Shanghai sea dikes are built to protect shorelines, factories, inland buildings and other vulnerable coastal areas against wave action and storm surge. For this reason, there is an urgent need to check whether sea dike in Shanghai can function well. The intuitive influences are sea level rise, severe storm surge, and land subsidence, which affect sea dike functions. Sea dikes should withstand many failure mechanisms, and the objective of this research is to investigate failure probability for the 3 most important failure mechanisms of a specific sea dike in Shanghai Pudong under present and future climate conditions, which are overtopping, revetment failure and macro instability. The applied rates for absolute sea level rise in this research are 2.5 mm/yr from 2011 to 2030 and 5.0 mm/yr from 2031 to 2100 respectively. Therefore, sea level rise in 2100 is 375mm from 2019. Land subsidence rate in Shanghai is estimated at 6 mm/year after 2010, so the subsidence value is 486 mm till 2100 from 2019. Based on previous research, 38 severe typhoon events from 1990s are chosen as reference events to study severe wave conditions in Shanghai. For Shanghai, a severe typhoon event means the maximum average wind speed near the center of bottom layer is larger than 41.5 m/s, and it lasts more than 24 hours in the area near Shanghai (25° N-35° N and 115° E-125° E). These events are simulated with the software package Delft3D, which is an important tool for modeling of severe wave conditions under current and future circumstances. Characteristics of wind and pressure fields for typhoons are important inputs for Delft3D to model wave characteristics. As a result, from a Delft3D a simulated record of water level, peak period and significant wave height during a typhoon event is obtained, and the maximum overtopping discharge and maximum significant wave height during a storm event can then be analysed. For future Shanghai boundary condition study, sea level rise and land subsidence value are considered directly in Delft3D. Boundary conditions around a sea dike are an important role when investigating the dike failure probability. Both in current and future conditions, uncertainty in maximum overtopping discharge can be described with Weibull distribution, maximum significant wave height can be described with a lognormal distribution. The predicted distributions are tested using the Kolmogorov-Smirnov (K-S) test. Based on fitted maximum overtopping discharge and maximum significant wave height, their probability density function is applied to calculate failure probability of overtopping and revetment failure. Failure is considered to be the probability that the overtopping discharge exceeds the critical value, which is defined mathematically by the limit state function. For overtopping and revetment failure mechanism, a probabilistic analysis called Monte Carlo simulation is used. Random samples of parameters are generated and are used with the limit state function to determine whether failure would occur. Fragility curves are constructed for each mechanism. A simplified method, called FOSM (first order second moment), is used to study the relative changes of macro instability between now and in the future. Although not all failure mechanisms for the Pudong sea dike are taken into account, this study proves that the lower bound of Shanghai sea dike failure probability is 0.31 per year in the condition of severe storm events in the year 2100. However, under current conditions, the lower bound of failure probability of Pudong sea dike is 0.23 per year. In terms of these three main failure mechanisms, revetment and overtopping mechanisms contribute more to Pudong sea dike failure.