B110-0006
Global gross primary productivity and terrestrial ecosystem respiration products derived from Breathing Earth System Simulator (BESS)
Global gross primary productivity and terrestrial ecosystem respiration products derived from Breathing Earth System Simulator (BESS)
Wednesday, 16 December 2020
Poster
Abstract:
Gross primary productivity (GPP) and terrestrial ecosystem respiration (TER) are the two largest carbon fluxes between the atmosphere and land. There is a great need to accurately quantify these fluxes for understanding the changes in the terrestrial carbon sink over large spatial scales. In this study, we presented a new set of global GPP and TER products at fine spatial (1-km) and temporal (daily) resolutions from 2001 to 2019, using a simplified remote sensing-forced and process-based model, the Breathing Earth System Simulator (BESS). BESS couples the processes of atmosphere and canopy radiative transfer, photosynthesis, evapotranspiration and ecosystem respiration, using multi-satellite products as inputs. We evaluated BESS products against ground-based eddy covariance measurements (i.e. FLUXNET2015 with 206 flux sites and 1496 site years) at the site scale, and against FLUXCOM products on a global scale. At the site scale, daily BESS products agreed with FLUXNET measurements with R2=0.60 and RMSE=2.70 gC/m2/d for GPP, R2=0.53 and RMSE=2.01 gC/m2/d for TER, and R2=0.27 and RMSE=2.26 gC/m2/d for net ecosystem exchange (NEE). They captured a majority of seasonal variability and about half of interannual variability in FLUXNET measurements. On a global scale, the mean annual sums of BESS GPP and TER patterns agreed with FLUXCOM products with R2 = 0.83 and RMSE=435.50 gC/m2/yr for GPP, R2=0.86 and RMSE=343.05 gC/m2/yr for TER, and R2=0.29 and RMSE=202.46 gC/m2/yr for NEE. Over the period of 2001–2015, BESS estimated the mean global GPP, TER and NEE as 148.68 ± 2.67 PgC/yr, 126.95 ± 1.26 PgC/yr and -21.32 ± 1.40 PgC/yr, respectively, with significant ascending trends by 0.52 PgC/yr (p < 0.05) for GPP, by 0.26 PgC/yr (p < 0.05) for TER and by -0.32 PgC/yr (p < 0.05) for NEE, respectively. Overall, BESS GPP and TER estimates were comparable with FLUXNET measurements and FLUXCOM products. BESS NEE showed more similar histogram distribution with FLUXNET measurements compared to FLUXCOM NEE. BESS GPP, TER and NEE can serve as a set of independent products from other global products.