A rule-based SDSS for integrated forest harvesting planning

The forestry sector in Italy had some problems in the last years: the general abandonment of mountains caused the uncontrolled growth of forests and some problems on their management for preventing hazards like wildfires, the increase of forest work salaries and the decrease of wood value, the concurrence of eastside European countries, the forest certification which protected more the ecological function of forests than their economical value, the increment of social and natural functions of forests as the carbon sinks. All these factors influenced the way of planning cuttings inside forests to the point that sometimes, due to technical difficulties or low wood value and amount, they are not economical and they are not done. But now, the increasing interest of the use of wood for heating or building purposes may increase again the demand and the value of this material. The Forest Operations Planning model helps the forester making decisions about which skidding system is the most viable according to stand assessmental data and geography. The model may also highlight areas which have low forest roads density. The skidding operations have high influence on the total cutting costs so the model considers only the skidding operation, the user will add unit costs for felling operations according to the system used (usually chainsaw or harvester). Five systems are here considered: the tractor with winch or skidder, the forwarder, the cable forwarder and two aerial systems, the mobile tower and the sledge yarder cable cranes. The model was built on a GIS environment with the ArcGis ModelBuilder. It is practically a tool which can be shared with other researchers and modified according to any needs. The input files required to run the model are five: the Digital Terrain Model, the soil classification or stability, the average yearly amount of rain, the forest road network and the assessmental forest stand data. The model was validated comparing results with real working sites done inside the study area or comparing results with other models on different study area. The model evaluation was done checking the influence of parameter variation on output results. The model outputs are several grid maps showing the feasible working area of each system, the technical and optimized distribution of systems with costs (evaluated cell-by-cell and per cubic meter). The statistic tools allow to make stand reports and deep analysis. Comparing model outputs it is possible to evaluate the accessibility of forest and plan the building of new roads to improve the infrastructure and reduce the skidding costs. At the end, two practical examples are reported and some discussion are done about the input data quality and a more site-specific planning.