Remote sensing has been proven to be very useful in the investigation of abandoned mine sites and other sources of environmental pollution where heavy metals and lethal acids have impacted the environment. Through the use of AVIRIS and land based chemical analysis, environmentally effected areas can be investigated to determine the point sources of pollution. This type of analysis provides the necessary information needed to conduct a clean up and restoration of an effected area. This project looks at one of the EPA's super fund sites and how AVIRIS was used in its initial investigation.
The geologic history of the Summitville area dates back to about 30 million years ago when several active volcanoes inhabited the region. The mine itself is situated at the top of what's left of an extinct volcano whose summit collapsed forming a coldera. Starting at the beginning of the Cenizoic and extending as recently as the Pliocene the region was subjected to several cycles of Magma intrusion, Hydrothermal alteration and mineral formation. This repeating process left a distinctive mark upon the landscape. The rocks through out this area exhibit signs of extensive corrosion due to the effects of sulfuric acid. As magma intruded towards the surface it carried along with it several dissolved gases and metals. The primary gases were sulfur dioxide SO2, hydrogen sulfide H2S and water. As the magma cooled the gases separated from the water and formed sulfuric acid. As the sulfuric acid moved towards the surface through cracks and fissures it dissolved the surrounding rock creating a porous surface for later mineral deposit. In the Summitville area these metal deposits consisted of gold, silver, iron, copper and arsenic. While in the Alum and Bitter Creek areas very little gold and silver were deposited. This difference in geology would be helpful in the later investigation of the Summitville area. Since little to no mining took place in these areas.
Gold was first discovered in this region in the 1870's by James L. Wightman. By 1883 Summitville had emerged as the third largest gold producing district in the State. In the early 1980's the area was mined by Summitville Consolidated Mining Company, Inc. From 1985 -1992 the company engaged in the extraction of low grade ores through the process of open pit cyanide heap leaching. Following the mines closure the mines operators began an environmental remedial process to restore the area to its pre-mined condition. They had begun treatment of the heap leach pit and had installed a water treatment facility. But in December of 1992 the company declared bankruptcy. Due to the hazardous condition of the mine, on December 16, 1992 the response branch of the EPA moved in to take responsibility of the site.
In the United States there are between 100,000 to 500,000 abandoned mine sites. Many of these sites are responsible for the degradation to the environment around them by hazardous run off of heavy metals and acids. Because of this enormous environmental hazard congress passed the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) in 1980. This act is commonly referred to as the Superfund. The purpose of this legislation was to identify sites that release hazardous substances into the environment and devise a plan of action for there ultimate clean up. Currently there are about 50 Superfund sites around the country with most being in the western portion of the country. In May 1994 the Summitville area was added to this list of superfund sites when the EPA placed it on its National Priorities List.
The USGS began an investigation of the Summitville area using the remote sensing capabilities of the AVIRIS system. The following data was collected on September 3,1993 and consisted of four data runs.
AVIRIS is an acronym for Airborne Visible and Infra-Red Imaging spectrometer. This system was designed by Nasa as a means to gather greater spectral and spatial data of the Earths surface. AVIRIS collects data in the spectral range from 0.4 micron to 2.45 microns in 224 continuous spectral channels. The instrument is flown onboard NASA's ER-2 a modified U-2 spy plane at an altitude of 60,000 ft. The normal pixel size for an image taken at this altitude is a 20 by 20 meter square for targets at sea level. Due to the altitude of the Summitville site, approximately 11,500 ft above sea level, each AVIRIS pixel depicts a much smaller square, in this case a unit of measure of 17 meters square. "Aviris scenes are often referred to as "Data cubes" since there are 224 images associated with each scene; hence there are three dimensions, just as there are in a cube, except in a data cube there are two spatial dimensions," representing the x and y axis "and one spectral." (Roger 1998) The spectral unit is recorded in 12 bits which can yield a range of values from 0 - 4095.
Traditionaly investigations such as the one that took place at Summitville have been geochemical in nature and required numerous collected samples which were identified and measured in a laboratory. Although optical wavelength remote sensing can not directly detect trace elements it can be used effectively to map the minerals that host these metals. Two groups of host minerals that were used in this investigation were the ferric oxides and oxyhydroxides.
In order to conduct the investigation it was essential that the data be reduced from its at sensor radiance to apparent surface reflectance. To achieve this a two step method was used. "The first step consisted of applying the ATREM radiative transfer based atmospheric correction code to each of the scenes. (Farrand 1995) Due to absorption at prominent atmospheric wavelength's a second ground base correction was needed in order to use the data. The ground base correction was the modified flat field (MFF) correction. "In the MFF correction the average radiance, or, in this case, the average ATREM derived apparent surface reflectance of a spatially and spectrally homogeneous area on the ground is divided into each pixel vector of the image cube. Each pixel vector of the resulting image cube is then multiplied by the laboratory or field measured reflectance of the target area." (Farrand 1995) The calibration site chosen was a plowed field about 18 km sw of Alamosa. Soil samples were obtained by the USGS on the date of the flight and the reflectance of the soil was measured using the USGS laboratory spectrometer.
For the Summitville study 64 minerals were mapped based on the presence of there absorption features. The wavelength regions investigated were the 0.45 - 1.0 microns, 1.5 micron, and 2.2 to 2.3 microns. Absorption bands between 0.4 - 0.8 microns is due to the reflectance associated with elements who are of the first transition series and have an outer unfilled d-shell. The element associated with this is iron and substances that contain iron in their mineral structure. Differences in the valance state and crystalline structure will create individual and distinct absorption bands with in this range.
