See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/324238721

Environmental Monitoring of Spatio-Temporal Changes in Northern Istanbul

using Remote Sensing and GIS

Article in International Journal of Environment and Geoinformatics · April 2018

DOI: 10.30897/ijegeo.410943

CITATIONS READS

11 793

3 authors:

Çiğdem Göksel Ruusa-Magano David

Istanbul Technical University Istanbul Technical University
87 PUBLICATIONS 633 CITATIONS 8 PUBLICATIONS 51 CITATIONS

SEE PROFILE SEE PROFILE

Ahmet Ozgur Dogru

Istanbul Technical University
101 PUBLICATIONS 458 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Integrating Ecosystem Services in Urban Planning (URBAN-NET) View project

Compilation of Digital Elevation Model for Turkey in 3-ARC-Second Resolution by Using SRTM Data Supported with Local Elevation Data View project

All content following this page was uploaded by Ahmet Ozgur Dogru on 15 July 2019.

The user has requested enhancement of the downloaded file.

 ISSN: 2148-9173 Vol: 5 Issue:1 March 2018

International Journal of Environment and Geoinformatics (IJEGEO) is an international,

multidisciplinary, peer reviewed, open access journal.

Environmental Monitoring of Spatio-Temporal Changes in

Northern Istanbul using Remote Sensing and GIS

Çiğdem Göksel, Ruusa Magano David and Ahmet Özgür Doğru

Editors

Prof. Dr. Cem Gazioğlu, Prof. Dr. Dursun Zafer Şeker, Prof. Dr. Ayşegül Tanık,

Prof. Dr. Şinasi Kaya, Assist. Prof. Dr. Volkan Demir

Scientific Committee (2018)

Assist. Prof. Dr. Abdullah Aksu, Prof. Dr. Bedri Alpar, Prof. Dr. Gülşen Altuğ, Prof. Dr. Lale Balas,
Prof. Dr. Can Balas, Prof. Dr. Levent Bat, Prof. Dr. Bülent Bayram, Prof. Dr. Nuray Çağlar, Prof.
Dr. Jadunandan Dash, Prof. Dr. A. Evren Erginal, Assoc. Prof. Dr. Ali Ertürk, Dr. Dieter Fritsch,
Dr. Amin Gharehbaghi, Prof. Dr. Melike Gürel, Assoc. Prof. Dr. Tolga Görüm, Dr. Hakan Kaya ,
Prof. Dr. Fatmagül Kılıç, Assoc. Prof. Dr. Maged Marghany, Prof. Dr. Nebiye Musaoğlu, Prof. Dr.
Masafumi Nakagawa, Prof. Dr. Haluk Özener, Prof. Dr. Erol Sarı, Prof. Dr. Elif Sertel, Prof. Dr.
Nüket Sivri, Assoc. Prof. Dr. Füsun Balık Şanlı, Prof. Dr. Uğur Şanlı, Assoc. Prof. Dr. Hasan
Özdemir, Prof. Dr. Taşkın Kavzoğlu Assoc. Prof. Dr. Oral Yağcı, Dr. Soofia Tahira Elias Özkan,
Prof. Dr. Seyfettin Taş, Assist. Prof. Dr. Ömer Suat Taşkın, Dr. İnese Varna, Assoc. Prof. Dr. İ.
Noyan Yılmaz, Assist. Prof. Dr. Baki Yokeş, Assit. Prof. Dr. Sibel Zeki

Abstracting and Indexing: DOAJ, Index Copernicus, OAJI, Scientific Indexing Services, JF, Google Scholar, Open Access
 International Journal of Environment and Geoinformatics 5(1): 94-103 (2018)

Environmental Monitoring of Spatio-Temporal Changes in Northern

Istanbul using Remote Sensing and GIS
Çiğdem Göksel1,* , Ruusa Magano David2, Ahmet Özgür Doğru1
1
Faculty of Civil Engineering, Geomatics Engineering Department, Istanbul Technical University, 34469 Maslak, Istanbul,
Turkey
2
Department Istanbul Technical University, Institute of Informatics, Geoinformation Technology Programme, 34469
Maslak, Istanbul, Turkey

