Jump to Main Content
A disaggregated dynamic model for predicting volume, biomass and carbon stocks in even-aged pedunculate oak stands in Galicia (NW Spain)
- Gómez-García, Esteban, Crecente-Campo, Felipe, Barrio-Anta, Marcos, Diéguez-Aranda, Ulises
- European journal of forest research 2015 v.134 no.3 pp. 569-583
- Quercus robur, aboveground biomass, autocorrelation, basal area, carbon sinks, climate change, decision support systems, dynamic models, equations, forests, heteroskedasticity, leaves, managers, prediction, trees, Spain
- A model was developed for predicting current and future disaggregated volume, biomass and carbon stocks in even-aged pedunculate oak (Quercus robur L.) stands in Galicia (northwestern Spain). In the model, the stand state at any point in time is defined by age (A), dominant height (H), number of trees per hectare (N) and stand basal area (G). A dynamic model project H, N and G over time by using transition functions in algebraic difference form. A number of static relationships are used to estimate other quantities as functions of the stand state. A disaggregation system, which comprises a diameter distribution function and a generalized height–diameter relationship, enables prediction of the number of trees in each diameter class and their average height. A taper equation, developed in a previous study, is used to estimate total and merchantable volume. A set of additive equations is used to predict total aboveground biomass and biomass of different components. Finally, carbon stocks are predicted from the average C content (%) in the different biomass components. The equations in the model were developed under tenable statistical assumptions: a base-age invariant method, which accounted for autocorrelation, correlated errors and different number of observations between the transition functions; and simultaneous fitting of the biomass system, which accounted for heteroscedasticity and inherent correlations between biomass components and the lack of leaf data in some trees. Critical errors of 13, 16 and 11 % were obtained in H, N and G predictions for a projection interval of 3 to 9 years. Biomass equations explained between 78 and 98 % of the observed variability. The average amount of C stored was approximately 48 % of total dry biomass. The proposed model can be used by forest managers as part of a decision support system that enables consideration of production and environmental aspects related to climate change (biomass and C stocks).