This paper presents measurements of momentum flux in moderate to strong winds using both the Reynolds flux and dissipation methods on a deep-water stable tower operated by the Bedford Institute of Oceanography. The dissipation method, which does not require explicit measurement of vertical velocity, is shown to be effective in high-wind conditions, while the Reynolds flux method is more suitable for low and moderate winds. The drag coefficients derived from 196 Reynolds flux measurements agree well with those reported by Smith (1980) based on independent observations at the same site. A comparison of dissipation and Reynolds flux results based on 192 runs shows excellent agreement for wind speeds from 4 to 20 m/s. The extensive dissipation data set (1086 hours from the tower and 505 hours from the ship CCGS Quadra) is used to investigate the dependence of the drag coefficient on wind speed, fetch, and stability. The results show that the drag coefficient in the neutral case (CDN) is independent of stability and fetch (for fetch/height ≥ 800) but increases with wind speed above 10 m/s. The paper also discusses the time variability of the momentum flux and the sources of variation in the drag coefficient, including the influence of rising and falling winds and changes in wind direction. The authors provide a method for calculating stress from the CDN and observations of wind speed and surface minus air temperature at heights other than 10 m.This paper presents measurements of momentum flux in moderate to strong winds using both the Reynolds flux and dissipation methods on a deep-water stable tower operated by the Bedford Institute of Oceanography. The dissipation method, which does not require explicit measurement of vertical velocity, is shown to be effective in high-wind conditions, while the Reynolds flux method is more suitable for low and moderate winds. The drag coefficients derived from 196 Reynolds flux measurements agree well with those reported by Smith (1980) based on independent observations at the same site. A comparison of dissipation and Reynolds flux results based on 192 runs shows excellent agreement for wind speeds from 4 to 20 m/s. The extensive dissipation data set (1086 hours from the tower and 505 hours from the ship CCGS Quadra) is used to investigate the dependence of the drag coefficient on wind speed, fetch, and stability. The results show that the drag coefficient in the neutral case (CDN) is independent of stability and fetch (for fetch/height ≥ 800) but increases with wind speed above 10 m/s. The paper also discusses the time variability of the momentum flux and the sources of variation in the drag coefficient, including the influence of rising and falling winds and changes in wind direction. The authors provide a method for calculating stress from the CDN and observations of wind speed and surface minus air temperature at heights other than 10 m.