Planet Earth Online

NERC = science links

JavaScript seems to be = disabled on=20 your browser - some functions on this site will not work without it (details= )=20

Features = &=20 special reports

Surveying rain forest clearance

Amazonia: pathways to policy-makers

29 September 2008

The complexity of change in tropical forests is challenging = researchers. But,=20 says Alan Grainger, so is communicating findings to governments and the = public.=20

Citation indices may be controversial, but they provide some measure = of how=20 effectively scientific knowledge is transferred to fellow scientists. = Assessing=20 how our discoveries are used by everyone else - governments, industry, = charities=20 and the public - is more difficult.

The challenge is particularly acute for knowledge about global = change, owing=20 to its complexity and spatial diversity. Our experiences at the School = of=20 Geography at the University of Leeds in communicating findings about = tropical=20 forests testify to that. In this article I use these experiences to shed = light=20 on the various paths along which knowledge travels to policy-makers, and = how=20 scientists can increase their chances of achieving meaningful = impacts.

Undisturbed forests

In 1998, while my colleague Oliver Phillips held a NERC Fellowship at = the=20 University of Leeds, he led an international team of researchers that = published=20 a groundbreaking paper in the American journal Science. Using detailed = evidence=20 collected in permanent forest plots, the team showed that undisturbed = forests in=20 Amazonia were continuing to take in, or sequester, carbon. Conventional = wisdom,=20 on the other hand, held that undisturbed forests were merely static = stores of=20 carbon.

Any challenge to accepted views is likely to spark off a vigorous = debate, and=20 this case was no exception. The paper has been cited over 200 times at = the last=20 count.

Oliver's response was to collect more data to consolidate his = findings, and=20 see whether this phenomenon was replicated elsewhere. Together with Tim = Baker=20 and Simon Lewis at Leeds, Yadvinder Malhi at Oxford, Percy Nu=C3=B1ez = Vargas in=20 Cusco, Peru, Niro Higuchi in Manaus, Brazil, and others, he established = a=20 tropics-wide network of permanent plots. The result was the Amazon = Forest=20 Inventory Network, more commonly known by its Spanish acronym = RAINFOR.

RAINFOR was recently awarded a $4.5 million grant from the US-based = Gordon=20 and Betty Moore Foundation to continue its research in Latin America, = and a =C2=A33=20 million grant from NERC to better quantify the carbon balance of = Amazonia. Simon=20 Lewis has also received a NERC New Investigator's Award to check if = undisturbed=20 African and Asian forests are also sequestering carbon.

Further support for undisturbed tropical forests being net carbon = sinks -=20 they store carbon - came from measurements of the global distribution of = carbon=20 dioxide in the atmosphere, reported in Science in 2007. Another member = of the=20 Leeds RAINFOR team, Jon Lloyd, was a co-author.

Any challenge to accepted views is likely = to spark=20 off a vigorous debate.

The RAINFOR network is prolific in publishing the results of its = research in=20 scientific papers. These have taken two routes to policy-makers. First, = by=20 reaching governments indirectly through major reports from the UN=20 Intergovernmental Panel for Climate Change (IPCC). The IPCC reports form = the=20 main evidence base for global climate policy.

The 1998 paper was mentioned in the IPCC's Special Report on Land = Use, Land=20 Use Change and Forestry, published in 2000, and in the physical science = section=20 of its Fourth Assessment Report, published in 2007.

Second, they have taken more direct paths. Assessing the effect of an = individual scientific paper on national policies on climate change is = tricky=20 because policy documents often cite IPCC synthesis reports, not = particular=20 articles.

But, RAINFOR has clearly influenced the British government, because = it=20 circulated a report by Simon Lewis synthesising RAINFOR's findings to = delegates=20 at the 23rd session of the Conference of the Parties to the UN Framework = Convention on Climate Change (FCCC), held in Montreal in 2005.

A golden lion tamarin of Brazil, an endangered species and among the = rarest=20 animals in the world.

RAINFOR's findings are important for policy-makers because they = strengthen=20 the case for including conservation of undisturbed forests in strategies = to=20 combat climate change. They have been taken on board by carbon-offset = schemes=20 and other private-sector initiatives. Brazil has also just launched an = =C2=A311=20 billion appeal to protect forests in Amazonia.

However, RAINFOR's work is still to become embedded in international = policy.=20 One reason is that developing countries are not yet required to = implement the=20 Kyoto Protocol. Another is that under the rules of the Protocol, even = full=20 partners can only gain carbon credits to offset their fossil fuel = emissions by=20 planting new forests, not sustainably managing or conserving existing = natural=20 forests. A further complication is that undisturbed forests are not = regarded as=20 sustainably managed.

So this and other pieces of knowledge about tropical forests are = currently in=20 a 'holding pattern' in IPCC's knowledge bank, waiting to be used when = the scope=20 of the Kyoto Protocol is extended.