Corresponding author* Received 09 March 2018

Tel : +90 212 285 3806 Accepted 31 March 2018
E-mail :göksel@itu.edu.tr

Abstract

This study aims to evaluate the landscape transformation and the effect of the third Bosphorus Bridge
construction works on the natural ecosystem over seven years’ time in two districts, Beykoz and Sarıyer,
located at the northern part of Asian and European Istanbul. The study applied supervised classification-
maximum likelihood algorithm to monitor land cover changes using multi-temporal satellite data of Landsat
covering study area for the years of 2009 and 2016. The overall classification accuracies are 92.36% with
Kappa statistic of 0.89 for 2009 and 91.25% with Kappa statistic of 0.89 for the year 2016. The current study
resulted with a rapid land cover change at the study area with approximately 4% and %1 decrease in forest and
green area, 2% increase in both agricultural and urban/build up areas. Additionally, the destroyed area
increased with approximately 1% of the study area in the 7 years period.
.

Keywords: Change detection, Land use/cover, Landsat 8 OLI, Supervised classification

Introduction alterations is the remarkable expansion

overseen in urban areas and the direct or
Change is an inevitable aspect of the millennial indirect population growth. The demand for
dynamic and ever growing cities. One of the LU/LC such as built up areas and roads
primary issues mentioned and recognized in increase according to the increase in the
global environmental approaches is land population in a specific area. Therefore,
use/land cover (LU/LC) changes. Nevertheless, research on how the land is transformed by
these changes can be considered in relation urbanization and especially land conversion of
with all aspects at once including the political, peri-urban environments has been identified as
socio economic and climatic conditions and a pivotal area of future land change research to
factors of the area in research. Many avoid certain negativities which may affect not
researchers from various disciplines have been only the environment but also its inhabitants.
exploring the extension and the over-population
of the cities along with LU/LC depending on The actuality, efficiency and certainty of the
the growth and the urban sprawl. LU/LC data and information are crucial for a
better understanding and evaluation of the
The transformation of forests and green areas to consequences of the above-mentioned changes.
urban development areas have severely altered Through the past decade, remote sensing
not only the quantity but also the quality of the techniques have been developed, which now
land leading to inevitable problems of soil have a proven record of accuracy in LU/LC
erosion and deforestation as an unquestionable maps and monitoring in a given time period
consequence of LU/LC (Wu, J. 2008). Another (Zhu, Z. and Woodcock, C. 2014; Ibrahim, M.,
noticeable factor that activates LU/LC 2014; Rahman, M. 2016). Moreover, the brief

DOI: 10.30897/ijegeo.
 Göksel et al., / IJEGEO 5(1): 5(1): 94-103 (2018)

