Contents lists available at ScienceDirect

Engineering Geology

journal homepage: www.elsevier.com/locate/enggeo

Establishing relationships between structural data from close-rangeterrestrial digital photogrammetry and measurement while drilling data

Sohail Manzoor⁎, Samaneh Liaghat, Anna Gustafson, Daniel Johansson, Håkan SchunnessonDivision of Mining and Geotechnical Engineering, Luleå University of Technology, Luleå, Sweden

A B S T R A C T

Geologists, mine planners, geotechnical, and mining engineers always strive for maximum information to get a better insight of the rock mass before interacting withit. Over the recent decades, close-range terrestrial digital photogrammetry (CRTDP) has been increasingly used for data acquisition and to support the conventionalmethods for rock mass characterization. It provides a safe, time-saving and contact-free way to gather enough data to minimize user dependent biases. However, itrequires an expensive camera, fieldwork and some software to extract the information from images. In addition, it can over-estimate the rock fracturing sometimesdue to weathering of the rock face or poor blasting practices. Measurement while drilling (MWD) data include the responses of different drilling parameters to thevariations in the rock mass. MWD data are produced in large quantity, as they come from every hole drilled. These data correspond to the inside variations of rockrather than the surface ones counted in photogrammetry.

In this paper, structural data are obtained from different bench faces of an open pit mine using a commercial software package, ShapeMetriX3D (by 3GSM). Thesedata are compared to the MWD data of the boreholes that were blasted to produce these bench faces to establish certain relationships between drilling parameters androck mass structures. Half casts of the boreholes with MWD data were visible on the bench faces of the pre-split wall that allowed a better correlation. The resultsshow abrupt changes in MWD parameters for open joints or cavities with some infilling material and overall increases or decreases in parameters for closely spacedbedding planes, fractures or foliations. The results are promising and suggest the method can be used to characterize the rock mass, modify the charging of explosivesin blasting operations and facilitate the geological modeling of the rock mass.

1. Introduction

Weak zones or discontinuities within a rock mass should be in-vestigated carefully in any rock engineering or geotechnical project.Knowing the physical and mechanical properties as well as the geo-metry of the rock mass is of prime importance in rock engineering asthese traits control rock failure mechanisms (Firpo et al., 2011; Gigliand Casagli, 2011; Chen et al., 2016; Buyer and Schubert, 2017;Riquelme et al., 2017). Without sufficient understanding, any sort ofrock failure analysis is prone to a high degree of uncertainty.

Physio-mechanical properties are usually determined from in-situ orlaboratory tests, while geometrical parameters are measured in the field(Firpo et al., 2011). Traditional methods for determining the geome-trical characteristics of the rock mass in the field require a geologicalcompass and physical access to the measurement locations (Gaich et al.,2006). Methods include one-dimensional scanline surveys and two-di-mensional window surveys (Sturzenegger and Stead, 2009a). Thesesurveys are typically performed by a geologist or field engineer andmay suffer from human error or measurement bias (Ferrero et al., 2009;Gigli and Casagli, 2011). Other constraints include safety issues, accessand time restrictions, difficulties in reproducing the results, limitedcoverage in large target areas, lack of user experience etc. These

constraints make traditional methods unpopular with users (Kemenyand Post, 2003; Ferrero et al., 2009), and they are being replaced byremote sensing techniques.

In the last decade, digital elevation modeling and geomorphologicalterrain analysis have undergone an evolutionary transformation(Westboy et al., 2012). Several techniques are now available to de-termine the geometrical properties of the rock mass without physicalaccess, thus increasing user safety and overcoming access and timerestrictions. These new techniques allow the user to determine thegeometry of the rock mass by developing 3D geometrical models of therock surface and analyzing them using specially developed computersoftware (Feng et al., 2001; Haneberg, 2008; Sturzenegger and Stead,2009a, b; Bonilla-Sierra et al., 2015; Riquelme et al., 2015). Althoughcontact-free techniques date back to the 1960s when researchers usedterrestrial photographs for geological mapping (Linkwitz, 1963;Rengers, 1967), image-based measurement has advanced rapidly inrecent years (Roberts and Poropat, 2000). Also, laser scanners or acombination of laser scanners and digital images can be used to con-struct the rock face (Kemeny et al., 2003; Feng and Roeshoff, 2004;Lemy and Hadjigeorgiou, 2004).

Today, photogrammetry and laser scanning are the most widelyused techniques for rock face characterization (Ferrero et al., 2009).

https://doi.org/10.1016/j.enggeo.2020.105480Received 27 March 2019; Received in revised form 4 November 2019; Accepted 13 January 2020

⁎ Corresponding author.E-mail address: sohail.manzoor@ltu.se (S. Manzoor).

Engineering Geology 267 (2020) 105480

Available online 13 January 20200013-7952/ © 2020 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/BY-NC-ND/4.0/).

