. In the case of DeepConv-DilatedNet, the deconvolution network is used to restore high-level feature maps into its original size, and the target information is further retained. On the other hand, DeepConv-DilatedNet uses a popular fully convolution architecture with computation shared on the entire image. Then, Soft-NMS is used to screen boxes and ensure sample quality. Also, K-Means++ is used to generate anchor boxes to improve the localization accuracy. The algorithm obtained 39.23% Mean Average Precision (mAP) on the X-ray image dataset from the Radiological Society of North America (RSNA) and got 38.02% Mean Average Precision (mAP) on the ChestX-ray14 dataset, surpassing other detection algorithms. So, in this paper, an improved algorithm that can provide doctors with location information of pneumonia lesions is proposed.

A Robust Prediction Model for Species Distribution Using Bagging Ensembles with Deep Neural Networks

Species distribution models have been used for various purposes, such as conserving species, discovering potential habitats, and obtaining evolutionary insights by predicting species occurrence. Many statistical and machine-learning-based approaches have been proposed to construct effective species distribution models, but with limited success due to spatial biases in presences and imbalanced presence-absences. We propose a novel species distribution model to address these problems based on bootstrap aggregating (bagging) ensembles of deep neural networks (DNNs). We first generate bootstraps considering presence-absence data on spatial balance to alleviate the bias problem. Then we construct DNNs using environmental data from presence and absence locations, and finally combine these into an ensemble model using three voting methods to improve prediction accuracy. Extensive experiments verified the proposed model’s effectiveness for species in South Korea using crowdsourced observations that have spatial biases. The proposed model achieved more accurate and robust prediction results than the current best practice models.

Genetic data improves niche model discrimination and alters the direction and magnitude of climate change forecasts

Ecological niche models (ENMs) have classically operated under the simplifying assumptions that there are no barriers to gene flow, species are genetically homogeneous (i.e., no population-specific local adaptation), and all individuals share the same niche. Yet, these assumptions are violated for most broadly distributed species. Here, we incorporate genetic data from the widespread riparian tree species narrowleaf cottonwood (Populus angustifolia) to examine whether including intraspecific genetic variation can alter model performance and predictions of climate change impacts. We found that (1) P. angustifolia is differentiated into six genetic groups across its range from México to Canada and (2) different populations occupy distinct climate niches representing unique ecotypes. Comparing model discriminatory power, (3) all genetically informed ecological niche models (gENMs) outperformed the standard species-level ENM (3-14% increase in AUC; 1-23% increase in pROC). Furthermore, (4) gENMs predicted large differences among ecotypes in both the direction and magnitude of responses to climate change and (5) revealed evidence of niche divergence, particularly for the Eastern Rocky Mountain ecotype. (6) Models also predicted progressively increasing fragmentation and decreasing overlap between ecotypes. Contact zones are often hotspots of diversity that are critical for supporting species' capacity to respond to present and future climate change, thus predicted reductions in connectivity among ecotypes is of conservation concern. We further examined the generality of our findings by comparing our model developed for a higher elevation Rocky Mountain species with a related desert riparian cottonwood, P. fremontii. Together our results suggest that incorporating intraspecific genetic information can improve model performance by addressing this important source of variance. gENMs bring an evolutionary perspective to niche modeling and provide a truly "adaptive management" approach to support conservation genetic management of species facing global change.

Where did they not go? Considerations for generating pseudo-absences for telemetry-based habitat models

BACKGROUND

Habitat suitability models give insight into the ecological drivers of species distributions and are increasingly common in management and conservation planning. Telemetry data can be used in habitat models to describe where animals were present, however this requires the use of presence-only modeling approaches or the generation of 'pseudo-absences' to simulate locations where animals did not go. To highlight considerations for generating pseudo-absences for telemetry-based habitat models, we explored how different methods of pseudo-absence generation affect model performance across species' movement strategies, model types, and environments.