This could happen soon if developing countries agree to implement the = successor to the Kyoto Protocol when its mandate runs out in 2012. There = are=20 also signs that restrictions on including existing forests could be = relaxed. At=20 the last FCCC meeting in Bali in December 2007 proposals were tabled to = launch a=20 scheme to avoid deforestation. This is called REDD - short for Reducing=20 greenhouse gas Emissions from Deforestation and forest Degradation.

Going, going, gone

How will climate change affect biodiversity in the Amazon? Oliver and = I=20 supervised NERC PhD student Lera Miles as she developed an innovative = technique=20 to model the potential impacts of global climate change on forests in = Amazonia.=20

Her research predicts that climate change will lead to major spatial = shifts=20 in species distributions. This has huge policy significance. It means = that=20 climate change must be tackled urgently if the existing rate of = degradation of=20 biodiversity is not to accelerate. Planners must also retain natural = forest=20 corridors over large areas to allow species to adapt as climate changes. = Otherwise many species could become extinct.

Lera published her findings in Global Ecology and Biogeography in = 2004. She=20 also contributed to a more comprehensive study published in Nature in = the same=20 year, coordinated by Chris Thomas, formerly of the School of Biology at = Leeds.=20

Publishing in prestigious journals is a recipe for high scientific = impact.=20 The Nature paper, which estimated that at least 18 per cent of species = could=20 become 'committed to extinction' by 2050, has since been cited more than = 500=20 times by fellow scientists. More detailed studies, like Lera's fine = paper in=20 Global Ecology and Biogeography, are often overshadowed by these = megapapers. But=20 the IPCC's Fourth Assessment Report in 2007 mentioned both studies in = its=20 impacts, adaptation and vulnerability section.

Top journals also bring massive media exposure, and policy-makers = could not=20 fail to notice the findings of the Nature paper given all the newspaper=20 headlines it received.
Extending the reach of science

Scientific knowledge can easily get distorted on its way from the lab = to=20 government offices, via scientific reports or the media. The risk is = even=20 greater when knowledge is disseminated globally. So the personal touch = is still=20 important.

Policy-makers, for example, often respond strongly to research = carried out by=20 their own scientists. This is one advantage of multinational scientific = networks=20 like RAINFOR. I am also coordinating efforts to train researchers in = South-east=20 Asia and Latin America to adapt Lera's model to predict the impact of = climate=20 change on domestic distributions of biodiversity.

Scientists can also increase their influence by direct contacts with=20 policy-makers. Addressing conferences attended by ministers is one way = to do=20 this. But greater impacts can be obtained through dialogue. According to = David=20 Guston's influential boundary organisation model, 'knowledge systems' = that=20 harness scientific discoveries are most effective when scientists and=20 policy-makers work together to translate these into a form best = understood by=20 politicians.

Lera has had the opportunity to interact with policy-makers by = representing=20 the UN Environment Programme's World Conservation Monitoring Centre = (WCMC) in=20 Cambridge at meetings of the Conventions on Climate Change and = Biodiversity,=20 including the Bali Climate Conference last December. She joined WCMC on = leaving=20 Leeds in 2002, and became acting head of its Climate Change and = Biodiversity=20 Programme last year. Her joint assessment of the REDD scheme appeared in = Science=20 in June.

Forest decline?

Success in communicating research findings to anyone outside the = scientific=20 community is always limited by their preconceptions. Policy-makers often = assume=20 that scientists can provide them with certain knowledge. So = communicating the=20 uncertainty associated with global change needs great care.

I encountered this problem when I challenged another long-held = assumption -=20 that tropical forest area is in long-term decline - in a paper published = in the=20 Proceedings of the National Academy of Sciences in January. This = concluded that=20 the evidence for forest decline provided by frequent estimates of = deforestation=20 rates is not, paradoxically, matched by convincing evidence for a net = downward=20 trend in forest area.

The paper did not deny that deforestation is still continuing on a = large=20 scale, but showed that there are doubts about the reliability of UN = forest=20 statistics, on which reports of deforestation are often based. It also = contained=20 a new time series, which included recent satellite data, that showed no = evidence=20 for a net decline in forest area in the humid tropics since 1970. My = analysis=20 was supported by a more comprehensive study in the Journal of Official=20 Statistics. This showed that inconsistencies in UN area statistics are = matched=20 by other inconsistencies in every stage of their production.

UN forest statistics have been widely used in modelling land-use = change,=20 calculating the global carbon budget, and predicting future trends in=20 biodiversity. So the paper has major implications for global - change=20 scientists, and has already been cited four times in the academic=20 literature.

One explanation for my discovery is that any net decline in tropical = moist=20 forest area is masked by the large errors involved in making global = estimates.=20 But another is that tropical forest change is more complex than = previously=20 imagined, and that deforestation is being offset by natural = reforestation in=20 abandoned clearances on a larger scale than shown by earlier estimates. =

To test my hypothesis about natural reforestation I am following the = example=20 of RAINFOR by forming a global network of scientists to collate = measurements of=20 this phenomenon throughout the tropics. The first synthesis of findings = could be=20 out later this year.
This may affect international policy on climate = change.=20 For if developing countries agree to implement 'Kyoto II' they could ask = to=20 receive credits for carbon being sequestered in regrowing forests, as = well as=20 that added in undisturbed forests.