T

Each has its own advantages and disadvantages (see Kemeny and Post,2003; Jaboyedoff et al., 2009; Sturzenegger and Stead, 2009a; Latoet al., 2010; Fisher et al., 2014; Li et al., 2016), but both can overcomethe limitations of traditional methods and have other advantages aswell, including the availability of excessive data, more objective results,the ability to work in magnetic environments, etc. The mining industryhas specific problems with geomechanical data collection, for example,limited exposure and unsafe rock face conditions. Therefore, a safe, fastand contact-free data acquisition technique is always preferred over atraditional geological mapping technique (Lemy and Hadjigeorgiou,2003). ISRM (1978) suggests photogrammetry can be a viable alter-native to traditional methods in potentially dangerous and inaccessibleareas.

In the last two decades, aerial and close-range terrestrial digitalphotogrammetry (CRTDP) have been widely used for 3D topographicmodeling (Remondino and El-Hakim, 2006; Matthews, 2008; Fraserand Cronk, 2009; Salvini et al., 2013; Assali et al., 2014). “Close-range”denotes a distance of up to 300 m from the camera/scanner location tothe analyzed object (Wolf and Dewitt, 2000). Photogrammetry is veryuseful in determining 3D structural parameters from stereoscopicimages (Vasuki et al., 2014). It has been applied for many years both insurface and underground mining operations (Fekete et al., 2010), but itis becoming even more powerful with advances in digital cameratechnology, accompanied by the use of unmanned aerial vehicles(UAVs) for digital photography in remote areas (Bemis et al., 2014).Use of UAVs can considerably improve data acquisition in inaccessibleareas where other techniques face various limitations (Salvini et al.,2017; Francioni et al., 2018). However, photogrammetry requires anexpensive camera, fieldwork that can hinder the production cycle inmining operation and a software to extract the information from theimages. In a weathered rock face or if blasting practices are poor, thetechnique may over-estimate rock fracturing.

Measurement while drilling (MWD) is another useful technique forgathering data to characterize the structural and mechanical propertiesof the penetrated rock mass (Monteiro et al., 2009; Khorzoughi andHall, 2016; Rodgers et al., 2018; Vezhapparambu et al., 2018). Itmonitors drilling parameters like penetration rate, rotation pressure,percussive pressure, feed pressure, flush pressure and damping pressureas the bit penetrates into the rock mass. The technique was introducedby Schlumberger Ltd. in 1911 to remove the blindness of drilling op-erations in the oil industry; later, in the 1970s, it was applied in themining industry to improve various mining operations (Segui andHiggins, 2002) and is considered a low cost approach to get high re-solution data in less time than other subsurface exploration techniques(Khorzoughi, 2013). Moreover, since data are recorded during thedrilling operation, the technique does not slow down the productionprocess. It permits fast and accurate insight into the geo-mechanicalparameters of the rock mass, far in advance of the downstream pro-cesses (Rai et al., 2016).

When combined with other exploration methods, the MWD tech-nique provides an improved description of the subsurface rock prop-erties (Segui and Higgins, 2002). It has been used to identify rockfractures (Schunnesson, 1996) and characterize the rock mass(Khorzoughi, 2013; Van Oosterhout, 2016; Ghosh et al., 2017;Khorzoughi et al., 2018; Van Eldert et al., 2019), but the results haverarely been validated by the established methods of rock mass char-acterization. This study aims to establish relationships between rockstructural data and MWD data using the well-established CRTD pho-togrammetry to harness the maximum benefits associated with MWDtechnique. It uses stereoscopic images of nine different bench faces fromthe Luossavaara Kiirunavaara AB (LKAB's) Leveäniemi iron ore mine toperform a structural analysis of the face using the commercial softwareShapeMetrix3D (by 3GSM). MWD data (penetration rate, percussivepressure, feed pressure and rotation pressure) collected from theboreholes before the exaction of these benches are analyzed in MATLABand Rock Manager (by Epiroc) to detect variations in drilling

parameters. Results from the analyses are compared and correlated toestablish relationships between them.

2. Methodology

This research is based on a literature review, close-range terrestrialdigital photogrammetry and measurement while drilling data. Datawere collected from Luossavaara-Kiirunavaara AB's (LKAB's)Leveäniemi open-pit mine in Sweden.

2.1. Site description

The Leveäniemi iron ore mine is located in Svappavaara, an oldmining village in Kiruna Municipality in the northern part of Sweden. Itis an open-pit mine, owned and operated by LKAB. The Leveäniemimine was first operated from 1964 to 1983. After that, the mine wasclosed and flooded. 1n 2015, the company received a permit to re-openthe mine, and the water was pumped out. Today, the planned annualproduction is 12 million tonnes of magnetite iron ore with an averagegrade of 44%. There are 15 years of remaining mine life as per reserveestimates. Because of the varying nature of the ore quality, both shortand long-term plans have been made using deposit models to maintaina good balance of iron concentration for the processing plant.

After blasting, the waste is dumped in a nearby dump area, and oreis hauled to the crushers. The crushed material is processed further toremove the waste from the ore. The processing phase has three stages:sorting, concentrating and pelletizing. Magnetic separation is used insorting and concentrating phases to separate iron from the waste. Thefinal product in the form of pellets is transported to customers forprocessing into high quality steel.

