Tholbecq et al. (2019:36–37) identified a destruction layer beneath the western staircase of the Temple’s peribole in Zone F of Qasr al-Bint, which they attributed to the southern 363 CE Cyril Quake. The dating is approximate (3rd–4th century CE) and is based on ceramic assemblages, including Schmidt Phase 4 material, North African North African Terra Sigillata, and oil lamps. The presence of colluvium above the destruction layer suggests partial abandonment of the area after the earthquake. Independent evidence for the same event was reported by Renel (2013:351–352), who documented a tumble of stone blocks in the southern sector of the apse monument within the southern temenos, near the entrance to the peribole. He attributed this tumble to collapse of the apse, plausibly caused by the same earthquake. At the northwest corner of Building B2 (Sondage P15), in the palatial sector adjacent to Qasr al-Bint, Tholbecq et al. (2023:38) and Tholbecq et al. (2024:37–38) documented a collapse layer with architectural debris, including Corinthian columns and facing blocks from a façade. This destruction was also attributed to the southern 363 CE Cyril Quake, with dating based on ceramic finds, notably amphorae and dishes in late North African sigillata.

Tholbecq et al. (2022:67) proposed that fallen blocks near the apse at Qasr al-Bint reflect damage from two distinct seismic events: the southern 363 CE Cyril Quake and a later quake such as the 419 CE Monaxius and Plinta Quake. In 2023, they revisited this interpretation and speculated that the second earthquake could instead correspond to one of the mid-8th century CE events ( Tholbecq et al., 2023:33). Tholbecq et al. (2022:67) reports that the stratigraphic evidence shows a collapse layer composed of intermixed blocks originating primarily from the apse monument's façade along with upper architectural elements of Qasr al-Bint, including the entablature and pediment or the cipher wall and marble veneers from the staircase. These materials are thought to have collapsed together and it was suggested that the fall of Qasr al-Bint’s upper portions during a fire caused the collapse of the facade of the apse monument and the adjacent stair complex. Tholbecq et al. (2022:67) opines that while the lowest collapse layer is attributed to the 363 CE earthquake, upper layers appear to correspond to a later event, such as the 419 CE earthquake after a period of abandonment. These two destruction phases are separated by a period of spoliation, which included the cutting up of a marble statue of Marcus Aurelius and recovery of architectural elements such as capitals. Tholbecq et al. (2022:67) characterizes absolute chronology as "more complicated than we thought: the northern part of the facade seems to have remained standing until the spoliations of the Byzantine era and the earthquake of 748 AD [JW: i.e. the 749 CE Sabbatical Year Quakes], while the southern part would be destroyed three centuries earlier" when the upper parts of Qasr al-Bint, weakened by fire, collapsed during the first earthquake.

Tholbecq et al. (2024:37–38) describe the final destruction phase in Building B2 (Sector P15) as marking the definitive abandonment of the structure. The collapse layer (US B2222) contains upper architectural elements from the ground floor, including the capital of an Ionic order cordiform pillar and fragments derived from the south façade of the north wing (wall B2201). Several fallen blocks had previously been reused during a reoccupation phase following an earlier episode of destruction. Tholbecq et al. (2024:37–38) further note that this reoccupation is indicated by installations such as low wall B2216 and tabun B2217, dated between the end of the 4th century and the first quarter of the 5th century CE. During this interval, numerous blocks from the initial collapse were reused. By contrast, elements of the ground-floor entablature are absent from the upper destruction debris, a pattern interpreted by the authors as evidence for prior looting.

The phase label (Phase 7b) for this earthquake is based on Augé et al. (2016) while the date range (3rd-4th century CE) is based on Tholbecq et al (2019).