收藏 ｜ 设为首页 广告服务 ｜ 联系我们 ｜ 在线留言
高级检索

干旱区地理 >

2020 , Vol. 43 >Issue 4: 1004 - 1013

DOI: https://doi.org/10.12118/j.issn.1000-6060.2020.04.15

地球信息科学

ARIMA-LSTM组合模型在基于SPI干旱预测中的应用——以青海省为例

展开
  • 1 青海水文水资源勘测局,青海 西宁 810000;
    2 华北水利水电大学,河南 郑州 450000
张建海（1968–）,男,高级工程师,学士,研究方向为水利工程与水文水资源. E-mail：1291065549@qq.com

收稿日期: 2019-11-05

  修回日期: 2020-06-08

  网络出版日期: 2020-11-18

基金资助

国家自然基金（51679089、51609082、51709107）资助

收起

Application of a combined ARIMA-LSTM model based on SPI for the forecast of drought：A case study in Qinghai Province

Expand
  • 1 Qinghai Hydrology and Water Resources Survey Bureau,Xining 810000,Qinghai,China;
    2 North China University of Water Resources and Electric Power,Zhengzhou 450000,Henan,China

Received date: 2019-11-05

  Revised date: 2020-06-08

  Online published: 2020-11-18

Fold

摘要

开展干旱预测是有效应对干旱风险的前提基础。利用1958—2017年青海省38个气象站点逐日降水量数据计算多尺度标准化降水指数（SPI）,并建立了SPI序列自回归移动平均模型（ARIMA）、长短时记忆神经网络模型（LSTM）和基于二者优点提出的ARIMA-LSTM组合模型;对模型参数进行率定和验证后,利用所建立的模型,以西宁站点为例,对多尺度SPI值进行预测,借助均方根误差（RMSE）、平均绝对百分比误差（MAPE）和决定系数R2对所有预测模型的有效性进行判定。结果表明：ARIMA-LSTM组合模型在SPI1和SPI12的RMSE值分别为0.159 7和0.181 0,均低于ARIMA模型的1.265 4和0.293 3,说明ARIMA模型与ARIMA-LSTM组合模型对SPI的预测精度都与时间尺度有关,ARIMA模型的预测精度随着时间尺度的增加而逐渐提高;结合GIS并利用实测数据与模型的预测数据相比较说明ARIMA-LSTM组合模型相比于单一ARIMA模型的预测精度更高,且能够很好拟合不同时间尺度的SPI值。

关键词： 干旱预测; SPI; ARIMA-LSTM组合模型; 青海省

本文引用格式

张建海, 张棋, 许德合, 丁严 . ARIMA-LSTM组合模型在基于SPI干旱预测中的应用——以青海省为例[J]. 干旱区地理, 2020 , 43(4) : 1004 -1013 . DOI: 10.12118/j.issn.1000-6060.2020.04.15

Abstract

Drought prediction is a precondition to effectively mitigate the risk of drought. Daily precipitation data obtained from 38 meteorological stations in Qinghai Province,China in the period from 1958 to 2017 were used to calculate the multiscale standardized precipitation index （SPI）. In addition,based on these data,the SPI sequence Autoregressive Moving Average model （ARIMA）,Long Short-Term Memory model （LSTM）,and ARIMA-LSTM combination model were constructed. After the calibration and verification of the model parameters,the model was used to predict multiscale SPI values using the Xining area as a case study. Moreover,the validity of all the prediction models was determined by root mean square error （RMSE） and mean absolute percentage error （MAPE）. The results indicated that the RMSE values of the ARIMA-LSTM combined model in SPI1 and SPI12 were 0.159 7 and 0.181 0,respectively,which were lower than those （1.265 4 and 0.293 3） of the ARIMA model. This indicates that the prediction accuracy of the ARIMA and LSTM models for SPI was related to the timescale. Comparing the measured data （using GIS） to the data predicted by the models,the combined ARIMA-LSTM model exhibited a higher prediction accuracy compared to the single ARIMA model. In addition,the combined ARIMA-LSTM model showed an ability to fit the SPI values of different timescales.

