Biodiversity Science Research Articles

GOTIT: A laboratory application software for optimizing multi‐criteria species‐based research

Abstract Storing, managing and sharing the ever‐increasing amount of species occurrence data and biological vouchers is a cornerstone of biodiversity science. The convergence of morphology‐ and DNA‐based methods for species delimitation has accentuated the need for optimizing laboratory workflow by generating new analytical processes, diversifying taxonomic reference systems and identification protocols and multiplying vouchers. Yet, laboratory information management systems dedicated to multi‐criteria species‐based research are rare or they have not been made openly available to biodiversity professionals. We developed GOTIT (Gene Occurrence and Taxa in Integrative Taxonomy), a database structure and web application for optimizing productivity at various steps of a laboratory's workflow including research project design, sampling gap analysis, sampling planning, species identification, DNA sequencing, information sharing and contributing data to global biodiversity facilities. GOTIT manages the different steps of a species‐occurrence data production process from sampling to sequencing, the storage of specimen lots, slides and DNA extracts and their assignment to morphology‐based and DNA‐based species hypotheses. The application also accommodates species occurrence data and DNA sequence metadata from external sources and the bibliographic referencing of information. GOTIT is designed to optimize research activities on the diversity and evolution of species which have a strong geographic component, such as biogeography and phylogeography. A key feature of GOTIT is the ability to assign multiple species hypotheses based on morphology and molecular delimitation methods to the same set of specimens. This integrative taxonomic approach allows to explore how different aspects of speciation, such as morphological distinguishability and genetic isolation, vary in strength relative to each other across taxa and geographic regions. Another key feature is traceability which promotes repeatability and minimizes duplicate work among multiple users. Any species data refer to a set of localities, sampling methods, biological vouchers, analytical protocols and delimitation methods that can be traced. The application is free and open‐source with extensive documentation and demo version. Developers can reuse the source code or develop new add‐ons to fulfill a laboratory's specific requirements although the application can be used as such to manage species occurrence data for a wide range of taxa.

Research applications of primary biodiversity databases in the digital age.

Our world is in the midst of unprecedented change—climate shifts and sustained, widespread habitat degradation have led to dramatic declines in biodiversity rivaling historical extinction events. At the same time, new approaches to publishing and integrating previously disconnected data resources promise to help provide the evidence needed for more efficient and effective conservation and management. Stakeholders have invested considerable resources to contribute to online databases of species occurrences. However, estimates suggest that only 10% of biocollections are available in digital form. The biocollections community must therefore continue to promote digitization efforts, which in part requires demonstrating compelling applications of the data. Our overarching goal is therefore to determine trends in use of mobilized species occurrence data since 2010, as online systems have grown and now provide over one billion records. To do this, we characterized 501 papers that use openly accessible biodiversity databases. Our standardized tagging protocol was based on key topics of interest, including: database(s) used, taxa addressed, general uses of data, other data types linked to species occurrence data, and data quality issues addressed. We found that the most common uses of online biodiversity databases have been to estimate species distribution and richness, to outline data compilation and publication, and to assist in developing species checklists or describing new species. Only 69% of papers in our dataset addressed one or more aspects of data quality, which is low considering common errors and biases known to exist in opportunistic datasets. Globally, we find that biodiversity databases are still in the initial stages of data compilation. Novel and integrative applications are restricted to certain taxonomic groups and regions with higher numbers of quality records. Continued data digitization, publication, enhancement, and quality control efforts are necessary to make biodiversity science more efficient and relevant in our fast-changing environment.

Review and Interpretation of Trends in DNA Barcoding

Interpretations and analytical practices surrounding DNA barcoding are reviewed from a compilation of 3,756 papers (as of December 31, 2018) with “DNA Barcode” in the title since 2004. By examining the practice of DNA barcoding in natural history and biodiversity science over this period, we explore the extent to which its purposes, premises, rationale and application have evolved. The number of studies involving identification, taxonomic decisions and the discovery of cryptic species has driven the publication of DNA barcode studies overall. Forensic studies and papers on biological conservation involving DNA barcodes have tracked the ensemble number of studies but rose sharply in 2017. Although neighbor-joining and graphic (tree-based) criteria for species delimitation have been preeminent, analytical paradigms have diversified slightly following the growing availability of tools in the Barcode of Life Database (BoLD). We conclude that the paradigms of DNA barcoding data are likely to persist and, in groups such as Lepidoptera, DNA barcoding has become a widely used tool in taxonomic science. The degree to which systematists will avail themselves of tools for extracting diagnostic data from barcodes remains to be seen.

