Suicide is the leading global cause of death, particularly challenging in adolescent health. The current findings on brain regions associated with suicide are often confounded by environmental factors. Moreover, the emergence and persistence of suicidality remain largely unknown. Using the Adolescent Brain Cognitive Development study, we analyzed neuroimaging, suicidality, and environmental measures from 11,220 participants at baseline and 1-year-follow-up. Participants with low family-neighborhood-school environmental risk (risk score below mean) were grouped by suicidality changes across two timepoints: risk to risk (R-R), risk to no risk (R-NR), no risk to risk (NR-R), and no risk to no risk (NR-NR). Propensity score matching was performed on demographic variables, comparisons of brain volumes, cortical thickness, and surface area were conducted between R-R and R-NR groups, as well as NR-NR and NR-R groups, with matched sample size. Our results showed reduced gray matter volume in temporal cortex, parahippocampal, pallidum and hippocampus in the comparisons between NR-R and NR-NR (emergence of suicidality). Conversely, comparisons between R-R and R-NR (persistent suicidality) showed reduced gray matter volume in superior temporal, visual cortex and default mode network. These findings suggest that baseline differences in brain regions are distinctly associated with the emergence and persistence of suicidality among adolescents.