Você está na página 1de 11

Journal of Research in Ecology An International Scientific Research Journal

ORIGINAL RESEARCH

Modeling the spatiotemporal pattern of farmland change in rural regions of


Ahvaz County by remote sensing and landscape metrics
Authors: ABSTRACT:
Journal of Research in Ecology

Hassanali Faraji Sabokbar1 Land-use and land cover change (LUCC) has engrossed much attention due to
Seyed Hassan Motiee
its effects on global and regional environmental change. The spatial pattern of LUCC
Langrood2 and
Hossein Nasiri3 can reflect underlying human activities, involving urbanization processes and policies
for social and economic development at local to region scales. Landscape pattern
metrics are important for evaluating ecological processes and effects of LUCC. This
study quantitatively analyzed spatiotemporal changes in land use and landscape
pattern in a Ahvaz region of southwest Iran by comparing classified satellite images
from 1986, 1998, and 2014, using a GIS, remote sensing, and landscape pattern
metrics. The results showed an increase in farm lands during 1986–2014. Over the
study period, 308% of newly-expanded farm lands were from bare lands and range
lands. This process has brought about noticeable land use changes and farming
Institution: growth at an unprecedented scale and rate, and consequently given rise to
1.Associate professor, substantial impacts on the landscape pattern. The results also disclosed that bare
University of Tehran. lands and range lands were the major resources that were converted for farming
2.Professor, University of development. Establishment of sugarcane agro-industry companies and policies of
Tehran. decision makers will increase positive significant impacts on agro ecosystem and
3.PhD Student, University of environment. To sum up, Trend analysis of landscape pattern metrics demonstrates
Tehran. integration of the farming landscape, with landscape pattern structure becoming
more homogeneous over the last three decades in the Ahvaz County.

Keywords:
Land use Changes, Landscape Pattern, Decision Tree, Spatial Metrics, Rural
Areas of Ahvaz county.

Corresponding author:
Hossein Nasiri Article Citation:
Hassanali Faraji Sabokbar, Seyed Hassan Motiee Langroodi and Hossein Nasiri
Modeling the spatiotemporal pattern of farmland change in rural regions of Ahvaz
County by remote sensing and landscape metrics
Email ID: Journal of Research in Ecology (2016) 4(1): 083-093
Dates:
Received: 05 May 2016 Accepted: 31 May 2016 Published: 11 July 2016

This article is governed by the Creative Commons Attribution License (http://creativecommons.org/


Web Address: licenses/by/4.0), which gives permission for unrestricted use, non-commercial, distribution and
http://ecologyresearch.info/ reproduction in all medium, provided the original work is properly cited.
documents/EC0080.pdf

Journal of Research 083-093 | JRE | 2016 | Vol 4 | No 1


in Ecology
An International www.ecologyresearch.info
Scientific Research Journal
Faraji Sabokbar et al., 2016

INTRODUCTION change, therefore, has attracted increasing attention to


Land Use and Land Cover Change (LUCC) remote sensing data for understanding spatiotemporal
significantly contributes to biodiversity loss, invasive changes, and managing and restoring, the rural-urban
species spread, changes in biogeochemical cycles, and ecosystems.
the loss of ecosystem services (Radeloff et al., 2012). Furthermore, spatial resolution of remote sensing
Planning for a sustainable future of Land Use and Land data and, image processing methods are also very crucial
Cover Change (LUCC) is a complex process (Irwin and for the technical limitation of LUCC detection.
Geoghegan, 2001; Lambin and Geist, 2006) and Developing an improved comprehending on how to
modeling these systems are challenging. match applications and change detection techniques are
Rapid urban development, especially in the still faced to several challenges. In spite of numerous
developing world, will continue to be one of the crucial criteria affecting the choice of appropriate change
issues of global change in the 21st century (Sui and detection methods, image differencing, Principal
Zeng, 2001). This has resulted in an unprecedented scale Component Analysis (PCA) and post classification are
and rate of urban expansion in Iran over the last three most commonly applied methods and usually
decades. However, rapid urban growth and exploitation demonstrate better performance compared with other
of natural resources have led to important impacts on technics (Collins et al., 1996); Yuan and Elvidge, 1998;
ecosystem structure, function and dynamics which have Lu et al., 2004; Jensen 2005).
made urban and rural as fragile regions. This is Much less known than remote sensing, spatial
particularly the issue in the economic-developed regions metrics can be a useful tool for objectively quantifying
where dramatic rural-urban expansion and LUCC have the structure and pattern of environment change from
induced serious environmental issues threatening rural- thematic maps (Herold et al., 2005). Spatial metrics are
urban sustainable development (Yeh and Li 1999; Weng, applied in landscape ecology, known as landscape
2001; Ji et al., 2001; Li and Yeh, 2004; Chen et al., metrics. Changes of landscape pattern can be detected
2005; Xiao et al., 2006; Liu et al., 2007). Accurate and and delineated by the landscape metrics, which quantify
timely information on the status and trends of ecosystem and classify complex landscape into identifiable patterns

