In area of water resources management, decision-makers usually need to make plans under various uncertainties in order to achieve the maximal total net benefits and prevent possible losses under risks arising from the indeterminacy. In this paper, we have incorporated the conditional value-at-risk, which has been widely used in portfolio selection problems for effective risk management, in the interval two-stage stochastic programming framework to perform trade-offs between economic objectives and associated risks. Moreover, based on the late progresses in Grey linear system, we have developed a new method to obtain a solution of the proposed model. In particular, this new approach can provide a number of decision alternatives under different coefficients, which help decision-makers make proper water supply plans according to their personal preferences. Finally, numerical experiments have been presented to demonstrate the validity and applicability of the proposed model for maintaining balance between profits and risks, as well as the efficiency of the new method for dealing with such model.