METHODS

We built habitat models for marine and terrestrial case studies, Northeast Pacific blue whales (Balaenoptera musculus) and African elephants (Loxodonta africana). We tested four pseudo-absence generation methods commonly used in telemetry-based habitat models: (1) background sampling; (2) sampling within a buffer zone around presence locations; (3) correlated random walks beginning at the tag release location; (4) reverse correlated random walks beginning at the last tag location. Habitat models were built using generalised linear mixed models, generalised additive mixed models, and boosted regression trees.

RESULTS

We found that the separation in environmental niche space between presences and pseudo-absences was the single most important driver of model explanatory power and predictive skill. This result was consistent across marine and terrestrial habitats, two species with vastly different movement syndromes, and three different model types. The best-performing pseudo-absence method depended on which created the greatest environmental separation: background sampling for blue whales and reverse correlated random walks for elephants. However, despite the fact that models with greater environmental separation performed better according to traditional predictive skill metrics, they did not always produce biologically realistic spatial predictions relative to known distributions.

CONCLUSIONS

Habitat model performance may be positively biased in cases where pseudo-absences are sampled from environments that are dissimilar to presences. This emphasizes the need to carefully consider spatial extent of the sampling domain and environmental heterogeneity of pseudo-absence samples when developing habitat models, and highlights the importance of scrutinizing spatial predictions to ensure that habitat models are biologically realistic and fit for modeling objectives.

Modeling recreational fishing intensity in a complex urbanised estuary

Urbanised estuaries, ports and harbours are often utilised for recreational purposes, notably recreational angling. Yet there has been little quantitative assessment of the footprint and intensity of these activities at scales suitable for spatial management. Urban and industrialised estuaries have previously been considered as having low conservation value, perhaps due to issues with contamination and disturbance. Studies in recent decades have demonstrated that many of these systems are still highly biodiverse and of high value to local residents. As a response, urbanised estuaries are now being considered by coastal spatial management initiatives, where assessments of recreational use in these areas can help avoid 'user-environmental' and 'user-user' conflict. The models of these activities need to be developed at a scale relevant to governments and regulatory authorities, but the few human-use models that do exist integrate fishing intensity to a regional or even continental scale; too large to capture the fine scale variation inherent in complex urban fisheries. Species Distribution Modeling (SDM) is a tool commonly used to assess drivers of species range, but can be applied to models of recreational fishing in complex environments, at a scale relevant to regulatory bodies. Using point-data from 573 visual surveys with recently developed Poisson point process models, we examine the recreational fishery in Australia's busiest estuarine port, Sydney Harbour. We demonstrate the utility of these models for understanding the distribution of boat and shore-based fishers, and the effects of a range of temporally static (geographical) and dynamic (weather) predictors on these distributions.

Effects of climate change and land cover on the distributions of a critical tree family in the Philippines

Southeast Asian forests are dominated by the tree family Dipterocarpaceae, whose abundance and diversity are key to maintaining the structure and function of tropical forests. Like most biodiversity, dipterocarps are threatened by deforestation and climate change, so it is crucial to understand the potential impacts of these threats on current and future dipterocarp distributions. We developed species distribution models (SDMs) for 19 species of dipterocarps in the Philippines, which were projected onto current and two 2070 representative concentration pathway (RCP) climate scenarios, RCP 4.5 and 8.5. Current land cover was incorporated as a post-hoc correction to restrict projections onto intact habitats. Land cover correction alone reduced current species distributions by a median 67%, and within protected areas by 37%. After land cover correction, climate change reduced distributions by a median 16% (RCP 4.5) and 27% (RCP 8.5) at the national level, with similar losses in protected areas. There was a detectable upward elevation shift of species distributions, consisting of suitable habitat losses below 300 m and gains above 600 m. Species-rich stable areas of continued habitat suitability (i.e., climate macrorefugia) fell largely outside current delineations of protected areas, indicating a need to improve protected area planning. This study highlights how SDMs can provide projections that can inform protected area planning in the tropics.

