Comparison of Landsat and ASTER in Land Cover Change

Detection within Granite Quarries

R. S. Moeletsi

1,2

and S. G. Tesfamichael

Department of Geography, Environmental Management and Energy Studies,

University of Johannesburg, Auckland Park, 2006, Johannesburg, South Africa

Mintek, 200 Malibongwe Drive, Randburg, 2125, South Africa

Keywords: Granite Quarries, Landsat, ASTER, Accuracy Assessment.

Abstract: This study evaluated and compared the utility of Landsat and ASTER in land cover change detection within

granite quarries. Landsat data used was acquired in 1998 and 2015 while ASTER data used was acquired in

2001 and 2013. Both Landsat and ASTER were classified using supervised and maximum likelihood

classification. Post-classification and Normalized Difference Vegetation Index change detection techniques

were applied to assess and measure changes in land cover caused by granite quarries. Overall classification

of ASTER was slightly higher than that obtained for Landsat (overall accuracy (OA) =79% and kappa 0.75vs.

OA=75% and kappa 0.71). Both Landsat and ASTER were able to assess land cover changes within granite

quarries. Change detection results revealed increase in granite quarries which subsequently resulted in

decrease in vegetation and bare land and increase in water bodies within the quarries. The study found ASTER

to be better at discriminating granite quarries from other land cover features and was able to detect small water

bodies within granite quarries due to higher spatial resolution of bands in the VNIR subsystem. On the

contrary, Landsat was found better at detecting changes in vegetation within granite quarries.

1 INTRODUCTION

Remote sensing techniques are useful in mapping,

monitoring and managing land cover changes related

to mining activities (López-Pamo et al., 1999).

Coupled with capabilities to cover large areas,

availability of historic data, availability of data at high

spatial and spectral resolution, the technology is

continuously contributing significantly to land

management initiatives (Rogan and Chen, 2004).

Several studies have used remotely sensed data

ranging from low to high spatial resolution sensors

such as MODIS, NOAA AVHRR, Landsat, ASTER,

SPOT, and IKONOS in change detection studies (Lu

et al., 2004). Even though the use of remotely sensed

data has been widely utilized in land use and land

cover change (LULCC), its applications in mapping

impacts of surface mining have not been extensively

explored (Latifovic, 2005). This paper therefore

compares utility of Landsat and ASTER satellite

sensors in land cover changes caused by granite

quarries located in the North West province of South

Africa.

Mining is an integral part of economic develop-

ment in many developing countries, however, it is

often associated with adverse environmental and

social impacts (Paull et al., 2006). Granite quarrying

in South Africa started in the late 1930s in

Bon-Accord area. Quarrying adversely affects

environment in various ways. Common

environmental impacts resulting from quarrying

activities include loss of vegetation, disruption and

destruction of natural habitat (Maponga and

Munyanduri, 2001), and can alter hydrological

systems (Darwish et al., 2011). It is therefore

important to monitor environmental variables related

to mining activities. Identifying and monitoring such

impacts contributes to sustainable development and

provides information regarding rehabilitation

measures, future site selection methods and

determining locations of abandoned and unreclaimed

quarries (Demirel et al., 2011).

2 STUDY AREA

Granite quarrying in South Africa occurs in the Main

Zone of the Rustenburg Layered Suite in the

Bushveld Igneous Complex. The area is dominated

Moeletsi, R. and Tesfamichael, S.

Comparison of Landsat and ASTER in Land Cover Change Detection within Granite Quarries.

DOI: 10.5220/0006675801870195

In Proceedings of the 4th International Conference on Geographical Information Systems Theory, Applications and Management (GISTAM 2018), pages 187-195

ISBN: 978-989-758-294-3

187

by gabbro and norite with interlayered anorthosite of

the pyramid Gabbro-Norite, Rustenburg Layered

Suite (Nex et al., 1998). The quarries are located

between Rustenburg and Brits towns in the North

West Province (Figure 1). Commercially, the word

granite refers to any crystalline rock exploited for use

in the construction and ornamental use (Dolley,

2007). Granite mining in the North West province

contributed 46 of the national mining of granite in

2008 (Lamprecht et al., 2011).

Figure 1: Maps showing (Top): Location of the study area

and granite quarries (Bottom): Landscape of quarries and

surrounding area.

