Multimodel ensembles improve predictions of crop-environment-management interactions
Wallach, Daniel1; Martre, Pierre2; Liu, Bing3,4; Asseng, Senthold4; Ewert, Frank5,6; Thorburn, Peter J.7; van Ittersum, Martin8; Aggarwal, Pramod K.9; Ahmed, Mukhtar10,11; Basso, Bruno12,13; Biernath, Christian14; Cammarano, Davide15; Challinor, Andrew J.16,17; De Sanctis, Giacomo18; Dumont, Benjamin19; Rezaei, Ehsan Eyshi5,20; Fereres, Elias21,22; Fitzgerald, Glenn J.23,24; Gao, Y.4; Garcia-Vila, Margarita21,22; Gayler, Sebastian25; Girousse, Christine26; Hoogenboom, Gerrit4,27; Horan, Heidi7; Izaurralde, Roberto C.28,29; Jones, Curtis D.29; Kassie, Belay T.4; Kersebaum, Kurt C.30; Klein, Christian31; Koehler, Ann-Kristin16; Maiorano, Andrea2,44; Minoli, Sara32; Mueller, Christoph32; Kumar, Soora Naresh33; Nendel, Claas30; O'Leary, Garry J.34; Palosuo, Taru35; Priesack, Eckart31; Ripoche, Dominique36; Roetter, Reimund P.37,38; Semenov, Mikhail A.39; Stockle, Claudio10; Stratonovitch, Pierre39; Streck, Thilo25; Supit, Iwan40; Tao, Fulu35,41; Wolf, Joost42; Zhang, Zhao43
Corresponding AuthorWallach, Daniel(daniel.wallach@inra.fr)
AbstractA recent innovation in assessment of climate change impact on agricultural production has been to use crop multimodel ensembles (MMEs). These studies usually find large variability between individual models but that the ensemble mean (e-mean) and median (e-median) often seem to predict quite well. However, few studies have specifically been concerned with the predictive quality of those ensemble predictors. We ask what is the predictive quality of e-mean and e-median, and how does that depend on the ensemble characteristics. Our empirical results are based on five MME studies applied to wheat, using different data sets but the same 25 crop models. We show that the ensemble predictors have quite high skill and are better than most and sometimes all individual models for most groups of environments and most response variables. Mean squared error of e-mean decreases monotonically with the size of the ensemble if models are added at random, but has a minimum at usually 2-6 models if best-fit models are added first. Our theoretical results describe the ensemble using four parameters: average bias, model effect variance, environment effect variance, and interaction variance. We show analytically that mean squared error of prediction (MSEP) of e-mean will always be smaller than MSEP averaged over models and will be less than MSEP of the best model if squared bias is less than the interaction variance. If models are added to the ensemble at random, MSEP of e-mean will decrease as the inverse of ensemble size, with a minimum equal to squared bias plus interaction variance. This minimum value is not necessarily small, and so it is important to evaluate the predictive quality of e-mean for each target population of environments. These results provide new information on the advantages of ensemble predictors, but also show their limitations.
Keywordclimate change impact crop models ensemble mean ensemble median multimodel ensemble prediction
Indexed BySCI
WOS Research AreaBiodiversity & Conservation ; Environmental Sciences & Ecology
WOS SubjectBiodiversity Conservation ; Ecology ; Environmental Sciences
WOS IDWOS:000447760300007
Citation statistics
Cited Times:6[WOS]   [WOS Record]     [Related Records in WOS]
Document Type期刊论文
Corresponding AuthorWallach, Daniel
Affiliation1.INRA, UMR AGIR, F-31326 Castanet Tolosan, France
2.Montpellier SupAgro, INRA, UMR LEPSE, Montpellier, France
3.Nanjing Agr Univ, Jiangsu Collaborat Innovat Ctr Modern Crop Prod, Key Lab Crop Syst Anal & Decis Making,Minist Agr, Natl Engn & Technol Ctr Informat Agr,Jiangsu Key, Nanjing, Jiangsu, Peoples R China
4.Univ Florida, Agr & Biol Engn Dept, Gainesville, FL USA
5.Univ Bonn, Inst Crop Sci & Resource Conservat, INRES, Bonn, Germany
