\(L^p\) estimates for \(\overline{\partial}\) in some weakly pseudoconvex domains in \(\mathbb{C}^n\) (Q1591412)

Let \(D\) be a connected pseudoconvex domain defined by \(\{ \rho<0 \}\cap \{ \sigma <0 \}\), where \(\{ \rho <0 \}\) and \(\{ \sigma <0 \}\) are strictly pseudoconvex domains in \(\mathbb C^n \) and \(\mathbb C^m\), \(1\leq m \leq n\). The authors prove the following result: there exists a linear operator \(\mathcal T\) and a constant \(C_D\) such that \(\mathcal T f \) solves the equation \(\overline \partial u = f \) provided that \(f\in L_{(0,q)}^p(D)\), \(1\leq p \leq \infty\), \(1\leq q \leq n\), is \(\overline \partial \)-closed in the sense of distributions; \(\mathcal T\) maps \(L_{(0,q)}^p(D)\) to \(L_{(0,q-1)}^p(D)\), \(1\leq p \leq \infty\), \(1\leq q \leq n\), with \(\|\mathcal T f \|_p \leq C_D \|f \|_p\), where \(C_D\) depends only on \(D\) and is independent of \(p ;\) moreover, the last inequality is sharp, if \(m<n ;\) if, in addition, the coefficients of \(f\) are in \(\mathcal C (\overline D)\), then so are those of \(\mathcal T f\). As an application of this result they also prove a theorem on approximating holomorphic \(L^p\) functions by functions holomorphic in a neighborhood of \(\overline D \) in \(L^p\) norm.