Journal of Himalayan Earth Sciences 46(1) (2013) 89-98
Land use change detection in the limestone exploitation area of Margalla Hills
National Park (MHNP), Islamabad, Pakistan using geo-spatial techniques
Muhammad Farooq Iqbal1, Mobushir Riaz Khan1 and Amir H. Malik2
1Department of Meteorology, COMSATS Institute of Information Technology (CIIT), Islamabad, Pakistan
2Department of Environmental Sciences, COMSATS Institute of Information Technology (CIIT),
The aim of the study is to assess the limestone (LS) exploitation area and its negative impacts on
natural resources using geo-spatial techniques. LS is an important constituent of cement manufacturing
which is extensively used in the construction of infrastructure such as roads and buildings. The Margalla
Hills National Park (MHNP) is situated around Islamabad and contains a large amount of LS reserves.
However, the exploitation of LS from the MHNP is causing harmful environmental impacts on the
surrounding areas. Geographic Information System (GIS) and Remote Sensing (RS) have been proved as
powerful tools for LS exploitation assessment. Four Landsat Thematic Mapper (TM)/Enhanced Thematic
Mapper (ETM+) satellite images have been acquired over a span of 17 years (1992-2009). The temporal
changes in the study area were detected by performing the digital image processing techniques of image
enhancement and supervised classification. The classification accuracy has been verified with high
resolution Google Earth images and by using error matrix. Advanced Space borne Thermal Emission and
Reflection Radiometer (ASTER) Global Digital Elevation (GDEM) has been used for topographic
information extraction. It was observed that LS exploitation area increased from 0.35% to 5.72% from
1992-2009, whereas, the vegetation was decreased from 23.46% to 12.12%. Urban development also
increased rapidly. The results showed that LS exploitation is deteriorating the ecosystem, biodiversity,
landscape and vegetation of the MHNP which was established in 1980 to protect, conserve and manage
the biodiversity and ecosystem in this region. We conclude that the LS exploitation in the MHNP should
be managed properly to secure the water, soil, air quality in the federal capital. Initiatives should be taken
for the rehabilitation of the LS exploited areas and then, to suggest alternative LS exploitation sites in the
near periphery of study area with EIA restrictions.
Keywords: Limestone exploitation; Landsat satellite; Remote Sensing; Land use; Change detection
Limestone (LS) exploitation is carried out for
the cement production which is obtained by grinding
clinker with Gypsum and used for construction
purposes (Akinbile, 2007). Stone crushing and
cement production is one of the significant industrial
activities that exists all over the world. This sector is
an important industrial segment for infrastructural
development and construction of roads, bridges,
canals, dams, building and housing projects.
However, during the process of crushing a
considerable amount of dust is emitted at almost
every stage (Akinbile, 2007). This process not only
effects the local environment but also the human
health in the surrounding areas. Furthermore, the
removal of vegetation exposes soil surface and
thereby enhances the chances of land erosion
(Ibrahim, 2002). Therefore, LS exploitation must
involve planning, conflict resolution, construction,
operation and close down (FMSMD, 1999).
Margalla hills contain high quality of
Limestone (LS including sand-stone and shale,
which is considered as best for construction
purposes (Nawaz et al., 2004). Increasing trend of
LS exploitation is harmful for vegetation, natural
ecosystem, biodiversity and human health. Thus,
environmental impact assessment (EIA) is required
before the execution of any project to mitigate the
harmful environmental impacts. Conventional
surveying and mapping techniques consume a lot
of time and are expensive as well for the
assessment of LS exploitation area while such
information is not readily available, especially in
developing countries. Satellite Remote Sensing
(SRS) and GIS have been widely applied in
identifying and analyzing for LS exploitation
assessment, environmental impact assessment, geo-
morphological analysis, land use mapping and
planning (Treitz et al., 1992; Harris and Ventura,
1995). Usually, remotely sensed data is used to
provide information on terrain surface whereas on
the other hand, GIS is a supporting tool to RS and
has the capability to manipulate and store data in
digital forms (Shaban, 2010). Using GIS,
Wandahwa and van (1996) implemented land
suitability evaluation and mapped climate, altitude,
soil type, and ecosystem. The topography of the
land surface is one of the most essential
characteristic of the earth used in GIS analysis
(Reuter et al., 2009).
