An operational workflow for producing periodic estimates of species occupancy at large scales

Abstract

Policy makers require high-level summaries of biodiversity change. However, deriving such summaries from raw biodiversity data is a complex process involving several intermediary stages. In this paper, we describe a workflow for generating annual estimates of species’ occupancy at national scales from raw species occurrence data, which can be used to construct a range of policy-relevant biodiversity indicators. We describe the workflow in detail: from data acquisition, data assessment and data manipulation, through modelling, model evaluation, application and dissemination. At each stage, we draw on our experience developing and applying the workflow for almost a decade to outline the challenges that analysts might face. These challenges span many areas of ecology, taxonomy, data science, computing and statistics. In our case, a key output of the workflow is annual estimates of occupancy, with measures of uncertainty, for over 5,000 species in each of several defined “regions” (e.g., countries, protected areas, etc.) of the United Kingdom from 1970-2019. This product corresponds closely to the notion of a species distribution “Essential Biodiversity Variable” (EBV). Throughout the paper, we note where the workflow can be adapted to other situations (e.g., geographic regions or data types). We also highlight areas where the workflow can be improved; in particular, we suggest incorporation of methods to diagnose biases in the species occurrence data, to understand whether and to what extent these bias downstream products, and to mitigate them if needed. Finally, we compare the data products generated using our workflow to the first generation of species distribution EBVs and the “idealized” product as defined by others. Going forward, we hope that this paper can act as a template for research groups around the world seeking to develop similar data products.

On the trade-off between accuracy and spatial resolution when estimating species occupancy from biased samples

“…The data were analysed at a 10 × 10 km scale-much coarser than the 1 × 1 km resolution used by others in the area (e.g. Boyd, August, et al, 2023)-and particular timeperiods were omitted, because of serious concerns about sampling biases affecting species data at finer scales across the 20 th century. For example, rarer and more critical taxa were more likely to be reported at finer scales in the early part of the time series.…”

“…Species occupancy, which we define as the proportion of areal units (sites) in some defined landscape occupied by the focal species, is often of interest to ecologists (Kéry & Royle, 2016). It is used to quantify species' range dynamics (Dennis et al, 2019;Outhwaite et al, 2020;Powney et al, 2019;Stroh et al, 2023), identify correlates and drivers of those range dynamics (Cooke et al, 2023;Woodcock et al, 2016), track the spread of invasive species and their effects on native taxa (Roy et al, 2012) and monitor progress towards (inter-) national biodiversity targets (Boyd, August, et al, 2023). Clearly, information on species occupancy has the potential to be useful, but realising this potential is conditional on available data being an accurate reflection of reality.…”

An operational workflow for producing periodic estimates of species occupancy at national scales

“…for all species in a taxonomic group). These files are generated and updated using the createMetadata function in the wrappeR package (Boyd et al ., 2022 b ), which summarises the input and output files from the Object Store. The run‐level metadata propagates metadata from the species level, where applicable, whilst also summarising higher level metadata, such as the number of species modelled.…”