Lower bound on the radius of analyticity of solution for fifth order KdV-BBM equation (Q2065594)

The authors are interested in the fifth order KdV-BBM type equation of the form \[ \begin{cases} \partial_t\eta+\partial_x\eta-\gamma_1\partial_t\partial_x^2\eta+\gamma_2\partial_x^3\eta+\delta_1\partial_t\partial_x^4\eta+\delta_2\partial_x^5\eta\\ =-\dfrac{3}{4}\partial_x(\eta^2)-\gamma\partial_x^3(\eta^2)+\dfrac{7}{48}\partial_x(\eta_x^2)+\dfrac{1}{8}\partial_x(\eta^3)\\ \eta(x,0)=\eta_0(x)\tag{1} \end{cases} \] where \(\eta:\mathbb R^2\longrightarrow\mathbb R\) is the unknown function, the parameters \(\gamma_1\), \(\gamma_2\), \(\delta_1\), \(\delta_2\) and \(\gamma\) are constants, and where the initial condition is analytic. This equation, introduced by \textit{J. L. Bona} et al. [J. Nonlinear Sci. 28, No. 2, 543--577 (2018; Zbl 1384.35111)], describes the unidirectional propagation of water waves. In a recent result, \textit{X. Carvajal} and \textit{M. M. Panthee} [``On the radius of analyticity for the solution of the fifth order KdV-BBM model'', Preprint, \url{arXiv:2009.09328}] establish that the uniform radius of spatial analyticity \(\sigma(t)\) of the solution in time \(t\) of Eq. (1) has an exponential decay for large \(t>0\). In the present paper, the authors significantly improve this bound by proving that the radius cannot decay faster than \(1/t\) for large \(t>0\).