Inclusion of Linearized Moist Physics in NASA’s Goddard Earth Observing System Data Assimilation Tools

Abstract

Inclusion of moist physics in the linearized version of a weather forecast model is beneficial in terms of variational data assimilation. Further, it improves the capability of important tools, such as adjoint-based observation impacts and sensitivity studies. A linearized version of the relaxed Arakawa-Schubert (RAS) convection scheme has been developed and tested in NASA's Goddard Earth Observing System data assimilation tools. A previous study of the RAS scheme showed it to exhibit reasonable linearity and stability. This motivates the development of a linearization of a near-exact version of the RAS scheme. Linearized large-scale condensation is included through simple conversion of supersaturation into precipitation. The linearization of moist physics is validated against the full nonlinear model for 6-and 24-h intervals, relevant to variational data assimilation and observation impacts, respectively. For a small number of profiles, sudden large growth in the perturbation trajectory is encountered. Efficient filtering of these profiles is achieved by diagnosis of steep gradients in a reduced version of the operator of the tangent linear model. With filtering turned on, the inclusion of linearized moist physics increases the correlation between the nonlinear perturbation trajectory and the linear approximation of the perturbation trajectory. A month-long observation impact experiment is performed and the effect of including moist physics on the impacts is discussed. Impacts from moist-sensitive instruments and channels are increased. The effect of including moist physics is examined for adjoint sensitivity studies. A case study examining an intensifying Northern Hemisphere Atlantic storm is presented. The results show a significant sensitivity with respect to moisture.

Assessing the impact of observations in a multi‐year reanalysis

“…The adjoint model of the forecasting model includes the adjoint of the cubed‐sphere hydrodynamical finite‐volume core (Jong Kim, Personal communication, July 2015), simple vertical diffusion and boundary layer, and the simplified large‐scale moist processes of Holdaway et al . (). All versions of the FSOI applications in GEOS ignore the effect of Incremental Analysis Update (IAU) in the adjoint model integration.…”

“…A remaining issue in the FSOI approach in general, and in the implementation for this work in particular, is associated with how well nonlinearities are approximated by the tangent linear and its corresponding adjoint components. When it comes to the adjoint of the nonlinear general circulation model, the study of Holdaway et al (2014), and the upgrades then implemented in the simplified physics, indicate that the linear impact is capable of recovering roughly 80% of the nonlinear error reduction. Their work validates the implementation of a 1 • adjoint, using a 1 • nonlinear (trajectory) model; the present work employs the 0.5 • MERRA-2 nonlinear trajectory model with a 1 • adjoint -to mimic the configuration of GEOS FP -thus introducing an approximation to the approach (and some inconsistency).…”

“…All nonlinear forecasts were generated at the full MERRA-2 cubed-sphere horizontal grid resolution of 0.5 • , roughly corresponding to 50 km, and with 72 vertical levels with a top pressure at 0.01 hPa. The adjoint model of the forecasting model includes the adjoint of the cubed-sphere hydrodynamical finite-volume core (Jong Kim, Personal communication, July 2015), simple vertical diffusion and boundary layer, and the simplified large-scale moist processes of Holdaway et al (2014). All versions of the FSOI applications in GEOS ignore the effect of Incremental Analysis Update (IAU) in the adjoint model integration.…”

“…Forecast sensitivities are generated on a 1 • horizontal grid resolution and the 72 vertical levels of the MERRA-2 nonlinear model. The matrix C corresponding to a linearized total energy norm that accounts for moist perturbations as in equation (10) of Holdaway et al (2014), with the latent heat term weighted by a factor of 0.3. The matrix P incorporates a projection operator that simply amounts to a diagonal matrix with ones along most of its diagonal except for some zeroes to exclude perturbation contributions above the 60th level of the model (roughly 0.6168 hPa) to steer results away from sponge-layer effects.…”

The Impact of Increased Frequency of Rawinsonde Observations on Forecast Skill Investigated with an Observing System Simulation Experiment

“…A moist adjoint of the GEOS-5 model has been developed to help estimate the impact of individual observation types on the quality of the 24-h forecast measured using a moist total energy norm, as described by Holdaway et al (2014). The moist total energy norm is calculated from the surface to 100 hPa as in Ehrendorfer et al (1999), with parameter chosen so that the contribution of temperature and specific humidity terms are of similar magnitude.…”