Biology subjects

Capinha, C.

Publications and source records attributed to Capinha, C..

At least 19 recordsLinked to original sources

bioRxiv2026

Two centuries of change in the traits and origins of non-native vertebrates

Globalization is redistributing species worldwide, yet whether the traits and origins of non-native fauna have changed through time remains unclear. We combined global first-record data for non-native species from 1800-2019 with harmonized information on body size, diet, habitat use, native range characteristics, and climatic niche characteristics for 1,910 non-native mammals, birds, reptiles, amphibians, and freshwater fishes. Across most groups, species recorded earlier originated disproportionately from higher latitudes, occupied broader native ranges, and had wider thermal niches. More recent first records increasingly involve species from warmer, lower-latitude regions with smaller and more restricted native distributions. Body size also declined through time in several groups, whereas shifts in diet and habitat use were more taxon-specific. Temporal changes in introduction pathways partly explained these patterns: declines in deliberate release and production-related pathways, together with increases in pet and ornamental pathways, were associated with shifts toward smaller-bodied, lower-latitude, and more range-restricted species. These results indicate that the functional and biogeographic composition of non-native vertebrates has been progressively reshaped over the past two centuries, weakening the historical dominance of widespread temperate species and increasingly incorporating tropical and range-restricted fauna into global redistribution. Anticipating future biological invasions will therefore require attention not only to the number of species being transported, but also to how the characteristics and pathways of transported species are changing through time.

ecology↗

bioRxiv2026

Data-driven forecasts of regional arrivals of non-native vertebrates worldwide

AimTo forecast near-future arrivals of non-native terrestrial and freshwater vertebrates at the regional level. LocationGlobal (geopolitical regions worldwide, including countries and main administrative divisions). MethodsWe compiled first regional record data and assembled functional and macroecological variables for 1,931 non-native vertebrate species. For each region, we identified recently arrived non-native species using retrospective windows of thirty and twenty years ending in 2015 (1986-2015; 1996-2015). We then fitted region-specific random-forest models classifying recently arrived species versus those not yet arrived using as predictors: (i) harmonised species traits (e.g., habitat, diet, body size and native-range attributes) and (ii) spread history, capturing time since first record elsewhere. Predictive performance was evaluated using leave-one-out cross-validation, comparing full models with trait-only and spread-only variants. We also assessed relationships between predictive accuracy, predictor importance, and the geographic positioning and trade connectedness of regions. Finally, we predicted region-specific probabilities of arrival for species not yet recorded. ResultsForecasting accuracy was consistently high across regions and taxa, with AUC values above 0.9 in more than half of the focal regions. Full models substantially outperformed models using either predictor set alone, and spread-history-only models typically exceeded trait-only models. Relative importance of spread-history predictors declined with geographic distance to the focal region, whereas predictability was lower in highly trade-connected regions. Predicted near-future high-risk arrivals were dominated by birds and freshwater fishes and showed strong regional structuring. A small set of species ranked highly across many regions (e.g., birds: Phasianus colchicus, Acridotheres tristis, Amandava amandava, Colinus virginianus, Corvus splendens and Lonchura malacca; fishes: Coregonus peled and Oreochromis mossambicus; mammal: Oryctolagus cuniculus), suggesting substantial unrealised spread potential. Main conclusionsNear-future regional arrivals of non-native vertebrates are predictable from spread history and species traits. This enables scalable, updateable regional watchlists to support prevention, early detection and horizon scanning.

