Environmental Drivers of Spatial Ecology in Juvenile Scalloped Hammerhead Sharks (Sphyrna lewini) in an Open-Coast Nursery Area in Jalisco, Mexico

Hammerhead sharks (Sphyrnidae) of southeast Queensland: habitat and movements

The aim of this study was to determine whether juvenile scalloped hammerhead sharks (Sphyrna lewini) use the south-eastern Gulf of California as a nursery and feeding area. This information could help lay the groundwork required for the conservation of this endangered species. To address this, we carried out stable isotope (δ15N and δ13C) and stomach content analyses of sharks caught between 2000 and 2004 in Mazatlan, Mexico. Stomach contents and δ13C values indicated that S. lewini is a predator that feeds on benthic prey near the coast. Differences in δ15N average values between sizes classes ( 100 cm) suggest that there was an ontogenetic change in this shark’s feeding habits and also in their living environment (from benthic areas to pelagic areas). The trophic position indicated that S. lewini is a tertiary consumer, but with a high degree of trophic plasticity, and thus, different trophic roles, highlighting the importance of this predator as a regulator of prey populations. Finally, the linear isotopic relationship between S. lewini and its prey indicates a long residency within the Mazatlan area. Our results demonstrate that the south-eastern Gulf of California is a nursery area that offers abundant food for juvenile scalloped hammerhead sharks.

Identification of nursery habitats and knowledge of the trophic ecology and habitat use of juvenile fishes within these habitats are fundamental in developing sound management and con- servation strategies. The brown stingray Dasyatis lata is a large benthic predator that inhabits the coastal waters of Hawai'i. Although abundant in these ecosystems, little is known about its basic eco - logy. Stomach content, bulk and amino acid stable isotope analyses were used to assess diet and habitat use of juvenile brown stingrays and to examine the possibility of competitive interactions with juvenile scalloped hammerhead sharks Sphyrna lewini that are sympatric with brown stingrays in K¯ ane'ohe Bay, Oahu. Based on stomach contents, brown stingrays fed almost exclusively on crus- taceans. An ontogenetic shift in stingray diet and an increase in relative trophic position (TP) were apparent from stomach content and stable isotope analysis. Stingray bulk δ 13 C and δ 15 N values indi- cated long-term foraging fidelity to subregions of the bay. Use of K¯ ane'ohe Bay as a nursery habitat was supported by nitrogen isotopic analysis of individual amino acids from stingray muscle samples. Our results clearly demonstrated that stingrays foraged within the bay for the majority of their juve- nile lives then shifted to offshore habitats with the onset of sexual maturity. Trophic enrichment fac- tors used to estimate TPs from amino acid analysis in previous studies may underestimate TPs in elas- mobranchs owing to urea retention for osmoregulation. Potential prey resources were partitioned between stingrays and juvenile scalloped hammerhead sharks, and TP estimates from each analyti- cal method indicated that juvenile scalloped hammerhead sharks forage on higher TP prey than do juvenile brown stingrays. These results show that the study of foraging ecology and habitat use of marine animals can greatly benefit from integrating traditional stomach content and bulk stable isotopic analyses with nitrogen isotopic analyses of individual amino acids.

There is often a trade-off between accuracy and interpretability in Machine Learning (ML) models. As the model becomes more complex, generally the accuracy increases and the interpretability decreases. Interpretable Machine Learning (IML) methods have emerged to provide the interpretability of complex ML models while maintaining accuracy. Thus, accuracy remains constant while determining feature importance. In this study, we aim to compare agnostic IML methods including SHAP and ELI5 with the intrinsic IML methods and Feature Selection (FS) methods in terms of the similarity of attribute selection. Also, we compare agnostic IML models (SHAP, LIME, and ELI5) among each other in terms of similarity of local attribute selection. Experimental studies have been conducted on both general and private datasets to predict company default. According to the obtained results, this study confirms the reliability of agnostic IML methods by demonstrating similarities of up to 86% in the selection of attributes compared to intrinsic IML methods and FS methods. Additionally, certain agnostic IML methods can interpret models for local instances. The findings indicate that agnostic IML models can be applied in complex ML models to offer both global and local interpretability while maintaining high accuracy.

This work aimed to characterize the nursery habitat use patterns of the scalloped hammerhead shark, Sphyrna lewini (SPL), in coastal areas of Jalisco and Colima, through the birth pattern, space-time distribution and relationship with environmental conditions. Information was combined from three sources: monitoring bycatch from the artisanal fishery, fishery-independent samplings, and acoustic tracking and monitoring. From September 2013 to May 2017, 408 juvenile SPL (41.6-100.1cm total length) were recorded. Births occurred between May and December (rainy-warmer season), within a radius of 2 km from river mouths in Marabasco, Navidad Bay, Rebalsito-Tecuan and Cuitzmala mainly in shallow (<20 m), turbid and soft-bottom areas. Some tagged SPL moved from Marabasco and Rebalsito to Navidad Bay. The peak of catch and births occurred in June-August. Tagged SPL remained near the river mouth in Rebalsito for up to 27 days, showing a mean residency index of 0.29, a home range of 5.55 km2 with a core area of 1.23 km2 located within a 1.5km radius from the river mouth. In December-January SPL left the river mouth areas and the catch was scarce until the following May-June, except in January 2016, when the catch was high due to El Niño 2015. SPL bycatch was significantly associated with temperature, precipitation and the Oceanic Niño Index.

