Read the full article in the Fall/Winter 2011 edition of Environment@Harvard.
In the Amazon, Gauging the Resilience of a Rainforest
By Alvin Powell
The tower lay on the forest floor, a twisted mess of metal, wires, leaves and branches.
Moments earlier, it had stood more than 200 feet high, a thin staff of crisscrossed metal loaded with scientific instruments, poking its nose above even the 160-foot tall trees of Brazil’s Tapajos National Forest, a vast tract of rainforest in the eastern Amazon more than 2,000 square miles in extent.
For nearly five years, from 2001 until its collapse in 2006, the tower’s instruments had sampled Amazonian air, measuring the forest’s inhalations and exhalations as it absorbed carbon dioxide during the day, when sunshine drives photosynthesis, and released the gas at night, when photosynthesis shuts down and respiration becomes dominant.
For Rotch professor of atmospheric and environmental science Steven Wofsy, the tower had been a scientist’s dream come true. Since the mid-1980s, when he first began tower-based data collection in the regenerating forests of Massachusetts, he had wanted to run similar experiments in the Amazon. An Amazonian tower, he had realized, could collect data in the world’s largest remaining tropical forests, and test hypotheses about the Amazon’s regional functioning and role in global climate. Such hypotheses filled scientific journals, magazines, and later, the weighty reports of the thousand-plus scientist Intergovernmental Panel on Climate Change.
Even in its ruin, though, the tower illuminated a key truth about the forest. The Amazon jungle is not a static biological feature, but rather a dynamic, shifting place, responding to inputs from the atmosphere and the soil, reacting to human chainsaws that convert vast tracts to field and pasture and growing new trees and vines in the suddenly sunny spots created by the regular fall of giant trees.
“This forest turns over really fast,” says Wofsy. “You have these large trees that…are growing fast, but they couldn’t have been doing so for very long because they’d be even bigger than they are. Then you realize they turn over quickly and they start falling on you and knocking your tower down.”
Though the Amazon is always in motion, concerns about a warming globe have raised scientific interest in just how it is changing. The forest covers 2.2 million square miles and contains a large portion of the world’s biodiversity, harboring perhaps one fifth of the Earth’s species of plants and animals.
This ecosystem exerts an enormous influence on the land, sea, and atmosphere. The Amazon river alone accounts for 18 percent of all the freshwater that flows into the world’s oceans. The Amazon basin, which receives more than twice the rainfall of the Northeastern United States, is a regional rainmaker: an estimated 30 percent of the rain that falls on the forest comes not from the ocean, but from evaporation from the trees themselves. In addition, a major atmospheric upwelling zone is anchored over the Amazon, pumping energy and water vapor into the atmosphere and linking the forest to global atmospheric circulation.
“The idea that changes to the Amazon will have global ramifications, I think that’s a very likely outcome,” Wofsy says. “What we couldn’t say is what that outcome will be, specifically. Something important will happen, but we don’t really know what.”
Today, scientists concerned about the changing global climate are interested in forests like the Amazon for their role in sequestering or emitting carbon. Carbon is a key ingredient in the greenhouse gases carbon dioxide and methane. The buildup of those gases in the atmosphere from human fossil fuel burning is thought to be the major culprit in global climate change.
The Amazon’s enormous size makes it a natural storehouse for carbon. Scientists know that carbon is a major element in the cellulose and lignin that make up the tissues in the forest’s riotous plant growth. Carbon dioxide is absorbed by the leaves and converted to carbohydrates during photosynthesis. Some of it is locked up in the plant’s woody tissues while a lesser amount is released again during the energy-producing process of respiration.
The carbon held in trunks, branches, and roots is stored away from the atmosphere until the tree dies and decomposes. But the Amazon rainforest is being cut down at an alarming rate, largely for conversion to pasture and agricultural fields. Already, 17 percent has been clear cut or burned; some estimates project as much as 30 percent will be gone by mid-century.
In addition to human-caused deforestation, there is potential for loss of the forest due to climate change: predictions suggest that the Amazon could see warmer, drier years, with more frequent and longer droughts. Scientists are concerned this will cause the forest to die off and be replaced by grasslands, fire-adapted woodlands, and savannah.
If that happens, it will be bad news for the planet. The Amazon is thought to store 70 to 80 billion metric tons of carbon in plant biomass. If the forest is burned to create fields and pastures, or if it dries out, dies off and decomposes, a significant portion of that stored carbon will be released into the atmosphere, speeding climate change.
The forest’s changes have therefore put scientists urgently to work on two simple-sounding but complex questions: “What is the forest doing now?” and “What will happen to it in the future?”
Wofsy and professor of organismic and evolutionary biology Paul Moorcroft are both hard at work on the problem. Wofsy, a member of the Harvard University Center for the Environment (HUCE) Steering Committee, is using instruments mounted on towers and flown on small planes to determine just what’s going on with the forest today. Moorcroft, a HUCE-affiliated faculty member, is using Wofsy’s measurements to help him build a new generation of computer models to predict what the Amazon will look like in the future.
A forest unbalanced
Though scientists understand photosynthesis and respiration, there is much they don’t know about how the forest handles carbon. The biggest question about the forest’s functioning today—at least in relation to climate change—involves the tally called the carbon balance: Is the forest, in the final analysis, absorbing carbon and storing it away, or giving it off?
This is a key question that figures into scientists’ calculations concerning climate change. Some research in recent years has indicated that the forest is absorbing carbon dioxide, perhaps because rising carbon dioxide levels in the atmosphere are acting to fertilize plant growth.
But Wofsy doesn’t think so. His years of data collection and analysis have led him to the conclusion that the forest is roughly in balance, and is even slightly emitting carbon dioxide, likely due to agricultural burning.
The forest tower that held Wofsy’s instruments was erected in 2001 near the Brazilian city of Santarem as part of a major international effort, led by Brazilian scientists, called the Large Scale Biosphere-Atmosphere Experiment in Amazonia (LBA).
The Harvard portion of the LBA included a second tower near the coast, managed by senior research fellow in atmospheric chemistry J. William Munger. This 30-foot tower was built on top of a roof in Maxaranguape, a coastal town in northeastern Brazil, in order to take readings of carbon dioxide and carbon monoxide in the trade winds blowing off the Atlantic. This data provided a baseline to better understand the results from the forest. The air is relatively clear of carbon monoxide when it reaches the first tower, says Munger. But as it flows toward the second set of instruments in Tapajos National Forest, it picks up small amounts of carbon monoxide. The carbon monoxide is likely from agricultural burning and human activities in the small towns in between; a small amount is probably naturally given off by the oxidation of hydrocarbons released by leaves.
