Early-Warning to the Impacts of Alluvial Mining on Sensitive Areas Using Earth Observation

The project is hosted and carried out under the auspices of University of Liège (GeMMe – Georesources & Geoimaging Lab) with the support of UN Environment and GRID-Geneva. It aims to provide the means for development agencies and authorities to identify priority areas to be preserved from the impacts of alluvial small-scale mining.

Researcher: Dr. Jingyi Jiang
Supervisor: Prof. Eric Pirard
Research Centre: University of Liège
Research partner(s): UN Environment

Project status: ongoing (2019-present)

Project Summary:

Development agencies and authorities active in studying, monitoring, planning, and creating policies related to the extractive industry have not yet fully tapped into the potential of Earth observation. The advantages of this technology are large coverage and temporal repetition of the acquisition. This allows integrated monitoring at a relatively low cost and with relatively high spatial resolution. Project EO-ALLert aims to provide the means for development agencies and authorities to identify priority areas to be preserved from the impacts of alluvial small-scale mining. It considers the case of gold mining in Colombia.

The project aims to provide impartial and integrated information based on stakeholder priorities and scientifically and geo-spatially based decisions. This information aims at suitable planning in the issuing of mining titles and licenses and efficient and well-designed interventions to stop illegal activities that may impact identified “sensitive areas”.

The applications of this work are not limited to Colombia. Small-scale and artisanal mining exist in many parts of the developing world and produce the majority of worldwide sapphire and about 20% of gold and diamond. While this type of mining provides livelihood to families in rural areas, it impacts the landscape, degrades the land, and can contaminate the food chain with heavy metals.

Project Objectives:

• Build on the CopX project to prominently transfer knowledge laterally between partners.
• Investigate stakeholder priorities to identify the characteristics of sensitive areas.
• Map land cover using a fusion of Sentinel-2 and Sentinel-1 based on “sensitive areas”.
• Create a spatio-temporal analysis of alluvial mining.
• Develop an automated product.

From blasting to tailings: integration of remote sensing and in-situ data for monitoring material streams in mining environments

The project is hosted and carried out under the auspices of the Helmholtz Institute Freiberg for Resource Technology. It aims to provide a continuous monitoring of mining activities using multi-temporal satellite data acquisitions.

Researcher: Dr. Louis Andreani
Supervisor: Prof. Richard Gloaguen
Research Centre: Helmholtz Institute Freiberg for Resource Technology

Project status: ongoing (2019-present)

Project Summary:

Each year the mining industry is extracting millions of tons of ore with low commodity/metal grades but also potential by-products that are frequently regarded as waste. The depletion of many in-situ mineral reserves, our needs to reach a circular economy as well as the environmental and legal barriers of access to primary mineral resources require a new paradigm of ore extraction.

Recent advances in extraction and processing technologies allow the recovery of complex materials but require adapted technologies and a precise characterization of the extracted products. The characterization of mining products and the geotechnical understanding required for extraction are usually based on information from in situ sampling, drilling, and geophysical and geochemical studies. An alternative solution would be to focus on a real time monitoring of material streams in mined areas. Such a solution would be cost effective and remote sensing techniques would provide a solid backbone for mapping both volumes and compositions. This would allow a detailed management of the extracted products as well as a record for the future re-processing of wastes. The potential market is rapidly growing as there is an increasing demand from mining companies for cost-effective tools dedicated to real time monitoring.

Satellite-based remote sensing offers a time-saving and cost-effective way of exploring mineral resources in support of mineral exploration and monitoring of mining activities. Unfortunately, there is a lack of solutions that can be directly used in the mining industry due to a lack of comprehensive tools for the non-specialist. Our objective is thus to develop an efficient and integrated monitoring workflow with the inclusion of multi-source remote sensing data from satellite to near-field scale and machine learning based characterization. Copernicus data includes sensors with particular characteristics that, combined, can provide valuable information for real time monitoring. On the other hand, the repetitive nature of the satellite orbits (e.g., Sentinel-1 and 2 constellations have a 2-5-day revisit cycle) can guarantee weekly acquisitions and the build up of time-series showing changes occurring on mining sites.

Project Objectives:

• Continuous monitoring of mining activities
• Composition and volume of material streams
• Environmental impacts

Copernicus for contamination monitoring and decision support in active mining and rehabilitation

CoCoMo is carried out and supervised by the Institute for Mine Surveying and Geodesy at the TU Bergakademie Freiberg. Support is provided by the EIT RawMaterials partner EFTAS and associated partners from the mining industry.

