LES ARM Symbiotic Simulation and Observation (LASSO) Workflow: Model-Observation “Data Cubes”
Poster PDF
Category
ARM next generation – Megasite and LES activities
Authors
|
Andrew M. Vogelmann — Brookhaven National Laboratory
|
Heng Xiao — Pacific Northwest National Laboratory
|
|
William I. Gustafson — Pacific Northwest National Laboratory
|
Zhijin Li — University of California
|
|
Tami Fairless — Brookhaven National Laboratory
|
Xiaoping Cheng — National University of Defense Technology
|
|
Satoshi Endo — Brookhaven National Laboratory
|
|
Description
The Atmospheric Radiation Measurement (ARM) Climate Research Facilities’ Large-Eddy Simulation (LES) ARM Symbiotic Simulation and Observation (LASSO) Workflow is currently being designed to provide output from routine LES to complement ARM’s extensive observations to support the study of atmospheric processes and support the improvement of the parameterization of these processes in climate models. An overview of LASSO is presented by Gustafson et al. and the model forcing methodology is presented by Endo et al. This presentation describes how the LES output will be combined with observations to construct multi-dimensional and dynamically consistent “data cubes,” aimed at providing the best description of the atmospheric state for use in analyses by the community. The megasite observations are used to constrain large-eddy simulations to provide representative spatial and temporal coverage of observables as well as information on processes that cannot be observed. Statistical comparisons of model output with their observables are used to assess the quality of a given simulated realization and its associated uncertainties. Thus, a data cube is a model-observation package that provides: (1) categorization of daily weather conditions and their specific attributes for the simulated period; (2) performance metrics of model-observation statistical summaries to assess the simulations and the ensemble spread; (3) statistical summaries of key model property output that cannot be or are very difficult to observe; and (4) snapshots of the 4-D simulated fields from the integration period. Searchable metrics and quicklooks will be provided for #1-3 to assist users in finding cases of interest. The data cubes ultimately will be accompanied by tools designed for easy access to cube contents from within the ARM archive and externally, the ability to compare multiple data streams within an event as well as across events, and the ability to use common grids and time sampling, where appropriate.
