Deep inference of seabird dives from GPS-only records: Performance and generalization properties.

At-sea behaviour of seabirds have received significant attention in ecology over the last decades as it is a key process in the ecology and fate of these populations. It is also, through the position of top predator that these species often occupy, a relevant and integrative indicator of the dynamics of the marine ecosystems they rely on. Seabird trajectories are recorded through the deployment of GPS, and a variety of statistical approaches have been tested to infer probable behaviours from these location data. Recently, deep learning tools have shown promising results for the segmentation and classification of animal behaviour from trajectory data. Yet, these approaches have not been widely used and investigation is still needed to identify optimal network architecture and to demonstrate their generalization properties. From a database of about 300 foraging trajectories derived from GPS data deployed simultaneously with pressure sensors for the identification of dives, this work has benchmarked deep neural network architectures trained in a supervised manner for the prediction of dives from trajectory data. It first confirms that deep learning allows better dive prediction than usual methods such as Hidden Markov Models. It also demonstrates the generalization properties of the trained networks for inferring dives distribution for seabirds from other colonies and ecosystems. In particular, convolutional networks trained on Peruvian boobies from a specific colony show great ability to predict dives of boobies from other colonies and from distinct ecosystems. We further investigate accross-species generalization using a transfer learning strategy known as 'fine-tuning'. Starting from a convolutional network pre-trained on Guanay cormorant data reduced by two the size of the dataset needed to accurately predict dives in a tropical booby from Brazil. We believe that the networks trained in this study will provide relevant starting point for future fine-tuning works for seabird trajectory segmentation.

ENSO Climate Forcing of the Marine Mercury Cycle in the Peruvian Upwelling Zone Does Not Affect Methylmercury Levels of Marine Avian Top Predators.

Climate change is expected to affect marine mercury (Hg) biogeochemistry and biomagnification. Recent modeling work suggested that ocean warming increases methylmercury (MeHg) levels in fish. Here, we studied the influence of El Niño Southern Oscillations (ENSO) on Hg concentrations and stable isotopes in time series of seabird blood from the Peruvian upwelling and oxygen minimum zone. Between 2009 and 2016, La Niña (2011) and El Niño conditions (2015-2016) were accompanied by sea surface temperature anomalies up to 3 °C, oxycline depth change (20-100 m), and strong primary production gradients. Seabird Hg levels were stable and did not co-vary significantly with oceanographic parameters, nor with anchovy biomass, the primary dietary source to seabirds (90%). In contrast, seabird Δ199Hg, proxy for marine photochemical MeHg breakdown, and δ15N showed strong interannual variability (up to 0.8 and 3‰, respectively) and sharply decreased during El Niño. We suggest that lower Δ199Hg during El Niño represents reduced MeHg photodegradation due to the deepening of the oxycline. This process was balanced by equally reduced Hg methylation due to reduced productivity, carbon export, and remineralization. The non-dependence of seabird MeHg levels on strong ENSO variability suggests that marine predator MeHg levels may not be as sensitive to climate change as is currently thought.

Machine and deep learning approaches to understand and predict habitat suitability for seabird breeding.

The way animals select their breeding habitat may have great impacts on individual fitness. This complex process depends on the integration of information on various environmental factors, over a wide range of spatiotemporal scales. For seabirds, breeding habitat selection integrates both land and sea features over several spatial scales. Seabirds explore these features prior to breeding, assessing habitats' quality. However, the information-gathering and decision-making process by seabirds when choosing a breeding habitat remains poorly understood. We compiled 49 historical records of larids colonies in Cuba from 1980 to 2020. Then, we predicted potentially suitable breeding sites for larids and assessed their breeding macrohabitat selection, using deep and machine learning algorithms respectively. Using a convolutional neural network and Landsat satellite images we predicted the suitability for nesting of non-monitored sites of this archipelago. Furthermore, we assessed the relative contribution of 18 land- and marine-based environmental covariates describing macrohabitats at three spatial scales (i.e. 10, 50 and 100 km) using random forests. Convolutional neural network exhibited good performance at training, validation and test (F1-scores >85%). Sites with higher habitat suitability (p > .75) covered 20.3% of the predicting area. Larids breeding macrohabitats were sites relatively close to main islands, featuring sparse vegetation cover and high chlorophyll-a concentration at sea in 50 and 100 km around colonies. Lower sea surface temperature at larger spatial scales was determinant to distinguish the breeding from non-breeding sites. A more comprehensive understanding of the seabird breeding macrohabitats selection can be reached from the complementary use of convolutional neural networks and random forest models. Our analysis provides crucial knowledge in tropical regions that lack complete and regular monitoring of seabirds' breeding sites.

Ecological trap for seabirds due to the contamination caused by the Fundão dam collapse, Brazil.

Human-induced rapid environmental changes can disrupt habitat quality in the short term. A decrease in quality of habitats associated with preference for these over other available higher quality is referred as ecological trap. In 2015, the Fundão dam containing iron mining tailings, eastern Brazil, collapsed and released about 50 million cubic meters of metal-rich mud composed by Fe, As, Cd, Hg, Pb in three rivers and the adjacent continental shelf. The area is a foraging site for dozens of seabird and shorebird species. In this study, we used a dataset from before and after Fundão dam collapse containing information on at-sea distribution during foraging activities (biologging), dietary aspects (stable isotopes), and trace elements concentration in feathers and blood from three seabird species known to use the area as foraging site: Phaethon aethereus, Sula leucogaster, and Pterodroma arminjoniana. In general, a substantial change in foraging strategies was not detected, as seabirds remain using areas and food resources similar to those used before the dam collapse. However, concentration of non-essential elements increased (e.g., Cd and As) while essential elements decreased (e.g., Mn and Zn), suggesting that the prey are contaminated by trace elements from tailings. This scenario represents evidence of an ecological trap as seabirds did not change habitat use, even though it had its quality reduced by contamination. The sinking-resuspension dynamics of tailings deposited on the continental shelf can temporally increase seabird exposure to contaminants, which can promote deleterious effects on populations using the region as foraging sites in medium and long terms.