UQ eSpace

Williamson, Hera; Ostréus, Rebecka

doi:10.14264/530d0f8

Cited by 1 publication

(1 citation statement)

References 19 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The choice of Stan has proven effective for this project, though it does not scale well to larger ensembles. Going forward, a new version of the tools defaulting to Maximum A Posterior (MAP) estimation and using efficient parallel implementation has just been released enabling millions of predictions in just a few seconds (Williamson & Volodina, 2020).…”

Section: High‐tune Explorer (Htexplo) a Statistical Tool To Calibrate...mentioning

confidence: 99%

Process‐Based Climate Model Development Harnessing Machine Learning: I. A Calibration Tool for Parameterization Improvement

Couvreux

Hourdin

Williamson

et al. 2021

J Adv Model Earth Syst

View full text Add to dashboard Cite

Atmospheric global or regional circulation models used either for numerical weather prediction (NWP) or climate studies encompass a dynamical core and a physical component. The dynamical core computes the spatio-temporal evolution of atmospheric state variables by solving a discrete version of the fluid dynamic equations. The physical component quantifies the impact on the resolved variables of radiative, thermodynamical, and chemical processes, as well as dynamical processes that occur at scales smaller than the computational grid. These processes are handled by a suite of sub-models, most often referred to as parameterizations, which provide source terms in the resolved-scale equations. Parameterizations (e.g., turbulence, convection, radiation, microphysics) are often based on a mixture of physical principles and heuristic description of the involved processes, of their interactions and of their impact on the larger resolved scales. Although it is difficult to trace back the origin of the term "parameterization" in climate modeling, it semantically points to the fact that the sub-models summarize the processes as functions of the model state vector x (typically the value of zonal and meridional wind, temperature, and water phases at each point of the three-dimensional [3D] model grid) that depends on some free parameters. These free parameters arise from the simplification of the complex nature of the subgrid processes (e.g., assuming a bulk thermal plume instead of a population of plumes, stationarity). The atmospheric model can be summarized as      () (,)

show abstract

Section: High‐tune Explorer (Htexplo) a Statistical Tool To Calibrate...mentioning

confidence: 99%