Space-borne sensors have shown a great
potential for delivering reliable estimates of the
extent and changes occurring in the LS exploitation
area along with mapping urban growth and forest
cover estimation (Foody et al., 1996). Multi-
temporal and multi-spectral satellite images are
used for characterizing land use and land cover
(LULC) change and deforestation rates (Lu et al.,
2004). SRS is a growing technology and is being
used as a fundamental component of conservation
planning and biodiversity assessment (Sesnie et al.,
2008; Stickler and Southworth, 2008). LS
exploitation area assessment from satellite data is
executed using either supervised and/or
unsupervised classification method. The spectral
complexity of predefined classes has further led to
numerous suggestions for procedures and
techniques to improve classifications including
topographic normalization, spatial filtering, image
segmentation, object-oriented classifications
(Foody et al., 1996), vegetation indices and multi-
temporal image data (Lucas et al., 1993).
LULC changes may have negative impacts on
environment, water resources, land and vegetation
if appropriate measures are not taken in time
(Sunday and Ajewole, 2006).The lack of proper
rehabilitation of the LS exploitation sites causes
serious environmental degradation and spread of
diseases (Kaliampakos et al., 1998). This
necessitates temporal monitoring of changes
occurring in the land use and land cover of an area.
The need for low-cost data resource is particularly
important for conservation and research in
developing countries where funding for mapping is
often limited. So research interests are being
directed to the mapping and monitoring using
RS/GIS techniques (Epstein et al., 2002). Landsat
onboard Multispectral Scanner (MSS), Thematic
Mapper (TM), and Enhanced Thematic Mapper
Plus (ETM+) is archiving and delivering free of
costimagesforabout30 years (USGS, 2008). It
meets a wide range of information requirements for
monitoring the conditions of earth’s land surface
(Williams et al., 2006; Chander et al., 2009). The
availability of multispectral and high resolution
data as well as advanced capabilities of digital
image processing techniques, in generating
enhanced and interpretable images, has further
enlarged the potential use of RS in delineating
lithological contacts and geological structure in
great detail (Drury, 1987; Yousif and Shedid, 1999;
Crippen and Blom, 2001). There are many
applications of RS in geology which involve the
delineation of structures, discrimination of different
rock, soil types and resource exploration (Kruse
and Dietz, 1991). Many geological studies have
employed TM and ETM+ data to discriminate the
various lithologies, lineaments limestone and
minerals (Abrams, 1984).
The main objective of the current study is to
assess the spatiotemporal changes in the limestone
exploitation area and its impacts over the past 17
years, using geospatial technique. Moreover, to
identify alternative LS querying sites in an attempt
to save the MHNP.
2. Study area
Margalla Hills National Park (MHNP) is
located in the north of the Pakistan’s capital city
Islamabad, whereas the study area lies between
33040'01" to 33042'43" N latitude, 72045'01" to
72052'32"E longitude as shown in (Fig. 1). MHNP
was declared as the National Park under
Islamabad Wild Life Ordinance in 1980. The step
was taken to conserve the natural resources from
injudicious human activities such as over
cultivation, grazing, mining and water pollution
(Ahmad, 2009). The foothills run approximately
from north to northwest direction and are about 40
km in length (Malik and Husain, 2003). The
current study focuses on a subpart of MHNP, where
limestone exploitation is taken place. MHNP, like
the Potwar Plateau and Azad Kashmir, also has a
distinct altitudinal range and lies at the junction of
Potwar Plateau and northern mountainous region of
Pakistan (Masroor, 2011). The soil of the study area
is derived from wind, water-laid deposits and
sedimentary rocks. Margalla Hills are largely
tertiary in age with smaller areas of formation
belonging to quartzitic sandstone calcareous shale
and limestone (Hijazi, 1984). The topography of
the area is rugged, varying in elevation comprising
mainly steep slopes and gullies (Fig. 2), where the
rock structure is basically limestone (Yasin and
Rubina, 1987). Natural springs and rainfall play an
important role for streams in the MHNP. The area
falls in the far end of monsoon zone and the mean
monthly 254 mm of monsoon precipitation occurs
in July and August (Maqsood, 1991). The mean
relative humidity for the same period varies
between 59 and 67% (Masroor, 2011). The hottest
months are May and June as the temperature then
rises up to 42°C and the coldest months are
December and January when temperature falls
below zero (Hussain, 1986). MHNP vegetation is
largely the result of monsoon and the foothills flora
is mostly tropical in origin (Shinwari and Khan,
1998). According to Chandio (1995) the water-
table has dropped from 12.5 to 22.9m in
Rawalpindi and from 10 to 19.8m in Islamabad.
