This review describes erosion at mine sites and provides information for miners and geotechnical engineers on evaluating, analyzing, and controlling sheet erosion, gully erosion, and long-term geomorphic-induced regional erosion that may affect the environment of reclaimed and closed mine sites.
Control of erosion is an ever-present mine reality. In this review, I will lead you through the papers, web sites, consultants, and commercial products available to control both short-term and long-term erosion at you mine site.
I define and address three types of erosion relevant to mine operations and closure:
- Sheet or surficial erosion that generates sediment and affects surface-water runoff quality.
- Gully erosion that results in exposure of the underlying layers of covers and contaminated materials.
- Topographic erosion that in the long term results in lowering of the landscape and undermining of the closed piles and dumps.
GENERAL WORKS ON EROSION
For those with a deep interest in erosion, see the monthly magazine Erosion Control with headquarters in Santa Barbara and a web site with current issues at www.erosioncontrol.com
Another general interest magazine is Soil Erosion and Hydro seeding, copies of which are available online at www.soilerosiononline.com.
Although it concentrates on the farming aspects of erosion, the basic description of the physics and types of erosion at www.farminfo.org is as good as any on the internet.
The manual for the revised universal soil loss equation as applicable to mines is available for download.
Software for erosion evaluation, engineering, and detailing is described at http://www.erosiondraw.com.
Essential publications on erosion in the mining industry that must be read in conjunction with this review include:
Incorporation of Natural Slope Features into the Design of Landforms for Waster Rock Stockpiles.
The best paper and presentation (for my money) at the SME conference in St Louis is that by Brian Ayres and his colleagues at O'Kane Consultants. I am much indebted to them for allowing me to post the presentation they made on InfoMine. I greatly and enthusiastically recommend you spend some time reading both the paper and the presentation.
On the basis that you may be busy, and because I believe in what they say and want to encourage you to read their paper and peruse their presentation, here is a precise of their main points:
- Design the final landform using natural analogues.
- The reclaimed landform can be no more stable than adjacent undisturbed areas; therefore use gentler slopes, higher density drainage, and smaller drainage basins.
- The preferred reclaimed slope is a "spur-end" slope plan with a concave of complex (convex-concave) profile.
- Avoid terraces and contour banks.
- Use computer codes to predict erosion profiles over at least 100 years.
- Avoid man-made materials such as pipes, gabions, and concrete.
- Design conservatively for extreme events
- Incorporate wetlands and small lakes into the design to attenuate runoff and peak flows.
Erosion Control, How Practical is it---Really? describes use of the computer code SIBERIA to quantify long-term erosion from tailings impoundments, waste rock dumps, and heap leach pads. Twenty-five figures illustrate the impact of erosion on well-shaped and ill-shaped mine waste disposal facilities. (The PowerPoint presentation for this paper contains good illustrations of all the salient points—download it but beware, the file is big and it may take a while) Recommendation include:
- Armor to limit erosion.
- Reshape piles to less erosion-prone forms.
- Contain eroded material in berms at the toe of the dump.
- Direct runoff and eroded sediment to the adjacent open pits.
Back To Top
CONSULTANTS ET AL.
Most environmental consultants will be able to assist you with the engineering of erosion control. Here are three companies that provide substantive information on their erosion control practices:
Water & Earth Technologies Inc. Fort Collins, Colorado ... states that they specialize in designing plans to control surface runoff and erosion. They use erosion and sediment control models, including RUSLE and SEDCAD and have developed and implemented programs for disturbed and reclaimed areas to evaluate erosion potential and sediment production, stream dynamics (of both natural and altered stream channels), and the implementation and effectiveness of Best Management Practices (BMPs).
