Although tree growth is mainly dependent on natural climate conditions, factors such as microclimate, growth of the previous year, tree age, and others may also affect tree growth. In dendroclimatological studies, it is generally assumed that the relationships between tree growth and climate factors are age-independent after removing the biological growth trends related to tree age. However, in some multi-aged forest stands, climate-growth relationships may be biased, because the trees' chronologies respond differently to climate factors depending on their age or size. To assess this age-dependent effect, in this study, we tested the consistency of climate-growth responses in tree-ring series from Picea purpurea trees of different age classes in the Wanglang Nature Reserve, Sichuan province. In the concentrated distribution area of the upper limit, spruces were grouped into three age classes: trees younger than 100 years (young age class, YAC), trees 100-200 years of age (middle age class, MAC), and trees older than 200 years (old age class, OAC). Residual chronologies of the three age classes were built to analyze the climate-growth relationships using correlation and response functions. Correlation analysis indicated that the radial growth of the YAC was significantly and positively correlated with mean monthly temperature of the previous November and current February, June, and September; mean monthly maximum temperature of the previous November and current February and September; and mean monthly minimum temperature of the current February, March, April, June, July, August, and September, and it was significantly and negatively affected by monthly precipitation of the previous December and current June. The ring width of the MAC was positively correlated with mean monthly minimum temperature of the current April and July, and was negatively correlated with monthly precipitation of the current June. The residual chronology of the OAC was significantly and negatively affected by mean monthly temperature of the previous August, mean monthly minimum temperature of the previous August and current September, and monthly precipitation of the previous December. Response analysis showed that radial growth of the YAC was significantly and positively correlated with mean monthly maximum temperature of the current February, and mean monthly minimum temperature of the current February and April. There was no significant correlation between radial growth of the MAC and climate factors. The ring width of the OAC was significantly and negatively affected by monthly precipitation of the previous December.Overall, our results revealed that with an increase in tree age, the sensitivity of spruce to climate factors was reduced. The responses of young spruce were significantly correlated with temperature before the growing season and in the growing season. The chronology of the middle-aged spruce showed a significant and positive correlation with monthly mean minimum temperatures in the current April and July. The ring-width index of old spruce was significantly negatively correlated with monthly mean temperature and monthly mean minimum temperature of the previous August. The "lag effect" of high temperatures in the previous growing season was prominent in the old spruce. Spruce within the young and middle-aged groups showed a significant negative correlation with current June precipitation. Adequate precipitation in December was not beneficial for the radial growth of young and old spruce. Our study may demonstrate that future studies should consider the adaptive chronology of age span for spruce in this area to avoid losing or magnifying the climate signals.