Documentation of Scalable Incremental MM5v3 4D-VAR System

Abstract

This report documents the changes made to the Penn State University/National Center for Atmospheric Research third revision of the Fifth Generation Mesoscale Model (MM5v3) four dimensional variational (4D-Var) analysis system to allow the execution of incremental driver for the system. The incremental driver allows the user to select resolution or choice of physics that are used in the minimization of the cost function that are different than those used to describe the basic state. In practice, this would mean either a coarser resolution model grid or less computationally intensive physics routines would be chosen for the cost function minimization while a finer resolution model grid or more computationally intensive phyiscs for the basic state. The effect would be to significantly reduce the computations needed to complete the analysis. This report presents examples of the use of the incremental driver for the Hurricane Isaac case and compares the results to the original driver that uses consistent resolution and physics for the minimization and basic state computations. The results show that with the use of less computationally complex physics, the minimization does not reduce the accuracy of the analysis or time to convergence. Using coarser resolution in the minimization does not reduce the accuracy of the analysis but does appear to increase the time to convergence.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Apr 14, 2004
Accession Number
ADA425240

Entities

People

  • D. X. Zhang
  • Xiaolei Zou

Organizations

  • Florida State University

Tags

DTIC Thesaurus Topics

  • Accuracy
  • Air Force
  • Air Force Research Laboratories
  • Assimilation
  • Boundary Layer
  • Computer Programs
  • Convergence
  • Directories
  • Four Dimensional
  • Hurricanes
  • Physics
  • Procedures (Computers)
  • Sea Level
  • Shell Scripts
  • Trajectories
  • Universities
  • Weather Forecasting

Readers

  • Computational Fluid Dynamics (CFD)
  • Finite Element Method (FEM) for solving Partial Differential Equations (PDEs)
  • Ocean-Atmosphere Mesoscale Modeling, Data Assimilation, and Flux Boundary Layers