West Consultants Inc. Tempe, Arizona ... provides copious information about their erosion control practice including this case history: The Tin Shed site is an alluvial terrace located along the Snake River in Hells Canyon National Recreation Area. Significant erosion of the riverbank adjacent to the Tin Shed site has occurred. The erosion has caused retreat of the bank line, undermining of riparian trees, and possibly loss of cultural artifacts. Continued bank erosion presents a serious threat to the historic and prehistoric cultural resources of the Tin Shed site. The Idaho Power Company contracted WEST Consultants to develop conceptual designs for alternatives to control bank erosion at the Tin Shed site.
North American Green. Evansville, Indiana ...has the most comprehensive site on erosion control including products, publications, design aids, and a list of the specialists who can help you. This is how they describe their company: Manufacturing quality rolled erosion control products is our only business. With this in mind, it is our goal to continue the effort of supplying and supporting our product line with the latest in erosion control technology.
I define short-term surficial erosion as the immediate and quick loss of the
upper layer of soil by those nuisance storms that plague all mines. Most often
you will not want or need to analyze such erosion; it is all too obvious when an
unprotected soil surface is vulnerable to erosion and when stabilization is
required. That is the time and the place to apply erosion-control best-management practices and commercial products.
In South Africa at the start of my consulting career I was called in to limit erosion from an old sandy pile of tailings close to the city. Vegetation would not grow in the acid materials that were cemented by negative pore pressures to a hard crust. We read all the literature of the seventies but found no answer. The gut-feel solution was to cut a series or benches, one-foot high by one-foot wide, with vertical and slightly inward-sloping near-horizontal surfaces. Rain fell on the near-horizontal benches, ponded, and seeped into the tailings. Nothing ran off and erosion was controlled.
On other nearby tailings piles, wind erosion gave rise to vast dust clouds that blanketed roads and buildings in the dry season. In vain we tried vegetation; soon enough it was dead and ineffective. We tried furrows and berms to break up the wind, but to no avail. Finally in desperation, we spread cement and ploughed it into the upper surface, making a crust of cementitious soil. It worked, the dust subsided, the surroundings were spared, and the cost was low. I do not know if this system is still used.
In late 2004 I was on a team sent to assess burnt areas of San Diego affected by brush fires. Our mission was to specify what should be done to prevent impending rains from washing down the soil, causing mud slides, and damaging homes, businesses, and traffic. It was obvious which areas were vulnerable to catastrophic erosion: those steep hillsides where no vegetation remained and all was ash, where there was little gravel and mostly sand and silt, and where there was both a large upstream catchment area and a vulnerable house or road downstream.
We specified and installed the following: spray commercial green stuff where the soil was vulnerable to washing away; sand bags to impede gully development; rock dams to create settlement ponds; and fences to catch the rocks that would otherwise have crashed onto busy roads. By the time we were halfway done, the countryside had turned an artificial green; every jute bag in the county was filled with sand and aligned along a ditch or road, and the clogged and filled sediment ponds emptied in anticipation of a new inrush of mud. It rained, and along a busy residential road, the mud and ash came down in a three-foot high wall of gunck, pushing trees and debris and covering driveways and ditches. An impressive sight, and to the credit of the county road folk, quickly cleared. Neither a house nor a car was mud besplattered.
A drive across country gives a good perspective on the commercial products well marketed to road departments. Depending, no doubt, on the preferences of the local engineers and the persistence of particular salesmen, you will see acres of jut matting, hay bales, silt fences, sand bags, and stakes. They are so neat and tidy and vulnerable. Here are some of the commercial products and suppliers you may wish to examine in deciding what to do at your property:
The Infomine Suppliers guide provides a complete list of suppliers of the many products used to control erosion at mine sites. For example, Tensar at www.nagreen.com contains copious information about erosion control products, computer programs that promise to analyze you situation and equally interesting a full section of the use of geosynthetics in underground mining.
At www.constr.com is a comprehensive list of suppliers of erosion control materials and services. Also on this site are listings of consultants specializing in erosion control, software to analyze erosion, and technical papers that they promise to change every month.
