<p> The Gulf of Alaska is a seasonally productive ecosystem surrounded by glaciated coastal mountains with high precipitation. With a combination of high biological production, inputs of suspended sediments from glacial runoff, and contrasting nutrient regimes in offshore and shelf environments, there is a great need to study particle cycling in this region. I measured the concentrations and size distributions of large marine particles (0.06-27 mm) during four cruises in 2014 and 2015 using the Underwater Vision Profiler (UVP). The UVP produces high resolution depth profiles of particle concentrations and size distributions throughout the water column, while generating individual images of objects >500 &mu;m including marine snow particles and mesozooplankton. </p><p> The objectives of this study were to 1) describe spatial variability in particle concentrations and size distributions, and 2) use that variability to identify driving processes. I hypothesized that UVP particle concentrations and size distributions would follow patterns in chlorophyll <i>a</i> concentrations. Results did not support this hypothesis. Instead, a major contrast between shelf and offshore particle concentrations and sizes was observed. Total concentrations of particles increased with proximity to glacial and fluvial inputs. Over the shelf, particle concentrations on the order of 1000-10,000/L were 1-2 orders of magnitude greater than offshore concentrations on the order of 100/L. Driving processes over the shelf included terrigenous inputs from land, resuspension of bottom sediments, and advective transport of those inputs along and across the shelf. Offshore, biological processes were drivers of spatial variability in particle concentration and size. High quantities of terrigenous sediments could have implications for enhanced particle flux due to ballasting effects and for offshore transport of particulate phase iron to the central iron-limited gyre. The dominance of resuspended material in shelf processes will inform the location of future studies of the biological pump in the coastal Gulf of Alaska. This work highlights the importance of continental margins in global biogeochemical processes.</p>

<p> The export of particulate carbon from the surface ocean into deeper water and to the seabed is a critical component of the carbon cycle. The concentrations and compositions of particulate pigments collected at different depths and sinking at different settling velocities can be used as a proxy for biologically mediated processes important to the early degradation of OM. By knowing what processes the compositional and quantitative changes in the particulate pigments represent, the POM cycle of the BaRFlux area can be better understood. It is important to understand the POM cycle because deposition of OM to the seabed is the only way that OM is sequestered. The removal of POM from the marine POM cycle is especially important to understand in subtropical gyre areas like the BaRFlux site because: 1. subtropical gyres are areas of downwelling, and therefore POM transport to the deep ocean and may increase as global warming continues. 2.the flux of CO₂ to the ocean is increasing from rising levels of atmospheric CO₂, and CO₂ removed by the biological pump will lessen processes like ocean acidification.</p><p> This thesis examines the early degradation of chloropigments in the sediment and water column in the Bermuda Rise area of the Sargasso Sea. Water column particulate samples were collected with in-situ pumps, Niskin bottles, and Indented Rotating Sphere (IRS) sediment traps, and sediment was collected by box cores during 2011-2013 to record seasonal patterns in quantity and quality of particulate pigments as a function of water column depth and particle size. Chl-a, Chl-b, and pheopigments were separated and quantified using reverse-phase High Performance Liquid Chromatography (HPLC).</p><p> The comparison of data from in-situ pumps and Niskin bottles indicates that collection method significantly affects particulate pigment data concentrations. Niskin bottle data showed total pigment concentrations 10 times greater than in-situ pump pigment concentrations at shallow depths. At depths below the euphotic zone, Niskin bottle and in-situ pump concentrations both appear similar because the particulate pigment concentrations were below the detection limit. For the BaRFlux study area, the differences in Niskin bottle and pump data are most likely from: 1. the biased particle distribution due to sampling a small volume of seawater with Niskin bottles in an area of dilute particle concentration; 2. the greater retention efficiency of picophytoplankton on Niskin GF/F filters than 1-<i>&micro;</i>m in-situ pump microquartz filters.</p><p> The compositional changes seen in small suspended particulate pigments over depth is consist with small suspended particles being consumed by shallow water zooplankton and then increasingly altered by microbial activity with increasing depth. The composition of small and large particulate pigments were compared to determine if aggregation-disaggregation was an important process. Larger suspended particulate pigments were nearly 100% Chl-a over depth and distinct from smaller suspended particulate pigments other than samples from May or June, during which particle exchange may be more important. The comparison of particulate pigment data to CTD beam transmissivity profiles suggests that the nepheloid layer consists of small suspended particulate matter rather than large particles.</p><p> Sediment trap samples were compositionally enriched in pheopigments relative to smaller bottle and pump samples, indicative of enrichment with more rapidly sinking larger zooplankton fecal pellets. The mole% of chlorophyll-a labile pigment increased with increasing settling velocity, suggesting aggregation may increase the settling velocity of particles enough to escape zooplankton feeding. The particulate pigment composition of seafloor sediment collected in August was compositionally distinct from that of suspended and sinking particulate pigments and was nearly 100% pheophorbide-a, indicating POM degradation by feeding macrobenthos.</p>