A way to look at the evolution of agriculture is as a progression from purely extractive hunting and gathering toward a system that produces as much as it consumes. This is a progression that is still proceeding and many of the threats of depletion heralded in activist screeds can be understood as aspects of this evolution that are currently being worked. Among those highly publicized concerns are the worries about soil carbon that are being exploited lately, and a related concern for water.

New space observations reveal that since October 2003, the aquifers for California's primary agricultural region – the Central Valley – and its major mountain water source – the Sierra Nevada – have lost nearly enough water combined to fill Lake Mead, America's largest reservoir. The findings, based on satellite data, reflect California's extended drought and increased pumping of groundwater for human uses such as irrigation. . .

"GRACE data reveal groundwater in these basins is being pumped for irrigation at rates that are not sustainable if current trends continue," Famiglietti said. "This is leading to declining water tables, water shortages, decreasing crop sizes and continued land subsidence. The findings have major implications for the U.S. economy, as California's Central Valley is home to one-sixth of all U.S. irrigated land and the state leads the nation in agricultural production and exports." . . .

The California results come just months after Matt Rodell of NASA's Goddard Space Flight Center, Isabella Velicogna of UCI, and Famiglietti found groundwater levels in northwest India declining by 17.7 cubic kilometers per year between October 2002 and August 2008, a loss attributed almost entirely to pumping and consumption of groundwater by humans.

"California and India are just two of many regions around the world where GRACE data are being used to study droughts, which can have devastating impacts on societies and cost the U.S. economy $6 billion to $8 billion annually," said Rodell, who was Famiglietti's doctoral student at the University of Texas at Austin.

The loss of soil carbon due to agriculture is a problem that has been in focus for several decades and a number of improved methods that reduce, eliminate or reverse loss have been developed and partially adopted.

Water has had less attention but there are methods to replenish aquifers. Seeing water as something to be actively managed rather than as only a resource to be extracted and exploited is a first step that continues and expands the evolution of agriculture from the hunter-gatherer mind set to a more productive one. It isn't the old empty ideas of reducing rates of extraction that are useful, it is ideas of replenishment that are needed.

There are two broad methods of replenishment: infiltration and injection. Both leverage the seasonal surpluses of water which can be destructive and so are a management problems of another sort. Managing surpluses can alleviate seasonal shortages by storing excess water for later use. Both floods and droughts are mitigated by the same management methods. And it isn't really surprising that managing water in wet seasons would also help reduce loss of soil carbon and increase sequestration of new carbon.

An earlier post, Water Pockets, discussed the use of various catchment systems - zai, bunds, Keyline - and the use of biochar to increase water holding capacity. They are methods to slow and spread water flow so that it lingers longer on the land before reaching water courses to hurry away to the sea, or flood downstream population centers. It's also a way to help replenish aquifers. There has been some use of infiltration ponds in the California southern central valley for this purpose, and there are lots of opportunities for more. If, as is predicted, a warmer winter climate in the area would bring more rain but less snow in the mountains this will become increasingly useful since more of the total water for the year will hit the valley in winter rather than being banked as snow in the mountains for year long supply.

This is also why I think that the trend toward "plasticulture" methods of reducing water use are counterproductive. Parsimonious dripping of water on crops rather than whole field sprinkling or flood irrigation may use less water for a given crop - more crop per drop as the advocates chant - but it is contrary to good water management when the whole system over time is considered. It's a way to go broke more slowly, but not a way to stay in business. The best method is flood irrigation but it must be done well so that water infiltrates aquifers. If it runs off fast then it accomplishes none of the water management objectives.

This is even more important at higher elevation, in the foothills of the mountains east of the valley, since water that lingers on the land or is otherwise caught and held at that level reduces floods and feeds into ground water systems. Much of it eventually ends up in the same water courses that feed the valley, but there's a delay since the water flows underground at a much slower rate. For every river above ground there is a larger but slower one flowing beneath it, fed by waters that infiltrated the ground during the wet season.

I'm currently investigating concepts of ground water injection. It's the opposite of pumping in that it puts water in the well rather than removing it, and would be done when there was too much surface water. Both activities take energy and equipment so they are more costly and complex than passive infiltration systems, but it may be that they make sense in some cases. I can imagine some system of water banking where stored water is withdrawn when required and deposited when available in surplus. It could work on either a credit or debit basis: repay what was borrowed or spend what was previously deposited.

Your limits would be defined by your ability to pay, which is ultimately the case whether we have a formal system or not. When the water is gone, it's gone. There's no free lunch, but payment is on the honor system. You can cheat for a while but eventually lunch will no longer be served and you will just have to go hungry. Literally.