The study selected one wall of the mine with poor rock conditionsand one with good rock conditions and pre-split blast holes (see Fig. 1).The wall with poor conditions was selected to check the response ofMWD data to highly fractured rock, and the wall with pre-split blastholes was selected for the following reasons:

i. It was intact with minimum blast-induced fractures because of pre-split blasting;

ii. Half casts of the pre-split blast holes were identified on the wall,permitting improved correlation with MWD data.

2.2. Close-range terrestrial digital photogrammetry

CRTD photogrammetry includes image collection, 3D model gen-eration and structural analysis. In the data collection phase of thisstudy, stereoscopic images of nine different bench faces were takenusing a DSLR camera following the standard procedures prescribed by3GSM, 2010. The basic steps involved are:

1. Setting up of control or reference points near the bench face in theimaging area.

2. Taking stereoscopic image pair i.e. two images of the bench facefrom different positions.

3. Generating a generic 3D model of the bench face from the 2Dimages.

4. Scaling and referencing the 3D model using the coordinates ofcontrol or reference points.

5. Analyzing structural features of the bench face.

2.2.1. Camera settingsThe camera used for this study was the Nikon D70s which could

cover an imaging area of approximately 40 m height and 50 m width.Fig. 2 shows the principal geometric arrangement for taking pictures ofa bench face.

Mean distance, D, should be enough to cover the area to be sur-veyed. The camera has two lenses: a Sigma lens and a Nikon lens. The

S. Manzoor, et al. Engineering Geology 267 (2020) 105480

2

Nikon lens is standard; it comes with the camera and has a focal lengthranging from 18 to 70 mm. The Sigma lens is used to capture images inmore confined or restricted spaces with a focal length ranging from 10to 20 mm. The focal length of the lens describes the view angle, i.e.,how much of the area will be photographed, and the magnification, i.e.,how much the image elements will be magnified. The shorter the focallength of the lens, the wider the angle of view and the lower themagnification and vice versa.

The Sigma lens has a shorter focal length than the Nikon lens and awider angle of view but lower magnification. If we are close to thebench face and cannot go further away because of restrictions likebench width and thus cannot cover the whole area of interest in theimage with the standard Nikon lens, the Sigma lens should be used.However, the Nikon lens should be used if the distance, D, does notrestrict the choice of lens, as this lens has less distortion than the Sigmalens and allows more light to enter the lens, making it more useful inlow light conditions.

The mining environment often restricts the distances and has moreconfined spaces, as in this study. Therefore, the Sigma lens was used tocapture the whole area with a shorter imaging distance, D. Severalcamera settings are pre-set by the software provider to ensure optimumuse with the software. Image resolution is one of those pre-set para-meters, with a value of 3008 × 2000. This image resolution was sui-table to identify the structures of interest in the images. Generally, theresolution should be better than the minimum size of the structure ofinterest in order to successfully identify structures in the photos. In thiscase, the images had a resolution of 300 dpi.

Camera calibration is another important procedure in photogram-metric analysis. In this procedure, a calibration object with knowngeometry is used. Many photos of that object are taken from differentangles. These photos are processed, and internal calibration parametersare determined, including lens distortion, principal point, and focallength (Qi et al., 2010). In our case, these details were delivered by thesoftware supplier as calibration sheets. The calibration sheets were

Fig. 1. Leveäniemi open pit mine with marked walls used for this study (modified from www.hitta.se).

Fig. 2. Basic geometric arrangement for data acquisition of a bench face (Modified from 3GSM, 2010).

S. Manzoor, et al. Engineering Geology 267 (2020) 105480

3

imported into the software before starting the 3D model construction tocalibrate the software for the camera used in this study. The systemtypically has a position accuracy in the centimeter range for geo-re-ferencing. So it is capable of producing quite accurate results.

2.2.2. Data acquisitionFig. 3 shows the principle of taking a stereoscopic image pair. These

stereoscopic images were used to generate 3D models of the bench facesusing a software module. A 3D model is developed by combining thespatial information with the information in 2D images. For our soft-ware, it is standard procedure to take two pictures instead of multiplepictures of the area under observation to generate a 3D model, as dataacquisition is easier and faster. However, several images were takenparallel to the bench face so that the best image pair giving maximumaccuracy in the 3D model generation could be used for analysis pur-poses.

Another crucial aspect of 3D model generation is marking andsurveying the control points in the model. This should be done carefullyto ensure the metric accuracy of the 3D model (3GSM, 2010). At leastthree control points are required to reference the model in a globalcoordinate system (geo-referencing). However, up to six points increasethe reliability of the measurements (3GSM, 2010). Therefore, six con-trol points, in the form of target disks (red-colored), were used in eachmodel for referencing purposes. The coordinates of these control pointswere determined using a GPS system. The distance between two controlpoints was fixed and measured and then used to scale the model.