and quick developments in remote sensing of information on LU/LC and the possibilities
technologies have led the LU/LC researches to for an optimization while meeting the
be more effective in terms of tools and data increasing demand for the human needs.
along with being actual and effective in Another point to make is the dynamic nature of
spotting the changes. This is why the Landsat LU/LC and the inevitable effect of the
Images are the main sources of data in anthropogenic factors’ altering power on the
considering and understanding deforestation issue discussed. To be more specific, LU/LC in
and urban growth patterns of the cities (Coppin the Northern parts of Istanbul’s metropolitan
P. and Bauer, M., 1996; Woodcock, C., 2001). area is constantly changing over space and time
as a result of the rapid growth of the city based
Spreading the issue to a more anthropogenic on various economic and industrialization
perspective, one can say that the efforts (Doğru et all., 2006; Göksel et all.,
transformations and impacts in the natural 2006; Şeker et all., 2016). Moreover, the mega
environment are caused by human activities. A construction projects including the third bridge,
more deepened perspective of anthropology in the sub-sea tunnels and a third airport are all
relation to the ecosystems affected in the urban traits of the increasing demand and growth of
areas has a profound relationship with LU/LC. the city. Even though these aspects are
Therefore, the above-mentioned aspects of the contributing to the social attributes of daily life
measured changes are crucially important and – and could also be considered as
valuable because of the spatial characteristics of improvements for citizens – their impacts on
LU/LC since the human activity cannot be the nature and the ecosystems are left inevitable
considered apart from the urban developments. and permanent which makes the monitoring of
Thus, detecting and mapping LU/LC change changes profoundly important and needed for
with an approach based especially on human effective decisions and to build a balance
activities using remote sensing and between the positive sides of the development
Geographical Information System (GIS) and the negative effects on the environmental
technologies aid in obtaining a clear resources caused for an increase in the socio-
understanding of the LU/LC alteration economic welfare.
processes due to land use change and their
consequences. The present study is focusing on The main objective of this study is to quantify
the megacity Istanbul, which is covering both the land use and land cover change trajectories
sides of the Bosporus and bridging Europe and with the multi-temporal Landsat imagery in the
Asia continents. Istanbul is a mega city with northern part of Istanbul Metropolitan area. The
both natural and historical texture at a strategic study evaluated the landscape transformation at
geographic location. Historical, cultural and the coastal line and the effect of the third
natural beings in Istanbul are subject to the Law Bosphorus Bridge (Yavuz Sultan Selim Bridge)
of Conservation and have been registered as on the natural ecosystem in two specific
'Cultural Heritage' or declared as "Protected districts; Beykoz and Sarıyer over a seven years
Area". In addition, the Bosphorus area is period using GIS and remote sensing
protected with the provisions of a special law. technologies.
However, even though the municipalities in
Turkey are known to have great powers, central Materials and Methods
government institutions (The Ministry of
Environment and Urban Development, The Study Area, Data and Methodology
Ministry of Forestry and Water Affairs, The Istanbul is a megacity with over 15 million
Ministry of Transport) are effective in inhabitants and the dominant center of
projecting and constructing major projects. The manufacturing, services, education and culture
study examines the land use/cover changes in which produces 40% of whole Turkey’s
Istanbul basically arising from such projects. industrial activities. Istanbul consists of 41
districts all belonging to Istanbul Metropolitan
Two of the main reasons of the land becoming Municipality. Beykoz and Sarıyer Districts
a scarce resource are the agricultural and have been selected as the study area in this
demographic pressure that requires extraction research, covering an area of approximately

95
 Göksel et al., / IJEGEO 5(1): 5(1): 94-103 (2018)

55258 ha (Figure 1). The Beykoz and Sarıyer region consisting of forests, green areas, basins,
Districts are located at the northern section of agricultural land and wetlands which are has
Asian and European sides of Istanbul biosphere reserve potential (Gazioğlu, et al.,
respectively, which is the entrance of the 1997; Simav et al., 2015). Among the areas
Bosphorus (the Strait of Istanbul) from the facing ecological disintegration are
Black Sea on both sides. The two districts have municapalities where risk levels must be
been affected by a mega 3rd Bridge urgently taken into account (Burak et al., 2004;
transportation project, as it is located in a very Gazioğlu et al., 2016).
precarious ecosystem in the Istanbul’s northern

Fig 1. The location of the study area

Two Landsat satellite imageries are used as the The reported study has been carried out using
main data for LU/LC classifications in the Erdas Imagine 2014 and Arcmap 10.x software.
study area for the years of 2009 and 2016. For The detailed methodology adopted in the study
the year of 2009, Landsat Enhanced Thematic is given in Figure 2.
Mapper (ETM+) at a resolution of 30 m and
Landsat Operational Land Imager (OLI) also at 2.1. Geometric Correction
a resolution of 30 m have been utilized. The An accurate registration of multispectral remote
Landsat data covering study area were obtained sensing data is essential for analyzing land use
from USGS Earth Resources Observation and land cover conditions of a particular
Systems data center (https://www.usgs.gov). In geographic location. The geometric correction
the study, 1:25000 scaled standard topographic error between two images is expressed in terms
maps and 1:5000 scaled ortho-photo maps and of total root mean square error (RMSE), which
google earth imageries have been used as the represents a measure of deviation of corrected
ancillary data. ground control point (GCP) coordinate values
from the original reference GCPs used to
Two main steps were involved in this research, develop the correction model (Balik Sanli et all.
which are the classification of the satellite data 2008). Change detection analysis is performed
for LU/LC and the change detection analysis in on a pixel-by-pixel basis; therefore, any mis-
the LU/LC types. Image registration, ortho- registration greater than one pixel will provide
rectification, classification, and change an anomalous result of that pixel. To overcome
detection using post-classification comparison this problem, several authors recommend that
were involved in the satellite data analysis. the root mean-square error (RMSE) between