What we (don't) know about global plant diversity

The era of big biodiversity data has led to rapid, exciting advances in the theoretical and applied biological, ecological and conservation sciences. While large genetic, geographic and trait databases are available, these are neither complete nor random samples of the globe. Gaps and biases in these databases reduce our inferential and predictive power, and this incompleteness is even more worrisome because we are ignorant of both its kind and magnitude. We performed a comprehensive examination of the taxonomic and spatial sampling in the most complete current databases for plant genes, locations and functional traits. To do this, we downloaded data from The Plant List (taxonomy), the Global Biodiversity Information Facility (locations), TRY (traits) and GenBank (genes). Only 17.7% of the world's described and accepted land plant species feature in all three databases, meaning that more than 82% of known plant biodiversity lacks representation in at least one database. Species coverage is highest for location data and lowest for genetic data. Bryophytes and orchids stand out taxonomically and the equatorial region stands out spatially as poorly represented in all databases. We have highlighted a number of clades and regions about which we know little functionally, spatially and genetically, on which we should set research targets. The scientific community should recognize and reward the significant value, both for biodiversity science and conservation, of filling in these gaps in our knowledge of the plant tree of life.

Measuring continuous compositional change using decline and decay in zeta diversity.

Incidence, or compositional, matrices are generated for a broad range of research applications in biology. Zeta diversity provides a common currency and conceptual framework that links incidence-based metrics with multiple patterns of interest in biology, ecology, and biodiversity science. It quantifies the variation in species (or OTU) composition of multiple assemblages (or cases) in space or time, to capture the contribution of the full suite of narrow, intermediate, and wide-ranging species to biotic heterogeneity. Here we provide a conceptual framework for the application and interpretation of patterns of continuous change in compositional diversity using zeta diversity. This includes consideration of the survey design context, and the multiple ways in which zeta diversity decline and decay can be used to examine and test turnover in the identity of elements across space and time. We introduce the zeta ratio-based retention rate curve to quantify rates of compositional change. We illustrate these applications using 11 empirical data sets from a broad range of taxa, scales, and levels of biological organization-from DNA molecules and microbes to communities and interaction networks-including one of the original data sets used to express compositional change and distance decay in ecology. We show (1) how different sample selection schemes used during the calculation of compositional change are appropriate for different data types and questions, (2) how higher orders of zeta may in some cases better detect shifts and transitions, and (3) the relative roles of rare vs. common species in driving patterns of compositional change. By exploring the application of zeta diversity decline and decay, including the retention rate, across this broad range of contexts, we demonstrate its application for understanding continuous turnover in biological systems.

The carrying capacity for species richness

AbstractThe idea that the number of species within an area is limited by a specific capacity of that area to host species is old yet controversial. Here, we show that the concept of carrying capacity for species richness can be as useful as the analogous concept in population biology. Many lines of empirical evidence indicate the existence of limits of species richness, at least at large spatial and phylogenetic scales. However, available evidence does not support the idea of diversity limits based on limited niche space; instead, carrying capacity should be understood as a stable equilibrium of biodiversity dynamics driven by diversity‐dependent processes of extinction, speciation and/or colonization. We argue that such stable equilibria exist even if not all resources are used and if increasing species richness increases the ability of a community to use resources. Evaluating the various theoretical approaches to modelling diversity dynamics, we conclude that a fruitful approach for macroecology and biodiversity science is to develop theory that assumes that the key mechanism leading to stable diversity equilibria is the negative diversity dependence of per‐species extinction rates, driven by the fact that population sizes of species must decrease with an increasing number of species owing to limited energy availability. The recently proposed equilibrium theory of biodiversity dynamics is an example of such a theory, which predicts that equilibrium species richness (i.e., carrying capacity) is determined by the interplay of the total amount of available resources, the ability of communities to use those resources, environmental stability that affects extinction rates, and the factors that affect speciation and colonization rates. We argue that the diversity equilibria resulting from these biodiversity dynamics are first‐order drivers of large‐scale biodiversity patterns, such as the latitudinal diversity gradient.

