Léo RamosFramed by a sunrise in the municipality of Senador Guiomard, in the state of Acre, a Brazil nut tree is in the foreground of the cover of the February 6, 2014 issue of the British scientific journal Nature, one of the most prestigious in the world. The tropical tree symbolized the Amazon, the central theme of an article whose principal author was Luciana Vanni Gatti, 53, coordinator of the Laboratory for Atmospheric Chemistry of the Nuclear and Energy Research Institute (IPEN). Gatti and her co-authors for this work calculated what is known as the carbon balance in the Amazon forest—a comparison between the quantity of carbon in the form of carbon dioxide (CO2) emitted and absorbed by the Amazon Basin—for two consecutive years in which temperatures were above the average for the last 30 years, but with a significant variation in rainfall.
The year 2010 was marked by an extreme drought, and 2011 had above-average rainfall. “We found that the Amazon acted as a carbon source in the dry year, when we took fires to clear land into account,” says Gatti, who published the article together with Emanuel Gloor, of the University of Leeds, and John Miller, of the University of Colorado, Boulder. “However, in the wet year, its carbon balance was close to neutral, as the quantity emitted and absorbed was more or less equivalent.” The data from the study on atmospheric gases was obtained as part of an initiative led by the Brazilian since 2010, and the research is part of an effort called Amazonica (Amazon Integrated Carbon Analysis), a large international project coordinated by Gloor. Every two weeks, small aircraft fly from four different locations in the Amazon (Santarém, Alta Floresta, Rio Branco and Tabatinga) and collect air samples all along a descending vertical trajectory from an altitude of 4.4 km to about 200–300 m from the ground. The samples are sent to Gatti’s laboratory at IPEN, where the greenhouse and other gases are measured. The study includes analyses of CO2, carbon monoxide (CO) and sulfur hexafluoride (SF6).
|Ribeirão Preto School of Philosophy, Science and Letters, University of São Paulo (USP)
National Oceanic and Atmospheric Administration (post-doctorate)
National Center of Atmospheric Research (post-doctorate)
|Nuclear and Energy Research Institute
The results were interpreted as worrisome, as they suggest that the Amazon’s capacity to absorb CO2—the principal greenhouse gas—from the atmosphere appears to be associated with the amount of rainfall. In dry years, such as 2010, there are more fire—both in the forest and in areas that have already been cleared—and they release large quantities of CO. Additionally, water stress apparently reduces plant photosynthesis levels and causes them to remove less CO2 from the atmosphere. In this interview, Gatti talks about the results and implications of her study and tells us a little about her career.
Had you imagined that the article would be on the cover of Nature?
I had thought it would be accepted by Nature more because of the importance of the subject than because of the quality of the work, but I never dreamed it would be the cover story. I attend many conferences and I meet people from all over the world talking about the Amazon. These people have no idea what the area is like. They have never come here and they perform modeling, extrapolating local data as if it were representative of the entire region. I have seen highly varied results from modeling, showing the Amazon as either a great source or sink of CO2. The Amazon is an important element in the global carbon balance. This is why determining its weight in this balance is tremendously important. What is the most popular topic nowadays? Climate Change. The planet is becoming hostile to humans. But, initially, we had intended to publish in Science.
That was my goal because [Simon] Lewis [a researcher at the University of Leeds] published a paper in Science in 2010 with conclusions that we wanted to contest. He said that, that year, the Amazon had emitted the same amount of greenhouse gases as all the fossil fuels burned in the entire United States. It was the result of modeling and the result was very exaggerated. I wanted to publish in Science in order to rebut Lewis. We even submitted a version of our article to the journal, containing just the data from 2010, which was a very dry year. The paper calculated the carbon balance for that year. Science said that it was a relevant study, but that its focus was too technical for its publication. They didn’t even send the article to be peer reviewed, and suggested that we send it to a more specialized journal. However, when we analyzed 2011 data, we realized that the situation was completely different from that in 2010. Understanding why the effects on the carbon balance were so different in 2010 and 2011 was what made Nature like the paper. For this reason, I favor long-term studies. If I had worked with just the 2010 data, I would have thought that the Amazon acted that way every year.
In an editorial, Nature said that the results of this paper are bad news.Do you agree?
I do. It’s very bad news. We did not expect that the Amazon would have such a low level of carbon absorption. No one had ever measured it in the way we have now done. There have been various studies that used local data and extrapolated an average for the entire region. But is taking an average valid? We already know that there is great variation in the Amazon.
