LIFE at the edges of extinction

Produced by an international research team, the LIFE metric helps confront the challenge of human-driven habitat loss and its severe impacts on biodiversity. By combining existing methodologies with powerful new computing technology, the team assessed how land-use changes impact extinction risk for over 30 000 vertebrate species. The LIFE metric is visualized through global maps offering an unprecedented level of  species-specific insight into how land-use changes – like converting forest to arable land, or restoring arable land to forest – affect species’ survival.

The team is enthusiastic about LIFE’s prospects:

“To date, very few metrics for biodiversity really get at the extinction risk for individual species; the great thing about LIFE is that this is exactly what it does”, commented co-author Jonathan Green, a Senior Research Fellow at SEI York. “LIFE looks at extinction risk, and it is elevating the importance of species for whom the loss of any further range is going to be most damaging”, he added.

Green worked alongside colleagues from the University of Cambridge and the Universidad de Austral de Chile to design the tool which generates these data-rich maps.

Bigger better computers?

LIFE’s foundations were formed back in 2016 through a project co-designed with WWF, which assessed the impact of agricultural commodity production on biodiversity in the Brazilian Cerrado, the world’s most biodiverse savanna. Working at a regional level, they developed a “persistence score” approach to assess extinction risk. Their model was so intelligent that they could calculate how land-use changes impact extinction risk probability for individual species. By this stage, they recognized the tool’s potential for more ambitious applications but lacked the necessary computing power.

Scaling up such a dynamic tool to a global scale required bigger, better computers and some heavy coding expertise: this is exactly what the team got when they linked up with colleagues from the University of Cambridge’s computer science department. Newly available high-performance computing and optimized coding enabled them to run their model using IUCN habitat data for over 30 000 vertebrate species. The technological leap significantly boosted processing speed:

“When we started it took us three weeks to calculate extinction risk of all species; we got it down to three days, and now we can do it in a day”, explained co-author Michael Dales, University of Cambridge.

Striking variations in global extinction risk

Mapping and measuring the impacts of land-use change is all a question of relative risk, the team’s efforts show. LIFE scores are generated by comparing a species’ original habitat (prior to large-scale human activity) against its current one. Using a relationship that gives greater weight to species that have, historically, lost a greater proportion of their ranges, the extinction risk of removing further habitat is calculated for each species. Repeating this input for over 30 000 vertebrate species, the metric adds later upon layer of biodiversity data to create high-resolution maps covering the entire globe. Each map cell measures just 3.4 km², an area similar in size to New York’s Central Park.

The LIFE score is the sum of extinction risk for all species affected by land conversion within a given cell. As a result, hotspots emerge, either because an area is rich in biodiversity or because it serves as a critical habitat for a particularly threatened species. These maps not only highlight regions where biodiversity will suffer most if land is converted to arable use, but also identify where restoration efforts could yield the highest biodiversity gains.

LIFE’s calculations reveal striking variations in global extinction risk: for instance, converting 1 km² of forest in South America to arable land could cause around 35 times more extinctions than converting 1 km² in Europe, on average.

The team’s research has shown that LIFE scores and the subsequent maps can be reliably applied to land-use changes affecting areas from 0.5 km² to 1000 km². This wide range of applications demonstrates the metric’s versatility and potential impact.

“I’d really like to see us making this data even more accessible…”

It isn’t just researchers who need access to the data, Green says. Policymakers, conservationists, and professionals working in land-use, especially agricultural commodity production, also need to be able to integrate LIFE scores into their decision-making: “I’d really like to see us making this data even more accessible, and look at having regularly updated maps; maps that are online and interactive.” The collaboration with the University of Cambridge computer scientists presents an exciting opportunity to advance these ambitious plans. 

Clearly, LIFE has matured since its beginnings nearly a decade ago. Green commented that the new momentum behind establishing the metric for regular use is heartening:

“It’s great to see it transform into something that can be relied upon by so many different actors and agencies across the world. It’s been a long time getting from our proof of concept to this global mapping that can actually support policy processes. I’m really excited to see the next stages as they take shape”, he concluded.

The tool is already making an impact: it has been added to the Global Environmental Impacts of Cconsumption (GEIC) indicator – a tool developed by SEI – which has just been recognized as a UK government Official Statistic. Getting tools like LIFE into regular use by governments, conservationists, commodity producers and researchers alike will facilitate evidence-based policy and decision-making at global and local scales: LIFE really could be the key to flourishing, biodiverse lives and livelihoods all over the world.