Global priorities for an effective information basis of biodiversity distributions

Abstract

Severe gaps and biases in digital accessible information (DAI) of species distributions hamper prospects of safeguarding biodiversity and ecosystem services and reliably addressing central questions in ecology and evolution. Accordingly, governments have agreed on improving and sharing biodiversity knowledge by 2020 (United Nations Convention on Biological Diversity’s Aichi target 19). To achieve this target, gaps in DAI must be identified, and actions prioritized to address their root causes. We take terrestrial vertebrates, an iconic and comparatively well-studied group, as a model and present the first globally comprehensive assessment of patterns and drivers of gaps in DAI, based on an integration of 157 million validated point records with 21,170 expert-based distribution maps. We demonstrate that outside a few well-sampled regions, DAI provides a very limited and spatially highly biased inventory of actual biodiversity. Coarser spatial grains result in more complete inventories, but provide insufficient detail for conservation and resource management. Surprisingly, large emerging economies are particularly under-represented in global DAI, even more so than species-rich, developing countries in the tropics. Multi-model inference reveals that completeness is mainly limited by distance to researchers, locally available research funding, and political participation in data-sharing networks, rather than transportation infrastructure, or size and funding of Western data contributors as often assumed. Our study provides an empirical baseline to advance strategies of enhancing the global information basis of biodiversity. In particular, our results highlight the need for targeted data integration from non-Western data holders and intensified cooperation to more effectively address societal biodiversity information needs.

Reconciling Biodiversity Indicators to Guide Understanding and Action

“…Third, there are substantial gaps in data and observations due to the accessibility, popularity, measurability, and even fundamental knowledge of different components of biodiversity. Observations sourced for the most widely used indicators are inevitably biased; generally toward recent decades, large-bodied and charismatic species, in terrestrial, temperate, economicallydeveloped, and easily-accessible environments (Boakes et al 2010;Hudson et al 2014;Pimm et al 2014;Geijzendorffer et al 2015;Meyer et al 2015;Newbold et al 2015;Gonzalez et al 2016). Certain areas of significant biodiversity, such as soils and oceans, especially involving invertebrate and microscopic organisms, are extremely poorly known and weakly sampled (Mora et al 2011).…”

Untitled

“…We considered four socio-economic factors that are particularly important for limiting mammalian assemblage-level occurrence information (Meyer et al 2015).…”

“…We estimated xiii) locally available financial resources from conservation funding data (Waldron et al 2013). Large, species-rich countries require more resources to attain high coverage for all species (Meyer et al 2015). We therefore first divided country-level conservation funds by the country's total area of overlapping mammal ranges, to calculate a country's available resources per species range size to-be-covered (in million USD/10,000 km² range size).…”

“…While such records represent vital fine-scale information on spatial and temporal occurrences of species, severe gaps and biases hamper broader application (Rocchini et al 2011). These data limitations have been mostly studied with a focus on geographical assemblages (Soria-Auza & Kessler 2008;Meyer et al 2015), whereas differences among species have received less attention (Cayuela et al 2009). Bias towards species with certain (bio-)geographical, phylogenetic, or ecological attributes can lead to biased ecological inference (Garamszegi & Møller 2011) and inefficient conservation (Gonzalez-Suarez et al 2012).…”

“…First, range geometry, i.e., the size and shape of a range, might affect the likelihood that a given range part is close or distant to a given record. Second, socio-economic factors, such as area appeal, proximity to research institutions, cooperation with data-sharing networks, and financial resources may limit occurrence information by affecting the likelihood that records from within a given range are collected, digitized, and shared (Meyer et al 2015). While all abovementioned factors might drive species-level variation in record count and range coverage, within-range geographical bias of records should be driven by range size and within-range variation in socioeconomic factors (see Box 1).…”