Chronology, duration, and periodicity of linear enamel hypoplasia at the late Iron Age site Non Ban Jak, Thailand: A quantitative microscopic analysis

We provide the first application of a quantitative microscopic approach that does not rely on the presence of perikymata for the identification and comprehensive analysis of linear enamel hypoplasia (LEH) to a large archaeological sample from Southeast Asia. Additionally, we introduce MicroPolySharp, a new computer program that automates the assessment of these stress episodes, and we present new dental crown width/height ratios for the region.Enamel surface depth profiles were measured from epoxy dental replicas generated from 48 individuals using a confocal microscope. The identification and analysis of LEH episodes was undertaken using MicroPolySharp, which combines three recently published methods specifically suited to worn archaeological samples. Ten pa-rameters were examined: frequency, prevalence, episode duration, stress duration, recovery duration, age at first onset, age at last onset, total growth disruption, proportion of available enamel affected, and periodicity.Results revealed a high prevalence of LEH: 97.92 % (47/48) individuals affected. LEH frequency averaged 2.5 episodes per individual (range = 1 to 4 episodes). The chronology of episodes averaged 3.66 years (range = 1.4 to 5.5 years). The age of earliest episode occurrence averaged 3.1 years (range = 1.4 to 4.2 years). While the age of last episode averaged 4.3 years (range = 2.9 to 5.5 years). Duration of growth disruptions (stress + recovery) averaged 103 days (range = 27 to 269 days). Consistent with other studies, the average duration of the stress portion of episodes was 56 days (range = 9 days to 6.4 months) while the duration of the recovery portion averaged 47 days (range = 8 days to 5 months). The total amount of growth disruption for individuals averaged 263 days (range = 87 to 543 days). The proportion of available crown height affected by growth disruption averaged 21 % (range = 6 % to 42 %), while LEH periodicity for the sample averaged 8.6 months (range = 2.4-23 months). Stress and recovery durations, along with age at first episode, were the most sensitive and useful parameters for discerning differences between the subgroups of interest. Finally, the effects of any amount of variable crown height wear within the study group was found to have a significant confounding effect on all aspects of LEH results and interpretation if not properly controlled. Given the high prevalence, all individuals were, for the most part, equally affected. Future analysis of samples from the periods prior to and leading up to the Late Iron Age from nearby sites will be required to provide much needed context to the high levels of systemic stress observed at Non Ban Jak.