2.2.3. Photogrammetric model generation3D models were constructed using high point densities, i.e., one

measurement for every pixel. Using higher point densities increases thedetails available in the 3D model. But there are drawbacks, includingincreased computation time and bigger file sizes. This was not a pro-blem for this study because of the availability of modern high speedcomputers. If there is a limitation due to hardware, lower point den-sities can be used, but this will compromise the details of interest. Oncea 3D model is constructed, the software displays the final result alongwith a quality indicator. Good results are shown as a green bar; bad orwrong results are in red. In this study, only the models with good resultswere used for further analysis. Another quality check for the 3D model

is the base length ratio indicated in the final results. The base lengthratio is the ratio of the base length, B, and the mean distance, D, asshown in Fig. 2. It should range between 12.5 and 20 for accuratemeasurements. The models complying with this condition were chosenfor further analysis.

The software provides an option to check the quality and accuracyof the model referencing as well. Statistics include the standard de-viation of easting, northing, and elevation, as well as total standarddeviation of coordinates. The 3D models showing accuracy in the lowercm range were used for accurate structural measurements. Range poleswith fixed length were used during data acquisition for scaling pur-poses. A special software tool, the ‘SMX normalizer’, was used to scalethe 3D model using the range pole length; this is very important for theaccuracy of the model. Fig. 4 shows a 3D model of one of the benchfaces with pre-split blast holes. These 3D models were analyzed usingthe Analyst module of the software package to determine the dip, dipdirection, trace length, exposed area and spacing of the discontinuitiesmarked on each bench face. To extract this information, scanlines weredrawn on the bench face.

The scanlines were drawn exactly on the location of the boreholes toachieve higher accuracy when comparing photogrammetric data withMWD data. For example, if a borehole had the coordinates with easting,182617.9, northing, 7504762.9, and elevation, 281.1 m, the scanlinewas added using the same easting, northing, and elevation values in the3D model. However, the depth of the scanline would be different fromthat of the borehole because the borehole is drilled deeper than thebench face height in blasting operations. Two parameters were used todraw scanlines at the borehole locations:

1. Coordinates of start and end points of the borehole.2. Half cast of the boreholes on the bench face.

Coordinates of the boreholes and range poles were measured in thesame coordinate system to have better correlation. Half cast of theboreholes were visible on the bench faces in pre-split wall so scanlineswere marked following the half cast so that scanline and borehole hadthe same location. However, only coordinates were used to matchscanlines and boreholes in poor rock conditions.

Fig. 3. Stereoscopic image pair. Two corresponding image points P(u,v) relate to one three-dimensional object point P(X,Y,Z) (3GSM, 2010).

S. Manzoor, et al. Engineering Geology 267 (2020) 105480

4

2.3. Measurement while drilling

MWD data were collected from the fully mechanized Epiroc (AtlasCopco) SmartROC D65 down-the-hole (DTH) drill rigs used in theLeveäniemi mine. The Epiroc drill monitoring system retrieves andstores MWD data. The recorded data include time (YYYY-MM-DDThh:mm:ss), depth (m), rotation pressure (bar), penetration rate (m/min), feed pressure (bar), and percussive pressure (water pressuremeasured by bar). The measured samples are recorded in files in theIREDES ‘.xml’ format. These files also record features like drill rigmodel, hole identification, plan identification etc., in addition to theabove-mentioned drilling parameters. The sampling interval along theborehole is set to 5 cm. MWD data are stored on each drill rig on a USBmemory stick. For this study, MWD data were collected from ninebench faces for a total of 473 blast holes. The length of the boreholeswas approximately 30 m for the pre-split blast holes and 15 m for theproduction blast holes. Fig. 5 shows both of bench faces with poor rockconditions and pre-split blastholes. The holes were drilled from 310 mcontour line i.e. the crest of the bench to 280 m contour line i.e. the toeof the bench in poor rock conditions. The length of the boreholes was15 m while the bench height was 30 m that means the bench wasblasted in 2 rounds to achieve a bench height of 30 m. The pre-split

blastholes were drilled from the contour line of 280 m i.e. the crest ofthe bench to 250 m contour line i.e. the toe of the bench as shown inFig. 4.

3. Data analysis

To establish the relationship between MWD data and rock structureslike joints, cavities, bedding planes and fractures, the data analysiscomprised the following two steps:

1- MWD data analysis2- Structural analysis

3.1. MWD data analysis

MWD data were collected from the drilling rigs. In the first step ofthe MWD data analysis, the recorded features were extracted in a tab-ular form. This table was used as the input data for the following steps.The measured features were placed in columns, and each sample was ona separate row. After importing the most significant features of thesamples into the table, the input data were filtered to remove incorrector faulty values. Noise and outliers can appear at different steps of the

Fig. 4. 3D model of a bench face with joints, exposed surfaces and scanlines (white dotted lines) marked on the surface.

Fig. 5. Location of the benches in both areas of the mine under consideration.

S. Manzoor, et al. Engineering Geology 267 (2020) 105480

5

data collection process, and should be removed before any analysis. Inthis study, the recorded noisy data were divided into three categories:

• Samples recorded in the beginning of each hole: The top of a bench isoften covered with loose rock fragments or fines due to poormucking of the material; alternatively, it can be fractured due to thesub-drilling and priming of the upper bench. In order to drill straightholes through this disturbed rock mass, the collaring is normallydone with great care, with reduced feed and percussive pressure.These recorded drilling parameters at the start of the borehole arenot related to the characteristics of the rock mass. Therefore theserecorded samples were removed from the table of input data.

