Objective This study aimed to compare the enamel surface roughness created by four polishing methods after debonding, by using scanning electron microscopy (SEM) and atomic force microscopy (AFM). Materials and Methods Four experimental polishing groups (Sof-Lex disc, SD; sandblaster, SB; tungsten carbide bur, TB; and white stone bur, WB) and one control group were selected from 100 premolars (n = 20/group). The experimental teeth were bonded with a bracket, thermocycled, and debonded. Residual adhesive was removed by either of the respective methods. Pre and postdebonding root mean square (Rq) values were obtained from AFM evaluations. All specimens were examined and evaluated with SEM using a modified enamel surface index (modified ESI). Statistical Analysis Differences among the polishing methods were compared with analysis of variance and Fisher’s least significant difference test at p < 0.05. Results Both microscopic evaluations indicated that the surface with the greatest roughness herein belonged to the SD group, followed by that for SB, TB, and WB groups. AFM measurements indicated a maximum postdebonding Rq herein for the WB group and a significantly greater surface roughness for the TB and WB groups than for the SD and SB groups. Among the experimental groups, SEM followed by modified ESI evaluations revealed similar data to those obtained with AFM. Significant differences were seen among all paired groups, except for that between the SB and TB groups. Conclusion Within the limitations of this study, all four polishing methods were concluded to be clinically acceptable for removing residual orthodontic adhesives.