SCIEPublish

Ecological Risk and Public Perception of Pinctada radiata Invasion in the Mediterranean: An Integrated SDM, SEM and Sentiment Analysis Framework

Article Open Access

Ecological Risk and Public Perception of Pinctada radiata Invasion in the Mediterranean: An Integrated SDM, SEM and Sentiment Analysis Framework

Author Information
Department of Ichthyology and Aquatic Environment, University of Thessaly, 38446 Volos, Greece
*
Authors to whom correspondence should be addressed.

Received: 20 June 2026 Revised: 09 July 2026 Accepted: 28 July 2026 Published: 10 August 2026

Creative Commons

© 2026 The authors. This is an open access article under the Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/).

Views:15
Downloads:9
Ecol. Divers. 2026, 3(3), 10011; DOI: 10.70322/ecoldivers.2026.10011
ABSTRACT: The rayed pearl oyster Pinctada radiata (Leach, 1814) was the first Indo-Pacific bivalve documented in the Mediterranean Sea and is now among the basin’s most widespread non-indigenous bivalves. Despite its long invasion history, basin-scale assessments integrating habitat suitability, anthropogenic exposure, and public perception remain limited. Here, we developed an integrated Mediterranean risk-screening framework that combines ensemble Species Distribution Modelling (SDM), Structural Equation Modelling (SEM), an Invasion Risk Index (IRI), and multilingual sentiment analysis for P. radiata. Occurrence data comprised 866 georeferenced records from five Mediterranean countries, of which 133 spatially thinned records were retained for SDM calibration. The ensemble SDM demonstrated high predictive performance (AUC = 0.934 ± 0.021, TSS = 0.812 ± 0.033, CBI = 0.883 ± 0.041) and identified mean annual sea-surface temperature, substrate type, and chlorophyll-a as the principal predictors of baseline habitat suitability. More than 81% of dated occurrence records originated from 2020–2026, and the spatially thinned dataset showed a significant westward redistribution through time (Spearman’s ρ = −0.241, p = 0.0059), although this pattern was interpreted cautiously because the dataset was dominated by opportunistic human observations and uneven reporting effort. Structural equation modelling supported consistent spatial associations among environmental suitability, anthropogenic exposure, and the observed distribution of P. radiata, while recognising that these relationships represent statistical associations rather than causal ecological mechanisms. The Invasion Risk Index identified High and Very High priority coastal sectors for surveillance, particularly where elevated environmental suitability coincided with intense maritime activity and relatively low current record density. Multilingual sentiment analysis of 1247 text units revealed predominantly neutral-to-positive public and stakeholder discourse, reflecting the species’ dual perception as both a non-indigenous organism and a commercially recognised resource. Collectively, these findings provide a spatially explicit framework for supporting surveillance, biosecurity planning, and risk communication for P. radiata and other marine non-indigenous species across the Mediterranean.
Keywords: Lessepsian migration; Non-indigenous species; Mediterranean Sea; Species distribution model; Structural equation modelling; Invasion risk; Sentiment analysis; Pinctada radiata
TOP