Near future prediction of rainfall distribution pattern by Adaptive Neuro Fuzzy Interference System in Central Java, Indonesia

Abstract

Rainfall distribution pattern is very important for sustainable production in the agricultural sector in Tropical region. The near future rainfall amount under changing climate was predicted from 2009 to 2028 by Adaptive Neuro Fuzzy Inference System (ANFIS), which was trained with the rainfall observation data from 1979 to 2013, spatiotemporally in the Central Java Province, Indonesia. Our analysis showed that the predicted rainfall data using ANFIS can represent actual rainfall conditions. Rainfall predicted from 2009 – 2028 in Central Java will experience a decrease in high rainfall in an area of 1615125,2 hectare, which can cause drought. The area that predicted to experience drought in the future are Kebumen, Jepara, Pati, Rembang, Kudus, Grobogan, Blora, Sragen, and parts of the regions of Purworejo, Cilacap, Banyumas, Brebes, Wonogiri, Karanganyar, and Demak.