3 METHODOLOGY

A minimum spatial coverage of 1 hectare and 200 m

distance between the quarries were specified for

quarry samples. The specification were set out to

avoid overlap of samples, to promote independent

comparison of quarries and to enhance detection with

spatial resolution of remote sensing data. As a result,

forty granite quarries were sampled for the study.

Since the launch of Google Earth

in 2005, people

have been using it to explore the world around them

(Cha and Park, 2007). It provides images with high

spatial resolution (<2.5 m)

that are useful for land use

and land cover mapping (Hu et al., 2013). Sampling

was achieved using geographi-cal coordinates of

known quarries. The coordinates were overlain on

Google Earth

to digitize the quarries and

subsequently convert the polygons to shapefiles in

ArcGIS

(ESRI 2016, ArcMap 10.4, Redlands,

California, USA). Digitization process was done with

the corresponding dates of acquired remote sensing

data and therefore, the year 2015 and 2013 were used

to digitize the quarries on Google Earth

. Quarries

corresponding to remote sensing data acqui-red in

earlier years before 2005 could not be used for

digitization due to lack of data in Google Earth

3.1 Remote Sensing Data

Data used in this study included Landsat images

acquired on the 16

March 1998 and 16

April 2015

while ASTER data was acquired on the 26

April

2001 and 11 October 2013. The data was acquired

from the United States Geological Survey

(https://earthexplorer.usgs.gov/). Attempts were

made to acquire images of same or close dates for

consistent comparison between Landsat and ASTER

data however, most of ASTER data was covered with

clouds and therefore only available cloud free data

was considered. The study preferred the use of data

acquired during high rainfall summer season when

vegetation is denser, however, unavailability of

suitable data necessitated the use of images outside of

this time period.

3.2 Data Processing

3.2.1 Image Registration of ASTER Data

Image registration process involves matching two or

more images which were taken from different sensors

at different times (Wahed et al., 2013). Accuracy in

image registration is important as this can

significantly affect the results of change detection

process. As a result, image registration accuracy

should be limited to half or one pixel in change

detection (Townshend et al., 1992). In this study, the

acquired ASTER data was firstly converted from

hierarchical data format (HDF) file to tagged image

file format (TIFF) file in ERDAS IMAGINE software

(ERDAS IMAGINE

2016, Hexagon Geospatial,

Norcross, USA). After converting ASTER data from

hdf to tiff format, image registration was performed

GISTAM 2018 - 4th International Conference on Geographical Information Systems Theory, Applications and Management

188

using automated registration technique in ArcGIS

10.4 on the 2001 and 2013 ASTER images using

Landsat 1998 and 2015 images as reference data.

Registration accuracy for both 2001 and 2013 images

were within one pixel in each dataset.

3.2.2 Radiometric Calibration

Radiometric calibration is an important step required

to improve quality of remotely sensed data by

removing factors that alters spectral properties of land

surface features (Pons et al., 2014). Both Landsat and

ASTER images were radiometrically calibrated using

absolute calibration method. This enables comparison

of images acquired at different times from different

sensors (Chander et al., 2009). Data was calibrated by

firstly converting the digital numbers (DNs) to at-

sensor spectral. The second step involved converting

at-sensor spectral radiance to exoatmospheric top of

atmosphere (TOA) reflectance (Chander et al., 2009).

Equations applicable Landsat data are explained by

(Chander et al., 2009) while those applicable for

ASTER data are provided by (Abrams and Hook,

2002).

3.2.3 Supervised Classification

Supervised classification was applied to multispectral

images created from Landsat and ASTER data. The

technique requires the user to select training samples

which are representative of the desired classes to be

identified. The quality of this classification method

depends highly on the quality of training classes

(Perumal and Bhaskaran, 2010). Supervised

classification involves three principle steps. The first

step involves defining training classes, the second

step is creation of signature file and the last step is

classification of the image (Lillesand et al., 2014).

Maximum likelihood classifier (MLC) algorithm was

used to classify multispectral images. This method

uses the training data by means of estimating means

and variances of the classes, which are used to

estimate probabilities and also consider the variability

of brightness values in each class (Perumal and

Bhaskaran, 2010). The effectiveness of MLC depends

highly on accuracy of training samples (Richards,

2012).

