1: \begin{abstract} 

2: Let $\mathcal{P}_\lambda:= \mathcal{P}_{\lambda\kappa}$ denote a Poisson point

3: process of

4: intensity $\lambda\kappa$ on $[0,1]^d, d \geq2$, with $\kappa$ a bounded

5: density on $[0,1]^d$ and $\lambda\in(0, \infty)$. Given a closed subset

6: $\mathcal{M}\subset[0,1]^d$ of Hausdorff dimension $(d-1)$, we consider

7: general statistics $\sum_{x \in\mathcal{P}_\lambda} \xi(x, \mathcal{P}_\lambda, \mathcal{M})$, where the score

8: function $\xi$ vanishes unless the input $x$ is

9: close to $\mathcal{M}$ and where $\xi$ satisfies a weak spatial dependency

10: condition. We give a rate of normal convergence for the rescaled statistics

11: $\sum_{x \in\mathcal{P}_\lambda} \xi(\lambda^{1/d}x, \lambda^{1/d} \mathcal{P}_\lambda, \lambda

12: ^{1/d} \mathcal{M})$ as $\lambda\to\infty$. When $\mathcal{M}$ is of class $C^2$, we

13: obtain weak laws of

14: large numbers and variance asymptotics for these statistics, showing

15: that growth

16: is surface order, that is, of order $\mathrm{Vol}(\lambda^{1/d} \mathcal{M})$. We use the

17: general results to deduce variance

18: asymptotics and central limit theorems for statistics arising in

19: stochastic geometry, including

20: Poisson--Voronoi volume and surface area estimators, answering

21: questions in

22: Heveling and Reitzner [\textit{Ann. Appl. Probab.} \textbf{19}

23: (2009) 719--736] and

24: Reitzner, Spodarev and Zaporozhets

25: [\textit{Adv. in Appl. Probab.} \textbf{44} (2012) 938--953].

26: The general results also yield the limit theory for the number of

27: maximal points in a sample.

28: \end{abstract}