This article describes the design of distributed Model Predictive Control (MPC) strategies based on a linear model of a water delivery canal, composed of three pools, each ending with a gate. The distributed MPC control algorithm relies on a Distributed Alternating Direction Method of Multipliers (D-ADMM) optimization framework. A non-linear model of the canal is used to identify a linearized model around an equilibrium point, which is then used to design the controllers. Since the plant states are not accessible, a Kalman filter is used to estimate them. Integral action is also included together with an anti-windup algorithm. Simulations using the nonlinear plant model are presented to show the impact of control algorithm parameters on performance.