3.2.4 Accuracy Assessment

Accuracy assessment was carried out on the Landsat

2015 and ASTER 2013 classified images. Error

matrix was used to evaluate the accuracy of

classification. A random set of 189 points were

selected for error matrix. These points were overlain

on Google Earth

; the name of each class was then

recorded using visual interpretation of land cover

features on Google Earth

. The recorded class

names in the reference data were then compared to

classes generated from each image and the supervised

classification. Google Earth

has been used in a

number of studies as a source of reference against

which classification could be compared (Cha and

Park, 2007; Rwanga and Ndambuki, 2017). Error

matrix was generated and accuracy assessment

parameters i.e. producer’s accuracy (measure of

omission errors), user’s accuracy (measure of

commission errors) and Kappa coefficient (measure

of agreement) were computed.

3.2.5 Change Detection

Change detection involves four major aspects: (1)

detecting that changes have occurred, (2) identify-ing

the nature of the change, (3) measuring the areal

extend of the change and (4) assessing the spatial

pattern of the change (Congalton and Green, 2008).

Various techniques used to perform change detec-tion

with digital imagery has been described by Singh

(1989). This study utilized post-classification and

Normalized Difference Vegetation Index (NDVI)

change detection techniques to assess land cover

changes within granite quarries. In post- classification

comparison, each image is classified independently

and then classification results are compared to

determine areas and magnitude of change (Singh,

1989). The NDVI has been widely used to measure

vegetation condition and biomass (Jiang et al., 2006).

It is defined as the difference between the near-

infrared band (NIR) and the red band divided by the

sum of these two bands (Tucker, 1979). The results of

NDVI range between -1 and +1, where negative

values correspond to absence of vegetation and

positive values correspond to vegetated zones.

The higher the index, the greater the chlorophyll

content of the target (Pettorelli, 2013).

4 RESULTS

4.1 Accuracy Assessment

Error matrix for Landsat image is presented in Table

1. Overall accuracy obtained for Landsat data was

75% with Kappa coefficient of 0.71. Error matrix

demonstrated that Water bodies had perfect producer’s

and user’s accuracy. Granite quarries had moderate

producer’s accuracy due to confusion with Exposed

rock formation, Bare land, Built-up land and Other

Comparison of Landsat and ASTER in Land Cover Change Detection within Granite Quarries

189

mining areas. Results for user’s accuracy however,

were very high with limited confusion from Other

mining areas. Results of other classes showed

misclassification with other classes i.e.: Bare land and

Vegetation had good producer’s accuracies but were

also confused with each other. Other mining areas were

confused with Bare land, Built-up land and Granite

quarries. Exposed rock formation resulted in low

producer’s accuracy due to confusion with Bare land

and Built-up land. Similarly, Built-up land was

confused with Bare land and that resulted in low

producer’s accuracy of Built-up land.

Overall accuracy obtained for ASTER imagery

was 79% with kappa coefficient of 0.75 (Table 2).

Similar to Landsat classification, there was also

confusion in classification of features. Water bodies

had perfect producer’s and user’s accuracy.

Producer’s accuracy for Granite quarries was high,

however, was confused with Vegetation. Misclassi-

fication of Exposed rock formation was observed due

to Granite quarries, Bare land, Built-up land and

Vegetation. Relatively low producer’s accuracy was

obtained for Other mining areas, due to mostly

confusion with Granite quarries, Built-up land, Bare

land and Vegetation. On the contrary, user’s accura-

cy for Other mining areas was perfect. Similarly,

Exposed rock formation had almost perfect user’s

accuracy with limited confusion observed with Bare

land. Granite quarries had high user’s accuracy, but

were confused with Exposed mining areas, Vegeta-

tion, and Other mining areas. Low user’s accuracy in

Bare land was caused by misclassification from Other

mining area, Built-up land, Exposed rock formation

and Vegetation. Similarly, low user’s accuracy in

Vegetation was a result of confusion caused by

Granite quarries, built-up land, Bare land, Other

mining areas and Exposed rock formation.

Table 1: Error matrix of classification derived from Landsat imagery in 2015.

Reference Data

Classified Data

WB GQ ER BUL BL V OMA Tot.

