Natural and urban forests worldwide are increasingly threatened by global change resulting from human-mediated factors, including invasions by lethal exotic pathogens. Ash dieback (ADB), incited by the alien invasive fungus Hymenoscyphus fraxineus, has caused large-scale population decline of European ash (Fraxinus excelsior) across Europe, and is threatening to functionally extirpate this tree species. Genetically controlled host resistance is a key element to ensure European ash survival and to restore this keystone species where it has been decimated. We know that a low proportion of the natural population of European ash expresses heritable, quantitative resistance that is stable across environments. To exploit this resource for breeding and restoration efforts, tools that allow for effective and efficient, rapid identification and deployment of superior genotypes are now sorely needed. Here we show that Fourier-transform infrared (FT-IR) spectroscopy of phenolic extracts from uninfected bark tissue, coupled with a model based on soft independent modelling of class analogy (SIMCA), can robustly discriminate between ADB-resistant and susceptible European ash. The model was validated with populations of European ash grown across six European countries. Our work demonstrates that this approach can efficiently advance the effort to save such fundamental forest resource in Europe and elsewhere.