• Samples recorded during the rod change process: Drill rods are usuallyshorter than the boreholes to be drilled. Therefore, extension rodsare used to drill deeper holes. Every time a new rod is added to thedrill string, the drilling parameters respond, but they resume theirbalanced state after few samples. These rod change signatures occurat regular intervals and are related to the drill rod length. Theserecorded responses don't represent the rock characteristics; rather,they represent changes in parameters caused by the drill rig. Sothese samples were eliminated from the input dataset.

• Samples with negative or abnormally high values: Faulty sampleshaving unrealistic values, e.g. negative or abnormally high valuesfor penetration rate or pressure, are not representative samples ofthe rock behavior during drilling. For example, a penetration rate of1000 m/min does not occur in reality. All these abnormal datasamples were removed from the input data.

Fig. 6 shows the measured parameters for one hole before filtering.The figure indicates that both feed pressure and percussive pressure arereduced during drilling of the first rod (6 m), and the reduction of feedpressure results in a reduction of the rotation pressure during the sameinterval. The figure also shows that the penetration rate reaches 150 m/min at the beginning of the hole which is a very unrealistic value andbecause of this high value, the rest of the curve is disappeared as themean value is less than 1 m/min. Finally, the pressure drops after aregular interval of 6 m, when a new rod is added to the drill string.

Information about the location of the holes is required to map theMWD datasets to the structural data obtained from the 3D models.According to the IREDES standard, another file type named DQ (DrillQuality) is recorded in addition to the MWD files. These files containthe specifications for each borehole: borehole identification, planidentification, borehole name, type of borehole, start point coordinatesof borehole, end point coordinates of borehole, etc. The start and endpoint coordinates of the borehole represent the x, y and z for the firstand last recorded samples in each borehole. Borehole identification andplan identification are two common features of the DQ files and MWD

files. The files can be combined based on their common features foreach sample in the MWD file, and the start and end point coordinates ofthe hole in which the sample is placed can be precisely determined.

In the MWD files, there is also a parameter named depth tag, re-presenting the depth from the collaring point where the parameters ofeach sample are recorded. The approximate coordinates of each samplecan be obtained by adding the depth value to the z coordinate of therelated borehole start point or by subtracting the depth value from the zcoordinate of the related borehole end point. Another feature recordedfor each borehole in the DQ files is status, representing the status of thespecified borehole with respect to the following values: undrilled, failand success. Since the recorded data from undrilled or fail holes are notreliable, the samples related to these holes were removed from the maindataset. Fig. 7 shows the data filtering, the start and end point co-ordinates and the status of the borehole.

3.2. Structural analysis

The structural analyses were carried out using ShapeMetriX 3D, acommercial software package developed by 3GSM and used to acquirestructural data from stereoscopic images (Gaich et al., 2006; Haneberg,2008; Vasuki et al., 2014; Beyglou, 2016; Buyer and Schubert, 2017).The SMX Analyst tool of the software was used to mark structuresmanually and add scanlines as shown in Fig. 3. Coordinates of theboreholes taken from the MWD data were used to draw scanlines at theborehole locations. The structural data were extracted for the desiredscanlines using ShapeMetriX3D which shows the coordinates of eachintersection between any structure and the scanline. Every intersectionbetween the structures and the scanline is presented as easting,northing and elevation. Easting and northing were used to match theresults with the corresponding borehole data in MWD files; elevationwas used to calculate the depth of the structures along the scanline tovisualize the variations in MWD data at the corresponding depth. Anexample of scanline data is shown in Fig. 8.

4. Correlation and validation

Coordinates of the scanlines and boreholes were matched to com-pare the variations in the structural data and the MWD data. Structuraldata and MWD data were plotted against the depth to identify the si-milarities, and the areas showing similarities in the data sets weremarked on the images manually. The behavior of the MWD data wasdifferent for different types of structures. Some examples of the resultsfrom the test walls are presented below.

Figs. 9 and 10 show the behavior of MWD data when the hole isdrilled through a rock with fractures or joints with very small aperture.Fig. 6. Noisy MWD data parameters plotted against depth of the hole.

Fig. 7. MWD data parameters vs. depth of the hole with specified locationcoordinates and status.

S. Manzoor, et al. Engineering Geology 267 (2020) 105480

6

In these conditions, it is primarily penetration rate (PR) and rotationpressure (RP) that respond to fractures or joints.

hese parameters show a slight increase in their values when thedrilling process encounters the above discontinuities. As the rock is ofcompetent nature and the aperture of these fractures/joints is not veryprominent, the increase in drilling parameters is also not very promi-nent. Statistically, the mean value of penetration rate is 0.85 m/min; itincreases up to 2 m/min, approximately, when it intersects the dis-continuities, as shown in the figures. Similarly, the mean value of ro-tation pressure is 60.78 bar, and it increases to 70 bar when drillingthrough.

the discontinuities.Fig. 11 shows another type of MWD data response indicated by

regions of noisy data with fluctuations for all drilling parameters. Thefigure indicates that this type of behavior occurs when there are cavitieswith removable material or a wedge which is no longer available on the

rock face after the blast. The fluctuations in the MWD data in this caselast longer than the fluctuations for fractures or joints. It is probablysticky or loose material that causes the MWD to fluctuate so much.

