Umeå universitets logga

Riparian buffer zones along watercourses provide important ecological function by contributing to temperature regulation, erosion control, and biodiversity. At the same time, harvesting restrictions within buffer zones involve an economic trade-off, as potentially harvestable forest is retained. Today, planning of harvesting within buffer zones largely relies on general guidelines and experience-based assessment.

The aim of this thesis was to investigate how a data-driven decision support system for riparian buffer zone planning can be designed using mathematical optimization. The objective was to identify which trees within a buffer zone can be harvested while maintaining a predefined shading requirement.

Three progressively developed optimization models were created. The base model optimized harvested timber volume subject to shading requirements along the watercourse. The model was then extended by including restrictions related to deciduous trees, wind exposure, and erosion risk, while also evaluating shading across the entire buffer zone. Finally, a grid-based model was developed in which shading was evaluated pointwise to provide a more spatially detailed representation. The models were formulated as binary linear optimization problems and implemented in Python.

The results showed that all models enabled improved economic outcomes while maintaining shading requirements. However, increased ecological consideration resulted in reduced harvested volume and economic return. Sensitivity analysis further indicated that the economic outcome was more sensitive to changes in shading requirements than to changes in buffer zone width.

The study indicates that optimization has the potential to function as a decision support tool for riparian buffer zone planning by clarifying trade-offs between ecological and economic objectives.