Drought forecasting in eastern Australia using multivariate adaptive regression spline, least square support vector machine and M5Tree model

详细信息    查看全文

• 作者：Ravinesh C Deo^a ; ^{ravinesh.deo@usq.edu.au} ; Ozgur Kisi^b ; Vijay P Singh^c ; ^d
• 关键词：Standardized precipitation index ; Drought forecasting ; Multivariate adaptive regression spline ; Least square support vector machine ; M5Tree model
• 刊名：Atmospheric Research
• 出版年：2017
• 出版时间：1 February 2017
• 年：2017
• 卷：184
• 期：Complete
• 页码：149-175
• 全文大小：4665 K
• 卷排序：184

文摘

Standardized Precipitation Index was modelled using MARS, LSSVM and M5Tree with rainfall, climate indices and SST as predictor variables. MARS and M5Tree models outperformed LSSVR, and highlighted the importance of periodicity as a predictor variable for SPI-modelling. Drought forecasting was dependent on proper combination of predictor variables and scaled with the geographic location of study sites.