Submarine cables are key carriers of electrical power and communication signals in submarine cable infrastructure， and the monitoring of such cables is essential to ensure the safety and stable operation of submarine cable infrastructure. In this study， experimental research on the detection of submarine cable disturbances was conducted according to the requirements of submarine cable monitoring. First， a submarine cable disturbance detection experimental system based on distributed optical fiber sensing technology was built and coupled to the dark fiber of the seafloor observation network in the South China Sea. Experiments were performed， and the results indicated that submarine cable disturbances induced by spade beating， vehicles， ocean waves， and typhoons could be detected by the proposed system. Following this， a submarine cable disturbance localization method based on the bidirectional long short-term memory neural network in terms of the submarine cable Rayleigh backscattering （RBS） amplitude signal features was developed， and its input features included the spectral entropy and instantaneous frequency of the submarine cable fiber RBS amplitude signal. These submarine cable fiber RBS amplitude signal samples were used to test the detection accuracy of the proposed method， and the results indicated that the average detection accuracy of disturbance and non-disturbance signals along the submarine cable was >99.6%. Thus， disturbances along submarine cables can be effectively detected with the combination of the proposed submarine cable disturbance detection experimental system and the proposed disturbance localization method.