 Modeling habitat connectivity in support of multiobjective species movement: An application to amphibian habitat systemsTimothy C Matisziw, Ashkan Gholamialam and Kathleen M Trauth PLOS Computational Biology, 2020, vol. 16, issue 12, 1-23 Abstract: Reasoning about the factors underlying habitat connectivity and the inter-habitat movement of species is essential to many areas of biological inquiry. In order to better describe and understand the ways in which the landscape may support species movement, an increasing amount of research has focused on identification of paths or corridors that may be important in providing connectivity among habitat. The least-cost path problem has proven to be an instrumental analytical tool in this sense. A complicating aspect of such path identification methods is how to best reconcile and integrate the array of criteria or objectives that species may consider in traversal of a landscape. In cases where habitat connectivity is thought to be influenced or guided by multiple objectives, numerous solutions to least-cost path problems can exist, representing tradeoffs between the objectives. In practice though, identification of these solutions can be very challenging and as such, only a small proportion of them are typically examined leading to a weak characterization of habitat connectivity. To address this computational challenge, a multiobjective optimization framework is proposed. A generalizable multiobjective least-cost path model is first detailed. A non-inferior set estimation (MONISE) algorithm for identifying supported efficient solutions to the multiobjective least-cost path model is then described. However, it is well known that unsupported efficient solutions (which are equally important) can also exist, but are typically ignored given that they are more difficult to identify. Thus, to enable the identification of the full set of efficient solutions (supported and unsupported) to the multiobjective model, a multi-criteria labeling algorithm is then proposed. The developed framework is applied to assess different conceptualizations of habitat connectivity supporting amphibian movement in a wetland system. The results highlight the range of tradeoffs in characterizations of connectivity that can exist when multiple objectives are thought to contribute to movement decisions and that the number of unsupported efficient solutions (which are typically ignored) can vastly outweigh that of the supported efficient solutions.Author summary: Biological studies and experiments have provided many insights as to the complex nature of the criteria (objectives) that species may evaluate in decisions regarding inter-habitat landscape traversal. Those insights are often used to develop models for identifying paths or corridors potentially supporting inter-habitat movement to characterize habitat connectivity from the species’ perspective. However, typically only a small proportion of the alternative paths that may also support species movement and be important to habitat connectivity are identified. Thus, the computational challenge is to develop methods that can more completely characterize the set of paths/corridors that may support habitat connectivity. To address this challenge, a modeling framework that better integrates the objectives thought to influence inter-habitat movement is outlined, facilitating the identification of a broader set of paths reflecting the tradeoffs among the objectives. Through an application of the developed framework to model habitat connectivity for multiobjective amphibian movement in a wetland system, it is demonstrated that an extensive and diverse set of efficient inter-habitat paths can be identified. The capability to characterize these additional dimensions of habitat connectivity supporting species movement can provide researchers and practitioners with a means to develop more robust representations of complex biological systems. Date: 2020 Downloads: (external link) Related works: Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text Persistent link: https://EconPapers.repec.org/RePEc:plo:pcbi00:1008540 DOI: 10.1371/journal.pcbi.1008540 Access Statistics for this article More articles in PLOS Computational Biology from Public Library of Science |
Is your work missing from RePEc? Here is how to contribute.
Questions or problems? Check the EconPapers FAQ or send mail to .
EconPapers is hosted by the
School of Business at Örebro University.