The coal mines that lie under portions of downtown Lafayette and Louisville, outskirts of Erie and most of Frederick, Firestone and Dacono were built to collapse. Coal mines on the Front Range from the 1860s when coal mining began in Colorado until roughly 30 years ago used the technique of “room and pillar” mining — tunneling into a coal seam and digging rooms off the side between pillars of coal. Miners began at the far end of the tunnel, shaving coal off the pillars until the shaft’s roof timbers cracked, the roof caved or pillars began squeezing into the opened space and the room was deemed unsafe and abandoned to cave in. Some caved immediately. Some are just doing so now, an event geologists refer to as a subsidence. They can take the shape of a bell-shaped pit, hole, trough or sag, and tension cracks can form around the perimeter of subsided ground.
The sinkhole that opened in a Lafayette road in January, nearly swallowing an SUV, was one such event — a mine shaft that accessed the Simpson Mine’s workings 200 to 250 feet underground opened up to the surface.
Where coal has been mined, voids have been left under ground. Mines at any depth, mined at any time, can collapse.
“The shallower the mine workings are, the higher the risk of subsidence, but there’s a risk when you live over any coal mine” says Rachael Nickless, project manager with the Colorado Division of Mining, Reclamation and Safety.
Underneath swaths of Lafayette, Louisville, Erie, Frederick and Firestone, U.S. Geological Survey maps show abandoned coal mine workings at 50 to 150 feet deep.
But few residents have enrolled in a state protection program to pay for damage to homes from coal mine subsidence, and oil and gas development is raising questions of increased concern as geologists have documented subsidence events tied to resource extraction.
An estimated 7,500 houses and 25,000 people in Colorado live in homes that are at risk. A subsidence can cause foundations and basements to pull away from the home structure, asphalt to buckle, brick and stone walls to crack, sag or tilt, roofs to sag, drywall, windows and doorframes to crack and fireplaces to separate from a building according to the state’s guidebook for homeowners. When a sudden pit or sag appears, a house can be heard popping and cracking. Water, sewer, phone and gas lines may be damaged, break and leak, and gravity flow systems may not drain properly.
Work to safeguard an area after a subsidence can take days or weeks to complete, Nickless says. Often, grout is used to mitigate a collapse under a house, and that work can take up to three weeks to finish. Home foundations can also be reconnected to bedrock with helical piers that could take up to four weeks to install, she says.
Many homeowners aren’t aware that their homes are over sites where subsidence may occur, and conventional homeowners insurance doesn’t cover the damage. The federal government gave Colorado $3 million to establish the mine subsidence protection program, which pays for damage to homes that results from coal mine subsidence. That money is kept in a trust fund designed to be self-sustaining.
Though thousands of people stand to be affected by mine subsidence, Nickless estimates there are 950 people currently enrolled in the state’s mine subsidence protection program.
The program includes houses built before 1989, when the Surface Mining Control and Reclamation Act of 1977 took effect. Before then, there was no oversight on the question of building over old mines.
“None,” Nickless says. “And there could still be more [oversight].”
Regulation is left to counties. Some do well with making sure builders and developers go through the geologic hazard process, she says, but not all counties are the same and they show different levels of caution.
“I just think that more mitigation measures could be taken before building over undermined areas than are now,” Nickless says.
Those mitigation measures could include filling in mines, blasting and making sure the mines are subsided before building proceeds.
There was little uniformity in how mining was done, making it difficult to evaluate the severity or timing of a room’s collapse. Entries were roughly 15 feet wide and led to a grid of pillars and rooms that could be 80 to 90 feet long. When the rooms and pillars were first established, the pillars that could be 20 to 30 feet wide and the rooms 10 to 12 feet wide but would become twice that width as coal in the pillars was mined away. The outline of those workings have become visible in aerial photos taken after a subsidence occurs. Mine shafts were often closed off with dumped mine debris and waste that was poorly compacted.
Natural caving and flooding have made accessing these underground workings now to map them impossible, and so predicting future subsidence events is “extremely difficult,” according to the guidebook for homeowners from the Department of Natural Resources, Colorado Division of Reclamation, Mining and Safety and Colorado Geological Survey.
Most mines collapse shortly after they’re abandoned, which begins the subsidence process, but subsidence can occur 100 years or more after mining stopped. How long the subsidence process continues and how it progresses depends how long the pillars and timbers left in place support the roof — the thinner the mined seam, the longer the roof is supported. The presence of groundwater can help buoy the pillars and prevent them from oxidizing, but may also lubricate certain soft types of rocks on the mine roof and floor, and periodic flooding or draining of the mine increases the chance of pillar failure.
