Dermal denticle assemblages in coral reef sediments correlate with conventional shark surveys

Erin M. Dillon; Kevin D. Lafferty; Douglas J. McCauley; Darcy Bradley; Richard D. Norris; Jennifer E. Caselle; Graziella V. DiRenzo; Jonathan P.A. Gardner; Aaron O’Dea

doi:10.1111/2041-210x.13346

USGS · 70208124

Dermal denticle assemblages in coral reef sediments correlate with conventional shark surveys

Erin M. Dillon·Kevin D. Lafferty·Douglas J. McCauley·Darcy Bradley·Richard D. Norris·Jennifer E. Caselle·Graziella V. DiRenzo·Jonathan P.A. Gardner·Aaron O’Dea

View original source ↗Export metadata ↓

Abstract

1. It is challenging to assess long-term trends in mobile, long-lived, and relatively rare species such as sharks. Despite ongoing declines in many coastal shark populations, conventional surveys might be too fleeting and too recent to describe population trends over decades to millennia. Placing recent shark declines into historical context should improve management efforts as well as our understanding of past ecosystem dynamics. 2. A new paleoecological approach for surveying shark abundance on coral reefs is to quantify dermal denticle assemblages preserved in sediments. This approach assumes that denticle accumulation rates correlate with shark abundances. Here, we test this assumption by comparing the denticle record in surface sediments to three conventional shark survey methods at Palmyra Atoll, Line Islands, central Pacific Ocean, where shark density is high and spatially heterogeneous. 3. We generally found a significant positive correlation between denticle accumulation rates and shark abundances derived from underwater visual census, baited remote underwater video, and hook and line surveys. 4. Denticle accumulation rates reflected shark abundances, suggesting that denticle assemblages can preserve a signal of time-averaged shark abundance in low-energy coral reef environments. We offer suggestions for applying this tool to measure shark abundance over long timescales in other contexts.

Explore related subjects

Methods in Ecology and Evolution Article

Source-reported bounding extent: 5.840763926791161° to 5.919995673041826° latitude; -162.17605590820312° to -161.97898864746094° longitude. This indicates report coverage, not an exact sampling location. View area on OpenStreetMap.

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

Erin M. Dillon, Kevin D. Lafferty, Douglas J. McCauley, Darcy Bradley, Richard D. Norris, Jennifer E. Caselle, Graziella V. DiRenzo, Jonathan P.A. Gardner, Aaron O’Dea. 2020-01-28. Dermal denticle assemblages in coral reef sediments correlate with conventional shark surveys. https://doi.org/10.1111/2041-210x.13346

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related USGS reports

USGS2026

A systematic review of avian influenza virus transmission models: Gaps and opportunities

Highly pathogenic avian influenza virus (HPAIV) increasingly threatens wild birds, poultry, non-human mammals and humans, yet gaps and opportunities in avian influenza virus (AIV) transmission modelling have not been comprehensively synthesized. We reviewed 182 articles published through 2025, categorized dimensions, including modelling approaches, data, host types, biological levels, spatial scales, temporal scales and model objectives, examined associations among these dimensions, and synthesized author-stated limitations and recommendations. Statistical and mechanistic models dominated. Model data centred on population dynamics and demography, disease surveillance, host epidemiology and competence, meteorology and land use and land cover. Most studies focused on poultry or wild birds at regional spatial scales, seasonal/annual and multi-year temporal scales, population or multi-population levels, and were retrospective. Machine learning and decision- making models were less common. Immunity, behaviour and life history, community composition, morphology, viral genetics and viral persistence received less attention. Non-human mammals, individual and community biological levels, local and broader spatial scales, daily temporal scales and prospective modelling were also less frequently modelled. The authors frequently reported limitations related to data availability, surveillance sampling and reporting bias, and model simplification, while commonly recommending improved data quality and greater model complexity and realism.