ecology↗

bioRxiv2026

Trait based assessment of the invasion potential of disease vector mosquitoes

Mosquito-borne diseases pose a growing global health threat, largely driven by the human-mediated spread of vector species beyond their native regions. Although only a few mosquito species historically established populations outside their native ranges, many have expanded rapidly in recent decades. Once established, these invaders are notoriously difficult to control, emphasizing the need for proactive identification before human-mediated spread occurs. Here, we present a framework to anticipate invasion potential for 184 mosquito species of medical importance based on their ecological, life-history, and macroecological traits. We first compiled a comprehensive dataset of 26 traits characterizing each species. We then used random forest models to relate these traits with the probability of species being introduced in new regions (before and after 1950, marking the onset of widespread trade globalization), and of establishment following introduction. Models achieved moderate to good predictive performance (AUC = 0.78-0.85) and revealed that species native to Asia and Australia, adapted to human-made breeding sites, and tolerant of climatic extremes are consistently more likely to be introduced and to establish in non-native regions. Among species with no known invasion history, we identified 24 with higher potential to become future spreaders, of which 17 also exhibit high establishment probabilities ( high-risk species). These results show that invasion potential can be inferred, to some extent, from intrinsic species traits and provide a quantitative basis for proactive surveillance, enabling prioritization of species most likely to become introduced in the future. Author SummaryMosquito-borne diseases threaten more than half of the worlds population and cause over 700,000 deaths each year. Only a small share of mosquito species can spread these diseases, but some of them are moving into new regions where they have never been seen before. The spread of these mosquitoes has led to increasing numbers of locally transmitted outbreaks in regions that previously, or in recent times, had no mosquito-borne diseases. Human activities like trade and travel help mosquitoes spread, and once they arrive, they are extremely difficult to eradicate. Therefore, it is crucial to understand which species may spread in the future and to identify those that should be closely monitored to prevent their introduction and establishment. In this work, we linked species characteristics with their known invasion history to identify the factors driving their introduction and establishment in new regions. We found that species from Asia and Australia, capable of using human-made breeding sites, and tolerant of climatic extremes are most likely to become invaders. We then used these findings to predict which species might spread next. We identified 24 species with high invasion potential, including 17 that also have high chances of establishing once introduced. These results demonstrate that invasion risk can be predicted from measurable species traits, providing a framework to guide early-warning surveillance and prioritize species for monitoring before they begin spreading and become widespread vectors of human disease.

ecology↗

bioRxiv2025

Impacts of human-introduced species on the geography of life on Earth

Human activities are increasingly transporting species beyond their native ranges, where they often establish and become permanent additions to recipient biotas. These introductions, documented for tens of thousands of taxa, are a departure from the natural constraints of dispersal that have long shaped species distributions. However, current knowledge on their impacts on global biogeography remains poorly understood and fragmented. Here, we review empirical evidence across 15 classical biogeographical rules to assess how non-native species are altering the spatial structure of global biodiversity. Our synthesis reveals that while some patterns, such as species-area relationships, are often reinforced under invasion, others, including the delineation of global biogeographic regions, relationships between island isolation-diversity and between body sizes and latitudinal gradient (i.e., Bergmans rule), already show profound reshaping in contemporary assemblages. Additionally, other patterns (e.g., latitudinal gradient of species diversity; body size and insularity) show context-dependent changes, shaped by factors such as spatial scale, taxonomic group, and introduction history. These transformations are often more pronounced at broad spatial scales and in highly invaded systems such as islands and temperate regions. Our findings demonstrate that biological invasions are selectively but profoundly reshaping the geography of life on Earth, with major implications for conservation, macroecology, and the future of biodiversity patterns in the Anthropocene.

ecology↗

bioRxiv2025

Distribution patterns of invasive alien plant species in mainland Portugal

Understanding the distribution and diversity of invasive alien species is increasingly vital to meet legal requirements and guide effective management. Portugal currently hosts a large number of invasive alien plant species, with significant environmental and socio-economic impacts. However, the distribution of most of these species in the territory and the factors driving their spread and diversity remain largely unexplored. To address this, we present the first atlas of invasive plants in mainland Portugal. A total of 96 terrestrial and aquatic invasive plants are represented, encompassing all species listed under Portuguese national legislation and the European Unions list of invasive species of concern. Occurrence data were collected from a broad array of sources, including national and international biodiversity observation databases, citizen science data, literature, and data collections owned by managers and researchers, totaling approximately 85,000 records (mean = 879 {+/-} 1,757 per species). Each species was characterized based on multiple distributional parameters, and a k-means analysis was used to group species with similar distribution patterns. The richness of invasive plants was mapped at the municipality level, and the drivers of their spatial variation were investigated using a comprehensive set of 30 variables representing multiple environmental and human factors. Results show that invasive plants occur in all mainland Portuguese municipalities, but with high variability. Four main patterns of distribution were identified: species that are primarily located along the coast (e.g., Acacia saligna), and a gradient of species with narrow (e.g., Reynoutria japonica), moderate (e.g., Ipomoea indica), and wide distribution ranges (e.g., Cortaderia selloana). Species richness was significantly higher in coastal and larger municipalities, particularly those closer to major urban centers and with a higher density of power lines. Our results provide the first comprehensive assessment of the distribution of invasive alien plants in mainland Portugal, establishing a much-needed baseline for future invasion prevention and management efforts.

