Potential consequences of draining Brazil’s Pantanal

By Sean Killian

The Pantanal, located in central South America, is the world’s largest continental wetland. Covering an area the size of Florida and home to nearly 2,000 bird and fish species, the Pantanal is one of the most productive ecosystems on the planet. While large parts of the Pantanal have remained pristine, the ecosystem is currently under unprecedented pressure from economic development, alterations of its water courses and conversion to other land uses, including a national push for ethanol production. These pressures not only threaten the biodiversity of the Pantanal, which is itself an enormous carbon sink, they also threaten to trigger the atmospheric release of massive levels of carbon dioxide if drained. This paper evaluates wetlands’ role in the carbon cycle, and attempts to estimate the carbon loss to the atmosphere if the Pantanal were drained.


Certain types of wetlands contain large historic, reservoirs of carbon in above ground biomass, litter, peats, soils and sediments [3]. Wetlands remove CO2 from the atmosphere by sequestering organic carbon in water-saturated soils where the reduced oxygen supply inhibits decomposition. Organic soils consequently accumulate carbon, unlike well-drained soils, which release carbon to the atmosphere by microbial oxidation [1]. Moreover, wetland plants convert atmospheric carbon into biomass and wetland soils store carbon-rich sediments. Net carbon sequestration occurs as long as rates of conversion exceed decomposition and external transport of materials.

While it is clear that wetlands significantly influence biogeochemical cycles, it is difficult to evaluate the net carbon sequestering role of wetlands because decomposition of organic matter, methanogenisis and sediment fluxes are extremely complex and there are gaps in scientific knowledge [3]. Estimates of the magnitude of carbon sequestration are uncertain, and depend on whether they are derived from measurements of gas fluxes above forests or of biomass accumulation in vegetation and soils [4]. Despite the lack of detailed data, the United Nations has estimated that wetlands hold more than 15 percent of the terrestrial biosphere carbon pools [14].

Wetlands are sensitive systems, and disruptions in hydrological regimes can significantly impact their size and profile. The principle human activity that could disrupt the functioning of wetlands as long-term carbon sinks is artificial drainage for agricultural development [1]. Drainage causes soil subsistence, or the lowering of the land-surface elevation from changes that take place underground, which releases CO2 as a result of microbial decomposition of organic carbon stored in plant biomass and soils. Biomass is also burned in the process of land clearing. After a rapid release from fire used to clear land or from the decomposition of leaves and fine roots, there is a prolonged period of carbon-dioxide release as coarse roots and branches decay and as wood products decay or burn [7]. Given the large potential of wetlands to store carbon, the aggregate release to the atmosphere of CO2 as a result of mass drainage of all wetlands would be catastrophic.


The drainage of a single wetland, the Pantanal, would also be catastrophic. The Pantanal, or “swampland” in Portuguese, is an immense alluvial plain within the Upper Paraguay River Basin in western Brazil, eastern Bolivia, and northeastern Paraguay. The Upper Paraguay Basin is 496,000 square kilometers, but the Pantanal itself includes only 140,000 square kilometers. The area is dominated by a matrix of seasonally-flooded savanna, streams, rivers, ponds, lakes and marshes. Principal vegetation consists of scrub forest and savanna characterized by native grasslands interspersed with gallery forest, humid semi-deciduous forest and wetland vegetation [5].

Average annual rainfall in the basin is approximately 1,300 mm, 80 percent of which falls from November to March. Beginning in November, up to 70 percent of the 450 kilometer long Pantanal basin is gradually inundated, turning it into a shallow inland sea, interspersed with higher areas that do not flood. Depending on local elevation, flooding lasts from three to nine months. More than 80 percent of the Pantanal floodplains are submerged during the rainy seasons, nurturing one of the world’s most biologically diverse collection of aquatic plants. During the transition period from wet to dry season, the daily average net ecosystem exchange rate ranges from a strong sink of -10 ± 5 kg C per ha per day to a situation close to balance in June. Within several months after peak floods, evaporation, evapotranspiration, absorption and outflow transform the area into a huge savanna, including rivers and tributaries, open grasslands, isolated pockets of cerrado forest vegetation, and many shallow water bodies with large numbers of trapped fish, attracting wading birds and other wildlife [5].