UA (%)

WB 10 0 0 0 0 0 0 10 100

GQ 0 20 0 0 0 0 1 21 95

ER 0 3 19 0 0 0 0 22 86

BUL 0 2 2 19 0 0 3 26 73

BL 0 1 9 11 24 3 4 52 46

V 0 0 0 0 6 27 0 33 82

OMA 0 3 0 0 0 0 22 25 88

Tot. 10 29 30 30 30 30 30 141

PA (%) 100 69 63 63 80 90 73

Overall accuracy = 75%, Kappa = 0.71

Key: WB=Water Bodies, GQ= Granite Quarries, ER= Exposed Rock Formations, BUL=Built-Up Land, BL=Bara Land,

V=Vegetation, OM= Other Mining Areas, Tot. =Total, PA=Producer’s Accuracy, UA= User’s Accuracy.

Table 2: Error matrix derived from ASTER imagery taken in 2013 (Key definitions similar as in Table 2).

Reference Data

Classified Data

WB GQ ER BUL BL V OMA Tot. UA (%)

WB 10 0 0 0 0 0 0 10 100

GQ 0 26 1 0 0 1 1 29 90

ER 0 0 24 0 1 0 0 25 96

BUL 0 0 1 21 0 1 3 26 81

BL 0 0 2 6 24 4 3 39 62

V 0 3 2 3 4 26 4 42 62

OMA 0 0 0 0 0 0 19 19 100

Tot. 10 29 30 30 29 32 30 150

PA (%) 100 90 80 70 83 81 63

Overall accuracy = 79%, Kappa = 0.75

GISTAM 2018 - 4th International Conference on Geographical Information Systems Theory, Applications and Management

190

4.2 Post-classification Change

Detection

Figure 2 shows an example of land cover change

within granite quarry boundaries on a zoomed portion

of the study area. Seven land cover types generated

from Landsat and ASTER classification analysis

included (1) Water bodies, (2) Vegetation (3) Other

mining areas (4) Granite quarries, (5) exposed rock

formation, (6) Built-up land and (7) Bare land. The

results from Landsat data classification in the whole

study area revealed that dominant land cover types in

1998 were Vegetation, Bare land, natural Water

bodies and Exposed rock formation with moderate

occurrences of Granite quarries. In the year 2015

there was an increase in Granite quarries, Built-up

land as well as Water bodies inside the quarries. A

decrease in Vegetation and Bare land was observed in

the year 2015. Figure 2 shows portions of the study

area where development of granite quarries evolved

(2015) on the land that did not have quarries before

(1998).

Results of classification of ASTER images

revealed that Granite quarries were lesser in 2001

compared to the year 2013. Land cover in 2001 was

dominated by Vegetation and Bare land, natural

Water bodies, Exposed rock formation and to a lesser

extent, covered with granite quarries. Classification

of 2013 image however, revealed an increase in

Granite quarries relative to those detected in 2001 as

well as an increase in water bodies inside the quarries.

The 2013 image also revealed loss in Bare land and

increase in Exposed rock formation. Similarly, Figure

2 shows a portion of land cover that did not have

granite quarries in 2001 but evolved in the year 2013.

4.2.1 Quantitative Measures of Land Cover

change

The measure of areal extent of land cover change

based on forty quarries between 1998 and 2015 for

Landsat is given in Table 3. The results reveal

significant increase in Granite quarries which

subsequently resulted in accumulation of Water

bodies. Increase in Granite quarries also resulted in

significant loss of Vegetation and Bare land.

Figure 2: Land cover distributions created using supervised classification. Left images: land cover change within granite

quarries derived from Landsat imagery (1998 and 2015). Right images: land cover change within granite quarries derived

from ASTER imagery (2001 and 2013).

Comparison of Landsat and ASTER in Land Cover Change Detection within Granite Quarries

191

Table 3: Measure of land cover change within granite

quarries based on Landsat classification data.

Classes

Area of classes

(ha)

Difference

(ha)

1998 2015 2015-1998

Water Bodies 0.2 2.07 1.9

Granite

Quarries

433.5 910.4 476.9

Bare

Land

19.2 2.7 -16.5

Vegetation 793.7 313.1 -480.6

Similarly, the measure of change within granite

quarries using ASTER data revealed more or less the

same as those obtained from Landsat data.

Table 4 presents quantitative measure of land

cover change based on ASTER data.

Table 4: Measure of land cover change using ASTER data.