Fig. 1. The location of the study area in the Margalla Hills National Park, Islamabad (Pakistan)
Fig.2. Digital Elevation Model (DEM) generated and extracted for topographic analysis, where study
area is divided into six zones with different elevations ranging from 449 to 883 meters.
3. Data and Methodology
Two types of data have been used in the
study. Elevation data is used for topographic
information extraction and multi-temporal
Landsat satellite images were used for Limestone
exploitation area assessment. Other LULC classes
i.e. bare land, vegetation, water, road network and
urban development were derived to observe the
impact of limestone exploitation on them.
Advanced Space borne Thermal Emission and
Reflection Radiometer (ASTER) Global Digital
Elevation Model (GDEM) have been released
with a 1 arc second and provided much extended
coverage. ASTER GDEM is an easy-to-use,
covering all the land on earth, and available to all
users regardless of size or location of the study
areas to extract topographic information. The
ASTER GDEM with 30 meter resolution is
expected to meet the requirements of many users,
for global topographic information (Reuter et al.,
Landsat data is a precious resource for
monitoring global changes and is a primary source
of medium spatial resolution used in decision-
making (Goward et al., 2006; Masek et al., 2008;
Vogelmann et al., 2008). Remote sensing
classification approaches such as supervised,
unsupervised and knowledge-based expert system
approaches (Sugumaran et al., 2003; Mundia and
Aniya, 2005; Lu and Weng, 2005), have been
used and are still being used for the land cover
change detection. The supervised classification is
a time consuming, however, an accurate system
used to extract surface features manually from
remote sensing images (Butt et al., 2011).
Heavy LS exploitation started in the study
area in 1980’s. This study is based on land cover
classification of four multi-temporal cloud free
images. The acquisition dates are 20 September
1992, 08 September 1998, 21 May 2002 and 30
May 2009 with having path 150 and row 37.
ERDAS Imagine 9.0 software was used to carry
out all image processing tasks while Arc GIS 9.1
was used for analysis and map development. To
minimize radiometric distortion due to different
atmospheric conditions and different acquisition
dates, a basic radiometric and image-based
atmospheric correction according to Chavez
(1996) was applied. Radiometric correction was
performed on all the four images due to variations
in scene illumination and sensor irregularities,
viewing geometry, atmospheric conditions and
sensor noise. The images have been geometrically
corrected using the UTM map projection (Zone
43N, datum WGS84). In order to geometrically
correct the original distorted image, the
resampling technique of nearest neighbor method
is used to determine the digital values, for placing
in the new pixel locations of the corrected output
image. This method uses the digital value from
the pixel in the original image which is average of
four nearest to the new pixel location in the
corrected image with RMS error of less than one
pixel. Our methodology is based on the Principal
Component Analysis (PCA) and supervised
classification methods. PCA are used to extract
specific training sites and finally, the Maximum
Likelihood (MLA) supervised classification
algorithmis used to produce satellite-derived maps
(Butt et al., 2012). PCA is used to convert raw
remote sensing data of multi-spectral imageries
into a new principal component image, which is
more easily interpretable (Singh and Harrison,
1985; Weng, 1993). Digital image processing
techniques are used to define unique training sites
for classification of the study area. The image
classification procedure is applied to automatically
categorize all pixels into land cover classes or
themes on the basis of training sites defined in an
image. In this study 40 training sites are identified
on the basis of filed survey and image
interpretation. The Maximum Likelihood (MLH)
classification algorithm is applied in the current
study, because it can be more effective than the
often used Minimum Distance Algorithm (MDA)
and Mahalanobis Distance Algorithm (MDA)
when the number of training sites per class is
larger (Jensen, 1996). As for as the spectral
reflectance concerns, water bodies generally
reflect high in the visible spectrum. For
vegetation, the spectral reflectance is based on the
chlorophyll and water absorption in the leaf. For
man-made materials, concrete and asphalt, both
display spectral curves generally increase from the
visible through the Near IR and Mid-IR regions.