On average, Munger reports, CO2 levels don’t change a lot on the overland journey from the ocean to the inland forest. The data show that “over that [land] interval, there’s neither a large source nor a large sink of carbon dioxide,” he says. “When we combine that with information we get from measurements, they all say the same thing: that over the whole region, the average level of carbon dioxide is about in balance.”
Wofsy’s and Munger’s results counter earlier studies which found that the forest was taking in more carbon dioxide than it was giving out—possibly a lot more. If those studies were accurate, this would mean that the Amazon rainforest is a large natural sink that helps humanity at least partly mitigate its climate change problem by taking carbon dioxide out of the air.
But Munger believes those earlier results were erroneous because researchers were not taking into account changing air circulation patterns and calmer conditions at night, which could cause an undercounting of carbon dioxide released by the trees during the process of respiration.
“If you don’t account for what goes on at night, you have a good accounting of the uptake [of carbon dioxide by the forest] but you miss the respiration,” Munger says. “Some of those reports of very large uptakes…are biased.”
Having two towers has helped the scientists understand the forest’s daily and annual uptake and release of carbon dioxide. Results published in 2005 showed that the seasonal variation in this cycle was actually the reverse of what people had assumed.
Scientists had thought the trees would be more active in the wet season, when there is plenty of water, than in the dry season. The dry season’s scarce water resources were thought to cause leaves to close their gas-exchanging pores, called stomata, to keep water vapor in, but with the collateral consequence of closing out carbon dioxide and slowing photosynthesis.
But Wofsy’s results showed that the trees actually take up more carbon dioxide in the dry season. The trees, Wofsy says, seem better able to manage water stress than originally believed and keep their stomata open and their photosynthetic machinery in full operation despite drier conditions.
“Most carbon models predicted that the dry season would be a time when carbon would be released…and then the reverse in the wet season,” Wofsy says. “The actual measurements showed the opposite. During the dry season more photosynthesis went on for one simple reason: that’s when the sun was out. The trees managed water stress much better than people thought they would.” (The measurements that made this discovery possible continue today, thanks to a 2007 grant from the National Science Foundation’s Partnerships for International Research and Education Program to Scott Saleska, a former postdoctoral fellow of Wofsy’s now at the University of Arizona. Saleska’s grant enabled reconstruction of the tower in 2008.)
While the towers provided in-depth profiles of carbon dioxide flows at two locations over time, the researchers wanted a snapshot of the whole Amazon basin.
In November 2008 and May 2009, Wofsy and a team of Brazilian and European collaborators, together with graduate students, converged on an airport in Manaus, Brazil. Together the team flew missions totaling 150 hours, measuring carbon dioxide in the atmosphere at one-second intervals from a small plane packed with instruments, while taking between 300 and 400 flask samples to be further analyzed in the laboratory.
The project, originally proposed as part of the LBA in 2001, had been another long-term dream of Wofsy’s. He first flew aircraft missions in the Amazon in 1985 and 1987, measuring the flux of carbon to and from the biosphere. Like the more recent tower measurements, those missions showed that the biosphere and the atmosphere in the region were roughly in balance, so when other researchers asserted that the forest was absorbing carbon, or releasing it, they got his attention.
“People have been publishing papers saying the Amazon is taking up a lot of CO2, or that the deforestation is a source of CO2. I knew, from the 1980s, that it was roughly in balance, so we could test some of those ideas,” Wofsy says. “The question is, is it any different now, 25 years later? As it turns out, probably not much.”
Though the Harvard contingent was banned by the Brazilian military from actually flying in the plane, they spent plenty of time in it on the ground, checking instruments and retrieving data as the sun beat down, causing temperatures to soar. Graduate student Victoria Chow, a Wofsy advisee, says the temperature in the plane was akin to the hottest days of a New England summer, but with higher humidity.
Each day, the team would get up at four or five a.m. to check the daily weather forecast and begin to prepare for the day’s flight. By six a.m. they had to make a fly or no fly decision, Chow says, so they could get to the airport and begin readying the plane and its instruments. The weather above the Amazon is tricky, she says, because storms pop up unexpectedly. There are also fewer airports where the plane can land in case of a problem.
While the plane flew without them, Chow, Wofsy and other members of the team monitored the weather, examined the previous day’s data to make sure the instruments were running properly, and prepared gear for future flights.
The plane monitored the entire basin in both the wet and dry seasons. Wofsy says the measurements showed that the lower atmosphere, at two kilometers, is much more affected by the forest’s daily fluctuations than higher up, at four kilometers.
“What we see is a really strong daily cycle, with a lot of CO2 coming out of the forest in the early morning after it builds up overnight. Then it gets sucked back down again during the day,” Wofsy says.
Together, the measurements show that, contrary to assertions that the Amazon is a moderate to large carbon sink, it is actually emitting a small amount of carbon dioxide. Wofsy says the measurements, which are still being analyzed, show the Amazon being a net source of carbon in the November 2008 flights, and a somewhat larger source in the May 2009 flights.
“The pluses have it. During that time, the Amazon was, on average, a small net source of CO2,” Wofsy says. “It wasn’t large, but it was noticeable.”
Wofsy says the findings agree with recently published results from the National Oceanic and Atmospheric Administration that did monthly measurements from space over a long period of time. Wofsy attributed the release of carbon dioxide to agricultural burning to clear land.
“The green stuff is in balance,” Wofsy says. “There is a source here. It’s a net source and the net source is coming during the burning season. A lot of the burning we’re seeing is agricultural burning. It still seems to be telling us that overall, the deforestation is adding carbon. We should [soon] be able to tell you how much.”
Wofsy says the findings, once they’re fully analyzed, might have implications for climate treaties, particularly for countries like Brazil, which has claimed it has reined in deforestation.
Wofsy’s results will likely find their way into computer models of the Amazon. Scientists seeking to understand the region and where it fits into the global climate puzzle often turn to computers and their complex simulations to put the pieces together and see where different management actions might lead in the future. Of course, the simulations’ forecasts will be most accurate if the models themselves accurately reflect the extraordinarily complex workings of the forest itself.
Until recently, computer models of the Amazon—and other forests as well—used what’s called a “big leaf” model, which essentially scales up the functioning of a single leaf to the size of a large forest tract and uses that as a representation of how the forest works.
While that may not be a problem if one wants a simplistic representation of forests as part of the planetary biosphere, the rise of climate change as a critical global problem has raised the stakes for accuracy in global computer models. It has put them front and center not just in scientific circles, but also in media explanations of what’s going on with the planet, in the public’s understanding of climate change and, perhaps most critically, in the debates engaged in by politicians around the globe that lead to law and regulatory change.
