Asto C. auteur aut IRD Bosse A. auteur aut IRD Pietri Alice auteur aut IRD Sauzède R. auteur aut IRD Graco M. auteur aut IRD Gutiérrez D. auteur aut IRD Colas François auteur aut IRD text journalArticle eng text/pdf reformatted digital access This study presents a regionally trained version of the "CArbonate system and Nutrients concentration from hYdrological properties and Oxygen using a Neural network" (CANYON) method, named CANYON-PU, for estimating primary macronutrients (phosphates, silicates, and nitrates) in the Peruvian Upwelling System (PUS). Using a neural network approach, the model was trained using extensive biogeochemical data spanning between 2003 and 2021, collected by the Peruvian Institute of Marine Research (IMARPE). Variables representing the low-frequency variability related to ENSO were introduced in the training and significantly improved the performance of the algorithm. The performance of CANYON-PU was validated against independent datasets and demonstrated an improvement in accuracy over the global CANYON model that struggled to represent the nutrient distribution in the PUS mainly due to the lack of samples in its training. Therefore, CANYON-PU successfully captured nutrient variability across different spatial and temporal scales, showcasing its applicability to diverse datasets, including high-frequency data such as profiling floats or gliders. This work highlights the effectiveness of neural networks for representing the nutrient distribution within highly variable ecosystems like the PUS. specialized Peruvian upwelling system nutrients neural network El Niño gliders profiling float PEROU PACIFIQUE 032 12 1558747 [18 ] 2025 https://www.documentation.ird.fr/hor/fdi:010094272 10.3389/fmars.2025.1558747 [F B010094272] https://www.documentation.ird.fr/hor/fdi:010094272 https://horizon.documentation.ird.fr/exl-doc/pleins_textes/2025-08/010094272.pdf IRD - Base Horizon / Pleins textes 2025-08-27 2025-11-04 fdi:010094272 fre