Individual asynchrony promotes population-level tree growth stability

Abstract

1 INTRODUCTION

A key challenge in ecology is to understand how the complexities of a biological system could stabilize the system's performance, such as the annual productivity of grasslands or forest communities (Jucker et al., 2014; Schindler et al., 2010; Yachi & Loreau, 1999). Although many studies have investigated this topic over the past decades, much of the research has been focused on the effect of biodiversity on community stability (Hector et al., 2010; Tilman, 1996; Walter et al., 2021), with little attention paid to the mechanisms of how individual-level differences affect population-level stability (Waddle et al., 2019). Filling this gap is important because it is the variation at the individual level that defines the variation at the population level, and subsequently, the higher ecological levels.

The stability of ecological communities is considered primarily determined by two major mechanisms: the portfolio effect (i.e. statistical averaging) (Doak et al., 1998; Tilman, 1996), where stability increases with the number of species; and the insurance effect, where stability increases with the temporal asynchrony among species (species asynchrony) (Blüthgen et al., 2016; Yachi & Loreau, 1999). One may draw a parallel between community stability and population stability, by which population size is analogous to species richness, and within-population asynchrony among individuals is analogous to species asynchrony. Take tree growth for example, as widely understood, it is strongly regulated by climatic conditions, especially the water–energy balance (Peltier & Ogle, 2020). The fluctuation in climate would thus inevitably cause variation in growth rate. In addition to climate, many other factors such as tree age, genetic and trait variations, and microhabitat conditions could all cause differences in growth rate within a population, thus leading to growth asynchrony of conspecifics (Cater & Chapin III, 2000; Peltier & Ogle, 2020; Takenaka, 2000; Tejedor et al., 2020). Such individual differences allow the growth rates of the faster-growing individuals to compensate those of the slower-growing individuals when averaged across the population, thereby promoting the stability in the population-level mean tree growth rate. This stabilization process has potential ecological significance in mitigating the negative impacts of global change on forest ecosystem. However, little is understood about the extent to which population size and within-population asynchrony can regulate the stability of population-level tree growth rate.

A subtle but important distinction between stabilizing processes at population and community levels lies in the different magnitudes of population size and species richness. In most communities (with exceptions for some extremely diverse systems, such as lowland tropical rainforests), species richness within a specific taxonomic group, such as trees or herbaceous plants, falls within a few dozen (Hector et al., 2010; Jucker et al., 2014), while a tree population can easily consist of hundreds of individuals. It has been observed that community-level stability of grass or wood growth rate generally levels off when species richness reaches around 20–30 (Jucker et al., 2014; Tilman, 1996; Walter et al., 2021). An interesting but unanswered question is how many trees are required to stabilize the average biomass growth of a population. Will this population-level tree growth also stabilize at a similar level of 20–30 trees? If this is the case, it means that the importance of portfolio effect in stabilizing population may diminish when this population size is reached while the effect of within-population asynchrony could still be strong.

Tree growth asynchrony can be affected by many factors and is not evenly distributed across different biomes (Defriez & Reuman, 2017; Shestakova et al., 2016; Tejedor et al., 2020; Walter et al., 2017). For example, Tejedor et al. (2020) shows that at the global scale, the within-population asynchrony in tree growth rate is positively correlated with mean annual temperature and precipitation, with tropics having the highest growth asynchrony and arid regions the lowest asynchrony. Such spatial heterogeneity in asynchrony could affect the global variation of tree growth stability. Understanding this process will help us comprehend the roles of tree growth asynchrony in mitigating the negative effects of climate change on forest ecosystem stability in different climate zones.

Using a global set of tree-ring data comprising 2133 populations, we examined the effects of the individual tree growth differences on population-level tree growth stability. Specifically, we (1) tested the population-level portfolio effect by quantifying the relationship between the population-level tree growth stability (STB_P) and population size (N); (2) tested the insurance effect by quantifying the relationship between population-level tree growth stability and the within-population tree growth asynchrony (WPA); (3) quantified the contributions of these two effects to population-level tree growth stability relative to the individual-level tree growth stability (STB_I); and (4) quantified the global variation of within-population tree growth asynchrony, and explored how it affected the variation of population-level tree growth stability across different climatic zones.