Frozen berries have been repeatedly linked to acute gastroenteritis caused by norovirus, the most common cause of foodborne illness in the United States. Many guidelines recommend that frozen berries be microwaved for at least 2 min, but it is unclear if this thermal treatment is effective at inactivating norovirus. The objective of this study was to model the effect of microwave heating at varying power levels on the survival of bacteriophage MS2, a norovirus surrogate, when inoculated onto frozen strawberries. Bacteriophage MS2 was inoculated onto the surface of frozen strawberries with a starting concentration of approximately 10 log PFU/g. Samples (either 3 or 5 whole strawberries) were heated in a 1300-Watt domestic research microwave oven (frequency of 2450 MHz) at power levels of 30, 50, 70, and 100% (full power), for times ranging from 15 to 300 s to determine inactivation. Temperatures at berry surfaces were monitored during heating using fiberoptic thermometry. All experiments were conducted in triplicate. The primary model for thermal inactivation was a log-linear model of logN vs. time. The secondary model was for a D-value decreasing linearly with temperature and an added term that was path-dependent on the thermal history. Parameters in the model were estimated using dynamic temperature history at the surface of the berry, via nonlinear regression using all data simultaneously. The root mean square error was ∼0.5 PFU/g out of a total 6-log reduction. Log reductions of 1.1 ± 0.4, 1.5 ± 0.5, 3.1 ± 0.1, and 3.8 ± 0.2 log PFU/g were observed for 30, 50, 70, and 100% microwave power levels when three berries were heated for 60 s. D-values were 21.4 ± 1.95 s and 10.6 ± 1.1 s at 10 and 60°C, respectively. This work demonstrates an approach to estimate inactivation parameters for viruses from dynamic temperature data during microwave heating. These findings will be useful in predicting the safety effect of microwave heating of berries in the home or food service.