Abstract To reduce the dependence on fossil fuel and imported energy resources, Taiwan has ever-increasing needs of renewable energy. With the rapid development of the technologies of wave energy converter, the wave energy source will be able to meet parts the demand. The Energy Research Laboratories of the Industrial Technology Research Institute, Taiwan (2005), based on the statistic of one-year wave data, stated that the mean wave energy at the northeast coast of Taiwan reaches 11.56 kW/m, giving it the potential of wave power utilization. However, one of the major obstacles with the wave energy utilization is lack of long-term ocean wave measurements. The long-term variations in wave parameters impose changes in wave energy converter outputs. Lack of long-term data makes it difficult to assess the cost-benefit of wave energy conversion projects for the policy and decision makers. The present study aims to quantitatively evaluate the wave climate variations of the northwestern Pacific and the Taiwan Waters based on long-term wave data base. Wave observations around Taiwan have been performed since 1998, thus, earlier data of wave climate are not available. This study reconstructs the wave data of the northwest Pacific over the past three decades based on the SWAN numerical wave model that driven by NECP global reanalysis wind fields. The simulation results are compared and validated with measured data. The results show that the long-term wave climate variations around Taiwan consist of oscillations of three different periods, i.e. the seasonal, inter-annual and decade oscillations. The seasonal oscillation has significant amplitude that leads the wave energy one order magnitude greater in winter than in summer. In addition to seasonal changes, the wave energy features inter-annual variations, which are highly related to the El Nino and La Nina phenomena. In the La Nina years, the annual averaged wave energy could be double than in El Nino years. Finally, this study adopted the Man-Kendall Non-Parametric Test and the Hilbert Huang EMD method to analyze the long-term wave variation trends. The results showed that the wave height experienced climbing trends during 1976–1985 and 1997–2006, and a descending trend during 1985–1997. The reasons for wave climate oscillations in the decadal variation should be further investigation.