Jump to Main Content
Identifying traits at crop maturity and models for estimation of lodging susceptibility in bread wheat
- Mirabella, N. E., Abbate, P. E., Alonso, M. P., Panelo, J. S., Pontaroli, A. C.
- Crop & pasture science 2019 v.70 no.2 pp. 95-106
- Triticum aestivum, breeding, breeding programs, cultivars, genotype-environment interaction, lodging, model validation, models, plant height, rapid methods, regression analysis, safety factor, screening, shoots, wheat, wind speed, Argentina, Pampas region
- Lodging is the permanent displacement of plant shoots from an upright position and represents a major obstacle to reaching yield potential in bread wheat (Triticum aestivum L.). Breeding programs would benefit from the identification of lodging-related traits amenable to easy and rapid screening, even in the absence of lodging. However, no locally tested lodging model is available for the Pampas region of Argentina, and most lodging models are based on measurements before crop maturity. We adapted two existing models and generated a new one, using easily measurable traits at crop maturity in 24 cultivars (14 for model fit and 10 for model validation) grown in plot trials with no nutritional, water or disease restrictions in three crop seasons at Balcarce, Argentina. Of 17 traits evaluated, 16 showed differences between cultivars (P<0.05), and in 11 of these traits, no genotype×environment interaction was detected (P>0.05). Estimations of the safety factor against stem lodging, proposed by Crook et al., and the wind velocity that produces lodging, proposed by Berry et al., showed a high correlation with lodging score (R2=0.60 and 0.72, respectively), but when the estimators were tested with another set of cultivars there was no association. A new empirical regression model was based on three traits measured at maturity: plant height, spike dry weight, and the inertia moment of the stem base (stem resistance to bending estimated from stem diameter and wall thickness). The model was then simplified by replacing the third trait with basal stem diameter, and it showed an even better fit (R2=0.90). These models were satisfactorily validated by rank correlations with a different cultivar set. The regression model proposed in this study can easily be applied to the evaluation of commercial cultivars and may be used to screen breeding materials. Measurements at maturity are convenient and easy to combine with other traits of possible selective advantage.