干旱区生态环境知识资源中心(Arid): Ability Assessment of the Stationary and Cyclostationary Time Series Models to Predict Drought Indices

Arid

DOI	10.1007/s11269-020-02710-5
	Ability Assessment of the Stationary and Cyclostationary Time Series Models to Predict Drought Indices
	Zarei, Abdol Rassoul; Mahmoudi, Mohammad Reza
通讯作者	Zarei, AR
来源期刊	WATER RESOURCES MANAGEMENT
ISSN	0920-4741
EISSN	1573-1650
出版年	2020
卷号	34 期号:15 页码:5009-5029
英文摘要	Drought forecasting and monitoring play a significant role in reducing the negative effects of global meteorological droughts caused by different intensities at different temporal and spatial scales in different regions, especially in regions with high dependency on rainwater. The present study tries to compare the accuracy of stationary time series (ST) models including autoregressive moving average (ARMA), moving average (MA) and autoregressive (AR) and cyclostationary time series (CT) models including periodic autoregressive moving average (PARMA), periodic moving average (PMA) and periodic autoregressive (PAR) to predict drought index (i.e. monthly reconnaissance drought index (RDI)) in periodic data series considering that CT models are more powerful and efficient than ST models by using data series of 8 synoptic stations with different climate conditions in Iran from 1967 to 2017. According to the results the monthly RDI was significantly periodic in all selected stations. The PAR (25) model was the best fitted CT model in data series at all stations and on the other hand, the following models were the best-fitted ST models in data series: the AR models at Babolsar and Rasht AR (25) and at Gorgan AR (24) and ARMA models at Tehran ARMA (2, 3), at Zahedan and Shiraz ARMA (2, 4) and at Esfahan and Shahre Kord ARMA (2, 5). Based on the best fitted CT and ST models, the results showed that the correlation coefficients (R) between observed and simulated RDI vary from 0.882 to 0.946 and from 0.693 to 0.874, respectively from January 1967 to December 2017. According to the best fitted CT and ST models, the validation test of the best fitted models indicated that the R between observed and simulated RDI vary from 0.634 to 0.883 and 0.585 to 0.847, respectively from January 2012 to December 2017. In total, it can be concluded that that the accuracy and capability of CT models in predicting the RDI were more than those of the ST models at all stations and the hypothesis of the study was confirmed.
英文关键词	ST models CT models RDI ARMA PARMA
类型	Article
语种	英语
收录类别	SCI-E
WOS记录号	WOS:000587666900001
WOS关键词	METEOROLOGICAL DROUGHT ; ARID REGIONS ; IMPACTS ; SPI
WOS类目	Engineering, Civil ; Water Resources
WOS研究方向	Engineering ; Water Resources
资源类型	期刊论文
条目标识符	http://119.78.100.177/qdio/handle/2XILL650/327412
作者单位	[Zarei, Abdol Rassoul] Fasa Univ, Dept Range & Watershed Management Nat Engn, Coll Agr Sci, Fasa, Iran; [Mahmoudi, Mohammad Reza] Fasa Univ, Dept Stat, Fac Sci, Fasa, Iran; [Mahmoudi, Mohammad Reza] Duy Tan Univ, Inst Res & Dev, Da Nang 550000, Vietnam
推荐引用方式 GB/T 7714	Zarei, Abdol Rassoul,Mahmoudi, Mohammad Reza. Ability Assessment of the Stationary and Cyclostationary Time Series Models to Predict Drought Indices[J],2020,34(15):5009-5029.
APA	Zarei, Abdol Rassoul,&Mahmoudi, Mohammad Reza.(2020).Ability Assessment of the Stationary and Cyclostationary Time Series Models to Predict Drought Indices.WATER RESOURCES MANAGEMENT,34(15),5009-5029.
MLA	Zarei, Abdol Rassoul,et al."Ability Assessment of the Stationary and Cyclostationary Time Series Models to Predict Drought Indices".WATER RESOURCES MANAGEMENT 34.15(2020):5009-5029.