Figure and table scripts, processed data from: NOAA observations, MLS measurements, GMI, UCI, and LMDz5 CTMs. This dataset includes processed numerical output from N₂O tracer experiments that quantify the stratospheric signal of N₂O, driven by atmospheric chemistry and transport, at Earth’s surface in observations and 3 CTMs.

Nitrous oxide (N₂O) is a long-lived greenhouse gas that affects atmospheric chemistry and climate. In this work we use satellite measurements of N₂O, ozone (O₃), and temperature from the Aura Microwave Limb Sounder (MLS) instrument to calculate stratospheric loss of N₂O, and thus its atmospheric lifetime. Using three chemistry transport models (CTMs) simulating the Aura period 2005-2017, we verify the stratospheric sink using MLS data and follow that loss signal down to the surface and compare with surface observations. Stratospheric loss has a strong seasonal cycle and is further modulated by the Quasi-Biennial Oscillation (QBO); these cycles are seen equally in both observations and the models.  When filtered for interannual variability, the modeled surface signal is QBO-caused, and it reproduces the observed pattern, highlighting the potential role of the QBO in tropospheric chemistry and composition, as well as in model evaluation.  The observed annual surface signal in the northern hemisphere (NH) matches well with the models run without emissions, indicating the annual cycle is driven mostly by stratosphere-troposphere exchange (STE) flux of N₂O-depleted air and not surface N₂O emissions.  In the southern hemisphere (SH), all three models disagree and thus provide no guidance, except for indicating that modeling STE in the SH remains a major model uncertainty.  Parallel model simulations of CFCl₃, which has greater stratospheric loss that N₂O and possibly surreptitious emissions, show that its interannual variations parallel those of N₂O, and thus the observed N₂O variability can identify the stratospheric component of the observed CFCl₃ variability.