Big leaf models that treat large forest tracts—typically thousands of square miles—as if they were a single entity are likely missing complex characteristics of a forest community that may be particularly important in a changing environment. Instead of having just one type of tree that may or may not be drought tolerant, for example, real communities have many species of trees whose response to changing conditions may be different from each other. That means a real forest may behave in entirely different ways from a simple cyber-forest.
Modeling trees as individuals
Moorcroft, whose office is just up the hallway from Wofsy’s in Harvard’s Geological Museum, has been a pioneer in adding a realistic complexity to computer models of forest functioning. Borrowing the concepts that physicists use for modeling fluids and gases, Moorcroft treats individual plants or animals as a physicist would a particle, and sets rules for how each particle responds to different environmental conditions. Rather than an entire ecosystem responding as if it were a single big leaf, in Moorcroft’s models it is the responses of each individual in the system to environmental conditions that determines the model’s overall behavior.
Moorcroft got his start modeling coyote movement patterns in Yellowstone National Park in advance of wolf reintroduction there, conducting fieldwork with radio-collared coyotes to check the model’s accuracy. He has more recently been improving models of how forests work, in many cases using Wofsy’s data and findings to sharpen his models, to check them against the current reality, and, hopefully, spin them forward to a more accurate understanding of the future.
Today, the terrestrial ecosystem model Moorcroft designed, called an ecosystem demography model, is in its second major iteration and continuing to undergo development. Moorcroft’s model treats trees as individuals of three different types that respond differently to levels of light and nutrients—key variables that are affected when a forest is disrupted by large treefall, for example. The model allows for gaps and disturbances in the forest, subjecting individuals to conditions dictated by the time since the last disturbance. Moorcroft’s model trees have the characteristics of early-, mid-, and late-successional species. Early succession trees are light-loving and fast-growing and the first to invade gaps in the canopy such as that left by Wofsy’s falling tower. Mid-succession trees have some shade tolerance, denser wood, and slower growth rates. Late-succession trees are the slow-growing giants, which eventually form the forest canopy and shade out competitors below.
“What we did was come up with a model, defined at the scale of individual plants competing dynamically for resources, like light, water, nutrients,” Moorcroft says. “The ecosystem you see is the outcome of that individual level competitive process. The ecosystem is this ensemble, this collection of individuals, competing for resources.”
Changes to the Amazon—and to the amount of carbon it stores—appear guaranteed. Though efforts are underway to preserve it, with continued cutting of roads through pristine forests, and the subsequent clearing of nearby land for cattle and soybeans, the shrinking of the forest seems inevitable.
Moorcroft’s research is geared toward understanding the regional ramifications of the changing Amazon, and how the dwindling forest will affect not only the flow of carbon through the system, but water as well. With so many trees drinking deeply through their roots and releasing water vapor during respiration, the forest is a key player in regional weather.
Early models of Amazon change looked at the effect of replacing the Amazon entirely with grassland, something unlikely to happen either soon or all at once. Those early models, though, showed that because the forest puts so much moisture in the air, should the forest disappear, rainfall in the region would drop by 30 percent and regional temperatures would climb two to three degrees Celsius.
“When you’re in the middle of the Amazon, you’re a long, long way away from the ocean,” Moorcroft says. “The modeling says because of that, the hydrologic cycle in the Amazon is really closed. Of the water that falls as rain there, a significant fraction was put there by the plants themselves.”
Moorcroft says he has always been intrigued by those early models and by the idea that the biosphere can affect the physical world. It has long been known that the physical world affects the living world—falling temperatures kill plants or send them into dormancy, spring rains cause a new blossoming of life, storms and earthquakes affect life in many ways, including our own—but the scientific understanding that the reverse can be true hasn’t been around very long.
Today, Moorcroft wants to apply that idea—that the forest affects the regional climate—to models of the forest’s future. With climate change models already projecting a warmer, drier Amazon and with forest conversion to agricultural land continuing, an important question is how the remaining forest will respond, and how that will further affect the atmosphere.
“What’s happening is people don’t just go clearing the entire Amazon. It’s being transformed slowly,” Moorcroft says. “And, as that transformation is happening, how does that change the climate? How does that feedback process play out? So we need much more realistic vegetation models, dynamic and responsive to climate, so you’re not prescribing it, you’re allowing it to evolve as climate evolves. As climate shifts, so the vegetation shifts.”
The answer is not just important to Moorcroft. Some of his research is funded by the Moore Foundation, a nonprofit that has traditionally been interested in conserving the Amazon’s immense biodiversity. With predictions that the forest might die back, they want Moorcroft’s help in deciding which tracts will still be forested 50 to 100 years from now and that might be most worth conserving.
Marcos Longo, a doctoral student being advised by Wofsy and Moorcroft, is working to understand how the atmosphere might respond as the forest is converted to pasture, as roads are cut through and the forest becomes more fragmented. Longo is using a coupled model simulation that brings together atmosphere and ecosystem models, to try to project those future effects. Longo and Moorcroft say fragmentation might initially increase rainfall along the forest edges by inducing small-scale atmospheric circulations as air over pasture land warms rapidly and interacts with moisture-laden air over the slower-warming forest.
“The idea is that this fragmentation process actually affects the atmosphere,” Moorcroft says. “Whenever you have contrast between two places, deep roots putting moisture into the atmosphere next to shallow roots, it can actually induce circulation that can affect rainfall.”
Longo says he hasn’t run the model long enough to see how the modeled atmosphere above a fragmented forest responds, but he continues to work on the problem.
With a drier Amazon a likely future scenario, Tom Powell, a doctoral student working with Moorcroft, is seeking to understand how drought affects the trees in order to incorporate that into the ecosystem demography model. If he’s successful, that will double the types of individual responses by the model’s trees, dividing each existing category into the additional categories of drought tolerance and intolerance.
“Hopefully, this will enable us to better represent how drought is going to affect the forest,” Powell says.
Powell is following up on a NASA-funded drought experiment at two Amazonian sites. One ran for five years and the second is still running after eight. The project imposed a 50 percent artificial drought on experimental plots by suspending solid panels connected to gutters just above the forest floor. Initial results reported that large trees were more susceptible to drought damage, dying with more frequency than smaller trees.
A more recent examination of the data, however, showed a more complex story, with differing responses from different genera of trees. If that is the case, Moorcroft says, it’s possible that the forest may have the capacity to resist a drought- and deforestation-induced conversion to grasslands. The forest might persist—at least the part we don’t cut down—by shifting the mix of trees from drought intolerant to drought tolerant.
