Kathmandu, the capital of Nepal, is under increasing water stress. A rising water demand in the Kathmandu Valley is initiated by urbanisation with a rate of 4.7 % per year and amplified by a growing trend of per capita water consumption. At the same time, water availability seems to deteriorate. However, the severity of the problem is hardly known due to immense data gaps. This research aims to (1) estimate the urban water demand from literature research, (2) collect new field data about spring and stream flow, water quality, and ecological stream health, and (3) analyse remotely sensed data in order to characterise the linkages and inter-dependencies between land use, hydrology, and water quality in the Kathmandu Valley. The field data were collected in August and September 2016 and helped to get a decent picture of the water situation. The landuse classification performed using Landsat 8 images had an accuracy of 88 % and proved to be a meaningful source of data to link to the other research topics. Automated SEBAL was attempted to be used for computing evapotranspiration estimates but cannot successfully be used in mountainous areas yet. Penman-Monteith equation was used as alternative. A simplified water balance was performed for the months of August and September 2015. The outcome showed estimates to be 2, 2 x 106 m3/d for ET, 5.6 x 106 m3/d for precipitation, 2.0 x 106 m3/d for the Bagmati outflow and 1.4 x 106 m3/d for groundwater infiltration. To be able to comment on long term groundwater depletion, an annual water budget needs to be performed. The clearest influence on water quality parameters (pH, DO, EC and Rapid Stream Assessment class) was found to be the presence of developed area. With more than 20 % developed area within the considered sub-watershed, the water becomes of such poor quality that it cannot be used for any purpose without extensive treatment. In the headwater areas of the Valley, a significant amount of water with a higher quality (i.e. drinking water) is found in the streams. In September 2016 the total inflow of drinkable water to the Valley floor was estimated to be at least 1.0 x 106 m3/d. This value cannot be considered as fully available for anthropogenic purposes, as water needs to remain in the streams in order to not jeopardise ecosystem services and other functions.