Abstract: The observed strongest global climate variation on the inter annual timescale is the El Niño-Southern Oscillation(ENSO) phenomenon, which is an irregular interannual oscillation resulting primarily from coupled ocean-atmosphere interaction in the tropical Pacific. Since 1970s the El Niño events have been stronger than the La Nina both on frequency and amplitude. Perhaps it links with the global climate warming. In this paper, sea surface temperature, wind stress and OLR data sets are diagnosed by using wavelet trans form. The ENSO cycle(2-7 years) signal and decadal variability(8-20 years) are filtered out to form the data sets in order to investigate the features of the decadal climate change and its effect on ENSO cycle. Then the numerical experiment sare designed to identify the effect byrunning a hybrid tropical Pacific coupled model. The main results of the paper can be outlined as follows:1) For the behavior of SSTA and wind stress in tropical Pacific by wavelet analysis, it is clear that not only the energy density of ENSO cycle(2-7 years) gets stronger during 1990s, but also does so the decadal oscillation(8-20 years). The long-term linear trend of climate change(more than 20 years) can beident ified with a catastrophe climate change in 1976 from cold climate state turning to warm state. 2) The main warming regions of the sea surfacet emperature after 1976 is the east tropical areas and extended west ward along the equator with the maximum value up to 0.6℃. The main features of wind stress pattern are the increasing westerly anomaly in west tropical Pacific and easterly anomaly at east tropical Pacific so that low level convection occurs near the date line. 3) However, the average state in the tropical Pacific in the 1990s is warmer than that in 1980s. This climate backgro und change in the 1990s may cause the asy mmetry of the warm and cold phases. On the ot her hand, the decadal oscillation has directly affected the ENSO cycle on both frequency and amplitude.