The arid Tarim basin, located in northwestern China, is strongly influenced by the changing environment triggered by growing irrigating agriculture and climate change. These changes have caused water scarcity in the lower and middle reaches of the Tarim and a shortening of the endorheic river, while an increased runoff has been observed at the head water of the Tarim. The Tarim River with its main source river Aksu, that originates in the kyrgyz part of the Central Tien Shan, is significantly fed from meltwater released by the glaciers in the Central Tien Shan. Downstram riparians are highly dependent on its water for irrigated agriculture. Thus, scenarios on future glacier changes and their consequences on the river runoff are highly relevant.\nExisting studies are often focusing on changes of glacier area which do not allow any direct conclusions regarding mass changes. Moreover, glacier area changes are only an indirect signal of climate change. Studies dealing with glacier mass changes in the Aksu catchment are mainly local considering single or small samples of glaciers. Region-wide glacier mass change investigations, comprising entire catchments and mountain ranges, are limited to the recent decades. They are evolving with the availability of satellite laser altimetry and satellite gravimetry, the latter accompanied by high uncertainties. In order to make statements regarding the temporal evolution of glaciers, long-term measurements of glacier mass budgets are needed. Therefore, the present thesis addresses in particular the determination of region-wide glacier mass changes since 1975 using declassified satellite imagery, SRTM data and SPOT-5 stereo imagery.\nGeodetic glacier mass balances were determined by DTM differencing using stereoscopic\nHexagon MC (Mapping Camera) imagery acquired between 1973 and 1976, 2000 SRTM data and 2009 SPOT-5 satellite imagery. Geodetic mass balance estimation using multitemporal digital terrain models required model co-registration to the SRTM master DTM.\nA two stage procedure was used for co-registration: first, to account for any tilt, the global trend over non-glaciated terrain was calculated; then horizontal offsets were minimized based on an analytical relationship between slope angle, aspect and elevation differences. The mean penetration depth of the SRTM-C-Band signal was determined at 2.2m using ICESat data and accounted for during DTM differencing. For outlier filtering an elevation dependent approach was developed, assuming decreasing variance of elevation difference over glaciated terrain towards the accumulation regions. This approach is more realistic than the previously used zero-assumption, that assumes stable conditions in the acccumulation regions. The resulting total glacier mass loss of 0.35±0.34 m w.e. a−1 between 1975–1999 was within the global average. However, in the recent years (1999–2009) a slightly decelerated mass loss of 0.23±0.19 m w.e. a−1 was observed. The decelerated mass loss might be explained by an increase of precipitation in the recent decades. Additionally, significant mass loss could also be observed for debris-covered glaciers, where melting is retarded when debris thickness exceeds a threshold of approximately 2 cm, whilst the glacier area shrinkage was comparatively low. Overall, glacier mass loss was heterogeneous with an increasing mass loss from east to west and from the inner to the outer ranges of the Central Tien Shan. Additionally, several surge-type glaciers could be identified which have not been reported in the literature so far. The contribution to the runoff of Aksu River due to glacier imbalance has been determined at approximately 20%.\nBesides knowledge about the glacier mass change, information about the total ice volume stored in glaciers are of high importance for sustainable water management. Data about glacier ice volumes are rare for remote mountain areas like the Central Tien Shan. Existing thickness measurements are often point-based and restricted to smaller glaciers only. Moreover, point-based thickness measurements using Ground Penetrating Radar (GPR) hardly permit any conclusions to be drawn about the spatial distribution of the glacier ice volume. \nBesides the surface slope the basal shear stress and the contribution of the basal velocity to the glacier surface velocity are the main influencing factors for ice thickness estimations. Using radar thickness measurements, glacier surface velocities and the glacier surface slope, the latter as an approximation for the slope on the glacier bed, an empirical relation could be found to estimate the contribution of the basal velocity to the glacier surface velocity. For the four largest glaciers in the Central Tien Shan ice thickness was successively estimated relying on Glen’s flow law based on the Shallow-Ice Approximation by applying the equation of laminar flow. The results indicated a mean ice thickness for the debris-covered glacier tongues of about 70 m to 140 m. This is significantly lower than previous studies suggested which assumed a constant basal shear stress for an entire glacier system or a constant contribution of the basal velocity. The estimated ice thickness values coincide roughly with GPR thickness measurements.\nAs a consequence of using a localized approximation of the basal velocity the mean deviation could be reduced from approximately 200 m to 50 m. Based on geodetic mass budgets it was estimated that the glaciers lost 6% to 28% of their volume since 1975. This is significantly lower than what has been estimated for non debris-covered glaciers in High Asia.