Researcher: Dr. Christian Köhler
Supervisor: Prof. Dr.-Ing. Jörg Benndorf
Research Centre: TU Bergakademie Freiberg
Research partner(s): EFTAS, MIBRAG

Project status: completed (2018-2019)

Project Summary:

Mining activities strongly influence the environment, e.g. by causing ground movements or producing contaminated waste dumps and water entities during the active operation and rehabilitation phases. Thus monitoring and managing the contamination is essential, as it reduces environmental impact and therefore increases social acceptance.

CoCoMo aims at the development of a contamination monitoring and decision support/early warning system for active and closed mining facilities. The project focuses on the vicinity of media pipelines of gas and oil storage caverns and waste dumps of active and abandoned mines. We target the integration and exploitation of multi-scale spatio-temporal data originating from Copernicus Sentinel satellites, unmanned aerial vehicles (UAVs) and ground based in-situ measurements.

This Remote Sensing (Sentinel, UAV) approach for environmental monitoring and early-warning systems, supported by ground based in-situ data, delivers spatially and temporally dense information. That, in turn, renders automatisation of the contamination analysis feasible, opening an avenue for cost effective and nearly real-time monitoring products.

Mine operators face the request of legal authorities for monitoring solutions as prerequisite for issuing operation permits. The primarily manual, terrestrial monitoring commonly in use today is costly in terms of money, time and man power, potentially dangerous for personal safety and, in addition, limited to small spatial areas and long measurement intervals. Thus, cost effective remote sensing-based monitoring solutions that deliver comprehensive data are highly desired by mine operators, engineering companies (offering monitoring and consulting services) or after care associations.

The public’s increased interest in mining related environmental impacts requires reassesment and readjustment of social acceptance – a considerable business risk in the mining sector. CoCoMo‘s open-source project results and publically-available Copernicus data increase transparency in communication between mining companies and public stakeholders (social and environmental initiatives, governmental bodies). Together with the education of future players in the mining industry through dissemination of results in the scientific community, this reduces environmental impact, increases social acceptance and ensures save and effective operation.

Project Objectives:

• Integrate multi-scale (Copernicus Sentinel, UAV, ground-based), multi- and hyperspectral spatio-temporal data into one comprehensive data product.
• Assess the potential of multidimensional spatio-temporal data for environmental contamination monitoring
• Development of workflows and algorithms for early, nearly real-time identification of potentially contaminated areas

Image 1: Redness Index (ri) as exemplary multi-spectral index in a region where contamination occurred. Characteristic changes in the temporal evolution of ri indicate potential contamination.

Image 2: Time series of the Redness Index (ri) as exemplary multi-spectral index in a region where contamination occurred. ‚Contaminated region‘ refers to the right red rectangle, ‚not contaminated region‘ to the left red rectangle in the figure above. Characteristic changes in the temporal evolution of ri indicate potential contamination.

Sampling optimization in stockpiles/tailings, for grade mapping of raw material using geostatistical analysis and Earth observation data.

The project is hosted and carried out under the auspices of University of Bologna (UNIBO) (Georesources Engineering Group) with the support of the “Geomatics” and “Geology” groups.

Researcher: Dr. Sara Kasmaeeyazdi
Supervisors: Prof. Roberto Bruno, Prof. Stefano Bonduà
Research Centre: University of Bologna
Research partner(s): Italian National Agency for New Technologies, Energy and Sustainable Economic Development (ENEA); National Technical University of Athens (NTUA); Bauxite processing plant-Greece (MYTILINEOS S.A.); Montenegrin Academy of Sciences and Arts (DANU).

