Conservation Prioritization Problems and their Shadow Prices

Kaim, Andrea
Abstract in English: 
Systematic conservation planning is an essential part of biodiversity preservation. In the context of conservation prioritization problems, the total cost of the entire reserve system is highly dependent on how big we set targets (e.g. 10% or 30%) for conservation features (e.g. species or habitats). Thus, it is of interest for conservation planners, how targets could be adjusted in a reasonable way in order to decrease total cost. The aim is to give a feature ranking based on their in uence on the latter. Focusing on the minimum set coverage problem { an integer linear optimization problem (ILP) { this thesis presents a method to rank features according to their in uence on total cost. Since the computation time is often too high to solve the ILP, its optimal solutions are approximated by the results of a linear optimization problem (LP). The shadow prices of the LP are used for the feature ranking which is compared to additional rankings. These are created by methods which used an ILP solver and the software Marxan which is based on a simulated annealing algorithm. The results showed that for the minimum set coverage problem shadow prices can be used to create an approximate feature ranking of impact on total cost. Furthermore many planning units selected for conservation by Marxan and the LP solver were the same. These results can be useful to improve Marxan. Additionally, the feature ranking provides a new supporting tool for decision makers in conservation planning.
kaim_diploma_thesis_links_signed.pdf1.2 MB