Using multi-model ensembles of CMIP5 global climate models to reproduce observed monthly rainfall and temperature with machine learning methods in Australia

doi:10.1002/joc.5705

	Using multi-model ensembles of CMIP5 global climate models to reproduce observed monthly rainfall and temperature with machine learning methods in Australia
	Wang, Bin 1; Zheng, Lihong 2; Liu, De Li 1,3,4; Ji, Fei 5; Clark, Anthony 6; Yu, Qiang 7,8,9
刊名	INTERNATIONAL JOURNAL OF CLIMATOLOGY
	2018-11-15
卷号	38 期号:13 页码:4891-4902
关键词	GCMs machine learning multi-model ensemble random forest support vector machine
ISSN号	0899-8418
DOI	10.1002/joc.5705
通讯作者	Wang, Bin(bin.a.wang@dpi.nsw.gov.au)
英文摘要	Global climate models (GCMs) are useful tools for assessing climate change impacts on temperature and rainfall. Although climate data from various GCMs have been increasingly used in climate change impact studies, GCMs configurations and module characteristics vary from one to another. Therefore, it is crucial to assess different GCMs to confirm the extent to which they can reproduce the observed temperature and rainfall. Rather than assessing the interdependence of each GCM, the purpose of this study is to compare the capacity of four different multi-model ensemble (MME) methods (random forest [RF], support vector machine [SVM], Bayesian model averaging [BMA] and the arithmetic ensemble mean [EM]) in reproducing observed monthly rainfall and temperature. Of these four methods, the RF and SVM demonstrated a significant improvement over EM and BMA in terms of performance criteria. The relative importance of each GCM based on the RF ensemble in reproducing rainfall and temperature could also be ranked. We compared the GCMs importance and Taylor skill score and found that their correlation was 0.95 for temperature and 0.54 for rainfall. Our results also demonstrated that the number of GCMs ensemble simulations could be reduced from 33 to 25 in RF model while maintaining predictive error less than 2%. Having such a representative subset of simulations could reduce computational costs for climate impact modelling and maintain the quality of ensemble at the same time. We conclude that machine learning MME could be efficient and useful with improved accuracy in reproducing historical climate variables.
WOS关键词	ORGANIC-CARBON STOCKS ; EASTERN AUSTRALIA ; RANDOM FORESTS ; CHANGE IMPACTS ; PROJECTIONS ; INDEPENDENCE ; PREDICTION ; SCENARIOS ; ALGORITHM ; NETWORKS
WOS研究方向	Meteorology & Atmospheric Sciences
语种	英语
出版者	WILEY
WOS记录号	WOS:000450222100015
内容类型	期刊论文
源URL	[http://ir.igsnrr.ac.cn/handle/311030/52486]
专题	中国科学院地理科学与资源研究所
通讯作者	Wang, Bin
作者单位	1.NSW Dept Primary Ind, Wagga Wagga Agr Inst, Wagga Wagga, NSW 2650, Australia 2.Charles Sturt Univ, Sch Comp & Math, Wagga Wagga, NSW, Australia 3.Univ New South Wales, Climate Change Res Ctr, Sydney, NSW, Australia 4.Univ New South Wales, ARC Ctr Excellence Climate Extremes, Sydney, NSW, Australia 5.NSW Off Environm & Heritage, Dept Planning & Environm, Sydney, NSW, Australia 6.NSW Dept Primary Ind, Orange Agr Inst, Orange, NSW, Australia 7.Univ Technol Sydney, Sch Life Sci, Fac Sci, Sydney, NSW, Australia 8.Univ Chinese Acad Sci, Inst Geog Sci & Nat Resources Res, Key Lab Water Cycle & Related Land Surface Proc, Beijing, Peoples R China 9.Northwest A&F Univ, State Key Lab Soil Eros & Dryland Farming Loess P, Yangling, Shaanxi, Peoples R China
推荐引用方式 GB/T 7714	Wang, Bin,Zheng, Lihong,Liu, De Li,et al. Using multi-model ensembles of CMIP5 global climate models to reproduce observed monthly rainfall and temperature with machine learning methods in Australia[J]. INTERNATIONAL JOURNAL OF CLIMATOLOGY,2018,38(13):4891-4902.
APA	Wang, Bin,Zheng, Lihong,Liu, De Li,Ji, Fei,Clark, Anthony,&Yu, Qiang.(2018).Using multi-model ensembles of CMIP5 global climate models to reproduce observed monthly rainfall and temperature with machine learning methods in Australia.INTERNATIONAL JOURNAL OF CLIMATOLOGY,38(13),4891-4902.
MLA	Wang, Bin,et al."Using multi-model ensembles of CMIP5 global climate models to reproduce observed monthly rainfall and temperature with machine learning methods in Australia".INTERNATIONAL JOURNAL OF CLIMATOLOGY 38.13(2018):4891-4902.