Abstract. Synthetic Aperture Radar (SAR) images from ERS-1 have been used to study the characteristics of internal waves northeast and south of Taiwan in the East China Sea, and east of Hainan Island in the South China Sea. Rank-ordered packets of internal solitons propagating shoreward from the edge of the continental shelf were observed in the SAR images. On the basis of the assumption of a semidiurnal tidal origin, the wave speed can be estimated and is consistent with the internal wave theory. By using the SAR images and hydrographic data, internal waves of elevation have been identified in shallow water by a thicker mixed layer as compared with the bottom layer on the continental shelf. The generation mechanism includes the influences of the tide and the Kuroshio intrusion

An internal solitary wave (ISW) encountering the shelf break, making the waveguide is compressed, can cause different joint coupling effect for acoustic modes. In this paper, the extended criterion for adiabatic invariance is developed by parameterizing the joint mode coupling effect of an ISW encountering the shelf break and is used for sensitivity studies considering various internal wave amplitudes and slope angles of shelf break, to examine the acoustic coupling effect resulting from both bathymetry and ISW. The modeling results are accurately predicted by the extended criterion of adiabatic invariance and compared to experimental observations from ASIAEX and NLIWI of the effect of acoustic waveguide being compressed by the shelf break and ISW. Results demonstrate that the coupling of acoustic energy to higher modes as the waveguide is compressed when the ISW encounters the shelf break. And as amplitude of the ISW and the incline of the sloping bottom increase, coupling strength for both adjacent- and non-adjacent modes is enhanced. [This research is sponsored by national science council.]

No abstract

Internal waves and bathymetric variation create time-and space-dependent alterations in the ocean acoustic waveguide, and cause subsequent coupling of acoustic energy between propagating normal modes. In this paper, the criterion for adiabatic invariance is extended to the case of an internal solitary wave (ISW) encountering a sloping bathymetry (i.e., continental shelfbreak). Predictions based on the extended criterion for adiabatic invariance are compared to experimental observations from the Asian Seas International Acoustics Experiment. Using a mode 1 starter field, results demonstrate time-dependent coupling of mode 1 energy to higher adjacent modes, followed by abrupt coupling of mode 5-7 energy to nonadjacent modes 8-20, produces enhanced mode coupling and higher received levels downrange of the oceanographic and bathymetric features. Numerical simulations demonstrate that increasing ISW amplitude and seafloor slope enhance the coupling of energy to adjacent and nonadjacent modes. This enhanced coupling is the direct result of the simultaneous influence of the ISW and its proximity to the shelfbreak, and, compared to the individual effect of the ISW or shelfbreak, has the capacity to scatter 2-4 times the amount of acoustic energy from below the thermocline into the upper water column beyond the shelfbreak in realistic environments.

General-purpose computing on graphics processing units (GPGPU) can provide orders of magnitude more computing power than general purpose processors (CPU) for highly parallel applications. For such parallel applications, the memory traffic pattern of GPGPUs behaves considerably different from that of CPUs. This gives rise to opportunities for optimizing the on-chip interconnection network (NoC) of GPGPUs. In this work, we first investigate the characteristics of GPGPU memory traffic of typical benchmarks and categorize the memory traffic patterns. Different traffic patterns require different throughput in the request and reply paths of the NoC to match the network load. To meet this requirement, we examine the feasibility of scaling the network frequency dynamically to balance the throughput of the request and reply networks. The decision is guided by monitoring some shader cores to identify the memory traffic pattern. Performance evaluation shows that this dynamic frequency tuning design can achieve up to 27% improvement in terms of execution speedup compared to a baseline setting and 7.4% improvement on average.978-1-4799-7615-7/14/$31.00 ©2014 IEEE