Predictability, bred vectors, and generation of ensembles in space-time chaotic systems
Physical Review E—Statistical, Nonlinear, and Soft Matter Physics, 2005•APS
Spatial configuration of initial errors strongly affects predictability of space-time chaotic
systems. The predictability of numerical models can be adjusted by using prepared
ensembles of initial conditions. We present a natural way of preparing ensembles based in
using finite-amplitude perturbations with varying correlation. This allows one to take into
account the underlying dynamics to generate initial perturbations with spatial correlations
varying from fully correlated (bred vectors) to random fluctuations.
systems. The predictability of numerical models can be adjusted by using prepared
ensembles of initial conditions. We present a natural way of preparing ensembles based in
using finite-amplitude perturbations with varying correlation. This allows one to take into
account the underlying dynamics to generate initial perturbations with spatial correlations
varying from fully correlated (bred vectors) to random fluctuations.
Spatial configuration of initial errors strongly affects predictability of space-time chaotic systems. The predictability of numerical models can be adjusted by using prepared ensembles of initial conditions. We present a natural way of preparing ensembles based in using finite-amplitude perturbations with varying correlation. This allows one to take into account the underlying dynamics to generate initial perturbations with spatial correlations varying from fully correlated (bred vectors) to random fluctuations.
American Physical Society以上显示的是最相近的搜索结果。 查看全部搜索结果