The solar tachocline — the thin shear layer between the Sun's differentially rotating convection zone and its rigidly rotating radiative interior — is treated as a structural constant. Models of solar dynamics place it at a fixed depth and use it as a boundary condition. It is the floor on which convection rests.
Thirty years of helioseismic data show the floor is moving.
Basu, Korzennik, and Tripathy find that at low latitudes, the tachocline has been steadily migrating inward, deeper toward the radiative zone, over three decades of observation. The shift is small in absolute terms but persistent and statistically significant. The boundary that solar physics treats as given is drifting.
The mechanism connects to an equally surprising finding about the tachocline's width: it is larger during periods of lower solar activity. Magnetic fields confine the tachocline. When the solar cycle weakens, the confining field weakens, and the shear layer broadens. The tachocline's geometry responds to the Sun's magnetic activity — it is not a structural feature but a magnetically maintained one.
This inverts the usual causal picture. Solar activity models typically compute the magnetic field from the tachocline's properties: the shear drives a dynamo that generates the field. But the field also maintains the tachocline's structure. The boundary condition and the dynamics it enables are coupled. The tachocline is not a fixed platform supporting the dynamo; it is a product of the dynamo it supports.
The inward migration adds a slow secular trend to this coupled system. Over decades, the shear layer creeps deeper. Whether this is part of a longer cycle, a response to the observed weakening of recent solar cycles, or a sign of a genuinely evolving interior structure is unclear. What is clear is that the “fixed” boundary was never fixed. It was held in place by a field that varies, and when the field varies, the boundary moves.
Thirty years of data was needed to see it. Not because the effect is small — it accumulates — but because helioseismology needed three decades of continuous observation to separate the trend from the cycle.