These files contain estimates of global CO2 fluxes, split into GPP, respiration, and air-sea components produced by the WOMBAT v2.0 flux-inversion system (see https://arxiv.org/abs/2210.10479). These fluxes are further decomposed into trend and seasonality. The file WOMBAT_v2_CO2_gridded_climatology_samples.nc4 contains the estimated spatial fields for each beta in the paper that define the trend/seasonality of the fluxes. The bottom-up estimates are provided for each beta, as well as samples from the posterior distribution. A second file, WOMBAT_v2_CO2_gridded_flux_samples.nc4, contains bottom-up estimates and posterior samples for the flux component fields. These are split into components and the parts of the decomposition, so for example the linear component of GPP is in gpp_linear_bottom_up/gpp_linear_posterior for the bottom-up/posterior. The different fields can be summed to get meaningful quantities, such as the NEE (sum of all GPP and respiration parts), or the net flux (sum of all parts). The final file, samples-LNLGIS.rds, contains samples from the posterior distribution of the model parameters. This file format may be read using the readRDS function in the R programming language.