Water Arithmetic for Biofuels

Some months ago, I set out to try to make a back of the envelope calculation of how much water is available in the U.S. for growing crops destined for processing into biofuels.  Unfortunately, the more I learn, the larger the envelope seems to get.

My interest was piqued at Synthetic Biology 2.0, where Steve Chu, Nobel Laureate and Director of LBL, suggested there was plenty of water available for growing rain fed crops on marginal agricultural land.  (I've written about this before: "The Impact of Biofuel Production on Water Supplies", and "Live from Synthetic Biology 2.0".)  I have spent most of my life living in Western states, and over the years the snow pack has gotten smaller, summer water shortages more frequent, and acrimony over water issues all the more intense.  So I am somewhat skeptical of the notion that we can somehow conjure up sufficient resources to simply farm our way into energy independence.

Because it seems very hard to sort out just how much water is available from rainfall, or from aquifers, I am going to punt on the calculation.  Perhaps someone else out there can figure out an easy way to make an estimate.  The simplest way to judge how much water can be used for growing biofuels may be to look at the broadest possible level and note how much effort Western states are putting into shoring up water rights, how many are building new pipelines, and how many are putting desalination plants into operation.  The New York Times has a nice story today on all of this, entitled "An Arid West No Longer Waits for Rain":

Some $2.5 billion in water projects are planned or under way in four states, the biggest expansion in the West's quest for water in decades.

..."What you are hearing about global warming, explosive growth -- combine with a real push to set aside extra water for environmental purpose -- means you got a perfect situation for a major tug-of-war contest," said Sid Wilson, the general manager of the Central Arizona Project, which brings Colorado River water to the Phoenix area.

New scientific evidence suggests that periodic long, severe droughts have become the norm in the Colorado River basin, undermining calculations of how much water the river can be expected to provide and intensifying pressures to find new solutions or sources.

..."The Western mountain states are by far more vulnerable to the kinds of change we've been talking about compared to the rest of the country, with the New England states coming in a relatively distant second," said Michael Dettinger, a research hydrologist at the United States Geological Survey who studies the relationships between water and climate.

Mr. Dettinger said higher temperatures had pushed the spring snowmelt and runoff to about 10 days earlier on average than in the past. Higher temperatures would mean more rain falling rather than snow, compounding issues of water storage and potentially affecting flooding.

Changes in rainfall are having very real consequences in the way state and regional planners think about how water is distributed in the West.  States are engaged in legal actions against each other to prevent new pipelines that might redistributed what water there is, and cities are paying for water now legally owned by farmers:

The great dams and reservoirs that were envisioned beginning in the 1800s were conceived with farmers in mind, and farmers still take about 90 percent of the Colorado River's flow. More and more,  [Robert W. Johnson, the Bureau of Reclamation commissioner], said, the cities will need that water.

An agreement reached a few years ago between farmers and the Metropolitan Water District of Southern California, the chief supplier of water to that region, is one model. Under the terms of the agreement, farmers would let their fields lie fallow and send water to urban areas in exchange for money to cover the crop losses.

"I definitely see that as the future," Mr. Johnson said.

Note that this means there will be less water available for crops presently grown as food.  Yet another complicating factor for figuring out how much water will be available for growing biofuels.  All across the globe, the demand for  food crops has increased dramatically as corn is used to make ethanol for fuel.  This has produced mass protest in Mexico, and prompted the Chinese government to curtail ethanol production.  For example, in the 21 December, 2006, Asian Times, "Biofuels eat into China's food stocks".

The story was more explicitly told in Red Herring a few months ago, "Corn Again: 3 Reasons Ethanol Will Be Back":

In more bad news, China on Wednesday halted the expansion of its ethanol industry, blaming it--and other industrial corn uses--for soaring grain prices, according to Xinhua, China's official news agency

Here is a recent column from Bloomberg on water and biofuels, by Andy Mukherjee.  He focuses on the trade-offs and odd cost structures used to encourage biofuel production in China and India.  The piece has some interesting numbers and is basically a tale of woe.

Oddly, near the end of the column, Mukherjee throws down the statement that, "The U.S. has plenty of water; the world as a whole doesn't."  Um, hasn't he heard the phrase "water wars"?  We have those today, every day, in the Western U.S., and they are only getting worse.  Food vs. electricity, waterborne commerce vs. fish?  Most of the fighting is done with words, but bullets and bulldozers come into play none too infrequently.  The only place on the west coast really flush with water is Los Angeles -- witness all the green lawns during the desert summer -- but that's because they just steal it all from somewhere else.                

The year end issue of New Scientist carried an interesting centerfold entitled "The State of the Planet", which, alas, doesn't seem to be available online.  There is a small map of groundwater withdrawal by country.  The U.S. withdraws somewhere between 251 and 500 cubic meters (1000 liters) per person per year, India between 101 and 250 cubic meters, and China less than 100 cubic meters.  Europe, Brazil, Russia, and Canada all fall between 100 and 250 cubic meters per person per year.  Interestingly, only the U.S., China, and India withdraw a total annual amount greater than what is recharged naturally.

Thus we are already operating at a significant, perhaps severe, water deficit, and I just don't see how we can avoid pushing further into negative territory by using yet more water for growing plants used as fuel.

Below are a few resources that may be of use in sorting out how much water is actually available for growing biofuels.

Here is a 1976 report suggesting the total annual precipitation in the US is 5759 cubic kilometers, which is 5759 billion cubic meters, and here is a page from Purdue University stating that:

The U.S. receives enough annual precipitation to cover the entire country to a depth of 30 inches...  Most of this precipitation returns to the water cycle through evapotranspiration. Of the 30 inches of rainfall, 21 inches returns to the atmosphere in this manner. Water loss by plants, the transpiration portion of evapotranspiration, is most significant. One tree transpires approximately 50 gallons of water a day. Approximately 8.9 inches of annual precipitation flows over the land in rivers and returns to the ocean. Only 0.1 of an inch of precipitation infiltrates into the ground water zone by gravity percolation.

A recent OECD report puts US water consumption at ~518 billion cubic meters annually, or ~1730 cubic meters per capita annually.

If you prefer thinking in old fashioned gallons, here is a report from the EPA entitled, "How We Use Water in These United States." 

Here is the USGS Groundwater Atlas of the United States, and Estimated Use of Water in the US in 2000.