Validation of an empirical model for grapevine leaf area estimation with data from simplified pruning systems

• Autor(a) principal: Lopes, C.M.
• Data de Publicação: 2005
• Outros Autores: Gruber, B., Schultz, H.R.
• Idioma: eng
• Título da fonte: Repositórios Científicos de Acesso Aberto de Portugal (RCAAP)
• Texto Completo: http://hdl.handle.net/10400.5/23024
• topic: grapevine; leaf area; statistical model; Tempranillo; Cabernet Sauvignon; Riesling; validation

The performance of a mathematical model developed for non-destructive estimation of primary leaf area per shoot of Tempranillo grapevines, was tested using independent datasets from two vineyards with simplified pruning techniques. The first dataset was collected in Portugal on Cabernet Sauvignon grapevines subjected to mechanical hedge pruning and the second one in Germany on minimal pruned Riesling grapevines. For both datasets the model presented a very good fit between observed and estimated values with the error increasing with the increase in leaf area per shoot. The mean absolute percent error for all systems was lower or equal to 10% with lower absolute values (7.7%) for the Riesling dataset. Both linear regression between observed (dependent variable) and estimated (independent variable) leaf area had high and significant R2 with an intercept not significantly different from zero. Fitted lines were not significantly different from 1 for Cabernet Sauvignon, but slightly yet significantly different from 1 for Riesling fitted line (1.03), indicating that the model underestimated the leaf area per shoot. The good results obtained with this validation test show that the model can be used to accurately predict primary leaf area per shoot independent of variety, training system and climatic conditions