Key words： drought forecast; SPI; ARIMA-LSTM; Qinghai Province

参考文献

[1] 高涛涛,殷淑燕,王水霞. 基于SPEI指数的秦岭南北地区干旱时空变化特征[J]. 干旱区地理,2018,41(4):761–770.
[GAO Taotao,YIN Shuyan,WANG Shuixia.Spatial and temporal variations of drought in northern and southern regions of Qinling Mountains based on standardized precipitation evapotranspiration index[J]. Arid Land Geography,2018,41(4):761–770.]
[2] 张乐园,王戈,陈亚宁. 基于SPEI指数的中亚地区干旱时空分布特征[J]. 干旱区研究,2020,37(2):331–340.
[ZHANG Leyuan,WANG Ge,CHEN Yanning.Spatial and temporal distribution characteristics of drought in Central Asia based on SPEI index[J]. Arid Zone Research,2020,37(2):331–340.]
[3] 李凤霞,伏洋,张国胜,等. 青海省干旱预警服务系统设计与建立[J].干旱地区农业研究,2004,22(1):1–5.
[LI Fengxia,FU Yang,ZHANG Guosheng,et al.The design and establishment of drought information service system in Qinghai Province[J]. Agricultural Research in the Arid Areas,2004,22(1):1–5.]
[4] 袁文平,周广胜. 干旱指标的理论分析与研究展望[J]. 地球科学进展,2004,19(6):982–991.
[YUAN Wenping,ZHOU Guangsheng.Theoratical study and research prospect on drought indices[J]. Advances in Earth Science,2004,19(6):982–991.]
[5] 沈国强,郑海峰,雷振峰. SPEI指数在中国东北地区干旱研究中的适用性分析[J].生态学报,2017,37(11):3787–3795.
[SHEN Guoqiang,ZHANG Haifeng,LEI Zhenfeng.Applicability analysis of SPEI for drought research in northeast China[J]. Acta Ecologica Sinica,2017,37(11):3787–3795.]
[6] 张菡,张喜亮,李金建,等. 基于SPEI的四川省盆地区季节性干旱时空变化特征分析[J]. 干旱地区农业研究,2018,36(5):242–250.
[ZHANG Han,ZHANG Xiliang,LI Jinjian,et al.SPEI-based analysis of temporal and spatial variation characteristics for seasonal drought in Sichuan Basin[J]. Agricultural Research in the Arid Areas,2018,36(5):242–250]
[7] 容锦盟,周丹,罗静,等. 4种干旱指标在华北地区气象干旱监测中的适用性分析[J].干旱地区农业研究,2019,37(1):295–276.
[RONG Jinmeng,ZHOU Dan,LUO Jing,et al.Applicability analysis of four drought indices for meteorological drought monitoring in Northern China[J]. Agricultural Research in the Arid Areas,2019,37(1):295–276.]
[8] 刘庚山,郭安红,安顺清,等. 帕默尔干旱指标及其应用研究进展[J]. 自然灾害学报,2004,13(4):21–27.
[LIU Gengshan,GUO Anhong,AN Shunqing,et al.Research progress in Palmer drought severity index and its application[J]. Journal of Natural Disasters,2004,13(4):21–27.]
[9] VASILIADES L,LOUKAS A,LIBERIS N.A water balance derived drought index for Pinios River Basin,Greece[J]. Water Resource Manage,2011,25:1087–1101.
[10] 林盛吉,许月萍,田烨,等. 基于Z指数和SPI指数的钱塘江流域干旱时空分析[J].水力发电学报,2012,31(2):20–26.
[LIN Shengji,XU Yueping,TIAN Ye,et al.Spatial and temporal analysis of drought in Qiantang River basin based on Z index and SPI[J]. Journal of Hydroelectricity,2012,31(2):20–26.]
[11] 郭伟,李莹,杜莉丽,等. 基于SPI的山西省1972—2012年春夏干旱特征及对玉米产量的影响分析[J]. 干旱地区农业研究,2018,36(1):230–236.
[GUO Wei,LI Ying,DU Lili,et al.Characteristic of spring and summer drought variations and its relation with maize yield in Shanxi Province in 1972—2012 based on SPI[J]. Agricultural Research in the Arid Areas,2018,36(1):230–236.]
[12] MCKEE T B,DOESKEN N J,KLEIST J.The relationship of drought frequency and duration to time scales[R]. Eighth Conference on Applied Climatology,American Meteorological,1993.
[13] 杨慧荣,张玉虎,崔恒建,等. ARIMA和ANN模型的干旱预测适用性研究[J]. 干旱区地理,2018,41(5):945–953.
[YANG Huirong,ZHANG Yuhu,CUI Hengjian,et al.Application of ARIMA and ANN models for drought forecasting[J]. Arid Land Geography,2018,41(5):945–953.]
[14] 许德合,张棋,黄会平. ARIMA-SVR组合模型在基于标准化降水指数干旱预测中的应用[J]. 干旱地区农业研究,2020,38(2):276–282.
[XU Dehe,ZHANG Qi,HUANG Huiping.Application of the combined ARIMA-SVR model in drought prediction based on the Standardized Precipitation Index[J]. Agricultural Research in the Arid Areas,2020,38(2):276–282.]
