Spring recession flows are analyzed from a Bayesian point of view. Two general equations are derived and it is shown that the classical formulas of recession flow are particular cases of both equations. It is shown that most of the recession equations reflect a non-Markovian process. That means that the groundwater storage exhibits a memory effect and that there is a nonlinear relationship between flow and storage. The Bayesian approach presented in this paper makes it possible to give a probabilistic meaning to recession flow equations derived according to a physical approach and can be an alternative to the study of complex reservoir for which the physical processes governing recession flow are unclear. Twelve spring recession flow series are analysed in order to validate the probabilistic approach presented in this paper and a conceptual model of storage-outflow is proposed.
展开▼
