Essential closures (Q1704490)

The aim of the paper is to propose a set of postulates for general essential closures and to develop a theory of essential closures. Let \((X,\tau)\) be a topological space equipped with an algebra \(\Omega\) over \(X\). The authors call an operator \(\mathcal{E}:\Omega\rightarrow\Omega\) an essential closure if \(\mathcal{E}\) is idempotent, \(\mathcal{E}(A)\) is a closed subset of the topological closure \(\overline{A}\) and \(\mathcal{E}(A\cup B)=\mathcal{E}(A)\cup\mathcal{E}(B)\) for \(A, B\in\Omega\). A typical example of an essential closure is a submeasure closure: If \(\mu\) is a \(\sigma\)-subadditive submeasure on a \(\sigma\)-algebra \(\Omega\) containing \(\tau\), then, for \(A\in\Omega\), the \(\mu\)-closure \(\overline{A}^\mu\) is defined as the set of all \(x\in X\) such that \(\mu(G\cap A)>0\) for all open neighbourhoods of \(x\). It is shown that \(A\mapsto \overline{A}^\mu\) is an essential closure if \((X,\tau)\) is either hereditarily Lindelöf or \(\mu\) is an inner regular measure. It is further shown that a ``nice'' essential closure must be a submeasure closure. Many examples of known and new submeasure closures are discussed.