4.1. Geographical and Environmental Significance of Parental Origin
The results indicated that the full-sib families that had higher levels of productivity (diameter, basal area, and volume) after 28 years of growth since initial plantation establishment in 1982 owed much of their high growth potential to having a maternal parent originating from a county close to the county of plantation establishment of the full-sib family (
i.e., Ingham county). In this study, a strong maternal environmental effect was also observed in terms of the effect of environmental conditions (annual and summer temperature and summer precipitation) of maternal county origin (relative to the plantation site) on family productivity. Maternal environmental effects have been reported for plants in general [
38,
39,
40] and trees in particular [
41,
42,
43]. In a greenhouse study, Reighard and Hanover [
44] reported a maternal effect on stem to root ratios of
P. × smithii. It has been hypothesized that the maternal effect acts partially through conferring on progeny the same phenological growth pattern as the mother [
41,
43]. Under this hypothesis, full-sib families with a mother originating far away from the planting site location would be expected to be phenologically mismatched for growth at that planting site. Namely, in the present study, it is speculated that full-sib families with a mother originating farther north from the planting site location are displaying earlier growth termination than their full-sib family counterparts with mothers originating closer to the plantation site location.
Geographic distance between counties of parents had no significant effect on measures of productivity (diameter, height, basal area, and volume) which was also confirmed by the study of Reighard and Hanover [
27] on biomass patterns of different full-sib families of
P. × smithii. Full-sib families did however show higher productivity, and thus suggesting better adaptation, to the plantation site when the maternal parent originated from a county that was warmer (annual and summer) or had higher summer moisture than environmental conditions of paternal county location.
Identity of the maternal parent (
P. tremuloides versus
P. grandidentata) did not affect overall cumulative growth in any of the growth parameters. Furthermore, identity of the maternal parent did not effect responses of interannual basal area growth patterns to climatic factors. In contrast, a prior greenhouse study indicated that interspecific hybrids in which the maternal parent was
P. tremuloides generally grew taller than hybrids in which
P. grandidentata was the maternal parent [
44].
4.2. Interannual Growth-Climate Relationships
Principal component analysis (PCA) indicated that there was strong common variance (63.5% growth variance captured by the first principal component) shared by the interannual patterns of basal area growth of 17 of the 18 full-sib families. However, the slowest growing full-sib family (MAR_CLA2) differed from the other families likely because of strong within stand competition related effects (suggested by the high slenderness coefficient) that affected interannual variations in basal area growth. Nonetheless, the high proportion of variance explained by the first principal component indicates that growth for most of the full-sib families is being affected by a common external environmental factor (i.e., climate).
Dendrochronological analyses of growth-climate relationships indicated that precipitation had a greater impact on interannual growth patterns of full-sib families of
P. ×
smithii than did temperature variables. In particular, the direct association of basal area growth of 14 out of 18 full-sib families to 3 month precipitation amounts ending in September of year prior to (t−1) ring formation indicated that growth was affected by moisture stress in this seasonal window (
Figure 5b). The lagged effect of late summer precipitation in the year prior to growth has also been noted in
P. tremuloides in northeastern British Columbia, Canada [
21] and in the eastern Canadian boreal forest in the provinces of Ontario and Quebec [
22]. This strong lagged effect of past precipitation amounts on diameter growth in the following growing season is likely due to a number of possible ecophysiological mechanisms. First, it is generally understood that favorable climatic conditions in late summer and early fall promote excess carbon production which is allocated to the accumulation of carbohydrate storage reserves in poplars in storage tissues such as roots, stems, and branches [
45,
46,
47,
48]. Carbohydrate reserves generally build up in late summer and reach their highest levels in the fall and are then depleted in the spring of the following growing season as these reserves are used to drive initial tree growth [
47,
49]. Consequently, low precipitation in late summer and early fall can thus limit the degree of carbohydrate reserve accumulation which in turn would limit the amount of reserves to drive diameter growth in the following growing season.