Classes

Area of classes

(ha)

Difference

(ha)

2001 2013 2013-2001

Water

Bodies

0.2 2.4 2.23

Granite

Quarries

213.1 745.1 531.9

Bare

Land

219.5 157.4 -62.1

Vegetation 704.2 236.5 -467.7

Increase in Water bodies as detected by ASTER is

more relative to results obtained from Landsat. There

was a significant increase in Granite quarries while

Vegetation and Bare land decreased significantly.

4.3 Change Detection using NDVI

Comparison of mean NDVI values within granite

quarries using Landsat and ASTER data is presented

in Figure 3. High mean NDVI values are observed in

the year 1998 indicating more presence of green

vegetation than in 2015. Mean NDVI values within

quarries in 1998 range from 0.17 to 0.54 while for

quarries in 2015 the range is between 0.05–0.3.

Majority of quarries in 2001 have mean NDVI values

above 0.25 while in 2013 majority have mean NDVI

values below 0.1. Quarry 27 shows lowest mean

NDVI value in the year 2001 whereas in 2013 and

2015, the quarry shows high mean NDVI values. This

is a typical example of an abandoned quarry where

revegetation is taking place. One quarry (Quarry #1)

was sampled to evaluate NDVI pattern based on

individual pixels within the quarry. NDVI histogram

of Landsat data based on 322 pixels revealed that 79%

of pixels in the year 1998 have NDVI values above

0.29. On the contrary, 45% of pixels in the year 2015

have NDVI values equal or less than 0. 148 while

other pixels show distribution across a range of NDVI

values.

Analysis of NDVI values using ASTER data was

based on 1296 pixels in the same quarry. Results showed

that 100% of the pixels in the year 2013 have NDVI

values between 0 and 0.233 while majority of pixels in

the year 2001 are distributed above 0.233 NDVI values.

Figure 3: Comparison of mean NDVI values within granite quarries using Landsat and ASTER data.

GISTAM 2018 - 4th International Conference on Geographical Information Systems Theory, Applications and Management

192

Figure 4: NDVI frequency distribution graphs for (left) Landsat data suing 322 pixels at 30 m spatial resolution and (right)

ASTER data using 1296 pixels at 15 m spatial resolution.

5 DISCUSSIONS

The aim of this study was to compare utility of

Landsat and ASTER in land cover change detection

within granite quarries. Both Landsat and ASTER

proved effective in mapping and detecting land cover

changes within granite quarries. Misclassifica-tion of

classes was encountered in both classification of

Landsat and ASTER imagery. Land cover change

detection using both satellite sensors revealed a

significant increase in Granite quarries. This increase

subsequently resulted in loss of Vegetation and Bare

land. Expansion of Granite quarries also resulted in

accumulation of Water bodies inside the quarries.

Mouflis et al. (2008) and Koruyan et al. (2012) have

also demonstrated capabilities of Landsat and

ASTER in monitoring land cover changes caused by

expansion in marble quarries.

NDVI change detection analysis revealed

decrease in green Vegetation cover within the

acquired data period for both Landsat and ASTER.

Results of NDVI derived from Landsat indicated that

mean NDVI comparison between 1998 and 2015

varies across all quarry samples. Results obtained

from ASTER data, showed that majority of quarries

(95%) in 2001 displayed mean NDVI values between

0.25-0.3 while for 2013, majority (90%) of quarries

had mean NDVI values below 0.1. Decrease in green

Vegetation within granite quarries indicates the

proliferation of quarrying activity over the acquired

data periods. On the contrary, an increase in

Vegetation on other quarries indicates revegetation

process in abandoned quarries.

Although Landsat and ASTER were able to map

land cover changes within granite quarries. ASTER

data was found to be more effective in discriminating

Granite quarries and small Water bodies within

granite quarries. This is attributed to the higher spatial

resolution of ASTER in the visible and near infrared

of electromagnetic spectrum than Landsat’s is (15 m

vs 30 m). On the other hand, analysis of NDVI change

detection revealed that Landsat sensor was better at

detecting green Vegetation compared to results

obtained using ASTER data.

Similar observations were recorded by Chevrel et

al. (2005) who demonstrated capabilities of ASTER

data in identifying and mapping surface disturbances

due to mining. Charou et al. (2010), also demonstrate-

ed the effectiveness of ASTER in monitoring

anomalies of water surfaces compared to Landsat and

SPOT. Similarly, Musa and Jiya (2011) demonstrated

the effectiveness of Landsat in assessing mining

activities impacts on vegetation using NDVI.