Though Powell’s ultimate goal is to improve computerized Amazonian forest models, he is heading back to the rainforest to visit the drought experiment plots and gauge how the trees manage water. The challenge is a difficult one, Powell says, because the trees’ response to drought is nonlinear. Instead of shutting down photosynthesis early to avoid losing too much water vapor through their leaf pores, trees seem to keep photosynthesizing as long as possible and then go through a rapid shutdown. He hopes to measure the turgor loss point of the trees—when they close their leaf pores—and use that as a proxy for a tree’s ability to handle drought stress. If it doesn’t work, he says he’ll have to find some other way to measure a trees’ drought tolerance, so it can be realistically incorporated into the model. If the best measure turns out to be a trait of the roots, he says he may end up with a shovel, digging deep into the Amazon’s soil.
“Plants live on the edge, trying to capture as much CO2 as possible” through opened leaf pores, balancing that against “trying not to lose so much water that they die.” Powell says. “Now the trick is to try to figure out…what traits make them drought tolerant versus intolerant. We’ll either [discover] this or go back to the drawing board.”
The Purest Air on the Planet
By David L. Chandler
Scot Martin has found a kind of atmospheric time machine. He’s been able to peer back in time to study the exact makeup of the finest particles in the air, called aerosols—the ones that form the basis of atmospheric cycles that are crucial to understanding everything from local rainfall to global climate change—more or less as they were before industrialization took hold. With an arsenal of science’s latest and most powerful tools, he’s been able to probe some of the planet’s cleanest air, to analyze its chemistry and composition as it was before humans altered it.
There are very few places you can go on Earth, over land, to find such clean air. But Martin, the McKay professor of environmental chemistry at Harvard, along with a team of about 40 scientists and students from Harvard and from Germany’s Max Planck Institute, found such a place in the Brazilian rainforest north of the relatively isolated but rapidly growing Amazonian inland port city of Manaus.
There, in a research preserve in a sparsely-settled region of the country where the prevailing winds blow in from the Atlantic across more than a thousand miles of mostly unbroken rainforest, Martin and his team put up a 130-foot tower to collect air samples. They planted it right in the middle of the forest, with as little disruption as possible, three hundred feet from the nearest dirt road. The metal lattice structure supported a set of vertical tubes, open at the top, running down to a small shipping container on the ground where an array of instruments were installed—all powered by a generator more than a mile away nd downwind, to avoid contaminating the samples. Once up and running, the system could theoretically operate autonomously 24 hours a day, but in practice almost always had a few members of the team on-site tinkering with the machinery and adjusting the instruments, Martin says. Their base camp was in a valley about 1,000 feet from the site.
After months of preparation, the long-sought result came during a few days in March of 2008, when the winds were slow and the rainfall had been strong. That’s when Martin and his colleagues sampled some of the most untainted air ever examined by science.
Traveling to one of the Earth’s remotest regions to study minute samples of microscopic particles, far smaller than the width of a human hair, might seem like an extraordinary effort for a tiny return. But this meticulous research could end up having profound impacts. In a series of recent papers, including one published in the journal Science in September 2010, Martin and his large team of collaborators describe how the pristine aerosol particles found in this air are dramatically different from any that have been examined in the lab before, and will provide an essential baseline of data that may, over time, help to solve one of the thorniest conundrums in the Earth’s changing climate system.
In this case, thinking microscopically may have effects globally. “We were interested in the connection between plants and rainfall, as regulated by atmospheric chemistry, and in particular, as regulated by particles,” Martin explained recently in his office at Harvard’s Pierce Hall. This connection is critical but surprisingly complex, and atmospheric scientists around the world have been trying to sort out for decades exactly how it works, and how it has been altered by the infusion of particulates caused by human activities.
When a mass of air becomes saturated with water vapor, it can’t just form water droplets and rainfall by itself; instead, it needs a little help in the form of nucleation centers—tiny particles that water molecules can adhere to and accumulate around until they form droplets large enough to be pulled by gravity so that they fall as rain. Nucleation centers can take many forms: particles of salt, especially over or near the oceans; particles of pollutants spewed into the air from the burning of fuel or biomass, as is the case almost everywhere in the northern hemisphere; or, as in Amazonia, particles of organic matter naturally given off by plants as part of their transpiration process, which then react chemically with oxygen in the air, energized by sunlight.
When there are more nucleation centers, they can produce a large number of droplets, which are small and therefore produce clouds that reflect more sunlight back into space and yield little rainfall; a lesser number of particles tends to result in fewer but larger droplets, making for darker clouds with less reflectivity, and ultimately more rainfall.
When human activities such as burning and land-clearing increase the concentration of nucleation centers in the air, this can mask the effects of global warming by producing more bright clouds, reflecting away more solar heat. But because carbon dioxide persists in the atmosphere for centuries, while the particles only last from weeks to a few years depending on their size, this masking is a temporary effect. The exact role of these aerosol particulates in causing clouds to reflect or absorb sunlight remains by far the biggest source of uncertainty in today’s global climate models.
That’s why trying to get a handle on exactly how much and what kinds of atmospheric particles are being produced by human activities, and how that compares with the baseline—the atmosphere as it would have been in the absence of human interference—remain among the most urgent open questions in climate science.
These are questions Martin has been studying for most of his career. Beginning in the fall of 2007, during a sabbatical year, he decided to try to find some answers in the Amazon basin. That’s when he joined a team of Max Planck Institute scientists who had been carrying out environmental monitoring in northern Brazil for more than 20 years. The arsenal of equipment the group brought with them in early 2008 was more extensive than any that had been assembled before for in-situ monitoring in this remote region, which is surrounded by many hundreds of miles of almost unbroken forest. They brought scanning electron microscopes and atomic-force microscopes, x-ray spectroscopes, ion and aerosol mass spectrometers, and several other cutting-edge tools to study samples of the air almost continuously over a period of weeks. And that’s how, for a few days in March, when the still air and recent rainfall had left the influence of outside air almost nonexistent (except for a small amount of dust from the Sahara), they got to study their sample of pristine air “as it was 500 years ago,” Martin says.
In the pre-industrial world natural nucleation systems, of which the particulates given off by plants were an important part, had evolved into a stable equilibrium, with the plants producing enough particulates to produce enough nucleation centers to make rainfall to sustain the plants. “In the Amazon, most of the rainwater is recycled – it’s not coming in from the Atlantic,” Martin explains. The cycle is so efficient that the water for the most part just keeps evaporating and raining down again in the same region.
