Anticipating species movement under climate change is a major focus in conservation. Bioclimate models are one of the few predictive tools for adaptation planning, but are limited in accounting for (i) climatic tolerances in preadult life stages that are potentially more vulnerable to warming; and (ii) local-scale movement and use of climatic refugia as an alternative or complement to large-scale changes in distribution. To assess whether these shortfalls can be addressed with field demographic data, we used California valley oak (Quercus lobata Nee), a long-lived species with juvenile life stages known to be sensitive to climate. We hypothesized that the valley oak bioclimate model, based on adults, would overpredict the species' ability to remain in the projected persisting area, due to higher climate vulnerability of young life stages; and underpredict the potential for the species to remain in the projected contracting area in local-scale refugia. We assessed the bioclimate model projections against actual demographic patterns in natural populations. We found that saplings were more constricted around surface water than adults in the projected contracting area. We also found that the climate envelope for saplings is narrower than that for adults. Saplings disappeared at a summer maximum temperature 3 °C below that associated with adults. Our findings indicate that rather than a complete shift northward and upward, as predicted by the species bioclimate model, valley oaks are more likely to experience constriction around water bodies, and eventual disappearance from areas exceeding a threshold of maximum temperature. Ours is the first study we know of to examine the importance of discrete life stage climate sensitivities in determining bioclimate modeling inputs, and to identify current climate change-related constriction of a species around microrefugia. Our findings illustrate that targeted biological fieldwork can be central to understanding climate change-related movement for long-lived, sessile species.