Bare land decreases as organic matter increases
(Iqbal et al., 2009).
The classification accuracy was verified using
high resolution Google Earth images which are
available and quite conveniently help to recognize
the studied natural objects. The study area shape
file was converted into KML layer which was
then imported to Google Earth. The classified
images and Google Earth images, all were
displayed simultaneously to cross refer the
classification. The areas that indicated limestone,
vegetation, urban development, water and bare
land on the classified images were zoomed in the
Google Earth images covered by KML layer and
the classification accuracy was checked. Accuracy
assessment was also performed by using the
original satellite images to avoid errors in the
reference data using Error matrix. Error matrices
were developed on all the four images to check
the accuracy as shown in Table 1. The
classification accuracy using error matrix was
found to be more than 98%. The higher level of
accuracy was obtained in the data of TM 1992 and
less accuracy was found in the data of TM 2009.
With the help of field survey and literature
review alternative sites would be identified and
their quantity and quality would be examined.
Few alternative querying sites have also been
recommended by the (Nawaz et al., 2004) other
Table 1. Accuracy analysis (percentage) of all the classes derived from Landsat data.
Landsat (1992) classification accuracy: 98%)
Landsat (1998) classification accuracy: 98%)
Landsat (2002) classification accuracy: 98%)
Landsat (2009) classification accuracy: 99%)
4. Results and discussion
Landsat TM and ETM+ satellite data was
classified using MLH algorithm to investigate the
LS exploitation area. Results of this classification
are shown in Table 2 and Figure3. Results show
that the LS exploitation area is increasing rapidly
from the past two decades. It can be deduced
from the results that in a short span of six years
from September 1992 to September 1998, the LS
exploitation area was increased by 1.47% (Fig. 4)
whereas, LS exploitation area increased by 3.08%
from September 1998 to May 2000. From May
2000 to May 2009 the increase observed 5.72%.
From the past two decades, we observed a
continuous increase in the LS exploitation area
and if it continues with the same rate in the
coming years, then some part of the Margalla
Hills can be disappeared. Prominent agencies of
the country including Environmental Protection
Agency (EPA) and Natural Resource Monitoring
(NRM) can be benefitted with such studies in
order to make plans for better management of the
limestone exploitation. The rapidly occurring LS
exploitation is in turn severely destroying the
natural ecosystem and vegetation as the results
shows that the vegetation is decreasing rapidly in
the study area. Total vegetation-covered area;
either sparse or dense decreased by about 12%
from September 1992 to 2009 (Table 2). The
obvious effect of LS exploitation would be on
vegetation and decreasing trend of vegetation
would result in bare land. Urban development
increased very rapidly in the study area from 1992
to 2009 as in September 1992 it was recorded to
be 0.44%, and finally in May 2009 urbanization was
estimated to be 3.86%. We monitored increasing and
decreasing trend of bare land in various years due to
other factors including vegetation, limestone
exploitation area, urbanization and water.
