Modelling Adaptive Responses Across Agricultural Environments as a Prerequisite for Identifying Adaptive Traits and Plant Ideotypes

Abstract

Modelling cultivar yield responses across agricultural environments by additive main effects and multiplicative interaction (AMMI) or factorial regression is useful for understanding genotype × environment interaction patterns, simplifying and improving the targeting of cultivars, and helping breeding programs in defining selection strategies, adaptive traits and plant ideotypes by itself and by acting as a benchmark for physiological studies. Examples are provided from results of joint field testing across several Mediterranean countries and further physiological work performed within the EU-funded project PERMED (Improvement of native perennial forage plants for sustainability of Mediterranean farming systems). Cultivar adaptive responses across Mediterranean sites depended mainly on the site drought-stress level for cocksfoot and sulla and on the level of drought, mowing frequency and soil salinity for lucerne. Specific adaptation of cocksfoot material was largely affected by its summer dormancy. Lucerne cultivars which showed contrasting adaptation pattern were evaluated for shoot and root traits in metal containers (55 cm × 12 cm × 75 cm deep) under different drought-stress levels. These artificial environments were able to reproduce the cultivar adaptive responses across agricultural environments (unlike another experiment using 30 cm-deep pots), revealing a conservative water strategy in a drought-tolerant Italian landrace.