Spatial capture-recapture models for the ecological study of common voles (microtus arvalis) in nw spain

  1. Romairone, Juan
Supervised by:
  1. Juan José Luque Larena Director
  2. François Mougeot Co-director

Defence university: Universidad de Valladolid

Fecha de defensa: 25 October 2019

  1. Vittorio Baglione Chair
  2. José María del Arco Montero Secretary
  3. Miguel Delibes Mateos Committee member
  1. Agroforestry Sciences

Type: Thesis


Gaining reliable estimates of population sizes is a fundamental aspect for many, if not most, ecological studies. The methods to do so, are continually adapting and as a result, these new methods must be tailored for particular groups or species to achieve these increasingly robust estimates of population sizes. Such methods, however, require substantial investments, both in time and money, to achieve fruition and alternative methods should be considered when they are sufficiently accurate depending on stakeholder requirements. As such, having rapid cost-effective surveying methods calibrated with the time-intensive methods is a crucial requirement in many applied systems. The common vole (Microtus arvalis) is a facultative agricultural pest that recently invaded farmlands in northwest Spain, causing crop damages and public health risks (e.g. transmission of tularemia to humans) during population outbreaks. There is thus, an urgent need to better understand how vole populations function to prevent and manage outbreaks. In this thesis, state-of-the-art models are applied to field data to fill important knowledge gaps related to field survey methods for common voles. Spatial capture-recapture (SCR) modelling offers a powerful tool to study elusive animals and precisely estimate population density and structure, and to study key basic ecological parameters. The thesis is based on the detailed study of a free-ranging common vole population in an experimental plot of 1.2 ha located in an agricultural landscape of NW Spain and on the use of capture-mark-recapture (CMR) methods applied every month over a two-year period during a density decline phase. I refined aspects of the CMR method and applied the SCR models to precisely estimate density and population structure (i.e. sex-ratio), and to obtain sex-specific insights on animal movement’s space use or behavior. SCR models resolve many of the issues faced by classic capture–recapture (CR) methods, which do not consider the spatial structure of the ecological processes. This thesis consists of three main chapters: (i) the design of a toolbox that allows the application of SCR modelling for the study of small mammals, with a case study on common voles; (ii) a validation and calibrating of an indirect method of estimating vole abundance based on vole activity signs, commonly used for the large-scale monitoring of this farmland rodent pest, and (iii) an assessment of capture biases that may influence estimates of population sex-ratio in the common vole. After testing and adjusting the trapping methodology to the model species, the highlight results were: (i) a 5-fold reduction in trap mortality through the inclusion of a nest box coupled to the trapping device, (ii) the optimization of traps configuration (minimum distance between traps) to accommodate the movement of individual that were more sedentary (females), (iii) a seasonal variation in the detectability of vole activity signs associated with increasing vegetation height and reduced vole activity, (iv) a validation of the use of vole activity signs measured inside the field to predict vole densities up to 100 voles per ha, though with a poorer performance of vole signs measured in field margins, (v) an assessment of bias in adult sex-ratio estimated using capture methods, which are affected by season, sexual differences in baseline detection probability and movement. The results of this thesis allow the improvement of current methods used to estimate common vole density and will enable the study further key ecological parameters in the future. They have also provided a better understanding of the heterogeneities in capturability or detectability and potential bias associated with survey methods, particularly crucial given the study species’ role as both pest and disease vector. Cumulatively, the outcomes of this thesis have the potential to help improve common vole monitoring, but similarly have a broader applicability to other similar species.