Mining is a complex activity that involves many kinds of risks. Among the most serious of these risks is collapse of the batters. Batters are the slopes that form the sides of open pit mines, in which ore is excavated from beneath the surface of the land. When these steep walls, which can be up to 15 meters high, give way, the mine can collapse. This puts miners’ lives at risk, causes material losses and interrupts work at the mine. In order to prevent these accidents, batters that show cracks, fissures or any type of earth movement are closely monitored by radar and other equipment on land. Soon, a new technology based on satellite images may help monitor these slopes, making mining safer in Brazil.
This innovation is the result of a research project developed under an agreement signed by Vale, FAPESP, Minas Gerais Research Foundation (FAPEMIG) and Pará Amazon Foundation (FAPESPA). The project began in 2012, and its goal is to detect and monitor the stability of the batters and surface deformations at Vale’s open pit iron mines located at the Carajás Mining Complex in the state of Pará, using an advanced method previously unknown in Brazil – DInSAR, or Differential Interferometric Synthetic Aperture Radar. Through this technique, high definition images taken by satellites provide information that allows millimetric detection of earth movements along the batters and other structures that make up the mining complex, such as access ramps, waste piles (unused material), conveyor belts, dams, roads and railroads, etc.
“The greatest innovation offered by this project, which was finalized last December, was to show that the DInSAR space technology can be used to monitor stability at active open pit mines, providing precise measurements of movements along the surface of mining and mining-related structures,” affirms geologist Waldir Renato Paradella, a researcher at the National Institute for Space Research (INPE) and coordinator of the FAPESP-Vale project. “This technique has been used in other countries to monitor deformations in oil and gas reservoirs on land and for urban engineering works, such as opening tunnels and building subway systems. For example, in Italy, orbital radar provides information that helps with earthquake monitoring. The study in Carajás was the first test at an active open pit mine under humid tropical conditions.”
Airbus
Image of Carajás, from the TerraSAR-X satellite in March 2012, shows the mines in fine parallel lines, with a dam at the larger dark pointAirbusOver the course of the project, this technique has been shown to be efficient in confirming the occurrence of deformations detected by Vale’s geotechnical corps at the N5W mine, one of the three open pit mines that make up the Carajás Iron Mines Complex. In addition to coverage from space, DInSAR technology has made it possible to quantitatively evaluate the cracks in the mine access ramp and the fractures on the batter faces. “This new technology is a predictive tool with a high degree of reliability. We are interested in incorporating it into our operations in the future and using it to continually monitor areas of interest,” notes geologist Luciano Mozer Assis of the Vale Ferrous Planning and Development Department. According to him, the use of SAR, or Synthetic Aperture Radar images onboard the satellite offers advantages over regular techniques because they cover large areas and can even be taken through cloud cover or at night.
While an image is being made, the SAR radar antenna installed on an orbital satellite emits millions of electromagnetic wave pulses towards the ground. When the irradiated electric field interacts with the target on land, part of the signal is reflected back and captured by the antenna. Processing techniques analyze two basic properties of this reflected signal: amplitude and phase. The amplitude indicates how much light the target reflects, while the phase shows how far the target is from the antenna. Therefore, the phase shows the distance between the radar sensor and a given point being imaged on land. SAR interferometry measures the change in phase of the signal between two images made of the same area at different times. The difference in phase between a pair of images is called an interferogram.
