Fluctuations in plant populations: role of exogenous and endogenous factors




What is the role of exogenous and endogenous factors in determining fluctuations in plant populations?


A total of 29 long-term time series from different plant species was analysed. Data were obtained from the literature as well as from our own surveys. For each series we calculated the autocorrelation function to identify the presence of cyclic dynamics. The existence and the order of density-dependent processes were examined by calculating the partial rate correlation function to obtain the global Lyapunov exponent to assess whether populations exhibited chaotic dynamics. To test the effect of exogenous factors on each population, we used regression analyses to relate mean annual temperature and total annual precipitation to population abundance. Finally, we employed a Mann-Whitney U-test to analyse whether populations that showed endogenous regulation occurred in less variable environments.


We found no evidence for periodicity in the long-term time series analysed. Endogenous factors appeared to contribute to abundance fluctuations in eight plant populations, with three showing evidence of direct density-dependent regulation (first-order negative feedback) and five showing evidence for delayed density dependence (second-order negative feedback). The majority of populations were characterized by stable dynamics, although in five cases a positive global Lyapunov exponent suggested the possibility of chaotic dynamics. The exogenous factors that were considered contributed to explaining the temporal dynamics for four populations. Populations that showed evidence for endogenous regulation were not more frequently found in less variable environments.


In this study the effects of direct and delayed density dependence were negative in eight of 29 populations, suggesting intra-specific competition, delayed competition or inhibitory non-resource-based effects as the regulating mechanisms. The climate variables considered helped to explain the temporal variation in the abundance of only four of 29 populations, and neither density dependence nor environmental control were detected for 19 of the 29 populations. Inability to detect exogenous regulation may be because key environmental drivers were not included in the analysis. Regarding endogenous processes, longer records could increase the probability of detection.