Spring rainfall prediction based on remote linkage controlling using adaptive neuro-fuzzy inference system (ANFIS)

Abstract

This paper aims to study the relationship between large-scale synoptic patterns and rainfall in Khorasan Razavi Province. The adaptive neuro-fuzzy inference system (ANFIS) was used in this study to predict rainfall in the period between April and June in Khorasan Razavi Province. We first analyzed the relationship between average regional rainfall and the changes in synoptic patterns including sea-level pressure, sea-level pressure difference, sea-level temperature, temperature difference between sea level and the 1,000-hPa level, the temperature of the 700-hPa level, the thickness between the 500- and 1,000-hPa levels, the relative humidity at the 300-hPa level, and precipitable water content. We have examined the effect of synoptic patterns in these regions on the rainfall in the northeast region of Iran. Then, the ANFIS in the period 1970–1997 has been taught. Finally, we forecast the rainfall for the period 1998–2007. The results show that the ANFIS can predict the rainfall with reasonable accuracy.