Some foliage of hybrid poplars arises from a determinate growth pattern particularly in short shoots commonly located near the tree crown base [
47,
50]. Therefore, a second possible ecophysiologically based explanation for the lagged effect of precipitation is that leaf primordia are generally formed in tree buds in the year prior to ring formation. Low precipitation levels would likely limit the number of leaf primordia produced and thereby curtail the amount of leaf area produced in the following growing season in the year of bud expansion.
The effect of moisture on interannual basal area growth was also present in the year of ring formation (t). Basal area growth was directly associated with monthly precipitation in October (t) (
Figure 4b) and in 3 month precipitation amounts ending in October (t) (
Figure 5b) in 11 out of 18 full-sib families of
P. × smitthii. These results suggest that basal area growth is strongly affected by the length of the growing season since greater precipitation in late summer and early fall was associated with greater growth. Precipitation amounts in the current year can effect the degree of leaf expansion of leaf primordia produced during bud set at the end of the prior growing season (determinate foliage development). Furthermore, hybrid poplars are generally known for producing more indeterminate foliage from long shoot regions of the upper crown [
47]. Water stress during the current year can limit the amount of leaf area development of indeterminate foliage development. Furthermore, the effect of drought stress in either the prior year or current year of ring formation can negatively impact growth by increasing the likelihood of xylem cavitation which is the blockage of these water conducting tissues with air thus forming embolisms [
47,
51].
Although diameter growth is most active in the summer (June and July), diameter growth of poplar and hybrid poplar clones in Washington state continued into October [
52] and November [
53]. Furthermore, growth of hybrid poplar (
P. tremula ×
P. tremuloides) in Finland continued into early and mid October [
54] and into November for
P. ×
canadensis Moench in Italy [
55]. The results in the current study thus indirectly suggest that the phenology of many of the full-sib families in southern Michigan continue diameter growth into the fall (
i.e., October) but further studies are required to confirm the phenology of diameter growth in these full-sib families of
P. × smithii.
The impact of moisture stress was also underscored by the fact that the lowest decrease in growth was observed in 1997 in the BAI growth chronology for all full-sib families combined and this year corresponded with the 1997 to 1998 El-Niño event [
56]. El-Niño events are generally associated with summer drought stress conditions in North America [
57].
The next climatic factor that impacted growth was the negative influence of winter precipitation (monthly December precipitation) in 7 out of the 18 full-sib families of
P. × smithii. The effect of winter precipitation appeared to affect the full-sib families with generally larger cumulative sizes. It is postulated that there is a tradeoff between large tree size and increased likelihood of crown breakage from high mechanical loads resulting from high winter precipitation levels [
58]. Damaged crowns would likely receive higher priority for carbohydrate allocation in the subsequent growing season which in turn would limit the amount of food resources for diameter growth which generally receives lower priority for carbohydrate allocation [
46]. By damaging tree crowns, high winter precipitation levels may also increase the risk of winter desiccation injury [
46,
59]. It is interesting that the slowest growing full-sib family with parents originating from Marquette and Clare (MAR_CLA2) showed a positive relation to 3 month precipitation amounts ending in December (
Figure 5b). It is speculated that this result suggests that the most competitively impacted full-sib family likely benefits from crown breakup of more dominant trees in the plantation stand.
While interannual basal area growth of full-sib families generally showed a very weak response to temperature variables, the most notable response was a positive response in 4 out of 18 families to March mean monthly temperature (
Figure 4a) or 3 month mean temperature ending in May (t) (
Figure 5a). These results suggested that growth in these full-sib families was generally controlled by the timing of the start of the growing season. Bud break and cambial reactivation in tree species is generally controlled by accumulated heat sums in the spring [
60]. Zalesny
et al. [
10] determined that belowground temperature heat sums was an influential factor controlling rooting of hybrid poplar clones in early spring in the Midwestern United States.