LONG-TERM EROSION CONTROL
Soon after joining the Uranium Mill Tailings Remedial Action (UMTRA) Project, I was sent to Washington, D.C. to meet with Ted Johnson of the U.S. Nuclear Regulatory Agency.
With a group of extraordinarily talented fellow engineers on the 14th floor of an old building in Albuquerque, we put together the equations and calculations to prove to Ted Johnson that what was proposed would control erosion on the remediated uranium mill tailings piles for 1,000 years at least.
He never let us off the hook regarding the tiniest detail. But he was intellectually penetrating and inquisitive. And we owed many of the advances we made to his persistence to delve to the depth of the issue.
I had one major disagreement with him. Somehow his colleagues in the NRC in Denver were, in my opinion, letting the private sector get away with lesser measures to control erosion than Ted was demanding of us. I manage to publish a paper that lost me a friend—my co-author accused me of betrayal and has not spoken to me since publication of the paper. Sometime after publication of this paper, responsibility for review of erosion control works was transferred from Denver to Washington. Ted won that battle.
Ted collated all the work we did regarding erosion on the UMTRA Project and all the work that was done by the consultants to the Title II (private sector) uranium mill tailings remediation site, into one major report: Design of Erosion Protection of Long-Term Stabilization (2002).
The full pdf of the report is available free for download by using the following link: Report.
In my opinion, this report should be on the desk of and should be used by every engineer involved in closing any mine waste disposal facility. This is a magnificent report. Not necessary well written; not necessarily well formatted; in fact it is pretty ugly. But it contains and distils an enormous amount of work, discussion, dispute, and trial & error. It encapsulates an enormous amount of wisdom. It stands as a monument to conservative practices that work.
That is the criticism that will be thrown at the report: the technical procedures that Ted mandates be applied to the closure and remediation of uranium mill tailings facilities are too conservative to be applied at ordinary mines that worked gold or copper or those less scary minerals and metals and commodities. Justifiably you may ask why set a design life of 1,000years for a gold mine tailings impoundment. What is wrong with 100 years?
In fact why not go for 30 years? All net present value calculation tells you it is better to postpone the expenditure to 31 years hence.
The criteria set for mine closure works are specific to each mine, to its jurisdiction, to its risks and rewards. At least take a look at what Ted Johnson has written to establish the baseline of excellence before you adopt lesser criteria and accept higher risks of failure.
Of much greater long-term concern to the mine is gully erosion. It eats into embankments, waste piles, tailings, and reclamation covers. Nature wants to reestablish equilibrium topographic forms where the artificial shapes of piles and covers stand proud of the surrounding landscape. To understand what will happen, I recommend going out into the field and observing what nature does and how natural soil and rock surface resist erosion and the development of gullies.
I have never come across a truly reliable equation or computer code to predict and analyze gully development. On the UMTRA Project we used a forgotten equation to prove to the US Nuclear Regulatory Commission that long-term erosion would not cause gullies in the uranium mill tailings piles we were closing. The details are available in the UMTRA Technical Approach Document. In that this document is difficult to obtain, rather just go into the field anywhere from New Mexico to Montana and you will see that talus slopes of rock grading 1.5 to 6 inches do not form gullies. That is the same gradation we used for the rock covers on the UMTRA piles from Texas to Oregon; the UMTRA website records the performance of these covers over the past decade and more. More recently others have advanced the state of the art; I refer you to Jody Waugh and his co-authors who write of what they did to cover the Monticello pile. I wish I could have worked with them and been part of their success.
