Microorganisms are important drivers of the carbon, nitrogen, and sulfur cycles on earth. They can adapt to various substrates and, thus, inhabit ecological niches too extreme for higher lifeforms, such as sub- and anoxic or even sulfidic waters. They follow a wide range of ecological strategies with variable levels of specialization to their environment. In the highly stratified water column of the Baltic Sea, respiration of organic matter in combination with sluggish ventilation causes the formation of anoxic zones. Here, organic material is decomposed anaerobically, which leads to the increased production of hydrogen sulfide – a highly toxic compound for higher lifeforms (e.g. multicellular organisms). The transition zone between sulfidic and suboxic conditions, the redox zone, is inhabited by two chemolithoautotrophic key organisms, both with the same ability to detoxify hydrogen sulfide via oxidation with nitrate. Interestingly, both organisms show an overlapping abundance. While the gammaproteobacterial SUP05 clade is most abundant in the suboxic zone, the epsilonproteobacterial Sulfurimonas GD17 subgroup dominates the sulfidic zone. This led to the question of how these two organisms can survive within the same habitat although they exhibit the same substrate requirements. In Paper I, I coupled phylogenetic identification with single-cell uptake measurements of SUP05 and Sulfurimonas GD17 in environmental samples from the Gotland Deep redox zone. I was able to identify niche separation due to different substrate utilization strategies: SUP05 is streamlined, non-motile, and slowly utilizing; essentially a K-strategist, adapted to low substrate conditions and omnipresent in most of the oxygen minimum zones worldwide. In contrast, Sulfurimonas GD17 is a fast utilizing r-strategist, specialized for high and fluctuating substrate conditions, which uses chemotactic behavior to move into regions of favorable substrate conditions. Together they drive a highly efficient detoxification machinery in the Baltic Sea redox zone. Remarkable microbial strategies which influence matter cycling can be found in other pelagic environments as well. Sinking particles are considered an important potential habitat for pelagic microorganisms due to their substrate richness and structural heterogeneity, as well as their omnipresence in the world oceans. Within individual porous particles, it is theorized that innumerable redox gradients should exist at the microscale, whose attributes in aggregate would drive and control significant elemental fluxes globally. However, marine particles still represent a major black box in microbial ecology due to their fragile nature, which makes them inaccessible for detailed micro-scale observations. Therefore, I present in Paper II a cryogel-based embedding and slicing approach, which enables detailed microscopic investigations of the microbial community within the intact particle structure. The approach is compatible with most structural and phylogenetic staining protocols, such as for different extracellular exopolymers and microbial identification using various fluorescent in situhybridization (FISH) protocols. It also allows the three-dimensional reconstruction of whole aggregates as well as precise porosity calculations and is, moreover, applicable to sediment traps for undisturbed in situ samplings. As Paper I clearly illustrates, cellular abundance and species distribution must be accompanied by cellular activity measurements to fully describe an organism’s ecological role. In Paper III, I therefore present an optimized embedding and slicing method based on soft and hard plastic resins, which enables single-cell uptake measurements using modern nano-scale secondary ion mass spectrometer (NanoSIMS) measurements across the complex microzone structures of marine particles. Embedded specimens were characterized by low outgassing and ablation properties within the ultra-high vacuum chamber (i.e. good conditions for NanoSIMS), but high secondary ion yields. Moreover, critical aspects of cellular biogeochemistry, such as the potential use of alternative electron acceptors by microorganisms, could be identified within particles for the first time, visible as 34S and 15N enrichments from stable isotope labelled sulfate and nitrate in single cells. Staining properties for structural compounds similar to those in Paper II, enabled three-dimensional reconstruction and porosity calculations as well. The combination of both methods presented in Paper II and Paper III opens up new ways to investigate the microbial ecology and their interaction with the particle structure in terms of phylogeny and activity. The influence of the particle-associated microbial community on matter cycling at larger scales depends both on particle structure (above) and on particle abundance and distribution in the water column. To scale up results measured by the methods developed above, we need to identify and measure the distributions of fragile particles at high vertical resolution, without physical distortion. In Paper IV, I present a coupled study of optical particle quantification, physical particle characterization, as well as molecular sequencing in the region of Fram Strait. Calculated particle sinking trajectories and microbial genetic source tracking revealed a strong vertical connectivity between the observed microbial communities. This connectivity was most pronounced in areas with sea ice coverage, where almost half of the particle-associated communities in the deep sea were linked to surface-derived microbes. In turn, it could be concluded that further sea ice decline in the Arctic Ocean may reduce vertical microbial connectivity, which possibly alters current biogeochemical cycling. This study exemplifies the huge potential of optical quantification coupled to microbiological and molecular methods for multiscale particle investigations. The abundance and sinking behavior of particles are highly influenced by biological processes, including microbial degradation and remineralization, as well as grazing and repackaging by zooplankton. However, physical forcing, particularly on the sub-meso and mesoscale, critically shape particle distributions in the water column. In Paper V, I present a combined investigation of optical particle counting and classification as well as Acoustic Doppler Current Profiler (ADCP) based current velocity measurements in a cyclonic eddy of the South Atlantic. I observed vertical propagation of presumably wind-driven inertial wave energy following the vorticity field at the eddy perimeter, a process known as ‘inertial chimney’ effect. The resulting zone of increased horizontal shear in the upper 1500 m caused increased upward vertical nutrient flux, supporting enhanced primary production and intensified particle formation in surface eddy. I could show that particles > 0.5 mm in diameter generally followed the relative vorticity field, leading to a sub-surface V-shape of the particle distribution that has not previously been observed. Repackaging and fragmentation by copepods in combination with low carbon-specific degradation led to a threefold increased carbon flux to the deep sea in the center of the eddy. I concluded that cyclonic eddies must regularly cause increased deep carbon export events, in the oligotrophic South Atlantic gyre, and globally. Global matter cycles, including entire pelagic food webs, are affected by the microbial dynamics of sinking particles. These, in turn, are shaped by a wide variety of physical and biological processes ranging from the microscale to mesoscale. Sinking particles and their complex communities thus represent a biogeochemical link between small- and large-scale processes. My work highlights how the global impacts of particle-associated microbial communities can only be understood through investigations using interdisciplinary approaches at multiple scales. In my thesis, I used cutting-edge methodologies to investigate microbial processes at the micro-scale, and built strategies to integrate these processes into a broader understanding of microbial dynamics at oceanographic scales of relevance to the global ocean.