The proposal by a subsidiary of Progressive Rail, a small, short-line railroad headquartered just south of the Twin Cities, would pump some 500 million gallons of water annually from wells for rail shipment to southwestern states.
The idea was strangled by state regulators even before a permit application was submitted, so it really isn’t a public issue. However, the idea itself gives us a good teaching moment in natural resource economics. And it is good for all of us to recognize some of the incongruities, if not outright hypocrisy, in our public attitudes about natural resource use.
Start with terms. The difference between “renewable” and “nonrenewable” resources is important in economics. These are commonly used terms and largely self-explanatory. Iron ore and coal are nonrenewable. Trees, solar power and surface waters all are.
One can quibble around the length of time considered for renewing. In geologic time, the volcano under Yellowstone National Park will erupt again. When it does, metal ores will be deposited in layers or veins. Yet in human time, this is negligible. Ores are nonrenewable. How about giant redwoods or sequoias? These trees grow, but so very slowly that remaining ones effectively are a nonrenewable resource.
Then there is the distinction between “stock” and “flow” resources. This sounds like renewable versus not-renewable, but does differ. If a stock resource is not used, it remains for use in a future period. If a flow resource is not used, it is gone forever. So copper ore is nonrenewable and a stock resource; it stays put until mined. Wind energy and hydropower are renewable and flows. If not used, that potential energy is gone forever. Trees are renewable, but a stock. If you don’t cut them down, they will be there tomorrow. Yes, some will die and rot, but for practical purposes they are a stock. Grasses in pastures are renewable and a flow. Unlike trees, if not used this season, what is there in September will not be usable next spring.
Understand that “unused” resources do not produce zero benefits for society. Trees not harvested, including those that die, benefit the whole ecosystem, often including other living trees of the same species. Standing prairie grass over winter benefits wildlife and diverse plant species that thrive better if given some protection during winter. Water cascading over falls and rapids benefits us. If not, we would run the entire Niagara River through hydroelectric turbines 24 hours a day.
Also understand that putting renewable resources to “beneficial use” can have external costs. Clear-cutting forests can cause flooding and sedimentation downstream and decimate wildlife. Dams on the Columbia and its tributaries wiped out an enormous salmon industry. Wind turbines kill some birds, and their noise disturbs people living nearby. Streams, wind and solar power all are renewable flows, but one must calculate the net possible benefits, subtracting off external costs, if one wants to tabulate the value of the “lost” resource.
Water has specific terms of its own. There are surface waters -- rivers, lakes, ponds and wetlands. Then there is groundwater in subterranean gravel and rock aquifers.
But the distinction is not so neat. Some aquifers are “recharged,” and some are not. There are aquifers, often large ones, that were filled with water in geological processes thousands of years ago and that now get no recharge from surface waters or rain. This “fossil water” is a nonrenewable stock resource like iron ore. Pump from one and levels will fall, never to rise again.
Other aquifers get recharged, but at a limited rate. Most of the California aquifers overpumped in recent droughts do get some recharge, mostly off the Sierra Nevadas. But it is a slow process. That it is easier to pump faster than the recharge rate causes precipitous declines in water tables in short periods of time.
You have free articles remaining.
Such aquifers could be sustainable steady-state resources over time if recharge equaled pumping. But some are sand that is stable only when saturated with water. If we pump levels down so far that this layer is dewatered, the ground will subside. The spaces between grains of sand will be reduced in size. No matter what future potential recharge might be, the aquifer will never be able to hold as much water as before. Damage is permanent. This has happened to significant extents in many California aquifers in the last decade.
It is possible that a body of surface water be “hydraulically connected” to an aquifer. This is extremely common in west-central Minnesota, starting with the sand plains on the northwest edge of the Twin Cities. For these, the top of the sand or gravel deposit that holds the groundwater runs all the way to the surface of the land. The lake or wetland is the top of the aquifer. In other cases, the lake or stream is separated from the aquifer itself, but by some permeable material that allows water to flow from surface to aquifer, perhaps rapidly, perhaps slowly.
Withdrawals of surface or ground water can be separated into “consumptive” and “nonconsumptive” uses. Water that is pumped from a river to run through the condensers of a steam electric plant and that is returned, albeit at a higher temperature, is a nonconsumptive use. Water withdrawn to run through a cooling tower and that evaporates is consumptive.
Older ethanol plants pump large quantities of well water, of which much is discharged as steam. Is this consumptive? It enters the atmosphere and blows downwind to other states. In the hydrologic cycle, it will be someone else’s rain.
At some point definitions need to be arbitrary. Our state’s Department of Natural Resources says “water withdrawn that is not directly returned to its original source” is consumptive use. Since the state health department prohibits most injections of water back into aquifers, virtually all pumping from wells, including those for irrigation, is deemed consumptive.
There is a final consideration, legal and political rather than economic. In general, and for varied reasons, there often are legal limits on uses of water that would result in a transfer of water between two different river basins. North Dakotans wanted to drain rising water in Devils Lake, but the easiest way to do so would have moved water from the Missouri River basin to one draining northward to the Arctic. Not allowed.
There is a Great Lakes Compact between all the states and Canadian provinces with drainage into these lakes. It forbids any transfer of water out of this enormous area. It was used to stop a proposed bottled water facility in Wisconsin, only a few miles into the covered watershed, that would have shipped product to the rest of the nation. If it had been a soda or beer plant of equal size, it would not have raised an eyebrow.
Debate on water use is plagued by the fact that different units and rates are thrown about -- gallons per minute, cubic feet per second, or acre feet. The average layperson struggles to put these in context. In the water-shipping case, the railroad company proposed pumping 500 million gallons per year. That sounds like a lot of water and it is. Yet it also is less than what some individual skyscrapers in Minneapolis pull out of deep sandstone aquifers under the city to run through air-conditioning condensers one time and then flush into sewers.
If converted into the “acre-feet” units used in crop irrigation, it would put 10 inches of water on 3 square miles of cropland, about what is needed on sandy soils in the Upper Midwest. And 500 million gallons would approximate the amount used per year in larger ethanol plants or by two midsize ones. No one would bat an eye if six farmers sought permits to irrigate 500 acres each. And if someone decided to build an ethanol plant instead of pump water into tank cars, local economic development authorities would celebrate their good fortune.
The proposed railroad water project would have pumped from the deepest aquifer currently used in that county. This aquifer does receive recharge and is not yet overtaxed, but involves more concerns than pumping out of a surficial gravel aquifer between the Twin Cities and Fargo. Once drilled, operating the wells would provide miniscule employment, mostly for shuttling railcars around. And it would have few environmental costs outside of the actual water used.
The whole business model was unclear. Why rail water from Minnesota when the much closer Missouri River now flows more than is convenient? The quantity is not enough to irrigate much land, and the cost would be prohibitive. And even if per capita water use was half the national average, it would only be enough for residential use in a town the size of Williston.
In any case, the project is a dead duck. But we can ruminate on these issues until the next time.
St. Paul economist and writer Edward Lotterman can be reached at firstname.lastname@example.org.