Daniel Elroi
Knight Piésold and Co.
1600 Stout Street, Suite 800
Denver, CO 80202

Introduction

Geographic Information Systems, or GIS, is a new "buzz-word" for the '90s, and not just in the computer industry. GIS software packages combine the power of database management systems (DBMS) with the benefits of computer-aided drafting and design (CADD) software, to create a versatile new way of managing and analyzing data. By adding the spatial dimension to data, GIS is well on its way to becoming a household term. Since the majority of data pertaining to a vast number of human endeavors relate to the location of objects and events, GIS has the potential to affect an almost unlimited range of disciplines.

As such, GIS may have great ramifications for the mining industry, and it is the purpose of this paper to discuss the principles of GIS, and to expand on its application to the production and post-production phases of a mine. This paper also explains some of the rudiments of GIS implementation, and discusses some of the potential obstacles specific to the mining industry. This paper does not portend to address the exploration and development phases of a mine's life span, nor does it deal with sub-surface aspects of an underground mine. Rather, it concentrates on the processing of a mine's output, its disposal, and the ultimate closure and reclamation of the mine site.

What is GIS and how is it different from CADD and database management software

GIS is a type of software which combines the capabilities of computerized mapping software and database management and analysis software. Conceptually, most GIS packages do so by first characterizing and recording all spatial phenomena as points, lines, or polygons in space. Examples of points might include monitoring wells, electricity poles, and spot elevations, while examples of lines might include roads, streams, and contours. Polygons might include soil groups, parcels of land, or the extent of waste rock disposal areas. In GIS the location of each feature is recorded in the computer as a series of coordinates in some recognized coordinate system (e.g. State Plane, Longitude-Latitude, or UTM), and a standard map projection (e.g. Mercator, Lambert, or Albers). Then, after each feature is recorded, GIS permits users to attach any amount of information about that feature in a linked database. Thus for each stored land parcel can be recorded its owner's name, address, and telephone number, each lessee's and lessor's relevant information, the assessed value of the land, the governing jurisdiction, and so on. Each groundwater monitoring well can have attached to it historical records of all the chemical constituents of water samples, their temperatures, pH factors, and depths below the surface.

This first attempt to define GIS should obviate the distinct differences between stand-alone CADD or database management software packages and GIS. Neither type of software can effectively store and display both tabular and graphical data. A GIS, on the other hand, can easily display maps based on the pre-selection of tabular data, such as displaying a map of all the mineral ownership leases about to expire in the next six months, or a map that shows all the monitoring wells whose cyanide content has exceeded acceptable levels in the past month. Likewise, a GIS enables users to select which parcels of land need to be leased or purchased based on their exact distance from a new mine or from a new planned access road.

A second important aspect of GIS, which can only be duplicated in a very primitive way with mere database management systems, is the ability to overlay different layers, or themes, of spatial data in order to analyze the relationships between them. For example, consider the effort required to produce a new map which depicts all areas around an existing processing mill which are: a) within 300 foot elevation difference from the elevation of the mill; b) overlying impermeable soils; c) are away from known faults; and d) are owned by the owners of the mill. Such a map, which might be required in order to site a new tailing facility, for example, might take days if not weeks to produce by manual methods or even using CADD. The level of effort is vastly increased if the source maps come in a typical assortment of sizes and scales. Once this data is entered into a GIS, however, such a task can take as little as a few minutes. Better yet, the criteria used to produce this map can very easily be changed (e.g. a 400 foot elevation difference), and a new map created. To extend this example further, suppose that the land required for the new facility and for access roads may be purchased or leased by the owners, if it does not already belong to it. Prior to GIS, comparing the total costs of construction versus land acquisition would have required complicated computations based on the use of a planimeter or CADD tools to measure the amount of land involved with each scenario. If the various possible locations of the new tailing are drafted into a GIS, though, those computations can be performed in a manner of minutes. And whether the measurements are made by one operator or another, they will remain the same, because they are not based on human interpretation, but rather on actual real world coordinates.