Increasing and decreasing trend of road network
has been found to be due to the condition of the
road. If the road would be repaired, then it would
not have been reflected as road and considered as
bare land. In this study, only surface water was
considered as water class and changing in the
water is due to the over consumption of water.
Water is directly related to the rainfall. If there
would have been rainfall prior to taking of image,
then we could have been observed more water-
covered area. Maps of the study area after
classification process are shown in Figures 5.
Table 2. Area in square kilometer (km2) obtained from Landsat satellite of the classes including
Limestone Exploitation, Bare Land, Vegetation, Water, Road network and Urban development.
A comparison was made between every two years to find the changes annually.
Fig. 3. Increasing and decreasing trend of all the six classes from 1992 to 2009, where limestone
exploitation area is increasing rapidly with the increasing speed of urbanization.
Fig. 4. Change detection of all the classes in percentage from 1992 and 2009.
Exploitation Bare Land Vegetation Water Road network Urban
1992 1998 2002 2009
Exploitation Bare Land Vegetation Water Road
Changes (1992-1998) Changes (1998-2002) Changes (2002-2009)
Fig. 5. Classified maps of satellite image from the year 1992 to 2009 (a-d).
Results show overall 98% average
classification accuracy using error matrices and
these results were also verified through high
resolution Google Earth images that are available
and quite conveniently help to recognize the
studied objects in the study area.
We have also identified various LS querying
sites other than the MHNP. These sites have been
identified using field survey and comprehensive
literature review. The alternative LS querying sites
are Khairi Murat, Kala ChittaDhar, Pathargarh,
Khanpur and Ganghar Ranges. It has been analyzed
that the recommended sites are sufficient for not
only present needs but also for future requirements.
These alternatives sites are not far away from
Islamabad and Rawalpindi and enough for present
and future requirements. Alternative recommend
sites would be beneficial to save the biodiversity
and ecosystem of MHNP and also support the
Government to collect the revenue.
On the basis of our results, we conclude that
i. Limestone exploitation is directly
affecting the vegetation-covered area that
might accelerate erosion.
ii. Due to excessive LS exploitation,
environment is also getting polluted and
as a result it is also infecting the
surrounding areas other than study area.
iii. LS exploitation is altering the natural
ecosystem and biodiversity.
iv. Geospatial techniques can be used to
assist the policy makers to identify
alternate sites for LS exploitation.
v. If both the LS exploitation and
urbanization keep on increasing with the
same rate, then environment of the study
area would be rigorously polluted.
Abrams, M.J., 1984. Landsat 4, Thematic Mapper
and Thematic Mapper Simulator data for a
Porphyry Copper Deposit. Photogrammetric
Engineering and Remote Sensing, 50, 1171-
Ahmad, S.S., 2009. Ordination and classification of
herbaceous vegetation in Margalla Hills
National Park Islamabad Pakistan. Biological
Diversity and Conservation, 2(2), 38-44.
Akinbile, L.A., 2007. Social impact assessment of
Limestone exploitation in Yewa North local
government area of Ogun State, Nigeria. Pakistan
Journal of Social Sciences, 4(1),107-11.
Butt, M.J., Waqas, A., Iqbal, M.F., Muhammad, G.,
Lodhi, M.A.K., 2012. Assessment of Urban
Sprawl of Islamabad Metropolitan Area Using
Multi-Sensor and Multi-Temporal Satellite
Data. Arab Journal Science and Engineering,
Chander, G., Markham, B.L., Helder, D.L., 2009.
Summary of current radiometric calibration
coefficients for Landsat MSS, TM, ETM+ and
EO-1 ALI sensors. Remote Sensing of
Environment, 113, 893-903.
Chandio, B.A., 1995. Water management policies to
sustain irrigation system in Pakistan.
Proceedings of regional conference on water
resources management, 119-128, Isfahan, Iran.
Chavez, P.S., 1996. Image-based atmospheric
corrections: Revised and improved.
Photogrammetric Engineering and Remote
Sensing, 62, 1025-1036.
