This paper first gives a brief review of the historical development of shear-wave splitting and P-wave azimuthal anisotropy in oil and gas seismic exploration, and focuses mainly on their distribution in the upper crust, the link between them, and their application requirements for seismic fracture detection. When a shear-wave enters a rock containing vertically aligned fractures, it splits into two travelling with different speeds, which is referred to as shear-wave splitting, or birefringence. Forty years of researches and applications of shear-wave splitting reveal that shear-wave splitting occurs widely in the upper crust, but most of the splitting is concentrated in the near surface (<1200 m). While P-wave azimuthal anisotropy in the upper crust varies considerably from weak to strong and shows no consistence. When applied to predict fractures, P-wave azimuthal anisotropy is only sensitive to gas-saturated fractures, and very insensitive to other fluid-saturated fractures. For fracture detection, theoretical analyses and experiment studies of the link between shear-wave splitting and P-wave azimuthal anisotropy verify that, it is essential to correct the near surface effects, which might limit the application to some extents. It is important to select target areas with relatively simple structure and relatively flat surface, as well as relatively thick reservoirs. Non-zero offset vertical seismic profiles (VSPs) or Walkaround P- or S-wave VSPs are also required to correct the near-surface effects and calibrate the reservoir response. In summary, the multi-component seismic data quality, the near-surface and reservoir conditions are three key factors affecting the application of shear-wave splitting and P-wave azimuthal anisotropy for fracture detection in oil and gas exploration.