Near Infrared Model Development for Pond-cypress Subjected to Chronic Water Stress and Botryosphaeria rhodina

Visible and near infrared (NIR) reflectance correlate with chemical and physical characteristics of forest canopies. Laboratory analysis of spectral reflectance in the NIR region has advantages over wet-chemical analysis. Spectral analysis of pond-cypress (Taxodium ascendens Brong.) under controlled water stress and Botryosphaeria rhodina (Cooke) Aux. inoculation has not been reported previously. Eight models were selected for preliminary multivariate evaluation of the relationship between reflectance (NIR 1100–2500 nm) and soluble carbohydrate constituents identified in a previous study. “Best fit models” for half of the constituents were first derivative, modified partial least squares (Mod PLS) models. The remainder were second derivative, Mod PLS models. The R2 values from the best fit models for branch-tip carbohydrates were (from greatest to least): arabinose (0.93), xylose (0.93), rhamnose (0.92), pentose:hexose ratio (0.91), total pentoses (0.87), total carbohydrates (0.83), galacturonic acid (0.81), total hexoses (0.81), mannose (0.78), glucose (0.67), galactose (0.47). Evaluation of dry, milled branch tips from natural stands of pond-cypress trees may provide better understanding of physiological impacts on trees associated with anthropogenic hydroperiod perturbations (e.g., from unsustainable aquifer yield, aquifer storage and recovery, and mining solid materials).