A proposal to inject wastewater from Consumers Energy’s J.H. Campbell coal plant deep underground raises a larger question: What does each source of electricity leave behind?

A recent Sierra Club announcement about an EPA public hearing sent me down another environmental rabbit hole. Consumers Energy is seeking federal permits for two deep injection wells at its J.H. Campbell Generating Station in West Olive. The company wants to dispose of wastewater from the coal plant and its eventual cleanup by injecting it into porous rock formations thousands of feet underground.

The Sierra Club encouraged people to attend the July 23 hearing in Holland after nearly 200 people submitted comments raising questions about the proposal. The announcement also linked to an MLive article examining the plan. By the time this blog is published, the hearing will be over, but the EPA will still be accepting written comments through Sunday, July 26, 2026.

More on that at the end. First, it is worth asking a broader question: How does coal waste compare with what other forms of electricity generation leave behind?

We Are Not Making the Waste Disappear

The EPA describes the proposed Campbell wells as Class I non-hazardous injection wells. That sounds reassuring, but “non-hazardous” is a regulatory classification. It does not mean the wastewater is harmless, the geology is perfectly understood or the system is free of long-term risk.

Deep injection wells are designed to place waste below aquifers used for drinking water. Steel and cement casings are intended to prevent it from escaping into groundwater. The basic idea is simple: drill deeply enough, isolate the material and keep it away from people. The problem is that the system must continue working long after the people who designed, approved and operated it are gone. Casings can deteriorate under pressure, chemical exposure and changing temperatures. Injection pressure can affect underground formations and, in some settings, contribute to seismic activity. Geology can be studied extensively, but never known with absolute certainty.

The permanence of the disposal may be its greatest weakness. Once wastewater has been injected thousands of feet underground, recovering it could be nearly impossible if future generations discover a problem. We call that disposal, but the material has not disappeared. We have simply put it somewhere and accepted the responsibility of hoping it stays there.

Campbell was scheduled to close on May 31, 2025, but federal emergency orders kept it operating. Whatever happens with the plant now, decades of accumulated waste remain. That is the long shadow of coal. The electricity may be gone in an instant, but its wastes can become someone else’s responsibility for generations.

Coal: The Waste Never Stops

Coal produces waste before it is burned, while it is burned and long after the power plant closes. Mining creates waste rock, contaminated drainage and damage to landscapes and waterways. Transportation requires trains, ports and other infrastructure. Combustion produces carbon dioxide, sulfur dioxide, nitrogen oxides, particulate pollution, mercury, ash and contaminated wastewater.

Coal combustion products remain one of the largest industrial waste streams in the United States. According to the latest national industry survey, power plants produced 63.6 million tons in 2024. About 46 million tons, or 72 percent, were reused in products such as concrete and wallboard, but nearly 18 million tons still required disposal or another form of management. It can contain mercury, cadmium and arsenic, all of which can threaten groundwater, drinking water and rivers when improperly managed. Some ash is reused in products such as concrete and wallboard, but enormous quantities still end up in landfills, ponds or other disposal facilities.

A large portion of coal’s waste is less visible because it goes directly into the atmosphere. Every day a coal plant operates, it creates additional carbon pollution, air pollution, ash and wastewater. Closing the plant stops the production of new waste, but it does not eliminate what has already accumulated.

The Campbell injection wells are part of that long cleanup bill. Coal generated electricity for Michigan for generations. We are now deciding where to put what it left behind.

Natural Gas: Less Ash, More Invisible Waste

Natural gas is cleaner than coal in several important ways. It produces almost no ash and releases much less carbon dioxide for each unit of electricity generated. In 2023, coal-fired power plants emitted about 2.31 pounds of carbon dioxide per kilowatt-hour, compared with about 0.96 pounds from natural gas plants. That is a meaningful improvement, but it is not zero.

Natural gas is primarily methane, a greenhouse gas far more powerful than carbon dioxide over shorter periods. Methane can escape during drilling, processing, transportation, storage and distribution. Gas plants also release nitrogen oxides and other combustion pollutants.

Hydraulic fracturing creates another waste stream. Large quantities of wastewater must be treated, reused or disposed of, sometimes through the same basic method now proposed at Campbell: underground injection. Drilling also disturbs land and requires pipelines, compressor stations and other infrastructure.

Natural gas therefore has a much smaller visible waste problem than coal. There are no mountains of natural gas ash piled beside a power plant. Much of the burden is simply harder to see because it enters the atmosphere or is created far from the plant where the gas is extracted.