Figure 1. Location of Ahvaz County at Khuzestan province, Iran


084 Journal of Research in Ecology (2016) 4(1): 083-093
Faraji Sabokbar et al., 2016

Table 1. Characteristics of remote sensing and ancillary data in this study


Sensor type Acquisition date WRS Path WRS Row Radiometric Spatial
Resolution Resolution
TM 08/05/1986 165 38 1,2,3,4,5,7 30m
TM 09/05/1998 165 38 1,2,3,4,5,7 30m
TM 05/05/2014 165 38 1,2,3,4,5,7 30m
Dem - - - 1 10m

and disclose some ecosystem characteristics that are not point of view and making preliminarily exploration of
directly observable (Antrop and Van Eetvelde, 2000; the influence of urban development on these changes.
Turner et al., 2001; Wseng, 2007). In last decades, there The overall objective of this study is to improve the
has been an increasing interest in using spatial metrics understanding of the effects of landscape patterns and
method in rural and urban studies because of their use in provide the basic information for distinguishing the
bringing out the spatial component in urban-rural ecological impacts and proper decision-making towards
structure and in the dynamics of change and growth environment sustainable development.
process (Deng et al. 2009). Furthermore, spatial metrics
have the potential for more detailed analyzes of the MATERIALS AND METHODS
spatiotemporal patterns of LUCC, and the assessment of Study Area
environmental process. So, the integrated application of Ahvaz is the capital of Khuzestan province, on
remote sensing and spatial metrics provide more spatially Iran’s south west and is located between 31°24′ and 31°
consistent and detailed information about environmental 38′ latitude and 50°58′ and 51°11′ longitude (Figure 1).
structure and change. The average level elevation of all residential areas is
Since Iran faced land reforms, urban located at the range of about 18 meters. This area is
development and a war in 1960-1988, tremendous land- located in the west of the central Zagros Mountain,
use change has occurred in many rural regions of Iran which is a region of complex geography located at the
such as the Khuzestan Province. As a case study in intersection of large-scale atmospheric circulations and
Ahvaz City, one of the most rapidly developing cities in major weather fronts effect on this area is from
Iran, this paper investigates the spatiotemporal variations Mediterranean Sea. Therefore less rainfall occurs in this
and evolution of land use and landscape pattern in region; consequently such a situation also occurs in the
response to the rapid urbanization and environmental Ahvaz County. However, the annual average
process by combining remote sensing technics and precipitation is about 213.4 mm, of which 90% occurs
spatial metrics. Bearing in mind the importance of between December and April. The annual average
accurate and consistent land use change information over maximum and minimum temperatures in this district are
a long-term, and finer resolution of remote sensing data 30° and 21°, respectively. Also, January is the coldest
in concert with the urban properties, this study proposes month and July is the hottest month in this region. The
a feasible and cost-effective land use change detection agricultural wells are the most important water resources
method by adopting a time series analysis of TM in this plain. In recent years, due to the rapid
imagery spanning the last 29 years (1986–2014). development of agricultural lands and long-term drought
Specifically, this study focuses on analysis of the in these areas, the groundwater storage volume has
spatiotemporal dynamics of LUCC, a comprehensive decreased sharply. Consequently, about half of the wells
investigation of landscape pattern from a comprehensive were dried and others have low water in this plain.

