Analysis of photo‐id data allowing for missed matches and individuals identified from opposite sides

摘要: Summary In many species, photo-identification could be used as an alternative to artificial marking provide data on demographic parameters. However, unless the population is very small or fragmented, software may required pre-screen and reject most image pairs potential matches. Depending species method obtain images, currently available falsely some matches. We estimate false rejection rate (FRR) of ExtractCompare (EC) program when images female grey seals. Filtering manually reduce FRR involves subjective assessment quality, reduces amount bias results in favour relatively well-marked individuals. The contain individuals identified only from left side right side, well both sides. Missed matches resulting rejections by pre-screening and/or inclusion opposite sides cause generate multiple encounter histories. We describe open model for this type which, given a measured risk missing match between randomly selected pair same individual, provides maximum likelihood (ML) estimates initial size, survival/emigration immigration/recruitment calculating expected frequency any history that generated. As case study method, we EC photographs seals breeding colony histories over five successive seasons. Allowing FRR, calculated ML comparison with previous studies. We also simulated using give mixture derived values drive simulation. With set at up 33%, gave abundance survival parameters simulation were biased 4·7% 3% down, respectively. The seal consistent apparent trends abundance.