At a mine in Idaho with which I am associated, it is clear even through the misted windows of the truck carrying us around the site, that steep slopes covered by a gravel-rich soil supporting abundant pines resists erosion. We will seek to replicate this natural form and its soil and vegetation mix when we put a cover on the pile. At this same site, it is obvious that gully erosion is endemic on 3:1 slopes any higher than 25 ft. Over one season, gullies up to three-feet deep and spaced at five to ten feet apart developed in a newly placed soil cover. Conversely, areas of a pile where ten years and more ago contour ploughing was done are still uneroded. What little rain falls and all the annual snowmelt ponds in the troughs created by the contour ploughing and that is where there is sparse vegetation. A few runs of UnSat-H and we convinced ourselves that the ponding water did not translate into infiltration or leachate (seepage), but instead the moisture entered the upper few feet of soil, stayed in the soil pores, and from there was evapotranspirated by the vegetation. What a good way to manage runoff and erosion!
In the Infomine.com library are papers on erosion; I refer to only one that strikes me as worth reading. It is the paper by Christopher Sanchez and Steven Anderson Designing for 1,000 years-Erosion Protection. The copy in the Infomine library gives no indication of where or when the paper was published. They describe the closure works for the Title II Uranium Mill Tailings Pile at the Bluewater Facility in New Mexico. They document the theoretical and practical considerations leading to placement of a rock layer over much of the pile to control long-term erosion. I confess sympathy for their conclusion that soil sideslopes cannot otherwise be protected from erosion. Calculations tell us that and so do observations of even the most common slopes across this broad country. Incidentally, this paper may not be readily downloadable if you are not an Infomine subscriber-in that case e-mail me at email@example.com
and I will get a copy to you.
Back To Top
Here is a detailed review of a paper called Thresholds, Triggers and Time –Erosion Risk on Evolving Reclaimed Landforms after Bauxite Mining in the Darling Range, Western Australia
by F.C. Mengler and R.J. Gilkes, both of the Center for Land Rehabilitation, The University of Western Australia. The paper is available in Mine Closure 2006
They describe their observations of gully development at reclaimed bauxite mines. Here is how they summarize their observations:
A combination of concentrated flow and trigger points caused gully headcut incision. Ponding, crusted soil and silt-like furrows were present upslope at many gully headcuts---providing a setting for “fill and spill” overland flow. Small pipe-like openings 2- to 15-cm wide on the face of tilled furrows upslope of a gully at Taipan [one of their sites] indicate that piping and subsurface flow may have contributed to some gullies at this site. The most common setting for gully erosion was at points in the landscape where water run-off from upslope forest areas was concentrated onto the steep shoulders of rehabilitated hill slopes. A poor standard of rehabilitation tillage (off contour, incomplete, or not seamlessly tied into adjoining areas) led to run-off concentration. Gullies incised wherever this concentrated run-off met a trigger such as concavities, thalwegs, knickpoints, boulders, poorly-installed fauna habitats or the remnants of waste rock dumps. Insufficient depths of returned topsoil and overburden (less than 20 cm combined) also triggered gully erosion. Steep gradients ort long slope lengths made gullies more severe but did not trigger gullies.
The authors propose that the critical factor in gully initiation is the relationship between critical slope gradient (Scr) measured in degrees, and the upslope drainage area (A) measured in hectares for the three sites they studies. They express the relationship as:
Scr = a A-b
These are the values of “a” and “b” for the three sites they studied:
On the basis of their observations, the authors propose two types of gullies that develop as follows:
Type A Gully: The gully is caused by the concentration of fast-moving, surface water in excess of infiltration capacity during “fill and spill” episodes involving the overflowing of ponds within contour furrows upslope. These gullies form when rapid flows cut down into weak topsoil and overburden materials. Once formed, they act as conduits for more upslope and lateral surface flow.
Type B Gully: The gully is caused by substrates that are saturated either by inherently slow-draining subsoil or by geological impediments to groundwater flow. Some Type B gullies develop from small pipe-like seepage outlets in upslope contour mounds. Type B gullies form more slowly than Type A gullies, mainly by seepage, slumping, and/or mass wasting. Once formed, Type B gullies evolve similarly to Type A gullies, but can act as conduits for both surface and subsurface flow.
