Methane sources and sinks in a periodically flooded South African savanna

Citation
Lb. Otter et Mc. Scholes, Methane sources and sinks in a periodically flooded South African savanna, GLOBAL BIOG, 14(1), 2000, pp. 97-111
Citations number
84
Categorie Soggetti
Earth Sciences
Journal title
GLOBAL BIOGEOCHEMICAL CYCLES
ISSN journal
08866236 → ACNP
Volume
14
Issue
1
Year of publication
2000
Pages
97 - 111
Database
ISI
SICI code
0886-6236(200003)14:1<97:MSASIA>2.0.ZU;2-I
Abstract
Methane (CH4) fluxes were measured over a 2-year period at subtropical sava nna and floodplain sites located in South Africa. No significant difference s were detected in fluxes from the nutrient-rich and nutrient-poor savannas . Savannas were not always sinks and showed some CH4 emissions during the w et, summer season with fluxes in the range of -1.6 to 1.68 mg m(-2) d(-1). Soil temperature did not have a significant effect on the flux, while CH4 c onsumption decreased and emissions increased with an increase in soil moist ure between 5% and 50% water-filled-pore-space (WFPS). Understanding the fa ctors which control the production and emission of CH4 in aerobic environme nts is extremely important for predicting net emissions from a region. Dry floodplain soil fluxes were near zero and did not differ significantly from savanna fluxes. During a flood, saturated sites emitted CH4 at an average rate of 69.4 mg m(-2) d(-1). Flooded areas with a water depth of 0.3 - 0.4 m emitted CH4 from the water surface at rates between 0.48 and 466.3 mg m(- 2) d(-1) with the highest emission occurring during summer. Emission rates were exponentially related to sediment temperature, which had a greater inf luence on the emission rate than the hood regime. The length of the dry per iod preceding the flood and the extent of a flood did not have a significan t effect on CH4 fluxes from saturated and flooded sites. Emission rates wer e highest when the water level was between 0.1 m below the soil surface and 0.4 m above the soil surface, with emission rates declining to near zero a s the water became deeper than 0.4 m. Savannas were estimated to consume an average of 0.04 g CH4 m(-2) yr(-1), with southern African savannas consumi ng 0.23 Tg CH4 yr(-1). Saturated and flooded sites were estimated to produc e 25.3 and 57.2 g CH4 m(-2) yr(-1), respectively. Southern African floodpla ins are estimated to produce between 0.2 and 10 Tg CH4 yr(-1) (excluding th e effects of vegetation-mediated emissions) and therefore produce more CH4 than the savannas consume.