Crippen, R.E., Blom, R.G., 2001. Unveiling the
lithology of vegetated terrains in remotely
sensed imagery. Photogrammetric Engineering
and Remote Sensing, 67(8), 935-946.
Drury, S.A., 1987. Image Interpretation in Geology.
Chapman and Hall, London.
Epstein, J., Payne, K., Kramer, E., 2002. Techniques
for mapping suburban sprawl. Photogrammetric
Engineering and Remote Sensing, 63 (9), 913-
FMSMD, 1999. A report and map of solid mineral
distribution in Ogun State. Federal Ministry of
Solid Mineral Development, Mines Department,
Abeokuta, Ogun State, Nigeria, 32-67.
Foody, G. M., Palubinskas, G., Lucas, R. M., Curran,
P. J., Honzak, M., Amaral, I.D., 1996. Identifying
terrestrial carbon sinks: Classification of
successional stages in regenerating tropical
forest from Landsat TM data. Remote Sensing
of Environment, 55, 205-216.
Goward, S., Irons J., Franks, S., Arvidson, T.,
Williams, D., Faundeen, J., 2006. Historical
record of Landsat global coverage: Mission
operations, NSLRSDA and international
cooperator stations. Photogrammetric
Engineering and Remote Sensing, 72, 1155-
Harris, P.M., Ventura, S.J., 1995. The integration of
geographic data with remotely sensed imagery
to improve classification in an urban area.
Photogrammetric Engineering and Remote
Sensing, 61, 993-998.
Hijazi, S., 1984. A phytosociological study of
Margallah Hills National Park. Unpublished M.
Phil thesis, Department of Biological Sciences,
Quaid-I-Azam University, Islamabad.
Hussain, M., 1986. Re-introduction of cheer
pheasant in the Margalla Hills National Park: A
report by World Wide Fund for Nature, Pakistan
and Capital Development Authority Islamabad.
Ibrahim, M.A., 2002. Introduction to environmental
problems and management. Wa’adallah
environment Consults (WADEC) Nigeria.
Iqbal, M.F., Lodhi, M.A.K., Butt, M.J., Ghaffar, A.,
2009. Landcover classification of Islamabad
region using Remote Sensing of Landsat Spatio-
Temporal images. Proceedings of First
International conference on Aerospace Science
and Engineering (ICASE-2009) Islamabad,
Jensen, J.R., 1996. Introductory Digital Image
Processing: A Remote Sensing Perspective.
Prentice Hall, Englewood Cliffs.
Kaliampakos, D., Bardani, I., Menegaki, M., 1998.
Environmental impacts of mining and social
attitude: A view through the records. Proceeding
of International Conference on Protection and
Restoration of the Environment, Halkidiki,
Kruse, F.A., Dietz, J.B., 1991. Integration of diverse
remote sensing data sets for geologic mapping
and resource exploration. Proceedings SPIE-The
International Society for Optical Engineering,
Earth and Atmospheric Remote Sensing, 1492,
Lu, D., Weng, Q., 2005. Urban classification using
full spectral information of Landsat ETM+
imagery in Marion County, Indiana.
Photogrammetric Engineering and Remote
Sensing, 71 (11), 1275-1284.
Lu, W., Mausel, P., Brondízio, E., 2004. Change
detection techniques. International Journal of
Remote Sensing, 25, 2365-2407.
Lucas, R.M., Honzak, M., Foody, G.M., Curran,
P.J., Corves, C., 1993.Characterizing tropical
secondary forests using multi-temporal Landsat
sensor imagery. International Journal of Remote
Sensing, 14, 3061-3067.
Malik, R.N., Husain, S.Z., 2003. Evaluating
deforestation using Landsat TM and SPOT XS
data in dry sub-tropical forest of Margalla Hills,
northwest of Pakistan. In: Benes (Ed.), Geo-
information for European-wide Integration, 429-
Maqsood, A., 1991. Mammals of Margallah Hills
National Parks: An Annotated List of Mammals,
Masek, J.G., Vermote, H.C., Wolfe, R., Cohen, W.,
Hall, F., Kutler, J., Nelson, P., 2008. North
American forest disturbance mapped from a
decadal Landsat record. Remote Sensing of
Environment, 112, 2914−2926.