96
 Göksel et al., / IJEGEO 5(1): 5(1): 94-103 (2018)

any two dates should not exceed 0.5 pixels rectification process, 22 GCP (Ground Control
(Kardoulas, N. et all 1996; Gong, P. and Xu, B. Points) distributed homogeneously over the
2003; Jensen, J. 2004) images were used. In terms of image
rectification, a nearest neighbor resampling
In this study, Landsat images were algorithm was used. The total RMSE for 2009
georeferenced to the Universal Transverse and 2016 images were 0.34 and 0.32 pixels,
Mercator projection system (UTM) using 1:25 respectively.
000 scaled standard topographic maps. For the

Figure 2. Flowchart of methodology for land use/land cover change

2.2. Image Classification 2.3. Accuracy assessment of image

classification
Image classification is usually performed by Accuracy Assessment of classified images was
per pixel labeling of raster data, in many carried out to determine the quality of
instances the desired result of image information derived from the data. This is
classification is a vector-based map of essential as the user of LU/LC maps has the
polygons (Lunetta R., et all. 2006). In this need to know how accurate the maps are in
study, maximum likelihood classification order to use the data more correctly and
(MLC) was used to derive LU/LC categories efficiently (Jensen, J. 2004). Error matrices as
of selected study area. In the supervised cross-tabulations of the mapped class vs. the
classification technique, the maximum reference class were used to assess
likelihood algorithm classifies the image classification accuracy (Smits, P. et all. 1999;
based on the training sets (signatures) Congalton, R. and Green, K. 1999). For the
provided by the user on the basis of their own accuracy assessment of land cover maps
knowledge or experiences. In this method, extracted from satellite images, stratified
the pixel is assigned to the class to which the random method was used to represent different
probability of the pixel belonging is the land cover classes of the area. The accuracy
highest (Weber, R. and Dunno, G.2005). assessment was carried out using 300 points
that were selected randomly and analyzed based
on ground truth data and visual interpretation.

97
 Göksel et al., / IJEGEO 5(1): 5(1): 94-103 (2018)

2.4. Land use/cover change detection The results reveals that from 2009 to 2016, the
Post-classification comparison change detection study area experienced land cover changes, the
algorithm was employed by the study to forest area occupied about 47.9% of the total
determine changes in land cover in a seven area in 2009 and has decreased to 44.6% in
years span. Two classified image pairs were 2010. Green areas were also reduced from 9.8%
compared using cross-tabulation in order to to 8.7%, respectively. Water bodies and bare
determine qualitative and quantitative aspects land also experienced a reduction in the seven-
of the changes for the periods from 2009 to year period. On the other hand, urban and build
2016. Thus, a change matrix was produced up land have increased from 6.6% to 8.7%,
from the two class maps, containing different agricultural land was 6.5% in 2009, this number
combinations of ‘‘from–to’’ change increased to 8.4% in 2016. As it can be seen in
information. Landscape transformations that Table 2, although there is no LU/LC class
have occurred i.e. the gains and losses in each named as destroyed area in the year of 2009,
category between 2009 and 2016 were then this class appeared with approximately 1% of
compiled and mapped. This information will the whole area in the year of 2016. Open
reveal both the desirable and undesirable mining area was 1.1% in 2009 but experienced
changes and classes that are “relatively” stable an increment of 1.7% in 2016. Other land cover
overtime. classes such as roads, sand dunes and open
mining areas have also experienced an increase
Results in the course of seven years. Table 2 reveals
The overall accuracy, along with the user and that both positive and negative changes
producer’s accuracies, and the kappa statistic occurred in the LU/LC pattern of the Beykoz
for each class were derived from the error and Sarıyer Districts.
matrix. The achieved overall classification and
Kappa coefficient are presented in Table 1. To have a further understanding of land
encroachment for different land categories
Table 1. Overall classification accuracies (%) during the seven years period, matrices of land
and kappa statistics cover changes from 2009 to 2016 (Table 3) was
Overall prepared. In the table, unchanged pixels are
Year Kappa Statistic located along the major diagonal of the matrix.
accuracy
2009 92.36 0.90 When the direction of the change has been
evaluated, out of 26504.9 ha that was forest in
2016 91.25 0.89 2009, 22368.7 ha was still forest in 2016 but
2030.4 ha was converted to green areas, 820 ha
Figure 3 shows the derived LU/LC maps was converted to agricultural land, 512.8 ha
showing the changes in land use and land cover was converted to bare land and 227.9 ha
of the study area. These images provide the became destroyed areas in 2016. Additionally,
information about the land use pattern of the at the same time the increase of agriculture,
Beykoz and Sarıyer. In the classification, nine from 2009 to 2016, was mainly from forest area
classes were produced for the 2009 image and green areas. The agricultural land gained
(water, forest, green areas, agricultural land, 1308.8 ha from green areas and 820 ha was
urban and build-up, roads, sand dunes, bare obtained from forest areas as mentioned before.
land and open mining areas). For 2016, ten The class that destroyed area mainly replaced in
classes were produced. With the nine classes as 2016 was forest, followed by green areas and
in 2009, an additional class of destroyed areas agricultural land. The destroyed area replaced
was added. This class is mainly consists of the around 227.9 ha of forest area and it took equal
transportation network of third bridge portion of land from the green areas and the
construction. The 2009 image and the 2016 agricultural land, 45.4 ha from each class. The
image were compared in terms of the total area results show that the most important land cover
of each LU/LC category. Table 2 present those types in the northern areas were affected by the
statistical results of classification. construction of the third bridge and so as a
large portion of the forests in comparison to
other classes.

