Genetics of fall, winter and spring cold hardiness in coastal Douglas fir seedlings

Genetics of fall, winter and spring cold hardiness were investigated in seedlings from two western Oregon breeding populations (Coast and Cascade) of coastal Douglas-fir (Pseudotsuga menziesii var. menziesii (Mirb.) Franco). Seedlings of forty open-pollinated families from each population were grown in nursery beds and subjected to either a wet or dry moisture regime. Cold hardiness was assessed in fall and winter after the second growing season, in winter after artificial warming, and in spring prior to budburst in the third growing season, by subjecting detached shoots to artificial freeze tests (AFT), and visually assessing needles, buds and stems for injury. Results of earlier investigations of the same families at age seven were used to compare genetic control of cold hardiness in seedlings and saplings.; Significant (p < 0.05) variation in cold hardiness was observed between the two populations in almost all fall assessments and in some cases in the spring assessment. Families within both populations varied significantly in all fall cold hardiness assessments and in the spring assessment. Estimates of individual heritabilities for cold hardiness were weak to moderate, although stronger in spring (h_i 2 = 0.57) than in fall (Oct h_i 2 = 0.37), stronger in the wet (h _i2 = 0.49) than in the dry regime ( h_i2 = 0.36), and stronger for stems (h_i2 = 0.41) than for needles or buds (h_i 2 = 0.33). Genetic correlations among the three tissues for cold injury in the same season were generally strong ( = 0.69). Moisture regime had little effect on family ranking for cold hardiness. Fall cold injury in freeze tests was fairly strongly associated with injury from a natural frost that occurred in November of the first growing season ( = 0.67). These results show that cold hardiness of Douglas-fir families can effectively be screened at the seedling stage using artificial freeze testing. Moderately negative genetic correlations between spring and fall cold injury ( = −0.47), however, indicate that fall and spring cold hardiness need to be considered separately.; Estimated genetic correlations between seedlings and saplings in stem cold injury were strong in both populations for both spring (r_B ≥ 0.78) and fall (r_B ≥ 0.80), indicating that cold hardiness is largely under similar genetic control at the two ages. This means that artificial selection for cold hardiness at either age will be quite effective in improving cold hardiness at the other age as well. Estimated genetic correlations between bud burst date and stem cold injury in the spring was strong in both seedlings (r_A ≤ −0.82) and saplings (r_A ≤ 5 −0.90), indicating that bud burst timing is a good predictor of stem cold hardiness prior to bud burst at both ages. Date of bud set, on the other hand, is a reliable predictor of fall cold hardiness only in seedlings (r_A ≥ 0.65). At the sapling stage, estimated genetic correlations between bud set and fall cold injury to stems were weak (r_A ≥ 0.28).; Genetic correlations of stem cold hardiness in winter after warming, with cold hardiness in winter prior to warming, and with spring cold hardiness, were strong (r_A = 0.68 and 0.75, respectively) in the Coast, but weak in the Cascade population (r