Large-Data Global Regularity for Three-Dimensional Navier--Stokes I: A Direct First-Threshold Continuation Proof for the Axisymmetric Swirl Class
Rishad Shahmurov

TL;DR
This paper establishes a novel direct continuation theorem for the axisymmetric Navier-Stokes equations with swirl, using lifted variables and a finite-threshold approach to prove global regularity for large data.
Contribution
It introduces a new proof technique employing lifted variables and a finite-threshold stopping time to demonstrate large-data regularity in the axisymmetric swirl class.
Findings
Proves a first-threshold continuation theorem for axisymmetric Navier-Stokes with swirl.
Develops a quantitative framework with a critical axis score envelope and local balance estimates.
Shows that no finite threshold leads to singularity, implying global regularity for the class.
Abstract
This is the first paper in a two-part direct-threshold series on large-data global regularity for the three-dimensional Navier--Stokes equations. We prove a direct first-threshold continuation theorem for the axisymmetric class with swirl. The proof is written entirely in the lifted variables \[ \Gamma=ru_\theta,\qquad G=\omega_\theta/r,\qquad d\mu_5=r^3\,dr\,dz, \] and uses the five-dimensional full-Dirichlet visibility \(\mathcal V_\chi\) as the local coercive quantity. The argument is organized by a finite first-threshold stopping time. We define a critical axis score envelope, follow it to a first possible threshold time, and prove that the corresponding normalized packet cannot exist. The proof has three quantitative ingredients. First, a small-envelope continuation theorem converts bounded score and regularized source size into smooth continuation. Second, a finite-overlap…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
