Labile carbon storage is an important functional trait in woody plants. When carbon uptake is interrupted, such as when deciduous species have dropped their leaves, soluble sugars and starch reserves – collectively referred to as nonstructural carbohydrates (NSCs) - act as a source to meet metabolic carbon demands. Genotypes subjected to episodic stress factors such as freezing, fire and herbivory may be under selection for enhanced allocation to NSC storage, with the potential tradeoff of reduced allocation to growth and reproduction. For example, plants that experience freezing temperatures have increased risk of xylem cavitation and tissue dieback, and NSC reserves could play a role in xylem refilling and providing substrates for constructing new tissues. To investigate variation in allocation patterns we constructed a common garden using genotypes of Tamarix, a non-native tree, that is abundant in riparian areas throughout the western US. Genotypes were selected across an altitudinal gradient where mid and high elevation populations are often exposed to freezing temperatures. Radial growth, flower output and tissue samples were analyzed to quantify the total NSC concentrations as well biomass accumulation to test our hypothesis that carbon should be allocated preferentially to storage in the populations from colder sites.
Results/Conclusions
We found significant differences in starch concentrations throughout the growing season (ANOVA F2,224 = 9.241, P < 0.001), with higher elevation populations maintaining higher levels of starch at each of the sampling periods. With respect to carbon demands for growth and reproduction biweekly census of flowers per genotype February until June resulted in allometric relationships to determine leaf area to basal area relationships in August. Low elevation populations had significantly more flowers (ANOVA: F2,8032= 138.7, p<0.001), and greater total leaf:basal area in August (ANOVA F2,26= 7.386, p = 0.003). The same low elevation populations tended to have lower starch levels suggesting allocation to growth and reproduction at the expense of storage. The results show that starch concentrations could be modelled using a combination of variables related to minimum winter temperatures, precipitation and elevation at the provenance sites (R2 = 0.21 F5,67 = 3.578, p = 0.006). Thus, stressful environmental conditions, such as cold temperatures, select for genotypes that allocate greater carbon to storage as a hedge against stress-related mortality. In taxa such as Tamarix spp. possessing high local genetic variation this selection can happen over relatively short evolutionary time, resulting in populations adapted to local conditions.