Facial morphing attacks have become a serious threat to identity verification systems, especially in security-critical environments such as Automated Border Control gates. These attacks exploit the vulnerabilities of face recognition systems by merging two distinct identities into a single image, allowing both individuals to bypass verification processes. Current Morphing Attack Detection (MAD) systems face significant challenges in accurately detecting such manipulations when test images present peculiar characteristics, not well represented in the training data. Even though the impact of these factors has been addressed to some extent in the literature, the effects of natural biometric variations like aging have not yet been widely investigated. This paper proposes a novel dataset, AMONOT, consisting of synthetic aged images to assess the robustness of MAD systems to aging. By generating realistic synthetic faces with a granular age progression, we enable a more comprehensive evaluation of MAD systems under conditions that mimic real-world scenarios. The proposed dataset and experimental protocol aim to enhance the detection accuracy of morphing attacks involving aged faces, offering a more robust benchmark for improving security in identity verification applications.