Phenology is a very effective indicator to detect and measure the impact of climate on vegetation. Remote sensing can play an important role by monitoring the phenological activity of vegetation communities at large spatial scales. The aim of this study is to identify modality and key phenological transition dates in a year, and to examine the spatial variation of these phenological patterns on a global basis. The transition dates in each vegetation phenological cycle are defined as the onsets of greenup, maturity, senescence, and dormancy. For this purpose, we utilize Moderate-Resolution Imaging Spectroradiometer (MODIS) data with a spatial resolution of one kilometer and a temporal interval of sixteen days for the period from 1 January to 31 December 2001. To capture the temporal variation of vegetation, we select an enhanced vegetation index (EVI) calculated from the MODIS Nadir Bidirectional Reflectance Distribution Function (BRDF) Adjusted Reflectance (NBAR) Product. A sigmoidal growth model is used to separately fit the increasing and decreasing sections of an annual trajectory of EVI. The extreme points of the curvature-change rate derived from these fitted growth models allow the automated determination of both phenological transition dates and of cycle modality. This resultant global phenology is reliable because (1) the MODIS NBAR-EVI dataset is stable and consistent; and (2) the algorithm is not only ecologically based but also treats each pixel individually without using regional thresholds and empirical constants.