A Physics-Aware Neural Network Approach for Flow Data Reconstruction From Satellite Observations

An accurate assessment of physical transport requires high-resolution and high-quality velocity information. In satellite-based wind retrievals, the accuracy is impaired due to noise while the maximal observable resolution is bounded by the sensors. The reconstruction of a continuous velocity field is important to assess transport characteristics and it is very challenging. A major difficulty is ambiguity, since the lack of visible clouds results in missing information and multiple velocity fields will explain the same sparse observations. It is, therefore, necessary to regularize the reconstruction, which would typically be done by hand-crafting priors on the smoothness of the signal or on the divergence of the resulting flow. However, the regularizers can smooth the solution excessively and will not guarantee that possible solutions are truly physically realizable. In this paper, we demonstrate that data recovery can be learned by a neural network from numerical simulations of physically realizable fluid flows, which can be seen as a data-driven regularization. We show that the learning-based reconstruction is especially powerful in handling large areas of missing or occluded data, outperforming traditional models for data recovery. We quantitatively evaluate our method on numerically-simulated flows, and additionally apply it to a Guadalupe Island case study—a real-world flow data set retrieved from satellite imagery of stratocumulus clouds.

Statistical modelling reveals spatial, temporal, and environmental preferences for white sharks at an oceanic aggregation site

Guadalupe Island, Mexico, is one of the most important white shark (Carcharodon carcharias) aggregation sites in the Eastern Pacific. In the waters surrounding Guadalupe Island, cage diving has been carried out since 2001 during August-November; however, there is scarce information regarding the factors associated with this seasonal aggregation. The purpose of this study was to describe the probability of occurrence of white sharks relative to spatial, temporal, and environmental factors in Guadalupe Island. Generalized additive models (GAMs) were used to describe the effect of sea surface temperature, water visibility, tide, moon phase, cloud cover, time of day, and location on white shark occurrence. GAMs were generated from a data set of 6266 sightings of white sharks, classified as immature males, mature males, immature females, and mature females. A sexual segregation related to month was observed, where females arrived after males during late September. GAMs evidenced a segregation of white sharks according to the analysed variables, which is consistent with previous observations in this locality. Environmental preferences for each white shark category are potentially influenced by feeding habits, sexual maturation, and reproduction. This study constitutes a baseline of the effect of the environment on the occurrence of white sharks in Guadalupe Island, which can be used in further studies regarding management and conservation in future climatic and anthropogenic scenarios. Its relevance is related to the understanding of its ecology in oceanic environments and the presence of this threatened species during the ecotourism season.

Evidence of interactions between white sharks and large squids in Guadalupe Island, Mexico

Shark-cephalopod interactions have been documented in trophic ecology studies around the world. However, there is little information about the encounters between white sharks Carcharodon carcharias and squids in the eastern North Pacific Ocean. Here we provide evidence of interactions between white sharks and large squids in the waters of Guadalupe Island, Mexico. Through the use of non-invasive techniques, we found the presence of evident scars made by large squids on the body of the white sharks, mainly on the head and trunk, of at least 14 sharks recorded during August–December in the years 2008, 2012, 2013, 2017 and 2019. The mean length of the white sharks was 3.7 m (SD ± 0.6; total length), although the majority of the sharks with scars were adult and subadult males (n = 9; 64%). One of these males was photographically recaptured during the same season in which the individual showed new scars, confirming that the squid-white shark interaction likely occurs near Guadalupe Island. Our results highlight the importance of the twilight zone for white sharks and the use of shared habitat and trophic interactions between squid and white sharks, in which future ecosystem studies should consider both species for management and conservation purposes.

Ontogenetic differences in muscle fatty acid profile of white sharks Carcharodon carcharias off Guadalupe Island, México

Carcharodon carcharias is a protected species occupying the apex of most marine foodwebs where they are present. Aggregation areas, such as Guadalupe Island, México, that have been related to feeding events, are of special interest for this species conservation. The aim of this study was to describe the fatty acid profile of C. carcharias' muscle for the first time on Guadalupe Island, using non-lethal biopsy methods to determine ontogenetic and sex differences in relation to diet and habitat use. Fatty acid profiles and biomarkers from different individuals are explored as a source of integrated information of their diet. Analysis of the fatty acid composition of individuals with varying total lengths (2.3-5.0 m) suggested a dietary shift between juveniles and adults occurring at aproximately 3 m. Fatty acid biomarkers indicated a higher degree of carnivorism in adults than in juveniles. Additionally, these ecological tracers suggested that juveniles feed in shallow waters close to the coast, while adults feed in deep waters along inshore and offshore areas. This study represents a first step towards using fatty acid composition as a relevant tool for further understanding dietary shifts and habitat use throughout the ontogeny of C. carcharias. However, to corroborate this, further studies with larger sample sizes are required.

Saildrone: Adaptively Sampling the Marine Environment

From 11 April to 11 June 2018 a new type of ocean observing platform, the Saildrone surface vehicle, collected data on a round-trip, 60-day cruise from San Francisco Bay, down the U.S. and Mexican coast to Guadalupe Island. The cruise track was selected to optimize the science team’s validation and science objectives. The validation objectives include establishing the accuracy of these new measurements. The scientific objectives include validation of satellite-derived fluxes, sea surface temperatures, and wind vectors and studies of upwelling dynamics, river plumes, air–sea interactions including frontal regions, and diurnal warming regions. On this deployment, the Saildrone carried 16 atmospheric and oceanographic sensors. Future planned cruises (with open data policies) are focused on improving our understanding of air–sea fluxes in the Arctic Ocean and around North Brazil Current rings.