Estimating above-ground biomass of trees: comparing Bayesian calibration with regression technique

The commitment to report greenhouse gas emissions requires an estimation of biomass stocks and their changes in forests. When this was first done, representative biomass functions for most common tree species were very often not available. In Germany, an estimation method based on solid volume was developed (expansion procedure). It is easy to apply because the required information is available for nearly all relevant tree species. However, the distributions of neither parameters nor prediction intervals are available. In this study, two different methods to estimate above-ground biomass for Norway spruce (Picea abies), European beech (Fagus sylvatica), and Scots pine (Pinus sylvestris) are compared. First, an approach based on information from the literature was used to predict above-ground biomass. It is basically the same method used in greenhouse gas reporting in Germany and was applied with prior and posterior parameters. Second, equations for direct estimation of biomass with standard regression techniques were developed. A sample of above-ground biomass of trees was measured in campaigns conducted previously to the third National Forest Inventory in Germany (2012). The data permitted the application of Bayesian calibration (BC) to estimate posterior distribution of the parameters for the expansion procedure. Moreover, BC enables the calculation of prediction intervals which are necessary for error estimations required for reporting. The two methods are compared with regard to predictive accuracy via cross-validation, under varying sample sizes. Our findings show that BC of the expansion procedure performs better, especially when sample size is small. We therefore encourage the use of existing knowledge together with small samples of observed biomass (e.g., for rare tree species) to gain predictive accuracy in biomass estimation.     

Financial optimisation of target diameter harvest of European beech (Fagus sylvatica) considering the risk of decrease of timber quality due to red heartwood

A mature, average stand of European beech was generated based on characteristic data of trial plots. Some 27 different strategies of target diameter harvest, were simulated for up to 80 years with the help of a distance-dependent single-tree growth simulator. The treatments were differing in the size of the target diameter, the beginning and the end of the harvest. Based on a statistical model, the probability of the occurrence of more than 30% of red heartwood at the front-side diameter was calculated for three sections of each log. Using the predicted probability, the decrease of timber quality due to red heartwood for different treatment strategies was assessed. The harvested volume and the predicted timber quality for different harvesting strategies were used to calculate the net revenue achieved in each simulation period with the help of a calculation program. The net present value for variable interest rates of the different harvesting strategies was calculated, assuming free land rent. Using a linear programming approach, optimal areas for different treatment strategies of a modelled forest of 100 ha were calculated under 4 different scenarios. The results of the optimisation showed how the increasing interest rates replaced higher target diameters out of the optimal solution. In contrast to that the treatments with higher target diameter became more important with increasing restrictions concerning budget or ecological constraints.     

Sensitivity of ecosystem goods and services projections of a forest landscape model to initialization data

Projections of indicators of forest ecosystem goods and services (EGS) based on process-based landscape models are critical for adapting forest management to climate change. However, the scarcity of fine-grained, spatially explicit forest data means that initializing these models is both a challenge and a source of uncertainty. To test how different initialization approaches influence the simulation of forest dynamics and EGS indicators we initialized the forest landscape model LandClim with fine resolution empirical data, coarse empirical data, and simulation-derived data, and evaluated the results at three spatial scales (stand, management area and landscape). Simulations were performed for a spruce (Picea abies) dominated landscape in the Black Forest, Germany, under current climate and a climate change scenario. We found that long-term (>150 years) projections are robust to initialization uncertainty. In contrast, shorter-term projections are sensitive to initialization uncertainty, with sensitivity increasing when EGS are assessed at smaller spatial scales, and when the EGS indicators depend on the spatial distribution of individual species. EGS dynamics are strongly influenced by interactions between the density, species composition, and age structure of initialized forests and simulated forest management. If EGS dynamics are strongly influenced by climate change, such as when climate change induces mortality in drought-sensitive species, some of the initialization uncertainty can be masked. We advocate for initializing landscape models with fine-grained data in applications that focus on spatial management problems in heterogeneous landscapes, and stress that the scale of analysis must be in accordance with the accuracy that is warranted by the initialization data.     

Tree species admixture increases ecosystem service provision in simulated spruce- and beech-dominated stands

European Journal of Forest Research - Climate-adaptive forest management aims to sustain the provision of multiple forest ecosystem services and biodiversity (ESB). However, it remains largely...