Description: Supralittoral and shallow water seaweed communities are particularly exposed to impacts such as climate change and disturbance by humans. Therefore, their classification, the study of composition, and the monitoring of their structural changes are particularly important. A phytosociological survey of the supralittoral and upper sublittoral vegetation of the South West Baltic Sea revealed eight phytobenthos communities with two variants comprising 35 taxa of macrophytes (18 taxa of Chlorophyta, 13 taxa of Rhodophyta and four taxa of Phaeophyceae, Ochrophyta). Five of the eight communities were dominated by Ulvales (Ulva intestinalis, Kornmannia leptoderma, and three Blidingia species), the other three by Fucus vesiculosus. Most Fucus vesiculosus-dominated communities contained U. intestinalis and U. linza as subdominants. Only one of the communities had until now been described as an association ( Ulvetum intestinalis Feldman 1937). The syntaxonomic composition of the investigated vegetation includes both phytocenoses with the domination of green algae ( Ulvetum intestinalis Feldman 1937 and communities of Blidingia marginata, unidentified Blidingia spp. and Kornmannia leptoderma), as well as a number of communities dominated by Fucus vesiculosus. Mainly boreal Atlantic species and cosmopolitans make up the bulk of the species in these associations.

Description: Seaweed bioinvasions increasingly affect coastal environments around the world, which increases the need for predictive models and mitigation strategies. The biotic interactions between seaweed invaders and invaded communities are often considered a key determinant of invasion success and failure and we here revise the current evidence that the capacity of seaweed invaders to deter enemies in newly reached environments correlates with their invasion success. Particularly efficient chemical defences have been described for several of the more problematic seaweed invaders during the last decades. However, confirmed cases in which seaweed invaders confronted un-adapted enemies in newly gained environments with deterrents that were absent from these environments prior to the invasion (so-called “novel weapons”) are scarce, although an increasing number of invasive and non-invasive seaweeds are screened for defence compounds. More evidence exists that seaweeds may adapt defence intensities to changing pressure by biological enemies in newly invaded habitats. However, most of this evidence of shifting defence was gathered with only one particular model seaweed, the Asia-endemic red alga Agarophyton vermiculophyllum, which is particularly accessible for direct comparisons of native and non-native populations in common garden experiments. A. vermiculophyllum interacts with consumers, epibionts and bacterial pathogens and in most of these interactions, non-native populations have rather gained than lost defensive capacity relative to native conspecifics. The increases in the few examined cases were due to an increased production of broad-spectrum deterrents and the relative scarcity of specialized deterrents perhaps reflects the circumstance that seaweed consumers and epibionts are overwhelmingly generalists.

Description: Ocean warming is one of the most important factors in shaping the spatial distribution and genetic biodiversity of marine organisms worldwide. The northwest Pacific has been broadly illustrated as an essential seaweed diversity hotspot. However, few studies have yet investigated in this region on whether and how past and ongoing climate warming impacted the distribution and genetic pools of coastal seaweeds. Here, we chose the invasive species Gracilaria vermiculophylla as a model, and identified multiple genetic lineages in the native range through genome-scale microsatellite genotyping. Subsequently, by reconstructing decadal trends of sea surface temperature (SST) change between 1978 and 2018, we found that SST in northern Japan and the East China Sea indeed increased broadly by 0.25-0.4°C/decade. The projections of species distribution models (SDMs) under different future climate change scenarios (RCP 2.6, RCP 4.5, RCP 6.0 and RCP 8.5) indicated that a unique genetic pool of G. vermiculophylla at its current southern range limit (i.e. the South China Sea) is at high risk of disappearance, and that the populations at its current northern range limit (i.e. in Hokkaido region) will undergo poleward expansions, particularly by the year 2100. Such responses, along with this species’ limited dispersal potential, may considerably alter the contemporary distribution and genetic composition of G. vermiculophylla in the northwest Pacific, and ultimately threaten ecological services provided by this habitat-forming species and other associated functional roles.

