Investigating landscape resistance using a connectivity modelling method
Cottet, R., 2021: Investigating landscape resistance using a connectivity modelling method. Master thesis. 83 p.
Cottet, R., 2021: Investigating landscape resistance using a connectivity modelling method. Master Thesis ETHZ, D-USYS. Supervisors: PD Dr. Janine Bolliger, Dr. Peter M. Bach.
The concept of blue-green infrastructure has been put forward as a tool to preserve biodiversity and ecosystem services from detrimental human activities and allow cohabitation. Among the different factors determining the efficiency of those infrastructures, their connectivity has been found to be of major importance. Many models have been created to assess the capacity of the landscape to favor or hinder movement of species. Among those, Circuitscape has been widely used in ecology and other fields. However, the resistance surfaces used as input by the software have sometimes been criticised because of the partly subjective knowledge included in estimating them. By applying an existing resistance modelling method to a network representing the landscape, I could explore numerous resistance models for various variables and investigate the role they played for connectivity. Five Swiss amphibian species were used as case study species. I found that the determinant variable underlying landscape connectivity were specific to each species. I could nonetheless identify the shortest distance to water and the slope to be dominant factors, closely followed by the traffic. The land-cover appeared to be a poor predictor of landscape connectivity due to a loss of information when transposing the data to the network. These results illustrate the crucial importance of carefully modelling resistance when carrying out a connectivity analysis and show how much can be learned about the role of the different variables for connectivity in the process. Overall, this study makes a strong case for the opportunities of the method for connectivity modelling in a blue-green infrastructure context.