The classical Navier-Stokes equations describe how ordinary viscous fluids flow, but many real materials like biological fluids, polymer solutions, and certain industrial liquids have a more complex "memory" - their resistance to flow depends not just on current conditions but on their entire deformation history. This paper studies a modified version of the Navier-Stokes equations where the viscosity carries a dual-scale hereditary memory, meaning the fluid remembers its past behavior across two different characteristic timescales. This makes the equations significantly harder to analyze because the dissipation mechanism no longer respects the clean scaling symmetries that mathematicians traditionally exploit to understand fluid equations.
The paper establishes two contrasting results depending on the regularity of the initial data. When the initial fluid velocity is measured in a relatively weak function space (the "supercritical" regime), the equations are ill-posed, meaning small changes in initial conditions can produce wildly different solutions instantaneously. The authors prove this by carefully tracking how energy is distributed across different frequencies and showing that certain frequency interactions cause the solution's size to blow up immediately. This kind of instability is a fundamental obstruction to any reasonable solution theory in those spaces.
On the positive side, the paper shows that the memory structure of the fluid actually helps in the most extreme critical function space, where classical Navier-Stokes equations are known to be ill-posed. The temporal smoothing introduced by the dual-scale memory suppresses the dangerous nonlinear interactions between high-frequency components that normally cause trouble. As a result, the authors prove that for small initial data in a slightly enlarged critical space, the equations have global solutions that depend continuously on initial conditions. This is a genuinely broader well-posedness result than what is achievable for ordinary Newtonian fluids, demonstrating that hereditary memory can be a stabilizing rather than complicating feature.