Journal of Research in Ecology (2016) 4(1): 083-093 085


Faraji Sabokbar et al., 2016

geometric correction, atmospheric effects were


normalized by dark object subtraction (DOS). Studies
have shown that these corrections are useful in the
classification and monitoring of LUCC (Liu et al., 2008;
Olthof et al., 2005; Röder et al. 2005; Song et al., 2001).
In this paper, all image processing was implemented
using ENVI 4.5 software. The information from satellite
images and groundwater resource data are shown in
Table 1.
Figure 2. General decision tree hierarchy
Land use change detection method
(Richards and Richards, 1999).
In the last decades, a variety of methods for land
Remote sensing data selection use/cover classification has been investigated including
Landsat Thematic Mapper (TM) satellite images maximum likelihood, minimum distance, and
(Path/Row 165 and 38) and ancillary data were used for mahalanobis distance. The classifiers mentioned above
analysis. have all been single stage in that only one decision is
The TM images of the study area were taken in made about a pixel, as a result of which, it is labeled as
May of the years 1986, 1998, and 2014. Preprocessing of belonging to one of the available classes or is left
the satellite images prior to image classification and unclassified. Multi-stage classification techniques are
change detection process is necessary and commonly also possible in which a series of decisions is taken in
comprises a series of sequential operations, containing order to determine the correct label for a pixel. The more
atmospheric correction or normalization, geometric common multi-stage classifiers are called decision trees
correction, and image enhancement (Coppin and Bauer (DT).
1996). Geometric registration of these data sets was They comprise of a number of connected
performed by image-to-image registration tie down classifiers (or decision nodes) none of which is expected
through a first-order polynomial fit. The dates of imagery to carry out the complete segmentation of the image data
were geo-referenced to a universal transverse Mercator set. Instead, each component classifier only performs
(UTM) projection with an RMSE of 0.5 pixels. After part of the task (Richards and Richards, 1999) (Figure 2).

Table 2. Spatial metrics selected in this study


Metrics Description Unit
Mean Shape Index (MSI) Measures the ratio between the perimeter of a patch and the perim- metres/
eter of the simplest patch in the same area hectare
Number of Patch (NP) Total number of patches in the landscape none
Mean Patch Size (MPS) The area occupied by a particular patch type divided by the num- square meter
ber of patches of that type
Patch Size Coefficient of Coefficient of variation of patches percent
Variance (PSCOV)
Area Weighted Mean Patch Area weighted mean patch fractal dimension is the same as mean none
Fractal Dimension (AWMFD) patch fractal dimension with the addition of individual patch area
weighting applied to each patch. Because larger patches tend to be
more complex than smaller patches, this has the effect of deter-
mining patch complexity independent of its size. The unit of meas-
ure is the same as mean patch fractal dimension.
Mean Patch Edge (MPE) Average amount of edge per patch. Example: Mean Patch Edge metres/patch
Conifer
086 Journal of Research in Ecology (2016) 4(1): 083-093
Faraji Sabokbar et al., 2016

Figure 3. Classified TM images of Ahvaz County in three years 1986, 1998, and 2014 and Changes of each land
use; Range land, Farm land, Water bodies, Built up land, and bare land in three periods: 1986–1998, 1998–
2014, and c 1986–2014.

Perhaps the simplest type is the binary tree in which each and configuration. Landscape composition refers to
component classifier, or node, is expected to perform a features associated with the presence and amount of each
segmentation of the data into only one of the two patch type within the landscape, but without being
possible classes, or groups of classes. In this study, the spatially explicit. Landscape configuration refers to the
attributes of the samples fell into two categories: A) physical distribution or spatial character of patches
remote sensing data, including the digital numbers of within the landscape. There are various indexes for
each TM bands (scaled reflectance values), and describing landscape patterns quantitatively (Mc Garigal
Normalized Difference Vegetation Index (NDVI) values et al., 2002).
and B) GIS data, including Digital Elevation Model In this paper, Mean Shape Index (MSI), Patch
(DEM) and slope map. Size Coefficient of Variance (PSCOV), Area Weighted
Landscape pattern indexes Mean Patch Fractal Dimension (AWMFD), Number of
Landscape pattern indexes are the quantitative Patch (NP), Mean Patch Size (MPS), and mean patch
descriptions of landscape patterns. Landscape patterns edge (MPE) at class level were chosen to be integrated
are recognized by spatial connection among component into the LUC model. Calculations of spatial metrics were
parts, and can be described by both their composition performed using the public domain software