Quantifying Global Potential Marginal Land Resources for Switchgrass

Switchgrass (Panicum virgatum L.) with its advantages of low maintenance and massive distribution in temperate zones, has long been regarded as a suitable biofuel feedstock with a promising prospect. Currently, there is no validated assessment of marginal land for switchgrass growth on a global scale. Although, on both regional and national scale there have been several studies evaluating the potential marginal lands for growing switchgrass. To obtain the first global map that presents the distribution of switchgrass growing in potential marginal land, we employed a boosted regression tree (BRT) modeling procedure integrated with released switchgrass records along with a series of high-spatial-resolution environmental variables. The result shows that the available marginal land resources satisfying switchgrass growing demands are mainly distributed in the southern and western parts of North America, coastal areas in the southern and eastern parts of South America, central and southern Africa, and northern Oceania, approximately 2229.80 million hectares. Validation reveals that the ensembled BRT models have a considerably high performance (area under the curve: 0.960). According to our analysis, annual cumulative precipitation accounts for 45.84% of the full impact on selecting marginal land resources for switchgrass, followed by land cover (14.97%), maximum annual temperature (12.51%), and mean solar radiation (10.25%). Our findings bring a new perspective on the development of biofuel feedstock.

Accessing habitat suitability and connectivity for the westernmost population of Asian black bear (Ursus thibetanus gedrosianus, Blanford, 1877) based on climate changes scenarios in Iran

Climate change, as an emerging phenomenon, has led to changes in the distribution, movement, and even risk of extinction of various wildlife species and this has raised concerns among conservation biologists. Different species have two options in the face of climate change, either to adopt or follow their climatic niche to new places through the connectivity of habitats. The modeling of interpatch landscape communications can serve as an effective decision support tool for wildlife managers. This study was conducted to assess the effects of climate change on the distribution and habitat connectivity of the endangered subspecies of Asian black bear (Ursus thibetanus gedrosianus) in the southern and southeastern Iran. The presence points of the species were collected in Provinces of Kerman, Hormozgan, and Sistan-Baluchestan. Habitat modeling was done by the Generalized Linear Model, and 3 machine learning models including Maximum Entropy, Back Propagation based artificial Neural Network, and Support Vector Machine. In order to achieve the ensemble model, the results of the mentioned models were merged based on the method of "accuracy rate as weight" derived from their validation. To construct pseudo-absence points for the use in the mentioned models, the Ensemble model of presence-only models was used. The modeling was performed using 15 habitat variables related to climatic, vegetation, topographic, and anthropogenic parameters. The three general circulation models of BCC-CSM1, CCSM4, and MRI-CGCM3 were selected under the two scenarios of RCP2.6 and RCP8.5 by 2070. To investigate the effect of climate change on the habitat connections, the protected areas of 3 provinces were considered as focal nodes and the connections between them were established based on electrical circuit theory and Pairwise method. The true skill statistic was employed to convert the continuous suitability layers to binary suitable/unsuitable range maps to assess the effectiveness of the protected areas in the coverage of suitable habitats for the species. Due to the high power of the stochastic forest model in determining the importance of variables, this method was used. The results showed that presence/absence models were successful in the implementation and well distinguished the points of presence and pseudo-absence from each other. Based on the random forests model, the variables of Precipitation of Driest Quarter, Precipitation of Coldest Quarter, and Temperature Annual Range have the greatest impact on the habitat suitability. Comparing the modeling findings to the realities of the species distribution range indicated that the suitable habitats are located in areas with high humidity and rainfall, which are mostly in the northern areas of Bandar Abbas, south of Kerman, and west and south of Sistan-Baluchestan. The area of suitable habitats, in the MRI-CGCM3 (189731 Km2) and CCSM4 (179007 Km2) models under the RCP2.6 scenario, is larger than the current distribution (174001 Km2). However, in terms of the performance of protected areas, the optimal coverage of the species by the boundary of the protected areas, under each of the RCP2.6 and RCP8.5 scenarios, is less than the present time. According to the electric circuit theory, connecting the populations in the protected areas of Sistan-Baluchestan province to those in the northern Hormozgan and the southern Kerman would be based on the crossing through the heights of Sistan-Baluchestan and Hormozgan provinces and the plains between these heights would be the movement pinch points under the current and future scenarios. Populations in the protected areas of Kerman have higher quality patch connections than that of the other two provinces. The areas such as Sang-e_Mes, Kouh_Shir, Zaryab, and Bahr_Aseman in Kerman Province and Kouhbaz and Geno in Hormozgan Province can provide suitable habitats for the species in the distribution models. The findings revealed that the conservation of the heights along with the caves inside them could be a protective priority to counteract the effects of climate change on the species.

