Uncertainty in identifying local extinctions: the distribution of missing data and its effects on biodiversity measures.

Elizabeth H Boakes,Philip J K Mcgowan,Richard A Fuller,Georgina M Mace

doi:10.1098/rsbl.2015.0824

Elizabeth H Boakes, Philip J K Mcgowan + Show 2 more

Open Access

https://doi.org/10.1098/rsbl.2015.0824

Copy DOI

Abstract

Identifying local extinctions is integral to estimating species richness and geographic range changes and informing extinction risk assessments. However, the species occurrence records underpinning these estimates are frequently compromised by a lack of recorded species absences making it impossible to distinguish between local extinction and lack of survey effort—for a rigorously compiled database of European and Asian Galliformes, approximately 40% of half-degree cells contain records from before but not after 1980. We investigate the distribution of these cells, finding differences between the Palaearctic (forests, low mean human influence index (HII), outside protected areas (PAs)) and Indo-Malaya (grassland, high mean HII, outside PAs). Such cells also occur more in less peaceful countries. We show that different interpretations of these cells can lead to large over/under-estimations of species richness and extent of occurrences, potentially misleading prioritization and extinction risk assessment schemes. To avoid mistakes, local extinctions inferred from sightings records need to account for the history of survey effort in a locality.

Highlights

Identifying local extinctions is central to documenting changing geographic ranges and informing assessments of species extinction risk
We hypothesized that the occurrence of data-absent cells would be affected by (i) biogeographic realm; (ii) land cover type; (iii) protected area (PA) status and (iv) mean human influence index (HII) [6] per cell
In Indo-Malaya, data-absent cells were significantly more likely to be found outside PAs but in contrast to the Palaearctic, data-absent cells were more likely to be found in areas of high mean HII and in grassland

Summary

Introduction

Identifying local extinctions is central to documenting changing geographic ranges and informing assessments of species extinction risk. We hypothesized that the occurrence of data-absent cells would be affected by (i) biogeographic realm (via a differing history of anthropogenic land conversion and scientific infrastructure); (ii) land cover type (via ease of access for both habitat conversion and conducting surveys); (iii) protected area (PA) status (local extinctions may be more likely to occur outside PAs, PAs may be more attractive to recorders owing to high biodiversity and greater accessibility) and (iv) mean human influence index (HII) [6] per cell (areas of high HII are likely to be both more accessible and more closely associated with local extinction).

Results

Conclusion