DATA REQUIREMENTS FOR SEASONAL DISCHARGE PROGRAMS - AN APPLICATION OF A REGIONALIZED SENSITIVITY ANALYSIS

The design of seasonal water quality management programs that allow different waste discharge rates during different periods of the year may require reliable long-term records of stream conditions. A modified regionalized sensitivity analysis (RSA) is applied for assessing the effect of unreliable stream records on the design of such programs. This approach is demonstrated for a seasonal waste discharge program that controls biochemical oxygen demand in a typical river basin. Here the uncertain input parameters include flow and temperature data at different times in the year and at different locations in the stream. An RSA is conducted for waste management scenarios that have different season length combinations and dissolved oxygen goals. Results indicate that the design of a seasonal waste discharge program for this case study is generally more sensitive to uncertainty in summer flow and temperature data than to uncertainty in winter data. Downstream flow data are shown to be more important than upstream data in low-flow periods, while upstream flows are also important in high-flow periods. In addition, the degree to which uncertain stream conditions affect the management model outcome depends on the water quality goals of the governing agency and the length of the seasons examined.