The near infrared radiation of the 1.5 micron band focuses on the vibrational energy of the investigated substances. The frequency absorption is dependent on the relative masses, geometry of the atoms and the forces between the bonds of the molecular structure. "There are two main types of molecular vibrations: stretching and bending. A stretching vibration is a movement along the bond axis which either increases or decreases the inter atomic distances. Bending vibrations consist of a change in the angle between bonds with a common atom or the movement of a group of atoms with respect to the remainder of the molecule, but without movement of the atoms in the group with respect to one another. Only vibrations that result in a change in the dipole moment of the molecule will be infrared active." (Roger 1998)
The absorption features of the 2.2 and 2.3 microns is associated with the -OH bond. In particular this range looks at the stretching and bending of these bonds in such molecules as AL-O-H and Mg - OH. Based on previously collected laboratory data this band has been broken down into 48 mineral standards that the Summitville data could be compared to.
The ferric oxide portion of the investigation of the Summitville site concentrated on looking for the presence of 22 minerals with absorption wavelengths at or near 1.0 microns. A sample of sludge was obtained from the Reynolds Tunnel and sent to the USGS laboratory for mineral identification. Based upon the comparison of the laboratory results and the AVIRIS data 7 of the 22 minerals being investigated were successfully identified. "The minerals detected include: amorphous iron-hydroxide, ferrihydrite, goethite, hematite, K-jarosite, Na-Jarosite, and an Fe bearing material." (Roger 1998) Figure 3 shows their distribution around the Summitville site. Note the natural presence of ferric oxides around the Alum and Bitter creek areas. Comparisons between these two sites and the Summitville mine give a good indication of the amount of damage the mine has caused in the area.
The oxyhydroxide part of the investigation took a look at the wavelengths between the 2.2 - 2.3 micron region of the electro magnetic spectrum. Minerals in this range were compared to a laboratory standard of 48 minerals and the test results of the Reynolds Tunnel sludge. From this comparisons 8 minerals were detected in or around the Summitville mine. Figure 2 shows the distribution of six of these eight minerals. Due to the limitations of the mapping algorithm and because of spectral similarities it is possible the some of the Na-montmorillonite shown in yellow may actually be muscovite or sericite. Once again the regions around the Alum and Bitter creek rivers give an indication of the extent of damage caused by the Summitville Mine.
Figure 2. Image from the USGS Spec Lab: Summitville Mine Drainage
In order to restore the Summitville mining area to its former state several projects must be completed. The long term remedial work involves plugging the Reynolds Tunnel and back filling the open pit with material from the Cropsy mine waste dump. Action must be taken to reshape the mine to prevent further run off. Portions of the site will need to be revegetated in order to prevent further erosion of the soil. The final site remedial investigation and feasibility study will be completed by the summer of 2001. Upon completion and review of the final analysis the EPA and CDPHE will give their recommendations to the public on the most effective means to restore the area and preserve the Alamosa Basins aquatic ecosystem.
Figures for the final clean up of the Summitville area range from $100 - 160 million dollars with the cost for the remediation being funded by the EPA and the State of Colorado. The EPA has pursued the parties responsible for this ecological disaster and have recovered a small percentage of the total cost mostly from parties who had limited exposure to the operation and whose responsibility for the damage was minimal. In December of 2000 the U.S. Department of Justice reached a settlement with Robert Friedland a former operator who agreed to pay the U.S. Government and the state of Colorado a sum of $27,750,000. This money will be used for further restoration of the site.
Remote sensing of the Summitville area was undertaken to assess the ecological impact of the Summitville mine. In figure 1 the AVIRIS data revealed that both the Summitville mine and the Alum and Bitter creeks areas were all sources of iron bearing sediments in the Alamosa River and potential carriers of heavy metals. In figure 2 the AVIRIS data revealed distinct distributions of Hydroxyl bearing materials at both the mine site and the Alum and Bitter creek areas. This distribution reveals that all three of these areas are sources of OH bearing minerals for the Alamosa River. One observation from the data is that the mine is not a significant source of OH pollution to the Wightman Creek. Any remedial efforts to clean up the area will have to take into account not only the mine site itself but the natural occurrence of these sediments. Finally the results obtained from this investigation reveal that remote sensing is an effective tool in the detection and mapping of acid mine drainage.
Emmons, W.H. 1937. Gold Deposits of the World. McGraw-Hill Book Company, New York, 562p. Jenson, J.R. 2000. Remote Sensing of the Environment, Prentice Hall, New Jersey, 544p. Eaton G.P. 1995. Environmental Considerations of Active and Abandoned Mine Lands : Lessons from Summitville, Colorado, United States Government Printing Office, 35p. Groat C.G. 1997. Relations Among Rainstorm Runoff, Streamflow, pH, and Metal Concentrations, Summitville Mine Area, Upper Alamosa River Basin, South Western Colorado 1995-97. United States Government Printing Office, 32p. (Dr. Roger N. Clark) 1998. USGS Spec Lab: Summitville Mine Drainage Mapping. http://speclab.cr.usgs.gov/PAPERS.summitv/summitv.html (Farrand H. Farrand) 1995. Mineralogic Mapping Over Summitville, Colorado Using AVIRIS Data: The Challenge of Ferric Oxide Minerals and Mineraloids. http://ltpwww.gsfc.nasa.gov/ISSR-95/mineralo.htm [CDH] Colorado Department of Health. 2001. Summitville Mine Site Summary. http://www.cdphe.state.co.us/hm/summitville.asp [EPA] Environmental Protection Agency. 2001. Summitville Mine. http://www.epa.gov/region8/superfund/sites/sville.html