Journal of Research in Ecology (2016) 4(1): 083-093 087


Faraji Sabokbar et al., 2016

Table 3. Changes of each land use; range lands, farm lands, built up lands and bare lands during 1986-1998.
Land use/Cover 1986 1998 Changes From 1986 to 1998
Classes Class Class Class Area Class Area Area Area
Area (Km) Area (%) (Km) (%) Changes Changes (%)
(Km)
Range lands 115.28 1.4 68.69 0.837 -46.58 -40.41
Farm lands 271.95 3.31 561.52 6.48 289.98 106.48
Water bodies 208 2.53 106.57 1.3 -101.35 -48.76
Built up lands 28.96 0.35 41.81 0.4 3.63 44.36
Bare lands 7588.32 92.39 7433.9 90.6 -145.58 -2.03
Total 8228.33 100 8228.33 100 - -

FRAGSTATS version 4.2. However, FRAGSTATS at a time in comparison with the other time are shown in
provides a large number of spatial metrics, a specific Figure 3
subset of them was specifically selected for this study Figure 3. Classified TM images of Ahvaz City in
which is described in Table 2. three years 1986, 1998, and 2014 and Changes of each
land use; Range land, Farm land, Water bodies, Built up
Results and Discussion land, and bare land in three periods: 1986–1998, 1998–
Spatiotemporal land use change 2014, and c 1986–2014.
The TM images was classified using decision According to Figure 3 and tables 3, 4 and 5, 40%
tree algorithm for the five land use; farm lands, range of range lands in 1986 compared with 1998 is converted
lands, built up lands, water bodies, and bare lands that to other land uses (farm land, built up land, and bare
are showed in Fig 3 in three years 1986, 1998, and 2014. land) and about 89% and 93% are decreased in over
The ‘Kappa’ coefficient of classification results for 2014 period of 1998–2014 and 1986–2014 respectively; hence,
image is about 80% and with overall accuracy 88%. the decreasing of range land leads to increased farm land.
According to Fig 3, with a quick look at the three images Furthermore, land use changes are quite different in farm
presented, an increase in farm land (bright green) area land than range land in over period of 1986-1998, 1998-
observed between 1986 and 1998. Generally, from 1986 2014, and 1986-2014. Temporal variation in farm lands
to 2014 the agricultural landscape trend towards the show a severe increasing in over period of 29 years
integration. Bare lands were the largest area among of (about 308%). This can be due to an increase in
land use types in the mentioned period. Also, decreasing sugarcane agro-industry companies in surrounding of
area of range lands is evident between 1986, 1998 and Ahvaz County.
2014. Spatiotemporal changes for each type of land use However, most of the net changes in land uses

Table 4. Changes of each land use; range lands, farm lands, built up lands and bare lands
during 1998-2014.
Land use/ 1998 2014 Changes From 1998 to 2014
Cover Classes Class Class Class Class Area Area
Area Area (%) Area Area Changes (Km) Changes (%)
(Km) (Km) (%)
Range lands 68.69 0.837 7.5 0.091 -61.4 -89.11
Farm lands 561.52 6.48 1113.97 13.52 550.19 97.93
Water bodies 106.57 1.3 138.32 1.68 -31.47 29.58
Built up lands 41.81 0.4 76.72 0.81 33.93 83.15
Bare lands 7433.9 90.6 6892.86 83.89 -554.04 -7.46
Total 8228.33 100 8228.33 100 - -