Characteristics and species composition of a small-scale shark fishery in Puerto Rico: Jurisdictional issues enable legal landings of prohibited and endangered species

This study incorporates a random forest (RF) approach to probe complex interactions and nonlinearity among predictors into an item response model with the goal of using a hybrid approach to outperform either an RF or explanatory item response model (EIRM) only in explaining item responses. In the specified model, called EIRM-RF, predicted values using RF are added as a predictor in EIRM to model the nonlinear and interaction effects of person- and item-level predictors in person-by-item response data, while accounting for random effects over persons and items. The results of the EIRM-RF are probed with interpretable machine learning (ML) methods, including feature importance measures, partial dependence plots, accumulated local effect plots, and the H-statistic. The EIRM-RF and the interpretable methods are illustrated using an empirical data set to explain differences in reading comprehension in digital versus paper mediums, and the results of EIRM-RF are compared with those of EIRM and RF to show empirical differences in modeling the effects of predictors and random effects among EIRM, RF, and EIRM-RF. In addition, simulation studies are conducted to compare model accuracy among the three models and to evaluate the performance of interpretable ML methods.

Food limitation in a nursery area: estimates of daily ration in juvenile scalloped hammerheads, Sphyrna lewini (Griffith and Smith, 1834) in Kāne'ohe Bay, Ō'ahu, Hawai'i

ABSTRACTWith the spread and rapid advancement of black box machine learning (ML) models, the field of interpretable machine learning (IML) or explainable artificial intelligence (XAI) has become increasingly important over the last decade. This is particularly relevant for survival analysis, where the adoption of IML techniques promotes transparency, accountability, and fairness in sensitive areas, such as clinical decision‐making processes, the development of targeted therapies, interventions, or in other medical or healthcare‐related contexts. More specifically, explainability can uncover a survival model's potential biases and limitations and provide more mathematically sound ways to understand how and which features are influential for prediction or constitute risk factors. However, the lack of readily available IML methods may have deterred practitioners from leveraging the full potential of ML for predicting time‐to‐event data. We present a comprehensive review of the existing work on IML methods for survival analysis within the context of the general IML taxonomy. In addition, we formally detail how commonly used IML methods, such as individual conditional expectation (ICE), partial dependence plots (PDP), accumulated local effects (ALE), different feature importance measures, or Friedman's H‐interaction statistics can be adapted to survival outcomes. An application of several IML methods to data on breast cancer recurrence in the German Breast Cancer Study Group (GBSG2) serves as a tutorial or guide for researchers, on how to utilize the techniques in practice to facilitate understanding of model decisions or predictions.

Traditional vital signs are an essential part of triage assessment in emergency departments (ED), and have been widely used in trauma prediction models. Previous researchers have studied the effect of vital signs scores on predicting traumatic injury outcomes and have found it to be significant. Based on the vital signs’ scores, an Interpretable Machine Learning (IML) method is proposed to predict patient outcomes and is compared with various ML algorithms. Results indicate that the IML method has a comparable performance with a mean AUC of 0.683, and its interpretability would help in the early identification of trauma patients at risk of mortality.

Light energy is necessary for vision, but ocular tissues are subject to photodamage, and many vertebrates sequester UV-absorbant pigments in their pre-retinal ocular tissues, in part to minimize such damage. In this study (21 May–1 July 2001), juvenile scalloped hammerhead sharks, Sphyrnalewini (Griffith and Smith, 1834), were exposed to higher levels of solar radiation than they had previously experienced in the source habitat in the turbid waters of Kane'ohe Bay, Hawai'i, USA. Light transmission through the ocular media was measured in two individuals shortly after capture and in other individuals after 7, 14, 20, 27, and 41 days exposure to high light levels in a shallow, outdoor pen. Sharks from their usual habitat filtered a small proportion of the UV spectrum, but sharks exposed to greater solar radiation showed increased UV blocking in their corneal tissues, particularly at wavelengths below 310 nm. The proportion of UV blocked was relative to the duration of exposure. There were no changes attributable to exposure duration in transmission through the whole eye or lens, nor was there any clear pattern to variation in transmission through dorsal, ventral, anterior, and posterior quadrants of the cornea. Further experiments will be needed to confirm that this apparently rapid corneal adaptation to high light was due to the increased UV exposure.

