A new species of deep-sea grunt, Rhonciscus pauco (Lutjaniformes: Haemulidae), from Puerto Rico.

A fourth species of the genus Rhonciscus (Lutjaniformes: Haemulidae) is described from various specimens collected by small-scale fishers from the insular upper slope of western Puerto Rico. The new species was molecularly recovered as sister to the Eastern Pacific R. branickii, to which it bears many morphological similarities. It is distinguished from other Rhonciscus species by the number of scale rows between the dorsal fin and the lateral line (7), larger and thus fewer scales along the lateral line (48-50), large eyes (9.4-10.4 times in SL), longer caudal peduncle (15.2-20% of SL), larger sized penultimate (14.7-19.1% in SL) and last (7.4-9.5% in SL) dorsal fin spines which translates to a less deeply notched dorsal fin, and its opalescent silver with golden specks live coloration. This grunt, only now recognized by ichthyologists, but well known by local fishers that target snappers and groupers between 200 and 500 m in depth, occurs in far deeper waters than any western Atlantic grunt.

Transfer learning for efficient classification of grouper sound.

A transfer learning approach is proposed to classify grouper species by their courtship-associated sounds produced during spawning aggregations. Vessel sounds are also included in order to potentially identify human interaction with spawning fish. Grouper sounds recorded during spawning aggregations were first converted to time-frequency representations. Two types of time frequency representations were used in this study: spectrograms and scalograms. These were converted to images, and then fed to pretrained deep neural network models: VGG16, VGG19, Google Net, and MobileNet. The experimental results revealed that transfer learning significantly outperformed the manually identified features approach for grouper sound classification. In addition, both time-frequency representations produced almost identical results in terms of classification accuracy.

Classification of red hind grouper call types using random ensemble of stacked autoencoders.

In this paper, a method is introduced for the classification of call types of red hind grouper, an important fishery resource in the Caribbean that produces sounds associated with reproductive behaviors during yearly spawning aggregations. For the undertaken task, two distinct call types of red hind are analyzed. An ensemble of stacked autoencoders (SAEs) is then designed by randomly selecting the hyperparameters of SAEs in the network. These hyperparameters include a number of hidden layers in each SAE and a number of nodes in each hidden layer. Spectrograms of red hind calls are used to train this randomly generated ensemble of SAEs one at a time. Once all individual SAEs are trained, this ensemble is used as a whole to classify call types of red hind. More specifically, the outputs of individual SAEs are combined with a fusion mechanism to produce a final decision on the call type of the input red hind sound. Experimental results show that the innovative approach produces superior results in comparison with those obtained by non-ensemble methods. The algorithm reliably classified red hind call types with over 90% accuracy and successfully detected some calls missed by human observers.

Recovery when you are on your own: Slow population responses in an isolated marine reserve.

Geographic isolation is an important yet underappreciated factor affecting marine reserve performance. Isolation, in combination with other factors, may preclude recruit subsidies, thus slowing recovery when base populations are small and causing a mismatch between performance and stakeholder expectations. Mona Island is a small, oceanic island located within a partial biogeographic barrier-44 km from the Puerto Rico shelf. We investigated if Mona Island's no-take zone (MNTZ), the largest in the U.S. Caribbean, was successful in increasing mean size and density of a suite of snapper and grouper species 14 years after designation. The La Parguera Natural Reserve (LPNR) was chosen for evaluation of temporal trends at a fished location. Despite indications of fishing within the no-take area, a reserve effect at Mona Island was evidenced from increasing mean sizes and densities of some taxa and mean total density 36% greater relative to 2005. However, the largest predatory species remained rare at Mona, preventing meaningful analysis of population trends. In the LPNR, most commercial species (e.g., Lutjanus synagris, Lutjanus apodus, Lutjanus mahogoni) did not change significantly in biomass or abundance, but some (Ocyurus chrysurus, Lachnolaimus maximus), increased in abundance owing to strong recent recruitment. This study documents slow recovery in the MNTZ that is limited to smaller sized species, highlighting both the need for better compliance and the substantial recovery time required by commercially valuable, coral reef fishes in isolated marine reserves.

Automatic classification of grouper species by their sounds using deep neural networks.

In this paper, the effectiveness of deep learning for automatic classification of grouper species by their vocalizations has been investigated. In the proposed approach, wavelet denoising is used to reduce ambient ocean noise, and a deep neural network is then used to classify sounds generated by different species of groupers. Experimental results for four species of groupers show that the proposed approach achieves a classification accuracy of around 90% or above in all of the tested cases, a result that is significantly better than the one obtained by a previously reported method for automatic classification of grouper calls.

An approach for automatic classification of grouper vocalizations with passive acoustic monitoring.

Grouper, a family of marine fishes, produce distinct vocalizations associated with their reproductive behavior during spawning aggregation. These low frequencies sounds (50-350 Hz) consist of a series of pulses repeated at a variable rate. In this paper, an approach is presented for automatic classification of grouper vocalizations from ambient sounds recorded in situ with fixed hydrophones based on weighted features and sparse classifier. Group sounds were labeled initially by humans for training and testing various feature extraction and classification methods. In the feature extraction phase, four types of features were used to extract features of sounds produced by groupers. Once the sound features were extracted, three types of representative classifiers were applied to categorize the species that produced these sounds. Experimental results showed that the overall percentage of identification using the best combination of the selected feature extractor weighted mel frequency cepstral coefficients and sparse classifier achieved 82.7% accuracy. The proposed algorithm has been implemented in an autonomous platform (wave glider) for real-time detection and classification of group vocalizations.

Marine managed areas and associated fisheries in the US Caribbean.

The marine managed areas (MMAs) of the U.S. Caribbean are summarized and specific data-rich cases are examined to determine their impact upon fisheries management in the region. In this region, the productivity and connectivity of benthic habitats such as mangroves, seagrass and coral reefs is essential for many species targeted by fisheries. A minority of the 39 MMAs covering over 4000km(2) serve any detectable management or conservation function due to deficiencies in the design, objectives, compliance or enforcement. Fifty percent of the area within MMA boundaries had no-take regulations in the U.S. Virgin Islands, while Puerto Rico only had 3%. Six case studies are compared and contrasted to better understand the potential of these MMAs for fisheries management. Signs of success were associated with including sufficient areas of essential fish habitat (nursery, spawning and migration corridors), year-round no-take regulations, enforcement and isolation. These criteria have been identified as important in the conservation of marine resources, but little has been done to modify the way MMAs are designated and implemented in the region. Site-specific monitoring to measure the effects of these MMAs is needed to demonstrate the benefits to fisheries and gain local support for a greater use as a fisheries management tool.