Shifts in vegetation phenology induced by climate change are substantially modifying various ecosystem processes, and those changes can in turn affect weather and climate systems. Realistic modeling of spring vegetation green-up is critical to improving process-based ecosystem models of the Tibetan Plateau (TP) and for better understanding of the coupling between TP terrestrial biophysical processes and the Asian monsoon system. However, no model is available for simulating the vegetation green-up date (VGD) across the entire TP. In this study, we first assessed the ability of several existing state-of-the-art phenological models to estimate VGD across the TP. We then modified the existing models by adding environmental constraints identified by partial least squares analyses. The modified models simulated VGDs with lower estimation errors than other models (Mean absolute error: 8.2 days vs. 8.7-12.9 days; P < 0.01, t-test). Moreover, although our model captured the inter annual variations in VGD better than any previous model, the correlation coefficient between predicted and remotely sensed VGDs was still low, especially in the western TP. This study revealed the necessity of considering multiple factors in VGD models and highlighted the challenge of developing models that will better represent phenology in future ecosystem models.