ecology↗

bioRxiv2025

Bioindicator-based mapping of peatland potential in the Western Mediterranean-Atlantic Realm

Ecosystems with a naturally accumulated peat layer at the surface and wetlands with or without a peat layer dominated by a vegetation that may produce peat (hereafter peatlands for simplicity) are critical assets in global change due to their large carbon content storage capacity and essential ecosystem services. A Global Peatland Assessment was initiated during the UNFCCC COP22; however, many countries still lack accurate peatland maps. This is particularly concerning in the Western Mediterranean, where these ecosystems are frequently overlooked or misclassified. To help address this issue, we developed predictive distribution models for vascular plant and bryophyte species typical of peatland-type systems in the Western Mediterranean. We identified 28 indicator mire-typical plant taxa and created a database using filtered data from six different data sources, including our own field surveys. Thirteen high-resolution spatial predictor variables, covering topographic, soil, and hydroclimatic factors, were selected to reflect potential relationships with species presence. For each species, a potential distribution model was built using MaxEnt software. The resulting potential-distribution maps were cross-validated, and models with an AUC score of [≥]0.75 were classified into suitable or unsuitable areas. The species richness map aligned well with areas known to support peatlands. The indicator species-based mapping also suggests a broad potential for yet unrecorded peatlands in the Western Mediterranean, including small areas, particularly along the western and southwestern sandy coastlines and in the southern mountains.

ecology↗

bioRxiv2025

Social and news media in mapping the endangered Mediterranean Monk seal: exceptional data gains but persistent bias

Social and news media are increasingly recognised as valuable sources of species distribution data, particularly for rare and endangered species. However, these data sources are prone to biases arising from uneven observation efforts, which pose challenges for their effective use. Despite this, the drivers and extent of such biases remain understudied. Here we investigate the contribution of social and news media to improving the availability of distribution data for a rare and endangered species, the Mediterranean Monk seal (Monachus monachus), in the Madeira and Porto Santo Islands. We also assess the relationship between the distribution data collected and socio-human factors likely to determine spatial and temporal variation in observation effort. We found that media sources provide a remarkable number of observation records for the species (n=302), which compared to the GBIF, iNaturalist and Observation.org combined (n=11), represent a 27.5-fold increase. The number of observations exhibited seasonal variation, with annual peaks occurring in the summer months. A generalised linear model assessing the spatial distribution of observation data also revealed a strong positive association with areas of high human concentration. Significant variables included population density, hotel density, and proximity to coastal recreational hotspots such as diving sites and bathing areas/non-open beaches. Our results underscore the potential of social and news media in documenting species distributions, but they also emphasise the importance of accounting for observational biases. Doing so is essential to ensure the reliable and effective application of social and news media sourced data in supporting species conservation efforts and management strategies.

