Switching to drip irrigation seems like a pretty commonsense way for farmers to save water. Instead of using giant sprinkler systems or ditches to get water to crops—losing much of it along the way to runoff, seepage, or evaporation—drip irrigation uses a system of hoses that delivers water directly to the plants' roots. It allows farmers to use no more water than their crops actually need, an alluring prospect in a world of diminishing usable water supplies and a growing number of mouths to feed. This promise of getting more "crop per drop" has caused drip irrigation to become the standard prescription for making agriculture work in arid regions, both in the American Southwest and the dry parts of the developing world.

But what if drip irrigation isn't actually that big of an improvement over traditional irrigation methods? That's the conclusion of a new paper by Frank Ward and Manuel Pulido-Velazquez that models the hypothetical results of varying levels of drip-irrigation subsidies on water use in the Rio Grande basin of Colorado, New Mexico, and Texas. The question is not whether drip irrigation is efficient—it unquestionably is—but whether traditional irrigation actually wastes as much water as critics say it does. Some of the water used in sprinkler or flood irrigation evaporates and is lost from the watershed. But a lot of it runs off into streams or percolates into the ground, where it helps recharge the basin's aquifers. If these return flows are taken into account, traditional irrigation doesn't look as bad.

The broader point is that in regions suffering from water shortages, there's a big difference between consumptive and non-consumptive uses of water. Letting the water run while you brush your teeth is not actually that big of a problem, because the water just ends up back in a local river to be used by people downstream. Irrigation water that flows back into creeks or soaks into the ground isn't lost, it's just temporarily relocated. Evaporation is the only way that a watershed really loses water.

Unfortunately, while some U.S. cities—like Las Vegas—get credit for water they return to its source, the end-users of water do not. A Las Vegas resident gets charged the same amount for a gallon of water used to take a shower, which eventually goes back into the Colorado River, or a gallon of water used to water the lawn, which probably does not. If water utilities wanted to get serious about discouraging the consumptive use of water, they would charge differential rates for water from outdoor spigots, which is far more likely to be lost to evaporation than water used indoors, or give customers some kind of credit for the water they return through their sewer pipes. But for now, at least, water is too cheap to make the extra metering costs worthwhile.

--Rob Inglis, High Country News