作者: Gabriele Villarini , Francesco Serinaldi , James A. Smith , Witold F. Krajewski
DOI: 10.1029/2008WR007645
关键词: Flood myth 、 Climate change 、 Water cycle 、 Generalized additive model 、 Catchment hydrology 、 100-year flood 、 Hurst exponent 、 Climatology 、 Geology 、 Drainage basin
摘要: [1] Annual peak discharge records from 50 stations in the continental United States with at least 100 years of record are used to investigate stationarity flood peaks during 20th century. We examine temporal trends and abrupt changes mean and/or variance distributions. Change point analysis for detecting distributions is performed using nonparametric Pettitt test. Two (Mann-Kendall Spearman) tests one parametric (Pearson) test detect presence trends. Generalized additive models location, scale, shape (GAMLSS) also parametrically model annual data, exploiting their flexibility account parameters distribution functions. Additionally, long-term persistence investigated through estimation Hurst exponent, an alternative interpretation results terms provided. Many drainage basins represented this study have been affected by regulation systems reservoirs, all experienced significant land use Despite profound that occurred throughout recognition elements hydrologic cycle being altered human-induced climate change, it easier proclaim demise than prove analyses data.