One of the attributes that can be assigned to geographical features is elevation. Therefore, most GIS packages have the capability to generate three-dimensional surface models and compute volumes. This capability is not unique to GIS. In fact there are powerful packages available that do little else but compute surfaces and volumes. The advantage of incorporating these capabilities into GIS is that the data that are used to generate these surfaces and volumes can be selected interactively from a computerized map, and that the results of such analyses can be translated back into maps, or integrated with other GIS layers. For example, a GIS database may contain contour information, spot elevations, seismic surveys which indicate the depth to bedrock, and borehole information also indicating depth to bedrock. Computing the volume of soil and loose aggregate available for excavation, as an example, requires the "development" of a three dimensional surface for the ground surface, and one for the bedrock surface, before volumetric computations can be made. Since not all depth and elevation data are of equal precision and reliability, GIS can be used to interactively select the data points used to create these surfaces, until the best possible surfaces are generated. Likewise, the results of the computations can be easily translated back into two-dimensional maps depicting the areas which will actually be excavated.

Two methods have been discussed so far by which GIS can view spatial data: as points, lines, and polygons, and as three-dimensional surfaces. An increasing number of packages also offer the option of viewing spatial phenomena in a grid-cell, or "raster", format, in which the "world" (at least that portion of the world considered in a particular database), is divided into a grid of equally shaped cells. Each cell is then coded with a single primary code, depending on the theme of the map. For example, in a soils map, a cell will be coded as either clay or sand, but never both. This particular method of representing spatial data is especially well suited for modeling environmental data. For example, drainage basins have automatically been delineated and the plumes of pollution in groundwater have been studied with this method.

GIS can represent the same spatial data in a variety of ways, each with its pros and cons in terms of the type of analysis and mapping that it is best suited to produce. But regardless of the method of storage, GIS data are stored in real world coordinates. Some of the benefits of this feature are that data from a variety of sources and in a variety of scales can easily be integrated with each other and compared for compatibility. A by-product of GIS spatial data handling capabilities is its strong ability to aid in quality controlling data. Two adjacent paper vegetation maps, for example, may give no indication of incompatibility. However, once digitized into a GIS, placed in the same coordinate system, and then superimposed on each other, even slight positional mismatches can easily be detected, which may prevent costly errors in the future. Also due to the fact that GIS data are stored in real world coordinates, the concept of scale (other than of the original source data), does not exist. This means than no restrictions exist on the scale at which data are viewed or plotted. A dozen adjacent map sheets, which may be difficult to view as a single entity in paper format or in a CADD system can be easily displayed in a GIS. As a side note to the issue of scale, unlike CADD, GIS annotations are not tied to a particular map. Therefore, whether zoomed in or zoomed out on a particular map, the text does not enlarge or reduce, but rather remains the same size as the one specified by the operator for that map.

The advantages of storing spatial data as real-world coordinates cannot be overemphasized. This feature enables GIS to easily translate data between various projections and coordinate systems, thereby making GIS a powerful data integration tool. A single GIS database might contain layers that have been digitized from paper maps, digitally converted from CADD drawings, automatically generated from spreadsheets (which contain coordinate data), scanned from documents, captured by remote sensing devices on satellites, or photographed from an airplane. This ability alone can sometimes justify the use of GIS, even if the GIS is not asked to perform any analysis with these data.

Having so defined GIS, it is important to clarify what exactly GIS is not. GIS is not synonymous with CADD. CADD software packages are designed specifically to handle any graphical data (not just spatial data), and to perform analysis with them which are often far from spatial analysis (material and structural performance, mechanical motion modeling, and systems control are examples that come to mind). Nor are graphical data usually stored in real world coordinates in a CADD system. Rather they are stored in the units in which they will be plotted on paper, or multiples of these units. Thus a CADD operator is really always manipulating the final plot on the computer, and then sending it to the plotter. A GIS operator, on the other hand, is usually working with simple graphical representations of spatial phenomena in real world coordinates on the screen, and only translates these into appealing maps and plots just before submitting them to the plotter. Thus colors, patterns, annotations, and other graphical communication tools are not a part of the database, but of the cartographic process associated with turning spatial database layers into maps.

GIS is also not synonymous with graphics or desktop publishing software. These packages too are concerned with the final plot, and the operator, very often an artist, manipulates the digital representation of the final plot at all times. Thus, although maps and plots produced by GIS may often appear beautiful or artistic, GIS is not an artistic software.

Fortunately, GIS packages usually come supplied with a variety of data translation tools which enable them to exchange data with CADD, graphics, desktop publishing, spreadsheet, database management, and word processing software. Such terms as DXF, TIFF, BIP, CGM, ASCII, and IGES are frequently used in conjunction with discussions of data exchange. Thus aerial photography, for example, can be photogrammetrically converted into contour maps using digital photogrammetry devices, then enhanced or manipulated in CADD, analyzed in GIS, and included in a report produced with a word processing software. Although this process may sound more convoluted than most firms or agencies might wish to engage in, the process described above is an extreme example, but one which is nevertheless viable, and which can yield substantial benefits, especially when associated with repetitive operations.

