Feasibility of forecasting duration extreme precipitation indices based on teleconnection phenomena. (case study: Gorgan and Rasht meteorology stations)

Articles in Press

Document Type : Original Article

Authors

1 Faculty member of Gorgan University of Agricultural Sciences and Natural Resources

2 Department of Water Engineering, Gorgan University of Agricultural Sciences and Natural Resource, Gorgan, Iran.

3 Department of Water Engineering, Gorgan University of Agricultural Sciences and Natural Resource, Gorgan, Iran

4 Department of Water Engineering, Faculty of Agriculture, Urmia University, Urmia, Iran.

5 Department of Desert Management, Faculty of Water and Soil Engineering, Gorgan University of Agricultural Sciences and Natural Resource, Gorgan, Iran.

10.22125/agmj.2024.436966.1164

Abstract

The study of extreme precipitation events is important due to the damage they cause. Teleconnection phenomena and their delayed effect on climate change can be suitable variables for forecasting. To investigate the possibility of forecasting consecutive dry and wet days, extreme precipitation indices were considered at two meteorological stations in Rasht and Hashem Abad, Gorgan. This study was conducted to explore the relationship between 24 teleconnection variables and the Extreme Precipitation indices during the statistical period of 1402-1362. Multivariate linear regression (MLR) and the M5 decision tree regression were employed to examine the linear and nonlinear relationships between the variables, respectively. The stepwise implementation of the M5 model was carried out to prevent greediness and to identify the most effective variables. It was found that the teleconnection indices had a delayed effect, which was confirmed by their higher correlation with extreme precipitation indices. The higher accuracy of the M5 model indicated a non-linear relationship between the teleconnection indices and the extreme precipitation indices. The M5 model predicted the extreme precipitation indices with a percentage error of less than 21%. The greater accuracy of the M5 model with only 5 variables highlights the importance of these variables and the greediness of the M5 algorithm. In conclusion, teleconnection indices have an impact on climate change. However, relying on just one index cannot explain all changes. To forecast extreme events, it is necessary to consider the combined effect of several indices and use a suitable model to determine their relationship with extreme indices.

Keywords

Main Subjects

Journal of Agricultural Meteorology

Articles in Press, Accepted Manuscript
Available Online from 07 March 2024

Files

Share

How to cite

Statistics