088 Journal of Research in Ecology (2016) 4(1): 083-093


Faraji Sabokbar et al., 2016
Table 3. Changes of each land use; range lands, farm lands, built up lands and bare lands
during 1986-2014.
Land use/Cover 1986 2014 Changes From 1986 to 2014
Classes Class Ar- Class Class Ar- Class Area Area Area
ea (Km) Area (%) ea (Km) (%) Changes (Km) Changes (%)
Range lands 115.28 1.4 7.5 0.091 -107.8 -93.51
Farm lands 271.95 3.31 1113.97 13.52 839.49 308.69
Water bodies 208 2.53 138.32 1.68 -69.9 -33.6
Built up lands 28.96 0.35 76.72 0.81 37.5 164.4
Bare lands 7588.32 92.39 6892.86 83.89 -699.04 -9. 34
Total 8228.33 100 8228.33 100 - -

are occurred in built up land. In over periods of 1986- changes in farm lands of case study (Figure 4).
1998, 1998-2014, and 1986-2014, amount of positive According to fig 4, the Number of Patch (NP) and Mean
changes for built up land has been calculated about Patch Size (MPS) both increase monotonically. Metrics
44.36%, 83.15%, and 164.4% respectively. In fact, the of NP, MPS and MPE at its lowest value in the 1986,
increasing of urban development, village to town indicating a fragmented farming landscape that is
conversion, developing of industries leads to increase composed of many small patches (Figure 4). For the
built up lands. Also, bare lands and water bodies have 1998, and 2014, NP, MPS and MPE increase gradually
swinging and negative changes for the entire periods with distance from the 1986, indicating integration in
study (1986-1998, 1998-2014, and 1986-2014). land parcels.
Therefore, results of application of remote sensing reveal MSI increases first and then decreases due to
that this plain faced with severe variation in LUCC. integration of farming parcels. Temporally, MSI value in
Hence, the successful identification of change the 1998 and 2014 has decreased, indicating that small
trajectories provides a critical component for land use, agricultural parcels have been transformed into large
conservation and restoration planning in this region. patches of developed lands. In contrast, Fractal
Farming landscape pattern analysis dimension describes the complexity and fragmentation of
The results from landscape metric analysis along a patch by a perimeter area proportion. Low values are
the Ahvaz city are shown in Figure 2. The diagrams acquired when a patch condensed to a rectangular from
show the changes of landscape pattern both spatially small perimeter areas comparatively. For more complex
along the Ahvaz city and temporally through three and fragmented patches, the perimeter increases, yielding
decades. On each of the diagram, the horizontal axis a higher fractal dimension. PSCOV and AWMPFD have
represents years and the vertical axis represents the experienced swinging condition in the 1986, 1998, and
metric value being plotted. A major theme behind the 2014. These metrics decreases first, afterward increases
following interpretations is how the changes of landscape due to drought condition and then decreases due to
pattern are related to the process of socioeconomically combination of farming patches.
and urbanization. The increasing growth of farm lands is evident in
In total, six commonly used landscape metrics all of the output diagrams (Figure 4). Spatially, PSCOV
are examined systematically. The mean patch size, and AWMPFD are positively related to the establishment
number of patches, mean shape index, mean patch edge, of sugarcane agro-industry companies while PSCOV and
patch size coefficient of variance, and area weighted AWMPFD are negatively related to the precipitation
mean patch fractal dimension all show a remarkably changes. Hence, a farmed landscape is characterized by a

Journal of Research in Ecology (2016) 4(1): 083-093 089


Faraji Sabokbar et al., 2016

Figure 4. Changes of class-level MPS, NP, MSI, AWMPFD, MPE, and PSCOV along the Ahvaz
County at three time points in farm lands

large patch size. Temporally, the most significant built up lands and subsequently integrated into farm
changes of landscape pattern in the past three decades lands.
occur at the center of plain, where landscape pattern
shifts from characteristics that represent bare landscapes CONCLUSION
to those represent farming landscapes. Based on these The interactions between human and physical
findings, it can be said that the results are consistent with parameters created land use/ cover change patterns.
research findings such as Karami and Feghhi (2012), Literature reviews of key factors of land use/ cover
Mirzaii et al (2014) and Talebi Amiri (2009). They change assisted to analyze which theory is suitable in a
believe that forest lands changes to grassland and specific region and increased the development of
agricultural landscape frequently. These changes theoretical knowledge. In this article an integrated
happened due to eliminating forest patches inside the method is displayed to analyze the pattern of LUCC that
allows a wide range of factors, from different disciplines,

