Three Datasets Agree: Amazon Deforestation Has Been Reduced

March 12, 2013 | 4:12 pm
Doug Boucher
Former Contributor

Has there been any progress at all in reducing global warming pollution? Is anybody doing anything to deal with climate change seriously? Is it all bad news, or are there at least a few rays of hope? As a scientist, I try to deal with these kinds of questions by looking at data rather than just listening to the radio or watching the TV news, which can be very discouraging. And in the last few months, three new datasets have been released that show us what has been happening to Amazon deforestation since the 1990s. Although they have lots of differences among them, they do agree that in the Amazon — the world’s largest expanse of tropical forest — there has indeed been some progress.

Amazon deforestation is important to climate because tropical forests are enormous storehouses of carbon, and when they are cut down, that carbon goes into the atmosphere as the principal greenhouse gas, carbon dioxide. Thus, reducing and eventually ending deforestation, in addition to its enormous value for biodiversity and the rights of Indigenous Peoples, can also be a very positive contribution to averting the worst consequences of global warming.

So, what are the new datasets, and what do they show?

The most “official” one is that of the Food and Agriculture Organization of the United Nations (FAO), which gives data on net forest conversion on the newly updated version of its FAOSTAT web site. This data is provided to FAO by government agencies in each country.

The RAISG data set is from the Red Amazónica de Información Socioambiental Georreferenciada (Amazon Network of Georeferenced Socio-Environmental Information), in tables TDF1 and TDF2 of its new atlas, Amazonía Bajo Presión (Amazonia Under Pressure), and comes mostly from the analysis of remote sensing data.

Finally, a dataset assembled by Tim Killeen, now at the World Wildlife Fund, comes from direct contacts with governmental and non-governmental experts in each of the Amazon countries as well as from both published and “gray literature” sources. (We have made this data available on the UCS website, but it is Tim’s dataset and anyone wishing to use it should contact him at [email protected].) This data was the basis of news stories on NPR and in the Washington Post last fall.

As you’d expect with data from different sources, in different time periods, with different methods and indeed different definitions of “forest” and “deforestation,” the numbers are not identical and in some respects not totally consistent. But the other side of this coin is that having three somewhat independent datasets, one can be much more certain of results if they show up in all three of them. A trend is much more likely to be a real one if we see it consistently, no matter which data source we look at.

Here are the graphs showing annual deforestation rates, by country, in hectares per year. (A hectare is about 2 ½ acres, or a little less than two U.S. football fields.)

Mean annual Amazon deforestation (ha/year) from the 1990s to the late 2000s according to the FAOSTAT dataset.


Mean annual Amazon deforestation (ha/year) from the early and late 2000s according to the RAISG dataset.

Mean annual Amazon deforestation (ha/year) from the 1990s to 2011 according to the Killeen dataset.

A few trends are clear and consistent in all three datasets:

  • Deforestation in the Amazon as a whole rose from the 1990s to the early 2000s, and has fallen since then. FAOSTAT shows a moderate decrease, while the RAISG and Killeen datasets indicate that the drop has been quite large, but all three agree that Amazon deforestation has been reduced in recent years. That’s the good news.
  • The not-so-good news is that essentially all of this reduction has come from the efforts of Brazil. In fact, in the Amazon outside of Brazil there’s no sign of progress – the FAOSTAT data show a slight increase, the RAISG data a slight decrease, and the Killeen data a more substantial increase.
  • But the slightly better news is that most of the Amazon forest, and most of the deforestation, is in Brazil, so that what happens in the Brazilian Amazon by and large determines what happens to the whole of the Amazon forest. Brazil has nearly 60% of the Amazon forest (Killeen) and until recently had more than 80 percent of the Amazon deforestation (RAISG), so its success has had an overwhelming effect.
  • Comparing the three datasets, it’s notable and a bit unexpected that the “official” data submitted by the countries to the FAOSTAT database actually shows the least progress. It indicates a considerably higher level of deforestation overall, and a lesser reduction, than is shown by the other two datasets.
  • Finally, apart from the question of whether deforestation has been reduced, are the levels high or low? By and large, they’re low compared to the global average for tropical deforestation, which is that about 0.5% of forest area is lost annually. It’s well below that in the most recent data from almost all the Amazon countries, ranging from 0.02% to 0.34% annually according to the RAISG figures and from 0.02% to 0.28% (except for Bolivia, where it is 0.76%) according to the Killeen data.

So, the overall story of the Amazon is a mildly hopeful one – relatively low rates of deforestation and a considerable reduction in less than a decade. There are two major challenges: to continue the progress made in Brazil, and to help the rest of the Amazon countries follow its positive example.