Determination of Mesoscale Predictability Limits with Respect to Uncertainty in the Larger-Scale Environment
Abstract
LONG-TERM GOALS. Our goal is to determine the limits of predictability inherent in atmospheric forecasts due to uncertainty in their initial conditions. These uncertainties are a consequence of inaccuracies of observational data and the algorithms that produce 3-dimensional analysis. Even if a model can simulate atmospheric behavior perfectly, since atmospheric flows exhibit instabilities leading to chaos, any errors in a forecast s initial condition will tend to grow, until information content of the forecast is negligible. The result is a limit to predictability. This predictability limit has been known for some time (Lorenz, 1963), although it continues to be ignored by some who make very optimistic claims (see Errico and Baumhefner, 1987 for some examples). Its character, especially regarding how various types of errors influence the predictive skill of various fields on various scales, has only been superficially explored to date (Lorenz, 1969; Errico et al., 1995). Since characterization of this limit has crucial implications regarding forecast reliability and possible observation system impacts, its determination is critical (Tribbia and Baumhefner, 1988).
Document Details
- Document Type
- Technical Report
- Publication Date
- Sep 30, 1999
- Accession Number
- ADA636464
Entities
People
- David Baumhefner
- Joseph Tribbia
- Ronald M. Errico
Organizations
- National Center for Atmospheric Research