Ecology, 94(8), 2013, pp. 1718–1728 Ó 2013 by the Ecological Society of America Landscape-scale density-dependent recruitment of oaks in planted forests: More is not always better EFRAT SHEFFER, 1,4 CHARLES D. CANHAM, 2 JAIME KIGEL, 1 AND AVI PEREVOLOTSKY 3 1 Robert H. Smith Institute of Plant Sciences and Genetics in Agriculture, Robert H. Smith Faculty of Agriculture, Food and Environment, The Hebrew University of Jerusalem, Rehovot Campus, P.O. Box 12, Rehovot 76100 Israel 2 Cary Institute of Ecosystem Studies, Box AB, Millbrook, New York 12545 USA 3 Department of Agronomy and Natural Resources, Institute of Plant Sciences, Agricultural Research Organization, the Volcani Center, P.O. Box 6, Bet Dagan 50250 Israel Abstract. Plant colonization studies usually address density-dependent processes in the narrow sense of recruitment constraints due to negative density-dependent seed and seedling mortality. However, complex density-dependent effects may be involved in additional stages of the recruitment process. We hypothesized that seed arrival and seedling establishment are influenced by density dependence acting at small scales at the site of colonization, and at larger scales as a function of the colonizing species’ landscape abundance. These hypotheses were tested in a study of colonization of pine forests by oaks in a heterogeneous Mediterranean landscape. Maximum-likelihood models show that density effects switch from positive to negative along the range of landscape-scale oak seed source abundance. Contrary to expectations, high seed source densities limited oak recruitment, suggesting a landscape-scale Janzen-Connell effect. We propose a range of mechanisms that generate positive or negative density dependence during colonization, resulting in nonlinear density-dependent feedbacks that can generate unexpected colonization patterns. Key words: animal-mediated dispersal; colonization; density dependence; dispersal limitation; Janzen- Connell; maximum likelihood; Mediterranean; Pinus halepensis; plantations; Quercus calliprinos; recruitment; seed predation. INTRODUCTION The process of colonization of a site by a plant species can be regulated by factors acting at a number of discrete stages, including propagule dispersal, germina- tion, seedling establishment, survival, and growth to reproductive maturity (Schupp 1995, Clark et al. 1998, Nathan and Casagrandi 2004, Zavala et al. 2011). Conceptually, colonization can be studied from the perspective of processes that determine both the colonization pressure, i.e., the number of propagules that arrive at a site, and the resistance to establishment at the site following propagule arrival. While density dependence can be involved in any of these stages, and involve processes acting at a wide range of spatial scales, most research has focused on negative density depen- dence during seedling establishment, particularly at highly localized spatial scales (e.g., Johnson et al. 2012). Density dependence, however, is not inherently a negative or a local-scale process. Here we propose a more comprehensive approach to examine how plant recruitment (seed arrival and seedling establishment) is influenced by the density of a species, both as potential seed sources in the surrounding landscape and as seeds and juveniles establishing in a new site. We argue that some recruitment processes are unaffected by the density of the colonizing species (neutral effect; Fig. 1A), while other processes can have either a positive (Fig. 1B) or a negative (Fig. 1C) response, or can even shift from positive to negative, depending on the absolute magni- tude of the species density (shifting density dependence, Fig. 1D). Moreover, these positive and negative density- dependent controls can change colonization patterns, sometimes in non-intuitive ways (Fig. 1B, C). Next we propose mechanisms that may lead to different density- dependent dynamics and present a study that supports some of these. Our approach examines density dependence in the two main components of the colonization process, seed arrival and seedling establishment, at two spatial scales: the abundance of seed sources in the landscape surrounding a site, and the density of seeds and newly established seedlings within a site. Seed arrival to a site is determined by the spatial distribution of seed sources, the density of seed sources at the landscape scale (e.g., proportion of land cover by stands of a given species), and dispersal patterns. We propose that the resulting colonization pressure is not a simple linear, additive function of the abundance of seed sources in the Manuscript received 6 December 2012; revised 19 February 2013; accepted 25 February 2013. Corresponding Editor: M. Fortin. 4 Present address: M31 Guyot Hall, Department of Ecology and Evolutionary Biology, Princeton University, Princeton, New Jersey 08544 USA. E-mail: eshefer@princeton.edu 1718