Background Identification of the factors affecting reactivation of varicella-zoster virus (VZV) largely remains an open question. Exposure to solar ultra violet (UV) radiation is speculated to facilitate reactivation. Should the role of UV in reactivation be significant, VZV reactivation patterns would generally be expected to be synchronous with seasonal UV profiles in temperate climates. Methods We analysed age and gender specific VZV notification time series data from Perth, Western Australia (WA). This city has more daily sunshine hours than any other major Australian city. Using the cosinor and generalized linear models, we tested these data for seasonality and correlation with UV and temperature. Results We established significant seasonality of varicella notifications and showed that while herpes-zoster (HZ) was not seasonal it had a more stable seasonal component in males over 60 than in any other subpopulation tested. We also detected significant association between HZ notifications and UV for the entire Perth population as well as for females and males separately. In most cases, temperature proved to be a significant factor as well. Conclusions Our findings suggest that UV radiation may be important for VZV reactivation, under the assumption that notification data represent an acceptably accurate qualitative measure of true VZV incidence.