NESTED THRESHOLD AUTOREGRESSIVE (NETAR) MODELS FOR STUDYING SOURCES OF NONLINEARITY IN STREAMFLOWS

Historical development of nonlinear stochastic modeling of streamflows is discussed. Physical considerations and graphical investigations of daily streamflows revealed that air temperature and state of basin st orage are the most important sources of nonlinearity in catchments wit h seasonal snow accumulation. The average temperature for the precedin g three days, and the flow one or two days earlier were adequate proxi es for the temperature and storage conditions of a catchment. The Nest ed Threshold Autoregressive (NeTAR) model, which considers these sourc es of nonlinearity simultaneously, was applied to two years of daily f lows of the Oldman River near Brocket in Alberta, Canada. A third year of daily data was used for validating the model. The final NeTAR mode l provided useful insights into the dynamics of this streamflow system .