Abstract: The roof-U rib welded joint in orthotropic steel bridge decks has become a fatigue-sensitive detail due to its complex geometric configuration and inherent welding defects. Welding residual stress and corrosive environment are two key factors affecting its fatigue performance, but the coupling mechanism between them has not been fully investigated. Full-scale joint specimens with different residual stress states (with/without annealing treatment) and corrosion degrees (0 d, 40 d, 80 d, and 120 d) were designed. The ultrasonic critical refracted longitudinal wave method was used to non-destructively measure the residual stress distribution. The neutral salt spray test was employed to simulate the corrosive environment, and three-dimensional scanning technology was adopted to quantitatively analyze the morphology of corrosion pits. Fatigue tests based on the hot spot stress method were conducted to monitor the strain response and identify crack initiation and propagation behavior. The results show that, compared with the uncorroded specimen, the equivalent fatigue lives of the high-residual-stress specimens after 40 d, 80 d, and 120 d of corrosion decrease by approximately 40%, 51%, and 56%, respectively. Under the same corrosion period, the residual stress relief increase the equivalent crack initiation life by approximately 7% to 17%. Both welding residual stress and corrosion significantly reduce the fatigue life of the roof-U rib welded joint, and there is an obvious coupling amplification effect between them. The effect of residual stress is to promote early crack initiation by increasing the local mean stress level. Corrosion forms pitting corrosion pits with significant stress concentration and shortens both the crack initiation and propagation stages. Residual stress affects the morphology of corrosion pits, and after severe corrosion, the residual stress field shows a certain degree of release and redistribution, further influencing the fatigue damage evolution mechanism. This study provides important experimental evidence and theoretical support for the fatigue life assessment and durability improvement of in-service orthotropic steel bridge decks.