Nitrous oxide emissions are presented from three independent atmospheric inversion frameworks. The frameworks are: 1) INVICAT: an inversion using the atmospheric transport model, TOMCAT and a 4D-Var optimisation method; 2) JAMSTEC: an inversion using the MIROC4-ACTM atmospheric transport model and a Bayesian analytical optimisation method; and 3) PYVAR: an inversion using the LMDZ5 atmospheric transport model and a 4D-var optimisation method. The emissions were optimised monthly and have been re-gridded from the model native resolution to 1.0 by 1.0 degrees. The files for TOMCAT and LMDZ5 (i.e. the inversion frameworks INVICAT and PYVAR, respectively) contain two flux variables: 1) the prior fluxes as estimated a priori, and 2) the posterior fluxes as estimated by the inversion. The file for the JAMSTEC inversion, contains five flux variables: 1) flux_apri_land: the prior fluxes over land, 2) flux_apri_ocean: the prior fluxes over ocean, 3) flux_apri_fossil: the prior estimate of emissions from combustion, 4) flux_apos_land: posterior fluxes over land estimated by the inversion, and 5) flux_apos_ocean: the posterior fluxes over ocean estimated by the inversion. Note that flux_apri_fossil was not optimised in the inversion but for the total posterior N2O emission, needs to be added to the flux_apos_ocean and flux_apos_land variables.

{"references": ["Thompson, R. L. et al. Nitrous oxide emissions 1999 to 2009 from a global atmospheric inversion, Atmos. Chem. Phys. 14, 1801-1817 (2014)", "Wilson, C. et al. Development of a variational flux inversion system (INVICAT v1.0) using the TOMCAT chemical transport model. Geosci Model Dev 7(5), 2485\u20132500 (2014)", "Patra, P. K. et al. Improved Chemical Tracer Simulation by MIROC4.0-based Atmospheric Chemistry-Transport Model (MIROC4-ACTM). SOLA 14, 91\u201396 (2018)."]}