Soil microbial communities and GHG emissions under different land-use types in Malaysian peatlands

Tropical peatlands are complex and globally important ecosystems which are increasingly threatened by anthropogenic disturbances such as logging and expansion of oil palm plantations, particularly in South East Asia. Microbes in peatlands play an important role in governing overall functions e.g. carbon and nutrient cycling, yet their characteristics in these habitats are poorly understood. This research aimed to elucidate the response of tropical peatlands to disturbance by quantifying the peat properties, soil microbial community structure, nutrient content and greenhouse gas (GHG) emissions, and the functional interactions between these components under different land uses. Specifically, differences between primary peat swamp forest, secondary peat swamp forest, different land-use types and cropping systems associated with oil palm plantations in Peninsular Malaysia were investigated. Sampling was conducted between November 2014 - October 2018, covering both wet and dry seasons. Bacteria at all depths overwhelmingly dominated all land-use types but microbial phenotypic community structure responded to land-use, depth and changes in season. CO2 emissions were lowest in the primary forest and highest in the secondary forest. CO2 emissions in agricultural plantations were lower than the secondary forest, due to very low autotrophic contribution to net emissions in those habitats. Second generation cropping systems exhibited distinct seasonal variations which were more pronounced in 2nd generation oil palm monocropping, where the CO2 emission was reduced to half in the dry season. CH4 emissions exhibited an opposite trend to that of CO2, with highest CH4 emissions in the primary forest and oxidation in the secondary forest, while the agricultural systems emitted low levels of CH4. However, CH4 emissions were very low in comparison to CO2 emissions. Both CO2 and CH4 emissions were significantly impacted by changes in moisture level, with 2nd generation oil palm exhibiting moisture limitation for CO2 emissions, unlike all the other peat land-use studied. Temperature and pH also significantly influenced microbial community structure and GHG emissions. Gram-positive and Gram-negative relative abundance were positively correlated with CO2 and CH4 emissions, respectively. Almost all the nutrients were substantially lower in the secondary forest to that of the primary forest. Cu and Mo concentration were negatively correlated with CO2 and CH4, respectively. Past drainage in secondary forest reduced nutrient contents, including some with antimicrobial properties, and favoured Grampositive bacteria over Gram-negative bacteria. In agricultural plantations the autotrophic contribution of the vegetation to CO2 emissions was significant only for mature oil palm and yam plants, implying that most of the CO2 emissions from other cropping systems is heterotrophic respiration, resulting in C loss. This is further supported by the 2nd generation mono-cropping having higher bulk density and the lowest organic content, while the 2nd generation intercropping maintained most of the organic content with relatively less change to other peat properties. Overall, the results demonstrate that peat properties, microbial communities and GHG emissions are affected by the land-use changes associated with oil-palm plantations. However, the impact was highest at prolonged oil palm mono-cropping, while the practice of intercropping ameliorated the damage caused by such plantations.