Temporal evolution of hydroclimatic teleconnection and a time-varying model for long-lead prediction of Indian summer monsoon rainfall

Automated predictive analytics tool for rainfall forecasting

Australia faces a dryness disaster whose impact may be mitigated by rainfall prediction. Being an incredibly challenging task, yet accurate prediction of rainfall plays an enormous role in policy making, decision making and organizing sustainable water resource systems. The ability to accurately predict rainfall patterns empowers civilizations. Though short-term rainfall predictions are provided by meteorological systems, long-term prediction of rainfall is challenging and has a lot of factors that lead to uncertainty. Historically, various researchers have experimented with several machine learning techniques in rainfall prediction with given weather conditions. However, in places like Australia where the climate is variable, finding the best method to model the complex rainfall process is a major challenge. The aim of this paper is to: (a) predict rainfall using machine learning algorithms and comparing the performance of different models. (b) Develop an optimized neural network and develop a prediction model using the neural network (c) to do a comparative study of new and existing prediction techniques using Australian rainfall data. In this paper, rainfall data collected over a span of ten years from 2007 to 2017, with the input from 26 geographically diverse locations have been used to develop the predictive models. The data was divided into training and testing sets for validation purposes. The results show that both traditional and neural network-based machine learning models can predict rainfall with more precision.

Analysis on the Characteristics of Dry and Wet Periods in The Yangtze River Basin

As China’s main grain producing region, the Yangtze River basin is vulnerable to changes in wet and dry conditions. In this study, the monthly scale of standardized precipitation evapotranspiration index (SPEI) was calculated, based on the Penman–Monteith equation from 239 meteorological stations in the Yangtze River basin, from 1960 to 2017. Water regime characteristic areas of the Yangtze River basin were extracted and divided using the rotating empirical orthogonal function (REOF). The linear trend of the drought and wetness indicators, the abrupt changes of the rotated principal component time series (RPCs), and the change periods of the drought/wetness intensity (DI/WI) in each subregion were analyzed and discussed. Subsequently, the effects of El Niño-southern oscillation (ENSO) and arctic oscillation (AO) on drought and wetness events were discussed. The results showed that the Yangtze River basin has the characteristic of coexistence of drought and wetness, and drought and wetness of similar severity tend to occur in the same region. There were six subregions extracted through REOF, based on the monthly scale of SPEI, of which the northwestern pattern had an aridization tendency. The stations with significantly increased wetness were located in the middle and eastern basin. The stations in the south of the northwestern pattern, and the west of the southern pattern, had a tendency of wetting in the first 29 years, however, there has been a significant tendency of drying in this region in the last 29 years, which was caused by an abrupt change in 1994. In addition, other patterns had multiple abrupt changes, resulting in multiple transitions between dry and wet states. The principal periods of WI in the southern pattern and northern pattern were longer than the DI, but in other subregions DI was longer than WI. ENSO and AO had the most obvious influence on DI and WI. Compared with the cold phase of ENSO, the DI/WI in the warm phase were higher/lower; compared with the negative phase of AO, both DI and WI were higher in the positive phase. The Hurst index showed that the current dry and wet conditions in the Yangtze River basin have persistent characteristics, the dry conditions in each subregion will continue in the future, and there were a few wetness indicators with weak anti-persistence.