The middle reaches of Yarlung Zangbo River (YR) and its two tributaries is a region in central Tibet Autonomous Region where the industrial and agricultural economy is most concentrative and developed. Remote sensing of land surface parameters has significant meaning for crop growth monitoring, crop yield assessment and disaster monitoring. It also has great theoretical importance for understanding energy and water cycle in local area and even for surrounding regions. In this paper, four scenes data in 2003 under nearly clear weather conditions are selected as the spring, summer, autumn and winter cases. Different algorithms for AVHRR and MODIS data will be applied to retrieve land surface temperature (LST), albedo, Normalized Difference Vegetation Index (NDVI), vegetation coverage, emissivity, and water vapour content, respectively. The derived LST, albedo are validated with field measurements and MODIS products. The estimations are in good agreement with in-situ measurements, with MODIS retrievals being slightly better than those of AVHRR. This means that the adopted remote sensing retrieval algorithms are not only applicable but also simple for the study area. As called the "water vapour inflow corridor", the valleys of YR and its two tributaries are proved to have higher LST and larger atmospheric water vapour content. Spatiotemporal variations of land surface parameters in the middle reaches of YR and its two tributaries can also be clearly identified.