Masroor, R.,2011. An Annotated Checklist of
Amphibians and Reptiles of Margalla Hills
National Park, Pakistan. Pakistan Journal of
Zoology, 43(6), 1041-1048.
Mundia, C.N., Aniya, M., 2005. Analysis of land
use/cover changes and urban expansion of
Nairobi city using remote sensing and GIS.
International Journal of Remote Sensing, 26
Nawaz, F., Hamidullah, Fayaz, A., 2004. The Effect
of Mining On Geomorphology (detection of
Changes By Using Remote Sensing
Techniques), Proceedings of ISPRS Congress
Commission VII, Geo-Imagery Bridging
Continents 12-23 July, Istanbul, Turkey.
Reuter, H.I., Nelson, A., Strobl, P. Mehl, W., Jarvis,
A.,2009.A first assessment of Aster GDEM tiles
for absolute accuracy, relative accuracy and
terrain parameters, Geoscience and Remote
Sensing Symposium, IEEE International,
Sesnie, S.E., Gessler, P.E., Finegan, B., Thessler, S.,
2008. Integrating Landsat TM and SRTM DEM
derived variables with decision trees for habitat
classification and change detection in complex
neotropical environments. Remote Sensing of
Environment, 112, 2145-2159.
Shaban, A., 2010. Support of Space Techniques for
Ground water Exploration in Lebanon. Journal
of Water Resource and Protection, 2, 469-477.
Shinwari, M.I., Khan, M.A., 1998. Ethnobotany of
Margalla Hills National Park of Islamabad.
PASTIC printing press, Islamabad.
Singh, A., Harrison, A., 1985.Standardized principal
components. International Journal of Remote
Sensing, 6, 883-896.
Stickler, C.M., Southworth, J., 2008. Application of
multi-scale spatial and spectral analysis for
predicting primate occurrence and habitat
associations in Kibale National Park, Uganda.
Remote Sensing of Environment, 112, 2170-
Sugumaran, R., Pavuluri, M.K., Zerr, D., 2003. The
use of high resolution imagery for identification
of urban climax forest species using traditional
and rule based classification approach. IEEE
Transactions on Geoscience and Remote
Sensing, 41(9), 1933–1939.
Sunday, O.A., Ajewole, A.R., 2006. Implications of
the Changing Pattern of Landcover of the Lagos
Coastal Area of Nigeria. American-Eurasian
Journal of Scientific Research, 1(1), 31-37.
Treitz, P.M., Howard, P.J., Gong, P., 1992.
Application of satellite and GIS technologies for
landcover and land-use mapping at the rural-
urban fringe: A case study. Photogrammetric
Engineering and Remote Sensing, 58, 439-448.
USGS, 2008. Imagery for everyone: timeline set to
release entire USGS Landsat archive at no
charge. Technical Announcement.
Wandahwa, P., Van, R.E., 1996. Qualitative land
suitability assessment for pyrethrum cultivation
in west Kenya based upon computer-captured
expert knowledge and GIS. Agriculture,
Ecosystems and Environment, 56, 187-202.
Weng, F., 1993. A knowledge-based vision system
for detecting land changes at urban fringes.
IEEE Transm. Geosciences Remote Sensing 31,
Williams, D.L., Goward, S., Arvidson, T., 2006.
Landsat: Yesterday, today, and tomorrow.
Photogrammetric Engineering and Remote
Sensing, 72, 1171-1178.
Yasin, Rubina, 1987. A check list of wild trees,
shrubs, and climbers of the National Park,
Margalla Hills, Islamabad. Biologia, 33, 149-
Yousif, M.S.M., Shedid, G.A., 1999. Remote
sensing signature of some selected basement
rock units from the central eastern desert of
Egypt. The Egyptian Journal of Remote Sensing
and Space Sciences, (1), 132-141.