To test whether the Cu, Ag and Au (and other metal) budget in arc lavas is controlled by the composition and thickness of the subducting slab we selected 65 glass samples from 17 Kermadec arc front volcanoes and several sites from the Havre Trough back-arc immediately west of the Kermadec arc front (e.g. Caratori Tontini et al., 2019). The samples that span ~1200 arc km and from arc front volcanoes, located above and to the north from the subducting Hikurangi Plateau (Fig. 1), were collected over the course of 20 years using the RVs Tangaroa and Sonne and the manned submersible Pisces V. Based on the new data produced during the GoldTrace project we developed a new model to explain what processes control Cu, Ag and Au beneath arc lavas (see details in WP 3). In this contributing we were able to link Cu, Ag and Au contents to temperature estimates and mantle depletion, requiring high degrees of partial melting of the Kermadec sub-arc mantle. Although the contribution This work is currently in review at Nature Geoscience. A second manuscript focussing on the behaviour of other mildly to highly chalcophile to metals (As, Mo, W, Tl, Re, Pt) is in preparation. Through this work it becomes more apparent how different individual metals behave during sub-arc magmatic processes and what the source of them are. A third publication about metal content and their transport to the seabed beneath Brothers volcano, Kermadec arc, is planned but has not commenced yet. These contributions will help to define regions on the seabed that preferably host Cu, Ag and Au-rich lavas on the seabed. Because it has been shown that a link exists between Cu, Ag and Au contents in arc lavas and the Cu, Ag and Au-grade of related mineralisation (Park et al., 2019). The results from our study will therefore help to develop vectors to guide explorers in their quest of discovering high grade ore deposits.