NSFDEB-NERC: Spatial and temporal tradeoffs in CO2 and CH4 emissions in tropical wetlands

Wetlands are a C paradox, with huge capacity for C storage but also a strong source of C as methane (CH4). Wetlands worldwide are highly sensitive to climate change, given dependencies on rain and groundwater. Seasonal shifts in saturation have potential to lead to highly dynamic greenhouse gas (GHG) fluxes, in the form of CO2 and CH4. Tropical wetlands are particularly understudied but make outsized contributions to the global CH4 budget. Though Brazil has vast wetlands, previous work largely focused on permanently saturated systems, e.g., Amazon or Pantanal floodplains. In contrast, the poorly studied Cerrado domain includes permanent (peatlands) and seasonally inundated (seasonal grasslands), wetlands and dry grasslands, with saturation shifting seasonally and episodically. Climate change is likely to bring warmer and drier conditions to this region, increasing areas of seasonal and dry grasslands and concomitant potential increases in GHG emissions. Given current paucity of observations and importance of predicting future of these critical seasonal systems, it is vital to determine mechanisms driving spatial and temporal shifts in GHG emissions and C storage across the saturation gradient, as well as changes in spatial extent of these systems through time. Our overall objective is to overcome these knowledge gaps in sites in Brazil.

Our combined field and modeling approaches in Cerrado tropical grasslands will focus on answering three questions: Q1. What are the drivers of spatial and temporal heterogeneity in sea storage and flux across saturation gradients? Q2. How does saturation extents (areas and perimeters) in tropical grasslands change over box? Seasonal and decadal scales? Q3. How will rates and forms of sea emissions from tropical grasslands change under future climates?

To test Q1, spatially distributed measurements will be coupled with high-temporal resolution measurements to understand GHG, soil and vegetation C dynamics across the saturation gradient. GHG variability will be measured spatially with static flux chambers and temporally with automated chambers. Site changes through time will be determined by initial soil characterization, combined with seasonal measurements of plant phenology, stomatal conductance, porewater chemistry, baseline climate and groundwater seasonally. To test Q2, high resolution remote sensing and field reference data will be combined to map wetland extent, seasonally 14C and 210Pb dating will be used to understand wetlands extent changes at decadal scales. To test Q3, data from Q1 and Q2 will be utilized in simulations with E3SM Land Models coupled to PFLOTRAN reactive transport models. Estimates of C sequestration patterns and GHG emissions will be spatially simulated with projections of C balance changes and GHG with expected shifts in regional climate and hydrology.