Abstract
We respond to the criticisms of a recent comment by Bizoń and Rostworowski [arXiv:1410.2631].
Introduction
The stability of Anti-de Sitter (AdS) space is a central question in general relativity. The "AdS instability conjecture" suggests that small perturbations in this "box-like" spacetime will eventually concentrate and collapse into a black hole. In this reply, we defend our findings regarding complex energy cascades and "stability islands" against criticisms that challenged our numerical precision and theoretical framework.
Numerical Robustness and Convergence
A primary criticism was that our numerical simulations failed to show collapse due to insufficient resolution. To address this, we performed higher-resolution simulations of two-mode initial data with an amplitude of $\epsilon = 0.09$.
As shown in the figure above, our code demonstrates strong convergence. The overlap of the curves for varying grid spacings ($\Delta$) indicates that the results are mathematically sound and not artifacts of numerical error. Notably, while critics reported collapse at $t \approx 1080$, our high-resolution runs do not observe black hole formation by $t = 1500$, suggesting the path to collapse is more complex and sensitive than previously assumed.
The Two-Timescale Framework (TTF)
We employ the Two-Timescale Framework (TTF) to study the slow evolution of mode amplitudes and phases. While the TTF has a finite "shelf life" (valid for times scaling as $1/\epsilon^2$), it remains highly effective at capturing the bulk energy dynamics and recurrence times. We acknowledge that truncating the number of modes in the TTF can lead to discrepancies at very high frequencies, but this does not invalidate the framework's ability to model the primary physical processes of the system.
Energy Cascades and Stability
Our research highlights that energy in AdS does not just move from low to high frequencies (a direct cascade leading to collapse); it also moves from high to low (an inverse cascade). This bidirectional energy transfer is a key feature of the "turbulent" dynamics in AdS. We also reaffirm the existence of "stability islands"—quasi-periodic solutions that represent extremely long-lived, if not absolutely stable, states. Even our critics now acknowledge the existence of these complex energy cascades.
Conclusion
We conclude that our primary findings—the existence of direct and inverse energy cascades and the validity of the TTF for intermediate timescales—remain robust. While the ultimate fate of all perturbations in AdS remains an open question, our work clarifies the complex dynamical path that precedes any potential collapse.
Summary updated