But how do such natural cycles change, exactly, as emissions increase? This experiment is about to be carried out in the real world. In the pristine forest near Manaus, for example, things are changing fast. The air is going through the same changes that air around the world has gone through as the process of industrialization took off and pollutants of various kinds began to build up in the atmosphere. Manaus, deep in the rainforest and still blessed with surroundings that maintain some of the most unpolluted air remaining in Earth’s tropics, is experiencing explosive fossil-fuel-powered growth, encouraged by government policies that have made it a tax-free zone to promote the growth of business and industry. And its air is deteriorating accordingly.
To illustrate the magnitude of the changes that are taking place, right now the air above the Amazon contains about 300 particles per cubic centimeter. Above Boston, the number is about 30,000: average for the industrialized world, and a concentration over a hundred times greater than in the pre-industrial world.
But near Manaus, one aspect will be different as things change. The whole process will be meticulously monitored and measured, so that researchers will at least know not only exactly what the baseline was, but how rapidly and in what ways change takes place.
If patterns seen elsewhere are duplicated here, the concentration of particles will increase dramatically, and rainfall could change accordingly, possibly altering circulation patterns over a wide region.
Just this October, Martin and a team of colleagues were awarded a grant for a year of detailed follow up research in the Amazon basin, using a large Department of Energy traveling laboratory, a shipping container full of equipment called the Atmospheric Chemistry Research Facility, to take place in 2014. They’ll be making small scouting trips this winter and spring to pick the research sites, and over time the expedition could end up involving a team of a few hundred researchers. “It’s a wonderful scientific opportunity,” Martin says. By studying side-by-side data from areas inside and outside the plume of pollution that blows from Manaus, “we’ll be able to do an apples-to-apples comparison,” he says.
If Brazilian officials can see exactly what is happening, chronicled in detail as it unfolds, they may have a chance to avert the worst of the consequences.
“One would hope, by being able to demonstrate how unique the Amazon is functionally”—with plants generating their own rainfall—“development there will be able to proceed accordingly,” Martin says. That is, they could make choices about the rate and the type of development that can mitigate or prevent the worst consequences.
And other nations facing the same growth patterns might then have a chance to learn the same lesson.
Projections are that over the next half-century, perhaps 50 new megacities with populations over 10 million will sprout up globally, and most of them will be in the tropics. This represents a very dramatic shift from past growth patterns, with most of the world’s largest cities in temperate and subtropical zones.
The planned Amazonian research, Martin says, “allows us to make an investigation of how the atmospheric chemistry of the world’s tropical regions will be affected by the development of these megacities.”
In a sense, then, his atmospheric time machine will be allowing a glimpse into the world’s future—perhaps in time to help avert some of its perils.
Managing Malaria, Beating the Mosquito in the Amazon Jungle
By Steve Bradt
In the developed world, malaria can seem like a relic of a bygone era. But in Brazil, home to 70 percent of the Amazon rainforest, it’s a scourge that in recent decades has been resurgent.
The Brazilian Amazon has been steadily settled—and deforested—since the 1970s, when the government created incentives to support agriculture, mining, and human settlement in this vast but once sparsely populated region. The massive environmental impact of this human influx is widely known: one-sixth of the Amazon Forest has been lost, much of it over the last 30 years.
But a less-recognized side effect of the clearing of the Amazon is that it has opened up massive new breeding grounds for mosquito species that are the primary carriers of malaria in South America. Because malaria in the Amazon takes a very different form than it does elsewhere in the world, there is a distinct name for the disease in Brazil and neighboring countries: “frontier malaria.”
The effect on public health has been catastrophic in the large swath of Brazilian territory officially classified as Amazon (the Amazônia Legal, or “Legal Amazon”), which has seen a tenfold increase in malaria since 1970. The number of cases peaked at more than 635,000 in 1999, before sliding to about half that since the implementation of comprehensive government programs to control the disease.
For more than 10 years, Marcia C. Castro, an assistant professor of demography at the Harvard School of Public Health, has been studying the thicket of factors contributing to frontier malaria that complicate the response to the disease in this area of the world.
“Malaria transmission in Brazil is extremely complex, characterized by the poorly understood interplay of social, environmental, economic, political, and behavioral factors,” she says. “Among the most troubling aspects of our limited understanding of frontier malaria is that we know very little about how climate change will impact transmission patterns. Infrastructure projects intended to promote further development of the region may worsen the burden of malaria if we fail to anticipate and mitigate the impacts.”
There’s long been an inherent tension between the environmental and public health implications of Amazon frontier expansion. Environmentalists would prefer to see as little of the Amazon clear-cut as possible. So when settlers move in to establish a homestead and farm, environmental concerns would dictate clearing as small a lot as possible.
But this, Castro says, flies in the face of what scientists know about malaria transmission: Namely, that proximity of poorly constructed houses to tropical forest is a surefire recipe for the spread of the disease.
“Malaria control is optimized when there’s a buffer of at least 100 meters between houses and the forest fringe, where mosquito density is highest,” Castro says. “When you have great proximity between man and the edge of the forest, the exposure to mosquito bites is maximized.”
Furthermore, the ramshackle homes in many newly settled parts of the Amazon—only about three-fifths of which offer their occupants sanitation and clean water—don’t offer much protection from the primary carriers of malaria, the Anopheles genus of mosquito. And compared to the malarial vectors prevalent in Africa and Asia, the Amazon species’ behavior seems optimized to infect people.
The mosquito carriers “are most active at 5 to 6 a.m. and at 5 to 6 p.m. They are also exophilic—they prefer to bite humans outdoors,” Castro notes. “Unfortunately, dawn and dusk are exactly the times when settlers are most likely to be outdoors, on their way to and from their work and school.”
Most of the migrants who have sought economic opportunity in the Amazon during the past four decades, tripling its population to 23.6 million, came from malaria-free areas in Brazil, and therefore have no immunity against the disease. They also know little to nothing about malaria protection or transmission.
But Castro’s research has helped develop new ways of modeling the determinants of malaria transmission in the Amazon. Specifically, her work supports a temporally- and spatially-targeted approach to combating malaria in newly settled areas.
In the initial years after an area is settled, she says, the Brazilian government needs to focus on environmental management to minimize malaria transmission. This does not necessarily include large scale vector and larval control, an almost impossible task given the sheer size of the region. But it does include strategies to improve the quality of houses and to reduce human-mosquito contact.
First, improved construction techniques would keep malarial mosquitoes at bay. Second, homes should be built with buffers of at least 100 meters from the woods and 500 meters from bodies of water, she says. The mosquitoes that carry the malarial pathogen in the Amazon exist in particularly dense numbers at the forest fringe, where the species find the partially shaded habitat they prefer for breeding. Such habitat is also found alongside lakes, rivers, and streams.
