With the world almost certain to miss a target to slow extinction rates by 2010, governments are looking to adopt new biodiversity targets next year. What those might be — and what science will be needed to underpin them — is yet to be established, researchers said last week at a conference on biodiversity held in South Africa.

"2010 may not be the year when we reverse the rate of loss of biodiversity, but needs to be the year when we reversed the response to that loss," says Achim Steiner, executive director of the United Nations Environment Programme (UNEP).

In 2002, more than 120 countries adopted a target to achieve a "significant reduction" in the rate of biodiversity loss by 2010. That aim will not be met, says Georgina Mace, director of the Centre for Population Biology at Imperial College London. But just how much the world will miss the target by is difficult to calculate, because the target does not specify a baseline extinction rate from which to start counting. "The lack of baselines and timescales are quite problematic," Mace says.

The next generation of targets will aim for a more positive outcome and set more easily measurable goals, says David Cooper of the Convention on Biological Diversity secretariat in Montreal. The targets are likely to aim for a complete halt to biodiversity loss by 2050, and to set more modest interim targets for 2020.

The new targets will be designed to encourage countries to address the underlying drivers of biodiversity loss, such as climate change and unsustainable land use, Cooper says. They will also recognize the socioeconomic value of 'biodiversity services', such as tourism revenue generated by coral reefs or the carbon sequestration value of a forest.

The new goals are being agreed through a set of international negotiations, to culminate in Japan in October 2010. There, governments will also consider whether to set up an Intergovernmental Platform on Biodiversity and Ecosystem Services to bridge the gap between science and policy.

Science advice is currently slow to reach policy-makers, says Steiner, and too piecemeal when it gets to them. "The number of scientific assessments, their focus and assumptions, are simply too bewildering and fragmented," he told the Cape Town conference.

Scientists also need to find better ways of estimating biodiversity loss, Mace says. Writing in Science last month, a team led by UNEP's Matt Walpole identified serious shortcomings in the indicators used to measure progress towards the 2010 target, such as monitoring changes in species' status on the International Union for Conservation of Nature's endangered 'red list' or in the size of protected areas. But the indicators do not include any measurement of the effects of climate change on biodiversity, and few address the societal benefits of safeguarding species (M. Walpole et al. Science 325, 1503–1504; 2009).

Cooper adds that scientists need to better integrate data to provide a more holistic picture of biodiversity trends. "We need to develop a monitoring system," he told Nature.

The Biodiversity Observation Network of the Group on Earth Observations offers such a system for 'joining the dots' of global biodiversity knowledge. Launched in 2008, the network aims to be a one-stop shop for information on biodiversity. It will release indicators, generate maps of priority areas, provide conservation plans and report on trends in uses of biological resources.

Chairman Bob Scholes, of South Africa's Council for Scientific and Industrial Research, says the network will start producing a "strong element" of this bigger picture in two to three years' time.