Canonical: https://knowlpedia.com/partial-differential-equations/finite-time-blowup/

# Finite-time blowup in a specified norm

Unbounded growth of a chosen solution norm as time approaches a finite endpoint.

Let \(u(t)\) be a solution for \(t<T_*<\infty\). It has **finite-time blowup in the norm \(X\)** if
\[
\limsup_{t\uparrow T_*}\|u(t)\|_X=\infty.
\]
The [one-sided lim sup](https://knowlpedia.com/real-analysis/limit-superior-at-an-endpoint/), norm, and finite endpoint are part of the assertion. Some authors reserve “blowup” for the stronger statement that the norm tends to infinity; the convention should be stated.

## Different norms measure different losses

Blowup of a derivative norm need not mean blowup of the velocity itself. Conversely, a bounded spatial \(L^2\) norm does not bound the \(L^\infty\) norm: increasingly narrow peaks may grow in height while retaining bounded \(L^2\) mass.

## Breakdown and extension

Failure to extend a classical solution is often called breakdown. A continuation theorem can relate breakdown to the divergence of a specified norm or space-time integral, but that implication is additional mathematical information.

## Explicit continuation obstructions

An [unbounded sequence approaching a finite space-time point](https://knowlpedia.com/real-analysis/unbounded-path-prevents-continuous-extension/) rules out a continuous extension there. A [Sobolev embedding](https://knowlpedia.com/partial-differential-equations/sobolev-continuation-obstruction/) can separately rule out continuous extension in a specified Sobolev norm.