6 CONCLUSION

Comparison of Landsat and ASTER data in change

detection within granite quarries was evaluated in this

study. Overall accuracy of classification using

supervised classification and MLC for Landsat was

75% with kappa coefficient of 0.71, while ASTER

returned a slightly better overall classification

accuracy (79%) and kappa coefficient (0.75). Land

cover mapping using Landsat data had limitation in

detecting water bodies within granite quarries due to

inadequate spatial resolution of the image relative to

water body sizes. Vegetation cover was well

discriminated in Landsat as compared to ASTER

data. ASTER was found more effective in delineating

granite quarries as compared to Landsat and this is

attributed to the high spatial resolution of ASTER in

Comparison of Landsat and ASTER in Land Cover Change Detection within Granite Quarries

193

the visible and near infrared of the electromagnetic

spectrum.

REFERENCES

Abrams, M. and Hook, S. (2002) ‘ASTER User Hand-book

Version 2’, Jet Propulsion, 4800, pp. 1–135.

Cha, S. and Park, C. (2007) ‘The utilization of google earth

images as reference data for the multitemporal land

cover classification with MODIS data of north Korea’,

Korean Journal of Remote Sensing, pp. 483–491.

Chander, G., Markham, B. L. and Helder, D. L. (2009)

‘Summary of current radiometric calibration

coefficients for Landsat MSS, TM, ETM+, and EO-1

ALI sensors’, Remote Sensing of Environment. Elsevier

Inc., 113(5), pp. 893–903.

Charou, E., Stefouli, M., Dimitrakopoulos, D., Vasiliou, E.

and Mavrantza, O. D. (2010) ‘Using remote sensing to

assess impact of mining activities on land and water

resources’, Mine Water and the Environment, 29(1), pp.

45–52.

Chevrel, S., Bourguignon, A., Cottard, F. and Itard, Y.

(2005) ‘Exploitation of ASTER imagery in mining-

related environmental management’, in Pecora 16 -

Global Priorities in Land Remote Sensing. Sioux Falls,

South Dakota, pp. 1–10.

Congalton, R. G. and Green, K. (2008) Assessing the

Accuracy of Remotely Sensed Data: Principles and

Practices. Second Edi. Boca Raton: CRC Press.

Darwish, T., Khater, C., Jomaa, I., Stehouwer, R., Shaban,

A. and Hamze, M. (2011) ‘Environmental impact of

quarries on natural resources in Lebanon’, Land and

degradation and Development, 22, pp. 345–358.

Demirel, N., Emil, M. K. and Duzgun, H. S. (2011) ‘Surface

coal mine area monitoring using multi-temporal high-

resolution satellite imagery’, International Journal of

Coal Geology. Elsevier B.V., 86(1), pp. 3–11.

Dolley, T. P. (2007) ‘Stone, Dimension’. USGS Minerals

Yearbook, 1, pp. 1–11.

Hu, Q., Wu, W., Xia, T., Yu, Q., Yang, P., Li, Z. and Song,

Q. (2013) ‘Exploring the use of google earth imagery

and object-based methods in land use/cover mapping’,

Remote Sensing, 5(11), pp. 6026–6042.

Jiang, Z., Huete, A. R., Chen, J., Chen, Y., Li, J., Yan, G.

and Zhang, X. (2006) ‘Analysis of NDVI and scaled

difference vegetation index retrievals of vegetation

fraction’, Remote Sensing of Environment, 101(3), pp.

366–378. doi: 10.1016/j.rse.2006.01.003.

Koruyan, K., Deliormanli, a. H., Karaca, Z., Momayez, M.,

Lu, H. and Yalçin, E. (2012) ‘Remote sensing in

management of mining land and proximate habitat’,

Journal of the Southern African Institute of Mining

and Metallurgy, 112(7), pp. 667–672.

Lamprecht, A. J. H., Cilliers, S. S., Götze, A. R. and Du

Toit, M. J. (2011) ‘Phytosociological description of

norite koppies in the Rustenburg area, North-West

Province and refinement of the distribution of the

Norite Koppies Bushveld on the national vegetation

classification map of South Africa’, Bothalia, 41(2), pp.

327–339.

Latifovic, R. (2005) Satellite RemoteSensing in Assessing

the Environmental Impactsof Large-Scale Surface

Mining. PhD Thesis, Department De Genie Des Mines,

De La Metallurgie Et Des, Univerite Laval Quebec.

