In thinking about drought, it's useful to have an understanding of the water cycle now. This morning, I've been studying a 2006 review published in Science: Global Hydrological Cycles and World Water Resources, by Oki and Kanae (subscription required). The above figure (click for a larger version in a new window) summarizes the present situation. The flows are in cubic kilometers per year, while the storage quantities are in cubic kilometers. Some things that seem noteworthy:
- The net transfer of about 45 km3/yr from ocean to land, returned to the ocean via river flow.
- Human usage of about 10% of that - 2.7 km3/yr for agriculture, 0.8 km3/yr for industry, and 0.4 km3/yr for domestic use.
- Globally, water input to cropland is still 80% from rain, with only 20% from irrigation.
The blue areas are where the water is. Obviously, in a lot of cases, this is not where the people are. A measure of water stress is the total amount of human withdrawals as a fraction of the available river flow. When this goes over 0.4, water stress is considered high (because at least in the dry season, usage is likely to be almost total with rivers failing to reach the sea). The map of this index is as follows:
Again, click for larger version in a separate window. The western US, southern Europe, the Middle East, northern India/Pakistan, and northern China are the epicenters of water stress. These regional variations lead people to undertake large engineering projects to move water from the unstressed places to the stressed. Perhaps the most famous is the South-North Water-Transport Project in China.
It's interesting to compare the water stress map above to the first principal component of the PDSI:
Apparently, the dry will be getting drier.
This post is part of a series on the future of drought.