The National Water Situation and What To Do About It

By Peter Gleick

Peter Gleick, a leading expert on water issues, presents a wide-ranging discussion of our circumstances and solutions to our problems.  The central theme is the need for taking the “soft path”, that is conservation, rather than indulging in massive water moving projects.

I'm going to do a lot of different things this morning. I'm going to talk about national water policy and recommendations for national water policy. I'm going to talk about the concept of peak water. I'm going to talk a little bit about drought and floods and climate change. I'm going to talk about the concept of the soft path for water, which I think of as the solution to our water problems. I am going to close with some recommendations at the national level but with the understanding that a lot of what we understand about water at the national level is relevant for the local level and relevant for the international challenges that we face.

Given the questions that we've already heard about agriculture, climate change, infrastructure, groundwater overdraft, and regional issues, I think you get a sense already about how complicated and interesting water is. So I'm going to try and touch on some of those issues as well and introduce you to some concepts that I think help us think about not just the problem of water but solutions around water.

There's great quote about drought and water in California. It comes from John Steinbeck’s East of Eden and in he writes: “it never failed that during the dry years the people forgot about the rich years, and during the wet years they lost all memory of the dry years, it was always that way.” We have wet years, we have dry years, we panic during the wet years and we forget about the drought. We panicked during the drought and we forget about the wet years.

Here’s something we should not forget: the concept of peak water. You have probably heard about the concept of peak oil. The concept of peak oil, which was presented in the late nineteen sixties by an oil geologist named M. King Hubbard, is that oil is a non-renewable resource. There's a certain amount of oil that's been laid down over millions of years and it's in these pools and geologic fields that we can tap. We tap them, we drill oil wells, and we start pumping them out. Over time as we pump more of this non-renewable resource out, it gets harder to reach. We use up the easy resources and the harder ones are more expensive to pump and eventually we will reach the point of peak oil, peak production of oil. The process shows that production begins, we have exponential growth in production, then a peak, and finally a decline.

In 1970 the United States reached peak oil: oil production in the United States peaked in 1970 and it’s declined since then. And in the last couple years with fracking and with expansion of oil drilling in the Bakken reserve in North Dakota and elsewhere, oil production in the US is starting to go up again. We're not at the peak of 1970 and I'm not sure we’ll reach that peak again. We might, but it will decline after that and globally it's going to decline. And that's the concept of the problem of peak oil.

Peak water is similar and different. Let me describe three forms of peak water. What I call peak renewable water, peak non-renewable water, and peak ecological water.

First of all think about an exponential curve, exponential growth. You all know what exponential growth is. It's what you want your retirement account to look like. It's what you want the stock market to look like. It’s what some economists want our overall economy to look like. It's what we see with population growth. It's behind a lot of our problems, not just our economic structure in a positive way, but a lot of our problems. Much of our water use worldwide has followed this exponential growth curve and it's followed it because it's been following our population growth curve. As population grows, as our economies grow, our demand for things grows including energy, minerals, and resources like water. And history has shown an exponential growth in the demand for water.

There are limits, however, to how much water we can tap on the planet. Water is a renewable resource: mostly. In the morning session you heard a little bit about the hydrologic cycle that you will remember from fifth grade. We have evaporation, the formation of clouds, condensation, precipitation of water back down to the ground, runoff back to the oceans, and evaporation again. That’s the hydrologic cycle. The hydrologic cycle feeds our rivers. The hydrologic cycle is why water mostly is a renewable resource. So we see river flows, we take water out of our rivers but that doesn't affect how much water we're going to get next year in those rivers. It's a renewable resource. It's a cycle. That’s the definition of a renewable resource. It's not limited by the amount of a stock, because it's renewable. But it is limited by the flow. This is the first definition: peak renewable water is our ability to take a renewable resource up to the limits that nature provides.   

Think about the Colorado River or any river. We tap that water, then we start exponentially taking water out of that river and we can continue to increase our demands from a renewable resource up until the limit of flow. And this is one of the characteristics of our water problems worldwide. We are reaching peak renewable limits on a variety of rivers around the world, on a variety of renewable water resources around the world. For example, this morning we heard about the Yellow River in China. We take the entire flow of the Yellow River now during parts of the year. We take the entire flow of the Colorado River. We take the entire flow, effectively, of the Nile River. We're reaching peak renewable limits on a number of water resources. That’s part of the definition of “peak water.”

