The Gravity Recovery and Climate Experiment (GRACE) satellites have collected data more than 11 years. With these data as well as the improvement of data processing methods, we can analyze the secular trends of mass variations rather than just the large-scale seasonal variations. Using these data we can study the uplift process and its mechanism of the Tibetan Plateau. In this paper, 11 years of satellite gravity data of GRACE Release 05 models (Center for Space Research (CSR), GeoForschungsZentrum (GFZ), and Jet Propulsion Laboratory (JPL)) from January 2003 to December 2013 were used to reveal the secular trends of gravity anomaly variation within the Tibetan Plateau by means of the least square method. A reduction of hydrological signals from the detected integral secular trends using global hydrological models (Global Land Data Assimilation System, GLDAS and Climate Prediction Center, CPC, averaged) is attempted. The glacier model result provided by Paulson is used to reduce the GIA (Glacial Isostatic Adjustment) effect. Also, we use the scaling factor method to weaken the GRACE post-process errors. At last we obtain the secular trends of gravity anomalies in China and adjacent areas. The results are largely similar to the previous research. Also, we find a remarkable increase signal of gravity in the Tibetan Plateau. Such a change cannot be attributed to the lake expanding and precipitation increase in the Tibetan Plateau. Instead it is possibly caused by the crustal thickening due to the subduction of Indian plate beneath the Eurasian plate. The continuing collision and compression between the India and Eurasia plates result in intense uplift and orogenic extension of the Tibetan Plateau. We built a forward model with 3D rectangular prisms according to the GPS results and Airy isostatic hypothesis. Using the Bangong-Nujiang suture zone as the boundary, the plateau was divided into southern and northern parts. The modeling results show the gravity anomalies are increasing at a rate of 0. 2 mu Gal . a(-1) in the north, which is a little less than the result from GRACE solution 0. 3 +/- 0.08 mu Gal . a(-1) (corresponding to a crustal thickening rate of 3 mm . a(-1)). The rest unexplained part is possibly related to lakes, glacier, permafrost and others. Meanwhile, our results show the southern part has a rate of -0. 62 +/- 0. 06 mu Gal . a(-1), which is presumably the combined influence of tectonics and groundwater and so on. It is a challenging to explain the positive signal in the Tibetan Plateau clearly with current knowledge. The new GRACE Follow-On satellites may provide us more means and evidence. With the continuing accumulation of observations, we can combine a variety of modern geodetic techniques (GRACE, GPS, absolute gravity, remote sensing, and so on) to study the geodynamics of the Tibetan Plateau more accurately.