Doubly Periodic SCREAM: A Tool to Connect High Resolution Global Modeling with ARM Observations



Peter Bogenschutz — Lawrence Livermore National Laboratory *
Yunyan Zhang — Lawrence Livermore National Laboratory
Peter Caldwell — Lawrence Livermore National Laboratory
* presenting author


General topics


In this poster we present the doubly periodic configuration of the Simplified Cloud Resolving Energy Exascale Earth System Model (SCREAM).  SCREAM is a global storm resolving model and is thus very expensive, which makes it difficult to analyze specific cloud regimes at the process level.  Therefore, we develop doubly periodic SCREAM (DP-SCREAM) to serve as our “efficient” configuration, which represents a standalone three-dimensional cloud resolving model with prescribed large scale forcing and boundary conditions periodic in both horizontal directions.  We detail the design characteristics of DP-SCREAM and demonstrate its computational efficiency and advantages, which includes the ability to explore the horizontal resolution sensitivity of SCREAM with ease.  The extensive case library of DP-SCREAM is also presented, which will include several recent ARM field campaigns such as GoAmazon, MAGIC, COMBLE, and LAFE.  We describe how we will use ARM observations to evaluate DP-SCREAM at the process level and how this will feedback to improvements in global SCREAM simulations.  Preliminary results for DP-SCREAM simulations and how they compare to observations for the aforementioned ARM cases will also be presented.

Lead PI

Yunyan Zhang — Lawrence Livermore National Laboratory