ecology↗

bioRxiv2024

Global invasion patterns and dynamics of disease vector mosquitoes

Mosquitoes are regarded as the most dangerous creatures on earth, spreading deadly pathogens through their bites. Human activities are driving range expansions of many mosquito species by unintentionally introducing them beyond their native ranges. Despite the often dire consequences for human health, a global picture of the introduction trends and the resulting range expansions of mosquitoes is missing. Here, we describe the global invasion patterns of mosquitoes that are vectors of human diseases and analyze the drivers shaping them. In addition, we provide the dataset compiled for these analyses which represent the most up-to-date standardized information on first records for this taxonomic group at a regional level. Our findings reveal that a total of 45 mosquito species have hitherto been introduced into regions outside their native range worldwide, representing 24% of those known to transmit human pathogens in the wild (i.e., outside experimental conditions), with 27 species successfully established. There has been a steep increase in introductions of emerging non-native mosquito species since the mid-20th century when 28 species (62% of all introduced species) were recorded for the first time. In just the last two decades, 12 new species have been identified. The geography of introductions largely mirrors global trade and transportation flows. Initially, most introduced species were native to Africa, but over time, Asian species have become more dominant. North America, Australia and Europe have consistently been the primary recipients. Our results provide a foundation for addressing the increasing threat of non-native vector mosquitoes globally, emphasizing the need for international cooperation and comprehensive control measures to mitigate their impact on public health.

ecology↗

bioRxiv2024

Drivers of temporal bias in biodiversity recording by citizen scientists

O_LICitizen science data is increasingly important for ecological research, biodiversity conservation and monitoring. However, these data often suffer from biases due to uneven recording efforts by citizen scientists. Biases caused by intra-annual differences in recording activity can be particularly severe, hindering the use of citizen science data in research areas such as population dynamics and phenology. Therefore, understanding the factors driving recording activity is essential. C_LIO_LIIn this study, we provide a detailed assessment of how weather and calendar-related factors influence biodiversity recording activity by citizen scientists at a daily resolution. To perform this, we analyse the recording patterns for six tree species in the Iberian Peninsula, which maintain a fairly consistent appearance throughout the year. Observation data were collected from iNaturalist, a leading platform for citizen-science data. We used boosted regression trees (BRT) to compare observed recording activity patterns with those expected by chance. Our analysis included a comprehensive set of explanatory variables, such as day of the week, month, holidays, temperature, accumulated precipitation, wind intensity, and snow depth. C_LIO_LIThe BRT models effectively identified the drivers of recording activity, with the correlation between predicted and observed temporal patterns (left out of model training) ranging from 0.47 to 0.96, depending on the species. The day of the week, daily temperature, and month of the year consistently emerged as the main drivers. Recording activity was higher on weekends, to some extent on Fridays, and during the spring months. Extreme low and high temperatures generally correlated with lower recording activity, although there were exceptions. Wind speed and precipitation had a moderate influence, with higher wind intensity and accumulated precipitation leading to decreased activity. Holidays and accumulated snow had very minor relevance across species. C_LIO_LIOur findings show that citizen scientists record more frequently on weekends, during mild weather, and in spring. By addressing these biases, we can maximize the utility of citizen-collected data for research and applied purposes, ensuring robust and reliable conclusions that enhance ecological understanding and conservation efforts. C_LI

ecology↗

bioRxiv2024

Supporting early detection of biological invasions through short-term spatial forecasts of detectability

Early detection of invasive species is crucial to prevent biological invasions. To increase the success of detection efforts, it is often essential to know the phenological stages in which the invasive species are found. This includes knowing, for example, if invasive insect species are in their adult phase, invasive plants are flowering, or invasive mammals have finished their hibernation. Unfortunately, this kind of information is often unavailable or is provided at very coarse temporal and spatial resolutions. On the other hand, opportunistic records of the location and timing of observations of these stages are increasingly available from biodiversity data repositories. Here, we demonstrate how to apply these data for predicting the timing of phenological stages of invasive species. The predictions are made across Europe, at a daily temporal resolution, including in near real time and for multiple days ahead. We apply this to detectability-relevant phenological stages of four well-known invasive species: the freshwater jellyfish, the geranium bronze butterfly, the floating primrose-willow, and the garden lupine. Our approach uses machine learning and statistical-based algorithms to identify the set of temporal environmental conditions (e.g., temperature values and trends, precipitation, snow depth, and wind speed) associated with the observation of each phenological stage, while accounting for spatial and temporal biases in recording effort. Correlation between predictions from models and the actual timing of observations often exceeded values of 0.9. However, some inter-taxa variation occurred, with models trained on several thousands of observation records performing consistently better than those based on a few hundred records. The analysis of daily predictions also allowed mapping EU-wide regions with similar phenological dynamics (i.e., phenoregions). Our results underscore the significant potential of opportunistic biodiversity observation data in developing models capable of predicting and forecasting species phenological stages across broad spatial extents. This information has the potential to significantly improve decision-making in invasion surveillance and monitoring activities.