Natural gas may be better than coal, but “better than coal” is not an especially high standard.

Nuclear: Very Little Waste, but Highly Toxic and an Enormous Responsibility

Nuclear power presents almost the opposite problem. Its physical waste stream is remarkably small. The Department of Energy estimates that all the commercial spent nuclear fuel produced in the United States since the 1950s could fit on a single football field stacked less than 10 yards high.

That is an astonishingly small amount of material compared with the mountains of coal burned to produce a similar amount of electricity. The problem, of course, is what is inside that small footprint. Spent nuclear fuel remains intensely radioactive and must be secured, monitored and isolated for extremely long periods. Most American spent fuel is stored at reactor sites, first in pools and later in thick steel and concrete dry casks.

Yet spent nuclear fuel is not entirely spent. Current American reactors extract less than 5 percent of uranium’s potential energy before the fuel is removed. Most of what remains is uranium, along with smaller quantities of plutonium and other radioactive elements. Some of that material can be separated and turned into new fuel.

France has commercially reprocessed spent nuclear fuel for decades. Uranium and plutonium are recovered, with some of the plutonium used to manufacture mixed-oxide, or MOX, fuel. About 10 percent of France’s nuclear electricity is generated using MOX. Russia has gone further with its BN-800 fast reactor, which began operating with a full core of recycled uranium-plutonium MOX fuel in 2023.

That makes the basic idea credible. It does not make it simple.

Most commercial recycling provides only one additional trip through a reactor. The MOX fuel eventually becomes spent fuel again and still requires disposal. Reprocessing is expensive and creates additional radioactive liquid and solid waste. Separating and transporting concentrated nuclear materials also creates security and weapons-proliferation concerns.

Advanced fast reactors could potentially use a much greater portion of the remaining energy. In a closed fuel cycle, uranium, plutonium and possibly other long-lived actinides would be recovered, fabricated into new fuel and repeatedly returned to reactors. Researchers are also studying transmutation, which could convert some long-lived elements into shorter-lived radioactive products while generating additional energy.

The science is sound, and parts of it have been demonstrated in laboratories, pilot facilities and experimental reactors. The commercial system remains another matter. Repeated recycling would require specialized processing plants, remote fuel-fabrication facilities, advanced reactors, secure transportation and extensive international safeguards. No country has yet established an economically reliable commercial breeder-reactor system operating with a fully closed fuel cycle.

Recycling would also not eliminate the need for a permanent repository. Nuclear fission creates radioactive products that cannot simply be reused forever. Some advanced reactors may produce unfamiliar wastes, including irradiated graphite, sodium-contaminated materials and used molten salts. The quantity and composition of nuclear waste could change, but some material would still require permanent isolation.

The balanced conclusion is that nuclear waste remains uniquely hazardous and long-lived, but part of it may also be valuable fuel. Coal ash cannot be placed back into a coal plant to generate decades of additional electricity. Spent nuclear fuel may eventually be able to do something close to that, although the United States does not yet have an economical industrial system capable of doing it safely at scale.

Hydropower: Low-Carbon Energy and Accumulating Consequences

Hydroelectric dams produce no ash, spent fuel or combustion pollution while operating, but they are not built without a carbon cost. Earthfill is actually the most common dam type, although many hydroelectric projects also use large amounts of concrete in their powerhouses, spillways, foundations and control structures.

Concrete is mostly sand, gravel and water. Cement, particularly the clinker used as its binding ingredient, is the carbon-intensive part. Producing cement requires heating limestone to extremely high temperatures, releasing carbon from both the fuel used and the limestone itself. Cement production accounts for roughly 7 to 8 percent of global carbon dioxide emissions.

A large dam can therefore begin with a substantial amount of embodied carbon. It can also generate electricity for a century or more, spreading those initial emissions across generations of power production. Lifecycle studies generally place conventional hydropower alongside other low-carbon sources and far below fossil fuels.

There is also an important difference between constructing a new dam and maintaining one that already exists. The carbon used to build Michigan’s century-old dams was released generations ago. Continuing to use them may preserve the value of that original investment without requiring the materials needed to build an entirely new source of power. That does not automatically mean every dam should remain.

Hydropower’s greatest environmental consequences are generally found in the river rather than in a waste container. Dams flood land, block fish passage, change water temperatures and interrupt the natural movement of sediment. Rivers constantly carry sand, gravel, clay and organic material downstream. When a dam blocks that movement, the material settles into the reservoir.

