Conditional stability for backward parabolic operators with Osgood continuous coefficients (Q2309434)

Backward parabolic equations are known to generate ill-posed Cauchy problems. Nevertheless, even for problems which are not well posed, the study of the conditional stability of the solution -- the surrogate of the notion of continuous dependence when existence of a solution is not guaranteed -- is of some interest. This paper gives a contribution to the study of the conditional stability of the Cauchy problem associated with a backward parabolic operator. In particular, the considered operator \(L\) is defined on the strip \([0, T] \times \mathbb{R}^n\), by \[ Lu = \partial_t u + \sum _{i, j =1}^{n} \partial_{x_i} \left (a_{ij} (t) \partial_{x_j} u \right ) +\sum _{ j =1}^{n} b_j(t) \partial_{x_j} u + c(t) u \] where all the coefficients are bounded. It is supposed that \( a_{i , j} (t) = a_{ j ,i }(t) \) for all \( i , j = 1, \dots , n\) and for all \( t \in [0, T ]\). It is also supposed that \(L\) is backward parabolic. The authors show that, if the coefficients of the principal part of \(L\) are at least Osgood continuous (i.e., their modulus of continuity \(\omega(s)\) satisfies the condition \[ \int \frac{1}{\omega(s)}\, ds = +\infty, \] then there exists a function space in which the associated Cauchy problem \[ Lu = f\text{ in }(0,T) \times \mathbb{R}^n,\quad u|_{t=0} = u_0\text{ in }\mathbb{R}^n, \] is conditionally stable.