a novel solution to the variable selection problem in window pane approaches of plant pathogen – climate models development, evaluation and application of a climatological model for brown rust of wheat论文.pdf

Agricultural and Forest Meteorology 205 (2015) 51–59 Contents lists available at ScienceDirect Agricultural and Forest Meteorology j o u rn al h omepage: w w w. elsev /locate/agrf ormet A novel solution to the variable selection problem in Window Pane approaches of plant pathogen – Climate models: Development, evaluation and application of a climatological model for brown rust of wheat David Gouache a,c,∗ , Marie Sandrine Léon a,b , Florent Duyme c , Philippe Braun d a ARVALIS – Institut du Végétal, Service Génétique Physiologie et Protection des Plantes, Rue de Noetzlin–Bât. 630, F-91405 Orsay Cedex, France b Université Paris-Descartes, France c ARVALIS – Institut du végétal, Station Expérimentale, F-91720 Boigneville, France d ARVALIS – Institut du végétal, Domaine de la Bastide, Route de Generac, F-30900 Nimes, France a r t i c l e i n f o a b s t r a c t Article history: A model for predicting brown rust severity in France was developed using the systematic screening of Received 2 October 2014 climatic variables of the Window Pane approach and data from 400 field trials spanning 30 years. The Received in revised form 12 February 2015 model was built using novel methods to manage the variable selection problem posed by the very large Accepted 20 February 2015 number of predictor variables generated by Window Pane, namely the elastic-net, and a systematic cross- Available online 27 February 2015 validation to determine the most frequently retained variables. The model predicts the final severity of brown rust