The authors do not bring all this insight together. Rather they simply conclude that a combination of these major triggers can cause gully erosion:
- Excess off-site waster supply into the eroding area
- Poor surface completion resulting in concentrated flow and/or poor infiltration
- Insufficient depth of returned topsoil and overburden
- Fauna habitats displacing contour lines
- Shallow groundwater.
I have walked large areas of the west with a geologist who has specialized in geomorphology. We have sought erosion barriers and baselines: those dikes and sills that impede erosion and that effectively establish a baseline to upstream topographic change. We have speculated on the events that will break and remove these erosional baseline features and tried to quantify the resulting impact on upgradient topography and hence the integrity of closed uranium mill tailings piles. We have recreated the past 10,000 years of geomorphic history of the site and its surroundings. I am surprised how easy this is to do and understand in the presence of an expert geomorphologist. In practice, in many areas of the west, not much change has occurred since the ice melted and the big animals disappeared. Or is it that those are the sites we visited and selected. I recommend a visit to the UMRTRA website at … where you can access the Environmental Impact Reports and engineering design calculations that detail our efforts.
I recognize that conventional mine closure plans do not address the issue of base level lowering around the pile. And maybe we do not have to as the time frame is long relative to the thirty years of conventional mine closure planning periods. If you are proving that the closed piles will remain stable for 200 years to 1,000 years and maybe even to 10,000 years then you do need to quantify site and region geomorphic change in that period.
A conventional practice on a waste dump is to create benches at regular intervals up the slope of the facility. Water that runs off the sideslopes is directed off the pile via the bench roads, sometime reinforced with gravel and sometime provided with concrete-lined channels. In the interim this works, but any water management facility needs maintenance and this is costly in the post-mining phase. In time, the ditches fail, the gravel is breached and the water tumbles over the edge of the road and down the subtended slope. A gully starts down the slope and nature seeks to reassert its control to flatten the pile to the softer contours of the land. If this is part of an accepted overall fate, so be it. If not, you can be sure the regulators will be back, the miners will be back, and the cost start to accrue again. I recommend deciding well in advance on the chosen course of prevention, response, and action.
Here is a summary of an article on mine erosion by an old friend. I find the article fascinating and thus repeat the relevant parts. It deals with a computer code SIBERIA applied to a mine in Australia.
1Based on a technical paper by McPhail and Wilkinson MINE WASTE DISPOSAL FACILITY EROSION CONTROL IMPROVEMENTS see Infomine Library
See the SIBERIA code paper and also http://www.publish.csiro.au/paper/SR99035.htm. I got more than 40,000 hits on Google using keywords, SIBERIA mine erosion, although only the first twenty or so deal with the compute code itself-from there on you get information about erosion at mines in Siberia, interesting enough, but a diversion nonetheless. Please find the article summary below.
Typical Dumps and Piles
Figure 1 shows a typical waste rock dump or
stockpile. In this example, parts of the dump have been benched and covered with
topsoil in preparation for closure while other parts remain in operation. Acid generating rock has been buried within the dump under the high area.
Figure 2 shows the dump after long-term
erosion as predicted using the program SIBERIA2:
2 Willgoose GR (2000) Erosion processes, catchment elevations and landform evolution modelling Conference on Gravel Rivers 2000, Christchurch, New Zealand
- Extensive erosion of the benched slopes down the sides of the high part of the dump particularly on the western side. This is
because the topography of the high surface sloped to the west.
- The extent of gulleying which could expose the
acid generating rock.
- Erosion of other slopes where the top surface has drained over the slopes.
- Re-deposition of eroded material away from the pit potentially off the mine property.
- Re-deposition of eroded material away from the pit and potentially off the mine property.
