Using the Association Rules to Forecast the Maximum Monthly Precipitation of Tabriz Synoptic Station

Long-term forecasting of maximum monthly precipitation (MMP) is very important for a variety purposes such as flood and runoff forecasting, irrigation scheduling and watershed management. In this study, the application of data mining technique (association rules) is offered to discover affiliation between MMP of Tabriz synoptic station and sea surface temperatures (SST) of the Black, Mediterranean and Red Seas considering the different lags. Data mining is a technology which helps extracting hidden predictive information from large data bases and thus it facilitates that decision makers to make proactive, knowledge-driven decisions. To examine the accuracy of the rules, support and confidence measures were calculated. The results show a relative correlation between the Mediterranean, Black and Red Sea SSTs and MMP of Tabriz synoptic station so that the confidence between the MMP values and the SST of seas is approximately 70 percent. Also the results indicate that the combination of the Mediterranean, Black and Red Sea SSTs to forecast the MMP of Tabriz synoptic station have a better performance.