Modelling plant trait variability in changing arid environments
von Liubov Zakharova
Datum der mündl. Prüfung:2020-09-14
Betreuer:Prof. Dr. Kerstin Wiegand
Gutachter:Prof. Dr. Kerstin Wiegand
Gutachter:Prof. Dr. Holger Kreft
EnglischCommunities in arid environments are especially vulnerable to global change because they experience highly unpredictable environmental conditions. The fate of communities in an uncertain future may be elucidated by understanding the drivers of these communities. The interplay between community drivers may be unravelled by using approaches based on functional traits because traits describe plant strategies and the responses of communities to environmental changes. Furthermore, inter- and intraspecific trait variability provides the necessary cues to identify survival strategies of desert plants under fluctuating environmental conditions. However, studying desert plant communities is challenging due to the spatial and temporal heterogeneity of arid environments. Modelling approaches support and complement empirical trait-based approaches in exploring desert plant communities and their drivers and dynamics in changing arid environments. The overarching aim of this thesis was to explore intra- and inter-specific variability of functional traits in arid environments and to investigate how this variability affects the ability of plants to tolerate aridity stress and succeed in competition with their neighbours. To address this aim, I developed, implemented and analysed a spatially explicit individual- and trait-based simulation model, conducted a simulation experiment, analysed data from model simulations and empirical experiments and synthesized the literature on trait-based models and metamodelling approaches. My research was focused on annual plant communities dominated by the True Rose of Jericho (Anastatica hierochuntica L.) in the Negev desert in Israel. According to the review in chapter 1, trait-based models are a suitable method to predict changes in community patterns under global change and to understand the underlying mechanisms of community assembly and dynamics. Combining modelling and trait-based approaches overcomes technical challenges, scaling problems, and data scarcity. Specifically, a combination of trait-based approaches and individual-based modelling was recommended to simplify the parameterization of models and to capture plant-plant interactions at the individual level, and to explain community dynamics. In chapter 2, in line with the major claim of chapter 1, the spatially explicit trait- and individual-based ATID-model was developed, implemented and analysed to explore how community dynamics arise from plant traits and the interactions among plants and with their environment. The sensitivity analysis of the model highlighted plant functional traits as key drivers of community dynamics and indicated that environmental factors were less important in the model. The outlined traits included both those traits that are involved in plant-plant interactions, such as relative growth rate and maximum biomass, and those that promote tolerance to abiotic stress, such as dormancy and germination probability. Among the environmental factors, the most influential factors were soil water availability and precipitation. The special role of functional traits in the community dynamics of desert annual plants indicates the importance of trait-based strategies as an adaptation to the stressful arid environment. Chapter 3 addresses the results from a simulation experiment that was conducted in the ATID-model. This experiment explored the influence of functional traits involved in two survival strategies defined in the study as ‘protective-competition’ and ‘escape-colonization’ strategies on community dynamics. These strategies differed not only in seed size and the number of seeds, but also in the plant functional traits related to competition and survival, which were highlighted in the sensitivity analysis of the model from chapter 2. Merging the colonization-competition trade-off with escape in time and space into one strategy set provided a more realistic representation of species because the merged strategies related to the entire plant life cycle. To gain more understanding on empirical trait distributions, in chapter 4 data on intraspecific trait variability and trait spaces of the desert annual plant A. hierochutica from a nethouse experiment were analysed. High salinity had significant effects on the average values of plant functional traits. Additionally, salinity stress affected the intraspecific trait spaces differentially with respect to the environmental conditions of the site of origin. Trait spaces of the populations originating from the same site but exposed to different salt stress levels became more dissimilar with increasing environmental aridity. Thus, intraspecific trait variability and salinity effects turned out to be essential in revealing population- and community-level processes in deserts and should be considered in future versions of the ATID-model. In support of the future development of the ATID-model developed in chapter 2, common metamodel types and the purposes of their usage for individual-based models were reviewed and evaluated in chapter 5. The review considered 40 metamodels applied for sensitivity analysis, calibration, prediction and scaling-up of individual-based models and can be used as a guide for the implementation and validation of metamodels. Overall, this thesis, and particularly the ATID-model analyses, highlights how trait-based modelling approaches can contribute to understanding the interplay between key drivers of desert plant communities in arid environments. The accompanying analysis of the nethouse experiment and critical literature reviews outline future extensions of the model and the ways to overcome the technical challenges and data scarcity identified in this thesis. Moreover, this thesis advocates for more intensive studies of the strategies of desert annual plants to survive in temporally and spatially heterogeneous environments with a focus on plant functional traits. Thus, the modelling framework presented in this thesis provides the basis for future research on the fate of communities in arid environments under global change.
Keywords: plant functional traits; ecological modelling; individual-based model; trait-based modelling; arid environment; annual plant community; interspecific and intraspecific trait variability