Apers, S., M. Bechtold, A. J. Baird, A. R. Cobb, G. Dargie, A. Katimon, R. D. Koster, M. Lampela, S. P. Mahanama, S. Page, R. H. Reichle, J. Vanderborght, and G. J. M. De Lannoy and collaborators:
"Integrating tropical peatland hydrology into a global land surface model (PEATCLSM)"
Presentation at the General Assembly of the European Geosciences Union, Online, 2020.

Abstract:
Tropical peatlands have a specific hydrology that regulates their internal processes and functioning. External disturbances such as drainage, land cover and land use changes, and climate change could disrupt the peat-specific hydrology and convert the immense peatland carbon stocks into strong greenhouse gas (GHG) emitting sources. The need for (more) accurate monitoring of GHG emissions has led to the development of complex biogeochemical models, which highly depend on proper representation of peat-specific land surface hydrology. However, the latter is often inadequately accounted for in global Earth system modeling frameworks.

In this research, we leverage the PEATCLSM modules recently developed for the Catchment land surface model (CLSM) of the NASA Goddard Earth Observing System framework (Bechtold et al., 2019). These modules were evaluated for northern peatlands, hereafter referred to as PEATCLSMN. Here, we present an extended version of PEATCLSM for tropical peatlands with literature-based parameter sets for natural (PEATCLSMT,Natural) and drained (PEATCLSMT,Drained) tropical peatlands. A suite of modeling experiments was conducted to compare the performance of PEATCLSMT,Natural, PEATCLSMT,Drained, PEATCLSMN, and the currently operational CLSM version that includes peat parameters but no peat-specific model structure (CLSMO). Simulations over major tropical peatland regions in Southeast Asia, the Congo Basin, and South and Central America were evaluated with a comprehensive and self-compiled dataset of groundwater table depth (WTD) and evapotranspiration (ET). Preliminary results show that the simulated WTD from CLSMO exhibits too much temporal variability and large biases, either positive or negative. The temporal correlation coefficient between simulated and observed WTD for both PEATCLSMT,Natural (over undeveloped peatlands only) and PEATCLSMT,Drained (over drained peatlands only) is similar to that of PEATCLSMN. However, both tropical versions reduce the average absolute bias to a few centimeters. Performance differences across the major tropical peatland regions are discussed.


Home

NASA-GSFC / GMAO / Rolf Reichle