Tropical wet season runoff mobilises younger carbon in rainforest streams but older carbon in agricultural streams

Le résumé fourni par la source

Abstract. Knowledge of the age of organic carbon (C) that is leached from soils to streams is key to understanding how C is mobilised within ecosystems. The tropics are characterised by significant C fluxes through streams, yet the time scales of organic C sequestration and export remain uncertain in these regions. Here we examined the concentration, composition and age of dissolved organic C (DOC) in 18 small mountainous catchments of the Australian humid tropics, including six rainforest and 12 agricultural catchments, sampled during both the dry and wet seasons. We found that DOC ages varied widely across sites but were generally centuries to millennia old (median ± standard deviation 1,553 ± 848 years BP), with no consistent differences between rainforest and agricultural catchments. However, the two land use categories diverged in their responses to high flow conditions, with DOC age in rainforest streams tending to decrease from 1,878 ± 604 years BP in the dry season to 708 ± 791 years BP in the wet season, whereas agricultural streams mobilised similarly aged or older DOC in the wet season (1,728 ± 641 years BP) than in the dry season (1,303 ± 1,036 years BP). A subset of dissolved inorganic C (DIC) samples collected from three of the catchments (both rainforest and agricultural) indicated that DIC was mostly modern (123 ± 136 years BP) and always younger than DOC. These differences in DIC and DOC ages suggest a partial decoupling between DOC and DIC export pathways, with DOC derived from older soil C pools, while DIC reflected recent C inputs from vegetation uptake and decomposition. Our results highlight the importance of seasonal shifts in the age of C export and the need to conduct sampling that encompasses seasonality in human-impact studies to better constrain C pools and sinks.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.