But not all water resources are renewable. Let's go back to oil for a minute. Oil was laid down over eons and we're pumping it faster than it’s naturally created. Much, much faster. That's also true for some of our water resources. Groundwater is the result of rainfall seeping into the ground and filling up the interstices in our geological formations. There are many different groundwater basins around the world. As Richard Hazlett noted about the Ogallala Aquifer and other basins around the world, they are in danger of overdrafting.

What does that mean? This is the term “peak non-renewable water.” Non-renewable resources are not flow limited like renewable resources, they are stock limited:  limited by the amount that's there. Like oil underground, you can have what’s underground but you can't have any more. Now the reality is that there are physical and economic constraints, so we can't even have all of any stock of resources. And that's true for water as well. But, when you pump out something faster than nature recharges it--oil or gas or water—it is non-renewable. Groundwater levels drop and it becomes more expensive to pump.

This is what is occurring in the Ogallala Aquifer and the Central Valley of California: overpumping, increasing costs, and ultimately limits on our ability to tap a non-renewable resource. Instead of this exponential curve, we have an exponential curve followed by a peak and a decline. We’re reaching peak non-renewable limits on a whole range of our water resources. And this is important because thirty to forty percent of global agricultural production comes from non-renewable groundwater, i.e., groundwater that we’re pumping faster than it is being recharged. We can continue to do that for a little while but it's not sustainable. Somewhere we're going to have to find a sustainable source of water, in the long run, to satisfy that demand for agricultural productivity.

The comment was made earlier about how important food is for us. Eighty percent of the water that we use worldwide goes to grow food. Eighty percent of the water that humans use in California goes to grow food. California turns out to be a pretty great place to grow food. Central Valley soils are fantastic. We have water. We have a remarkable climate. But the water resources we're applying to those demands are not always sustainable and that’s the challenge of peak non-renewable water.

The third concept is “peak ecological water.” Our ecosystems are incredibly dependent on water, of course, and they're hurt by human withdrawals of water. The water that we take out for the things we want to do is water that ecosystems used to use for something. The Central Valley used to be a huge wetland and it supported an incredibly diverse set of fish populations, and a huge bird population, and a huge natural ecosystem of ruminants. As we took water out of that system, it shrank and the water available for ecosystems, shrank as well. Fisheries started to suffer; migratory bird populations plummeted, and the entire ecosystem was transformed. This was a cost of human development.

We see similar consequences around the world. So the concept of peak ecological water is the idea that as we take water out of a system, the economic benefit we get starts to grow. We can produce food. We can make semiconductors. We can do the things that we want with water and the use of that water produces an economic benefit. But every unit of water we take out of an ecosystem causes a little bit of harm, ecological harm. And at first we don't really notice, or we don't really care, or we don't really know we ought to care and notice, and that was the reality of eighteenth century and nineteenth century and 20th century water development. We didn't know or we didn't see or we didn't care about the ecological harm that was accompanying this exponential growth and economic benefit from using water. But there comes a point when the next unit of water you take out of a system causes more harm then it provides benefit.

We're really good at measuring the economic benefit of using a gallon of water. We’re not as good at measuring ecological harm in economic terms.  We measure this economic growth but it is hard to measure the economic harm caused by the ecological damage were causing. That doesn't mean it isn't real: our inability to measure it doesn't mean it isn't real. The concept of peak ecological water is that in many places around the world we're reaching a point where our additional use of water now causes more harm than it provides benefit. That's the simplest way to think about it. I would argue we are past the point of peak ecological water in places like the Aral Sea of the former Soviet Union where we used the entire flow of the Amu Darya and the Syr Darya rivers. The Aral Sea started to shrink and all 24 species of fish endemic to the Aral Sea are now extinct. And one could make that same argument for the Colorado River or the Central Valley of California or all sorts of places where we're beginning to understand the ecological damage of our water use.

Now, here is my one slide. On its y-axis is the gross domestic product of the United States in billions of dollars, in inflation-adjusted terms. On the x-axis is time, from 1900 to 2005. There are two lines. There's a red line and there's a blue line. The red line is the gross domestic product of the United States, revealing the exponential growth curve that economists love to see. Our economy has grown enormously over the last 105 years. You see there are little ups and downs, and you know these are recessions and depressions. But in general the graph captures our exponential growth curve. The blue line is an estimate of total water withdrawals in the United States, for everything: irrigation, domestic use, washing your clothes, making semiconductors, flushing our toilets, cooling power plants, everything. This is our estimate of total water withdrawals in the United States. The y-axis over here measures water use in cubic kilometers per year. The units measuring economic growth are on the left.  

