Leaves on the line and in the plane (Q2184619)

The dead leaves model (DLM for short) provides a natural way of generating a stationary random tessellation of the plane with non-convex cells having possibly curved boundaries, see [\textit{C. Bordenave} et al., Adv. Appl. Probab. 38, No. 1, 31--46 (2006; Zbl 1095.60004)]. A collection of leaves arrives as an independently marked homogeneous Poisson point process \(\mathcal{P} = \sum_{i=1}^\infty \delta_{(x_i, t_i)}\) in \(\mathbb{R}^d\times\mathbb{R}\) of unit intensity with marks \((S_i)_{i\ge 1}\) taking values in \(\mathcal C\) (\(\mathcal C\) is the space of compact subsets of \(\mathbb R^d\)) with common mark distribution \(Q\). Each point \((x_i, t_i, S_i)\) of this marked point process is said to have arrival time \(t_i\) and the associated leaf covers the region \(S_i+x_i\subset\mathbb R^d\) from that time onwards. At a specified time, say time 0, and at spatial location \(x\in\mathbb R^d\), the most recent leaf to arrive before time 0, that covers location \(x\) is said to be visible at \(x\). For each \(i\in\mathbb N\), the visible portion of leaf \(i\) at time 0 is the set of sites \(x\in\mathbb R^d\) such that leaf \(i\) is visible at \(x\) at time 0. The connected components of visible portions of leaves (at time 0) form a tessellation of \(\mathbb R^d\), which is called the DLM tessellation. In this paper, the DLM for \(d=1\) (Section 2) and for \(d=2\) (Section 3) are considered. In Section 2, the second order theory for the point process of cell boundaries and a spatial central limit theorem (CLT) are established. In Section 3, a similar program for the surface measure of cell boundaries within a large window is considered. In Section 4, for general \(d\ge 3\) an extension of the DLM which the author calls the dead leaves random measure (DLRM) is considered. In Sections 2--5 all the results are stated, and in Sections 6--8 they are proved.