Protecting designated areas of ecological value is one of the most popular conservation tools for safeguarding biodiversity. As deforestation advances, the theory goes, these protected areas offer sanctuary for threatened species and natural ecosystem processes. By 2020, the 193 nations that have signed up to the Convention on Biological Diversity — the United States is not among them — hope to protect at least 17% of the planet's biodiversity-rich areas in this way. But is drawing artificial lines around ecologically valuable land an effective method for protecting biodiversity?

As a conservation mechanism, protected areas have a chequered history, and have faced particular criticism because some exclude poor local populations who want to gather food, wood and other resources from the forests on which they depend. Management of the thousands of protected areas varies around the world, and biodiversity scientists have long suspected that areas with better safeguards to stop illegal encroachment for logging, agriculture or other activities are more likely to be a safe haven for biodiversity. However, demonstrating that such exclusion is effective is not straightforward.

Around half of the reserves are experiencing a severe loss of biodiversity.

Assessments of ecosystems and biodiversity are hampered by piecemeal data collection that uses incomparable methodologies, and the data have large gaps. That makes it difficult to draw broad conclusions about the global health of biodiversity, which many scientists say is a key reason why biodiversity has failed to climb political agendas, even as extinctions of animals and plants continue to rise. Around one-fifth of all 5,490 described mammal species are at risk of extinction, according to data from the International Union for Conservation of Nature.

These difficulties are driven by funding constraints and the intricacies of monitoring and assessing complex ecosystems. Nevertheless, some informative evaluations do exist, including the biennial Living Planet Index drawn up by the conservation group WWF, which follows trends in populations of species around the world to give indications of the state of global biological diversity.

More of these data gaps have now been filled, thanks to an assessment of protected tropical forest areas led by William Laurance, a conservation biologist at James Cook University in Cairns, Australia, which offers a snapshot of global biodiversity and reviews the success of this tool.

In a Herculean effort, Laurance and his team, along with more than 100 co-authors, systematically collected standardized data on environmental changes over the past 20–30 years in 60 protected areas across the world's major tropical regions of Africa, America and Asia. The data include changes in the abundance of 31 groups of species, including primates, freshwater fish and exotic plants, and 21 potential drivers of environmental change, such as road building and hunting.

To gather the information, the team conducted 262 interviews with field biologists and environmental scientists, and asked them each to complete a detailed 10-page questionnaire. It took the team around four years to gather the data, but Laurance reckons that to do so from scratch would cost billions of dollars and take 20–30 years. The effort paid off, and the results appear online on Nature's website this week (W. Laurance et al. Nature http://dx.doi.org/10.1038/nature11318; 2012).

The team found that around half of the reserves are experiencing a severe loss of biodiversity. Crucially, the researchers also demonstrated the negative impact that environmental changes immediately outside the areas have on the health of the habitat inside the reserves. And they found that destructive activities such as forest clearance, fires and logging increasingly reach up to the edge of the protected areas. The results show that 85% of the reserves experienced declines in surrounding forest cover over the study period, whereas only 2% gained forest. These findings will not surprise biodiversity scientists, who have long been aware of such trends, but now have the data to show it.

The work highlights, yet again, the growing challenges that threaten the success of the protected-reserves model. Many reserves simply do not function as intended, but there are few alternative approaches. The results underscore the importance of better management of areas around the reserves. Rather than treating these ecological havens as islands, Laurance and his team recommend implementing buffer zones around protected areas to cushion the blow. The reserves can protect wildlife, but we must first protect the reserves.