Unfortunately, the environmental transformation undertaken by many new arrivals to the Amazon often ends up creating a perfect mosquito-breeding habitat. Settlers generally leave taller trees standing when they slash-and-burn a parcel of land, maintaining the partial shade Anopheles finds desirable. Burning increases soil pH, and the mosquitoes prefer breeding in alkaline standing water.
Malaria outbreaks during the first two years after an area is settled tend to be especially severe. Castro points to the experience of a settlement called Machadinho, where migrants first arrived in late 1984. By 1986, more than 90 percent of the population had malaria at least once, and 56 percent of residents were sickened in at least five months of the year.
Longer-term, after an area has been settled for about 10 years, urbanization and a degree of community cohesion tend to supplant erratic migration and highly variable land clearing practices. Paving and improved drainage create environments that are less hospitable to mosquito larvae, and health clinics become more prevalent.
Castro’s research suggests that at this later stage of settlement the government could manage malaria by focusing on human behavior. Key contributors to the virulence of frontier malaria, she says, include the many asymptomatic carriers of the disease—essentially, outwardly healthy people who are unlikely to seek medical treatment for symptoms but who can nonetheless infect mosquitoes when bitten.
“We really think the two critical reasons transmission remains stubbornly high in the Amazon are the large number of asymptomatic carriers and the high rate of human mobility,” she says. “Put these two pieces together and you have a recipe for sustained transmission of malaria.”
The distinction between frontier malaria and malaria elsewhere in the world was first recognized in the 1980s, as the opening of the Amazonian frontier accelerated during the Brazilian government’s launch of aggressive efforts to encourage settlement there. Frontier malaria’s distinctive characteristics include low levels of immunity, lack of community and social isolation, and exacerbation of problems by poor government planning.
A way for the Brazilian government to respond, according to Castro, would be frequent testing of everyone in frontier areas of the Amazon, treating those found to be asymptomatic. But the feasibility of this approach is questionable.
There are practical and logistical hurdles: How do you find and treat everyone in an area where the population is highly transient?
“The cost here is in searching for and finding settlers, and in monitoring their long-term health,” she says. “Once you’ve done that, the cost of actual treatment is very manageable.”
But in a vast jungle spread across nine nations, borders are permeable and a single government can’t solitarily coordinate testing and treatment.
All across the Amazon basin, rivers rise dramatically during the rainy season, flooding areas all along their banks. When the rains end and the rivers subside, pools of water—ideal for mosquito breeding—are left along the Amazon and its many thousands of tributaries.
“The sheer size of the region, and its unique climate and hydrology, are important challenges to controlling malaria,”
Castro says. “Control of mosquito larvae is not feasible, given the extremely difficult task of identifying, or even reaching, most breeding habitats.”
During the first half of the 1900s in coastal areas of Brazil, engineering projects—such as the construction of drainage systems and the filling of marshes—were successfully used to combat malaria, which is thought to have come to South America with Europeans in the sixteenth century. But the drainage work never extended inland to the vast Amazon region.
These early engineering projects were later augmented by nationwide anti-malaria campaigns starting in 1940, when some four to five million Brazilians—then one tenth of the population—contracted malaria annually. Beginning in 1947, the insecticide DDT was used to spray the interior of houses across the country every six months, reducing the number of cases by 99 percent in just six years. (Interestingly, the tremendous success of this approach may explain why the Anopheles mosquitoes that transmit Amazon malaria have now evolved to become an almost exclusively outdoor species.)
By 1970, cases of malaria nationwide reached their low point: just 52,371 individuals, about 60 percent of them in the Amazon. But throughout the 1960s, trends had been building that would soon send the rate of infection skyrocketing. In 1960, the Brazilian capital was moved inland from Rio de Janeiro to Brasília, marking the launch of a broad effort to integrate the Amazon region with the rest of the nation. Construction of highways into the rainforest intensified during the late 1960s and early 1970s, bringing with it an influx of impoverished migrants seeking land and employment.
With this effort to link the coast with Brazil’s interior, to promote national development, and to increase the nation’s industrial power, the modern Amazon frontier expansion began. But it didn’t take long for the government’s strategy to upset the ecosystem balance of the once-thinly settled region. By the early 1980s, the Brazilian Amazon was already experiencing high rates of deforestation, conflicts with indigenous populations, and severe malaria outbreaks. The lack of acquired immunity among most settlers, combined with precarious living conditions—houses with partial walls, or with walls and roofs made of tree leaves—led to a resurgence of the disease.
Unlike in Africa, where mortality rates are high, the disease causes relatively few deaths in the Amazon. Primarily due to improved clinical treatments, fewer than 100 people a year die of malaria in Brazil. The hundreds of thousands of cases annually in the Amazon result mostly in missed work and school, due to flulike symptoms that commonly include fever, chills, headache, sweats, fatigue, nausea, and vomiting. Despite the low mortality rate, the disease imposes a heavy social and economic burden, since most residents of the Amazon are farmers who are unable to hire others to work on the land.
“Government permits to settle should be based, in part, on the risk of malaria in those areas,” Castro says. “In many parts of the Amazon, people should not have been settled in the first place.”
One of the biggest question marks in the ongoing control of malaria in the Amazon is what role climate change will play in future transmission patterns. While in temperate North America heavy rains usually promote mosquito reproduction by creating standing pools of water for breeding, the opposite can actually be the case in the humid tropics.
During periods of drought, Castro says, river and lake levels fall in the Amazon, creating areas of stagnant water along their margins that are optimal mosquito breeding grounds. For example, 2005 saw a drought, followed by a jump in cases of malaria. But, she cautions, the relationship between precipitation and malaria outbreaks is probably more complicated than this, an issue she is currently exploring.
“The variation in water levels, combined with irregular river and stream margins, helps to create pools of water,” Castro says. “But if a drought gets really severe and the area dries out, there are no breeding habitats. It all depends on how fast the process is, and if the water stays there long enough to allow larvae to mature into adult mosquitoes.”
At least half a million cases of malaria were recorded annually in the Amazon between the late 1980s and early 2000s. Since a 2005 spike—the last time cases of malaria numbered more than 600,000—the incidence of the disease has been on a steep downward trajectory: only about 300,000 cases were documented in 2008 and 2009.
Whether Brazil and its neighbors can sustain this progress remains to be seen. A malaria control program implemented by the Brazilian government in 2000 and bolstered in 2003—targeting the communities that are home to the largest numbers of cases—has cut outbreaks dramatically. Castro says the government has improved its monitoring of pathogen resistance to existing malaria drugs, as well as its tracking of water levels so as to step up mosquito control when needed. But it’s possible, Castro says, that the government might instead curtail the program in an economic slowdown, as has happened during previous malaria control efforts.