There's a lot going on in this graph, a lot of really interesting things. First of all, if you look from 1900 to 1975 you'll see these two curves are pretty much lined up. They go up together to about here and that's the idea as our economy grows, so does our demand for water. One thing that's not on this graph but could easily be, is the rate of growth in the U.S. population. Our population follows this red line, exponential growth. So, for the first three quarters of the twentieth century, the population curve, the GNP curve, and the water curve went up together. Economy grows, population grows, and demand for water grows very substantially. That seventy-five year period was really the period of time when we put in place our water infrastructure, in a concrete sense, in a literally concrete sense, our water management systems, and our water education systems. This was a time when hydrologists and engineers were trained to assume exponential growth in water demand would accompany our growth in population and economy and trained to build structures to deal with that. They built reservoirs to store water in wet years, so we could use it in dry years. They built aqueducts to move water from the northern part of California to the southern part, from the Sierra Nevada to the coast, from where we have water to where we demand water. This is when we built the big dams in the Colorado River that now store more than four years of average flow of the Colorado River, the big dams on the Columbia River that powered the aluminum industry that helped us win World War II, and much of the rest of the water-related infrastructure that has brought enormous benefits to this country.

Then something strange happened, in the late 1970s/early 80s, those two curves split apart. Now total withdrawals in the United States of water are less today than they were in 1980. Our population continued to grow, our GNP continued to grow, but our total demand for water, our total use of water in the United States is less today than it was thirty years ago. On a per-capita basis, the amount of water each of us uses per person has gone down much more because populations continue to grow. What happened is a remarkable thing because it's not what we were taught and trained to assume and expect and it's still not the way water managers think.

Here's another thing to think about. Suppose that blue line had continued to grow exponentially. This is how much additional water we would be using today: twice what we're actually using. Where would that water have come from? Where would we have found another 600 or 700 cubic kilometers of water? We are at peak water on the Colorado River. We can't have any more water out of the Colorado River.

So the next question one might ask is what's going on here? That's a complicated question and I don't have a complete answer. But let me give you some hint of what's going on. First of all we’re reaching peak water. It was easy to continue to expand water use when we weren't completely tapping the Colorado River. When we weren’t over drafting the Ogallala aquifers or the Central Valley aquifers or when we could take more water out of the Sacramento-San Joaquin. But as we approach peak water, peak renewable water, peak non-renewable water, and peak ecological water, it gets harder and harder economically, physically, politically, and environmentally to build more infrastructure to tap water systems. So, there is a physical constraint on our previous pattern of consumption. And in that sense, this is an indication of peak water.

The 1960s and 1970s were when the serious environmental problems associated with our economic development and our industrial revolution started to become clear. This helped drive the awakening of the environmental movement. Lake Erie was dying. The Cuyahoga River caught fire (again) in 1969 but this time on national TV. We had a massive oil spill off the Santa Barbara coast that same year. There was a growing awareness about the air-quality problems in our major cities and so the United States Congress passed a number of fundamental environmental laws at the national level. The Clean Water Act and the Safe Drinking Water Act in particular created new regulatory authorities and mandates to protect our drinking water and rivers and streams, and they have had considerable positive impact.

One was to substantially clean up discharges in our rivers and streams. Many of our rivers and streams are much cleaner now than they used to be. Another is that industries discovered that they had to cut their wastewater discharges and they had to stop dumping uncontrolled garbage into our rivers and streams. One way to do that is to collect all that stuff and treat it but it turns out that a better way to do it is not to produce the stuff in the first place. It turns out that one of the ways not to produce it in the first place is to change the way we use water. For example, in the 1920s it used to take 200 tons of water to make a ton of steel, and much of that two hundred tons of water ended up as about 190 tons of wastewater heavily contaminated with heavy metals. In the 1980s the steel industry had cut that to twenty tons of water to make a ton of steel and they did it not because they were trying to save water but because they were trying to reduce their pollution, their wastewater discharges. We're still making steel and yet demand for water goes down.  