6.Leibniz Ctr Agr Landscape Res, Muncheberg, Germany
7.CSIRO Agr & Food Brisbane, St Lucia, Qld, Australia
8.Wageningen Univ, Plant Prod Syst Grp, Wageningen, Netherlands
9.BISA CIMMYT, CGIAR Res Program Climate Change Agr & Food Secur, New Delhi, India
10.Washington State Univ, Biol Syst Engn, Pullman, WA 99164 USA
11.Pir Mehr Ali Shah Arid Agr Univ, Dept Agron, Rawalpindi, Pakistan
12.Michigan State Univ, Dept Earth & Environm Sci, E Lansing, MI 48824 USA
13.Michigan State Univ, WK Kellogg Biol Stn, E Lansing, MI 48824 USA
14.German Res Ctr Environm Hlth, Helmholtz Zentrum Munchen, Inst Biochem Plant Pathol, Neuherberg, Germany
15.James Hutton Inst Invergowrie, Dundee, Scotland
16.Univ Leeds, Sch Earth & Environm, Inst Climate & Atmospher Sci, Leeds, W Yorkshire, England
17.Int Ctr Trop Agr CIAT, ESSP Program Climate Change Agr & Food Se, CGIAR, Cali, Colombia
18.European Food Safety Author, GMO Unit, Parma, Italy
19.Univ Liege, Gembloux Agrobio Tech, Dept Terra & AgroBioChem, Liege, Belgium
20.Ctr Dev Res ZEF, Bonn, Germany
21.CSIC, IAS, Cordoba, Spain
22.Univ Cordoba, Cordoba, Spain
23.Dept Econ Dev, Jobs Transport & Resources, Agr Victoria Res, Ballarat, Vic, Australia
24.Univ Melbourne, Fac Vet & Agr Sci, Creswick, Vic, Australia
25.Univ Hohenheim, Inst Soil Sci & Land Evaluat, Stuttgart, Germany
26.Univ Clermont Auvergne, INRA, UMR GDEC, Clermont Ferrand, France
27.Univ Florida, Inst Sustainable Food Syst, Gainesville, FL USA
28.Univ Maryland, Dept Geog Sci, College Pk, MD 20742 USA
29.Texas A&M Univ, Texas A&M Agrilife Res & Extens Ctr, Temple, TX USA
30.Leibniz Ctr Agr Landscape Res, Inst Landscape Syst Anal, Muncheberg, Germany
31.German Res Ctr Environm Hlth, Helmholtz Zentrum Munchen, Inst Biochem Plant Pathol, Neuherberg, Germany
32.Potsdam Inst Climate Impact Res, Potsdam, Germany
33.IARI, PUSA, Ctr Environm Sci & Climate Resilient Agr, New Delhi, India
34.Grains Innovat Pk, Dept Econ Dev, Agr Victoria Res, Jobs Transport & Resources, Horsham, Vic, Australia
35.Nat Resources Inst Finland Luke, Helsinki, Finland
36.INRA, US Agroclim, Avignon, France
37.Univ Gottingen, Trop Plant Prod & Agr Syst Modelling TROPAGS, Gottingen, Germany
38.Univ Gottingen, Ctr Biodivers & Sustainable Land Use CBL, Gottingen, Germany
39.Rothamsted Res, Computat & Syst Biol Dept, Harpenden, Herts, England
40.Wageningen Univ, Water & Food & Water Syst & Global Change Grp, Wageningen, Netherlands
41.Chinese Acad Sci, Inst Geog Sci & Nat Resources Res, Beijing, Peoples R China
42.Wageningen Univ, Plant Prod Syst, Wageningen, Netherlands
43.Beijing Normal Univ, Fac Geog Sci, State Key Lab Earth Surface Proc & Resource Ecol, Beijing, Peoples R China
44.EFSA, Parma, Italy
Recommended Citation
GB/T 7714
Wallach, Daniel,Martre, Pierre,Liu, Bing,et al. Multimodel ensembles improve predictions of crop-environment-management interactions[J]. GLOBAL CHANGE BIOLOGY,2018,24(11):5072-5083.
APA Wallach, Daniel.,Martre, Pierre.,Liu, Bing.,Asseng, Senthold.,Ewert, Frank.,...&Zhang, Zhao.(2018).Multimodel ensembles improve predictions of crop-environment-management interactions.GLOBAL CHANGE BIOLOGY,24(11),5072-5083.
MLA Wallach, Daniel,et al."Multimodel ensembles improve predictions of crop-environment-management interactions".GLOBAL CHANGE BIOLOGY 24.11(2018):5072-5083.
Files in This Item:
There are no files associated with this item.
Related Services
Recommend this item
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Wallach, Daniel]'s Articles
[Martre, Pierre]'s Articles
[Liu, Bing]'s Articles
Baidu academic
Similar articles in Baidu academic
[Wallach, Daniel]'s Articles
[Martre, Pierre]'s Articles
[Liu, Bing]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Wallach, Daniel]'s Articles
[Martre, Pierre]'s Articles
[Liu, Bing]'s Articles
Terms of Use
No data!
Social Bookmark/Share
All comments (0)
No comment.

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.