Jack Caldwell - Mining Engineer - Robertson GeoConsultants

Here’s what I wrote in 1991 about criteria for covers and closure of waste facilities. This is from my book Principles and Practice of Waste Encapsulation. With Charles Reith I wrote this book on the basis of what we did to construct the encapsulation facility at the Weldon Spring site near St Louis, Missouri. What we wrote was put into practice and as far as I know the site is still stable and a tourist attraction.

In this book, we use five periods, defined as follows, when thinking about and talking about the longevity and design life and performance period of encapsulation systems:

  • The Construction Period: This is the period during which the cell is constructed and the wastes are placed.
  • The Short Term: This is the period during which we can rely on active oversight, maintenance, and repair. I belive that this period is no longer than 50 years.
  • The Medium Term: This is the period for which it seems reasonable to anticipate some institutional or societal control over the facility. I suppose this period is no longer than 100 to 200 years.
  • The Long Term: This is the period to which we should turn our attention and think how the encapsulation system will perform without human intervention, as it responds quietly to the forces of nature. Surely this period extends for 200 to 1,000 years.
  • The Very Long Term: This is the period from 1,000 to 10,000 years for which humans have built dwellings, cultivated the land, and had diverse languages. In considering this degree of longevity, we must contemplate the possibilities of climate change, loss of institutional control, and intrusion by humans who have adopted languages and cultural perspectives far different from those of today.

We need to distinguish between: (a) the design life of an encapsulation system; (b) the performance period of a cell; and (c) and the recurrence interval of a design event. Confusing the three leads to misconceptions and generally muddled discussions.

The design life may be defined as the period for which the disposal cell will encapsulate the wastes with a desired degree of certainty. The design life may be conceived of as a reasonable estimate of the median period for which the majority of the components that make up the encapsulation system will function as intended. The component performance period need not be the same as the encapsulation system design life.

The performance period of a components may be very short or very long, and it need not have a one-to-one correlation with the cell’s design life. For example, a geomembrane in a cell may have a performance period of only fifty years. The design life of the cell may still be more than 1,000 years if the design provides for a clay layer to control seepage when the geomembrane fails.

Neither the design life nor the performance period should be confused with a particular design event recurrence interval. For example, the recurrence interval of a site design earthquake in not the same as the design life of the disposal cell. The design earthquake may have a recurrence interval on one million years; that does not mean that the design life of the cell is one million years, or that the sideslopes will perform satisfactorily for so long a period. All that it means is that there is a one in a million chance the earthquake will occur in any given year. In brief, there is a one in a million chance the sideslopes will be subjected to the forces imposed by the design earthquake each and every year of the 1,000 or more years of the design life.

The design life for disposal facilities may range from as short as thirty years to as long as 10,000 years. The required design life may be established by law, regulation, technical desirability, or custom. Design lives as short as thirty years have been specified for the cover of municipal landfills. Design lives of up to 500 years are required for low-level radioactive waste disposal facilities. Design lives of up to 1,000 years, to the extent reasonably achievable, are mandated for the covers of uranium mill tailings piles. A design life of 10,000 years is applied to deep-geologic disposal facilities for certain high-level or transuranic radioactive wastes. Designers of near-surface disposal facilities for transuranic contaminated soil have adopted the standard of 10,000 years as a general design objective.

In some European countries where the works of man constructed 2,000 and more years ago still stand, there is considerable sentiment to adopt even longer design lives. Desirable as this may be, it is simply not technically possible or feasible to make accurate or substantive conclusions about the design life of a disposal cell cover. Precedent and technical experience to make provable claims about the life of alternative covers does not exist. Natural analogues provide ample evidence for optimism and confidence that engineered covers will endure for long period, but there are no direct, long-lasting examples or replicates of the cover and liners that are proposed and now being built to encapsulate wastes. Therefore, conservative designs should be provided which are likely to resist both extremes and cumulative destructive events throughout their design lives.

The following design life is a reasonable objective for most waste encapsulation facilities: “1,000 years to the extent reasonable achievable, and at any rate for 200 years.” This design life is consistent with the philosophy that very little, if any, maintenance should be needed to achieve such a design life.

A design life of 1,000 years to the extent reasonably achievable, means that within the bounds of reasonableness and technical practicability, the great majority of the disposal cell components will function as intended for a median period of 1,000 years, In practice most of the cell components will function satisfactorily for considerably longer than 1,000 years. For example none of the construction materials, the sand and clays, will “deteriorate” in that period. The rocks will still be rock and the gravels will still be gravel. The slopes will not fail and unacceptable deformation will not occur. Run-off will still occur as planned and vegetation will continue to grow.