Another factor associated with the problem of peak water was the realization that we could actually do what we want with a lot less water. And that's the issue of water-use efficiency. Some people say “conservation” but I want you to think about it as “efficiency” and “productivity.” The steel example I just gave is a really good one, and it turns out the most efficient steel plants today use three or four tons of water to make a ton of steel. We're doing what we want with less water.

That’s true in the domestic arena, too. It used to take six gallons of water to flush a toilet. That's when we didn't care because there was plentiful water. In the 1980s and 90s, California passed a number of regulations about appliance efficiency for showerheads, washing machines, and toilets that required three and a half gallon-per-flush toilets. In the early 1990s, President George H.W. Bush signed the 1992 Energy Act into law that included water-efficiency regulations requiring that all toilets sold in the United States use only 1.6 gallons per flush. In fact, there are now toilets that use 1.2 gallons per flush or 0.8 gallons per flush. We have low-flow showerheads. We have washing machines that are much more efficient at doing what we want. This is not deprivation, it’s efficiency, it’s productivity.

California agriculture is another good example of this. California farmers today grow much more food and make much more money per unit of water used than they did 20 years ago. In some crops they have moved from flood irrigation to sprinklers, and then from sprinklers to drip. They’re monitoring soil moisture so that instead of watering on a rotation whether the crop needs it or not they are being more careful about when water is applied. All of these changes help explain the drop in levels of water consumption.

Where will the gains come from in the 21st century? Probably not from the 20th century’s hard path of infrastructure solutions tapping into new supply, because of peak water constraints and concerns about ecological devastation. I think we need a new approach. There are still parts of the world where we need new infrastructure, new dams, new reservoirs, new aqueducts, although they must be built to a different standard than they were built when we didn't care -- or didn't know we ought to care -- about the environment or downstream communities or international politics. But we need more than new supply.

We are nowhere near the limits of our ability to improve water-use efficiency. We've made enormous progress at reducing our water use through more efficient appliances and fixtures. Many of our industries are more efficient than they used to be. Many of our farmers are more efficient than they used to be. But there are still enormous untapped efficiency improvements out there that could help us push this curve down even farther and these options are cheaper than new supply. It's cheaper than trying to find a way to tap farther and more distant watersheds for water or drill another well into an over-tapped aquifer. Here's another example in California's Central Valley, where, as I said, we've made enormous progress in moving from flood irrigation sprinklers and sprinklers to drip. Yet 15 or 20 percent of our orchards and vineyards remain on flood irrigation. They ought to be on drip. Eighty percent of them are on drip. That's a tremendous improvement but there are still untapped efficiency improvements in the Central Valley.

The soft path for water also means building smarter institutions. I have an engineering degree. I was trained how to construct a dam and how to think about hydrology and extreme droughts and floods and how to size a dam for these extreme events. I wasn't trained how to replace toilets for a hundred thousand people -- but you get more water today replacing toilets then you do building new dams in California. This is an institutional and educational challenge we are still coming to grips with.

We need smart national water policies as well. The role of the federal government in solving our water problems is rightly limited. Water is usually a local issue or a watershed issue, or a state issue. But there are also federal and international roles and responsibilities. Getting the management side of things right is important, and that includes addressing our water problems at the right level.

The responsibility to develop and implement responsible national water policy is isn't being adequately fulfilled at the moment by the appropriate federal agencies. Part of the problem is confusion over authority. Part of the problem is the failure of the executive branch and the legislative branch of congress to appropriate resources to deal with water issues and to manage our water issues. Part of the problem is old legislation that hasn't been updated for the 21st century. Part of the problem is that there are gaps in our scientific understanding. So let me offer eight recommendations:

1. Federal water-related agencies and programs are fragmented and they require better coordination. Let me preface this by saying some colleagues of mine and I wrote a book a couple years ago called A 21st century US Water Policy (Oxford University Press, New York) and in it we note that that there are twenty-odd federal agencies that have some responsibility for water. For example, the US Geological Survey is responsible for monitoring and measuring groundwater and water use. USDA is responsible for the agricultural side of things. The Bureau of Reclamation manages water through federal infrastructure in the west. The Army Corps of Engineers builds and manages infrastructure in the Tennessee Valley area and other parts of the US. The National Oceanic and Atmospheric Administration conducts atmospheric modeling and weather forecasting and climate modeling. The challenges of the 21st century require a better integration of those responsibilities and that is really hard to do because nobody likes to give up authority.

