Analytical and Numerical Investigation
Published online by Cambridge University Press: 25 May 2016
In a recent paper, Morbidelli and Giorgilli (1995) proved the superexponential stability of invariant tori. As usual in the theory of dynamical systems, the results are rigorously proved assuming that the perturbation is small enough. The numerical experiments show, however, that invariant tori persist up to much larger perturbation magnitudes. Therefore, it is interesting to check numerically if the superexponential stability and the other properties outlined in Morbidelli and Giorgilli's theorem persist up to the value of the perturbation for which the torus actually breaks up. Moreover, one would like to have a numerical indication about the size of the super-exponentially stable region existing around a torus. Is the superexponential stability just an asymptotic result, or does it concern a macroscopic region of physical interest?