Does weighting presence records improve the performance of species distribution models? A test using fish larval stages in the Yangtze Estuary

To obtain realistic forecasts of the impacts of climate change on species habitat suitability, novel approaches based on species distribution models (SDMs) are being developed and scrutinized. We argue here that, when dealing with data from long-term monitoring programmes, incorporating a temporal weight on the occurrence points may result in a more realistic prediction of a species' potential distribution. Using larval fish presence records collected from 1999 to 2013 in the Yangtze Estuary, China, we compared the performance of ensembles of standard SDMs versus SDMs constructed with weighted time-series presence records in predicting the present and future distributions of the larval stages of two dominant fish. The results of the ensemble SDMs showed that weighted presence records can significantly improve SDM performance, as measured through standard validation metrics. The SDM projections suggest that suitable habitat for both species will decrease under future climate scenarios, with one species (Stolephorus commersonnii) predicted to be more susceptible to climate change than the other (Engraulis japonicus). In addition to range contraction, model projections suggest that the future habitats of both species will shift northward-an implication of climate change that should be considered in future management and conservation strategies for the Yangtze Estuary.

Zeeshan Ali, Dasgpta N, Uniyal VP, Johnson JA, Hussain SA. Evaluating performance of four species distribution models using Blue-tailed Green Darner Anax guttatus (Insecta: Odonata) as model organism from the Gangetic riparian zone

In this paper we evaluated the performance of four species distribution models: generalized linear (GLM), maximum entropy (MAXENT), random forest (RF) and support vector machines (SVM) model, using the distribution of the dragonfly Blue-tailed Green Darner Anax guttatus in the Gangetic riparian zone between Bijnor and Kanpur barrage, Uttar Pradesh, India.  We used forest cover type, land use, land cover and five bioclimatic variable layers: annual mean temperature, isothermality, temperature seasonality, mean temperature of driest quarter, and precipitation seasonality to build the models.  We found that the GLM generated the highest values for AUC, Kappa statistic, TSS, specificity and sensitivity, and the lowest values for omission error and commission error, while the MAXENT model generated the lowest variance in variable importance. We suggest that researchers should not rely on any single algorithm, instead, they should test performance of all available models for their species and area of interest, and choose the best one to build a species distribution model.  

Climate-induced distribution dynamics of Plebeia flavocincta, a stingless bee from Brazilian tropical dry forests

AIM

The objective of this study is to estimate the current potential geographic distribution of Plebeia flavocincta and to evaluate the influence of climate on the dynamics of suitable habitat availability in the past and in the future.

LOCATION

Northeast region of Brazil and dry forest areas.

METHODS

The habitat suitability modeling was based on two algorithms, two global circulation models, and six different scenarios. We used this tool to estimate the areas of occurrence in the past (Last Interglacial and Last Glacial Maximum), in the present, and in the future (years 2050 and 2070).

RESULTS

According to the models, P. flavocincta had great dynamics in the availability of suitable habitats with periods of retraction and expansion of these areas in the past. Our results suggest that this taxon may benefit in terms of climate suitability gain in Northeast Brazil in the future. In addition, we identified high-altitude areas and the eastern coast as climatically stable.

CONCLUSION

The information provided can be used by decision makers to support actions toward protecting and sustainably managing this taxon. Protection measures for this taxon are particularly important because this insect contributes to the local flora and, although our results indicate that the climate may favor this taxon, other factors can negatively affect it, such as high levels of habitat loss due to anthropogenic activities.