The aspect of repeated actions is key to the benefits of GIS. As with many software technologies, GIS can yield substantial benefits as outlined above, due its unique tools and method of viewing spatial phenomena. Where GIS truly benefits its users, however, is in automating repetitive processes. GIS packages usually fall under the "toolbox" classification of software. Which means that they contain various commands that enable users to perform discreet actions, which if strung together in a logical progression enable users to execute a task. In this aspect GIS is similar to word processing. Word processing software contains tools to enter, erase, and move words, to center paragraphs, and make text appear bold or italic. Word processing software, however, does not write letters, books, or reports. People do. This too is the case with GIS. The difference between the two, though, is that GIS can be trained to perform the same actions with different data with the aid of macro programs. In this way GIS can be customized so that it can be used by engineers, scientists, and managers, rather than by GIS specialists alone. Thus, for example, a GIS specialist will be required to use the GIS tools that select land parcels by value, that create all the elements of a map, that insert the desired parcels into the map, that generate a report relating to the value of the surrounding parcels, and that place that report on the map. However, if this action needs to be performed on a repeated basis for various parcels, or with updated assessed values, then the GIS can be customized to perform that "application" in a simple, user-friendly fashion, so that the actual user of the data, such as a manager, can with no more that a few key-strokes ask the GIS to generate that map. This places powerful tools in the hands of the professionals who need them most. In the sections that follow, several examples of GIS applications are given as they relate to the processing of a mine's output, the disposal of its wastes, and the ultimate remediation of the mine site itself.

Applications of GIS to the production phase of a mine

Whenever new software is introduced on the market, these questions are frequently asked: "Why do we need this software if we have managed to do without it this far?" and "Is this new way better than the old way?" These are fair questions and ones which ought to be asked. Anyone observing others or themselves start up a word processing software, creating a new file, typing, sending to print and then retrieving from a printer a mere memo, which could have been scribbled on a piece of scratch paper in a fraction of the time, or anyone who kicks themselves for replacing all the office typewriters with computers when it comes time to fill out some important forms, understands the importance of these questions. What follows is by no means a comprehensive list of GIS applications, but rather one which should stimulate answers to these questions.

Site selection

An example of how GIS might be used to select a new site for a waste rock disposal area or a tailing facility has already been given earlier in the paper, where various thematic layers are first manipulated singularly (e.g. selecting alternative elevations from the topography layer), and then combined by overlaying them on each other. This methodology can also be applied to siting such things as new roads. In addition to overlaying layers of information, GIS can also be queried for the distance between various features, their areas and volumes. Thus, once potential sites for a tailing facility are identified, the volume to surface area ratio of the sites can be computed and compared. Also, buffers can be created around selected features, such as ecologically sensitive areas, to identify zones of no interference. Other GIS tools which can be brought to bear in the process are the ability to compute the slope angle and direction of the surface, and the ability to compute the inter-visibility between points. Potential objections to the expansion of a facility based on the visual impact of the construction on a nearby resort, for example, can be eliminated by determining that the expansion (e.g. a tall stack) will not be visible from the resort. Other factors which play into the process of site selection such as property ownership, lease holding, and mineral rights, can also be successfully managed within a GIS database.

Environmental Quality Monitoring

Environmental quality indicators such as water and air pollution levels are often recorded in large spreadsheets, which are then all too often archived and stored, but not used. When they are used, they are either analyzed with standard spreadsheet tools, or imported into specialized software tools. These analyses ordinarily expose variations in pollutant levels at a particular site over time in the form of a graph or a chart. The integration of these data into a GIS enables the analyst to query monitoring locations on digital maps, as well as to display the results of spreadsheet or database type queries on a map. It also permits the creation of interpolated surfaces or plumes, indicating the rising and falling levels of the pollutants through space. Such horizontal slices across the data can be stored as snapshots in time, and then displayed in sequence to help identify trends which might indicate the spreading of a plume. This particularly applies to water-borne pollutants. Air-borne pollutant analysis can benefit from GIS by overlaying interpolated plumes on layers showing concentrations and characteristics of human populations. This can be used to assess risk and in certain cases enable an appropriate and rapid response to emergency situations. In either case, GIS can also be helpful in predicting where and when various pollution levels might exceed permissible levels, which can mitigate costly remediation activities under emergency situations.