Drawing attention to the uncertainty of our knowledge about the trend = in=20 tropical forest area was risky, since policy-makers could use this as an = excuse=20 not to take action on tropical forests. To guard against this I stressed = the=20 importance of errors in forest estimates, and at the end of the paper = called for=20 a new global monitoring scheme to reduce these.

Balancing the assessment of scientific uncertainty with a positive = policy=20 message paid dividends. It has got people all over the world talking = about the=20 need for a World Forest Observatory.

To ensure that my message got across clearly I spent a lot of time = before=20 publication working on a press release with our excellent Leeds = University press=20 office.
The potential to influence policy-makers and the public was = boosted=20 by an interview on BBC radio. The BBC then listed my findings on its = website.=20 This, and the press release, led to worldwide coverage, in newspapers as = far=20 afield as Australia, Brazil and the Philippines.

From policy to practice

It is one thing to influence policy. It is quite another for policies = to lead=20 to practical outcomes on the ground.

So if carbon credits are to be given for forest regrowth and = protecting=20 forests under the REDD scheme, this will need meticulous checking. This = means=20 that tropical forests must be monitored with greater accuracy than = current=20 national and international mechanisms permit. So the idea for a World = Forest=20 Observatory could be quite timely.

From science to policy

Devising a social impact index comparable to a citations index is not = impossible, but it is made difficult by the many pathways research = findings take=20 to reach governments, and by the frequent gulf between government = policies and=20 outcomes on the ground.

Our experiences at Leeds show that scientists can communicate their = findings=20 effectively by exploiting as many of these routes as possible. Keeping = the=20 scientific message clear is even more important in complex global-change = studies, and having a distinct policy message also helps.

While scientific papers and media reports have a huge impact, the = personal=20 touch is still important. That is why international networks of = scientists like=20 RAINFOR are not only useful for collecting data - every scientist is a = potential=20 communicator to their own government. NERC recently recognised this by = extending=20 its Knowledge Exchange Programme to include international partners.

One thing that scientists have little influence over, however, is the = long=20 time-lag between the publication of research findings and their = translation into=20 policy documents and practical outcomes. Governments should remember = this when=20 evaluating the impacts of their science spending!

Dr Alan Grainger and Professor=20 Oliver Phillips are members of the School of Geography at the = University=20 of Leeds.

In March 2008, NERC launched the =C2=A33m Amazonia consortium led by = Emanuel Gloor=20 from the University of Leeds.

Amazon in numbers

Amazonia's tropical forests are some of the most important ecosystems = on=20 Earth. They account for 45 per cent of the world's tropical forests and = store=20 around 40 per cent of the global carbon held in vegetation on land.

Relatively small changes in the structure and function of these = forests are=20 likely to have far-reaching affects on biodiversity, the carbon cycle = and the=20 rate of climate change.
In a series of groundbreaking publications = going back=20 to 1998 a team of researchers from the University of Leeds has shown = large-scale=20 patterns and changes in the structure, metabolism, function and = composition of=20 one of the world's richest ecosystems.

A key finding is the discovery that trees in the Amazon are getting = bigger=20 across the basin. Most species of trees have begun growing faster. = Existing=20 trees are also dying faster and being replaced by young new trees. This = may seem=20 obvious because plants grow by fixing carbon dioxide from the air by=20 photosynthesis. So higher carbon dioxide levels will act as a fertiliser = promoting plant growth. But the discovery has surprised most tropical = ecologists=20 as they had assumed mature forests to be carbon-neutral, giving off as = much=20 carbon as they absorb.

These increases in biomass suggest the Amazon is a sink for = atmospheric=20 carbon dioxide, taking in 0.5 to 1 billion tonnes of carbon per year -=20 equivalent to the fossil-fuel emissions of the major economies of = Western=20 Europe. Or put another way, Amazonia's biomass, at least for now, is = absorbing=20 15 per cent of carbon dioxide emissions from human activities.

Zones of tension

Jon Lloyd from the University of Leeds leads the =C2=A31.6m NERC = consortium=20 Tropical Biomes in Transition (TROBIT). Jon and colleagues are = interested in=20 'zones of tension' in Africa, Australia and South America. These are = where=20 rainforest and savannah, the other major vegetation system in the = tropics, grow=20 side by side.

Oliver Phillips and Jon Lloyd lead the global Amazon Forest = Inventory=20 Network (or RAINFOR network). RAINFOR, which is funded by a range of=20 international partners, coordinates research on tropical forest ecology = across=20 four continents.

Simon Lewis leads a NERC New Investigator's Award to extend = evaluations of=20 change in undisturbed tropical forests into Africa and=20 Asia.

Planet Earth Online

Features = &=20 special reports

Amazonia: pathways to policy-makers

Undisturbed forests

Going, going, gone

Forest decline?

From policy to practice

From science to policy

Amazon in numbers

Zones of tension

Tools

Related links

News & information

Features & special reports

Most popular = features