2. The nation must develop a better understanding of the science of how much water we have and where it is and how much water we use. Every state evaluates water use differently so there are no standard measurements for how much water is used. As a result, our estimates of water use are rarely complete or accurate. This is a federal responsibility. We ought to have really good estimates of how much water we have, where it is, and who's using it to do what. There is a USGS project underway called the National Census on US water. The whole thing costs a mere ten million dollars yet I'm not even sure Congress has fully appropriated the money for it.

3. More appropriate economic strategies can create more sustainable water use patterns. Water pricing is often thought about as a local concern or a state concern. Your water bills and your water rates ought to be set by your local water agency and not by the federal government. But, the largest wholesaler of water in the West is the federal government through the US Bureau of Reclamation through a system that was built with our taxpayer dollars, including the Central Valley Project, the big dams on the Colorado River, and the big dams on the Columbia River. The rates charged for those systems are inadequate for repaying the nation for their construction. Those are subsidies. There was a time when we thought that was the right thing to do, but forty years ago the last National Water Commission recommended discontinuation of subsidies for new irrigation projects and said “direct beneficiaries of federal irrigation developments should pay in full the costs of new projects allocated to irrigation.” Now, another four decades later, that key recommendation remains unfulfilled. The US should reform water-pricing policies that subsidize inefficient use of water and continue to cost taxpayers money.

A quick example in California: we have the Central Valley Project and we have the State Water Project. Water at the State Project is more expensive to users than water at the federal projects because of federal subsidies. Farmers who get water from the state projects are more efficient than farmers who get water from the federal projects. That’s not a coincidence.

4. Water policies and infrastructure should be designed to evolve with changing climatic conditions. Climate change is real. It's happening because of human emissions of greenhouse gases. Scientists are as sure of that as we are as that smoking tobacco can cause cancer. What's going to happen precisely with rainfall in California? We don't know. We're learning. But climate change’s impacts on water resources will be significant, and we had better figure out how to manage the federal piece of this puzzle.

5. Existing federal water laws should be updated and adequately enforced. I talked earlier about the Clean Water Act and the Safe Drinking Water Act. They need to be reformed and they need to be brought up to date. There are pollutants that we know about now that we didn't know about then. We've addressed point-source pollution from big industries but have failed to tackle nonpoint source pollution such as runoff from agricultural fields. When we update these federal laws we ought to provide the resources to enforce them. The disaster a couple weeks ago in West Virginia with the drinking water supply was a failure of regulation at the state, federal, and local level, was a failure of not having certain kinds of regulations in place, and was a failure of inadequately enforcing regulations that we do have.

6. Water management in the 21st century has to encompass decentralized solutions such as demand management in efficiency, storm water capture and reuse, recycled water, grey water, and other non-traditional approaches. I said we're running into peak water limits on new supply but there are many interesting new supply options that aren’t traditional. In California, we collect a lot of wastewater, we treat it to a high standard, and we dump it in the ocean. We're starting to reuse more of that for groundwater recharge, for industrial cooling, for landscape irrigation and the like. I made a recommendation earlier this week to the State Water Resources Control Board that within five years we ought to have a policy that prohibits dumping highly treated wastewater into the ocean. It should be considered an asset, not a liability.

7. Federal water policies must be integrated with other policies including energy, agriculture, and climate change. Too often we’ve thought about energy separately from water and yet things we do in energy policy affect water. Things we do on the water side affect energy, so we need to integrate those policies. The ethanol policy that we implemented in the US because we want to boost domestic production of liquid fuels took a tremendous amount of our corn crop and it had implications that we did not understand for global food prices and for water as well.

8. Incorporate environmental justice principles into federal water policy and local communities. Many federal agencies including EPA and the Department of the Interior already have the statutory ability to address equity issues in their policies. Concerns about unequal impacts on low-income communities of environmental challenges, problems in the Central Valley with nitrate pollutions, and the fact that there are populations in California that don't have access to safe drinking water because of nitrate pollution is unbelievable in the 21st century. We have to expand our efforts to integrate these kinds of policies into our water strategies.

I really believe there are solutions to our water problems: there are new supply options; there are ways of integrating the soft path for water; there are ways of dealing with the issue of peak water. I think we're moving in the right direction, though too slowly in many cases and places. I think inevitably we're going to move to a sustainable use of water -- the problem is how fast we can get there.

© Peter Gleick, 2014