Seasonal dynamics of non-structural carbon pools and their relationship to growth in two boreal conifer tree species.

In an attempt to comprehensively study the dynamics of non-structural carbon compounds (NCCs), we measured the seasonal changes of soluble sugars, starch, lipids and sugar alcohols in the leaves, branches, stem and roots of the fast-growing Pinus contorta (Loudon) (pine) and slow-growing Picea glauca (Moench) Voss (spruce) trees growing in a boreal climate. In addition to measuring the seasonal concentrations of these compounds, the relative contribution of these compounds to the total NCC pool within the organs of trees (~8 m tall) was estimated and compared across different phenological and growth stages. Both species showed large seasonal shifts from starch to sugars from spring to fall in nearly all organs and tissues; most likely an adaptation to the cold winters. For both species, the total fluctuation of sugar + starch across the year (i.e., the difference between the minimum and maximum observed across collection times) was estimated to be between 1.6 and 1.8 kg for all NCCs. The fluctuation, however, was 1.40 times greater than the minimum reserves in pine, while only 0.72 times the minimum reserves in spruce. By tissue type, NCC fluctuations were greatest in the roots of both species. Roots showed a large build-up of reserves in late spring, but these reserves were depleted over summer and fall. Storage reserves in needles and branches declined over the summer, and this decline may be linked to the sink strength of the stem during diameter growth. Some notable highlights of this holistic study: a late winter build-up of sugars in the stem xylem of both species, but especially spruce; and an increase in sugar alcohols in the bark of spruce in very late winter, which could indicate mobilization to support early growth in spring and high lipid reserves in the bark of pine, which appeared not to be impacted by seasonal changes between summer and winter. Collectively, these observations point toward a more conservative NCC reserve strategy in spruce compared with pine, which is consistent with its stress tolerance and greater longevity.

Viscosity Prediction of Lubricants by a General Feed-Forward Neural Network.

Modern industrial lubricants are often blended with an assortment of chemical additives to improve the performance of the base stock. Machine learning-based predictive models allow fast and veracious derivation of material properties and facilitate novel and innovative material designs. In this study, we outline the design and training process of a general feed-forward artificial neural network that accurately predicts the dynamic viscosity of oil-based lubricant formulations. The network hyperparameters are systematically optimized by Bayesian optimization, and strongly correlated/collinear features are trimmed from the model. By harnessing domain knowledge in the selection of features, the quantitative structure-property relationship model is built with a relatively simple feature set and is versatile in predicting the dynamic viscosity of lubricant oils with and without enhancement by viscosity modifiers (VMs). Moreover, partial dependency, local-interpretable model-agnostic explanations, and Shapley values consistently show that the eccentricity index, Crippen MR, and Petitjean number are important predictors of viscosity. All in all, the neural model is reasonably accurate in predicting the dynamic viscosity of lubricant solvents and VM-enhanced lubricants with an R2 of 0.980 and 0.963, respectively.