Assessing frost damages using dynamic models in walnut trees: exposure rather than vulnerability controls frost risks.
As observed for most stresses, tree frost resistance can be split into two main processes: avoidance and tolerance. Avoidance of freezing is achieved by introducing species only in the climatic context in which the probability of freezing events is very low for the sensitive stages of buds or stems; i.e., when good synchronism exists between the annual cycle and the critical climatic periods. Buds become able to grow only after chilling requirements have been satisfied (endodormancy released) during winter; they subsequently break after heat requirements have been completed (end of ecodormancy) in early spring. Actually, this period is often subject to more or less severe freezing events. Trees are also able to adjust their freezing tolerance by increasing their capacity of extracellular freezing and decreasing the possibility of intracellular freezing through the process of frost acclimation. Both freezing resistance processes (avoidance and tolerance) are environmentally driven (by photoperiod and temperature), but there are also genotypic effects among species or cultivars. Here, we evaluated the degree to which differences in dormancy release and frost acclimation were related to environmental and genetic influences by comparing trees growing in common garden conditions. This investigation was carried out for two winters in lowland and mountain locations on different walnut genotypes differing significantly for budburst dates. Chilling requirement for endodormancy release and heat requirement during ecodormancy were evaluated in all situations. In addition, frost acclimation was assessed by the electrolyte leakage method on stems from the same trees before leaf fall through budburst. No significant differences were observed in chilling requirements among genotypes. Moreover, frost acclimation dynamics were similar between genotypes or locations when expressed depending on chilling units accumulated since 15 September as a time basis instead of Julian day. The only exception was for maximal frost hardiness observed during winter with the timber-oriented being significantly more resistant than fruit-oriented genotypes. Heat requirement was significantly different among genotypes. Thus, growth was significantly faster in fruit-oriented than in wood-oriented genotypes. Furthermore, among wood-oriented genotypes, differences in growth rate were observed only at cold temperatures. Frost acclimation changes differed significantly between fruit- and wood- walnuts from January through budburst. In conclusion, from September through January, the acclimation dynamic was driven mainly by environmental factors whereas from January through budburst a significant genotype effect was identified in both frost tolerance and avoidance processes.