The third type of behavior shown by the MWD data occurs when thedrilling process is carried out in foliated rock or in a rock mass withclosely spaced bedding planes. There is a region of increased penetra-tion rate for the part of the hole with either of these traits. Fig. 12presents MWD data for a highly fractured rock face, with correspondingimage.

As the figure shows, the foliations do not cause any abrupt fluc-tuations in MWD data. There are four spikes in the penetration rate as aresult of four fractures/joints marked as black lines in the rock faceimage, but the regions between them show almost constant behavior forthe foliations or bedding planes. Statistically, the mean value of thepenetration rate is increased, i.e. 0.95 m/min, showing the low strengthor fractured nature of the rock. The end part of the hole is actually sub-drilling, so it cannot be seen in the images. The rock was blasted with anormal production blasting method, so the actual borehole location wasremoved due to back break.

5. Concluding remarks

Measurement while drilling (MWD) is increasingly applied inmining activities around the world. However, the use of MWD data toimprove mining processes is still not fully implemented, as the varia-tions in MWD parameters are not fully understood yet. Rock masscharacterization using traditional methods or remote sensing techni-ques to describe the rock behavior is mostly based on the visible portionof the rock mass. If applied correctly, MWD data can be used to greatlyimprove this description of the rock behavior, as the data represent thehidden volume of the rock.

Two of the drilling parameters, feed pressure and percussive pres-sure, are controlled by the drill control system and therefore do notrespond to rock properties, as can be seen in Figs. 6 and 8. The othertwo parameters, however, penetration rate and rotation pressure, vary

Fig. 8. Scanline data showing coordinates of different structures intersectingthe scanline.

Fig. 9. Rise in penetration rate (PR) and rotation pressure (PR) at fracture/joint location.

S. Manzoor, et al. Engineering Geology 267 (2020) 105480

7

with the variations in the rock mass being drilled.The pre-split wall has good rock conditions, so the drilling para-

meters do not fluctuate much and only show some variations whenmajor joints or cavities are encountered. The penetration rate and ro-tation pressure normally rise when the drill bit intersects a joint. In case

of a single joint, this shows up as a peak with one sample. With a cavity,variations appear in multiple samples; this appears as a noisy region onthe MWD plot. The other wall has joints and closely spaced beddingplanes, fractures and foliations, so it shows abrupt variations, as well assystematic increases or decreases in MWD data plot. The close joints

Fig. 10. Variations in penetration rate (PR) and rotation pressure (RP) along with corresponding structures.

Fig. 11. MWD data from two different holes showing fluctuations in all the drilling parameters.

S. Manzoor, et al. Engineering Geology 267 (2020) 105480

8

and fractures or bedding planes do not cause variations in drillingparameters, but they can change the mean values for penetration rateand rotation pressure due to the loss of material strength.

This information can be incorporated into the blast design to adjustthe amount of explosives in the blast holes to improve rock fragmen-tation through better material handling and increase the stability of theremaining wall. However, the MWD system needs to be improved toacquire better quality data with minimum noise and false information.

Declaration of Competing Interest

None.

Acknowledgments

The authors acknowledge LKAB and the staff and management ofthe Svappavaara mine for valuable input and support during field stu-dies. Epiroc is also acknowledged for its valuable input into the project.Finally, Vinnova, the Swedish Energy Agency and Formas are ac-knowledged for financing this project through the SIP-STRIM program.The authors would like to acknowledge the support from the SLIMproject funded by the European Union's Horizon 2020 research andinnovation program under grant agreement no. 730294 and the supportfrom Center for Advanced Mining and Metallurgy (CAMM2), Sweden.

References

3GSM, 2010. ShapeMetriX 3D, User Manual. 3G Software and Measurement GmbH., Graz.Assali, P., Grussenmeyer, P., Villemin, T., Pollet, N., Viguier, F., 2014. Surveying and

modeling of rock discontinuities by terrestrial laser scanning and photogrammetry:semi-automatic approaches for linear outcrop inspection. J. Struct. Geol. 66,102–114.

Bemis, S.P., Micklethwaite, S., Turner, D., James, M.R., Akciz, S., Thiele, S.T., Bangash,H.A., 2014. Ground-based and UAV-based photogrammetry: a multi-scale, high re-solution mapping tool for structural geology and paleoseismology. J. Struct. Geol. 69(A), 163–178.

Beyglou, A., 2016. On the Operational Efficiency in Open Pit Mines, Licentiate Thesis.Luleå: Luleå Tekniska Universitet, Sweden. ISBN: 978-91-7583-699-7 (electronic).

Bonilla-Sierra, V., Scholte’s, L., Donze´, F.V., Elmouttie, M.K., 2015. Rock slope stabilityanalysis using photogrammetric data and DFN–DEM modelling. Acta Geotech. 10 (4),497–511.

