Islands of Sea : the evolution, ecology and conservation of marine lake invertebrates

The marine realm is seemingly boundless, and the absence of physical barriers to dispersal has led to assumptions of high dispersal abilities and subsequent slow speciation rates for marine organisms. Recently, these assumptions are being overturned by studies showing high genetic structuring for marine populations even at small spatial scales, suggesting isolative mechanisms may be at play. In this thesis, four modes of isolation are considered, including neutral and selective processes. Neutral processes of dispersal limitation include isolation-by-distance and isolation-by-resistance. Selective processes that affect establishment are isolation-by-environment and historical contingency. The relative importance of processes shaping genetic structure on different spatial and temporal scales in the marine realm remains unclear.

Islands are ideal systems to test for the modes of isolation. Anchialine systems such as marine lakes provide a marine analogue to oceanic islands by being bodies of seawater surrounded by land but maintaining a subterranean connection to the surrounding sea. For this thesis, marine lakes in Indonesia were selected with a range of connection to the surrounding sea and variable local environments, but in the same spatial and temporal context. Two types of genetic marker panels were used: a single marker (cytochrome oxidase subunit I) and a reduced representation genomic strategy (double digest restriction site-associated DNA sequencing, or ddRAD). Three marine invertebrate species with different life histories were sampled: a jellyfish (Mastigias papua), a sponge (Suberites diversicolor) and a mussel (Brachidontes sp.). The objectives were to 1) set environmental and biological baselines in the marine lakes of Indonesia, 2) assess how benthic communities and three invertebrates conform to expectations of Island Biogeography Theory, 3) test whether invertebrate populations show panmixia or if they show patterns of genetic structuring, and 4) quantify the relative importance of neutral and selective factors shaping population differentiation.

Chapter 2 presents a comprehensive overview of marine lakes in Raja Ampat, West Papua, Indonesia. Thirty-two 32 marine lakes were identified, of which 17 were new to science. The lakes showed large variability in local environments (e.g., temperature ranging from 30.0° to 36.8°C) and the connection to the surrounding sea (relative tidal amplitude ranged from 5% to 89% compared to the sea). Connection to the surrounding sea and temperature were important predictors of diversity distributions. Lake area did not influence benthic group coverage or diversity. It is concluded that marine lakes can be seen as marine analogues of island-like systems and can be used in ecological and evolutionary studies.

In chapter 3 focusses on marine lakes in Raja Ampat, West Papua, containing dense populations of the golden jellyfish (Mastigias papua) as a case study to include peripheral ecosystems in Marine Protected Areas (MPAs). While these lakes receive increasing tourism attention, they are not yet included in conservation management plans. Strong genetic differentiation between lakes was found using COI, including one subspecies defined by previous research and potential unique subspecies, with concordant morphological differences between lakes. Major fluctuations in jellyfish abundance were observed, with no temporally consistent pattern across lakes. Identified threats from stakeholder interviews included tenure disputes, conversion to aquaculture, introduced species and unregulated tourism. A 30-fold increase of tourism to Raja Ampat was observed since 2007. Chapter 3 provides the scientific basis in the form of unique genetic and morphological variation to distinguish the lakes as individual management units.

In chapter 4 population structure and associated drivers of the sponge Suberites diversicolor was assessed and results of genome-wide sequencing were compared to previously published results using single markers. Both low- and high-resolution markers were able to detect two major genetic lineages which might represent distinct species. However, with the high-resolution marker panel, new evidence of strong population structure within one of the lineages was found, even at spatial scales of <10km. The discrepancy in genetic structure assumed for high- versus low-resolution markers perhaps calls for the reassessment of other benthic marine organisms where previous panmixia was reported based on low-resolution markers.

The mussel Brachidontes sp. was sampled for ddRAD sequencing from seven marine lakes to assess whether isolation-by-distance, isolation-by-environment or historical contingency contribute most to population structure in chapter 5. Clear patterns of strong genetic differentiation among marine lake populations and limited gene flow was found, with a clear isolation-by-distance pattern on both large (>1,400km) and small (<200km) spatial scales. Within West Papua (spatial scale <200km), this pattern was accompanied by an association of genetic divergence to the degree of connection to the surrounding sea. It was concluded that the isolation provided by even incomplete barriers to dispersal could be enough to establish long-lasting population divergence, perhaps through eco-evolutionary dynamics such as priority effects.

In chapter 6, the effects of ecological speciation versus neutral processes for Brachidontes sp. mussels was explicitly tested by combining genomic and phenotypic data. A unique dataset is provided consisting of genomic data (ddRAD) and phenotypic data representing environmental acclimatization including morphological shell characteristics, mussel-associated microbial communities, and trophic niche space. In contrast to most population genomic studies in the marine realm, strong clustering per marine lake was found regardless of only using neutral loci or also including loci under putative selection. Geographic location and connection to the surrounding sea were found to be important predictors of genomic variation. No associations between genomic and phenotypic differentiation were found.

The last chapter (chapter 7) discusses the utility of marine lakes as insular systems, and patterns are compared to other population genomic studies in the marine realm. The population genomic data in this thesis consistently showed strong genetic structuring regardless of the marine invertebrate studied, and regardless of using neutral loci or loci under putative selection. Eco-evolutionary dynamics such as priority and monopolization effects may have effectively been fixed, underlining the often-neglected importance of historical contingency in shaping population genomic patterns. This thesis takes a first step in elucidating population genomic patterns of marine lake invertebrates. It paves the way for future studies to assess how insular systems can inform fundamental questions underlying generation and maintenance of biodiversity, and how they can function as models to assess conservation genetics and conservation.