A representation for super Brownian motion (Q1336969)

Let \((X_ t)\) be a super-Brownian motion (i.e., a strong Markov process with continuous paths taking its values in the space \(M_ F\) of finite measures on \(R^ d\) with the topology of weak convergence) such that, for some \(m \in M_ F\), \(X_ 0 = m\) \(P^ m\)-a.s. Let \(S_ t\) denote the closed support of \(X_ t\). It is known that at a fixed time \(t>0\) the measure \(X_ t\) can be recovered from \(S_ t\). The main result is the following alternative representation of super-Brownian motion in terms of its support process \((S_ t)\) which is analogous to a representation for Brownian local time due to \textit{J. F. C. Kingman} [J. Lond. Math. Soc., II. Ser. 6, 725-731 (1973; Zbl 0283.60080)]: Let \(A \subset R^ d\) \((d \geq 3)\) be bounded. Fix \(t>0\) and \(m \in M_ F\). Then there exists an absolute constant \(c_ 0\) (only depending on \(d)\) such that \[ \lim_{\varepsilon \to 0} \varepsilon^{2-d} | S^ \varepsilon_ t \cap A | = c_ 0 X_ t (A) \text{ a.s. and in } L^ 2. \] Here, \(| B |\) denotes the Lebesgue measure of a set \(B \subset R^ d\), and \(S^ \varepsilon_ t\) is the set \(\{x : d(x,S_ t) \leq \varepsilon\}\) where \(d(x,S_ t) = \inf \{| x-y | : y \in S_ t\}\).