Buyer, A., Schubert, W., 2017. Calculation the spacing of discontinuities from 3D Point

clouds. Proceedia Eng. 191, 270–278.Chen, J., Zhu, H., Li, X., 2016. Automatic extraction of discontinuity orientation from

rock mass surface 3D point cloud. Comput. Geosci. 95, 18–31.Fekete, S., Diederichs, M., Lato, M., 2010. Geotechnical and operational applications for

3-dimensional laser scanning in drill and blast tunnels. Tunn. Undergr. SpaceTechnol. 25 (5), 614–628.

Feng, Q., Roeshoff, K., 2004. In-situ mapping and documentation of rock faces using a fullcoverage 3D laser scanning technique. Int. J. Rock Mech. Min. Sci. 41 (1), 139–144.

Feng, Q., Sjögren, P., Stephansson, O., Jing, L., 2001. Measuring fracture orientation atexposed rock faces by using a non-reflector total station. Eng. Geol. 59 (1–2),133–146.

Ferrero, A.M., Forlani, G., Roncella, R., Voyat, H.I., 2009. Advanced geostructural surveymethods applied to rock mass characterization. Rock Mech. Rock. Eng. 42 (4),631–665.

Firpo, G., Salvini, R., Francioni, M., Ranjith, P.G., 2011. Use of digital terrestrial photo-grammetry in rocky slope stability analysis by distinct elements numerical methods.Int. J. Rock Mech. Min. Sci. 48 (7), 1045–1054.

Fisher, J.E., Shakoor, A., Watts, C.F., 2014. Comparing discontinuity orientation datacollected by terrestrial lidar and transit compass methods. Eng. Geol. 181, 78–92.

Francioni, M., Salvini, R., Stead, D., Coggan, J., 2018. Improvements in the integration ofremote sensing and rock slope modelling. Nat. Hazards 90 (2), 975–1004.

Fraser, C.S., Cronk, S., 2009. A hybrid measurement approach for close-range photo-grammetry. ISPRS J. Photogramm. Remote Sens. 64 (3), 328–333.

Gaich, A., Pötsch, M., Schubert, W., 2006. Acquisition and assessment of geometric rockmass features by true 3D images. In: The 41st U.S. Symposium on Rock Mechanics(USRMS), 17-21 June, 2006. Amreican Rock Mechanics Association, Golden,Colorado.

Ghosh, R., Schunnesson, H., Gustafson, A., 2017. Monitoring of drill system behavior forwater-powered In-The-Hole (ITH) drilling. Minerals 7, 121.

Gigli, G., Casagli, N., 2011. Semi-automatic extraction of rock mass structural data fromhigh resolution LIDAR point clouds. Int. J. Rock Mech. Min. Sci. 48 (2), 187–198.

Haneberg, W.C., 2008. Using close range terrestrial digital photogrammetry for 3D rockslope modelling and discontinuity mapping in the United States. Bull. Eng. Geol.Environ. 67 (4), 457–469.

ISRM, 1978. Suggested methods for the quantitative description of discontinuities in rockmasses. Int. J. of Rock Mech. Min. Sci. Geo Mecha. Abstr. 15 (6), 319–368.

Jaboyedoff, M., Couture, R., Locat, P., 2009. Structural analysis of Turtle mountain(Alberta) using digital elevation model: toward a progressive failure. Geomorphology103 (1), 5–16.

Kemeny, J., Post, R., 2003. Estimating three-dimensional rock discontinuity orientationfrom digital images of fracture traces. Comput. Geosci. 29 (1), 65–77.

Kemeny, J., Mofya, E., Handy, J., 2003. The use of digital imaging and laser scanningtechnologies for field rock fracture characterization. In: Culligan (Ed.), Soil and RockAmerica. 2003. pp. 117–122.

Khorzoughi, M.B., 2013. Use of measurement while drilling techniques for improved rockmass characterization in open-pit mines. In: Master’s Thesis. Vancouver: University ofBritish Columbia, Canada. https://doi.org/10.14288/1.0073823.

Khorzoughi, M.B., Hall, R., 2016. Processing of measurement while drilling data for rockmass characterization. Int. J. Min. Sci. Technol. 26 (6), 989–994.

Khorzoughi, M.B., Hall, R., Apel, D., 2018. Rock fracture density characterization usingMeasurement While Drilling (MWD) technique. Int. J. Min. Sci. Technol. 28 (6),

Fig. 12. MWD data plot of a borehole and rock face image for poor rock conditions.

S. Manzoor, et al. Engineering Geology 267 (2020) 105480

9

859–864.Lato, M.J., Diederichs, M.S., Hutchinson, D.J., 2010. Bias correction for view-limited lidar

scanning of rock outcrops for structural characterization. Rock Mech. Rock. Eng. 43(5), 615–628.

Lemy, F., Hadjigeorgiou, J., 2003. Discontinuity trace map construction using photo-graphs of rock exposures. Int. J. Rock Mech. Min. Sci. 40 (6), 903–917.

Lemy, F., Hadjigeorgiou, J., 2004. A field application of laser scanning technology toquantify rock fracture orientation. In: 53rd Geomechanics Colloquium; EUROCK2004, Salzburg, Austria, 06-08 October 2004. Austrian Society of Geomechanics, pp.435–438.

