An autoregressive error model has been tested on the residuals of the conceptual HBV-model for the Emån catchment. The autoregressive model gives considerable improvements for real shorttime forecasting, but for long range (10 days or more) forecasting no improvement is achieved compared to the conceptual model. Separation of the error functions for high and low discharges does not give any further improvement.

This content is only available as a PDF.
You do not currently have access to this content.