Research

My research happens at the interface between ecology, conservation, statistics and mathematics. My general interest is the development and application of quantitative tools to answer ecological questions and address conservation problems. My main lines of research are encompassed by the three overlapping themes below.


1. Occupancy-detection modelling

The detection of species during surveys is often imperfect, and therefore accounting for detectability can provide a more reliable estimation of species occupancy rates and species distributions. Within the area of occupancy-detection modelling, I have pursued work along different lines, with the main topics being:

  • Model developments: models for data collected along transects or during an interval of time (e.g. camera traps), which describe the detection process as a point process [6,9 in ‘Publications’]

  • Study design: trade-off between number of sampling sites and survey effort per site [4,10,15]; impact of sampling with replacement in studies with spatial replication [7]

  • Model performance: evaluation of how the occupancy-detection model performs under different conditions [4,14,17]  <- check out my little SODA program!

  • Applications: analysis of data from different taxa including lemurs, tigers, tapirs, and birds [2,8,12,13]

I largely developed my interest in these topics working with Byron Morgan and Martin Ridout during my PhD. Since then, I have continued researching in this area including new collaborators such as José Lahoz-Monfort, Mick McCarthy and Brendan Wintle. Although most of my work has been in occupancy modelling, I am as well interested in related hierarchical models for abundance, richness and multiple states.


2. Species distribution modelling (SDM)

Beyond the implications of imperfect detection, I also carry out work in the area of species distribution modelling from a broader point of view. My main interests lie in promoting the right use of existing modelling tools, as well as in the interface between species distribution modelling and conservation applications such as spatial prioritisation or optimal monitoring. My main collaborators in this area are José Lahoz-MonfortJane Elith and Brendan Wintle


3. Detectability and its implications 

Apart from its relevance for estimating species occurrence and distributions, accounting for detectability is important in other contexts. For instance, knowledge about detectability is fundamental to set appropriate minimum survey effort requirements for impact assessments. My work around detectability and its implication include the following recent or ongoing projects:

  • setting minimum survey effort requirements for impact assessments considering the cumulative effects of development, with Brendan WintleMick McCarthy and José Lahoz-Monfort,

  • developments of a method for setting the optimal surveillance of invasive species, with Cindy Hauser and Mick McCarthy [16]

  • an explicit model that links species detectability and abundance, with Mick McCarthy and Madeleine Gorsuch (Master’s student)

Advertisements