98
 Göksel et al., / IJEGEO 5(1): 5(1): 94-103 (2018) 2018)

Figure 3. LU/LC in 2009 and 2016

99
 Göksel et al., / IJEGEO 5(1): 5(1): 94-103 (2018)

Table 2. Summary of statistics for 2009 and 2016

Land Cover Type 2009 Percentage 2016 Percentage Sum of Change
(TM) of the area (OLI) of the area the area Between
area area (ha) (ha) Both
(ha) Years
Water 11733.8 21.2 11513.1 20.8 -220.7 -0.4
Forest 26504.9 47.9 24654.6 44.6 -1850.3 -3.3
Green areas 5463.3 9.8 4838.3 8.7 -624.9 -1.1
Agricultural land 3625.2 6.5 4653.5 8.4 1028.3 1.9
Urban and build up 3675.6 6.6 4851.1 8.7 1175.5 2.1
Roads 723.9 1.3 825.5 1.5 101.6 0.2
Sand dune 107.5 0.2 122.4 0.2 14.9 0.03
Bare land 2835.6 5.1 2340 4.2 -495.6 -0.9
Open mining area 589 1.1 984.3 1.7 395.3 0.7
Destroyed Area 0 0.0 476 0.8 476 0.9
TOTAL 55258.8 100 55258.8 100

Table 3. Matrices of land use land cover (ha) from 2009 to 2016
TOT.
LC Type W F GA AL U&B R SD BL OA DA
2016
Water-W 11474.4 18.1 0.30 4.1 0.9 2 1 0.4 11.9 0 11513.1
Forest-F 13.8 22368.7 1478.1 200.4 13.3 36.2 0 541.6 2.5 0 24654.6
Green areas_GA 1.00 2030.4 1531.4 661.1 49 68.7 5.8 458.9 32 0 4838.33
Agricultural Land-AL 2 820 1308.8 1271.1 293.9 115.4 14.7 768.8 58.8 0 4653.5
Urban &Build up-
11.7 203 301.8 640.1 2875.5 225.5 15.3 441.6 136.6 0 4851.1
U&B
Roads-R 71.2 77.7 79.3 110.7 179.4 145.7 5 72.2 84.3 0 825.5
Sand Dunes-SD 43.2 10.1 3 10 11.1 5.8 17.6 2.6 19 0 122.4
Open Mining Areas-
10.5 512.8 624.5 525.4 117.2 53.2 10.8 421.7 63.9 0 2340
OA
Bare land- BL 98.6 236.2 90.7 156.9 92.5 54.9 24.2 95 135.3 0 984.3
Destroyed Areas- DA 7.4 227.9 45.4 45.4 42.8 16.5 13.1 32.8 44.7 0 476
TOTAL 2009 11733.8 26504.9 5463.3 3625.2 3675.6 723.9 107.5 2835.6 589 0 55258.8