These diverse hydrological zones and phases have made the Pantanal both biologically diverse and extremely complex to study. By virtue of being a large wetland, the Pantanal is undoubtedly rich in soil organic carbon and dissolved aquatic carbon dioxide. Because if its productivity, the Pantanal also boasts large carbon stocks in aboveground biomass. If the entire Pantanal were to be cleared and drained, a significant amount of CO2 and CH4 would be released to the atmosphere, immediately and over time. The question is how much.


Estimates of carbon pool sizes have been based either on soil maps or on summarization of field data for different ecosystem types [1]. Mapping data for the tropical ecosystems is scarce, and data inadequacies using the biome-type approach require the extrapolation of results from a small number of studies to a large geographical area. Given its hydrological diversity and sheer size, the Pantanal is therefore particularly difficult to study. As a result, no comprehensive estimates have been conducted to date of the average carbon stored across the Pantanal.

Bernoux’s estimates of soil organic carbon across Brazil are detailed, yielding 36.4 ± 3.4 Pg C at a depth of 30 cm for all Brazilian ecosystem types [2]. Yet, many wetland deposits of carbon far exceed 1 m in depth, making the study inadequate for the purposes of this paper [1]. In the Pantanal’s largest alluvial fan spanning a third of the region, for example, sediments reach 400 m deep [12]. In addition, concentrations of dissolved CO2 in various water profiles deviate from atmospheric equilibrium under the influence of biotic activity, including photosynthesis, aerobic and anaerobic respiration, and methanogenesis, which complicates measurement of CO2 levels. Within the Pantanal, even the most alkaline waters do not carry detectable concentrations of carbonate solids in suspension, leaving waters undersaturated with respect to calcium carbonate. Given the absence of carbonate solids, dissolved CO2 produced by the biota will accumulate [9]. The aggregate volume of dissolved CO2 will change dramatically between wet and dry seasons, and measurement of dissolved CO2 as a part of the Pantanal’s overall carbon profile becomes difficult.

One approach is to extrapolate from the historical carbon emissions from the Everglades, which in its earlier history resembled the ecological conditions in the Pantanal today [5]. Although the Pantanal watershed is several times larger than that of the Everglades, both are large, globally significant freshwater wetland systems dependent on larger watersheds. Yearly and wet season rainfall totals appear to be approximately equivalent in each system. Both ecosystems depend on water regimes with yearly wet and dry cycles, although the Pantanal experiences greater differences in water levels between dry and wet seasons [5]. For the purposes of a rough estimate, the profile of the Everglades decades ago can serve as a proxy for the Pantanal.

Drainage of freshwater peat was begun in 1912 in the Everglades agricultural area. Subsidence reached 1.8 m by 1950 and about 3.3 m by 1970 [1]. The rate of carbon loss associated with subsidence can be estimated from the bulk density of the Everglades soils (about 0.23), the soil organic matter content (about 88 percent), and the carbon content of the organic matter (about 56 percent). At a subsidence rate of 2.54 cm yr-1, of which 75 percent is attributed to oxidation, Everglades peat would release 0.007 x 109 t C yr-1 [1]. This translates into 85.56 Mg of CO2 ha-1 yr-1 released from Everglade soils. This is the value used to estimate carbon loss in the Pantanal.


Both land clearing for agricultural production and changes in the hydrological regime due to climate change threaten the ecological balance of the Pantanal and will result in carbon loss to the atmosphere. Using the aforementioned historical estimates of soil carbon loss from the Everglades to roughly estimate the potential of those for the Pantanal and assuming a subsidence rate of 2.54 cm yr-1, we get a total of:

85.56 Mg of CO2 ha-1 yr-1 x 140,000 km2 x 100 ha per km2 = 1.197 Pg of CO2 yr-1

This amount is equal to 19.6 percent of CO2 emissions in the United States in 2007, which were 6.103 Pg CO2 [13]. Applying a 20 percent margin of error still results in a low-end value of 0.957 Pg CO2 yr-1 or 15.6 percent of the U.S. total in 2007 — a significant release of carbon dioxide. This is under a scenario that does not consider alternative uses of the land and assumes a consistent rate of carbon loss over all 140,000 km2.