[15] 于家瑞,艾萍,袁定波,等. 基于SPI的黑龙江省干旱时空特征分析[J]. 干旱区地理,2019,42(5): 1059–1068.
[YU Jiarui,AI Ping,YUAN Dingbo,et al.Spatial-temporal characteristics of drought in Heilongjiang Province based on standardized precipitation index[J]. Arid Land Geography,2019,42(5):1059–1068.]
[16] 韩萍,王鹏新,王彦集,等. 多尺度标准化降水指数的ARIMA模型干旱预测研究[J]. 干旱地区农业研究,2008,26(2):212–218.
[HAN Ping,WANG Pengxin,WANG Yanji,et al.Drought forecasting based on the standardized precipitation index at different temporal scales using ARIMA models[J]. Agricultural Research in the Arid Areas,2008,26(2):212–218.]
[17] 程俊. 基于ARIMA-LSTM混合模型的机械传动件制造企业销售预测方法研究与应用[D].成都:电子科技大学,2018.
[CHENG Jun.Research and application of sales forecasting method for mechanical transmission parts manufacturing enterprises based on ARIMA-LSTM hybrid model[D]. Chengdu:University of Electronic Science and Technology of China,2018.]
[18] ZHANG J F,ZHU Y,ZHANG X,et al.Developing a Long Short-Term Memory (LSTM) based model for predicting water table depth in agricultural areas[J]. Journal of Hydrology,2018,561:918–929.
[19] 苏夏羿. 基于多源遥感数据的青海省干旱监测模型研究[D]. 杨凌:西北农林科技大学,2017.
[SU Xiayi.Study on drought monitoring model of Qinghai province based on multi-source remote sensing data[D].Yangling:Journal of Northwest A & F University (Natural Science Edition),2017.]
[20] 戴升,李林,刘彩红,等. 青海夏季干旱特征及其预测模型研究[J].冰川冻土,2012,34(6):1433–1440.
[DAI Sheng,LI Lin,LIU Caihong,et al.Characteristics and prediction model of summer drought in Qinghai Province[J]. Journal of Glaciology and Geocryology,2012,34(6):1433–1440.]
[21] 张强,胡隐樵,曹晓彦,等. 论西北干旱气候的若干问题[J].中国沙漠,2000,20(4):357–362.
[ZHANG Qiang,HU Yinqiao,CAO Xiaoyan,et al.On some problems of arid climate system of northwest China[J]. Journal of Desert Research,2000,20(4):357–362.]
[22] 谢金南,李栋梁,尹东,等. 甘肃省干旱气候变化及其对西部大开发的影响[J]. 气候与环境研究,2002,7(3):359–369.
[XIE Jinnan,LI Dongliang,YIN Dong,et al.Effects of Gansu arid climate change on developing of the western China[J]. Climate and Environment Research,2002,7(3):359–369.]
[23] 林慧,王景才,黄金柏,等. 基于SPI和SPEI的淮河中上游流域气象干旱时空分布特征对比研究[J].水资源雨水工程学报,2019,30(6):59–67.
[LIN Hui,WANG Jingcai,HUANG Jinbai,et al.Comparative study on spatial and temporal distribution characteristics of meteorological drought in the upper and middle research of Huai River Basin based on SPI and SPEI[J]. Journal of Water Resources & Water Engineering,2019,30(6):59–67.]
[24] 中华人民共和国国家质量监督检验检疫总局. GB/T 20481—2006中华人民共和国国家标准:气象干旱等级[S]. 北京:中国标准出版社,2006.
[General Administration of Quality Supervision,Inspection and Quarantine of the People’s Republic of China. Classification of meteorological drought(GB /T20481–2006)[S]. Beijing: Standards Press of China,2006.]
[25] 刘晓璐,周延刚,温莉,等. 基于VSWI和SPI的2000—2016年河南省干旱特征研究[J]. 干旱区地理,2018,41(5):984–991.
[LIU Xiaolu,ZHOU Yan’gang,WEN Li,et al.Characteristics of drought in Henan Province from 2000 to 2016 based on VSWI and SPI[J]. Arid Land Geography,2018,41(5):984–991.]
[26] XU Dehe,ZHANG Qi,DING Yan,et al.Application of a hybrid ARIMA-SVR model based on the SPI for the forecast of drought:A case study in Henan Province,China[J]. Journal of Applied Meteorology and Climatology,2020.
[27] 曾妍,王迪,赵小娟,等. 基于支持向量回归的关中平原冬小麦估产研究[J].中国农业信息,2019,31(6):10–20.
[ZENG Yan,WANG Di,ZHAO Xiaojuan,et al.Study on yield prediction of winter wheat in Guanzhong Plain based on SVR[J]. China Agricultural Informatics,2019,31(6):10–20.]
Options
文章导航

/

〈 〉