ABSTRACT
The assessment of extreme hydrological events relies on the assumption that time series are independent and identically distributed. However, empirical evidence contradicts this assumption, indicating the presence of nonstationarity, including trends, step changes, or both, in the river records of the Argentinean Central Andes. This study proposes the use of generalized additive models in location, scale and shape (GAMLSS) to model the annual maximum and minimum flow of seven rivers. Firstly, the presence of trends and step changes was assessed. Subsequently, the models were formulated considering time and various climatic indices as covariates. The findings reveal declining trends and negative jumps over the past 50 to 70 years. The location parameters of the models exhibit linear or smoothing dependence with time or climate indices covariates. Different modes of climate variability are associated with hydrological extremes. The nonstationary models provide a novel complementary framework for water resource planning in the region.
Editor S. Archfield; Associate Editor M. Batalini de Macedo
Editor S. Archfield; Associate Editor M. Batalini de Macedo
Disclosure statement
No potential conflict of interest was reported by the author(s).
Data availability statement
Daily streamflow series: https://snih.hidricosargentina.gob.ar/Inicio.aspx
Sea surface temperature in the Equatorial Pacific Ocean: Monthly ERSSTv5 (1991–2020 base period) Niño 1+2 (0–10°S)(90–80°W) Niño 3 (5°N–5°S)(150–90°W) Niño 4 (5°N–5°S) (160°E–150°W) Niño 3.4 (5°N–5°S)(170–120°W)
https://www.cpc.ncep.noaa.gov/data/indices/ersst5.nino.mth.91-20.ascii
PDO index: http://research.jisao.washington.edu/pdo/PDO.latest
TPI index: https://psl.noaa.gov/data/timeseries/IPOTPI/Tripole