A hybrid spatio-temporal modelling: an application to space-time rainfall forecasting
- PDF / 950,966 Bytes
- 12 Pages / 595.276 x 790.866 pts Page_size
- 4 Downloads / 160 Views
ORIGINAL PAPER
A hybrid spatio-temporal modelling: an application to space-time rainfall forecasting Amit Saha 1 & K. N. Singh 2 & Mrinmoy Ray 2 & Santosha Rathod 3 Received: 6 March 2020 / Accepted: 3 September 2020 # Springer-Verlag GmbH Austria, part of Springer Nature 2020
Abstract Efficient and reliable forecasting techniques for various climatic conditions are indispensable in agricultural dependent country like India. In this context, rainfall forecasting is one of the most challenging tasks because of the existence of three patterns, viz., temporal, spatial, and non-linear, simultaneously. Space-Time Autoregressive Moving Average (STARMA) model is one of the promising and popular approaches for modelling spatio-temporal time series data. However, the observed features of many space-time rainfall data comprise complex non-linear dynamics and modelling these patterns often go beyond the capability of conventional STARMA model. Moreover, despite the popularity of artificial neural network (ANN) and support vector machine (SVM) for modelling complex non-linear dynamics, they are not capable to deal with spatial patterns. To overcome the problem, a new spatio-temporal hybrid modelling approach has been proposed by integrating STARMA, ANN, and SVM as well. The proposed approach has been empirically illustrated on annual precipitation data of six districts of northern part of West Bengal, India. The study reveals that proposed spatio-temporal hybrid approach has better modelling and forecasting precision over conventional STARMA as well as most widely used Autoregressive Integrated Moving Average (ARIMA) model. Keywords ANN∙ . Hybrid model . Rainfall forecasting . STARMA . SVM
1 Introduction Forecasting rainfall in a systematic and scientific way can play a crucial role for better agricultural management and disaster mitigation in the developing countries like India. Fluctuations in rainfall influence agriculture, drinking water, energy sector, and livelihood of a huge number of individuals living in the nation. Consequently, rainfall forecasting has become a challenging area in the parlance of time series forecasting as its high changeability has noteworthy effect on the economy of the nation. There are various statistical models available in the literature for time series analysis. The most popularly employed time series model is Autoregressive Integrated
* Amit Saha [email protected] 1
CSRTI, Central Silk Board, GOI, Mysore, India
2
ICAR-Indian Agricultural Statistics Research Institute, New Delhi, India
3
ICAR-Indian Institute of Rice Research, Hyderabad, Hyderabad, India
Moving Average (ARIMA) (Box and Jenkins 1970). There are instances of application of ARIMA model for forecasting hydrological and meteorological phenomena (Leite et al. 1996; Tektay 2010). However, there are two main disadvantages of ARIMA model. Firstly, it is a univariate model; hence, it cannot deal with spatial interrelationships existing in the time series data and secondly is the assumption of linearity. Therefore, if the
Data Loading...