What was the general feeling about the carbon balance in the region?
The estimate was that the Amazon absorbed around half a petagram of carbon per year. Everyone thinks that the Amazon is an enormous carbon sink. However, in 2010, because of water stress, the plants performed less photosynthesis and many more died. The forest, on average, absorbed only 0.03 petagram of carbon. Very little. This is equivalent to 30 million metric tons of carbon. The amount is equal to the study margin of error. Due to clearing using fire and forest fires, the Amazon emitted 0.51 petagram of carbon (510 million metric tons). Therefore, in the carbon balance, much more was emitted than absorbed. This is horrible news. In 2011 it was more humid and the carbon balance was practically zero [0.30 petagram of carbon was emitted, but 0.25 petagram was absorbed, eight times more than in the previous year].
Is the amount of rainfall the principal factor for understanding the carbon balance in the Amazon?
Not exactly. Our study shows that water availability is a more important factor than temperature. It’s a matter of weight. But that doesn’t mean that the temperature is not important. The big difference between 2010 and 2011 was the water issue, but this is also linked to temperature variations. It is difficult to give a definitive answer. This indicates that you can’t develop a climate forecasting model taking into account only increases in temperature. You need to include all of the consequences of this increase in temperature. A very simplistic model will not be able to predict what will happen in the future.
Were the 2010 drought and the 2011 rains abnormal for the Amazon?
We can’t say that the rains of 2011 were extreme, because they were not above the historical maximum. It was a rainy year, above average, but not unusual. It is a matter of definition. There have been other years with similar rainfall. The 2010 drought was extreme, unusual, below the historical minimum. Therefore, I cannot say that the 2011 absorption capacity was typical for a rainy year. In 2010, the forest suffered greatly from drought. The following year, the vegetation could have still been reeling from the impact of that extreme stress. The events of one year could have influenced the following one. It may be that, after a rainy year, carbon sequestration is greater if there is then a second rainy year.
Data for a single year should not be analyzed in isolation.
Exactly. This is why long-term studies are so important. When I participated in large projects and saw that there was such variability from year to year, I gave up on doing this type of study. I think it is great to bring together researchers from many different areas [in large projects] because the studies complement each other. Advances in some aspects of knowledge frequently occur in this type of situation, but not in the sense of finding a significant number that represents all of the Amazon. There is great variability in this aspect. You can not study a month in the dry season and another in the rainy season and assume that these periods represent everything that occurs in the dry season and the wet season and extend the result to the whole year. This number could be double or half the real value, for example. During our 10-year study in Santarém, Pará, we saw this wide variability. I’m a perfectionist. If I know that my number might be very wrong, I am not satisfied.
With data from only two years, can you come to a conclusion about the carbon balance in the Amazon?
Since 2010 was so different from 2011, we concluded that, even with four or five years of data—our original plan—we would not be able to calculate a conclusive average. Now we are looking for funds to finance the continuation of this project for a decade. Will the average of 10 years be enough? Yes, studies on the carbon cycle are more conclusive if they last a decade. But it is important to understand that the Amazon is being changed, both by man and by the climate, which is also being altered by man. Therefore, the average result could be different from what took place in the prior decade, or in the decade before that. We are going to submit a proposal to continue this study. However, in addition to resources to carry out the measurements, we need resources to pay a team to do the research. I am the only IPEN employee working on the project; all the others are paid by the projects involved in this study. And, without this fine team, this incredible project would not exist. Many people are putting in a lot of effort.
Some studies suggest that an increase in greenhouse gases could lead some plants to perform more photosynthesis.Couldn’t this change the carbon balance in the Amazon in the long-term?
And not only this. More CO2 in the atmosphere stimulates plants to perform more photosynthesis. There are other mechanisms, too. When less water is available, roots absorb less. The plant reduces its metabolism and thus absorbs less carbon. What we know for sure is that the forest’s ability to absorb carbon falls when water is scarce.
How can air collected in four locations in the Amazon represent the atmosphere of all of this enormous region?
At any of the points, the samples collected in flight represent the mass of air that passed over various regions of the Amazon, from the Brazilian coast to the collection point, and in the case of Santarém, even sections of the Northeast. If it took seven days to reach the collection point, it represents a week, and not just the time at which it was obtained. It stores all the history of the path it took through the Amazon during those seven days, all emissions and absorptions that occurred on that route. We are not, therefore, collecting an air sample for a given hour. We are collecting the history of a column of air that traveled all the way from the Brazilian coast. We calculate the route that each air sample took before being collected at each sampled altitude.