Li, X., Chen, J., Zhu, H., 2016. A new method for automated discontinuity trace mappingon rock mass 3D surface model. Comput. Geosci. 89, 118–131.

Linkwitz, K., 1963. Terrestrisch-photogrammetrische kluftmessung. Rock Mech. Eng.Geol. I, 152–159.

Matthews, N.A., 2008. Aerial and Close-Range Photogrammetric Technology: ProvidingResource Documentation, Interpretation, and Preservation. Bureau of LandManagement, Denver, Colorado.

Monteiro, S.T., Ramos, F., Hather, P., 2009. Conditional Random Fields for RockCharacterization using Drill Measurements. In: International Conference on MachineLearning and Applications, Miami, USA, pp. 13–15 December 2009.

Qi, W., Li, F., Zhenzhong, L., 2010. Review on camera calibration. In: Chinese Control andDecision Conference, Xuzhou, China, May 26–28.

Rai, P., Schunnesson, H., Lindqvist, P., Kumar, U., 2016. Measurement-while-drillingtechnique and its scope in design and prediction of rock blasting. Int. J. Min. Sci.Technol. 26 (4), 711–719.

Remondino, F., El-Hakim, S., 2006. Image-based 3-D modelling: a Review. Photogramm.Rec. 21 (115), 269–291.

Rengers, N., 1967. Terrestrial photogrammetry: a valuable Tool for engineering geolo-gical purposes. Rock Mech. Eng. Geol. V, 150–154.

Riquelme, A.J., Abellán, A., Tomás, R., 2015. Discontinuity space analysis in rock massesusing 3D point clouds. Eng. Geol. 195, 185–196.

Riquelme, A., Cano, M., Tomás, R., Abellán, A., 2017. Identification of rock slope dis-continuity sets from laser scanner and photogrammetric point clouds: a comparativeanalysis. Procedia Eng. 191, 838–845.

Roberts, G., Poropat, G., 2000. Highwall joint mapping in 3-D at the moura mine usingSOROJOINT. In: Beeston, J.W. (Ed.), Bowen Basin Symposium 2000 Proceedings.Rockhampton, Australia, 22–24 October 2000, pp. 371–377.

Rodgers, M., McVay, M., Horhota, D., Hernando, J., 2018. Assessment of rock strengthfrom measuring while drilling shafts in Florida limestone. Can. Geotech. J. 55 (8),1154–1167.

Salvini, R., Francioni, M., Riccucci, S., Bonciani, F., Callegari, I., 2013. Photogrammetryand laser scanning for analyzing slope stability and rock fall runout along theDomodossola–Iselle Railway, the Italian Alps. Geomorphology 185, 110–122.

Salvini, R., Mastrorocco, G., Seddaiu, M., Rossi, D., Vanneschi, C., 2017. The use of anunmanned aerial vehicle for fracture mapping within a marble quarry (Carrara,Italy): photogrammetry and discrete fracture network modelling. Geomatics Nat.Hazards Risk 8 (1), 34–52.

Schunnesson, H., 1996. RQD prediction based on drill performance. Tunneling Undergr.Space Technol. 11 (3), 345–351.

Segui, J.B., Higgins, M., 2002. Blast design using measurement while drilling parameters.Fragblast: Int. J. Blasting Fragm. 6 (3–4), 287–299.

Sturzenegger, M., Stead, D., 2009a. Close-range terrestrial digital photogrammetry andterrestrial laser scanning for discontinuity characterization on rock cuts. Eng. Geol.106 (3–4), 163–182.

Sturzenegger, M., Stead, D., 2009b. Quantifying discontinuity orientation and persistenceon high mountain rock slopes and large landslides using terrestrial remote sensingtechniques. Nat. Hazards Earth Syst. Sci. 9 (2), 267–287.

Van Eldert, J., Schunnesson, H., Johansson, D., Saiang, D., 2019. Application of mea-surement while drilling technology to predict rock mass quality and rock support fortunnelling. Rock Mech. Rock. Eng. https://doi.org/10.1007/s00603-019-01979-2.

Van Oosterhout, D., 2016. Use of MWD data for detecting discontinuities. In: Master’sThesis. Delft: Delft University of Technology, Netherland.

Vasuki, Y., Holden, E., Kovesi, P., Micklethwaite, S., 2014. Semi-automatic mapping ofgeological structures using UAV-based photogrammetric data: an image analysisapproach. Comput. Geosci. 69, 22–32.

Vezhapparambu, V.S., Eidsvik, J., Ellefmo, S.L., 2018. Rock classification using multi-variate analysis of measurement while drilling data: towards a better samplingstrategy. Minerals 8, 384.

Westboy, M.J., Brasington, N.F., Glasser, M.J., Hambrey, J.M., Reynolds, J.M., 2012.‘Structure-from-Motion’ photogrammetry: a Low-cost, effective tool for geoscienceapplications. Geomorphology 179, 300–314.

Wolf, P.R., Dewitt, B.A., 2000. In: Elements of Photogrammetry with Applications in GIS,Third. McGraw-Hill, Boston, pp. 608.

S. Manzoor, et al. Engineering Geology 267 (2020) 105480

10