Operations of reservoir groups face significant challenges due to uncertainties in inflow and water demand arising from various risk factors, including extreme climate, demand fluctuations, and operational experience. Traditional risk operations have rarely addressed uncertainties from multiple sources over current and future periods comprehensively. This study innovatively applies economic Newsboy theory to reservoir risk operations to fill this gap. Specifically, taking the reservoir group in the Pearl River Basin as the research object and comprehensively considering multiple risk sources from inflow uncertainty and demand uncertainty in both current and subsequent operating periods, a Newsboy-based model of real-time risk operation is established, and the influence law of multiple risk sources on the real-time salty-tide risk process of reservoirs in this basin is then revealed. The results demonstrate that the optimized risk process, fluctuating between 0 and 0.053, offers increased stability and lower risk compared to the measured process, which fluctuates between 0 and 0.093. Under single risk source with uncertain inflow, the greater the runoff forecast error, the greater the risk of suppressing salt tide at Wuzhou section. Under single risk source with uncertain water demand, with the increase of inflow, opportunity risk of suppressing salty tide gradually changes into excess risk. And the inflow corresponding to the minimum risk is 2200 m3/s at Wuzhou section. Under multiple risk sources with uncertain inflow and demand, the risk increases significantly compared with single risk source, and the impact of demand uncertainty is more significant on risk. Specifically, multi-source risk is 1.78 times of the single-source risk with inflow uncertainty, and is 1.07 times of the single-source risk with demand uncertainty. The study findings advance real-time risk operation theories and methodologies for reservoirs.