Species Distribution Modelling Using Deep Learning

Species distribution modelling, or ecological niche modelling, is a collection of techniques for the construction of correlative models based on the combination of species occurrences and GIS data. Using such models, a variety of research questions in biodiversity science can be investigated, among which are the assessment of habitat suitability around the globe (e.g. in the case of invasive species), the response of species to alternative climatic regimes (e.g. by forecasting climate change scenarios, or by hindcasting into palaeoclimates), and the overlap of species in niche space. The algorithms used for the construction of such models include maximum entropy, neural networks, and random forests. Recent advances both in computing power and in algorithm development raise the possibility that deep learning techniques will provide valuable additions to these existing approaches. Here, we present our recent findings in the development of workflows to apply deep learning to species distribution modelling, and discuss the prospects for the large-scale application of deep learning in web service infrastructures to analyze the growing corpus of species occurrence data in biodiversity information facilities.

Integrated Biodiversity Infrastructure for Decision Making

The Atlas of Living Australia (ALA) is an Australian Government supported collaborative partnership of organisations that have stewardship of Australian biodiversity data. The ALA (www.ala.org.au) provides research infrastructure that enables delivery of biodiversity information to over 45,000 unique users in research, industry and government per annum. It delivers impact and enables research excellence in fields such as biodiversity, environmental management, ecology and genetic sciences. Integrated and consistent infrastructure and processes are fundamental to increasing value of collections and associated data. The Atlas of Living Australia has a mature industry engagement program that provides data standardisation, quality and analytical services to decision makers in all tiers of Australian government (local, state and federal). This program is built on formal partnerships between data providers (collection institutions) and analytical services (such as Virtual Laboratories and Research and Science Clouds www.ecocloud.org.au). The provision of high quality, authoritative data is critical to utilisation and uptake of these services and sector sustainability. This presentation will showcase data service and analytical methods for decision makers within the Australian context. It will also explore how international efforts such as DiSSCo assist in data stewardship, cultural change and system enhancement.

Tropical bird species richness is strongly associated with patterns of primary productivity captured by the Dynamic Habitat Indices

Biodiversity science and conservation alike require environmental indicators to understand species richness and predict species distribution patterns. The Dynamic Habitat Indices (DHIs) are a set of three indices that summarize annual productivity measures from satellite data for biodiversity applications, and include: a) cumulative annual productivity; b) minimum annual productivity; and c) variation in annual productivity. At global scales and in temperate regions the DHIs predict species diversity patterns well, but the DHIs have not been tested in the tropics, where higher levels of productivity lead to the saturation of many remotely sensed vegetation indices. Our goal was to explain bird species richness patterns based on the DHIs in tropical areas. We related the DHIs to species richness of resident landbirds for five guilds (forest, scrub, grassland, generalist, and all resident birds) based on a) species distribution model (SDM) maps for 217 species, and b) range map for 564 species across Thailand. We also quantified the relative importance of the DHIs in multiple regression models that included two measures of topography, and two climate metrics using multiple regression, best-subsets, and hierarchical partitioning analyses. We found that the three DHIs alone explained forest bird richness best (R2adj 0.61 for both SDM- and rangemap based richness; 0.15–0.54 for the other guilds). When combining the DHIs with topography and climate, the richness of both forest birds and all resident bird species was equally well explained (R2adj 0.85 and 0.67 versus 0.81 and 0.68). Among the three DHIs, cumulative annual productivity had the greatest explanatory power for all guilds based on SDM richness maps (R2adj 0.54–0.61). The strong relationship between the DHIs and bird species richness in Thailand suggests that the DHIs capture energy availability well and are useful in biodiversity assessments and potentially bird conservation in tropical areas.

