In an ambitious study that represents the latest merger between big data approaches and the quest to conserve the planet, scientists have found that across a majority of the Earth’s land surface — including some of its most important types of terrain and its most populous regions — the abundance or overall number of animals and plants of different species has fallen below a “safe” level identified by biologists.

The reason is not exactly a surprise — from grasslands to tropical forests, humans are using more and more land for agriculture, to live on, to build roads and infrastructure upon. When we take over, we clear the land or otherwise convert it for our purposes. This doesn’t always cause extinctions, but it does reduce the abundance of species and what researchers call the “intactness” of ecosystems — and when biodiversity levels fall too low, it can mean that larger ecosystems lose their resilience or even, at the extreme, cease to function.

“Exploitation of terrestrial systems has been vital for human development throughout history, but the cost to biosphere integrity has been high,” notes the study published Thursday in Science, which was led by Tim Newbold of the United Nations Environment Programme and University College London with a large group of colleagues representing several British, Australian, Danish and Swiss universities and institutions.

The researchers compiled 1.8 million separate measurements of the abundance of species (39,123 of them) at 18,659 locations across the globe — a volume of data that an accompanying essay in Science, by ecologist Tom Oliver of the University of Reading, calls the “most comprehensive quantification of global biodiversity change to date.”

The researchers then extrapolated across the rest of the planet, and compared the results to a “Biodiversity Intactness Index” to determine where species declines were too great.

The research is based on a “planetary boundaries” concept that “attempts to set some sort of safe limit to the amount of biodiversity we can lose, while biodiversity still supports important ecosystem functions,” said Newbold, the study’s lead author. And it is important to note that in the context of this analysis, safety actually means safe for humans, in significant part.

The concern is that species-anemic ecosystems will struggle or fail, and so become unable to provide us what we actually need in the form of stored carbon, filtered water, fertile soils and much else. Animals need these ecosystem “services,” to be sure, but so do humans.

“Biodiversity supports a number of functions within ecosystems, things like pollination, nutrient cycling, soil erosion control, maintenance of water quality,” Newbold said. “And there’s evidence that if you lose biodiversity, that these functions don’t happen as well as they would have done in the past.”

As a conservative or precautionary standard, the researchers therefore assumed that a decline of more than 10 percent of species abundance in a given area (compared with what that abundance was before human interference) represented crossing into a danger zone for biodiversity. But their study found that overall, across the globe, the average decline is already more like 15 percent. In other words, original species are only about 85 percent as abundant (84.6 percent to be precise) as they were before human land-use changes.

Some places are, of course, better off than others — for instance, northern tundras and boreal forest ecosystems were still relatively intact, the study found. So was much of the Amazon rain forest. In contrast, central North America showed a huge gash on the researchers’ maps, representing a large region with less than 60 percent of its original biodiversity intact, stretching all the way from Canada to Texas.

Overall, 58 percent of the Earth had declined below 90 percent biodiversity intactness and, in effect, into the danger zone. And this was strongly correlated with human population — that 58 percent of the Earth is the home to 71 percent of its human inhabitants, the study reported.

There are, to be sure, some major uncertainties (and matters of interpretation) in this analysis, ones that the authors freely acknowledge. For instance: Who is to say that 90 percent “intactness,” or abundance of the original species that lived in an ecosystem, is the right number in all cases?

And moreover, it isn’t just that ecosystems have been losing original species — they have also been gaining, in many cases, non-original or “invasive” species. So is that a net plus to them, or a net minus?

The study considered these options and, not surprisingly, found that if new species are considered to benefit ecosystems, or if ecosystems can go down to 80 or 70 percent of their original species abundance, then considerably less of the world is in trouble. In the end, then, this really boils down to a decision about how much risk you want to take with nature.

It is not, therefore, that ecosystems are about to start collapsing all around us because of passing this 90 percent threshold. However, the research does mean that less-intact ecosystems will be less able to withstand future challenges like ongoing global warming, Newbold said.

“We’re entering a space where things become more uncertain, and we expect that things will be less resilient in the face of other changes,” he said.

Immediate reactions to the work were mixed from two experts consulted by the Post to respond to the study.

“Newbold and colleagues find sobering evidence that we have already crossed that line in terrestrial ecosystems,” said Mark Urban, who directs a newly founded Institute of Biological Risk at the University of Connecticut, which focuses on biodiversity losses. “Human land use has reduced local populations to 85 percent of original abundances on average. What this means is we have not only crossed a planetary boundary, but have kept going. At least now we’re looking back.”

Urban added that while the new study stresses the impact of changes in land use, “this result ignores the accelerating threat from a warming climate. Climate change is about to make things more complicated as we try to pull back from the edge of the Earth’s resilience.”

But Erle Ellis, who directs the Laboratory for Anthropogenic Landscape Ecology at the University of Maryland, Baltimore County, had a different take. While Ellis said the new study “should be widely read and thought about,” he challenged the basis for the 90 percent figure used in the paper, suggesting the threshold was “arbitrary.”

“There is no rigorous scientific basis behind the concept that there is a single measurable ‘safe space’ for biodiversity change,” Ellis said by email.

“As a policy message, with numbers and maps of how much and where ‘safe limits’ on biodiversity change are exceeded, this paper goes way beyond the science — and this could lead to very faulty decisions on strategies and priorities for conservation,” he added.

There is no dispute, then, that humans have caused a significant retraction of non-human biological life on land — that they have, in effect, acted very effectively as its competition. Nor is there much dispute that human-caused climate change will act on top of this preexisting disturbance as an exacerbating factor.

The question, then, remains how close we are to a point where global ecosystems could see major tipping points, or dramatic changes of a sort ultimately traceable to us.