Review of capture–recapture methods applicable to noninvasive genetic sampling



    Corresponding author
    1. Colorado Cooperative Fish and Wildlife Research Unit, Department of Fishery and Wildlife Biology, Colorado State University, Fort Collins, CO 80523, USA,
    Search for more papers by this author

    1. US Geological Survey, Colorado Cooperative Fish and Wildlife Research Unit, Colorado State University, Fort Collins, CO 80523, USA
    Search for more papers by this author

  • Box 1 The Robust Design

    The robust design (Pollock 1982; Kendall et al. 1997) is one of the most useful sampling designs for capture–recapture studies. The robust design refers both to a sampling design and the class of models used for its analysis. The design consists of two levels of sampling periods. The primary sampling occasions are separated by a relatively long period of time over which the population is assumed to be demographically and geographically open, this means that birth, death, immigration and emigration all may occur. At each primary sampling occasion, there are several secondary sampling occasions. Secondary sampling occasions occur over a very short time interval. The population is assumed to be closed between secondary sampling occasions, although generalizations to this exist. The robust design allows abundance, temporary emigration/immigration rates, and survival to be estimated. Capture probability is estimated by secondary sampling occasion, therefore allowing a robust estimate of abundance for each primary occasion.

    Across secondary sampling occasions within a primary occasion, it is best to actively try to give all individuals a chance of being detected. For example, it could be useful to change the type of scent lure used on different occasions to appeal to a broader range of the population. In addition, it is useful to have multiple sampling locations within an individuals home range to increase the probability of detecting an individual.

  • Box 2 Software available for capture–recapture analysis

    mark — Software for open- and closed-population capture–recapture and patch occupancy analysis. mark has taken the place of many older programs such as capture, release, and brownie. mark includes models specifically designed to estimate genotyping error rate. It is freely available at

    m-surge— Software for multistate or multistrata problems or other open-population capture–recapture for which the multistrata model is a generalization. m-surge is freely available at

    popan— Software for the analysis of Jolly–Seber model problems. popan is freely available at and also available within mark.

    presence— Software for the analysis of patch occupancy data. presence includes variations on patch occupancy models such as multispecies occupancy, robust design and latent abundance. It is freely available at

    Analysis Software Forum— An online forum and other electronic resources for capture–recapture software and analysis are available at

Paul M. Lukacs, US Geological Survey, Patuxent Wildlife Research Center, 1201 Oakridge Dr., Suite 250, Fort Collins, CO 80525, USA. Fax: 970-225-3597; E-mail:


The use of noninvasive genetic sampling to identify individual animals for capture–recapture studies has become widespread in the past decade. Strong emphasis has been placed on the field protocols and genetic analyses with fruitful results. Little attention has been paid to the capture–recapture application for this specific type of data beyond stating the effects of assumption violations. Here, we review the broad class of capture–recapture methods that are available for use with DNA-based capture–recapture data, noting the array of biologically interesting parameters such as survival, emigration rates, state transition rates and the finite rate of population change that can be estimated from such data. We highlight recent developments in capture–recapture theory specifically designed for noninvasive genetic sampling data.