Towards the construction of a standard adjoint GEOS-Chem model

  • Authors:
  • Kumaresh Singh;Paul Eller;Adrian Sandu;Daven Henze;Kevin Bowman;Monika Kopacz;Meemong Lee

  • Affiliations:
  • Virginia Tech, Blacksburg, VA;Virginia Tech, Blacksburg, VA;Virginia Tech, Blacksburg, VA;Columbia University, New York, NY;NASA Jet Propulsion Laboratory, Pasadena, CA;Harvard University, Cambridge, MA;NASA Jet Propulsion Laboratory, Pasadena, CA

  • Venue:
  • SpringSim '09 Proceedings of the 2009 Spring Simulation Multiconference
  • Year:
  • 2009

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper we present the construction of a standardized adjoint of the global GEOS-Chem model. Adjoint models are powerful tools for estimating the sensitivity of (a function of) the model output with respect to a large number of inputs. The construction of adjoint models is a difficult, labor intensive, and error prone task. The adjoint of gas phase chemistry is generated using the Kinetic PreProcessor (KPP) and parsers to automatically interface KPP with GEOS-Chem are constructed. The adjoints of other individual science processes are generated manually and using automatic differentiation tools, and are validated extensively against finite-differences. The standardized adjoint framework discussed here will allow the world-wide community of GEOS-Chem users to perform sensitivity analyses and 4-D variational data assimilation for initial tracer concentrations, boundary conditions, and emission sources.