Simulation of atmospheric N<sub>2</sub>O with GEOS-Chem and its adjoint: evaluation of observational constraints

Abstract

Abstract. We describe a new 4D-Var inversion framework for nitrous oxide (N2O) based on the GEOS-Chem chemical transport model and its adjoint, and apply it in a series of observing system simulation experiments to assess how well N2O sources and sinks can be constrained by the current global observing network. The employed measurement ensemble includes approximately weekly and quasi-continuous N2O measurements (hourly averages used) from several long-term monitoring networks, N2O measurements collected from discrete air samples onboard a commercial aircraft (Civil Aircraft for the Regular Investigation of the atmosphere Based on an Instrument Container; CARIBIC), and quasi-continuous measurements from the airborne HIAPER Pole-to-Pole Observations (HIPPO) campaigns. For a 2-year inversion, we find that the surface and HIPPO observations can accurately resolve a uniform bias in emissions during the first year; CARIBIC data provide a somewhat weaker constraint. Variable emission errors are much more difficult to resolve given the long lifetime of N2O, and major parts of the world lack significant constraints on the seasonal cycle of fluxes. Current observations can largely correct a global bias in the stratospheric sink of N2O if emissions are known, but do not provide information on the temporal and spatial distribution of the sink. However, for the more realistic scenario where source and sink are both uncertain, we find that simultaneously optimizing both would require unrealistically small errors in model transport. Regardless, a bias in the magnitude of the N2O sink would not affect the a posteriori N2O emissions for the 2-year timescale used here, given realistic initial conditions, due to the timescale required for stratosphere–troposphere exchange (STE). The same does not apply to model errors in the rate of STE itself, which we show exerts a larger influence on the tropospheric burden of N2O than does the chemical loss rate over short (< 3 year) timescales. We use a stochastic estimate of the inverse Hessian for the inversion to evaluate the spatial resolution of emission constraints provided by the observations, and find that significant, spatially explicit constraints can be achieved in locations near and immediately upwind of surface measurements and the HIPPO flight tracks; however, these are mostly confined to North America, Europe, and Australia. None of the current observing networks are able to provide significant spatial information on tropical N2O emissions. There, averaging kernels (describing the sensitivity of the inversion to emissions in each grid square) are highly smeared spatially and extend even to the midlatitudes, so that tropical emissions risk being conflated with those elsewhere. For global inversions, therefore, the current lack of constraints on the tropics also places an important limit on our ability to understand extratropical emissions. Based on the error reduction statistics from the inverse Hessian, we characterize the atmospheric distribution of unconstrained N2O, and identify regions in and downwind of South America, central Africa, and Southeast Asia where new surface or profile measurements would have the most value for reducing present uncertainty in the global N2O budget.

Top-down constraints on global N<sub>2</sub>O emissions at optimal resolution: application of a new dimension reduction technique

“…This has the advantage of avoiding any aggregation errors associated with traditional clustering methods. However, our previous work (Wells et al, 2015) has shown that the degrees of freedom for atmospheric N 2 O inversions is typically much less than the native model grid dimension and, furthermore, that native resolution optimizations have limited ability to resolve any temporal (e.g., seasonal) N 2 O emission biases. We therefore apply two alternate approaches to reduce the dimension of the inverse problem: (1) a 4D-Var inversion solving for emissions on aggregated, geographically defined land and ocean regions and (2) a 4D-Var inversion solving for emissions on a reduced emission basis set defined using an SVD-based information content analysis.…”

“…The N 2 O simulation employed here, previously described by Wells et al (2015), is based on the GEOS-Chem CTM (www.geos-chem.org) with GEOS-5 assimilated meteorological data from the NASA Goddard Earth Observing System. We use a horizontal resolution of 4 • × 5 • with 47 vertical levels from the surface to 0.01 hPa as well as time steps of 30 min for transport and 60 min for emissions and chemistry.…”

“…This type of variability poses a major challenge to bottom-up and top-down efforts to quantify N 2 O surface fluxes and attribute them to specific times, locations, and mechanisms. The relatively sparse coverage of measurement sites and low atmospheric variability (because of the long N 2 O lifetime, surface mixing ratios typically vary by < 10 ppb on a ∼ 325 ppb background) compound the challenge and limit the spatial and temporal resolution at which emission fluxes can be inferred (Wells et al, 2015). As a result, global N 2 O inversions often employ some aggregation strategy to optimize emissions for a small set of geographic regions (e.g., Hirsch et al, 2006;Huang et al, 2008;Saikawa et al, 2014).…”

“…The global N 2 O source is reasonably well constrained (15.7 to 20.1 Tg N yr −1 for years 1999-2009Prather et al, 2012;Saikawa et al, 2014;Thompson et al, 2014a, c) by its atmospheric abundance and estimated lifetime. However, attribution of this source to specific regions and sectors has been hindered by the sparse global observing network and by the weak variability in N 2 O mixing ratios (e.g., Wells et al, 2015). Quantitative interpretation of atmospheric N 2 O measurements in terms of globally resolved emissions thus first requires a rigorous assessment of how results hinge on the modeling framework employed.…”

“…Past global N 2 O inversion studies have established the initial conditions in a variety of ways: from a forward model spinup that is then evaluated against observations (e.g., Huang et al, 2008); by including the initial condition as a separate adjustable parameter in the source optimization (e.g., Saikawa et al, 2014;Thompson et al, 2014a); or from interpolation of atmospheric observations (e.g., Wells et al, 2015). To our knowledge there has not yet been a detailed evaluation of these different methods and their impacts on N 2 O source inversions.…”