The variation of the vegetation growing season in the Three-Rivers Headwater Region of the Tibetan Plateau has recently become a controversial topic. One issue is that the estimated local trend in the start of the vegetation growing season (SOS) based on remote sensing data is easily affected by outliers because this data series is short. In this study, we determine that the spring minimum temperature is the most influential factor for SOS. The significant negative linear relationship between the two variables in the region is evaluated using Moderate Resolution Imaging Spectroradiometer-Normalized Difference Vegetation Index data for 2000-13. We then reconstruct the SOS time series based on the temperature data for 1960-2013. The regional mean SOS shows an advancing trend of 1.42 d (10 yr)(-1) during 1960-2013, with the SOS occurring on the 160th and 151st days in 1960 and 2013, respectively. The advancing trend enhances to 6.04 d (10 yr)(-1) during the past 14 years. The spatiotemporal variations of the reconstructed SOS data are similar to those deduced from remote sensing data during the past 14 years. The latter exhibit an even larger regional mean trend of SOS [7.98 d (10 yr(-1))] during 2000-13. The Arctic Oscillation is found to have significantly influenced the changing SOS, especially for the eastern part of the region, during 2000-13.