In the Front Range, coal seams are often surrounded on either side by sedimentary rocks — sandstone, claystone or conglomerate layers that are more porous and more prone to water seepage.
“If we have major changes in water that could definitely affect things,” Nickless says. “A prolonged drought can cause things to move. A big flood event like we experienced a few years ago, that can definitely affect the way things move. And piping, the soil can pipe down into the mine workings and eventually work its way up to the ground surface. Colorado does not have the best soil in any area for construction for houses. We have swollen soils and collapsible soils, which all can tie into subsidence, too, and exasperate the issues with subsidence.”
The state collected maps of mines while they were in operation, but mining may have continued after the maps were drawn, changes in land surveys and landmarks may mean the map isn’t accurately positioned relative to the surface, maps of smaller mines were never made or no longer exist and some mined areas were never drawn on maps — “coal poaching” was common. The Colorado Geological Survey maps include a buffer zone around known mine workings, Nickless says.
“Anything that’s within that buffer zone, we consider to be undermined. It may not be, but it may be,” she says.
There are also mines that submitted records for production, but no actual location.
The state commissioned more detailed subsidence hazard investigations for coal fields in Colorado Springs and Boulder- Weld that included depth of mining, thickness of the mined coal seam, existence of multiple mining levels and remaining coal pillars, and used that data to estimate subsidence potential. Subsidence investigations locally may include examining records as well as drilling exploratory holes to confirm the existence of mine workings and whether void space still exists under ground or if a subsidence has already taken place.
In the Raton Basin on the Colorado-New Mexico border, the U.S. Geological Survey has found evidence of subsidence in satellite images that is connected to underground water or gas withdrawal. The area had experience induced seismicity — increased seismic activity that was caused by wastewater injection in the area — and had begun to study surface deformation that followed a magnitude 5.3 earthquake that occurred there in 2011, which U.S. Geological Survey research had determined was induced.
In the satellite images used to study the ground surface deformation — which was occurring in the range of 5 mm to 10 mm per year — researchers could also see substantial ground subsidence that was likely tied to extraction of natural gas, says William Barnhart, a former research geophysicist for the U.S. Geological Survey and lead author of a study on the area that was recently released in the Journal of Geophysical Research.
“I should point out that down in that basin, there is no hydrofracturing going on,” Barnhart says. “The subsidence that we see is most likely due to them pulling out natural gas, which with that comes a lot of fluids, a lot of water, so the same sort of idea would apply to a region where they’re doing hydrofracturing because you’re removing fluids from the ground.”
Ground subsidence of a couple centimeters or more can occur from a lot of different human activities, over-tapping of aquifers being one of them, he adds, and there have been cases where ground water extraction has led to meters of subsidence over the course of decades.
“We certainly see it in oil and gas fields, simply because as you remove material, the ground surface settles and you can have subsidence due to that,” he says. “The amount of subsidence really depends on two things, one is how much fluid is being withdrawn and then what is the local geology. So if you live on hard bedrock you won’t have that much subsidence necessarily but if say you live in a region where there’s a lot of sedimentary rock that has a lot of fluid in it, then you can get a lot of subsidence.”
The Front Range is thick with sedimentary rock, and coal is often bookended with layers of sandstone, claystone and conglomerate.
Given that information, he says, “You could see a fairly large amount of subsidence.”
“Colorado does not have the best soil in any area for construction for houses,” Nickless says. “We have swollen soils and collapsible soils, which all can tie into subsidence, too, and exasperate the issues with subsidence.”
Concentrated areas of oil and gas activity could lead toward greater subsidence, whereas spreading the activity over a few square miles may reduce the effects.
Though there have been a couple instances of earthquakes that are suspected of having been induced by fracking — wastewater injection is the more frequent source — most of the seismic activity tied to fracking is on the lower end of the scale, magnitude three or magnitude four, so small that it may not trigger a ground deformation, he says, and “any earthquake-specific subsidence would come from larger events of magnitude, say four and a half or greater.”
As Front Range communities see increased pressure to permit oil and gas extraction in the midst of neighborhoods, how much these two issues of subsidence and resource extraction interlock may call for additional research.
The Colorado Mine Subsidence Information Center is a repository for all known existing maps of inactive or abandoned coal mines in the state and can be reviewed by appointment with a Colorado Geological Survey staff geologist. For an appointment, call 303- 866-2611.