Statistical predictability in two-dimensional turbulence

The predictability problem in the inverse energy cascade of two-dimensional turbulence is addressed by means of high resolution direct numerical simulations. The analysis is done in terms of the finite size Lyapunov exponent (FSLE) which is a measure of the growth rate at fixed error level. The numerical results are compared with classical closure predictions and good quantitative agreement is found. Finally, it is shown that the inertial range predictability properties are not affected by the presence of noise induced by the small, not resolved, scales.