090 Journal of Research in Ecology (2016) 4(1): 083-093


Faraji Sabokbar et al., 2016
to contribute to the explanation of land-use change. Coppin Pol R and Marvin E Bauer. (1996). "Digital
Hence, this paper applies an integrated method of change detection in forest ecosystems with remote
DT-landscape metrics for satellite images classification sensing imagery." Remote sensing reviews, 13(3-4):207-
and land use changes monitoring. The landscape metrics 234.
is used in order to quantitating of farming changes. From
Deng Jin S, Ke Wang, Yang Hong, and Jia GQi.
the results of research, regardless of the changes
(2009). "Spatio-temporal dynamics and evolution of land
occurred during three periods (1986-1998, 1998-2014
use change and landscape pattern in response to rapid
and 1986-2014) can be found that the land use changes
urbanization." Landscape and urban planning , 92(3-4):
has increasingly occurred into farm lands class.
187-198.
Compared to former classes, farm lands have changed a
tremendous during the periods. With respect to the Herold Martin, Helen Couclelis and Keith CClarke.
landscape metrics Such as NP, MPS, MSI, AWMPFD, (2005). "The role of spatial metrics in the analysis and
PSCOV, and MPE, the agricultural lands have shown modeling of urban land use change." Computers,
simpler and homogenous changes. This means currently Environment and Urban Systems, 29(4): 369-399.
the area of agricultural lands have increasingly growth
Irwin Elena G and Jacqueline Geoghegan. (2005).
during the periods. It could be attributed to a good event
"Theory, data, methods: developing spatially explicit
that has been happening in the agricultural sector.
economic models of land use change." Agriculture,
Accuracy and Kappa ratio from hybrid model of DT-
Ecosystems and Environment, 85(1):7-24.
landscape metrics display the metrics as a powerful tool
in quantitating of landscape, to improve understanding of Jensen John R. (2005). Introductory digital image
the variation in farm lands and different land use types. processing: a remote sensing perspective. University of
South Carolina, Columbus.
REFERENCES
Ji CY, Qinhuo Liu, Danfeng Sun, Sheng Wang, Pei
Antrop Marc and Veerle Van Eetvelde. (2000).
Lin and Xiaowen Li. (2001). "Monitoring urban
"Holistic aspects of suburban landscapes: visual image
expansion with remote sensing in China." International
interpretation and landscape metrics." Landscape and
Journal of Remote Sensing 22(8): 1441-1455.
urban planning, 50(1-3): 43-58.
Karami Arash and Feghhi Jahangir. (2012).
Chen Shui-sen, Liang-fu Chen, Qin-huo Liu, Xia Li,
"Quantitative analysis of landscape metrics for protecting
and Qiyu Tan. (2005). "Remote sensing and GIS-based
sustainable land use pattern (Case study: Kohgiluyeh and
integrated analysis of coastal changes and their
Boyerahmad Province)" Journal of Environmental
environmental impacts in Lingding Bay, Pearl River
Studies, 37(60): 79-88 .
Estuary, South China."Ocean and coastal
management, 48(1): 65-83. Lambin Eric F and Helmut JGeist. (2006). Land-use
and land-cover change: local processes and Global
Collins John B and Curtis E, Woodcock. (1996). "An
Impacts. Springer Science and Business Media,1-2.
assessment of several linear change detection techniques
for mapping forest mortality using multi temporal Li Xia, and Anthony Gar-On Yeh. (2004)."Analyzing
Landsat TM data." Remote sensing of environment, 56 spatial restructuring of land use patterns in a fast growing
(1) : 66-77. region using remote sensing and GIS. "Landscape and

Journal of Research in Ecology (2016) 4(1): 083-093 091


Faraji Sabokbar et al., 2016
Urban planning 69(4): 335-354. sensing digital image analysis. 3 Berlin et al: Springer.