“Despite accounting for half the cases of malaria in the Western Hemisphere, Brazil has been cited in recent years as a country where malaria eradication should, in theory, be feasible,” Castro says. “I’m not sure I agree with that optimistic assessment, given the likely high prevalence of asymptomatic individuals. Aggressive and active surveillance—and treatment—of these carriers might be essential to containing malaria in the Brazilian Amazon.”
Renaissance Training in the Water Worlds of Brazil
By David L. Chandler
Brazil, says John Briscoe, is “basically four different worlds, when it comes to water.” About the size of the United States, the country has very different regions whose needs and resources cannot easily be lumped together when thinking about appropriate policies on agriculture, forestry, and energy production—all of which are connected to water.
Briscoe, who grew up in South Africa, has spent significant time living in Brazil, mostly in its capital city of Brasília. During that time he represented the World Bank but he has now assumed a new role: director of the Harvard University Water Security Initiative.
The program, which began with a focus on Australia, Pakistan, Mexico, the U.S., South Africa, and Brazil, aims to build on the model set by the renowned Harvard Water Project of the 1960s, which pioneered the integration of engineering, government, and economics in the interests of water policy. The new, revived initiative will work to perform integrated, interdisciplinary research across the areas of technology, economic growth, law, business, environmental responsibility, and public health.
Few people are better qualified to be running such a sweeping collaborative program. Trained originally at the University of Cape Town as a civil engineer, Briscoe has spent four decades working on water issues in dozens of countries, and has lived in South Africa, Mozambique, Bangladesh, India, and Brazil. Formerly a senior water advisor to the World Bank, as well as the bank’s Country Director for Brazil, he speaks five languages. Once a student in Harvard’s earlier water program (he earned his doctorate in environmental engineering in 1976), he has now returned as Gordon McKay professor of the practice of environmental engineering in the School of Engineering and Applied Sciences—with a joint appointment in the School of Public Health and the Harvard Kennedy School.
The new initiative is operating in a very changed world, Briscoe points out. The time when it may have made sense to send experts from developed nations to try to teach lessons to the locals is long gone, he says. Now, what is essential is to develop programs that are highly collaborative from the get-go, working with local scientists, engineers, businessmen, and officials, setting the priorities in accordance with the local needs. “Everyone understands this intellectually, but I don’t think they understand it viscerally.”
Brazil is an interesting and promising arena for such research, Briscoe explains, in part because of its enormous natural and economic resources and solid, stable government. “They’ve made enormous strides,” he says. “They have a quite stable political system—there’s zero chance of a Chavez or a Morales there.” The country’s economy has likewise been relatively stable, and its largely urban population—some 80 percent of Brazilians live in cities—has achieved great gains economically over the last decade, with poverty declining by about 70 percent. The fertility rate has gone down during the past 25 years from about 5.5 to about 2 children per woman, he says: “a precipitous drop.”
And even for those still living in poverty, Briscoe says, the country has made great strides in innovative systems for providing water and sanitation, even in vast urban slum areas.
Briscoe is clearly deeply attached to his second home of the last two decades or so, citing its incredibly friendly people. “Anyone who works on Brazil never stops working on Brazil,” he says.
The water issues there are quite different from those in many other parts of the world. In much of the country, vast rivers supply an abundance of water that can provide power, transportation, and a host of other resources for both people and industry; there the issues have more to do with protecting the quality of the water that’s there, and protecting the urban populations from the vagaries and variations of the raging waters. In São Paolo, for example, a city of 20 million, a hard rain can produce flooding that slows cars and trucks to a near standstill, producing traffic jams nearly 100 miles long.
Brazil is “an agricultural superpower,” Briscoe says, and about 90 percent of the country’s agricultural output is dependent on rainfall.
In the country’s sparsely populated northeast, however, where the climate is relatively dry, developing agriculture requires extensive new irrigation systems. Already, several states in the region are beginning to battle over rights to water from the São Francisco, the major river there, Briscoe says. From the rhetoric being bandied about regarding diversion of some of that river’s water for agriculture, “you’d think they were going to suck it dry,” he says, when in fact,
currently proposed irrigation systems would divert only about a half-percent of the river’s annual flow.
In the south and southeastern regions, by contrast, water is abundant. The vast majority of the country’s supply of electricity, well over 90 percent of it, comes from hydropower, with most of the installed capacity located in the southeast. The country now has more than 600 hydroelectric dams, including one that generates the most electricity in the world (more than five times that of the Hoover Dam): the Itaipu complex on the Paraguay border.
But despite these impressive numbers, the country’s hydro resources are still vastly underused by global standards. Only about 35 percent of Brazil’s hydroelectric potential has been tapped, compared to 80 percent on average in the industrialized world. “Almost all of that underused portion is in the Amazon,” Briscoe says. “So development of hydropower in the Amazon is a large issue—but there are concerns about environmental impact.” The southern rim of the Amazon borders the center-west, the new growth area for Brazilian agriculture.
The scale of the potential for power generation is vast. One single tributary of the Amazon, the 2,000-mile-long Madeira River that few outside Brazil have even heard of, is twice the size of the Mississippi in terms of annual flow.
Outgoing president Luiz Inácio Lula da Silva, described by President Obama as “the world’s most popular politician,” used much of his considerable political capital to push for new dams in the Amazon basin, and his newly-elected successor Dilma Rousseff is likely to continue those efforts, Briscoe says. These projects, in Briscoe’s view, make a great deal of sense for Brazil and for the environment. They are run-of-the-river projects with small footprints, they mean that Brazil will reduce its growing use of fossil fuels, and they open the way for transporting agricultural products by water rather than by destructive and expensive roads.
Good as these projects are, Brazil faces a major challenge in planning them more strategically. For decades, Briscoe explains, hydroelectric development was entirely under the control of one huge public-private company, Electrobras, which was the planner, financer, contractor, generator, and distributor. Following global good practice, the company’s operations have been unbundled with Electrobras now focusing on distribution, while private companies have become the main generators of new capacity.
“What dropped off the table was planning,” he says. “Projects are now being done essentially without planning, without taking non-energy needs into account and all over the place instead of concentrating in a few places.” In the absence of any overall strategy for development and environmental protection, individual dams have begun popping up everywhere.
This rapid growth has led to an urgent need for analysis of existing resources and ecosystems, something that the Brazilian government recognizes and has started to implement. Policies and potential directions for development also need to be re-examined. That’s one place where Harvard’s Brazilian partners see a potential role for Harvard’s new initiative, working with local researchers, regulators, and institutions to study plans and evaluate alternatives that integrate development, energy, transportation, water supply, and the environment.