ecology↗

bioRxiv2024

Large language models overcome the challenges of unstructured text data in ecology

The vast volume of currently available unstructured text data, such as research papers, news, and technical report data, shows great potential for ecological research. However, manual processing of such data is labour-intensive, posing a significant challenge. In this study, we aimed to assess the application of three state-of-the-art prompt-based large language models (LLMs), GPT 3.5, GPT 4, and LLaMA-2-70B, to automate the identification, interpretation, extraction, and structuring of relevant ecological information from unstructured textual sources. We focused on species distribution data from two sources: news outlets and research papers. We assessed the LLMs for four key tasks: classification of documents with species distribution data, identification of regions where species are recorded, generation of geographical coordinates for these regions, and supply of results in a structured format. GPT 4 consistently outperformed the other models, demonstrating a high capacity to interpret textual data and extract relevant information, with the percentage of correct outputs often exceeding 90% (average accuracy across tasks: 87-100%). Its performance also depended on the data source type and task, with better results achieved with news reports, in the identification of regions with species reports and presentation of structured output. Its predecessor, GPT 3.5, exhibited reasonably low accuracy across all tasks and data sources (average accuracy across tasks: 81-97%), whereas LLaMA-2-70B showed the worst performance (37- 73%). These results demonstrate the potential benefit of integrating prompt-based LLMs into ecological data assimilation workflows as essential tools to efficiently process large volumes of textual data.

ecology↗

bioRxiv2023

Biodiversity monitoring in Europe: user and policy needs

To implement the goals of the 2030 Global Biodiversity Framework, the European Biodiversity Strategy and the EU Green Deal, biodiversity monitoring is a pivotal instrument to achieve accountability and progress in conservation. Monitoring efforts in Europe, however, suffer from gaps and biases in taxonomy, spatial coverage, and temporal resolution, resulting in fragmented and disconnected data which does not provide sufficient evidence for policy making. To assess user and policy needs in biodiversity monitoring, we employed with EuropaBON a four-step user-centred stakeholder engagement process, including an international public stakeholder workshop, a standardised online survey, semi-structured interviews, and an expert meeting with representatives of EU member states, the European Commission and the European Environment Agency. The resulting insights into national and European biodiversity monitoring schemes identify policy needs, current challenges and potential solutions. Based on this in-depth policy and science stakeholder assessment, we recommend the establishment of a European Biodiversity Observation Network through a permanent Biodiversity Monitoring and Coordinating Centre to optimise existing observation efforts, harmonise data, and enhance our ability to predict and respond to key challenges related to biodiversity loss in a changing climate in Europe.

ecology↗

bioRxiv2023

Recent land use and land cover pressures on Iberian peatlands

Iberian peatlands have been severely affected by land use and land cover (LULC) changes. Despite these pressures, some peatlands persist in the region, although their susceptibility to LULC change remains poorly understood. This study presents the most detailed and extensive distribution data for Iberian peatlands to date and analyzes the dynamics and drivers of LULC in Iberian peatlands and their surrounding areas. We compiled peatland records from various sources and used Corine Land Cover Change layers to determine LULC shifts for 1990, 2000, 2006, 2012, and 2018. Environmental and socioeconomic variables were used to create Boosted Regression Tree models explaining spatial variations in the mean percentage of changed area. Analysis of 270 peatland locations in the Iberian Peninsula revealed that forests and seminatural areas constituted over 80% of the peatlands surroundings. Agricultural areas expanded the most, except between 2006 and 2012 when the artificial areas showed more gains. While most areas experienced an average change of 0%-9.51% of the total area, between 1990 and 2018, lowland peatlands (littoral and sublittoral) suffered more intense changes (9.51% to 38.43%). Our models showed that only elevation and agricultural area density were relevant predictors of spatial distribution changes. Upland Iberian peatlands showed lower susceptibility to LULC changes, while lowland peatlands underwent remarkable transformations. This study substantially expands previous knowledge about the distribution and conservation needs of these ecosystems in the Iberian Peninsula, especially those in littoral and sublittoral lowlands.

