RESUMO
Rehabilitation of injured or immature individuals has become an increasingly used conservation and management tool. However, scientific evaluation of rehabilitations is rare, raising concern about post-release welfare as well as the cost-effectiveness of spending scarce financial resources. Over the past 20 years, events of juvenile Eurasian lynx presumably orphaned have been observed in many European lynx populations. To guide the management of orphaned lynx, we documented survival, rehabilitation and fate after the release and evaluated the potential relevance of lynx orphan rehabilitation for population management and conservation implications. Data on 320 orphaned lynx was collected from 1975 to 2022 from 13 countries and nine populations. The majority of orphaned lynx (55%) were taken to rehabilitation centres or other enclosures. A total of 66 orphans were released back to nature. The portion of rehabilitated lynx who survived at least one year after release was 0.66. Release location was the best predictor for their survival. Of the 66 released lynx, ten have reproduced at least once (8 females and 2 males). Conservation implications of rehabilitation programmes include managing genetic diversity in small, isolated populations and reintroducing species to historical habitats. The lynx is a perfect model species as most reintroduced populations in Central Europe show significantly lower observed heterozygosity than most of the autochthonous populations, indicating that reintroduction bottlenecks, isolation and post-release management have long-term consequences on the genetic composition of populations. The release of translocated orphans could be a valuable contribution to Eurasian lynx conservation in Europe. It is recommended to release orphans at the distribution edge or in the frame of reintroduction projects instead of a release in the core area of a population where it is not necessary from a demographic and genetic point of view. Rehabilitation programmes can have conservation implications that extend far beyond individual welfare benefits.
Assuntos
Lynx , Humanos , Masculino , Animais , Feminino , Lynx/genética , Europa (Continente) , Ecossistema , Centros de ReabilitaçãoRESUMO
This paper presents the evaluation of 36 convolutional neural network (CNN) models, which were trained on the same dataset (ImageNet). The aim of this research was to evaluate the performance of pre-trained models on the binary classification of images in a "real-world" application. The classification of wildlife images was the use case, in particular, those of the Eurasian lynx (lat. "Lynx lynx"), which were collected by camera traps in various locations in Croatia. The collected images varied greatly in terms of image quality, while the dataset itself was highly imbalanced in terms of the percentage of images that depicted lynxes.