Are earthworms really in decline? Representative data and rigorous models are needed to assess large-scale, long-term trends in earthworm populations

Aidan M Keith, Frank Ashwood, Rob James Boyd, Kevin Butt, Kelly Mason, Fiona Seaton, Olaf Schmidt

Declines in aboveground invertebrates have been reported widely but similar assessments belowground are rare. A recent quantitative study aimed to address this gap, presenting findings which suggested alarming long-term declines in earthworm abundance in the UK. Estimating temporal trends in abundance from diverse sources of data presents many challenges and there is a need for rigour toward inference. However, a comprehensive review of the published dataset revealed numerous issues including extensive errors, omissions and various problematic aspects of the data and modelling. For the present study, we have revised the published dataset, with data from 70 of 91 original sources being corrected and/or amended, and augmented with data from additional sources. Reanalysing the revised and augmented datasets and assessing the risk of bias for estimating temporal trends, we demonstrate that the original conclusions are not robust and that the large-scale declines suggested are not supported by the corrected data. The importance of earthworms for soil functional processes and as a food source for animals is well established and therefore significant declines would have critical ecological implications. It is conceivable that earthworm populations in the UK may have declined to some extent in particular environmental contexts, however sound conclusions on temporal trends must be based on correct and representative data and rigorous analyses. We highlight factual errors, omissions and methodological flaws in the original study and appeal to improve data collation, enhance meta-data and encourage long-term monitoring to develop a sound understanding of trends in earthworm populations and typical variability observed under different environmental contexts.