The training workshop will focus on Occupancy Modelling.  It will be run by Drs. Riddhika Kalle and T. Ramesh from the Salim Ali Centre for Ornithology and Natural History, India.

The workshop is aimed at beginners and will cover the basic methods for modelling species occurrence using simple presence/absence data. We will focus on the practical use of field data in the freely available program, PRESENCE. The workshop will begin with a lecture followed by practical exercises on how to estimate species occupancy and detection probability using program PRESENCE taking real field data, followed by the selection of appropriate models, and interpreting results. We will focus only on the basic occupancy model i.e single-season model and demonstrate how to analyze detection/non-detection data to determine occupancy and detection probability. We will then model the impact of environmental parameters on species detection and occupancy. By the end of the training workshop, participants will be comfortable in identifying their scientific questions applicable to this technique. Participants will be able to analyze their own data sets from point counts, camera-trap surveys, transects, trail walks, call counts and many more in an occupancy modelling framework.

Occupancy models can be used to estimate the probability that a site is occupied (and other parameters related to changes in occupancy) despite false negatives. In ecological research, simple ‘presence/absence’ data can give useful results, if corrected for detection probability. Such analyses rely on survey data that result from efforts to detect the species of interest and measure site characteristics (variables/covariates). These models use information from repeated observations at each site to estimate detectability. The presence or absence of a species in a set of units (polygons, landscape units, territories, patches, and quadrats) is a fundamental concept in many ecological studies. Visits to the sampled units can result in a positive detection of a species or non-detection of the species in that unit. Therefore, occupancy models were developed to solve the problems created by imperfect detectability (for eg: cryptic species, bad weather conditions, thick bush, level of expertise, and tough terrain while recording the data). However, a species may not always be detected if present which results in “false absences” causing parameter estimates to be biased if unaccounted for, possibly giving misleading results and conclusions.

Single-season occupancy tools are ideal for beginners and other occupancy models are an extension of these models. Our workshop will cover the basic methods for modelling species occurrence using simple presence/absence data. We will focus on the practical use of field data in the freely available program, PRESENCE. The workshop will begin with a lecture followed by practical exercises on how to estimate species occupancy and detection probability using program PRESENCE taking real field data, followed by the selection of appropriate models, and interpreting results. We will focus only on the basic occupancy model i.e single-season model and demonstrate how to analyze detection/non-detection data to determine occupancy and detection probability. We will then model the impact of environmental parameters (categorical as well as continuous factors) on species detection and occupancy. By the end of the course, participants will be comfortable in identifying their scientific questions applicable to this technique. Participants will be able to analyze their own data sets from point counts, camera-trap surveys, transects, trail walks, call counts and many more in an occupancy modelling framework.

Outline:

  • What is occupancy modelling?
  • The role of detectability
  • Including covariates/predictor variables
  • Model selection and multi-model inference

 

The main goals of this workshop are 

  • To conduct basic occupancy modelling in program PRESENCE and R,
  • To give you an overview of the single season occupancy models currently available and
  • To offer hands on experience to help get you started with occupancy studies and one-on-one consultations to address questions regarding your specific project and associated data. This will be done if there is sufficient time available.

 

Prerequisites for the workshop

  • The Occupancy Modelling course will be presented on Wednesday morning following on from, and as an extension of, the camera trapping special session on Tuesday. We expect significant interest in this course but the number of places are limited. Please provide a short (2-3 paragraph) written motivation with your registration as to the reason and benefit of you attending the course. In the case of demand exceeding available places the course conveners will use the motivation to select the attendees. All participants on the course must be registered for both Tuesday 7 November (camera trapping session) and Wednesday 8 November (course) of the Symposium.
  • The target audience: Participants are beginners to occupancy analyses who have little or no experience. No previous experience with program PRESENCE, R or occupancy models is assumed. However, some experience with R coding and programming language might be useful.
  • Bring your own laptop. Before the workshop begins make sure you have installed the latest version of program PRESENCE from http://www.mbr-pwrc.usgs.gov/software/presence.html by clicking on PRESENCE 12.1 in the Downloads sub-section. Install the latest version of program R from https://cran.r-project.org/bin/windows/base/ depending upon your system requirements. After this install the package unmarked from https://cran.r-project.org/web/packages/unmarked/index.html
  • Before attending our workshop we suggest participants to read the following papers/book chapters on occupancy modelling to gain familiarity with the methods.

 

Recommended reading prior to the workshop

  1. Program Presence http://www.mbr-pwrc.usgs.gov/software/presence.html
    http://www.uvm.edu/rsenr/vtcfwru/spreadsheets/?Page=occupancy/occupancy.htm
  2. MacKenzie et al. 2006. Occupancy Estimation and Modeling: Inferring Patterns and Dynamics of Species Occurrence. http://www.uvm.edu/rsenr/vtcfwru/spreadsheets/?Page=occupancy/occupancy4.htm
  3. Ramesh, T, and Colleen T. Downs 2014. Land use factors determining occurrence of Red-necked Spurfowl (Pternistis afer) in the Drakensberg Midlands, South Africa.Journal of Ornithology155(2): 471-480.
  4. Ramesh, T., Kalle, R., Sankar, K., & Qureshi, Q. 2013. Dry season factors determining habitat use and distribution of mouse deer (Moschiola indica) in the Western Ghats. European Journal of Wildlife Research, 59(2), 271-280.
  5. MacKenzie, D.I. 2006. Modeling the probability of resource use: the effect of, and dealing with, detecting a species imperfectly. Journal of Wildlife Management. http://ag.arizona.edu/classes/wfsc578/MacKenzie%202006.%20Modeling%20probability%20of%20resource%20use.pdf
  6. MacKenzie, D.I. 2005. Was it there? Dealing with imperfect detection for species presence/absence data. Australian and New Zealand Journal of Statistics 47: 65-74. http://onlinelibrary.wiley.com/doi/10.1111/j.1467-842X.2005.00372.x/epdf