Statistical Atmospheric Predictability at Varying Time Scales

New York University, New York NY

Investigators

Abstract

Traditionally the aspect of predictability in atmospheric research has been concerned with its dependence on initial conditions for weather prediction and on boundary conditions for prediction. The investigators will study the intermediate range where important areas of predictability may be found at weekly timescales. Two numerical models that are sufficiently complex but computationally efficient will be used to explore statistical prediction. The influence of initial and boundary conditions, ensemble size and the optimal temporal averaging strategies will be studied. Being able to make a reasonable a-priori estimate of an ensemble forecast's skill and the sources for that skill would be of considerable interest in the forecasting community. The project will involve the participation of a graduate student.