Assessing modelled methane emissions over northern wetlands by the JULES-HIMMELI model

Abstract

Northern wetlands are considered to be one of the most significant natural sources of methane (CH4) emissions. The default wetland CH4 emission scheme in JULES, a current state-of-art land surface model, only takes into account the CH4 emissions from inundated wetland areas in a simple manner based on soil temperature and substrate availability. In this work, a process-based peatland CH4 emission model HIMMELI was integrated with JULES, and the HIMMELI parameters were optimized with measured CH4 flux at six northern wetland sites for each site separately or multi-sites simultaneously. The simulated CH4 emission was significantly improved when using the optimized parameter values, with the bias of 54.88 mg m−2 d−1 averaged across all the studied sites in the simulation using the default parameter (DPR) values being reduced to −0.70 mg m−2 d−1 in the simulations using parameters values derived from the single site optimization (SSO) for each site. In the simulations using parameters values from the averages of single site optimization (SSO_AVG) and the multi-site optimization (MSO), the biases averaged across all the studied sites were −7.39 mg m−2 d−1 and −8.36 mg m−2 d−1, respectively. The MSO simulations demonstrated more stable root mean square error (RMSE) between the simulated and observed methane emissions than the SSO_AVG simulations over the studied sites, when the RMSEs of SSO simulations were used as reference points. To further reduce the uncertainties in the simulated CH4 emissions by the JULES-HIMMELI model, model processes related to the environment conditions (e.g. water table, soil carbon and vegetation) of wetland and northern wetland CH4 emission processes (e.g. snow and ice covering effect) are suggested to be improved in JULES and HIMMELI, respectively. This study presents a comprehensive analysis of the impact of different parameters on the CH4 emission in the JULES-HIMMELI model and obtains optimal parameter values for modelling CH4 emissions at the studied northern wetlands. These findings pave the way for accurate regional estimates of northern wetland CH4 emission.