Ten of the most widely used datasets to measure biodiversity changes and inform conservation policies significantly underestimate the level of uncertainty, and could thus lead to unreliable estimations.
That’s according to a study led by researchers at the University of Sheffield and published in the journal Nature, which stressed the need for more accuracy in biodiversity trend assessments.
National and global policies are informed by the estimation of biodiversity loss calculated using specific datasets. However, according to the study, these datasets often fail to monitor biodiversity properly over time, leading to “uncertainty, incorrect trends, and poorly resolved prediction”, which could undermine the current interpretation of wildlife abundance trends.
The study examined the approaches of ten influential abundance datasets: BioTIME, Living Planet, the North American Breeding Bird Survey, FishGlobe, RivFishTime, UK Fish Counts, ReSurveyGermany, TimeFISH, CaPTrends (on large carnivores), and the European biodiversity dataset.
“The abundance datasets we analyse are influential in policy, tracking progress towards biodiversity targets at national and international scales, and so it is vital that any inference gained is both valid and reliable,” the report said.
To evaluate the diversity of approaches used to model abundance change over time, researchers analysed 44 relevant studies identified in a literature search spanning 282 published papers.
Results showed a “pronounced shift” in abundance trends between the examined datasets and the model developed by researchers, finding that existing approaches all underestimate the level of uncertainty.
“This imposes a risk that past estimates of abundance change – pointing to declines, no net change, and recovery – may be unreliable,” the researchers claimed.
The underestimation happened because the datasets analysed ignored temporal and spatial factors and the evolutionary history of the species, the study said, showing that changes in the abundance of wild species populations vary greatly across the planet.
“This does not mean that biodiversity is not declining, simply the uncertainty is too high to detect trends in biodiversity,” said Thomas Frederick Johnson, researcher at the University of Sheffield and lead author of the paper.
“Despite this uncertainty at vast scales, we reveal improved local-scale prediction accuracy by accounting for spatial, temporal, and phylogenetic structures,” the authors said.
“Improved prediction offers hope of estimating biodiversity change at policy-relevant scales, guiding adaptive conservation responses.”
Uncertainty in measuring biodiversity change remains one of the main issues in addressing nature loss, since it could lead to misguided conservation measures and thwart efforts to meet the Global Biodiversity Framework targets.
According to a separate study authored by two professors at Canada’s University of McGill and published in the journal Science Advances earlier this year, a global biodiversity monitoring initiative is needed to address “patchy” data problems.
By Giada Ferraglioni – giada@carbon-pulse.com
** Click here to sign up to our twice-weekly biodiversity newsletter **
