Description: Seismological models of upper mantle structure provide important constraints on the Earth"s convection system. Resolving the details of the upper mantle discontinuities is important for modelling the composition of the mantle and for understanding the effect that the discontinuities may have on mantle convection. Recently, numerous permanent and temporary seismic stations and networks have been set up around the world. It is possible to get the seismic records for the research needs from data management systems like IRIS, GEOFON, GEOSCOPE, FREESIA, etc. The use of seismograms collected from a large number of stations and earthquakes around the world enable us to study the global and the regional structure of the Earth. In this work, the receiver function technique (e.g. Owens et. Al., 1995) is applied to study the upper mantle structure in the northwest Pacific subduction zone and in the Hawaiian hotspot area. In the northwest Pacific, the Pacific plate is subducted into the upper mantle to more than 600 km depth, indicated by seismicity. In Hawaii, the volcanic edifice of the Hawaiian Islands and seamounts are believed to result from the passage of the oceanic lithosphere over a stationary mantle hotspot (Wilson, 1963; Morgan, 1971; Morgan et. al., 1995). In both regions the upper mantle structure is affected by the cold and warm materials, respectively. To study the extension of the temperature anomaly is important for understanding the Earth"s convection system. The olivine component of the mantle material is intensively studied in laboratories (e.g. Ito and Takahashi, 1989; Irifune, 1987). With increasing temperature and pressure, the olivine crystal undergoes a series of phase transformations which will result in a variation of the seismic structure. The effect of the temperature anomaly on the main upper mantle discontinuities will be discussed in chapter 2. Recently, the receiver function technique is increasingly applied to investigate the upper mantle discontinuities. To isolate the upper mantle conversion phases, newly developed moveout correction and migration methods are applied to separately distributed seismic stations as well as station arrays. The receiver function method used in this study will be introduced in chapter 3. In chapter 4 and 5, receiver function studies in the area of the northwest Pacific subduction zone and the Hawaiian mantle plume are presented. Regional tectonic background and the previous seismological works in these two areas will be first introduced in each chapter, and followed by description of data, processing steps, results and interpretations. In chapter 6, I will summarize the observations of the 410 and 660 topography in the northwest Pacific subduction zone and in the area around the Hawaiian mantle plume.