Earlier dam designs often assumed a predictable sedimentation rate and a useful reservoir life of perhaps a century. Those estimates did not always account accurately for changing land use, erosion, floods, wildfire and watershed conditions. At some reservoirs, sediment accumulated faster or caused operational problems sooner than expected. As it builds, it reduces water-storage capacity and can interfere with intakes, outlets, recreation and power generation. If the sediment has absorbed decades of industrial contamination, removing or disturbing it can create another expensive environmental problem.

Drought can shorten a dam’s useful life even when the structure remains sound. Hydropower depends on adequate and reasonably predictable water flow. Longer droughts, shifting precipitation and increased evaporation can reduce reservoir levels and leave less water available to turn turbines. A solar panel can continue producing electricity during a drought. A wind turbine can continue turning. A hydroelectric plant depends on the future river behaving enough like the river its engineers studied decades earlier.

Yet dams can also create unexpected ecological relationships. Michigan’s Net River Dam is a wildlife-management dam rather than a hydroelectric facility, but it illustrates the dilemma. Built in the early 1960s, it created a wetland impoundment that now supports wildlife habitat and wild rice beds valued by the Keweenaw Bay Indian Community. When part of the dam failed in 2022, the state chose to rebuild it and restore the water levels supporting that ecosystem.

Tippy Dam on the Manistee River presents an even stranger case. Thousands of bats hibernate inside its spillway. The colony has survived white-nose syndrome far better than many populations using traditional caves and mines, leading researchers to study whether conditions inside the dam could offer clues for protecting bats elsewhere.

Conservation is dam complicated! Removing a structure can restore a free-flowing river, reconnect fish habitat and allow sediment to move downstream. It can also eliminate wetlands, recreational lakes and wildlife habitats that developed after the dam was built.

That debate now surrounds Consumers Energy’s proposed sale of 13 Michigan hydroelectric dams on the Au Sable, Grand, Kalamazoo, Manistee and Muskegon rivers. Consumers has proposed selling the dams for $1 each to Confluence Hydro, an affiliate of the private equity firm Hull Street Energy, then buying their electricity through a 30-year agreement.

Consumers argues that the sale is the lowest-cost option for customers and would preserve renewable electricity, reservoirs, recreation and local economies. Opponents question whether the proposed buyer has the resources and long-term commitment needed to maintain century-old dams and thousands of acres of surrounding land. Plans to place each dam in a separate limited liability company have added concerns about who would ultimately pay if one of those companies could not afford a major repair or failed financially.

River advocates raise a different concern. A 30-year agreement could commit Michigan to preserving all 13 dams before the state has completed a dam-by-dam evaluation of safety, energy production, river ecology, public access, recreation and long-term costs. In June 2026, an administrative law judge recommended that the Michigan Public Service Commission reject the proposed sale, concluding that Consumers had not demonstrated that the transaction was in the public interest or that Confluence could reliably meet the dams’ full lifecycle needs. That recommendation was not a final decision.

Consumers and Confluence later proposed additional protections, including a Hydro Safety Fund of up to $270 million from Consumers shareholders, continued community engagement, retention of the existing hydro workforce and commitments not to reduce associated lands during relicensing unless doing so serves the public interest.

Those changes address some concerns, but not the larger question. These dams are not simply power plants. They shape rivers, communities, property values, recreation, wildlife habitat and public lands. Michigan should not decide their future only as part of a single business transaction.

Hydropower’s waste problem is therefore less about material thrown away every day and more about consequences accumulating over time. Sediment builds, infrastructure deteriorates, climate conditions change and communities and wildlife adapt. Eventually, someone must decide whether to repair, modify, rebuild or remove the structure, and someone must pay for that decision.

Solar: A Waste Problem We Still Have Time to Solve

Solar panels generate electricity without producing ash, wastewater or combustion pollution during operation. Their principal waste challenge comes when they are damaged, replaced or reach the end of their useful lives, generally after several decades.

About 95 percent of the material in a conventional panel can be recycled using existing technology and roughly 99 percent is considered non-hazardous. The problem is economics. Recycling in the United States can cost approximately $15 to $45 per panel, compared with roughly $1 to $5 for landfilling, before transportation is included. Recycling companies also face a chicken-and-egg problem: there are not yet enough retired panels in most regions to support large, efficient processing facilities, but without those facilities, collection and transportation remain expensive.