Does this method have some limitation?
The major limitation is that we have only taken samples at altitudes of up to 4.4 kilometers. What happens at higher altitudes is beyond our reach. A convective cloud can move air upwards or downwards. This means that our air column could be partially elevated to an altitude above our flight limit. In this case, we lose information. This is the largest source of error in our method. It would be ideal if we could fly up to 8 or 12 kilometers high. We have already begun to do this, beginning at the start of 2013 for the data collection point near Rio Branco, and the results are very encouraging. In this altitude range, in one year, we did not observe a significant variation that would indicate a large error. This is very exciting.
Do the four air sample collection points behave the same?
The point near Santarém is different from the others in terms of results. We need to think about its area of influence. There is a high population density all along the coast. This is the point in the Amazon region in which the area/population ratio is lowest. Our data collected at that point suffer from urban influence, human activities and fossil fuels that do not appear so strongly in other regions in the Amazon. Even the pollution coming from the cities in the Northeast affects it. Sometimes, during the rainy season, we observe carbon emission at this point, whereas the other three monitored points indicate absorption.
What explains this difference?
It could be due to human activities in areas near Santarém. The Guianas are north of the Equator. During the rainy season in Brazil it is dry there. There are also fires to clear land and human activities along our coast. They say I should stop taking measurements in Santarém, because they do not represent the Amazon. But I have 14 years of data from there. Brazil does not have time series of measurements of this type. If we stop taking measurements in Santarém… It’s a dilemma.
But doesn’t Santarém represent an important part of the eastern Amazon?
In the Santarém area of influence, 40% is forest. If you look at the entire area of the Amazon forest, the Santarém measurements represent a “slice,” in quotes because it is gigantic, of about 20%. We only discovered this when we started to study the other side of the Amazon. The samples obtained in-flight are the result of everything that happened before the air gets there. With the exception of carbon monoxide, which comes from burning the forest to clear it, there is no way to know the contribution of each carbon source. In the case of Santarém, this approach does not work very well. We believe that a portion of the carbon monoxide comes from other biomass burning, perhaps from fossil fuels and not primarily from burning forest vegetation.
How did your work on the Amazon get started?
I began participating in the LBA [Large-Scale Biosphere-Atmosphere Experiment in Amazonia] at the beginning, in 1998. I did field work. Ten years ago I began taking systematic measurements in Santarém. Until then, air profile samples were sent to the United States to be analyzed in a NOAA [National Oceanic and Atmospheric Administration] laboratory. In 2004 I set up my laboratory at IPEN and we stopped sending samples to the United States. My laboratory is a replica of the NOAA lab, the best lab in the world for greenhouse gases. I did everything exactly the same and imported a replica of their laboratory. We put everything in boxes and brought it here. Really, everything. From the computer mice to the shelves. The entire system in a box. We can measure CO2, CH4, N2O, CO, SF6 and H2. NASA paid for the laboratory and we use it for the LBA. Everything I learned in this area was from the NOAA Global Monitoring Division team. I visited there three times. We are always together. They have access to Magic, our analysis system. Everything is done in partnership with them. I have been working with these guys for 10 years. It is like I am their daughter. After 2004, our measurement frequency in Santarém increased. That year, we flew in both the dry and the rainy season for the first time. We also tried to take measurements in Manaus, but of the three years of data collection we had flight authorization problems in one year and CO2 analysis problems the following year. Then the money ran out. As we only had full data for a single year, I never ended up publishing the data. But this is a problem I need to correct. We collected data only in Santarém until 2009, when we obtained funding from FAPESP and NERC [a UK research funding agency] and began taking measurements at three additional points.
When studies were restricted to Santarém, were you able to reach any conclusions?
We observed that there is great variability in the carbon balance during the rainy season in the Amazon. We published this data in a paper in 2010. We saw that there was no point in carrying out a short-term study over one or two years. It must cover many years. This was the first important lesson I learned from that study.
What is the next step in your work in the Amazon?
Calculating the carbon balance in 2012 and 2013. We already have the data. 2012, for example, is half way between 2010 and 2011. It rained like crazy in the northwestern region and the rest was drier than in 2010. This is why I like data collected using an airplane, which allows us to calculate the result [carbon emissions and absorptions]. If I worked only with an emissions tower, and it were on the dry side, I would come to one conclusion. If it was on the rainy side, I would come to a different one. With the kind of data we use, we can take everything into consideration and see which predominated. We can calculate everything and take the average. And in 2012, on average, the entire basin was dry. In terms of forest burning, the amount was between the values seen in 2010 and 2011.