Ecological niche models display nonlinear relationships with abundance and demographic performance across the latitudinal distribution of Astragalus utahensis (Fabaceae)

The potential for ecological niche models (ENMs) to accurately predict species' abundance and demographic performance throughout their geographic distributions remains a topic of substantial debate in ecology and biogeography. Few studies simultaneously examine the relationship between ENM predictions of environmental suitability and both a species' abundance and its demographic performance, particularly across its entire geographic distribution. Yet, studies of this type are essential for understanding the extent to which ENMs are a viable tool for identifying areas that may promote high abundance or performance of a species or how species might respond to future climate conditions. In this study, we used an ensemble ecological niche model to predict climatic suitability for the perennial forb Astragalus utahensis across its geographic distribution. We then examined relationships between projected climatic suitability and field-based measures of abundance, demographic performance, and forecasted stochastic population growth (λs). Predicted climatic suitability showed a J-shaped relationship with A. utahensis abundance, where low-abundance populations were associated with low-to-intermediate suitability scores and abundance increased sharply in areas of high predicted climatic suitability. A similar relationship existed between climatic suitability and λs from the center to the northern edge of the latitudinal distribution. Patterns such as these, where density or demographic performance only increases appreciably beyond some threshold of climatic suitability, support the contention that ENM-predicted climatic suitability does not necessarily represent a reliable predictor of abundance or performance across large geographic regions.

Improving prediction of rare species' distribution from community data

Species distribution models (SDMs) have been increasingly used to predict the geographic distribution of a wide range of organisms; however, relatively fewer research efforts have concentrated on rare species despite their critical roles in biological conservation. The present study tested whether community data may improve modelling rare species by sharing information among common and rare ones. We chose six SDMs that treat community data in different ways, including two traditional single-species models (random forest and artificial neural network) and four joint species distribution models that incorporate species associations implicitly (multivariate random forest and multi-response artificial neural network) or explicitly (hierarchical modelling of species communities and generalized joint attribute model). In addition, we evaluated two approaches of data arrangement, species filtering and conditional prediction, to enhance the selected models. The model predictions were tested using cross validation based on empirical data collected from marine fisheries surveys, and the effects of community data were evaluated by comparing models for six selected rare species. The results demonstrated that the community data improved the predictions of rare species' distributions to certain extent but might also be unhelpful in some cases. The rare species could be appropriately predicted in terms of occurrence, whereas their abundance tended to be underestimated by most models. Species filtering and conditional predictions substantially benefited the predictive performances of multiple- and single-species models, respectively. We conclude that both the modelling algorithms and community data need to be carefully selected in order to deliver improvement in modelling rare species. The study highlights the opportunity and challenges to improve prediction of rare species' distribution by making the most of community data.

Meta-replication, sampling bias, and multi-scale model selection: A case study on snow leopard (Panthera uncia) in western China

Replicated multiple scale species distribution models (SDMs) have become increasingly important to identify the correct variables determining species distribution and their influences on ecological responses. This study explores multi-scale habitat relationships of the snow leopard (Panthera uncia) in two study areas on the Qinghai-Tibetan Plateau of western China. Our primary objectives were to evaluate the degree to which snow leopard habitat relationships, expressed by predictors, scales of response, and magnitude of effects, were consistent across study areas or locally landcape-specific. We coupled univariate scale optimization and the maximum entropy algorithm to produce multivariate SDMs, inferring the relative suitability for the species by ensembling top performing models. We optimized the SDMs based on average omission rate across the top models and ensembles' overlap with a simulated reference model. Comparison of SDMs in the two study areas highlighted landscape-specific responses to limiting factors. These were dependent on the effects of the hydrological network, anthropogenic features, topographic complexity, and the heterogeneity of the landcover patch mosaic. Overall, even accounting for specific local differences, we found general landscape attributes associated with snow leopard ecological requirements, consisting of a positive association with uplands and ridges, aggregated low-contrast landscapes, and large extents of grassy and herbaceous vegetation. As a means to evaluate the performance of two bias correction methods, we explored their effects on three datasets showing a range of bias intensities. The performance of corrections depends on the bias intensity; however, density kernels offered a reliable correction strategy under all circumstances. This study reveals the multi-scale response of snow leopards to environmental attributes and confirms the role of meta-replicated study designs for the identification of spatially varying limiting factors. Furthermore, this study makes important contributions to the ongoing discussion about the best approaches for sampling bias correction.