Would It Be Worthwhile Dozing The Slopes
Here is a brief examination of dozing the slopes to reduce their inclination and control erosion. Erosion of the slope is about the energy of the eroding medium - the water. Since we are interested in the erosion over time we need to focus on energy over time or, more specifically, look at stream power since power is the rate of use of energy. Stream power has the equation ½ρQv2 where ρ is the density of the water, Q is the flow rate and v is the velocity. But Q = vA where A is the catchment area. For a 1-m wide slope of length L the area A is equal to Lx1 = L. This means that stream power is proportional to Lv3. Manning’s equation for open channel flow states that the velocity, v, is proportional to the root of the slope, S. Applying this proportionality to stream power means that the stream power is proportional to LS1.5. The table in Figure 3 shows the relative changes in erosion potential with slope angle and contributing slope length. For a slope at 37.5 degrees (approximately natural angle of repose) the erosion potential is taken as unity or base case. If the slope is flattened to 20 degrees, notwithstanding the increased slope length, the erosion potential is reduced to 0.6, a reduction of 40%. However, if the top surface above the 35 degree slope begins contributing to the flow the erosion potential increases from unity to 1.9 ie the top slope can almost double the erosion potential – hence the importance of ensuring that top surfaces or bench areas retain integrity.
Consider a slope at 37.5 and 500 years of erosion as predicted using SIBERIA and indicated in Figure 4 below and compare this with the 20 degree slope and 500 years of erosion indicated. It is evident that the long term erosion profiles for the two slopes are almost identical.
This indicates that while the volumes of material that are ultimately moved by natural erosion processes are less by virtue of the pre-dozing, ultimately the slope will adopt its own profile. The difference will be largely in the initial erosion rates where what would have taken place naturally have been affected mechanically instead.
In fact over the long term the net difference in the volume of eroded material between the natural angle of repose slope and the 20 degree slope is only 15% - significantly less than is indicated by the ratio of stream power as indicated in Figure 4 above.
Figure 5 shows potentially modified dump geometry where the modifications would involve controlled additional dumping and some re-shaping of existing areas.
Features of the modified geometry are:
- A drainage channel at a gradient of 1 in 25 from the central area of the dump to the pit area so that runoff water is directed to the open pit.
- The flat surfaces of the dump modified to gently slope towards the drainage channel even from the previous high dump area.
- A crest berm around the southern perimeters.
- The berms on the northern and eastern faces dozed away to form a continuous slope.
Figure 6 indicates the long-term erosion of the modified dump as predicted using SIBERIA based on the same parameters and time period as for the eroded dump indicated in Figure 2.
- Erosion is significantly reduced on all slopes as the contribution of the flat areas has been eliminated.
- Erosion on the previously benched slope is vastly reduced
Notwithstanding the above improvements, there is still opportunity to improve and optimise. For example:
- There is significant erosion in the “vee’s” that are generated where slopes meet. Erosion is higher in these areas because the flow aggregates down the “vees”. This effect can be reduced by rounding the “vees” to an extent where water continues to spread down the slope intersection.
- The slopes to the drainage channel could be flattened.
- The slope in the south eastern corner should be reinstated to form a continuous slope
In the case of a new dump it would be possible to incorporate the above features from the outset. In that case, rather than doze down the entire slope at closure it would be more cost-effective to plan ramps so as to form a stepped outer profile before hand and then simply doze between the berms. This will very considerably reduce the volume of material to be dozed especially if one considers the fact that on high slopes it is necessary to re-doze the same material several times in order to bring the slope down to the required angle. Moreover, it will be possible to generate a slope profile that approximates that of the long term erosion profile thereby further reducing the volume of eroded material. This is illustrated in Figure 7:
By “knocking the tops off” the benches at closure it is possible to approximate the 500 year profile of Figure 7. SIBERIA analyses indicate that the erosion volume from such a profile would be approximately 50% of the erosion that would occur from a slope of equivalent height but at natural angle of repose. This should be compared with the 85% figure for the 20 degree slope indicated in Figure 4 above.
If benches or step-backs are unavoidable it would be prudent to make their width approximately 3 times the height of the slope and to provide a crest bund of the geometry illustrated in Figure 8 show suggested criteria.
| Back To Top |