When a GIS is used to view and analyze environmental data, it often becomes apparent that the existing distribution of monitoring stations is inadequate. This can then serve to aid in the placement of new stations, which would result in better coverage and more reliable plume surfaces.

Volume Computations

Assuming that a topography layer was created in a GIS prior to the commencement of production, GIS can be used to compute the increasing volumes of waste rock heaps, and in the case of open-pit excavations, the volumes removed. This is done with periodic flights over the mine site, and the photogrammetric conversion of the resulting photographs into new layers, which when compared to older layers using the volumetric tools of the GIS, yield the volume differences between the old and new layers.

Applications of GIS to the post-production phase of a mine

Vegetation Characterization

The remediation of an old mine site can take several forms. These include the returning of the site to a state that most closely resembles the original lay and appearance of the site prior to development, as well as the conversion of the site to a new appearance and use, as in turning an old rock quarry into a recreational lake. When a site is to be returned to as close to a natural state as is possible, it is preferable to obtain a careful record of the local biomes, preferably prior to the development of the site. Since this is usually not possible, it is important to characterize the surrounding vegetation and wildlife, so as to best predict which remediation plan to embark upon. For example, it is important to match vegetation and top soil not only to the slope of the reclaimed land, but also to its aspect. It is a fact that vegetation zones (e.g. deciduous forest, alpine meadows, etc.) occur at lower elevations on north facing slopes than they do on south facing ones. Therefore, an attempt to plant vegetation of the same type at the same elevation on both sides of a valley may prove troublesome. GIS is ideal for storing and analyzing vegetation and wildlife characteristics.

Slope-Aspect Characterization

In addition to determining the correct biome based on the slope and aspect of the land, as explained above, slope and aspect characteristics also indicate the levels of erosion control efforts required at any particular portion of a site. Slope and aspect calculations are direct derivatives of the surface modeling tools built into most GIS packages. Coincidentally, these two characteristics have been used in GIS to determine the amount of insolation (duration and intensity of sunlight), which are also important determinants in the success or failure of any particular remediation plan. Occasionally slope and aspect are important when planning a new use for an old mine site. For example, a recreational lake is of limited use if accessing the lake poses dangers due to steepness or surface instability.

Volume Computations

No less important in the post-production phase as in the production phase, the ability to compute volumes is essential to a good remediation plan. Most remediation plans require considerable volumes of soil to be moved, and where these can be obtained locally, the cost of remediation is reduced. Thus it is important to determine the amounts of soil which are available for removal nearby. This also applies if concavities need to be filled in with crushed rock, or if quarries need to be filled in with water.

Visualization

One of the more important aspects of site remediation planning is becoming the process of "selling" the plan to the governing bodies as well as to the public, and this is more and more frequently done with the aid of visualization tools. To be convinced, people wish to see what the planned remediation will look like when it is complete. By itself, GIS can produce impressive visual simulations of proposed remediation plans. Together with graphics packages the results can be truly astounding, and can often make very strong statements in and of themselves.

Implementation of GIS

Although the term GIS has been used thus far exclusively to mean software, the implementation of GIS in fact comprises of many interrelated components. The most tangible ones, and the ones most often thought to constitute the major expense in the implementation of GIS are hardware and software. Hardware can consist of as little as a mid-powered PC to a very high powered engineering workstation (UNIX being the most popular operating system for such). Fully configured GIS shops, however, also include large format digitizing boards, color plotters (pen-based or electrostatic), printers, various computer peripherals such as tape backup devices, and occasionally scanners and other esoteric equipment. The part of hardware and software in a GIS installation, for a single operator, can cost anywhere between $5,000 and $50,000. This cost, however, is not the most significant portion of GIS implementation. The collection of data and its placement in the GIS database, the hiring and training of personnel, and the creation of custom GIS applications for the purpose of bringing GIS technology down to the level of everyday use, as well as the nearly constant upgrading that is associated with computer technology, far outstrip the expenses associated with hardware and software.