Phylogenetics and population structure of the steppe species Hycleus polymorphus (Coleoptera: Meloidae: Mylabrini) reveal multiple refugia in Mediterranean mountain ranges

Many continental species distributed in the Eurasian steppe occur as relict populations in the mountains of Western Europe. Their biogeographical responses to Quaternary climate changes have been poorly studied; however, they could have responded as cold-adapted species. We investigated the biogeographic history of a steppe beetle, Hycleus polymorphus, using mitochondrial and nuclear DNA sequences (COI, CAD, ITS2), and species distribution modelling (SDM) under present and past bioclimatic envelopes. We first performed a phylogenetic assessment to define species boundaries within the H. polymorphus species group. Specimens previously treated as Hycleus humerosus on morphological grounds are assigned to H. polymorphus, and those identified as Hycleus zebraeus assigned to Hycleus atratus. ITS2 data analyses revealed a strong phylogeographical structure of H. polymorphus populations, with four haplogroups corresponding to the (i) Italian Alps, (ii) French Alps and Pyrenees, (iii) South Balkan and Pontic mountains, and (iv) North Dinaric Alps. Based on these analyses and the SDM, we propose that during a glacial period, following the spread of steppic habitat, H. polymorphus underwent a range expansion from Asia to South-West Europe. Within the Mediterranean area, during the last interglacial the climatic suitability for the species was limited to mountains that acted as refugia and prompted allopatric divergence into four main lineages.

Integrated Methods for Monitoring the Invasive Potential and Management of Heracleum mantegazzianum (giant hogweed) in Switzerland

Biological invasions are a major driver of human-induced global environmental change. This makes monitoring of potential spread, population changes and control measures necessary for guiding management. We illustrate the value of integrated methods (species distribution modelling (SDM), plant population monitoring and questionnaires) for monitoring and assessing invasions of Heracleum mantegazzianum (giant hogweed) over time in Switzerland. SDMs highlighted the potential spread of the species, uncovered ecological mechanisms underlying invasions and guided monitoring at a regional level. We used adaptive and repeat plant sampling to monitor invasive population status and changes, and assess the effectiveness of H. mantegazzianum management over three periods (2005, 2013 and 2018) within the pre-Alps, Vaud. We also conducted questionnaire surveys with managers and the public. Multiscale modelling, and integrating global and regional SDMs, provided the best predictions, showing that H. mantegazzianum can potentially invade large parts of Switzerland, especially below 2 000 m a.s.l. Over time, populations of invasive H. mantegazzianum in the Vaud pre-Alps have declined, which is most likely due to a sharp rise in management uptake post 2007 (7% of municipalities before 2007 to 86% in 2018). The level of known invasive populations has decreased by 54% over time. Some municipalities have even successfully eradicated H. mantegazzianum within their borders. However, a few areas, particularly in the rural, higher-altitude municipalities, where management was not implemented effectively, populations have expanded, which could hamper control efforts at lower altitudes. We provide encouraging evidence that control measures can be effective in reducing plant invasions with long-term commitment, as well as a good template for using integrated methodological approaches to better study and monitor invasive alien species.

Risk factors and predicted distribution of visceral leishmaniasis in the Xinjiang Uygur Autonomous Region, China, 2005-2015

Background

Visceral leishmaniasis (VL) is a neglected disease that is spread to humans by the bites of infected female phlebotomine sand flies. Although this vector-borne disease has been eliminated in most parts of China, it still poses a significant public health burden in the Xinjiang Uygur Autonomous Region. Understanding of the spatial epidemiology of the disease remains vague in the local community. In the present study, we investigated the spatiotemporal distribution of VL in the region in order to assess the potential threat of the disease.

Methods

Based on comprehensive infection records, the spatiotemporal patterns of new cases of VL in the region between 2005 and 2015 were analysed. By combining maps of environmental and socioeconomic correlates, the boosted regression tree (BRT) model was adopted to identify the environmental niche of VL.

Results

The fitted BRT models were used to map potential infection risk zones of VL in the Xinjiang Uygur Autonomous Region, revealing that the predicted high infection risk zones were mainly concentrated in central and northern Kashgar Prefecture, south of Atushi City bordering Kashgar Prefecture and regions of the northern Bayingolin Mongol Autonomous Prefecture. The final result revealed that approximately 16.64 million people inhabited the predicted potential infection risk areas in the region.

