Spatiotemporal Variations in Hydrology Drive Greenhouse Gas Emissions in Tropical Peatlands

In the previous issue, Hirano et al. (2025) assess the impacts of land use change and drought on greenhouse gas (GHG) emissions from tropical peatlands of Southeast Asia (Figure 1). This region is responsible for nearly half of all global peatland-related greenhouse gas emissions despite containing only 5% of global peatlands (UNEP, 2022). These tropical peatland ecosystems were naturally forested with underlying soils perennially saturated by high groundwater levels (GWLs). Anaerobic conditions resulted in net uptake of carbon through burial of partially decayed woody vegetation, albeit partially offset by methane (CH4) emissions (Swails et al., 2021). This carbon sink built up peat soils over thousands of years, several meters deep in many locations, across parts of Borneo, Sumatra, and peninsular Malaysia. More than half of peat swamp forests have been cleared in recent decades, with much of the area converted to land uses such as plantations, agriculture and infrastructure. Land conversion in peatlands requires construction of extensive drainage canals to lower groundwater levels. Once drained, peatlands experience rapid subsidence during the first year, with slower rates thereafter (Hooijer et al., 2012). Much of the initial peat elevation drop is compaction as water is removed and the overlaying weight compresses the drying peat, but draining also oxygenates the upper peat layers. Global peatland distribution. The black oval highlights the tropical peatland region of study described by Hirano et al. (2025). Figure adapted from UNEP (2024). Hydrology governs carbon storage dynamics in tropical peatlands (Cobb et al., 2017). Peatland drainage initiates aerobic microbial respiration that reduces anoxic CH4 fluxes while greatly increasing CO2 emissions (Swails et al., 2021). Changes in persistent flooding and anoxic conditions due to climatic shifts, peatland drainage, or both, thereby destabilize peat carbon stocks, increasing the flux of GHGs to the atmosphere. Net carbon emissions will continue perennially until a new stable peatland morphology is reached with the established hydrologic regime (Cobb et al., 2020). Unless peatland hydrology in these regions is restored, net emissions from drained tropical peatlands could consume 7%–9% of the remaining greenhouse gas budget necessary to keep global temperature rise below 2°C (Leifeld et al., 2019). Greenhouse gas emissions from tropical peatlands are tightly associated with water table depths but these levels vary greatly in space and time. GWLs fluctuate annually with seasonal rainfall, interannually with drought conditions—particularly El Niño-Southern Oscillation (ENSO) events, and spatially with land cover, land use and management practices. Despite these known factors that influence greenhouse gas emissions, to date, regional estimations of changing carbon emissions have followed Intergovernmental Panel on Climate Change (IPCC) protocols, using fixed, land cover-specific emission factors (EFs) (e.g., Miettinen et al., 2017; Sasmito et al., 2025) that cannot address the specific locations and timing of carbon fluxes. Now, Hirano et al. (2025) present a methodology for using remotely sensed precipitation (Global Satellite Mapping of Precipitation (GSMaP)) to accurately estimate spatiotemporal variations in GWLs. Based upon this improved depiction of regional hydrology, it was possible to model fluctuating carbon dioxide and methane emissions by month, year and across major land covers and land use practices present in the peatlands of Southeast Asia. By integrating precipitation, GWL and eddy covariance flux tower data from several locations across the region, the authors created empirical models specific to peat swamp forest, drained-peat swamp forest and managed peatlands that estimate net ecosystem exchanges (NEEs) for CO2 and CH4 of these three main land cover types. Applying these models to historical precipitation observations from 2011 to 2020 provided estimates of the spatial distributions of monthly mean GWL and the associated NEECO2 and NEECH4 that could be tied to both land cover and the regional impacts of El Niño droughts. Subsequently, future projections of monthly GWLs and likely carbon fluxes for 2041–2060 were made based upon projected precipitation changes associated with climate change scenarios depicted by five models in the Coupled Model Intercomparison Project Phase 6 (CMIP6). Establishing a methodology capable of monitoring the slow, continuous but cryptic emissions of greenhouse gases (NEECO2eq) from major peatland landcover types is an important accomplishment that can help regional governments to evaluate landscape restoration activities, quantify the effects of reduced emissions from deforestation and degradation (REDD) projects, and achieve nationally determined contributions (NDCs) for GHG emissions. Hirano et al. (2025) provide a tour de force treatment of non-fire-related CO2 and CH4 emissions from the tropical peatlands of Southeast Asia but microbial respiration makes up less than half of the carbon escaping from these landscapes. Perennial emissions from microbial decay processes are exacerbated by immense short-term emissions of gaseous carbon from episodic peat fires that occur whenever and wherever drought enables flaming surface fires to ignite smoldering combustion in the underlying peat (Yokelson et al., 2022). Such fires may burn for weeks or months and, in aggregate over the years, wildfire emissions may equal or exceed those from microbial decay (Silvius et al., 2006). Peatland disturbance and drainage may cause another 10% (Lennartz et al., 2022) of carbon to leach into effluent waters from peat soils as dissolved and particulate organic carbon (Gandois et al., 2020). Tracking spatiotemporal variations in GWLs has brought us closer to being able to make complete landscape-level estimates of carbon fluxes from tropical peatlands. In addition to facilitating estimation of emissions from aerobic decay processes, spatiotemporal GWLs could provide governments with information for effective management of fire use in peatlands. Fire is a cost-effective landscape management tool in cleared and drained peatlands. As such, land clearing fires will continue to be frequently used, but these fires are not the real problem. Large carbon losses and associated smoke impacts only begin when conditions arise (e.g., GWLs >30 cm (Putra et al., 2018)) where these fires frequently transition into peat fires that slowly burn through the underlying soils. Having near-real time spatiotemporal estimates of GWLs would enable a much more granular approach to regional management of fire risk. Monitoring policy-set GWL thresholds would enable timely and spatially relevant fire management activities that could reduce the likelihood or spread of peat fires like establishing burn bans or prepositioning of fire response forces. While not a panacea, the methodology for estimating the spatial and temporal GWL fluctuations in tropical peatlands that Hirano et al. (2025) provide may also prove useful for addressing carbon emissions from fires and leaching, since they too are tied to hydrologic changes. M. A. Cochrane thanks E. A. Davidson and C. V. Cochrane for helpful comments that improved the draft manuscript, S. Bomar for help with the figure and acknowledges support from NASA Grant 80NSSC20K0408. The author declares no conflicts of interest relevant to this study. Data were not used, nor created for this research.