Simulating carbon and water fluxes using a coupled process-based terrestrial biosphere model and joint assimilation of leaf area index and surface soil moisture

[摘要] Reliable modeling of carbon and water fluxes is essential for understandingthe terrestrial carbon and water cycles and informing policy strategiesaimed at constraining carbon emissions and improving water use efficiency.We designed an assimilation framework (LPJ-Vegetation and soil moistureJoint Assimilation, or LPJ-VSJA) to improve gross primary production (GPP)and evapotranspiration (ET) estimates globally. The integrated model, LPJ-PM (LPJ-PT-JPL SM Model) as the underlying model, was coupled from theLund–Potsdam–Jena Dynamic Global Vegetation Model (LPJ-DGVM version 3.01)and a hydrology module (i.e., the updated Priestley–Taylor Jet PropulsionLaboratory model, PT-JPL SM ). Satellite-based soil moisture products derived from the Soil Moisture and Ocean Salinity (SMOS) and Soil Moisture Active and Passive (SMAP) and leaf area index (LAI) from the Global LAnd and Surface Satellite (GLASS) product were assimilated into LPJ-PM to improve GPP and ET simulations using a proper orthogonal decomposition (POD)-based ensemble four-dimensional variational assimilation method (PODEn4DVar). The joint assimilation framework LPJ-VSJA achieved the best model performance (with an R 2 ( coefficient of determination) of 0.91 and 0.81 and an ubRMSD (unbiased root mean square deviation) reduced by 40.3 % and 29.9 % for GPP and ET, respectively, compared with those of LPJ-DGVM at the monthly scale). The GPP and ET resulting from the assimilation demonstrated a betterperformance in the arid and semi-arid regions (GPP: R 2   =  0.73,ubRMSD  =  1.05 g C m −2  d −1 ; ET: R 2   =  0.73, ubRMSD  =  0.61 mm d −1 ) than in the humid and sub-dry humid regions (GPP: R 2   =  0.61, ubRMSD  =  1.23 g C m −2  d −1 ; ET: R 2   =  0.66; ubRMSD  =  0.67 mm d −1 ). The ET simulated by LPJ-PM that assimilated SMAP or SMOS data had a slight difference, and the SMAP soil moisture data performed better thanSMOS data. Our global simulation modeled by LPJ-VSJA was comparedwith several global GPP and ET products (e.g., GLASS GPP, GOSIF GPP, GLDASET, and GLEAM ET) using the triple collocation (TC) method. Our products,especially ET, exhibited advantages in the overall error distribution(estimated error ( μ ): 3.4 mm per month; estimated standard deviationof μ : 1.91 mm per month). Our research showed that the assimilationof multiple datasets could reduce model uncertainties, while the modelperformance differed across regions and plant functional types. Ourassimilation framework (LPJ-VSJA) can improve the model simulationperformance of daily GPP and ET globally, especially in water-limitedregions.

[学科分类] 妇产科学