Field of application

Issues that can be addressed with EFISCEN are for example effects of expected changes in wood demand, forest area, forest management or increment level. Effects can be expressed in the form of various characteristics or indicators of the forest resource. The main model outputs consist of:

  • Forest characteristics (e.g. forest area, tree species, stemwood volume, increment, age-class distribution and natural mortality) by forest type;
  • Harvested wood/biomass by forest and harvest type (thinnings, final fellings, logging residues and stumps);
  • Carbon stocks in forest biomass (stem, branches, foliage, coarse roots, fine roots) and soil by forest type;
  • Ecosystem services (carbon sequestration, biodiversity preservation (dead wood), recreation, moderation of wind and fire risk.

EFISCEN is suitable for the projection of forest resources for periods up to about 50-60 years. The model is designed for large forest areas, such as provinces or countries. Application to smaller areas is possible, but there have been no studies yet to determine the minimum size and effects of scale on uncertainty of the projections. Generally, several thousand hectares could be regarded as a safe minimum.

EFISCEN has been developed for even-aged, managed forests. Deviations from this situation (e.g. uneven-aged forests, unmanaged forests and shelterwood systems) make the application of EFISCEN less suitable. Furthermore, the model is currently not suited to simulate fast growing tree species with very short rotations, due to the 5-year time step.

As with all models, uncertainties in EFISCEN depend largely on the quality of the input data. Especially a correct estimation of the increment functions is important for the model outcomes. Initial uncertainties propagate through the model with every simulated time step, and thus the overall uncertainty increases. For 10-12 time steps (50-60 years) the model is believed to give reasonable projections. With increasing projection length, observed patterns become more important than absolute values.