Flores, Oswaldo ORCID: https://orcid.org/0009-0006-8620-7847, Rigby, Robert ORCID: https://orcid.org/0000-0003-4787-623X, Segura, Marceliano, Ledesma, Jesús, Carlos, Paulino, Chacón, Germán and Valles, Juan (2026) Effect of oceanographic variables and space-time factors on the density of Peruvian anchovy (Engraulis ringens): an approach with GAMLSS. Progress in Oceanography, 243:103688. ISSN 0079-6611 (Print), 1873-4472 (Online) (doi:10.1016/j.pocean.2026.103688)

PDF (Author's Accepted Manuscript) 52608 RIGBY_Effect_Of_Oceanographic_Variables_And_Space-Time_Factors_On_The_Density_(AAM)_2026.pdf - Accepted Version Restricted to Repository staff only until 6 February 2027. Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (469kB) | Request a copy

Abstract

This study examines how environmental and spatiotemporal factors modulate the acoustic density (area backscattering coefficient, sA) of the Peruvian anchoveta (Engraulis ringens), using data from 25 IMARPE surveys conducted between 2007–2017. Given the semicontinuous nature of sA (characterized by an excess of zeros and strong positive skewness) GAMLSS models with zero-adjusted Box–Cox t and Generalized Inverse Gaussian distributions (ZABCTo and ZAGIG) were fitted. These models allow sA to be analyzed in its original scale while jointly estimating the location (μ), scale (σ), skewness (ν), kurtosis (τ), and probability of absence (π0) parameters, thus providing a complete distributional characterization of aggregation processes. The results revealed a consistent spatial structure, with the highest densities occurring within 40 nm of the coast and between 6°–15°S, under salinities of 33.75–35.2 PSU, sea surface temperatures of 14–25 °C, and dissolved oxygen > 5 mL L⁻1. Modeling sA on its original scale uncovered nonlinear environmental thresholds, diel variability, and changes in spatial heterogeneity that are not detected when using Gaussian and Tweedie GAMs or GAMLSS models based on logarithmic transformations. Seasonal models further showed that upwelling intensity, ENSO-related anomalies, and short-term fluctuations influence not only the magnitude and dispersion of sA but also reflect the extreme densities that characterize anchoveta aggregation dynamics. By demonstrating that distributional modeling of sA without transformations preserves ecological information that is lost under traditional approaches, this study provides a robust statistical basis for interpreting aggregation responses in the highly dynamic upwelling system of the Humboldt Current.

Item Type:	Article
Uncontrolled Keywords:	Peruvian anchovy, oceanographic variables, GAMLSS, zero-adjusted distributions, fishery assessment
Subjects:	G Geography. Anthropology. Recreation > GE Environmental Sciences Q Science > Q Science (General) Q Science > QA Mathematics > QA75 Electronic computers. Computer science S Agriculture > SH Aquaculture. Fisheries. Angling
Faculty / School / Research Centre / Research Group:	Faculty of Engineering & Science Faculty of Engineering & Science > School of Computing & Mathematical Sciences (CMS)
Last Modified:	06 Mar 2026 16:58
URI:	https://gala.gre.ac.uk/id/eprint/52608

Actions (login required)

View Item

Downloads