Advances in machine learning have changed many fields of study and it has also drawn attention in a variety of remote sensing applications. In particular, deep convolutional neural networks (CNNs) have proven very useful in fields such as image recognition; however, the use of CNNs in large-scale remote sensing landcover classifications still needs further investigation. We set out to test CNN-based landcover classification against a more conventional XGBoost shallow learning algorithm for mapping a notoriously difficult group of landcover classes, wetland class as defined by the Canadian Wetland Classification System. We developed two wetland inventory style products for a large (397,958 km2) area in the Boreal Forest region of Alberta, Canada, using Sentinel-1, Sentinel-2, and ALOS DEM data acquired in Google Earth Engine. We then tested the accuracy of these two products against three validation data sets (two photo-interpreted and one field). The CNN-generated wetland product proved to be more accurate than the shallow learning XGBoost wetland product by 5%. The overall accuracy of the CNN product was 80.2% with a mean F1-score of 0.58. We believe that CNNs are better able to capture natural complexities within wetland classes, and thus may be very useful for complex landcover classifications. Overall, this CNN framework shows great promise for generating large-scale wetland inventory data and may prove useful for other landcover mapping applications.

1. Characterizing variation in predator behaviour and, specifically, quantifying kill rates is fundamental for parameterizing predator-prey and food web models. Yet, current methods for recording kill rates of free-ranging predators, particularly those that consume small-bodied (<2 kg) prey, present a number of associated challenges.2. In this paper, we deployed custom-adapted acoustic recorders and tri-axial accelerometers on free-ranging Canada lynx Lynx canadensis to assess the capacity of biologging devices to continuously document individual hunting behaviour, including prey selection and kill rates, on a predator that specializes on prey weighing <2 kg.3. Automated classification of acoustic recordings captured 87% of snowshoe hare kills that were identified through snow-tracking (26 of 31 kills). Classification of detailed acceleration recordings summarized over minutes, instead of seconds, captured consumption of snowshoe hare Lepus americanus, but not smaller species, at high accuracy (F1 = 0.96). 4. By summarizing acoustic and accelerometer data from free-ranging lynx, we demonstrate the capacity of these devices to document within-and betweenindividual variation in diet composition (ranging from 40% to 80% snowshoe hares) and daily feeding bouts (ranging from 0 to 3.5 bouts per day).5. We suggest that acoustic recorders provide a promising method for characterizing several aspects of predator hunting behaviour including prey selection and chase outcomes, while broad-scale accelerometer-based behavioural classifications provide hare kill rates and fine-scale non-hunting behavioural information.Combined, the two technologies provide a means to remotely document both kills and feeding events of small-bodied prey, allowing for individual-based exploration of functional responses, predator-prey interactions and food web dynamics at temporal scales relevant to environmental change.

No abstract

No abstract