Lillesand, T., Kiefer, R. W. and Chipman, J. (2014) Remote

Sensing and Image Interpretation. 7th edn. New York:

John Wiley & Sons.

López-Pamo, E., Barettino, D., Antón-Pacheco, C., Ortiz,

G., Arránz, J. C., Gumiel, J. C., Martínez-Pledel, B.,

Aparicio, M. and Montouto, O. (1999) ‘The extent of

the Aznalcollar pyritic sludge spill and its effects on

soils’, Science of the Total Environment, 242(1–3), pp.

57–88.

Lu, D., Mausel, P., Brondizio, E. and Moran, E. F. (2004)

‘Change detection techniques’, International Journal of

Remote Sensing, 25(February), pp. 2365–2407.

Maponga, O. and Munyanduri, N. (2001) ‘Sustainability of

the dimension stone industry in Zimbabwe—challenges

and opportunities’, Natural Resources Forum, 25(3),

pp. 203–213.

Mouflis, G. D., Gitas, I. Z., Iliadou, S. and Mitri, G. H.

(2008) ‘Assessment of the visual impact of marble

quarry expansion (1984-2000) on the landscape of

Thasos island, NE Greece’, Landscape and Urban

Planning, 86(1), pp. 92–102.

Musa, H. D. and Jiya, S. N. (2011) ‘An Assessment of

Mining Activities Impact on Vegetation in Bukuru Jos

Plateau State Nigeria Using Normalized Differential

Vegetation Index (NDVI)’, Journal of Sustainable

Development, 4(6), pp. 150–159.

Nex, P. A., Kinnaird, J. A., Ingle, L. J., Van Der vyver, B.

A. and Cawthorn, R. G. (1998) ‘A new stratigraphy for

the main zone of the Bushveld Complex, in the

Rustenburg area’, South African Journal of Geology,

101(3), pp. 215–223.

Paull, D., Banks, G., Ballard, C. and Gillieson, D. (2006)

‘Monitoring the Environmental Impact of Mining in

Remote Locations through Remotely Sensed Data’,

Geocarto International, 21(1), pp. 33–42.

Perumal, K. and Bhaskaran, R. (2010) ‘Supervised

classification performance of multispectral images’,

Journal of Computing, 2(2), pp. 124–129.

Pettorelli, N. (2013) The Normalized Difference Vegetation

Index. New York, USA: OUP Oxford.

Pons, X., Pesquer, L., Cristóbal, J. and González-Guerrero,

O. (2014) ‘Automatic and improved radiometric

correction of landsat imageryusing reference values

from MODIS surface reflectance images’,

International Journal of Applied Earth Observation

and Geoinformation. Elsevier B.V., 33(1), pp. 243–

254.

Richards, J. A. (2012) Remote Sensing Digital Image

Analysis: An Introduction. Berlin Heidelberg: Springer

Science & Business Media.

Rogan, J. and Chen, D. (2004) ‘Remote sensing technology

for mapping and monitoring land-cover and land-use

change’, Progress in Planning, 61, pp. 301–325.

GISTAM 2018 - 4th International Conference on Geographical Information Systems Theory, Applications and Management

194

Rwanga, S. S. and Ndambuki, J. M. (2017) ‘Accuracy

Assessment of Land Use/Land Cover Classification

Using Remote Sensing and GIS’, International Journal

of Geosciences, 8(4), pp. 611–622.

Singh, A. (1989) ‘Review Article Digital change detection

techniques using remotely-sensed data’, International

Journal of Remote Sensing, 10(6), pp. 989–1003.

Townshend, J. R. G., Gurney, C., McManus, J. and Justice,

C. O. (1992) ‘The Impact of Misregistration on Change

Detection’, IEEE Transactions on Geoscience and

Remote Sensing, 30(5), pp. 1054–1060.

Tucker, C. J. (1979) ‘Red and photographic infrared linear

combinations for monitoring vegetation’, Remote

Sensing of Environment, 8(2), pp. 127–150.

Wahed, M., El-tawel, G. S. and Gad El-karim, A. (2013)

‘Automatic Image Registration Technique of Remote

Sensing Images’, International Journal of Advanced

Computer Science & Applications, 4(2), pp. 177–187.

Comparison of Landsat and ASTER in Land Cover Change Detection within Granite Quarries

195