100
 Göksel et al., / IJEGEO 5(1): 5(1): 94-103 (2018)

Urban and build up increased from 3675.6 ha in into the “destroyed area class” which represent
2009 to 4851.1 ha in 2016. It replaced 640.1 ha the ring of the Yavuz Sultan Selim Bridge.
of agricultural land and 301.8 ha of green areas.
Around 99 ha of water was replaced by the This study elucidates the significance of
open and mining areas, whereas the area of incorporating Remote Sensing and GIS for
other land cover classes replaced by water was change detection study of land cover/land use
relatively small. of an area as it offers essential information
about the spatial distribution as well as nature
A While the above change statistics shed a little of landscape transformation.
light on the question of where land use changes
are occurring, constructing a change detection Discussion and Conclusion
map Figure 4 can show the exact changes in the The land cover changes between different times
land cover types and can be much more Interval have been visualized and analyzed by
appreciated.
using remote sensing technology. The results
The third bridge construction along the clearly reveal that LU/LC changes were
Bosphorus was referred to as (destroyed areas significant during the period between 2009 and
class) in the study and the most important 2016. Also, Landsat classifications can be
conversion in the Beykoz and Sarıyer areas. It utilized to produce accurate land cover change
can also be seen in the Figure 4 that the major maps and statistics that can be used to monitor
area of forest was converted to the the changes occurred and its impacts on the
transportation ring along the Bosphorus on the environment have been demonstrated. It can be
European and Asian side of Istanbul city. The seen that there is a significant decrease in
dark brown colour in Figure 5 along the forest, green areas density in the northern part
Bosphorus trait represent the forest area turned

Figure 4. Forest areas in Beykoz and Sarıyer Districts

of Istanbul where main changes appeared from dramatic LU/LC changes that took place in the
2009 to 2016. It was observed that the study area.
developmental activities and the construction of
new roads and bridges as a result of It is obvious that Bosphorus bridges linking
urbanization were the major reasons for the Asia and Europe serve a high portion of urban

101
 Göksel et al., / IJEGEO 5(1): 5(1): 94-103 (2018)

traffic with a high capacity due to increasing Congalton, RG. and Green, K. (1999)
population and intercontinental traffic. In this Assessing the accuracy of remotely sensed
sense, the Third Bosphorus Bridge will data: principles and applications. Lewis
contribute to reducing this density with 8 lane Publishers, Boca Raton, Fla, 43-64.
roads and 2 high-speed railway lines. Beykoz Coppin, PR. and Bauer, ME. (1996). Digital
and Sarıyer, the study areas, are the most change detection in forest ecosystems with
populated areas of forest and shrubbery in remote sensing imagery. Remote Sensing
Istanbul. Based on the previous experiences on Reviews, 13(3–4), 207–234.
first and second bridge construction projects in Doğru, OA.; Bektas Balcık, F.; Göksel, Ç. and
Istanbul, it is known that such projects Uluğtekin, N. (2006). Monitoring coastal
promoted the urban growth around the project dunes by using remote sensing and GIS
sites and large amount of agricultural and forest integration in northwest Turkey: A case
lands were damaged. The study introduced that study of Kilyos dunes. Fresenius
the change arisen from the construction of the Environmental Bulletin PSP,Volume 15 –
3rd Bridge and its service roads is not large No 9b, pp 1216-1220.
enough, for both districts, rent-focused Gazioğlu C, Yücel Y Z, Burak S., Okuş, E. and
unplanned urbanization episodes are considered Alpar, B. (1997). Coastline change and
as a potential danger. Decisions and sanctions inadequate management between Kilyos
for protection to be taken by local governments, and Karaburun shoreline. Turkish Journal of
as well as government agencies' decision- Marine Sciences, 3(2): 111–122.
making mechanisms, will contribute to further Gazioğlu, C., Akkaya, M.A., Baltaoğlu, S. and
deterioration and protection of the natural areas Burrak, S.Z. (2016). ICZM and the Sea of
and land of the region. Marmara: The İstanbul Case. The Sea of
Marmara: Marine Biodiversity, Fisheries,
These results have highlighted the increasing Conservations and Governance (Editors:
rate of modification of forest ecosystem by Özsoy, E., Çağatay, M.N., Balkıs, N.,
anthropogenic activities and the need to Balkıs Çağlar, N., Öztürk, B.): 935-957.
apprehend the situation to ensure sustainable Göksel, C.; Musaoğlu, N.; Gürel, M.;
environmental management. The quantification Uluğtekin, N.; Tanık, A. and Şeker, DZ.
of LU/LC changes of Beykoz and Sarıyer area (2006). Determination of land-use change in
is very useful for environmental management an urbanized district of Istanbul via remote
groups, policy makers and for public to better sensing analysis. Fresenius Environmental
understand the impacts the developmental Bulletin (FEB), 15(8A): 798-805.
activities have on the green areas and forested Gong, P. and Xu, B. (2003). Remote sensing of
land in the area and the surrounding. The forests over time: changes types, methods,
general LU/LC changes trends determined in and opportunities. In Remote Sensing of
this study, and the variable outputs associated Forest Environments: Concepts and Case
with these LU/LC, can be used as input by Studies, Woulder M, Franklin SE (ed), 301–
decision makers during the planning phase of 333, Amsterdam: Kluwer Press.
the similar mega projects. İbrahim, Mahmoud, M.; Duker, A.; Conrad, C.;
Thiel, M.; Shaba and Ahmad, H. (2016).
References Analysis of settlement expansion and urban
growth modelling using geoinformation for
Balık, Şanlı, F.; Bektas, Balcık, F. and Göksel, assessing potential impacts of urbanization
Ç. (2008). Defining temporal spatial on climate in Abuja City, Nigeria. Remote
patterns of mega city Istanbul to see the Sensing, 8(3): 220.
impacts of increasing population. Jensen, JR.. (2004). Introductory Digital Image
Environmental Monitoring and Assessment, Processing: A remote sensing perspective,
146: 267–275 (3rd ed.), Upper Saddle River, NJ: Prentice
Burak, S., Doğan, E. and Gazioğlu, C. (2004). Hall.
Impact of urbanization and tourism on Kardoulas, NG.; Bird, AC. and Lawan, AI.
coastal environment. Ocean and Coastal (1996). Geometric correction of spot and
Management 47(9–10): 515–527. Landsat imagery: A comparison of map-

