Hmm, I don't know what happened. I really meant to say that it is really a matter of mass balance. As I was saying, at steady state (inputs=outputs), the volume of water stored in aquifers and surface water bodies will remain relatively constant. If inputs >outputs, then there will be an increase in water being stored. If output>inputs, then there will be a decrease in water being stored. When we are talking about water that can be used by humans, as stoo_pid pointed out, we are limited to relatively shallow groundwater and fresh surface water (rivers and lakes), which is a very small percentage of the total water budget. The main input is precipitation, and outputs include evapotranspiration (evaporation and transpiration by plants), runoff to the ocean via rivers, and human consumption (drinking water, agricultural uses, industrial uses, etc.). There is not a lot we can do about input to the system on a short term time scale. Whether or not we can effectively change the climate on a larger time scale is another discussion all together. As andy-hughes pointed out, some of the outputs are also effected by climate, but human usage is a very large output and is increasing as the population increases. Another effective output is water contamination, which has become a very serious problem world wide. I say it is an effective output because it causes a certain volume of water to be unusable by humans. Again, most water contamination is caused by humans. So I would say that the main cause of a water shortage is humans.