You didn’t begin your career in atmospheric chemistry.How did you start?
My undergraduate and master’s research was in electrochemistry. But I was really frustrated and asked myself what it was good for. Then the first environmental chemistry meeting was held in Brazil, organized by Wilson Jardim [a professor at Unicamp] in Campinas in 1989. I attended and fell in love. That was what I wanted to do with my life.
Where are you from?
I was born in Birigui, but I left the city when I was 3. I spent most of my childhood in Cafelândia, which had 11,000 inhabitants at the time. Everyone knew each other there. That is why I am the way I am. I talk to everyone. I also talk a lot with my hands. My students say that if they were to tie my hands, I would not be able to teach class. The staff at the IPEN gates do not even ask me for my badge. That is how people act in non-urban areas. People who live in the capital can be alone in a crowd. My family moved from Cafelândia to Ribeirão Preto because my father didn’t want his children to leave home to study. He chose a city with many universities and moved the whole family there. He represented Mobil Oil in Brazil. It did not matter if he lived in Cafelândia or Ribeirão Preto. The funny thing is that my sister went to Campinas, I went to the Federal University of São Carlos, my brother went to FEI in São Bernardo, and the only one that stayed home was the third daughter. I had health problems and had to go back to my parents’ house before I graduated. I transferred to the USP Ribeirão Preto campus, but there the chemistry program had almost twice as many credits as the program at the Federal University of São Carlos. It took me another three and a half years to take all those credits, whereas I would have finished in just one and a half years at the Federal University. Everything had prerequisites. But it was really great because in São Carlos I had studied well only during the first year. Then I joined a semi-clandestine militant party, as a student leader. I spent more time on politics than studying. We were prohibited from attending classes. When it came time to take exams, I photocopied the notes of friends, studied all night, without sleeping, took the exam and passed. But you can imagine how much I could remember later. So it was good I had to practically restart my bachelor’s degree from the beginning. What kind of professional would I be if I had not had to do it all over and learned to study hard?
What were the classes like at USP?
I left the student movement, which had disappointed me soundly. I wanted a more just world. But my boyfriend was part of the national leadership of the revolutionary party. I left him, and he took revenge by using his power. I realized that the problem was not in the mode of production, communist or socialist, but in man’s level of evolution. So I decided that the only person I could change was myself. I became zen, spiritual, and began my internal revolution. I realized that I could not change the world, but I could make myself a better person. So then I began my career, practically from the beginning, because USP Ribeirão Preto was very tough. If you don’t study, you don’t pass. I did undergraduate research, I obtained a FAPESP scholarship, I focused on my studies and I fell in love with environmental chemistry.
What was your doctorate like?
It was what I was able to do. When I began my doctorate with [Antonio Aparecido] Mozeto, my intention was to study reduced sulfur compounds, in the area of gases. At that time, nobody was working with gases. There was only one person working in the area in Brazil, Antonio Horácio Miguel, but he had gone abroad. I had to do everything. For example, I had to develop a permeation tube standard. I had to develop the tube, purchase the liquid to permeate, everything. When I got everything working and put the tube in the chromatograph, the device broke. The professor had bought a chromatograph to measure sulfur compounds that a professor at the University of Colorado had decided to manufacture. Everything came out wrong. It had a stainless steel crossarm with a flame that produces hydrogen and water when it burns. The flame heated the crossarm and water leaked onto the photomultiplier and burned the apparatus. It lasted just one day. The problem was that I had been working on the project for two years and needed a new device in order to continue with my doctorate. Wilson Jardim then asked why I thought no one in Brazil was working with gases. “It’s difficult! The only person working in the area had left Brazil. Forget about gases and focus on something else,” he told me. But, at that point, I had already lost two years and I was the only professor at the Federal University of São Carlos without a doctorate. A professor then told me that I was on probation and that, if I didn’t crank out a doctorate in one year to solidify my position, they would not allow me to continue. So I quickly looked for a topic that I could do and that I would not be ashamed to have done. I performed an analysis of the sediment at the bottom of some small lakes near the Mogi-Guaçu river. I used analysis methods from aerosol studies to discover the origin of the sediments and also date them. In this way we can learn about the history of occupation of the river basin. I finished my PhD at the Federal University of São Carlos and began working in the area of atmospheric chemistry, which is what I wanted, and in which there are few Brazilian researchers.