Methods in Ecology and Evolution↗

USGS2026

Spatially realistic dynamic N-mixture models

Dispersal is a complex ecological process that can be driven by the abiotic environment, density-dependent processes and species mobility. Traditionally, dispersal has been studied with capture-based data that are often limited in space and time. More recently, spatial dynamic N-mixture models (i.e. N-mixture models accounting for dispersal among local populations and changes in abundance over time) allow the study of dispersal with count data. However, current spatial dynamic N-mixture models focus mainly on distance-based dispersal, limiting our ability to evaluate realistic patterns and drivers of dispersal at relevant spatial and temporal scales. We developed a spatially realistic dynamic N-mixture model that allows dispersal to be driven by environmental conditions and population density. Using simulations, we evaluated the inferential performance of this model when large proportions of sites are unsurveyed and the survey period is short, situations that are typical of real-world studies. We then used two case studies to illustrate the applications of this model, one representing a species with limited mobility in fragmented habitat (Chiricahua Leopard Frogs, Lithobates chiricahuensis , in isolated ponds) and the other representing a species with great mobility in continuous heterogeneous habitat (Common Nighthawks, Chordeiles minor , in grasslands). Our simulation study showed that the model provides valid inference even when the number of surveyed sites is small (𝑛 =10) or when the survey period is short for dynamic models (5 years), indicating that the model could have great practical value in real-world studies. The case study of Chiricahua Leopard Frogs showed that the model is capable of revealing pond-type-specific dispersal and predicting dispersal among ponds. The case study of Common Nighthawks showed the ability of the model to reveal the effect of primary productivity on dispersal and to map dispersal across the study area. Our effort demonstrates the potential of using count data to gain insights about complex yet realistic dispersal processes. The model that we introduce is flexible and adaptable for a broad range of situations in population and conservation ecology

Arizona, Kansas↗

USGS2025

Simulated soundscapes and transfer learning boost the performance of acoustic classifiers under data scarcity

1. The biodiversity crisis necessitates spatially extensive methods to monitor multiple taxonomic groups for evidence of change in response to evolving environmental conditions. Programs that combine passive acoustic monitoring and machine learning are increasingly used to meet this need. These methods require large, annotated datasets, which are time-consuming and expensive to produce, creating potential barriers to adoption in data- and funding-poor regions. Recently released pre-trained avian acoustic classification models provide opportunities to reduce the need for manual labelling and accelerate the development of new acoustic classification algorithms through transfer learning. Transfer learning is a strategy for developing algorithms under data scarcity that uses pre-trained models from related tasks to adapt to new tasks. 2. Our primary objective was to develop a transfer learning strategy using the feature embeddings of a pre-trained avian classification model to train custom acoustic classification models in data-scarce contexts. We used three annotated avian acoustic datasets to test whether transfer learning and soundscape simulation-based data augmentation could substantially reduce the annotated training data necessary to develop performant custom acoustic classifiers. We also conducted a sensitivity analysis for hyperparameter choice and model architecture. We then assessed the generalizability of our strategy to increasingly novel non-avian classification tasks. 3. With as few as two training examples per class, our soundscape simulation data augmentation approach consistently yielded new classifiers with improved performance relative to the pre-trained classification model and transfer learning classifiers trained with other augmentation approaches. Performance increases were evident for three avian test datasets, including single-class and multi-label contexts. We observed that the relative performance among our data augmentation approaches varied for the avian datasets and nearly converged for one dataset when we included more training examples. 4. We demonstrate an efficient approach to developing new acoustic classifiers leveraging open-source sound repositories and pre-trained networks to reduce manual labelling. With very few examples, our soundscape simulation approach to data augmentation yielded classifiers with performance equivalent to those trained with many more examples, showing it is possible to reduce manual label-ling while still achieving high-performance classifiers and, in turn, expanding the potential for passive acoustic monitoring to address rising biodiversity monitoring needs.

Methods in Ecology and Evolution↗