102
 Göksel et al., / IJEGEO 5(1): 5(1): 94-103 (2018)

and GPS-derived control points.

Photogrammetric Engineering and Remote
Sensing, 62: 1173–1177.
Lunetta, RS.; Knight, JF.; Ediriwickrema, J.;
Lyon, JG. and Worthy, LD. (2006). Land-
cover change detection using multi-
temporal MODIS NDVI data. Remote
Sensing of Environment, , 105(2): 142–154.
Rahman, MT. (2016). Detection of Land
Use/Land Cover Changes and Urban Sprawl
In Al-Khobar, Saudi Arabia: An Analysis of
Multi-Temporal Remote Sensing Data.
ISPRS International Journal of Geo-
Information. 5(2): 15.
Şeker D.Z., Tanık, A., Citil, E., Ozturk, I.,
Ovez, S. and Baycan, T. (2016). Importance
and Vulnerability Analyses for Functional
Zoning in a Coastal District of Turkey.
International Journal of Environment and
Geoinformatics. 2016. Vol 3 (3): 76-91.
Simav, Ö., Şeker, D.Z., Tanık, A., Gazioğlu, C.
(2015). Determining the endangered fields
of Turkish coasts with coastal vulnerability
index. Journal of Map, 153: 1-8.
Smits, PC.; Dellepaine, SG. and Schowengerdt,
RA. (1999). Quality assessment of image
classification algorithms for land cover
mapping: A review and a proposal for a cost
based approach. International Journal of
Remote Sensing, 20: 1461–1486.
Weber, R.M and Dunno, G.A. (2001). Riparian
vegetation mapping and image processing
techniques, Hopi Indian Reservation.
Photogrammetry Engineering Remote
Sensing. 67(2), 179–186.
Woodcock, CE.; Macomber, SA.; Pax-Lenney,
M. and Cohen, WB. (2001). Large area
monitoring of temperate forest change using
Landsat data: generalization across sensors,
time and space.Remote Sensing of
Environment. 78(1–2): 194–203.
Wu, J. (2008). Land Use Changes: Economic,
social, and environmental impacts.
Choices., 23(4):6–10.
Zhu, Z. and Woodcock, CE. (2014).
Continuous change detection and
classification of land cover using all
available Landsat data. Remote Sensing of
Environment, 144, pp. 152–171.

103

View publication stats