Conclusions

Our results provide a better understanding of the potential endemic foci of VL in the Xinjiang Uygur Autonomous Region with a 1 km spatial resolution, thereby enhancing our capacity to target the potential risk areas, to develop disease control strategies and to allocate medical supplies.

The use of classification and regression algorithms using the random forests method with presence-only data to model species' distribution

Random forests (RF) is a powerful species distribution model (SDM) algorithm. This ensemble model by default can produce categorical and numerical species distribution maps based on its classification tree (CT) and regression tree (RT) algorithms, respectively. The CT algorithm can also produce numerical predictions (class probability). Here, we present a detailed procedure involving the use of the CT and RT algorithms using the RF method with presence-only data to model the distribution of species. CT and RT are used to generate numerical prediction maps, and then numerical predictions are converted to binary predictions through objective threshold-setting methods. We also applied simple methods to deal with collinearity of predictor variables and spatial autocorrelation of species occurrence data. A geographically stratified sampling method was employed for generating pseudo-absences. The detailed procedural framework is meant to be a generic method to be applied to virtually any SDM prediction question using presence-only data.

•

How to use RF as a standard method for generic species distributions with presence-only data

•

How to choose RF (CT or RT) methods for the distribution modeling of species

•

A general and detailed procedure for any SDM prediction question.

Climate change impact on ecosystem functions provided by birds in southeastern Amazonia

Although the impacts of climate change on biodiversity are increasing worldwide, few studies have attempted to forecast these impacts on Amazon Tropical Forest. In this study, we estimated the impact of climate change on Amazonian avian assemblages considering range shifts, species loss, vulnerability of ecosystem functioning, future effectiveness of current protected areas and potential climatically stable areas for conservation actions. Species distribution modelling based on two algorithms and three different scenarios of climate change was used to forecast 501 avian species, organized on main ecosystem functions (frugivores, insectivores and nectarivores) for years 2050 and 2070. Considering the entire study area, we estimated that between 4 and 19% of the species will find no suitable habitat. Inside the currently established protected areas, species loss could be over 70%. Our results suggest that frugivores are the most sensitive guild, which could bring consequences on seed dispersal functions and on natural regeneration. Moreover, we identified the western and northern parts of the study area as climatically stable. Climate change will potentially affect avian assemblages in southeastern Amazonia with detrimental consequences to their ecosystem functions. Information provided here is essential to conservation practitioners and decision makers to help on planning their actions.

Multi-Scenario Species Distribution Modeling

Correlative species distribution models (SDMs) are increasingly being used to predict suitable insect habitats. There is also much criticism of prediction discrepancies among different SDMs for the same species and the lack of effective communication about SDM prediction uncertainty. In this paper, we undertook a factorial study to investigate the effects of various modeling components (species-training-datasets, predictor variables, dimension-reduction methods, and model types) on the accuracy of SDM predictions, with the aim of identifying sources of discrepancy and uncertainty. We found that model type was the major factor causing variation in species-distribution predictions among the various modeling components tested. We also found that different combinations of modeling components could significantly increase or decrease the performance of a model. This result indicated the importance of keeping modeling components constant for comparing a given SDM result. With all modeling components, constant, machine-learning models seem to outperform other model types. We also found that, on average, the Hierarchical Non-Linear Principal Components Analysis dimension-reduction method improved model performance more than other methods tested. We also found that the widely used confusion-matrix-based model-performance indices such as the area under the receiving operating characteristic curve (AUC), sensitivity, and Kappa do not necessarily help select the best model from a set of models if variation in performance is not large. To conclude, model result discrepancies do not necessarily suggest lack of robustness in correlative modeling as they can also occur due to inappropriate selection of modeling components. In addition, more research on model performance evaluation is required for developing robust and sensitive model evaluation methods. Undertaking multi-scenario species-distribution modeling, where possible, is likely to mitigate errors arising from inappropriate modeling components selection, and provide end users with better information on the resulting model prediction uncertainty.