The recognition of these costs is obviously not enough to deter organizations from implementing GIS, as is evidenced by growth patterns in the GIS industry which match, if not exceed, nearly all of the fastest sectors of computer technology today. The implementation of GIS seems to sweep across whole sectors of the public and private portions of our society as soon as a critical mass of organizations in that sector implement GIS. Thus, for example, the number of county governments, and federal agencies which have not implemented GIS is shrinking rapidly. Package delivery companies, the likes of UPS and Federal Express, have taken to GIS in rapid succession. So what catalyst can make GIS technology take hold in the mining industry?

There are many reasons which in an ideal world would justify the implementation of GIS technology at a mining firm. But such seemingly logical reasons as long term savings and increased worker productivity do not always work in the real world. First there is fear of the unknown which translates to perceived risk in investing in this technology. Second there is the need to retrain, replace, or hire personnel to manage this new technology. Associated with that is the momentum associated with an entrenched workforce and with established routines and practices. There is the preference for quick returns instead of long term benefits. The list continues.

The types of catalysts that do seem to motivate the average firm often appear external. For example, more and more regulatory agencies, lending institutions, and consultants use GIS, and either require compatibility, or at any rate make it difficult to function without it. Competition with other firms, especially among consultants, tends to force some firms into the technology race. Lawsuits are another important reason firms obtain a GIS, whether to assist in the defense of property-related claims, for example, or to mitigate the consequences of a pollution- related case. It is in situations like these that the bottom line of implementing a GIS seems relatively insignificant, and a long term cost-recovery program is dispensed with. These are not ideal ways to enter this technology.

Fortunately, internal motivations do prevail on occasion. Usually a single forward-thinking and influential individual realizes the vastly increased number of scenarios that can be modeled with a GIS. Or the level of frustration and wasted effort associated with retrieving and analyzing vast amounts of environmental monitoring data. Or the superior performance of a GIS-equipped consultant. Then this individual is sometimes able to sell the idea to the rest of the firm. However, even should a firm have the good fortune to have reached the point of wanting a GIS, there remains the question of how best to implement it.

One excellent way to implement GIS is to wisely select a project which has to be carried out as part of the normal operation of the business, such as preparing an Environmental Impact Report or Environmental Impact Statement, or getting new mapping for a basemap. Then a relatively small expense can be added to at the very least make the database which results from the project compatible with GIS. This does not necessarily mean that it has to be carried out with a GIS, or even that a future GIS has to be selected at this stage, although of course each of these steps would bring about economic benefits at a later time. If GIS-capable consultants are selected to perform the work, then they ought to be using a GIS package that the firm is potentially going to want to purchase at a later date. An off- brand or home-grown system employed by the consultants will probably not be a wise investment. If the consultants are using GIS, as much technology-transfer should occur as is affordable. For example, the database created by the consultants should be meticulously designed, quality controlled, and documented. If the same analyses performed by the consultants might one day be performed internally by the firm, should they obtain a GIS, then the consultants should be encouraged to clean up and document any of the custom programs they wrote to help them in the their work, in such a way that they could later be transferred to the firm. Learning should occur from the experience, even if the consultants must be paid a little extra to explain to the client firm exactly what they are doing.

If a project is used to set a firm on its way to GIS implementation, the next step might be, if the firm is yet unwilling to commit to the technology, to ask the consultants to continue maintaining the database, and producing analysis or maps from it on demand. This would be akin to renting GIS time, and would alleviate the firm from having to obtain hardware, software, and experienced personnel until the technology proves itself worthy.

At the point the firm is ready to install its own GIS, one of the common mistakes is to buy the hardware and software, and just try to figure it all out. Beyond learning the intricacies of the tools, GIS requires at least a certain level of understanding of spatial analysis principles. These can be learned with time, but the use of a GIS consultant to set things up, train staff to perform those tasks of greatest urgency to the firm, and to customize the software so that it performs money- or time-saving tasks right away, is invaluable. Communication and education are probably the next most important tasks. All too often a GIS is purchased and then left untouched because people do not know about it, people do not understand it, people do not want to share it, or people do not know how to use it. Only good communication and education can avoid these pitfalls. Another realization that is important to make among management, is that the implementation of GIS often brings in its wake a rethinking of many of the functions of the firm and its employees. Reducing redundancy in data also reduces redundancy in people. And the acceleration of data update and transfer, often unearths illogical or at least inefficient methods of data handling which had entrenched themselves into the organization. Therefore, capitalizing on the potential benefits of GIS can mean as little as or as much as an organization is able and willing to undertake.

Conclusions