Patterns of temporal changes in species composition and diversity in island ecosystems

• Κύριος συγγραφέας: Καραδήμου, Ελπίδα
• Άλλοι συγγραφείς: Δημόπουλος, Παναγιώτης
• Μορφή: Thesis
• Γλώσσα: English
• Έκδοση: 2017
• Θέματα: Functional diversity; Species functional traits; Alpha diversity; Beta diversity; Βeta functional diversity; Spatial scale; Temporal biodiversity changes; Functional diversity - area relationship (FDAR); Assembly rules; Volcanic islands; Λειτουργική ποικιλότητα; Λειτουργικά χαρακτηριστικά των ειδών; Άλφα ποικιλότητα; Βήτα ποικιλότητα; Βήτα λειτουργική ποικιλότητα; Χωρική κλίμακα; Σχέση λειτουργικής ποικιλότητας – έκτασης; Κανόνες συνάθροισης ειδών; Χρονικές μεταβολές ποικιλότητας; Ηφαιστειογενή νησιά; 577.52

Patterns of spatial and temporal changes in biodiversity, natural or anthropogenic, and their importance in understanding the dynamics and restoration of natural ecosystems, have long been among the major questions in ecology. These changes are rather important to be understood in the early stages of communities’ development. Young oceanic islands offer the unique opportunity of a natural laboratory for the investigation of such processes. The interest in using different facets of biodiversity, in order to understand ecological processes rose over the past years, since there is an increasing awareness that species’ traits influence species coexistence and ecosystem function. Ecologists now quantify trait variation or multivariate trait differences within a community, generally referred to as ‘functional diversity’ (FD). This diversity facet technically represents the diversity of traits but is taken to represent the diversity of species’ niches or functions. This thesis consists an endeavor to investigate small spatial scale patterns as well as short- and long-term temporal patterns of vegetation, functional diversity and community assembly on two young islands (Palea Kameni, PK and Nea Kameni, NK) of Santorini Archipelagos in Greece, which consist the result of subaquatic volcanic activity. These two islands differ on geological age, history of volcanic activity and total area. For this purpose three data sets were compiled: i) a vegetation data set comprised of species abundance data, collected on both islands in three consecutive years (from 2010 to 2012) using different spatial scales, ii) a data set comprised of presence-absence data collected from 1911 to 1990 (four censuses conducted in 1911, 1933, 1967 and 1990) on NK and iii) a functional traits data set comprised of 26 functional traits (vegetative, habitat-related and reproductive) for each recorded taxon, collected from databases and bibliography. More specifically, this thesis examines the distribution of both plant communities and habitat types as well as temporal changes in vegetation patterns form 1984 to 2015 on both islands. Seven plant communities in total have been defined on both islands (four on PK and three on NK). Four of these communities comprise the data set used in the current thesis’ analysis. Six habitat types were recognized and described on both islands in 2015 (five on PK and three on NK). PK is geologically more than 1800 yrs. older ii than NK and no eruptions have disturbed the island’s vegetation since 726 AD. PK's actual vegetation consists partly of an evergreen sclerophyllous shrub community, most probably representing the climax vegetation on the island, thus no significant changes were noted since 1984. After several outbursts of the island’s volcano in the past century (the last one recorded in 1950), the re-establishment of NK's plant cover has reached different stages in different parts of the island, according to age and ashes deposition. Significate variations have been identified in the island’s vegetation even after 1984, the most tricking in the floristic composition of one of the two therophytic communities, which has changed substantially. The exploration of both taxonomic and functional aspects of biodiversity simultaneously can lead to a better understanding of the ecological processes. The analysis of functional beta diversity represents a relatively new approach in community ecology. The aim of this thesis is to track possible temporal changes in taxonomic and functional alpha and beta diversity within the short time period from 2010 to 2012 and to investigate, if the temporal patterns change with spatial scale. No evidence for a general temporal variation trend was found in the two facets of diversity according to the results, neither for the alpha nor for the beta component. Low functional turnover among communities as well as a small change in functional beta diversity was observed. Thus, it is important for ecologists studying biotic homogenization to quantify also changes in functional beta diversity. Positive correlation among different components of diversity was detected among some of their facets. It is indeed crucial to determine whether the observed species replacement and corresponding changes in taxonomic richness induce functional turnover (i.e. whether ‘loser’ and ‘winner’ species are functionally different) and/or changes in functional richness. Moreover, the results highlight the importance of spatial scale when studying variation in the functional composition among a set of communities, since changes are more evident at fine spatial scales, where only a part of the total community’s species pool is recorded. There are only a few recent studies investigating the relationship between functional diversity and area, the Functional Diversity - Area Relationship (FDAR). In this thesis the pattern of the relationship between functional diversity and area is examined, along with the question, whether this pattern differ, when different facets of FD are used, as well as different trait groups (in order to examine the effect of the data set used in the analysis on these patterns). A typology of FDARs was identified, depending on the iii facet of FD analyzed: (i) strong positive for indices that quantify the range of functional traits in the community, (ii) negative correlation for indices quantifying the evenness in the distribution of abundance in the trait space, (iii) no clear pattern for indices reflecting the functional similarity of species and (iv) idiosyncratic patterns with area (either positive or negative or non-significant correlations) for the index that reflects the divergence of individuals in the functional space occupied by the species assemblage. As area increases, the range of traits observed in the community increases, but the abundance of traits does not increase proportionally and some traits become dominant, implying a reliance on some functions that may be located in either the center or the periphery of the trait space. This pattern is not dependent on the group of functional traits used or the type of SAR (island species-area curve or speciesaccumulation curve). The two sea-born islands offer the opportunity to study community assembly at two different phases: a very early and a late stage of vegetation succession and to determine whether: i) the mechanisms driving community assembly differ between two islands of different age and history of vegetation development, ii) sampling scale affect the strength that each assembly mechanism poses on the assembly of plant communities and iii) the intensity of the mechanisms of assembly change during short time periods. Null model analysis was used to test for two different assembly mechanisms: habitat filtering and limiting similarity against random assembly. Functional diversity of the communities on the younger island implied no specific assembly mechanism; perhaps due to its age, the community is still at the early stages of colonization (i.e. stochastic events, such as arrival of new species, prevail). On the older island, the lack of disturbances for a long time period allowed the establishment of communities driven by certain assembly mechanisms, such as competition and habitat filtering. Changes in communities, regarding species composition, species richness and species abundances during short time periods do not reflect changes in the mechanisms of the communities’ assembly. The existence of a long-temporal vegetation data series for NK provides the opportunity to study colonization (immigration and extinction events) as well as the biotic assembly and the successional processes following the destruction of vegetation after a volcanic eruption, using alpha and beta functional diversity and null model analysis to test for two different assembly mechanisms. The results indicate that alpha diversity increases iv with time interval since disturbance, while beta diversity decreases. The patterns of temporal changes in taxonomic and functional dissimilarity do reveal functional redundancy, at least to some extent. Long – term temporal changes in assembly processes revealed that the effect of abiotic filtering decreases over time. Overall, the use of functional diversity at different facets and components provided valuable insights into biodiversity patterns and proved to be able to provide important complementary information to the taxonomic component of diversity. Spatial scale, even when small local scales are examined, affected all patterns investigated. The two sea-born islands proved to be a valuable tool for the investigation of different stages in succession and the assembly of communities, despite the lack of repeatability.