Spatial variability of organic matter properties determines methane fluxes in a tropical forested peatland

Abstract

Tropical peatland ecosystems are a significant component of the global carbon cycle and feature a range of distinct vegetation types, but the extent of links between contrasting plant species, peat biogeochemistry and greenhouse gas fluxes remains unclear. Here we assessed how vegetation affects small scale variation of tropical peatland carbon dynamics by quantifying in situ greenhouse gas emissions over 1 month using the closed chamber technique, and peat organic matter properties using Rock-Eval 6 pyrolysis within the rooting zones of canopy palms and broadleaved evergreen trees. Mean methane fluxes ranged from 0.56 to 1.2 mg m^-2 h^-1 and were significantly greater closer to plant stems. In addition, pH, ranging from 3.95 to 4.16, was significantly greater closer to stems. A three pool model of organic matter thermal stability (labile, intermediate and passive pools) indicated a large labile pool in surface peat (35-42%), with equivalent carbon stocks of 2236-3065 g m^-2. Methane fluxes were driven by overall substrate availability rather than any specific carbon pool. No peat properties correlated with carbon dioxide fluxes, suggesting a significant role for root respiration, aerobic decomposition and/or methane oxidation. These results demonstrate how vegetation type and inputs, and peat organic matter properties are important determinants of small scale spatial variation of methane fluxes in tropical peatlands that are affected by climate and land use change.

Keywords: Carbon dioxide; Geochemistry; Methane; Organic matter; Rock-Eval pyrolysis; Tropical peat.

Fig. 1

Peat sampling strategy at five points around each individual plant stem, replicated across six C. panamensis and six R. taedigera plants

Fig. 2

a CO₂, and b CH₄ fluxes at 0.5 m and 1.5 m from C. panamensis and R. taedigera plant stems. Means ± 1 SEM (n = 6)

Fig. 3

Proportions of peat labile (C_l), intermediate (C_i) and passive (C_p) carbon pools for C. panamensis and R. taedigera in peats 0.5 m and 1.5 m from plant stems. Means ± 1 SEM (n = 6)

Fig. 4

Carbon stocks associated with labile (C_l), intermediate (C_i) and passive (C_p) pools for C. panamensis and R. taedigera in peats 0.5 m and 1.5 m from plant stems. Means ± 1 SEM (n = 6)

Fig. 5

I and R indices for C. panamensis (white) and R. taedigera (grey) peats at 0.5 m (filled square) and 1.5 m (filled circle) from plant stems

Fig. 6

a Scores for C. panamensis (white) and R. taedigera (grey) peats at 0.5 m (filled square) and 1.5 m (filled circle) from plant stems. b Loadings from PCA for all species and peats. Combined PC1 and PC2 account for 51% of variance

Fig. 7

Linear regression of S2 and logged mean CH₄ fluxes for C. panamensis (white) and R. taedigera (grey) peats at 0.5 m (filled square) and 1.5 m (filled circle) from plant stems