In this thesis, high resolution ocean models are used to evaluate and forecast coastal ocean variability in two different applications. In the first study, the 2-km resolution ocean circulation model for the Eastern Bering Sea is utilized to understand whether slope-interior exchange along the path of the Aleutian North Slope Current (ANSC) helps maintain the subsurface temperature maximum on the isopycnal surface 26.8 kg m⁻³, approximately 300-400 m deep. The simulation period is June-October of 2009. At the abovementioned isopycnal surface, the model shows the warmer pattern extending westward along the southern slope of the Aleutian Islands and then eastward along the northern slope as the season progresses. The direct exchange from the south to the north through Amukta Pass on this isopycnal surface is very limited. The model does not exhibit vigorous eddy shedding along the ANSC. However, there are several topographic features where the warm slope current separates into the basin, particularly at 178˚ W (just east of Amchitka Pass) and 174˚ W (Atka Island). Currents on the 26.8 kg m⁻³ isopycnal surface are too slow to account for the warming pattern along the ANSC reaching the Bering Canyon and the Bering Slope Current. The warming can be explained as a combination of faster advection of warmer waters above and downward vertical turbulent transport due to intensive tides. This hypothesis is confirmed by the heat equation term balance analysis and two-dimensional Lagrangian particle tracking on the 26.8 kg m⁻³ surface and a shallower, 26.4 kg m⁻³ surface. In the second study, a team of four graduate students, including two ocean modelers, a cartographer, and a social scientist, work together as part of the National Science Foundation Research Trainee (NRT) program to develop new products based on ocean forecasts, quantify their uncertainty and communicate this knowledge to commercial fishermen. A 2-km resolution ocean prediction system for the Oregon and Washington coasts produces three-day forecasts of surface velocity, temperature, and salinity. Based on the social scientist’s communications with the commercial fishermen on their perceptions of risk and uncertainty, uncertainty in the surface current forecast is quantified by calculating the root mean square error of the forecast with high frequency radar observations for each forecast horizon. This calculation reveals that the model performs better in the northern portion of the domain where high frequency radar observations are available, with a noticeable source of error being near the Columbia River Estuary. Additionally, the depth of the thermocline is calculated with two different methods: as a depth at which the temperature is 2˚F less than the surface temperature (a definition provided by the commercial fishermen) and as a depth of the maximum buoyancy frequency squared. Overall, the two parts of our thesis study complement each other showing that coastal ocean models can be used both for basic oceanographic research and for operational prediction.
description.provenance : Made available in DSpace on 2017-10-17T00:02:01Z (GMT). No. of bitstreams: 1thesis_MatthewMauch.pdf: 2868238 bytes, checksum: d94a28bf3691cb2392f7a85f6cfeec8b (MD5)
description.provenance : Submitted by Matthew Mauch (mauchm) on 2017-09-25T18:03:21ZNo. of bitstreams: 1thesis_MatthewMauch.pdf: 2868238 bytes, checksum: d94a28bf3691cb2392f7a85f6cfeec8b (MD5)
description.provenance : Approved for entry into archive by Margaret Mellinger(email@example.com) on 2017-10-17T00:02:01Z (GMT) No. of bitstreams: 1thesis_MatthewMauch.pdf: 2868238 bytes, checksum: d94a28bf3691cb2392f7a85f6cfeec8b (MD5)
description.provenance : Approved for entry into archive by Julie Kurtz(firstname.lastname@example.org) on 2017-09-25T19:41:56Z (GMT) No. of bitstreams: 1thesis_MatthewMauch.pdf: 2868238 bytes, checksum: d94a28bf3691cb2392f7a85f6cfeec8b (MD5)