Project status: completed (2018-2019)

Project Summary:

The mine industry produces millions of tons of waste with low grade contents of minerals, usually referred to as stockpiles/tailings. New technologies allow recovering valuable materials from these stockpiles/tailings coupling economic and environmental benefits. To obtain maps of the grade concentration of stockpile/tailing, survey sampling is necessary. Sampling can be an economical and time expensive phase and in general more samples are available, maps with higher accuracy can be provided. Hence, design of an optimum sampling plan is essential. On the other hand, data coming from Earth observation data like Copernicus land services should be joined with the collected in situ data, in order to improve map accuracy by reducing the more expensive in situ sampling methods. Hence, using both sampling data and Earth observation data, maps of grade and environmental indexes can be achieved. The raw material contained in stockpiles/tailings can be a potential economic resource. Economical evaluation and/or identification of hazardous areas of the resource can be obtained by maps of grade, using both in situ sampling method and Earth observation data. Optimal sampling plans can be designed by geostatistical methods. Several statistical tool can be used, like the estimation variance. The data sampling variogram model can be used for evaluating the estimation variance for a wide variety of possible sample patterns, without actually doing the sampling and hence to find the grid that just gives the required accuracy. Therefore, the geostatistical approaches with combined use of Earth observation data can lead to a high reconstruction accuracy maps with a given number of samples.

The objectives of the project are:

• Collection of information and data from two applications and selection of Greece bauxite residuals as the target case study.
• Integration of in-situ samples and Copernicus data (Sentinel-2 images) and performing Gaussian simulation using turning bands algorithm to map Aluminium, Iron and Vanadium variability within the bauxite residuals.
• Optimization of sampling with market and environmental scenarios.

Integration of space-borne remote sensing data with geophysical and geological data for exploration of IOCG mineralizations – IOCG-SENSE

The project is hosted and carried out under the auspices of Luleå University of Technology with the support of EIT Raw Materials. The aim is on development of exploration methods for iron oxide copper-gold (IOCG) class of deposits. IOCG is an important resource of minerals.

Researcher: Dr.Mehdi Abdolmaleki
Supervisor: Prof. Thorkild Maack Rasmussen
Research partner(s): Geological Survey of Denmark and Greenland, GEUS; Helvetica Exploration GmbH

Project status: completed (2018-2019)

Project summary:

The project establishes optimum procedures for integrating space-borne remote sensing data with other geoscientific data and information for Iron Oxide Copper Gold (IOCG) exploration. In focus of this project is Sentinel-2 data from the high Arctic, where problems of interpreting remote sensing data due to vegetation cover in general is at a minimum and where geological exposures are excellent.

The developed procedures are verified and evaluated based on data from Greenland. Greenland offers very good geological conditions with respect to ground truthing and has in addition excellent geological potential for securing the supply of critical raw materials. Areas selected in Greenland are (1) Inglefield Land in Northwest Greenland and (2) South Greenland. Inglefield Land has been subject to a number of investigations with the aim of providing data for mineral exploration. A combined airborne magnetic and transient electromagnetic survey was performed 1994 followed by detailed aeromagnetic and an airborne gravity survey funded by industry. Geochemical data are furthermore available together with results from geological mapping campaigns. The south Greenland area is covered with regional aeromagnetic survey data and detailed combined airborne magnetic. A large geochemistry data-set from stream sediments and rock samples are furthermore accessible.

There is in general a need among many universities to teach remote sensing techniques in order to ensure that students get properly introduced to the subject. The project shall exemplify possibilities by providing results from case studies.

The project shall contribute to securing the supply of raw materials to the society. Inclusion of Sentinel data analysis will allow a more focused exploration programme to be designed at a reduced cost compared to present day procedures. The Government of Greenland is actively promoting its mineral sector and the developments provided in the project is an important contribution towards reaching the goal of a mineral producing country.

Data integration is in general acknowledged as a key parameter for successful and optimised exploration, and this project shall ensure that procedures for extracting and utilization of information from remote sensing data are developed. The fact that Sentinel data has no depth penetration furthermore emphasises the importance of performing joint interpretation with data providing information at depth.

IOCG deposits are very important sources of raw-materials such as e.g. a source for copper, but the structural control and genesis of these deposits are far from being well understood. This project will provide additional input for a better understanding of these and related types of mineralization.

Project Objectives:

In general:

• Exploration of iron oxide copper-gold (IOCG) class of deposits.

In details:

• Establish optimum procedures for integrating space-borne remote sensing Sentinel-2 data with other geoscientific data and information for Iron Oxide Copper Gold (IOCG) exploration.
• Development and verification of new data feature extraction methods.
• The developed procedures shall be verified and evaluated based on data from Greenland.
• The experience gained from this project will be applicable in the development of a strategy for inclusion of Copernicus data in mineral exploration.

