Based on the China Meteorological Administration homogenized dataset, the spatial and temporal variations of monthly mean temperature at 71 stations with elevations above 2000ma.s.l. in the eastern and central Tibetan Plateau (TP) during 1961–2004 are examined. Using principal component analysis (PCA) in S-Mode, four main subregions of temperature variability in the TP are identified. Trend analyses are then carried out on mean series calculated for each subregion and for the eastern and central TP as a whole at annual, seasonal and monthly resolutions. The NE subregion has the most significant warming trends especially in winter and autumn. The PCA method is also applied to sea level pressure (SLP) over the domain of 20°–60°N and 60°–130°E. Different atmospheric circulation patterns are classified in summer and winter. Moreover, temperature series in the TP are often correlated with SLP and associated with different atmospheric circulation patterns. There are more water vapor flux and total cloud cover in the warmer summers and winters. Therefore, we suggest that change in atmospheric circulation is one of the important factors contributing to the recent climate warming of the TP.