Liu Kai, Xia Li, Xun Shi and Shugong Wang. (2008). Röder Achim, Tobias Kuemmerle and Joachim Hill.
"Monitoring mangrove forest changes using remote (2005). "Extension of retrospective datasets using
sensing and GIS data with decision-tree learning." multiple sensors. An approach to radiometric
Wetlands, 28(2): 336-346. intercalibration of Landsat TM and MSS data." Remote
Sensing of Environment, 95(2): 195-210.
Liu Yong, Xiaojian Lv, Xiaosheng Qin, Huaicheng
Guo, Yajuan Yu, Jinfeng Wang and Guozhu Mao. Song Conghe, Curtis E Woodcock, Karen C Seto,
(2007). "An integrated GIS-based analysis system for Mary Pax Lenney and Scott A Macomber.
land-use management of lake areas in urban (2001)."Classification and Change Detection Using
fringe." Landscape and urban planning, 82(4): 233-246. Landsat TM dData: When and How to correct
atmospheric effects?." Remote sensing of
Lu Dengsheng, P Mausel, E Brondizio and Emilio
Environment, 75(2): 230-244.
Moran. (2004). "Change detection
techniques." International journal of remote sensing, 25 Sui Daniel Z and Hui Zeng. (2001). "Modeling the
(12): 2365-2401. dynamics of landscape structure in Asia’s emerging
desakota regions: a case study in Shenzhen." Landscape
Mc Garigal, Kevin, Sam A Cushman, Maile C Neel,
and urban planning, 53(1-4): 37-52.
and Eduard Ene. (1995)."FRAGSTATS: spatial pattern
analysis program for categorical maps". Talebi Amiri Shima, Azari Dehkordi Forood, Sadeghi
Seyed Hmidreza and Soofbaf Seyed Reza. (2009).
Mirzaii Mohsen, Riahi Bakhtiyari, Alireza Mahini,
"Study on Landscape Degradation in Neka Watershed
Abolrasul, Gholamali Fard and Mehdi. (2014). " Land
Using Landscape Metrics" Environmental Sciences, 6(3):
cover change detection using landscape metrics in
133-144.
Mazandaran province." Applied Ecology, 4 :37-54.
Turner Monica G, Robert H Gardner and Robert V
Olthof Ian, Darren Pouliot, Richard Fernandes and
O'neill. (2001). Landscape ecology in theory and
Rasim Latifovic. (2005). "Landsat-7 ETM+ radiometric
practice. New York: Springer,401
normalization comparison for northern mapping
applications." Remote Sensing of Environment, 95(3): Weng Qihao. (2001). "A remote sensing? GIS
388-398. evaluation of urban expansion and its impact on surface
temperature in the Zhujiang Delta, China." International
Radeloff Volker C, Erik Nelson, Andrew J Plantinga,
journal of remote sensing 22(10): 1999-2014.
David J Lewis, Davis Helmers, JJ Lawler, JC Withey,
F Beaudry, S Martinuzzi, V Butsic, E lonsdorf, D Xiao Jieying, Yanjun Shen, Jingfeng Ge, Ryutaro
White and S Polasky (2012). "Economic-based Tateishi, Changyuan Tang, Yanqing Liang and
projections of future land use in the conterminous United Zhiying Huang. (2006). "Evaluating urban expansion
States under alternative policy scenarios." Ecological and land use change in Shijiazhuang, China, by using
Applications, 22(3): 1036-1049. GIS and remote sensing." Landscape and urban
planning, 75(1-2): 69-80.
Richards John A and JA Richards. (1999). Remote
Yeh Anthony Gar-On and Xia Li. (1999). "Economic

092 Journal of Research in Ecology (2016) 4(1): 083-093


Faraji Sabokbar et al., 2016
Development and Agricultural Land lLoss in the Pearl
River Delta, China." Habitat International, 23(3): 373-
390.

Yuan Ding and Chris Elvidge. (1998). "NALC Land


Cover Change Detection Pilot Study: Washington DC
area experiments." Remote Sensing of Environment, 66
(2): 166-178.

Submit your articles online at ecologyresearch.info


Advantages
 Easy online submission
 Complete Peer review
 Affordable Charges
 Quick processing
 Extensive indexing
 You retain your copyright

submit@ecologyresearch.info
www.ecologyresearch.info/Submit.php.

093 Journal of Research in Ecology (2016) 4(1): 083-093

Você também pode gostar