Swamp forests on deep tropical peatlands have undergone extensive deforestation and draining for agriculture and plantations, consequently becoming globally significant carbon (C) sources. To study the effects of land-use change on peat as a biological environment, which directly affects decomposition dynamics and greenhouse gas emissions, we sampled peat from four common land-use types representing different management intensities in Central Kalimantan, Indonesia. The near-pristine swamp forest was used to describe unmanaged conditions, and the three other sites in order of increasing management intensity were reforested; degraded; and agricultural. We examined peat substrate quality (total C & nitrogen (N), dissolved organic C (DOC) and N (DON)), organic matter quality characterized by infrared spectroscopy, and microbial biomass and extracellular enzyme activity, to describe both biotic and abiotic conditions in peat. We found that the peat at altered sites was poorer in quality, i.e. decomposability, as demonstrated by the higher intensity of aromatic and aliphatic compounds, and lower intensity of polysaccharides, and concentration of total N, DOC, and DON compared to the peat in the swamp forest. The observed differences in peat properties can be linked to changes in litter input and decomposition conditions altered after deforestation and draining, as well as increased leaching and fires. The quality of the peat substrate was directly related to its biotic properties, with altered sites generally having lower microbial biomass and enzyme activity. However, irrespective of management intensity or substrate quality, enzyme activity was limited primarily to the first 0–3 cm of the peat profile. Some differences between wet and dry seasons were observed in enzyme activity especially in swamp forest, where the most measured enzyme activities were higher in dry season. Reforestation 6 years before our measurements had not yet restored enzyme activity in the peat to the level of the swamp forest, although the topmost peat characteristics in the reforested site already resembled those in the swamp forest. This is likely contributed by the limited capacity of the young tree stand to produce litter to support peat formation and restore the quality and structure of the peat, and the chemical weed control performed at the site. Therefore, we conclude that intensive land management, including deforestation and draining, leads to the surface peat becoming poorer biological environment, and it may take long time to restore the peat properties.

Vast areas of deforested tropical peatlands do not receive noteworthy shading by vegetation, which increases the amount of solar radiation reaching the peat surface. Peat temperature dynamics and heterotrophic carbon dioxide (CO2), nitrous oxide (N2O) and methane (CH4) fluxes were monitored under four shading conditions, i.e. unshaded, 28%, 51% and 90% shading at experiment sites established on reclaimed fallow agricultural- and degraded sites in Central Kalimantan, Indonesia. Groundwater tables on the sites were at about 50 cm depth, the sites were maintained vegetation free and root ingrowth to gas flux monitoring locations was prevented. Half of the four shading areas received NPK-fertilization 50 kg ha−1 for each of N, P and K during the experiment and the other half was unfertilized. Increases in shading created a lasting decrease in peat temperatures, and decreased diurnal temperature fluctuations, in comparison to less shaded plots. The largest peat temperature difference in the topmost 50 cm peat profile was between the unshaded and 90% shaded surface, where the average temperatures at 5 cm depth differed up to 3.7 °C, and diurnal temperatures at 5 cm depth varied up to 4.2 °C in the unshaded and 0.4 °C in the 90% shaded conditions. Highest impacts on the heterotrophic CO2 fluxes caused by the treatments were on agricultural land, where 90% shading from the full exposure resulted in a 33% lower CO2 emission average on the unfertilized plots and a 66% lower emission average on the fertilized plots. Correlation between peat temperature and CO2 flux suggested an approximately 8% (unfertilized) and 25% (fertilized) emissions change for each 1 °C temperature change at 5 cm depth on the agricultural land. CO2 flux responses to the treatments remained low on degraded peatland. Fertilized conditions negatively correlated with N2O efflux with increases in temperature, suggesting a 12–36% lower efflux for each 1 °C increase in peat temperature (at 5 cm depth) at the sites. Despite the apparently similar landscapes of fallow agricultural land and degraded peatland sites, the differences in greenhouse gas dynamics are expected to be an outcome of the long-term management differences.

Degraded tropical peatlands in Southeast Asia are a major challenge for reforestation. Often treeless, drained and several times burnt, these peatland areas are nutrient-poor hostile environments prone to droughts, heavy flooding and extreme diurnal temperature changes. In order to succeed in establishment of a viable tree stand, careful selection of species and management techniques is needed. In this study we investigated the suitability of five native tree species for reforestation of tropical peatlands with three site preparation treatments for potentially enhancing seedling success: weeding, mounding and fertilizing. The study area was a clear-cut, drained and repeatedly burnt former tropical peat swamp forest in Central Kalimantan, Indonesia. Seedlings were grown in a field nursery, planted in the field and their growth and survival were monitored regularly for 1.5 years. Seedling growth in response to environmental variables and treatments was studied by linear mixed models and seedling survival with Cox regression models. In most cases, weeding and fertilizing proved beneficial for the growth and survival of the seedlings, whereas mounding only had a minor impact on seedling performance. The seedlings of Shorea balangeran performed the best and can be recommended for reforestation of heavily degraded areas. Alstonia pneumatophora and Dacryodes rostrata performed relatively well depending on the treatments, whereas Dyera polyphylla had mixed results with problems in seedling production, and Campnosperma squamatum performed rather poorly. The effects of wildfires which engulfed the study area two years after planting were also monitored and are discussed.