Search

Increased human-induced environmental changes and global warming alter bird migration timing and routes. Recently, many Holarctic species, including white storks, Ciconia ciconia , were reported to overwinter at higher latitudes, closer to breeding grounds. We aimed to understand the causes and implications of this phenomenon by examining bird survival and behaviour during overwintering in Europe versus Africa. We compared GPS and body acceleration data of 54 juvenile (first-year) white storks that originated from the same European natal population and overwintered in Europe or Africa. All six juveniles that overwintered in Europe survived through their first year, which was significantly higher than only 38% of the 48 overwintering in the species' traditional grounds in Africa. During overwintering, storks in Europe differed from those in Africa by (1) reducing movement and foraging range, (2) spending less time flying and more time resting, thus using less energy (estimated from overall dynamic body acceleration) and (3) reducing foraging effort, while relying more on anthropogenic resources (landfills and agricultural areas). Timing affected overwintering site as juveniles that overwintered in Europe hatched and started migrating later. We emphasize, however, that late hatching by itself did not yield a survival benefit as not all late juveniles curtailed their migration. We suggest that wintering in Europe was less demanding compared to Africa which may explain the increased survival of juveniles that wintered in Europe. Our findings correspond to the general increase in the European wintering population of white storks, and shed light on the contemporary trend of shortened bird migration; a phenomenon with potentially broad ecological implications. [ABSTRACT FROM AUTHOR]

The implementation of intelligent software to identify and classify objects and individuals in visual fields is a technology of growing importance to operatives in many fields, including wildlife conservation and management. To non-experts, the methods can be abstruse and the results mystifying. Here, in the context of applying cutting edge methods to classify wildlife species from camera-trap data, we shed light on the methods themselves and types of features these methods extract to make efficient identifications and reliable classifications. The current state of the art is to employ convolutional neural networks (CNN) encoded within deep-learning algorithms. We outline these methods and present results obtained in training a CNN to classify 20 African wildlife species with an overall accuracy of 87.5% from a dataset containing 111,467 images. We demonstrate the application of a gradient-weighted class-activation-mapping (Grad-CAM) procedure to extract the most salient pixels in the final convolution layer. We show that these pixels highlight features in particular images that in some cases are similar to those used to train humans to identify these species. Further, we used mutual information methods to identify the neurons in the final convolution layer that consistently respond most strongly across a set of images of one particular species. We then interpret the features in the image where the strongest responses occur, and present dataset biases that were revealed by these extracted features. We also used hierarchical clustering of feature vectors (i.e., the state of the final fully-connected layer in the CNN) associated with each image to produce a visual similarity dendrogram of identified species. Finally, we evaluated the relative unfamiliarity of images that were not part of the training set when these images were one of the 20 species "known" to our CNN in contrast to images of the species that were "unknown" to our CNN.