Some of the disposal cell components may change in 1,000 years. Hence, we may talk of the performance of individual components or elements of the disposal cell. Foundation soils may consolidate, the cell components may rust, and the geomembrane at the base of the cell will disintegrate. For that reason, there is a second part to the design life requirement: to the extent that 1,000 years of assured performance is not reasonable achievable or demonstrable, the components (construction materials) should last, and the cell should perform satisfactorily, at any rate, for 200 years. Design details should be chosen so that the disruptive forces reasonably foreseeable in a 200-year period will not disrupt the integrity of the cell even though certain components may not perform for a period equal to the longest design life. For example, if the precipitation run-off from the design storm, which has a reasonable likelihood of occurring in the first 200 years of operation of the cell, may just initiate erosion in the soils and gravel of the top part of the cover, then the planned maintenance of the cover should provide for repair.

An indefinite design life is theoretically possible if the facility operator undertakes an active maintenance philosophy and can establish with assured credibility that it will continue active maintenance indefinitely.

The U.S. Nuclear Regulatory Agency considers 100 years the maximum period that a promise of active institutional control may be the basis for complying with the U.S. EPA regulations governing waste disposal. It is likely that society, in general, will accept a promise of active maintenance to be reasonable for no more than 100 years. In practice, 100 years may be the maximum period to which a philosophy of indefinite active maintenance may apply.

Now that is what I wrote nearly twenty-five years ago - so much for a thirty-year design life. Last year I wrote the following. Nothing has changed:

I was in yet another of those arguments about what constitutes “perpetual” in mining and mine closure. I had heard all the arguments, smart and cynical, more than thirty years ago when we debated them on the UMTRA Project. But the arguments continue, for everyone has an opinion and wants to be heard.

“The BC and Canadian governments will always be here,” is the first profound statement. “They will always be around to provide institutional control of the closed mine. You need not examine how the closed mine will perform in the absence of responsible surveillance and maintenance.” There is always a cock-eyed optimist in the crowd!

“Unless, of course, they have mined out the oil sands, and everybody is starving and they have cut all the timber and have decaying houses. Then they will not give a damn about your closed mine that relies on the government to keep it in shape and prevent its failure.” There is always a cynic in the crowd!

What does “perpetual” mean in the context of mine closure?

Maybe it means what the dictionary says it means, namely forever.

But how long is forever? The lawyer instinct is strong in us all regardless of how much we deprecate lawyers.

Maybe forever is as long as the universe has existed. Say 14 to 15 billion years. Or maybe it is only as long as there has been mining. They were mining at Bomvu Ridge in Swaziland 43,000 years ago.

Or maybe it is as long as the Holocene; that geological age that begun about ten thousand years ago with the retreat of the last glaciers.

“No. It is only as long as the Bible says the earth has existed. God created the earth only about 6,800 years ago. So why try to out-time God?”

My Spanish friends say perpetual is 2,000 years. That was when they begun mining in Spain to provide the needs of the Roman Empire.

Not so, say the Greeks. “Athens flourished long before that on the output of the Laurentian silver mines. They never closed them. Just build a temple to some goddess and nobody will care to go see the damage wrought by 2,800-year old mines.”

Today I got a paper from Mark Logsdon. He notes these examples of old mining:

  • Example 1: There is a record of sulfide mining in the Iberian Pyrite Belt that extends at least 4,500 years. Massive-sulfide copper deposits of the Eastern Mediterranean were major sources of copper in the Homeric Bronze Age, and they continue to be reactive today
  • Example 2: In erosional scars exposed in Quartz-Sericite-Pyrite altered volcanic and intrusive rocks in the Red River Valley, northern New Mexico, the radiometric dating of jarosite and alunite formed by oxidation of pyrite in these scars indicates that pyrite has been reacting in situ for periods of 30,000 to 1.5 million years, yet such altered rocks still retain much of their original pyrite. Calculated depletion times for waste rock there range from ca. 1,500 a to more than 100,000 a, based on plausible alternative models for the hydrodynamics of the waste-rock piles.

Mark then summarizes current thinking on the UMTRA standard of longevity/perpetuality, namely “1,000 years to the extent reasonable, and at any rate for 200 years. “ He writes:

There is no industry standard for such a long-term performance of closed mine facilities, nor are there established regulatory criteria for rock piles. The only geotechnical systems for which there has been extensive analysis of long-term periods of performance is for the mining and milling residues controlled under the Uranium Mill Tailings Radiation Control Act (UMTRCA) of 1978 (P.L. 95-604). The regulatory standard requires that control of tailings “shall be effective for up to 1,000 years to the extent reasonably achievable and, in any case, for at least 200 years” (EPA, 1983). The UMTRCA time frames were established to consider periods over which climatological and geomorphic processes could reasonably be predicted, given current knowledge of earth science and engineering. In a review of the technical basis for the regulations, the National Research Council, the contracting-review arm of the U.S. National Academy of Sciences, concluded that the 200-year period was consistent with our knowledge of the longevity of engineered systems, but that estimates looking forward 1,000 years must be thought of as qualitative and inherently uncertain.