Description: Increased pollution in the coastal areas may cause changes in the biodiversity of marine organisms depending upon their physiological capacity and resilience to thrive under stressing environmental conditions. The present research evaluates the heavy metals pollution degree of coastal waters using the macroalgae Ericaria selaginoides as bioindicator along the Atlantic coast of Morocco. Eight stations were chosen: two located near Eljadida city, three nearby Safi city and three around the city of Essaouira. Results showed that the heavy metal content in the thalli of E. selaginoides, in seawater and sediment varied seasonally. At the same time, it was negatively correlated with algal biodiversity onsite. However, the Chemical Oxygen Demand was significantly higher at the polluted station S5 than at other stations, while Dissolved Oxygen and Biological Oxygen Demand were lower. E. selaginoides accumulated metals in the following order Fe 〉 Zn 〉 Mn 〉 Cu 〉 Ni 〉 Pb 〉 Cr 〉 Cd. In conclusion, E. selaginoides is overall more resilient to heavy metal pollution than other marine organisms in the Atlantic coast of Morocco, as indicated by substantially elevated concentrations of heavy metals in some sites. Our results support that E. selaginoides would be a suitable bioindicator for monitoring of heavy metals in polluted coastal areas.

Description: Despite an increasing awareness of disease impacts on both cultivated and native seaweed populations, the development of marine probiotics has been limited and predominately focused on farmed animals. Bleaching (loss of thallus pigmentation) is one of the most prevalent diseases observed in marine macroalgae. Endemic probiotic bacteria have been characterized to prevent bleaching disease in red macroalgae Agarophyton vermiculophyllum and Delisea pulchra; however, the extent to which probiotic strains provide cross-protection to non-endemic hosts and the influence of native microbiota remain unknown. Using A. vermiculophyllum as a model, we demonstrate that co-inoculation with the pathogen Pseudoalteromonas arctica G-MAN6 and D. pulchra probiotic strain Phaeobacter sp. BS52 or Pseudoalteromonas sp. PB2-1 reduced the disease risks compared to the pathogen only treatment. Moreover, non-endemic probiotics outperformed the endemic probiotic strain Ralstonia sp. G-NY6 in the presence of the host natural microbiota. This study highlights how the native microbiota can impact the effectiveness of marine probiotics and illustrates the potential of harnessing probiotics that can function across different hosts to mitigate the impact of emerging marine diseases.

Description: Ongoing global climate changes leads to an increased water temperature and a higher probability of extreme events. This thesis focuses on (i) the characterisation of heatwaves and cold-spells in the Kiel Fjord, (ii) using this characterisation for treatment development, (iii) running mesocosm experiments to unravel the importance of frequency, duration, and intensity of environmental stress (i.e., heatwaves, upwelling, and warming) for the impact on the two important species Asterias rubens and Zostera marina, and (iv) the coping mechanisms of these species with (succeeding) environmental stress. Starfish were substantially impacted by heatwaves, with stronger and longer-lasting impacts in amplified and prolonged heatwaves. The impacts reached from 100% mortality (simulated future intensity) to decreased feeding rates, lower growth, and reduced activity (today’s intensity). Upwelling reduced the starfish's activity. However, the succession of extreme events led to an alleviated impact during the second event. Eelgrass was not impacted by short heatwaves or upwelling alone but had less leaves when the heatwave was extended, or a short heatwave was followed by upwelling. The latter combination of stressors also reduced the number of eelgrass shoots. When exposed to long-term warming, aboveground and belowground biomass of non-heat-selected eelgrass was strongly reduced. However, eelgrass individuals from a potentially heat-selected population grew more than those from a non-heat-selected population. Overall, this thesis highlights that extreme events are important drivers of temperate benthic ecosystems but that their impact depends on the nature of the event, their succession, duration, and intensity. Yet, species may have the potential to cope with recurring stress events via an ecological memory or via heat-selection in situ.

Description: High-quality data is necessary for modern machine learning. However, the acquisition of such data is difficult due to noisy and ambiguous annotations of humans. The aggregation of such annotations to determine the label of an image leads to a lower data quality. We propose a data-centric image classification benchmark with ten real-world datasets and multiple annotations per image to allow researchers to investigate and quantify the impact of such data quality issues. With the benchmark we can study the impact of annotation costs and (semi-)supervised methods on the data quality for image classification by applying a novel methodology to a range of different algorithms and diverse datasets. Our benchmark uses a two-phase approach via a data label improvement method in the first phase and a fixed evaluation model in the second phase. Thereby, we give a measure for the relation between the input labeling effort and the performance of (semi-)supervised algorithms to enable a deeper insight into how labels should be created for effective model training. Across thousands of experiments, we show that one annotation is not enough and that the inclusion of multiple annotations allows for a better approximation of the real underlying class distribution. We identify that hard labels can not capture the ambiguity of the data and this might lead to the common issue of overconfident models. Based on the presented datasets, benchmarked methods, and analysis, we create multiple research opportunities for the future directed at the improvement of label noise estimation approaches, data annotation schemes, realistic (semi-)supervised learning, or more reliable image collection.