Environmental drivers of marine temperatures extremes: a function on function quantile regression approach.

A. Sarra, T. Di Battista, A. Evangelista, E. Nissi, N. Di Deo

Understanding how environmental drivers shape the full distribution of marine temperature across depth is crucial for characterising thermal extremes in coastal ecosystems. Standard regression approaches either overlook the functional nature of vertical profiles or focus solely on mean behaviour, leaving the distributional tails poorly described.
We introduce a penalised Function-on-Function Quantile Regression (FFQR) framework that models temperature and environmental predictors as depth-dependent functions, estimating a bivariate coefficient surface capturing how predictors at one depth influence temperature quantiles at another. An anisotropic roughness penalty separately controls smoothness along predictor and response depths, reflecting distinct physical coupling mechanisms.
The framework is applied to a multi-year dataset from stations along the Abruzzo coastline (Central Adriatic Sea), comprising monthly vertical profiles of temperature and environmental covariates.

Palabras clave: functional data analysis, quantile regression, function-on-function regression, roughness penalty, marine temperature extremes, Central Adriatic Sea

SI Sesión Hispano-Italiana
3 de septiembre de 2026  11:10
Aula 20

G. Loffredo, E. Romano, A. M. Aguilera del Pino, F. Maturo, M. Vidal

F. Porro, M. Restaino, J. E. Ruiz Castro, M. Zenga