This estimate also does not take into account the carbon loss as a result of clearing aboveground biomass. Unfortunately, little data exists for the Pantanal. One study of the Pantanal, which claims to be the first estimation of carbon dynamics in aboveground biomass , was published in 2008 and covers only aboveground course wood biomass in the northern region [10]. Another, published in 2006, uses allometric correlations to estimate the aboveground biomass and wood volume of a part the Pantanal, focusing on only woodland savanna areas [11]. As a result, estimates of aboveground biomass loss as a result of clearing are difficult to calculate. Nonetheless, studies that approximate the carbon debt incurred from clearing different types of ecosystems can also be used as proxies. Fargione et al estimate that aboveground biomass for rainforest peatland and cerrado forest are roughly 10 Mg of CO2 ha-1 yr-1 and 1 Mg of CO2 ha-1 yr-1, respectively [7]. Cerrado grassland aboveground biomass is minimal. Using these values, we get an initial carbon release from clearing across the of 0.154 Pg, which yields a grand total of 1.351 Pg. Again, even with a substantial margin of error, this volume of CO2 would be catastrophic.


Drainage of the Pantanal for agricultural purposes and for human development is gradual, and there is little evidence that the rate of subsidence in the Pantanal is the same as that of the Everglades during the 20th century. Yet, the threat is real. Population growth, demand for food crops, and a push for biofuel production could conspire to eventually encroach on the Pantanal at subsidence rates exceeding 2.54 cm yr-1. If this were to happen, the combined effort to reduce carbon emissions from anthropogenic sources would pale in comparison to the carbon released from the Pantanal. The Pantanal is, however, upwards of 99 percent privately owned. It will therefore require economic incentives and international policy regimes to prevent drainage of the Pantanal .


1. Armentano, Thomas (1980): Drainage of Organic Soils as a Factor in the World Carbon Cycle. Bioscience, 30(12):825-830

2. Bernoux, Martial et al (2002): Brazil’s Soil Carbon Stocks. Soil Science of America, 66:888-896

3. Kunsler, John (1999): Climate Change in Wetland Areas Part II: Carbon Cycle Implications. Acclimations, August.

4. Richey, Jeffrey et al (2002): Outgassing from Amazonian Rivers and Wetlands as a Large Tropical
Source of Atmospheric CO2. Nature, 416:617-620

5. Wade, Jeffry et al (1993): Comparative Analysis of the Florida Everglades and the South American Pantanal.

6. Knipling, E. B. et al (1970): CO2 Evolution from Florida Organic Soils. Soil Crop Science Society of Florida Proceedings, 30:320-326

7. Fargione, Joseph et al (2008): Land Clearing and Biofuel Carbon Debt. Science, 319:1235-238

8. Fahmuddin, Agus et al (2007): Carbon Climate Human Interaction: Carbon Pools, Fire, Mitigation, and Restoration. Proc. of International Symposium and Workshop on Tropical Peatland, Indonesia, Yogyakarta.

9. Hamilton, S.K. et al (1995): Oxygen Depletion and Carbon Dioxide and Methane Production in the Waters of the Pantanal Wetland of Brazil. Biogeochemistry, 30:115-141

10. Schoengart, J. et al (2008): Carbon Dynamics in Aboveground Course Wood Biomass of Wetland Forests in the Northern Pantanal. Biogeosciences Discussion, 5:2103-2130

11. Salis, Suzana et al (2006): Estimating the Aboveground Biomass and Wood Volume of Savanna Woodlands in Brazil’s Pantanal Wetlands Based on Allometric Correlations. Forest Ecology and Management, 228: 61–68

12. Pott, Arnildo and Pott, Vali (2004): Features and Conservation of the Brazilian Pantanal Wetland. Wetlands Ecology and Management, 12:547-552

13. U.S. Environmental Protection Agency (2008): Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990–2007, ES-4

14. United Nations University (2004): Inter-Linkages Approach for Wetland Management: The Case of the Pantanal Wetland. UNI-IAS, November


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: