Constraints on treeline advance in a warming climate: a test of the reproduction limitation hypothesis

Journal of Plant Ecology VOLUME 11, NUMBER 3, PAGES 411–422 June 2018 doi: 10.1093/jpe/rtx009 Advance Access publication 14 February 2017 available online at academic.oup.com/jpe Constraints on treeline advance in a warming climate: a test of the reproduction limitation hypothesis Dasvinder Kambo1,* and Ryan K. Danby1,2 1 Department of Geography, Queen’s University, 99 University Avenue, Kingston, Ontario K7L 3N6, Canada School of Environmental Studies, Queen’s University, 99 University Avenue, Kingston, Ontario K7L 3N6, Canada *Correspondence address: Department of Geography, Queen’s University, 99 University Avenue, Kingston, Ontario K7L 3N6, Canada. Tel:+1-613-572-3530; E-mail: 2 Aims Several mechanistic hypotheses have been developed to explain the existence of alpine treelines worldwide. The reproduction limitation hypothesis (RLH) postulates that reproductive processes such as seed production and dispersal are restricted in treeline environments, thereby limiting the establishment of new individuals in advance of extant treelines. Despite its popularity, no study has tested this hypothesis in a comprehensive fashion. In this experiment, we attempted to answer whether there are enough viable seeds being dispersed beyond treeline for sexually mediated treeline advance, and what the implications of climate change might be on these processes. Methods We established 30 plots across two aspects (north vs. south) and three elevational habitats (forest, treeline and tundra) in a white spruce (Picea glauca) boreal forest-alpine tundra ecotone in southwest Yukon, Canada. In each plot, tree characteristics, seed production and predispersal damage were measured. Additionally, eight dispersal trays were positioned in each plot to measure seed rain, and germination trials with and without predation exclosures were constructed in a subset of plots to quantify dispersal and germination success. Important Findings Results were highly variable both temporally and spatially. In 2014, a mast year, 69% of adult trees produced cones compared to 0.4% in the following year. Higher density of trees in forest plots INTRODUCTION Global temperatures have been rising since the 1850s, culminating in a 0.85°C global average rise between 1880 and 2012 (IPCC 2013). As temperatures continue to increase, thermal influence will continue to impact global ecological compared to treeline and tundra resulted in greater seed production at lower elevations. Across all plots, 88% of seeds were damaged before dispersal or were not viable. Treeline plots had significantly greater predispersal damage. Seed rain was greater in south-facing plots than north-facing plots. Less than 2% of seeds produced on the landscape were dispersed into Tundra plots, located 50 m above treeline. There was a net movement of seeds from the north-facing slope to the south at our study site, likely due to prevailing winds during the dispersal period. Germination counts were more than double on north-facing slopes and one-third higher inside exclosures. Cumulatively, the results provide some evidence for the RLH. Collectively, the high amount of predispersal damage and nonviable seeds, variability associated with dispersal and significant seed predation can functionally influence treeline dynamics. These findings suggest that global treeline distribution models, which rely largely on temperature, may not be entirely accurate for predicting treeline advance—at least at finer temporal scales. Many stochastic factors need to align temporally for successful advance, which is likely to result in a lag of many decades between the period of temperature amelioration and an increased number of trees beyond extant treelines. Keywords: altitudinal gradients, treeline dynamics, predispersal seed damage, seed predation, seed masts, ecotones Received: 21 July 2016, Revised: 7 February 2017, Accepted: 10 February 2017 communities and biome boundaries (ACIA 2004; Chapin et al. 1995; Kaplan and New 2006; Sturm et al. 2001). The single largest biome boundary that is predicted to be impacted is between the boreal forest and tundra, an ecotone referred to as treeline. Position of the treeline ecotone is largely correlated with a growing season mean root-zone temperature of © The Author(s) 2017. Published by Oxford University Press on behalf of the Institute of Botany, Chinese Academy of Sciences and the Botanical Society of China. All rights reserved. For permissions, please email: Abstract 412 Predispersal damage to seeds, including underdeveloped seeds and insect predation, can act as a reproductive constraint to treeline advance by limiting the amount of viable seeds (Jameson et al. 2015; Sirois et al. 1999; Sirois 2000; Wilmking et al. 2012). However, despite its potential importance, few studies have quantified predispersal damage at treeline and its constraints on advance are not well understood (Payette et al. 1982). The third reproductive constraint is dispersal. Limited seed dispersal from source populations can, when combined with other factors, lead to stagnant conifer treelines (DufourTremblay et al. 2012). In long-term models, dispersal limitation has been shown to significantly decrease rate of potential treeline advance (Dullinger et al. 2004). In some studies, suboptimal dispersal has been shown to limit treeline advance to as little as 30 m over 1000 years, despite ideal climate conditions (Slatyer and Noble 1992). Conversely, some alpine sites have demonstrated long distance spruce seed dispersal greater than 300 m elevation (Piotti et al., 2009), suggesting a high degree of local and regional differences. The fourth seed-related mechanism associated with the RLH is predation. Seed predation can negatively influence treeline by decreasing the number of viable seeds on the landscape available for germination and subsequent recruitment. It has also been shown to damage germinants, potentially leading to changes in plant community composition at treeline (Herrero et al. 2011; Moen et al. 2008). At some treeline sites, predation has been shown to account for the removal of 10–30% of tree seeds on the landscape (Munier et al. 2010; Wheeler et al. 2011). Each of the four reproductive constraints influencing the number of seeds available for germination have been studied individually in various treeline systems, but there has been no study to-date that has examined all of these factors at one site. This is clearly insufficient, as the relative seed losses attributable to each factor can only be identified when all factors are studied simultaneously. This study aimed to test the treeline RLH at a white spruce treeline in Canada’s Yukon Territory by quantifying seed losses from production through to germination, while also measuring dispersal, all in one location. The study was guided by one central research question: Are there ‘enough’ viable seeds being dispersed beyond treeline for sexually mediated advance? To answer th (...truncated)