Search Results (5)

With in situ imaging systems becoming more common, precise, and economically viable, use of these systems has grown dramatically, including both automated classification and biomass estimations. However, a rather large and overlooked portion of these efforts is reliable detection and classification of these organisms as they pass through the imaging device. This paper focuses on the development of an end-to-end classification CNN-based algorithm for marine zooplankton using the in situ Ichthyoplankton Imaging System (ISIIS-DPI) from Bellamare (La Jolla, CA, USA). Our novel approach considers many issues with automated segmentation and classification, including over-segmentation, noise segmentation, and organism size input. This allows for classifications in diverse water types, demonstrated by the comparison of three datasets created in conjunction with this project, each with very different water properties and zooplankton communities (Florida Gulf coast; Trondheimsfjord, Norway; Sargasso Sea). Our segmented image dataset contains 70,624 regions of interest (ROIs) across four organism classes—Chaetognath, Crustacean, Gelatinous, and Larvacean—with two classes dedicated to detritus. Four common network architectures—Resnet, Xception, GoogleNet, and Darknet—are trained on this dataset, with final test accuracies in the range of 95.94–96.09%. Following this initial training, a secondary level of classification is introduced. The base Gelatinous class is further divided into six groups. The same four CNN architectures are used once again, with final accuracies in the range of 86.12–90.40%, showing the ability to taxonomically classify down to the order level. The present work introduces a versatile, adaptable, scalable and autonomous segmentation and classification algorithm using niched networks mirroring taxonomy, and is fully contained in a publicly available MATLAB R2025a custom graphical user interface. Full article

Appendicularians are important but remain poorly studied groups of zooplankton in polar regions. The present research is based on samples collected in Admiralty Bay (King George Island) during a year-long period. Six larvacean species were noted, among which Fritillaria borealis and Oikopleura gaussica were found to be the most numerous, while the other species were relatively rare. Fritillaria borealis was a dominant part of the late summer (warm water) community, while O. gaussica had the highest presence in the winter (cold water) community. The abundance of appendicularians recorded in the bay was less numerous than that described by other authors. The most important factors influencing annual changes in the larvaceans in the bay was season, but only in the case of the two species. These facts were probably linked to the very dynamic changes in the abiotic conditions in the fjord, and the influx of specific masses of water. Full article

Appendicularians are one of the most common animals found within zooplankton assemblages. They play a very important role as filter feeders but are, unfortunately, inconsistently reported in the Antarctic literature. The present paper attempts to describe the zonal diversity of appendicularians and the main environmental factors influencing their communities in the Drake Passage. Samples were collected during Antarctic summer in 2009–2010. A total of eight species of larvaceans were identified. Fritillaria borealis was the species found in the highest numbers in almost the entire studied area, and was observed at all sampling stations. The distributions of other taxa were limited to specific hydrological zones and hydrological conditions. F. fraudax and Oikopleura gaussica were typical of the areas between the Polar Front and the Subantarctic Front zones, and their distributions were significantly correlated with temperature and salinity, likely making them good indicator species. The F. fusiformis distribution was strictly related to South American waters. In summary, temperature was the strongest environmental factor influencing the larvacean community structure in the Drake Passage, and we also found that testing environmental factors on larvaceans as a whole group did not give entirely reliable results. Full article

Flexible propulsors are ubiquitous in aquatic and flying organisms and are of great interest for bioinspired engineering. However, many animal models, especially those found in the deep sea, remain inaccessible to direct observation in the laboratory. We address this challenge by conducting an integrative study of the giant larvacean, an invertebrate swimmer and “fluid pump” of the mesopelagic zone. We demonstrate a workflow involving deep sea robots, advanced imaging tools, and numerical modeling to assess the kinematics and resulting fluid transport of the larvacean’s beating tail. A computational model of the tail was developed to simulate the local fluid environment and the tail kinematics using embedded passive (elastic) and active (muscular) material properties. The model examines how varying the extent of muscular activation affects the resulting kinematics and fluid transport rates. We find that muscle activation in two-thirds of the tail’s length, which corresponds to the observed kinematics in giant larvaceans, generates a greater average downstream flow speed than other designs with the same power input. Our results suggest that the active and passive material properties of the larvacean tail are tuned to produce efficient fluid transport for swimming and feeding, as well as provide new insight into the role of flexibility in biological propulsors. Full article

Tunicates or urochordates—comprising ascidians, larvaceans, and salps—are the only metazoans that can synthesize cellulose, a biological function usually associated with bacteria and plants but not animals. Tunicate cellulose or tunicine is a major component of the outer acellular coverage (tunic) of the entire body of these organisms. Previous studies have suggested that the prokaryotic cellulose synthase gene (CesA) was horizontally transferred into the genome of a tunicate ancestor. However, no convenient tools have been devised to determine whether only tunicates harbor CesA. ORTHOSCOPE is a recently developed tool used to identify orthologous genes and to examine the phylogenic relationship of molecules within major metazoan taxa. The present analysis with this tool revealed the presence of CesA orthologs in all sequenced tunicate genomes but an absence in other metazoan genomes. This supports an evolutionary origin of animal cellulose and provides insights into the evolution of this animal taxon. Full article