Biodiversity data is distorted by past inequities. Scientists are wrestling to get a clearer picture.

The stakes are high. A new international agreement promises $30 billion a year to protect biodiversity. Can data-driven decisions avoid previous traps?

This article, initially featured on Future Earth’s Anthropocene journal website on January 4, 2024. Future Earth is an affiliated body of the ISC.

Maps of biodiversity around the world don’t just illuminate the plants and animals living everywhere. They also trace layers of human history that shape our picture of the planet’s inhabitants, including legacies of injustice.

Species are spotted disproportionately in wealthier countries with more scientists scouring the landscape, meaning North America, Europe and Australia get an outsized amount of attention.

Social turmoil can skew things. Ecological observations coming from Cambodia in southeast Asia collapsed during the 1970s and 80s, a time of civil war and the murderous Khmer Rouge regime.

Even at the local level, past discrimination can influence which areas are more nature-rich. In the U.S., discriminatory racial restrictions on who could buy houses in certain neighborhoods—known as redlining—meant that whiter, wealthier neighborhoods had more green space and, consequently, roughly double the number of birds spotted.

As efforts to slow the decline of biodiversity wins backing from governments and conservation groups, some scientists are warning that unless care is taken, this legacy of inequity could be reinforced by the biodiversity data on which many rely.

“Biodiversity data trace not only the distributions species but also cities and roads, the rise of surveillance technology, shadows of colonial histories, and echoes of contemporary racial and economic inequity,” Millie Chapman, an ecologist and postdoctoral researcher at the National Center for Ecological Analysis and Synthesis, at the University of California, Santa Barbara, wrote in an email.

The stakes are high. The 2022 adoption of a new global biodiversity agreement, known as the Kunming-Montreal Global Biodiversity Framework, included a pledge to increase financing for biodiversity work to $30 billion per year by 2030. Scientific data about species can influence where that money gets spent.

The growing attention to biodiversity helped spur Chapman and a group of other ecologists, sociologists, computer scientists, and political ecologists to wrestle with how to address shortcomings in the data and how it’s used. Some of their key findings are spelled out in a new paper in Science, published today.

The Global Biodiversity Information Facility is a prime example of the ways ecological data collide with social history. The government-funded international database compiles more than 2.6 billion observations of species spanning the globe. The data is meant to help inform policy decisions about a host of conservation-related projects, such as managing endangered or invasive species. But even a glance at the facility’s data map shows that it doesn’t match with biodiversity hotspots. While the United States and Europe are replete with observations, the rainforests of central Africa—places far richer in species – are relatively blank.

This problem is well known among ecologists and can be partly corrected by statistical models. But Chapman and her collaborators warn that the challenges run much deeper.

“Without directly addressing and correcting for social and political disparities in data, the conservation community will likely fall into the same traps that other domains do—entrenching the inequities of the past and present in future decision-making through data,” they write.

One possibility is increasing the number of observations with new tools, including programs that recruit non-scientists to help gather data, new sensors that can gather environmental data with less effort, and environmental DNA (eDNA), which detects species from bits of DNA floating in the air or water. But these tools can also be pitfalls. While they hold the promise of filling data gaps, there is evidence that new data sources are echoing the imbalances of the past, the authors warned.

More nuanced modeling could also help. But again, the researchers caution that it will be hard to account for so many social variables. While it’s one thing to control for factors like how close an area is to roads or towns, it’s far more difficult to trace the effects of decisions such as who gets scientific funding.

The researchers write that one solution is a richer understanding of the contexts in which data is being collected. That includes working with local people and institutions to better understand the social and historical conditions in a place, and how they might influence information about biodiversity. As an example of such community-based systems, the scientists point to International Forestry Resources and Institutions, an alliance of more than a dozen research institutions around the world that conduct local forest-related research using share methods. That research isn’t just about counting species, but also examining the social factors shaping the forests.

The shortcomings of more data or fancier models doesn’t mean there is no hope for building a clearer picture of global biodiversity, the authors write. But it does mean it will take labor-intensive work and close attention to local circumstances. “It means,” they write, “there are no shortcuts.”

Disclaimer
The information, opinions and recommendations presented in this article are those of the individual contributors, and do not necessarily reflect the values and beliefs of the International Science Council.