Project results/findings:

The output of the project can be listed as:

• Using Sentinel 2 (free satellite data) for extracting spectral features related by IOCG deposits is adapted for the first time in this field.
• The delivery of cost-effective, timely and comprehensive solutions for mineral exploration is considered a key driver for this project.
• Multi-source data integration has been applied (Remote sensing data, airborne magnetic and airborne electromagnetic data).
• Linking surface sensitive Sentinel data to geophysical data with depth information was done.
• Unsupervised Self Organizing Map (SOM) classification techniques is applied for discrimination lithological and structural.
• Fuzzy Analytic Hierarchy Process (FAHP) as a hybrid knowledge driven method is applied to calculate the weights of each evidential layer and then fuzzy operators are used to integrate them. Favorability map outlining high, moderate and low potential IOCG potential has been produced.
• The use of a knowledge-driven approach allows the treatment of systemic uncertainties in mineral potential mapping in a flexible and consistent way.
• High areas identified with high potential include previously mapped mineral occurrences, high concentration of Fe sulfide and soil samples of the study area.

Integration of Copernicus data in a multi-scale and multi-source exploration scheme

The project is hosted and carried out under the auspices of the Helmholtz Institute Freiberg for Resource Technology with the support of two other partners of the EIT RawMaterials community, Geological Survey of Finland (GTK), and Geological Survey of Denmark and Greenland (GEUS). It aims to create new and adapted concepts for a combined use of satellite- and drone-based multi-sensor observations and ground monitoring methods.

Researcher: Dr. Louis Andreani
Supervisor: Prof. Richard Gloaguen
Research Centre:Helmholtz Institute Freiberg for Resource Technology
Research partner(s): Geological Survey of Finland (GTK), Geological Survey of Denmark and Greenland (GEUS)

Project status: completed (2017-2018)

Project Summary:

Mineral exploration needs time-saving, cost effective, and, particularly in Europe, environmentally friendly and socially acceptable techniques to ensure sustainable access to raw materials in the EU. The objective of this research project is to develop an efficient and integrated exploration workflow with the inclusion of multi-source remote sensing data at satellite to near-field scale. Geological remote sensing offers a cost- effective way of exploring mineral resources but its full synergetic potential remains to be demonstrated. So far, a variety of multi- and hyperspectral, SAR, air- and satellite-based sensors are usually used individually to obtain qualitative and (semi-)quantitative information on the composition or structure of the surface. However, new methods are needed for geological applications as a consequence of the limitations of space-borne sensors with respect to their spatial and/or spectral resolution. We propose to overcome these problems by integrating satellite data, drone-borne data, geomorphic analyses and ground campaigns into new and adapted concepts for ground monitoring methods.

The objectives of the project are:

• To test the integration of satellite multispectral (Sentinel-2) and terrestrial hyperspectral data for mineral exploration.
• To test the suitability of different Machine Learning algorithms in order to define classification schemes adapted to mineral exploration.
• To create approaches that have high-performance in terms of environmental impact and cost.

Project results/findings:

• Sentinel-2 proves to be a powerful tool for mapping iron-bearing and alteration minerals. Specific mineral assemblages can be efficiently targeted.
• We have successfully tested different approaches, from the simple ones which provide a broad distribution of iron and alteration minerals but are fast to compute and require no prior knowledge to the more complex ones which allow to target specific minerals but require field data and knowledge.
• These approaches can find applications not only for mineral exploration but also for monitoring mining activities and their environmental impact.

Earth’s Critical Zone Early Warning System (CZ-EWS) by integrating SAR and seismic data in a mining context (Minas Riotinto, SW Iberia)

The project is hosted and carried out under the auspices of Institute of Geosciences (IGEO), Spanish National Research Council (CSIC) with the support of DARES (remote sensing) and Atalaya Mining (site owner). It aims to improve safe and sustainable supply of mineral resources by reducing ground stability risk.

Researcher: Dr. Ignacio Marzan
Supervisor: Prof. Jose Fernandez Torres
Research Centre: Spanish National Research Council (CSIC)
Research partner(s): DARES and Atalaya Mining

Project status: completed (2017-2018)

Project Summary:

This project focuses on the impact of the mining activities on the near Earth’s surface (Earth’s Critical Zone, CZ). Earthworks, digging and pumping affect ground mechanical integrity and may cause landslides, subsidence or runoff drifting; disasters that often have serious consequences, economically, environmentally and in terms of human life. As a consequence, public perception of the mining industry is usually negative. Our project aims to improve mining viability by guaranteeing ground/underground CZ integrity. Our strategy is to approach the target from two sides, from above by using satellite data, from below by permanent recording geophysical data. The solution of these two dataset should correlate in the soil surface. Our main outcome will be a CZ Early Warning System (CZ-EWS) protocol affordable for the mining industry, combining satellite and geophysical monitoring techniques.

