Methane emissions to the troposphere from the Amazon floodplain

Abstract

Methane emissions to the troposphere from the Amazon River floodplain have been determined. In total, 94 individual flux measurements were made along a 1,700-km stretch of the river during the early falling water period of the flood cycle (July-August 1985). The overall average rate of CH 4 emission from wetlands was 390 mg CH4 m -2 d-x. Water surfaces covered by aquatic macrophytes had the highest emissions, 590 mg CH½ m -2 d -•, while emissions from flooded forests were 110 mg CH,• m -2 d -• and open lake areas averaged 120 mg CH½ m-2 d-•. Ebullition was the dominant mechanism of emission, accounting for 85% of the total. Surface water dissolved oxygen was always undersaturated, while dissolved CO2 concentrations were significantly supersaturated with respect to atmospheric equilibrium. Surface water CH½ concentrations were highly supersaturated, averaging 6.4 pM. The g•3C of the CH½ varied from -42%0 to -73%0 with an apparent trend toward heavier values downstream. Extrapolation to the entire Amazon floodplain results in estimated annual emissions in the troposphere of 10 Tg yr-• and indicates the importance of the area in the global atmospheric CH4cycle. Recent evidence indicates that the atmospheric concentration of methane has been increasing at a rate of between 1-1.5% per year [Rasmussen and Khalil, 1986; Khalil and Rasmussen, 1986; Pearman et al., 1986; Stauffer et al., 1985]. As CH,• is an important "greenhouse" gas and plays a role in the atmospheric cycles of O 2, C1, and OH radical [Crutzen, 1973; Wofsy, 1978; Logan et al., 1981], this observed increase could be of significant environmental consequence. Current global CH,• budgets place inputs to the atmosphere at between 370 and 1000 Tg yr-•, with the majority of the input being biogenic in origin [Ehhalt and Schmidt, 1978; Khalil and Rasmussen, 1983; Sheppard et al., 1982; Seiler, 1984]. Although estimates of the magnitude of the various components of the biogenic emissions vary between budgets, CH,• emissions from wetlands are generally thought to account for 20 to 40% of the total. These estimates, however, are based on only a few field measurements, and virtually no data are available from the extensive wetland areas of the tropics, e.g., the Sudd in Sudan and the Amazon River floodplain and Pantanal regions of Brazil. Here we present the results of a study of mechanisms and magnitudes of CH,• emissions to the troposphere from the Amazon River floodplain. During its annual flood cycle the Amazon River rises 10 m, inundating its approximately 100,000 km 2 floodplain or "varzea." Depending on the flood stage, during a normal year between 40 and 80% of the floodplain is covered by shallow lakes, swamps, and dense beds of floating aquatic macrophytes [Junk, 1985]. As the varzea is a region of high primary production and uniformly warm temperatures, these wetlands constitute a significant potential for CH,• production [Richey et al., ment and as part of the Carbon in the Amazon River Experiment (CAMREX) project, during the early falling water period ...

Seasonal variability in methane and nitrous oxide fluxes from tropical peatlands in the western Amazon basin

“…Third and last, our data probably underestimate net CH 4 fluxes for the PMFB because we chose to include fluxes with strong negative values (i.e., more than −10 mg CH 4 -C m −2 day −1 ) in our calculation of mean diffusive flux rates. These observations are more negative than other values typically reported elsewhere in the tropical wetland literature (Bartlett et al, 1988(Bartlett et al, , 1990Devol et al, 1988Devol et al, , 1990Couwenberg et al, 2010). However, they represent only a small proportion of our dataset (i.e., 7 %, or only 68 out of 980 measurements), and inspection of our field notes and the data itself did not produce convincing reasons to exclude these observations (e.g., we found no evidence of irregularities during field sampling, and any chambers that showed statistically insignificant changes in concentration over time were removed during our quality control procedures).…”

“…Diffusive CH 4 flux, averaged across all vegetation types, was 36.1 ± 3.1 mg CH 4 -C m −2 day −1 , spanning a range from −100 to 1510 mg CH 4 -C m −2 day −1 . This mean falls within the range of other diffusive fluxes observed in Indonesian peatlands (3.7-87.8 mg CH 4 -C m −2 day −1 ) (Couwenberg et al, 2010) and other Amazonian wetlands (7.1-390.0 mg CH 4 -C m −2 day −1 ) (Bartlett et al, 1988(Bartlett et al, , 1990Devol et al, 1988Devol et al, , 1990. Although the ebullition data must be treated with caution because of the sampling methodology (see below), we observed a mean net ebullition flux of 973.3 ± 161.4 mg CH 4 -C m −2 day −1 , spanning a range of 27 to 8082 mg CH 4 -C m −2 day −1 .…”

Diffusive emission of methane and carbon dioxide from two hydropower reservoirs in Brazil

“…The method was described by Devol (1988Devol ( , 1990 and Bartlett et al (1988Bartlett et al ( , 1990. All the samples were taken in vegetation-free areas both in the middle of reservoir and near the edges.…”

Atributos polarimétricos de imagem radar na inferência de parâmetros morfológicos de macrófitas

“…Em contrapartida, são habitats favoráveis à reprodução de vetores de doenças de veiculação hídrica (malária, dengue, febre amarela, esquistossomose) com sérias conseqüências para a saúde pública (PEDRALLI, 2003;VASCONCELOS et al, 2006). Respondem pelas maiores taxas de emissão de metano quando comparadas às áreas de floresta inundada ou águas abertas (DEVOL et al, 1988;BARTLET et al, 1990).…”