The Role of Natural History Collections in Public Engagement for Enhancing Biodiversity Research

Citizen science biodiversity monitoring projects are becoming very common. It is generally accepted that these joint projects, of scientists and the public, have a positive effect on biodiversity and conservation education programs as well as on policy-makers opinion Ganzevoort et al. (2017). Yet, there is still a debate on the quality of the data collected in citizen science monitoring schemes, and especially on the benefits to high-quality research. Here, I present an example of how collection-based research and involvement of the public (non-taxonomists) in taxonomical education, i.e., advanced citizen science, can enhance research on scorpion diversity in Israel. Furthermore, the process of public involvement in monitoring and especially the prerequisites needed for this process, contributed to high-quality research, that in turn is enhancing biodiversity science. Considering this, I will discuss the basic stages required for successful public engagement in high-quality biodiversity research and monitoring schemes.

Methods, New Software Tools, and Best Practices for Developing High-quality Training Data for Machine Learning-based Image Analysis in Biodiversity Research

Recent progress in using deep learning techniques to automate the analysis of complex image data is opening up exciting new avenues for research in biodiversity science. However, potential applications of machine learning methods in biodiversity research are often limited by the relative scarcity of data suitable for training machine learning models. Development of high-quality training data sets can be a surprisingly challenging task that can easily consume hundreds of person-hours of time. In this talk, we present the results of our recent work implementing and comparing several different methods for generating annotated, biodiversity-oriented image data for training machine learning models, including collaborative expert scoring, local volunteer image annotators with on-site training, and distributed, remote image annotation via citizen science platforms. We discuss error rates, among-annotator variance, and depth of coverage required to ensure highly reliable image annotations. We also discuss time considerations and efficiency of the various methods. Finally, we present new software, called ImageAnt (currently under development), that supports efficient, highly flexible image annotation workflows. ImageAnt was created primarily in response to the challenges we discovered in our own efforts to generate image-based training data for machine learning models. ImageAnt features a simple user interface and can be used to implement sophisticated, adaptive scripting of image annotation tasks.

Introducing ‘The bdverse’: a family of R packages for biodiversity data

The bdverse is a collection of packages that form a general framework for facilitating biodiversity science in R. We build it to serve as a sustainable and agile infrastructure that enhances the value of biodiversity data by allowing users to conveniently employ R, for data exploration, quality assessment, data cleaning, and standardization. The bdverse supports users with and without programming capabilities. It includes six unique packages in a hierarchal structure — representing different functionality levels (Fig. 1). Major features of three core packages will be highlighted and demonstrated: (i) bdDwC provides an interactive Shiny app and a set of functions for standardizing field names in compliance with Darwin Core (DwC) format; (ii) bdchecks is an infrastructure for performing, filtering and managing various biodiversity data checks; (iii) bdclean is a user-friendly data cleaning Shiny app for the inexperienced R user. It provides features to manage complete workflow for biodiversity data cleaning, including data upload; user input - in order to adjust cleaning procedures; data cleaning; and finally, generation of various reports and versions of the data. We are now working on submitting the bdverse packages to rOpenSci software review, and as soon as the packages meet core requirements, we will officially release the bdverse. The bdverse project won the 2nd prize in the 2018 Ebbe Nielsen Challenge.

Perspectives on biodiversity informatics for ecology

AbstractBiodiversity informatics is the application of informatics techniques to ecology and biodiversity sciences. The premise is utilizing natural history collections/data, such as specimens and biodiversity observations, with information and communication technology. During the past two decades, biodiversity informatics has improved dramatically and has been applied increasingly in ecology science. In this paper, I review biodiversity informatics for ecology with a particular focus on the treatment of data. First, I discuss the traditional perspective of data collection and its usage in the ecological community. Then, I provide an overview of the trajectory of changing perspectives of data treatment relative to developing biodiversity informatics (i.e., the infiltration of the data reuse and sharing concepts). Specifically, I discuss the significance of data reuse, which offers numerous benefits for research, and data sharing with their supporting mechanisms and case studies. Finally, I discuss potential future developments in biodiversity informatics relative to the field of ecology.

Biodiversity Informatics in Brazil: A personal perspective

Biodiversity informatics (BI) plays an important role in helping us know, protect and use biodiversity sustainably. It encompasses activities from data digitization, standardization, sharing and aggregation, to supporting decision and policy making. In a country like Brazil, with a large continental geographic area containing ca. 15% of the planet’s biodiversity, the challenge is even greater: stakeholders are widely distributed over a large country and the amount of data is huge. Brazil has been a part of the international BI community, including Biodiversity Information Standards (TDWG), for around two decades. Initially represented solely by the Centro de Referência em Informação Ambiental (CRIA), gradually other groups from universities, museums and institutions joined the arena. Despite the broader group of stakeholders now involved, the local community is not strong enough. From a human resources point of view, the country has very good universities that train competent professionals both in information technology (IT) and in biology or related fields. Concerning the IT professionals, not surprisingly, other industries and job opportunities are usually more attractive and few people even know about BI. This is probably not unique to Brazil. Biological sciences professionals, for their part, usually have little literacy in computing and are equally unaware of BI as a field. On the institutional level, museums, universities and other biological data owners often lack IT support for biological data management, including digitization, and systems development/maintenance. This may reflect the lack of appreciation of the importance of data and of BI as a foundation for good biodiversity science and management. The same happens when it comes to funding. Biological collections are not adequately funded and lack more than a few episodic programs to support collection and museum maintenance and digitization. This lack of infrastructural funding highlights the tragedies of the fires at the Butantan Museum in 2010 (80,000 snakes, 180,000 spiders) and the Museu Nacional in 2018, (20 million biological specimens and objects of the Brazilian and world history and art were lost). The exception is the São Paulo Research Foundation (FAPESP), which has been supporting projects since 1999 on biodiversity and BI via its successful Biota-FAPESP program, the first to tie biodiversity projects to data digitization and sharing in Brazil. The lack of institutional engagement and support, and funding affects the sustainability of many initiatives and puts at risk the long term data availability. Due to political reasons, Brazil only joined the Global Biodiversity Information Facility (GBIF) in 2012 as an associate (non-contributing financially and non-voting) member with a commitment to become a voting member within five years. Until recently, the Brazilian Biodiversity Information System (SiBBr), Brazil’s GBIF node, was also hindered by politics from having a solid, stable national governance and funding to help “consolidate a solid national infrastructure on biodiversity data”, and to unite the growing Brazilian BI community around it. In the international scenario, while political, cultural and funding reasons may have hindered more equitable collaborations (e.g., tools development and sharing) with countries in the Global South, competing Global North-centric projects have prevailed. Although most remaining biodiversity is in the Global South, where local engagement is crucial, in many cases southern partners still often only act as data providers. Collaborative work is required with genuine co-creation, empowering all parties. Initiatives like the Living Atlases community must be recognized and welcomed as a positive shift. Despite all of these challenges, it can be surprising how much Brazilian biodiversity science has achieved throughout the years, and it gives us hope that in the future, if some of these issues are addressed, a lot more can be done. Education and training, continued funding and institutional support, governance, and international collaboration are essential.

Virtual Laboratories for Biodiversity Modelling: An Australian perspective

Recent technologies have enabled consistent and continuous collection of ecological data at high resolutions across large spatial scales. The challenge remains, however, to bring these data together and expose them to methods and tools to analyse the interaction between biodiversity and the environment. These challenges are mostly associated with the accessibility, visibility and interoperability of data, and the technical computation needed to interpret the data. Australia has invested in digital research infrastructures through the National Collaborative Research Infrastructure Strategy (NCRIS). Here we present two platforms that provide easy access to global biodiversity, climate and environmental datasets integrated with a suite of analytical tools and linked to high-performance cloud computing infrastructure. The Biodiversity and Climate Change Virtual Laboratory (BCCVL) is a point-and-click online platform for modelling species responses to environmental conditions, which provides an easy introduction into the scientific concepts of models without the need for the user to understand the underlying code. For ecologists who write their own modelling scripts, we have developed ecocloud: a new online environment that provides access to data connected with command-line analysis tools like RStudio and Jupyter Notebooks as well as a virtual desktop environment using Australia’s national cloud computing infrastructure. ecocloud is built through collaborations among key facilities within the ecosciences domain, establishing a collective long-term vision of creating an ecosystem of infrastructure that provides the capability to enable reliable prediction of future environmental outcomes. Underpinning these tools is an innovative training program, ecoEd, which provides cohesive training and skill development to enhance the translation of Australia’s digital research infrastructures to the ecoscience community by educating and upskilling the next generation of environmental scientists and managers. Both of these platforms are built using a best-practice microservice model that allows for complete flexibility, scalability and stability in a cloud environment. Both the BCCVL and ecocloud are open-source developments and provide opportunities for interoperability with other platforms (e.g. Atlas of Living Austalia). In Australia, the same technical infrastructure is also used for a platform for the humanities and social science domain, indicating that the underlying technologies are not domain specific. We therefore welcome collaborations with other organisations to further develop these platforms for the wider bio- and ecoinformatics community. This presentation will showcase the tools, services, and underpinning infrastructure alongside our training and engagement framework as an exemplar in building platforms for next generation biodiversity science.

Developing a Data-Literate Workforce through BLUE: Biodiversity Literacy in Undergraduate Education

The biodiversity sciences have experienced a rapid mobilization of data that has increased our capacity to investigate large-scale issues of critical importance (e.g., climate change and its impacts, zoonotic disease transmission, sustainable resource management, impacts of invasive species, and biodiversity loss). Several initiatives are underway to aggregate and mobilize these biodiversity, environmental, and ecological data resources (iDigBio, NEON, GBIF, iNaturalist, etc.). This requires a new set of skills for the 21st century biodiversity scientist; who is required to be fluent in integrative fields spanning evolutionary biology, systematics, ecology, geology, and environmental science and possess the quantitative, computational, and data skills to conduct research using large and complex datasets. The biodiversity science community has recognized a need to unite biodiversity and data sciences and improve data literacy in the emerging science workforce. The NSF-funded Biodiversity Literacy in Undergraduate Education (BLUE; biodiversityliteracy.com) is working to bridge a gap between efforts that currently exists to promote data literacy pre-college and professional development for those pursuing careers in biodiversity science. The BLUE network is developing strategies and materials to infuse biodiversity data into the core of the undergraduate science curriculum, facilitating broad-scale adoption of biodiversity data literacy competencies, and improving undergraduate biology training to meet increasing workforce demands in data and biodiversity sciences. The BLUE network has four major goals: 1) Cultivate a diverse and inclusive network of biodiversity researchers, data scientists, and biology educators focused on undergraduate data-centric biodiversity education; 2) build community consensus on core biodiversity data literacy competencies; 3) develop strategies and exemplar materials to guide the integration of biodiversity data literacy competencies into introductory undergraduate biology curricula; and 4) extend the network to engage a broader community of undergraduate educators in biodiversity data literacy efforts. The BLUE community continues to grow and build new partnerships and initiatives across the biodiversity science community. In year two of the BLUE network we have been focusing efforts building the community, developing and disseminating exemplar educational materials, and defining core biodiversity data literacy skills and competencies. We will present our current and ongoing work and ways in which members of the biodiversity_next community can be involved in shaping the biodiversity science of the future, while addressing the needs of a changing planet.

A More Precise Living Planet Index (LPI)

Reliably measuring biodiversity change is of major interest both ecologically and politically. Thoroughly testing the reliability of biodiversity metrics, meaning their ability to present a precise, accurate, and unbiased measurement of biodiversity trends, is vital to avoid misinforming decision makers when selecting management strategies, and misleading the public’s view of biodiversity issues. Developed by Loh et al. (2005), the World Wildlife Fund (WWF), and the Zoological Society of London, the Living Planet Index (LPI) measures the overall global trend in vertebrate abundances since 1970 (Loh et al. 2005). The LPI has become a popular indicator of global biodiversity change due to its intuitive association with biodiversity targets, which makes it a powerful tool to communicate the status of biodiversity to the public, and to decision makers tasked with the management of biodiversity (Collen et al. 2009). Importantly, the Convention on Biological Diversity (CBD) selected the LPI as one of four indices approved to monitor progress towards the Aichi Biodiversity Targets (CBD 2016). The LPI was also selected as an Essential Biodiversity Variable, which are essential tools for the harmonized study, reporting, and management of biodiversity change worldwide (Pereira et al. 2013). Due to the LPI’s pivotal role in policy, conservation, and communication, ensuring its reliability as a biodiversity change metric is crucial for both biodiversity science and management at a global scale. Despite the LPI’s influence on the global stage of biodiversity management, the index’s reliability as a measure of biodiversity loss has not been rigorously tested for limitations. The LPI’s capacity to accurately estimate global population trends depends in part on undetected errors in its supporting data, including process errors (i.e. random population fluctuations), and observation errors (i.e. poor detectability of certain species) (Clark and Bjørnstad 2004, Gotelli and Colwell 2001, Dennis et al. 2006, Buckland et al. 2004). These errors are often undocumented and cannot be directly assessed, making it difficult to determine their impact on LPI trends. Interpretation of an LPI trend also heavily relies on its precision, which establishes the uncertainty surrounding the estimated magnitude and direction of biodiversity change. If uncertainty is high, the LPI is a less reliable measure of biodiversity change, and must be interpreted with greater caution when communicating biodiversity trends to the public and when choosing management strategies (Hui et al. 2008). The LPI’s confidence interval progressively widens as uncertainty accumulates over time from the baseline in 1970 (WWF 2018, McRae et al. 2017, Collen et al. 2009). This growing imprecision could impede our ability to confidently interpret both the direction and magnitude of biodiversity change, which has serious implications for monitoring progress towards global biodiversity targets, and for maintaining public engagement in biodiversity issues. Here, we suggest a methodological approach to address this growing uncertainty, in order to improve the reliability of the LPI. We propose a methodology for integrating the covariation between population trends into the computation of the LPI, in order to bolster confidence in related trends and ultimately decrease imprecision. Rather than assuming all population trends are varying in isolation, we explicitly identify covariation between population trends, and weight populations according to this covariation. We then compare the confidence intervals of the reported LPI trend to the trend obtained using our proposed methodology, and consider the potential political, ecological, and communicational implications of this approach on the interpretation of biodiversity metrics.

Designing an Ecosystem Services Ontology within GEOBON

Assessing, mapping and quantifying ecosystem services is a growing priority within local, national and international policy and decision-making processes. That requires robust and scientifically sound data and information on biodiversity, including ecological and socio-economic aspects that interact within the social-ecological systems. Ecosystem services are quantified and mapped across spatio-temporal scales, socio-political contexts and for different policy objectives, which leads to an immense variety of approaches, methods, tools, modelling and mapping outputs. Within the Group on Earth Observation-Biodiversity Observation Network (GEOBON) working group on Ecosystem Services (https://geobon.org/ebvs/ecosystem-services/), we acknowledge the significance of maintaining this diversity, but also the need to bring these different approaches together, to improve data and information sharing across a broad range of scientific fields. Given the high diversity of topics, terms, and classification frameworks that exist, harmonizing everything under one core set of data standards proved to be both a challenge and a questionable option in terms of usability. We therefore work towards developing and proposing an ontological system that can be used for monitoring ecosystem services across space and time. We designed ESOnto, the ecosystem services ontology, which uses the principles of Linked Data (Berners-Lee 2006) to connect the concepts of the ontology, while providing a flexible structure that hosts multiple approaches/definitions and ecosystem services classification systems. ESOnto takes into account the most widely used ecosystem services classification frameworks that exist are all used in ESOnto, since they all address different objectives and approaches. Their core concepts are all included in the ontology and the relationships among them are established allowing the user to navigate among the different classification systems. as well as reRelevant ontologies from both the environmental and social o-economic domains, such as ENVO (https://www.ebi.ac.uk/ols/ontologies/envo) and OBO (www.obofoundry.org) . These are also used to define build the basic concepts that ESOnto covers such as, 'map' or 'habitat'. This use of existing knowledge allows for interoperability among ontological systems, and guarantees a more usable ontology that does not aim to reinvent the wheel. Although the need for setting up data standards for ecosystem services has been expressed among researchers (Palomo et al. 2018, Drakou et al. 2015), this is the first time such an attempt has been made. Within this talk we will present the workflow we followed: we made an overview of existing data standards and ontologies from the fields of biodiversity, ecology and social sciences, which were used to inform the ESOnto; we identified the gaps and proposed a set of data standards for ecosystem services; we developed a set of task-based queries and user-based evaluations to validate those data standards before their final establishment as an ontology. ESOnto is now applied and tested within existing platforms, namely the Bon-in-a-Box (https://boninabox.geobon.org/). The final objective is that ESOnto will be used to facilitate the process of data collection through remote sensing within the GEOBON community.