The Academy has recently returned to this topic in an updated review of the performance of engineered barriers for waste management (NRC, 2007). The general conclusion of the NRC is that up to 20 years of field observations indicate that engineered waste-containment systems designed, constructed and maintained appropriately meet or exceed their intended performance. However, NRC notes that the demonstrated period of performance for such systems remains only a few decades, and that longer–term monitoring will be needed to show that performance over hundreds of years can be achieved reliably across the range of waste-management alternatives currently in service. A key finding of the recent study is that on-going maintenance is required (Mitchell, 2008).

Industry-standard practice by mining companies and their technical advisors considers that engineered covers, such as a stable, very low erosion slope with a low-net infiltration, sustainable re-vegetated cap can be established. Given the exploration, development and operation history of the mine and an expected construction and reclamation activity on the order of 50 years, there will be nearly 100 years of geologic and hydrogeologic data available for most modern mines. It seems entirely reasonable to project that time period forward and to establish a goal of 200 years performance for the engineered closure system at modern mines, in keeping with the rational of the National Academy.

I and my engineers designed and oversaw the construction of those cells. They will last a thousand years. If you doubt me just go see how the Cahokia Mounds have performed over a thousand years. Or see the Myan Mound at my Guatemalan site. Great stuff for idle and unemployed consultants to blather about. But guys, it has all been researched and said. You are perpetually repeating the obvious.

Now I know many hate the UMTRA formula of 1,000 years to the extent reasonable and at any rate for 200 years. Andy Robertson can expound for hours on why it is wrong, muddle-headed, and absolutely inapplicable in Canada. No proud Canadian would accept or seek to implement a U.S, standard. We must reinvent the wheel ourselves.

Even Mark cannot bring himself to accept and validate it. Instead he take refuge in this wordy restatement of the same idea:

Consideration of these periods, the scientific basis for understanding closure risk, and established engineering practice leads us to suggest that a reasonable, total planning period for management of mine wastes should be in the range of 200 years, and it probably should include a semi-quantitative assessment of whether or not major changes in performance are likely to occur between approximately 500 and 1,000 years. Plans should include (1) identification of risks to surface and ground water in terms of adverse impacts to beneficial uses, (2) presentation of a case that mine-waste structures would be stable with respect to erosion by flooding or deep-seated shear failures, and (3) presentation of a case that reactive wastes as disposed for those periods will remain physically stable.

There we go again: risk assessment is the magic wand!

Of course, on UMTRA we did not do risk assessments. They became fashionable long after we had closed 24 sites. Now we can “explain” away almost every decision on the basis of a spurious risk assessment. If you do not believe me, talk to Franco Oboni and his son Cesar Oboni at Riskope.

The consultants and operators in the oil sands industry have recently provided us with another definition of perpetual. They conclude that before you can de-list an oil sand tailings facility you must prove that is will respond to long-term geomorphic forces in the same way as the natural landscape around the closed and e-listed facility.

Now that is innovative and superbly bold: make the closed facility just like the hills & valleys, just like the wetland & rivers, just like the local arboreal forests of northern Alberta. An amazing concept. I would have thought this would come from a regulator. But is comes from the consultants and operators of the oil sand mines. So we must presume they honestly believe they can do it, and achieve this immeasurably ambitious goal, and can close their mines so that an alien in an arriving spaceship could not tell the difference between nature and man-made.

Their naivety is stunning. Not the aliens! But the humans!

OK, I concede that it is impossible to get an answer. I concede that the reality is that every closed mine will need institutional controls which implies the presence of government. I concede that it is good to believe that government, as we now know, it will be there. And to close mines on the basis that the government will be able and willing to do whatever is necessary in the long term. Look at Giant Mine and Faro—perpetual make-work projects by any definition.

But in my heart, I cannot believe that governments will always be there and willing to undertake surveillance and maintenance of inadequately closed mines. Maybe I have read & thought too much of history to believe that what is now will always be.

Thus I must conclude that the simple, professional thing to do is close mines to be stable for 1,000 years to the extent reasonable and at any rate for 200 years. I know it can be done; I have done it.

The rest is repetition, avoidance, dodging, and persiflage.