The performance of a hydrogen-blended gasoline direct injection engine under various second gasoline direct injection timings is investigated on a modified commercial gasoline direct injection engine. All tests are carried out at a typically congested city-driving condition using lean burn and split gasoline direct injection modes. Results show that under pure gasoline fuel supply mode, brake mean effective pressure, brake thermal efficiency, flame development period, flame propagation period, and maximum in-cylinder pressure vary obviously with various second gasoline direct injection timings and the best second gasoline direct injection timing for engine performance is 130 CAD before top dead center. Blending hydrogen could weaken variations of engine performance related parameters caused by changing second gasoline direct injection timing compared to that under pure gasoline mode. Besides, blending hydrogen could also reduce flame development and propagation periods, and enhance the maximum in-cylinder pressure under the same second gasoline direct injection timing. With hydrogen addition, the coefficient of variations in indicated mean effective pressure drops to less than 1.1%, but second gasoline direct injection timing has no obvious effect on it any more. Besides, adding hydrogen could reduce hydrocarbon and carbon monoxide emissions by 33.10% and 18.28% in average, respectively. Particulate number has a reduction of order of magnitudes (similar to 10(8) to similar to 10(6) n/cm(3) in average). With hydrogen addition, nitrogen oxides emissions increase.