This is where economic incentives and smart policy can make a difference.

The European Union uses extended producer responsibility, requiring manufacturers and importers to help finance collection and recycling for the products they sell. For photovoltaic panels placed on the European market after August 13, 2012, the producer generally bears responsibility for end-of-life management.

France has taken that model further. A government-approved nonprofit organization called Soren collects an “eco-participation” fee when a panel is sold and uses the money to finance future collection, sorting and recycling. France also adjusts the fee according to the panel’s weight, technology and environmental design. Beginning in 2025, panels meeting stronger environmental criteria became eligible for lower fees. That approach not only pays for recycling but rewards manufacturers for making panels easier to repair, disassemble and reuse.

Japan expects discarded solar panels to reach approximately 500,000 tons annually during the latter part of the 2030s and is studying a system that could include mandatory recycling. Australia announced a $24.7 million national recycling pilot in January 2026 that aims to collect as many as 250,000 panels from roughly 100 sites while measuring collection, transportation and processing costs. The results are intended to guide a permanent national product-stewardship program.

The United States could take similar steps. A modest recycling fee could be collected when panels are sold. Manufacturers and importers could be required to participate in approved collection programs. Utility-scale solar developers could include recycling costs in their decommissioning plans and financial assurances. Fees could reward panels designed for easier repair and material recovery. Public grants could help regional facilities grow until the volume of retired panels becomes large enough to support the industry without subsidies. Landfill charges or disposal restrictions could also stop landfilling from remaining artificially cheaper than recovering glass, aluminum, copper, silicon and silver.

The point is not to pretend solar has no waste problem. It is to build the system before millions of panels reach the end of their lives. A solar panel produces electricity for decades and becomes waste once. Coal creates new pollution and waste every day it burns. With the right policies, much of yesterday’s solar panel can become raw material for tomorrow’s energy system.

Wind: Mostly Valuable Materials Waiting to Be Reused

Wind turbines also generate electricity without producing combustion waste. Approximately 85 to 90 percent of a turbine’s mass can already be commercially recycled. Steel, copper, aluminum and iron all have established markets.

The difficult component is the blade. Wind-turbine blades are generally made from fiberglass or carbon-fiber composites designed to be light, strong and capable of surviving decades of stress and weather. Those qualities also make the materials difficult and expensive to separate. New recycling technologies and more recyclable blade designs are being developed, while some retired blades are repurposed or processed for use in cement and other products.

Wind projects need clear decommissioning plans, financial responsibility and a stronger recycling network. Their waste is still fundamentally different from coal’s. A turbine does not create a new pile of waste every time its blades turn. Most of its material remains valuable at the end of its operating life.

Coal Is Different

Every source of electricity has consequences. Natural gas produces less carbon pollution and almost no ash, but still contributes to climate change, methane leakage and drilling wastewater. Nuclear power leaves a tiny but intensely radioactive waste stream, although some of that material may eventually be reused as fuel. Hydropower generates low-carbon electricity but reshapes rivers and creates infrastructure decisions that can last for generations. Solar panels and wind turbines eventually wear out, but most of their materials can be recovered if we build the systems and incentives to do it.

Coal manages to combine nearly all the worst characteristics. It produces enormous quantities of solid waste, contaminated water, continuous air pollution and carbon emissions. Its environmental damage begins at the mine and continues through transportation, combustion, disposal and cleanup.

That is what the Campbell injection-well proposal is really about. The electricity has already been generated and the bills have already been paid. Now we are deciding where to place what is left behind and how much risk future communities should be expected to accept.

The Hearing Is Over. The Decision Is Not.

The EPA held its public hearing on the two proposed Campbell injection-well permits on July 23, but written comments are being accepted through Sunday, July 26, 2026.

You do not have to be a geologist, engineer or groundwater expert to submit a meaningful comment. You can ask what chemicals will be injected, how the wells will be monitored, how long that monitoring will continue and who will pay for it. You can ask what happens if a casing fails decades from now, whether the waste could migrate through underground faults and how Consumers or the EPA would respond if contamination were detected.

You can also ask the most basic question of all: Why should future generations be expected to manage another permanent waste site created by burning coal?

Submit your written comment through the EPA public-notice page before July 26.

The EPA identifies the proceeding as Docket No. EPA-R05-OW-2025-2730. People who already submitted comments do not need to submit them again.

The hearing may be over, but the public still has a chance to be heard.