GAMLSS-based nonstationary modeling of extreme precipitation in Beijing-Tianjin-Hebei region of China.

Due to the climate variability and the intensification of human activities, the hydrological time series no longer satisfies the hypothesis of stationarity. In this study, a framework for precipitation frequency analysis is developed based on the Generalized Additive Models for Location, Scale and Shape parameters (GAMLSS), a tool for modeling time series under nonstationary condition. Based on the 12 stations in Beijing-Tianjin-Hebei region of China, two approaches to nonstationary modeling in GAMLSS were applied to the annual maximum daily precipitation records. The results of the first approach, in which the parameters of the selected distributions are modeled as a function of time only, show the presence of clear nonstationarities in the annual maximum daily precipitation. In the second approach, the parameters of the precipitation distributions are modeled as functions of seven climate indices. The results show that the model using the second method captures more adequately the dispersion of precipitation values than the model using the first method. The application of nonstationary analysis shows the differences between the nonstationary quantiles and their stationary equivalents, which suggests the urgent need for nonstationary modeling of extreme precipitation in Beijing-Tianjin-Hebei region of China. [ABSTRACT FROM AUTHOR]