Spring phenologies are advancing in many ecosystems associated with climate warming causing unpredictable changes in ecosystem functioning. Here we establish a phenological model for Daphnia, an aquatic keystone herbivore based on decadal data on water temperatures and the timing of Daphnia population maxima from Lake Constance, a large European lake. We tested this model with long-term time-series data from two lakes (Müggelsee, Germany; Lake Washington, USA), and with observations from a diverse set of 49 lakes/sites distributed widely across the Northern Hemisphere (NH). The model successfully captured the observed temporal variation of Daphnia phenology in the two case study sites (r(2) = 0.25 and 0.39 for Müggelsee and Lake Washington, respectively) and large-scale spatial variation in the NH (R(2) = 0.57). These results suggest that Daphnia phenology follows a uniform temperature dependency in NH lakes. Our approach--based on temperature phenologies--has large potential to study and predict phenologies of animal and plant populations across large latitudinal gradients in other ecosystems.