ecology↗

bioRxiv2023

Predicting the timing of ecological phenomena across regions using citizen science data

Spatial predictions of intra-annual ecological variation enhance ecological understanding and inform decision-making. Unfortunately, it is often challenging to use statistical or machine learning techniques to make such predictions, due to the scarcity of systematic, long-term observational data. Conversely, opportunistic time-stamped observation records, supported by highly informative data such as photographs, are increasingly available for diverse ecological phenomena in many regions. However, a general framework for predicting such phenomena using opportunistic data remains elusive. Here, we introduce a novel framework that leverages the concept of relative phenological niche to model observation records as a sample of temporal environmental conditions in which the represented ecological phenomenon occurs. We demonstrate its application using two distinct, management-relevant, ecological events: the emergence of the adult stage of the invasive Japanese beetle (Popillia japonica), and of fruiting bodies of the winter chanterelle mushroom (Craterellus tubaeformis). The framework accounts for spatial and temporal biases in observation data, and it contrasts the temporal environmental conditions (e.g., in temperature, precipitation, wind speed, etc.) associated with the observation of these events to those available in their occurrence locations. To discriminate between the two sets of conditions, we employ machine-learning algorithms (boosted regression trees and random forests). The proposed approach can accurately predict the temporal dynamics of ecological events across large geographical scales. Specifically, it successfully predicted the intra-annual timing of occurrence of adult Japanese beetles and of winter chanterelle mushrooms across Europe and North America. We further validate the approach by successfully predicting the timing of occurrence of adult Japanese beetles in Northern Italy, a recent hotspot of invasion in continental Europe, and the winter chanterelle mushroom in Denmark, a country with a high number of records of this mushroom. These results were also largely insensitive to temporal bias in recording effort. Our results highlight the potential of opportunistic observation data to predict the temporal variation of a wide range of ecological phenomena in near real-time. Furthermore, the conceptual and methodological framework is intuitive and easily applicable for the large number of ecologists already using machine-learning and statistical-based predictive approaches.

ecology↗

bioRxiv2022

Forecasting the numbers of disease vectors with deep learning

Arboviral diseases such as dengue, Zika, chikungunya or yellow fever are a worldwide concern. The abundance of vector species plays a key role in the emergence of outbreaks of these diseases, so forecasting these numbers is fundamental in preventive risk assessment. Here we describe and demonstrate a novel approach that uses state-of-the-art deep learning algorithms to forecast disease vector numbers. Unlike classical statistical and machine learning methods, deep learning models use time series data directly as predictors and identify the features that are most relevant from a predictive perspective. We demonstrate the application of this approach to predict temporal trends in the number of Aedes aegypti mosquito eggs across Madeira Island for the period 2013 to 2019. Specifically, we apply the deep learning models to predict whether, in the following week, the number of Ae. aegypti eggs will remain unchanged, or whether it will increase or decrease, considering different percentages of change. We obtained high predictive accuracy for all years considered (mean AUC = 0.92 {+/-} 0.05 sd). We also found that the preceding numbers of eggs is a highly informative predictor of future numbers. Linking our approach to disease transmission or importation models will contribute to operational, early warning systems of arboviral disease risk.

ecology↗

bioRxiv2022

Predicting species distributions with environmental time series data and deep learning

Species distribution models (SDMs) are widely used to gain ecological understanding and guide conservation decisions. These models are developed with a wide variety of algorithms - from statistic-based approaches to machine learning approaches - but a requirement almost all share is the use of predictor variables that strongly simplify the temporal variability of driving factors. Conversely, novel architectures of deep learning neural networks allow dealing with fully explicit spatiotemporal dynamics and thus fitting SDMs without the need to simplify the temporal and spatial dimension of predictor data. We present and demonstrate a deep learning based SDM approach that uses time series of spatial data as predictors using distribution data for 74 species from a well-established benchmark dataset. The deep learning approach provided consistently accurate models, directly using time series of predictor data and thus avoiding the use of pre-processed predictor sets that can obscure relevant aspects of environmental variation.