“When a target on land shows movement between two images, the distance between the radar sensor and the target changes; this is expressed as a difference of phase. When this difference is related to a point located on the batter of a mine, it probably indicates that the ground has moved at that location,” Paradella explains. The presence of clouds, humidity and pollution do not cause significant interference in signal collection, but the accuracy of the measurements can be affected by delays in the signal reaching the ground. These delays are normally associated with the behavior of the troposphere, the lowest layer of the Earth’s atmosphere. “These ‘noises’ are corrected by using the PSI (Persistent Scatterer Interferometry) technique, developed by scientists from the Polytechnic University of Milan and patented by the Italian company TRE as PSInSAR modelling,” says Paradella. To carry out this work, scientists worked with 33 images from the Carajás mines generated by the German satellite TerraSAR-X, which orbits the Earth at an altitude of 514 kilometers and passes over Brazil once every eleven days. If the interval between images were shorter, the chance of detecting deformation on the ground would be higher. Based on the good results obtained in the study, Vale is already planning to implement a new project in the field, this time focused on the mines in the Quadrilátero Ferrífero (Iron Quadrangle) area of Minas Gerais. This time, data from two satellites, the TerraSAR-X and the Italian satellite Cosmo-Skymed will be used. “From the standpoint of methodology, the study in the Quadrilátero Ferrífero, which is expected to last two years, will complement the Carajás study. Today, with the Cosmo-Skymed system, the interval for the satellite to return to the mine would fall to eight days,” says Paradella. With other future satellites, the trend is for the data acquisition period to become increasingly shorter, improving the ability to detect deformations in a shorter period of time.
Vale plans to use the DInSAR technique together with field radar and total stations with reflecting prisms, the two conventional methods of geotechnical surveying and monitoring in use today at its mines. “Our idea is for the method that uses satellite information to be added to the conventional field techniques. It will complement what we are already doing, with the advantage of being predictive. With it, we will be able to optimize our operations,” says geologist Luciano Assis, of the Vale Ferrous Planning and Development Department. Besides the innovative DInSAR, two other monitoring techniques are currently in use. Field radars, installed in the mine itself, send microwave signal pulses to a given target, for example, a bench, and then collect these signals back, comparing them to the signals previously sent. If any change in the phase is detected between the signals sent at different times, surface deformations may be occurring on the target. This type of radar is used at different observation points. In Carajás, Vale uses three SSRs (Slope Stability Radar) from the Australian company GroundProbe to monitor the mine sectors.
Total stations are instruments that measure vertical and horizontal angles and linear distances. They must be positioned at a location free from obstacles and be aimed at reflecting prisms, which are attached to a pole and placed over the location to be measured – in the specific case of Vale, the bench surfaces of the batters. A laser beam emitted by the total station is reflected back by the prism and returns to the station. An internal computer calculates the distances based on the response time and the angle measurements. The reflected signals are stored and compared to previously measured signals, showing whether the points moved closer together or farther apart where the prisms are installed. The limitation of this system and of field radars is that they only provide information on the locations where they are installed, unlike the DInSAR technique, which provides much broader coverage, encompassing the entire mining complex. “This study showed that the complementary and synergistic use of DInSAR together with the two traditional systems is the optimal configuration,” says INPE electrical engineer and researcher José Claudio Mura, assistant coordinator of the project.
In addition to generating knowledge that will lead to the adoption of a new technology in Brazilian mining, this study contributed to the education of graduate degree candidates in remote sensing at INPE: one student has completed a PhD and two others are working on their masters’ degrees. The project approved at FAPESP was conducted by a multidisciplinary team composed of geologists, cartographers, electrical and environmental engineers from INPE, the School of Sciences and Technology of Unesp Presidente Prudente, and the Geosciences Institute of Federal University of Pará (UFPA).
Project
Detection and monitoring of batter stability and surface deformations at an open pit mine using advanced differential radar interferometry techniques: an assessment at the N4 iron mine (Carajás) using data from the TerraSAR-x satellite (no. 2010/51267-9); Grant mechanism – Research Partnership for Technological Innovation (PITE); Principal investigator Waldir Renato Paradella (INPE); Investment R$626,273.35 and US$433,463.20 (FAPESP).
Scientific articles
Mura, J. C. et al. Monitoring of surface deformation in open pit mine using DInSAR Time-Series: A case study in the N5W iron mine (Carajás, Brazil) using TerraSAR-X data. Sept. 2014. Proceedings SPIE RS 2014. V. 9243. Oct. 2014.
Hartwig, M. E. et al. Detection and monitoring of surface motions in active open pit iron mine in the Amazon region, using persistent scatterer interferometry with TerraSAR-X satellite data. Remote Sensing. V. 5, n. 9, p. 4719-34. Sept. 2013.
Paradella, W. R. et al. Radar interferometry in surface deformation detection with orbital data. Revista Brasileira de Cartografia. V. 64, n. 6, p. 797-811. Dec. 2012.