“Obviously the ideal would be that you co-define what you want to do” together with Harvard’s local partners, Briscoe says. “How can we pool the intellectual capacity and co-produce research?” The approach he envisions is “much less hub-and-spoke, and much more mutuality.” Briscoe’s years of working in Brazil and building relationships with officials and academics will help in setting up these working partnerships.
For example, in looking at ideas for development of areas of the Madeira and Amazon basins, Harvard can help with the development of an “integrated approach to how you do this,” he suggests, selecting specific areas for intensive development and leaving others untouched to maintain biodiversity.
The importance of hydropower to Brazil’s growth and economic strength provides an avenue for demonstrating to people the importance of protecting their resources, Briscoe says. For example, while global climate change may eventually exacerbate some water resource problems, that’s an issue too far removed from the day-to-day concerns of most people to play a role in planning. But in Brazil there is a far more tangible and immediate impact, since the Amazon forest is the “water pump” that fuels the benign hydrology on which so much of Brazil’s electricity and agriculture depends. If deforestation and climate change are shown to alter the rainfall cycle and affect the abundance of water in the entire region, that’s a more immediate concern. “The role of the forest in regulating the hydrological cycle becomes a big reason for Brazilians to pay attention to maintaining the forest,” he says. “There’s a clear local benefit, much more visible to many people than the more abstract global benefits that they don’t really see.”
The tradeoffs involved in such decisions can be complex and can span a wide variety of economic, technical, cultural, and political factors. Briscoe hopes this will provide a fertile area for research in which students will have a chance to dig deeply and become broadly educated.
“At Harvard, our strength is not that we have a big engineering school. We don’t. We’re renaissance engineers,” Briscoe says. “Across this great university we have people who specialize in economics, laws, business, biology, government, public health, and so on—we can bring together a range of disciplines, to converge around environmental issues and economic development.” A kind of renaissance man himself, Briscoe says he still thinks of himself as an engineer, but “in my career, I’ve worked as an epidemiologist, an anthropologist, and as a banker, always with a core of working on water.” He hopes the new water program will develop a generation of people devoted to water, who will delve deeply within their own disciplines, but will also learn about the contribution of other disciplines and be able to grasp the complexities and relationships among issues. They’ll look at “sociological aspects, regulatory institutions, technology issues—the fusion of all of these,” he says. By bringing together experts from a variety of fields around specific water-resource issues, Briscoe hopes the program will educate a new generation of ‘specialized integrators.’ “For a biologist, to be the best biologist they can be; for an economist, to be the best economist they can be; but with a broad understanding so they don’t just stay in that domain. We want them to understand that the particular domain in which they work exists in a broader context.”
“Each country that deals with water deals with it in a very particular way,” Briscoe says. But learning about these complexities in a context like Brazil, with its diversity of regions, resources and issues, and its strong partner universities, should provide a broad swath of experience that will stand Harvard’s new specialized integrators in good stead wherever their work takes them.
Balancing Brazil's Economic & Environmental Needs
By Steve Bradt
“The Amazon is not just a collection of trees,” says Roberto Mangabeira Unger, Pound professor of law at Harvard Law School. “It is also a collection of close to 25 million residents.”
This view of the Amazon as habitat not only for flora and fauna, but also for humanity, informed Unger’s work from 2007 to 2009 as Brazil’s minister of strategic affairs. Popularly known as the “minister of ideas,” he grappled with the challenges of balancing environmental concerns and the economic needs of those who call the Amazon home.
“There is no precedent for the sustainable development of a major rainforest region,” says Unger, who has remained involved in the politics of his native land since joining the Harvard Law School in 1971. “We cannot copy; we must initiate a model of development that is environmentally sustainable and socially inclusive.”
“Without economic sustainability,” he adds, “the Amazon will be driven inexorably toward destruction.”
Starting in the 1960s, the government encouraged relocation to the Amazon, as it saw the settlement of Brazil’s vast interior as a benefit to national security. Those settlers were required to clear land to stake a claim and obtain credit.
Then, as North American and European nations recoiled at the Amazon’s deforestation, Brazil did a U-turn, putting in place some of the world’s strictest environmental regulations. This environmental straitjacket left less than a tenth of the Amazon open to productive use.
“We need environmental regularization,” Unger says, “not the choice between a free-for-all versus genuflection to the rich North Atlantic nations. We need a real framework, not one developed for show.”
An urgent issue underlying the environmental pressures on the Amazon, Unger says, is confusion over who owns it: less than four percent of the Amazon has a clear legal title. Unger has called for legal changes to regularize land titles.
Also key to addressing the environmental needs of the Amazon, he says, is the realization that it is not a vast monolith. The 70 percent of Brazil that’s legally considered Amazon is divided between inland rainforest and savannah. This savannah, or cerrado, is home to some of the world’s most productive agricultural land.
But it’s an agricultural dynamo under greater pressure than any other part of Brazil. Key to the sustainable development of the cerrado, Unger says, is the recovery of large areas degraded by unregulated grazing. For every acre that is farmed, five are turned over to grazing.
According to Unger, the technology needed to return this despoiled pasture land to productive farmland is affordable. It costs just 1,500 Brazilian reais—less than $900—to return a hectare (about 2.5 acres) of abandoned pasture to high-intensity, high-value agriculture.
The forested part of the Amazon is subject to a similar dynamic, in that many environmentally destructive practices—soy farming, low-intensity ranching, lumbering—are economically efficient. Settlers have little financial incentive to abandon these pursuits, and the only way to prevent them is policing that’s difficult to impossible in such a massive territory.
Unger argues that new technologies, better suited to the heterogeneity of the rainforest, must replace those developed for use in temperate forests. Economic development in the Amazon, he says, must also mitigate the temptation to encroach on the natural environment. He points to the city of Manaus, where cell phones and motorcycles are manufactured, so the economy doesn’t rely on the extraction of natural resources. Finally, Unger says, alternative legal paradigms are needed for managing forests, “so the only arrow in our quiver is not the concession of large forests to corporations.”
When he was tapped by President Luiz Inácio Lula da Silva in 2007 as Brazil’s first minister of strategic affairs, Unger’s objective was to formulate a future direction for the government and the nation. He quickly realized that long-term planning needs to begin in the short term.
“We need a succession of moves, not a blueprint,” he says. “This should be music, not architecture.”
Unger was succeeded as minister in 2009 by a Harvard law student, Daniel Barcelos Vargas.
“I will be engaged as I can, directly or indirectly,” says the man who, from Cambridge, has been a behind-the-scenes player in Brazilian politics for four decades. “I have great hope that this will go on. We must protect and preserve the Amazon for the benefit of humanity.”