The purpose of this study is to develop a linear reservoir model with Kalman filter using Kalman filter theory which removes a physical uncertainty of rainfall-runoff process. A linear reservoir model, which is the basic model of Kalman filter, is used to calculate runoff from rainfall in river basin. A linear reservoir model with Kalman filter is composed of a state-space model using a system model and a observation model. The state-vector of system model in linear reservoir model with Kalman filter takes the ordinate of IUH from linear reservoir model. The average value of the ordinate of IUH for a linear reservoir model with Kalman filter is used as the initial value of state-vector. A linear reservoir model with Kalman filter shows better results than those by linear reserevoir model, and decreases a physical uncertainty of rainfall-runoff process in river basin.