Project Objectives:

Overall objective: to reduce the hazards of mining activities affecting the soil/subsoil mechanical integrity developing a monitoring system that combines SAR and passive seismic data.
Specifics objectives:

• Historical review of ground deformations in Riotinto mine by using SAR Copernicus data and correlating it with the mining company records.
• Assess the potential of Sentinel-1 for monitoring ground deformation at mining scale.
• Characterize underground stability by means of a local seismic network.
• Monitoring system combining Sentinel-1 and seismic data for ground stability hazards.

Project results/findings:

Overall objective: to reduce the hazards of mining activities affecting the soil/subsoil mechanical integrity developing a monitoring system that combines SAR and passive seismic

• The deformation field of the Riotinto mine area has been characterized using Copernicus data, and four main deformation zones are identified.
• Three mine waste deposits show evidence of subsidence. In the fourth zone, the abandoned open pit, evidence of instabilities is only shown in the eastern corridor limited by a fault zone, in good correlation with GPS observations.
• Preliminary results from the seismic network shows that the mining ambient noise can be used to monitor variation of the soil mechanics.

Spatiotemporal mapping of dust dispersion around mining sites using remote sensing

The project is hosted and carried out under the auspices of University of Liège, GeMMe – Georesources & Geoimaging Lab with the support of the “Remote Sensing for Mineral Mapping” group at CSIRO (Perth, Australia) and industrial partners of the EIT RawMaterials community, namely ERAMET and ArcelorMittal.

Researcher: Dr. Elsy Ibrahim
Supervisor: Prof. Dr. Ir. Eric Pirard
Research Centre: University of Liège
Research partner(s): CSIRO, ERAMET and ArcelorMittal

Project status: completed (2017-2018)

Project Summary:

MINeoDUST aims to explore the potential of Copernicus data in mine site monitoring and especially in mapping dust dispersion patterns around active sites. Previous studies of similar problems in Australia and in Europe have been performed with airborne hyperspectral data (HyMap) or discontinued spaceborne data (ASTER) offering both higher spatial and spectral resolutions. Nevertheless, the proven potential of Sentinel 2 for mapping iron oxides and the use of the most recent multi-sensor fusion/classification techniques open a window of opportunities to serve the raw materials community. This project also considers the high revisiting time of the Sentinel 2 programme as another very interesting asset when it comes to mapping the dynamic of dust dispersion around extractive sites, but also transportation (rail, road) and loading sites (ports). MINeoDUST will be hosted by the University of Liège and more specifically the GeMMe – Georesources & Geoimaging Lab. This lab is a member of EARSEL and has a team made up of both field geologists and digital imaging specialists. ULiège also hosts the famous CSL (Centre Spatial de Liège), one of ESA’s four test sites for satellites with specialization in optics and imaging, offering a unique opportunity for this project to also come up with recommendations for future missions. The project builds on a close partnership with the “Remote Sensing for Mineral Mapping” group at CSIRO in Perth (Dr Ramanaidou – Commodity Research Leader Fe and Lateritic Ni). This group is currently collecting field data on test sites in New Caledonia. Two major mining companies, Eramet, and ArcelorMittal, have already provided letters of support to the project, and confirming that other EIT RawMaterials industrial partners with test sites in iron, manganese, bauxite and Ni-laterite mining will be involved.

Project Objectives:

• Assessment of the potential of combined Sentinel 2 and other sources of remotely sensed data like LANDSAT to evaluate the spatiotemporal distribution of iron oxides containing dust particles
• Development of fusion and classification methodologies to take full advantage of the combined Sentinel 2 and other sources of remote sensing data
• Definition of future specifications for spaceborne missions to address the needs of environmental monitoring of mining sites, mineral transportation sites (railway, ports, etc.) and storage sites
• To explore the potential for sharing remote sensing data with all stakeholders of a mining project through open data platforms (Map-X a joint UNEP-WorldBank initiative for the extractive industry)