ecology↗

bioRxiv2021

Invasive alien mammals of European Union concern

O_LIBiological invasions have emerged as one of the main drivers of biodiversity change and decline, and numbers of alien species are rapidly rising. The European Union established a dedicated regulation to limit the impacts of invasive alien species (IAS), which is focused on a Union List of IAS of particular concern. However, no previous study has specifically addressed the ecology of invasive alien mammals included in the Union List. C_LIO_LIWe performed a systematic review of published literature on these species. We retrieved 262 studies dealing with 16 species, and we complemented these with the most up-to-date information extracted from global databases on IAS. C_LIO_LIWe show that most of the study species reached Europe as pets that escaped from captivity or were intentionally released. On average, 1.2 species new first records/year were documented in European countries in the period 1981-2020, and most species are still expanding their alien ranges colonising neighbouring territories. France, Germany, Italy, and The Netherlands are the most invaded nations, and the muskrat (Ondatra zibethicus), the raccoon dog (Nyctereutes procyonoides), and the American mink (Neovison vison) are the most widespread species, having invaded at least 27 countries each. Invasive mammals of European Union concern are threatening native biodiversity and human well-being: worryingly, 81.3% of the study species are implicated in the epidemiological cycle of zoonotic pathogens. C_LIO_LIContaining the secondary spread to further countries is of paramount importance to avoid the establishment of new populations of invasive mammals and the related impacts on native communities, ecosystem services, and human health. C_LIO_LIOur results offer the most updated compendium on the ecology of invasive mammals of European Union concern, that can be used to assist environmental policies, identify and subsequently fill knowledge gaps, and inform stakeholders. C_LI GRAPHICAL ABSTRACT O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=143 SRC="FIGDIR/small/440832v1_ufig1.gif" ALT="Figure 1"> View larger version (57K): org.highwire.dtl.DTLVardef@934bcborg.highwire.dtl.DTLVardef@551036org.highwire.dtl.DTLVardef@9b519aorg.highwire.dtl.DTLVardef@10a7699_HPS_FORMAT_FIGEXP M_FIG C_FIG Graphical abstract: Invasive alien mammals of European Union concern. The figure illustrates how the introduction of a species in few new areas, followed by a lag phase of adaptation and sometimes enriched by further subsequent releases, can rapidly lead to the colonisation of large parts of a continent. On the top left, a heat map with species richness in countries of Europe. On the top right, a word cloud with the main keywords of our literature search and some of the study species names. On the bottom left, four out of 16 study species: in clockwise order, the raccoon dog (Nyctereutes procyonoides), the muskrat (Ondatra zibethicus), the American mink (Neovison vison), and the raccoon (Procyon lotor). On the bottom right, the temporal distribution of the first records of the study species in the countries of Europe.

ecology↗

bioRxiv2020

Deep learning for time series classification in ecology

Temporal data is ubiquitous in ecology and ecologists often face the challenge of accurately differentiating these data into predefined classes, such as biological entities or ecological states. The usual approach consists of transforming the time series into user-defined features and then using these features as predictors in conventional statistical or machine learning models. Here we suggest the use of deep learning models as an alternative to this approach. Recent deep learning techniques can perform the classification directly from the time series, eliminating subjective and resource-consuming data transformation steps, and potentially improving classification results. We describe some of the deep learning architectures relevant for time series classification and show how these architectures and their hyper-parameters can be tested and used for the classification problems at hand. We illustrate the approach using three case studies from distinct ecological subdisciplines: i) insect species identification from wingbeat spectrograms; ii) species distribution modelling from climate time series and iii) the classification of phenological phases from continuous meteorological data. The deep learning approach delivered ecologically sensible and accurate classifications demonstrating its potential for wide applicability across subfields of ecology.

ecology↗