The impact of El Nño-southern oscillation on the rainfall temporal variability in the Piquiri watershed, Paraná state, Brazil

This research was developed considering the impacts caused by El Niño - Southern Oscillation (ENSO) on the variability of precipitation. It presents a suggestion to use an inferential statistical model to identify the participation of such phenomenon on rainfall temporal variability in the Piquiri watershed (Paraná State, Brazil), based on the Southern Oscillation Index (SOI). The ENSO phenomenon was analyzed using Southern Oscillation Index data provided by the Australian National Climate Centre and monthly rainfall data from 41 rain gauge stations in the period from 1976 to 2010, provided by Instituto das Águas do Paraná. A total of 31.4% of the analyzed historical series showed a negative Southern Oscillation Index (≤ –7), while 21.1% presented a positive Southern Oscillation Index (≥ +7), and 47.5% of the months were considered neutral regarding the Southern Oscillation Index. The Generalized Additive Models for Location, Scale and Shape (GAMLSS) were applied to identify the influence of ENSO on monthly rainfall average. A lag of up to one month was observed between Southern Oscillation Index deviations and rainfall in the Piquiri river watershed. Average monthly rainfall was found to decrease when each Southern Oscillation Index unit increases in the inferential model. This decrease varied between the three drainage areas considered in the study. (author)