Juvenile scalloped hammerhead sharks, Sphyrna lewini, are apex predators within their nursery ground in Kāne‘ohe Bay, Ō‘ahu, Hawai‘i. Understanding daily maintenance requirements of a top-level predator is an important step toward understanding its ecological impact within a nursery ecosystem. Juvenile S. lewini were fed a range of daily ration levels to examine the effect of feeding rate on growth and gross conversion efficiency. The von Bertalanffy growth model yielded the best fit to the data, predicting a maintenance ration of 115 kJ kg−1 day−1 (3.4% body weight (BW) day−1) and a maximum growth rate of 38 kJ kg−1 day−1. This finding is in agreement with the previous prediction of high energetic requirements for S. lewini. In combination with the hypothesized food limitation within Kāne‘ohe Bay, this result may explain the observed high mortality rates of S. lewini. Gross conversion efficiency, K 1, ranged from −36% to 34%, with maximum efficiency at feeding levels of 5.1% BW day−1. The growth conversion efficiency of S.␣lewini is similar to that of lemon sharks and teleost fishes. Growth rates of juvenile S. lewini are possibly restricted by their high metabolic rate, limited food availability and foraging inexperience. By directly examining the effect of ration size on growth and food conversion, it was possible to resolve discrepancies between earlier studies, which used respiratory metabolism and gut content analyses.

We present a brief history of the field of interpretable machine learning (IML), give an overview of state-of-the-art interpretation methods, and discuss challenges. Research in IML has boomed in recent years. As young as the field is, it has over 200 years old roots in regression modeling and rule-based machine learning, starting in the 1960s. Recently, many new IML methods have been proposed, many of them model-agnostic, but also interpretation techniques specific to deep learning and tree-based ensembles. IML methods either directly analyze model components, study sensitivity to input perturbations, or analyze local or global surrogate approximations of the ML model. The field approaches a state of readiness and stability, with many methods not only proposed in research, but also implemented in open-source software. But many important challenges remain for IML, such as dealing with dependent features, causal interpretation, and uncertainty estimation, which need to be resolved for its successful application to scientific problems. A further challenge is a missing rigorous definition of interpretability, which is accepted by the community. To address the challenges and advance the field, we urge to recall our roots of interpretable, data-driven modeling in statistics and (rule-based) ML, but also to consider other areas such as sensitivity analysis, causal inference, and the social sciences.

The scalloped hammerhead shark is an endangered species that is listed in CITES Appendix II. Information on the biological aspects, exploitation rate, and spawning potential ratio of scalloped hammerhead sharks are very limited, especially in Lampung Bay waters. These data were important to find solutions and the best management to sustain the scalloped hammerhead sharks in Indonesia, especially in Lampung Bay waters. The aim of this study is to investigate the biological aspects, exploitation rates, and spawning potential ratios of scalloped hammerhead sharks in Lampung water. Scalloped hammerhead shark samples were collected from the catch of bottom gillnet in Kalianda Fish Landing Place, Lampung Province. Enumerators collected biometric data (length and weight) of 332 scalloped hammerhead sharks from July to November 2020. Biological aspects include distribution of length, length-weight (L-W) relationships, length at first capture, and length at first maturity. The utilization rate was calculated using the exploitation rate (E) and the estimated SPR based on length data. The results show that the length of scalloped hammerhead sharks ranged from 37.5 to 173.0 cm FL. Most of the samples fish had not spawned yet or were immature (Lc =47.1 cm FL; Lm =89.4 cm FL). The growth pattern of scalloped hammerhead sharks was allometrically negative. Scalloped hammerhead sharks have natural mortality of 0.18/year, while fishing mortality was 1.08/year. The fishing status of scalloped hammerhead sharks in Lampung Bay was fully exploited (E=0.85), and the stocks were in a recruitment overfishing condition. These findings indicate that effective management is required to ensure the sustainability of scalloped hammerhead sharks in Lampung Bay waters, such as adjusting mesh size, fishing season, and avoiding capture in the nursery area.

The age of 296 juvenile scalloped hammerhead sharks Sphyrna lewini caught by several fisheries in the Mexican Pacific Ocean from March 2007 to September 2017 were estimated from growth band counts in thin-sectioned vertebrae. Marginal-increment analysis (MIA) and centrum-edge analysis (CEA) were used to verify the periodicity of formation of the growth bands, whereas elemental profiles obtained from LA-ICP-MS transect scans in vertebrae of 15 juveniles were used as an alternative approach to verify the age of the species for the first time. Age estimates ranged from 0 to 10+ years (42-158.7 cm total length; LT ). The index of average percentage error (IAPE 3.6%), CV (5.2%), bias plots and Bowker's tests of symmetry showed precise and low-biased age estimation. Both MIA and CEA indicated that in the vertebrae of juveniles of S. lewini a single translucent growth band was formed during winter (November-March) and an opaque band during summer (July-September), a period of faster growth, apparently correlated with a higher sea surface temperature. Peaks in vertebral P and Mn content spatially corresponded with the annual banding pattern in most of the samples, displaying 1.19 and 0.88 peaks per opaque band, respectively, which closely matched the annual deposition rate observed in this study. Although the periodicity of growth band formation needs to be verified for all sizes and ages representing the population of the species in the region, this demonstration of the annual formation of the growth bands in the vertebrae of juveniles should lead to a re-estimation of the growth parameters and productivity of the population to ensure that it is harvested at sustainable levels.