The increasing effort to develop and apply nonstationary models in hydrologic frequency analyses under\nchanging environmental conditions can be frustrated when the additional uncertainty related to the\nmodel complexity is accounted for along with the sampling uncertainty. In order to show the practical\nimplications and possible problems of using nonstationary models and provide critical guidelines, in this\nstudy we review the main tools developed in this field (such as nonstationary distribution functions,\nreturn periods, and risk of failure) highlighting advantages and disadvantages. The discussion is supported\nby three case studies that revise three illustrative examples reported in the scientific and technical\nliterature referring to the Little Sugar Creek (at Charlotte, North Carolina), Red River of the North (North\nDakota/Minnesota), and the Assunpink Creek (at Trenton, New Jersey). The uncertainty of the results is\nassessed by complementing point estimates with confidence intervals (CIs) and emphasizing critical\naspects such as the subjectivity affecting the choice of the models� structure. Our results show that (1)\nnonstationary frequency analyses should not only be based on at-site time series but require additional\ninformation and detailed exploratory data analyses (EDA); (2) as nonstationary models imply that the\ntime-varying model structure holds true for the entire future design life period, an appropriate modeling\nstrategy requires that EDA identifies a well-defined deterministic mechanism leading the examined process;\n(3) when the model structure cannot be inferred in a deductive manner and nonstationary models\nare fitted by inductive inference, model structure introduces an additional source of uncertainty so that\nthe resulting nonstationary models can provide no practical enhancement of the credibility and accuracy\nof the predicted extreme quantiles, whereas possible model misspecification can easily lead to physically\ninconsistent results; (4) when the model structure is uncertain, stationary models and a suitable assessment\nof the uncertainty accounting for possible temporal persistence should be retained as more theoretically\ncoherent and reliable options for practical applications in real-world design and management\nproblems; (5) a clear understanding of the actual probabilistic meaning of stationary and nonstationary\nreturn periods and risk of failure is required for a correct risk assessment and communication.
Loading....