TaxonWorks: An experience of migrating large datasets into the new cybertaxonomic infrastructure

TaxonWorks (http://taxonworks.org) in an integrated, open-source, cybertaxonomic web application serving taxonomists and biodiversity scientists. It is designed to facilitate efficient data capture, storage, manipulation, and retrieval. It integrates a wide variety of data types used by biodiversity scientists, including, but not limited to, taxonomy (with validation based on codes of zoological, botanical, bacterial, and viral nomenclature), specimen data, bibliographies, media (images, PDFs, sounds, videos), morphology (character/trait matrices), distribution, biological associations. Available TaxonWorks web interfaces currently provide various data entry forms for simple and advanced querying of the database. TaxonWorks has integrated batch uploader functionality. But, for larger datasets, specialized migration scripts were used. Several projects, historically build in 3i (http://dmitriev.speciesfile.org), MX (http://mx.phenomix.org), SpeciesFiles (http://software.speciesfile.org), and other databases, have been or are being migrated into TaxonWorks. Of the projects moving into TaxonWorks, it is worth mentioning several: 3i World Auchenorrhyncha Database, LepIndex, Universal Chalcidoidea Database, Orthoptera SpeciesFile, Plecoptera SpeciesFile, Illinois Natural History Survey Insect Collection database, and several others. An experience of the data migration will be shared during the presentation.

Scaling Up Data Literacy and Computing Skills Training in Biodiversity Science, Lessons Learned from The Carpentries

The expanding availability of access to data about museum specimens, species occurrences, trait data, genetics, and landscapes is revolutionizing biodiversity research. But mobilizing, evaluating, and synthesizing these data to address research questions requires domain-specific computing skills and knowledge. The Carpentries is a global non-profit organization teaching researchers computational and data science skills for data management and analysis in all domains of research. The Carpentries' teaching practices are based on evidence from the science of learning. Current domain-specific lessons include life, physical, and social sciences. Hands-on exercises build on the existing knowledge of learners, quickly enabling them to apply skills learned to their research. The Carpentries' train-the-trainer model, its globally established community of ~2,000 instructors and its network of regional coordinators have allowed the organization to scale up rapidly, and over 40,000 learners have participated in Carpentries workshops since 2012. The Carpentries assessment program evaluates the impact of attending a workshop on the participants' perception of their skills and confidence. After attending a Carpentries workshop, participants feel more confident in their ability to program and to overcome issues with their analysis. The Carpentries' mentoring program ensures that newly certified instructors receive support and guidance to successfully engage in leading and teaching workshops. Pairing new and experienced instructors, in combination with the development of local communities of instructors, have contributed to the success of The Carpentries workshops. The Carpentries collaboratively seeks to develop new curricula to address the training needs of the biodiversity data community. Establishing training partnerships between The Carpentries and the various organizations involved in building biodiversity data science skills capacity would increase reach and efficacy. We seek to foster institutional and organizational collaborations that can support the establishment and growth of the infrastructure needed to provide the computational and data literacy skills needed to empower our stakeholders. These include those creating and mobilizing data (e.g., collectors, ecologists, conservationists, curators, collection/data managers), for all scientists, policy-makers, educators, and students who want to use biodiversity data.