B I O D I V E R S I T A S
ISSN: 1412-033X
Volume 15, Number 1, April 2014 E-ISSN: 2085-4722
Pages: 53-59 10.13057/biodiv/d150108
Structure and spatial pattern of land uses patches in the Zagros
Mountains region in the west of Iran
ARASH KARAMI
1
, KIOMARS SEFIDI
2,♥
, JAHANGIR FEGHHI
1
1
Department of Forestry, Faculty of Natural Resources, University of Tehran, Karaj, Iran
2
Faculty of Agriculture Technology and Natural Resources, University of Mohaghegh Ardabili, Ardabil, Iran. Tel./Fax.: +98-451-5512081;
♥
email: kiomarssefidi@gmail.com
Manuscript received: 28 August 2013. Revision accepted: 18 November 2013.
ABSTRACT
Karami A, Sefidi K, Feghhi J. 2014. Structure and spatial pattern of land uses patches in the Zagros Mountains region in the west of
Iran. Biodiversitas 15: 53-59. Landscape ecology as a new interdisciplinary science, concepts, theories and methods provides for
evaluation and management of land. Quantification of landscape patterns has key role in the interpretation and modeling of spatial and
chronically variation of land uses. This study carried out in the Zagros vegetative region in the west of Iran to quantify structure and
spatial pattern of land uses and forest fragmentation in the Zagros Mountains region. The mosaic analysis method was used for
quantifying landscape metrics. Totally 2783 land use patches were recorded in the study area. The most of the patches were agricultural
area and the lowest number of patches recorded for rivers. Diversity indices analysis showed agricultural land use has highest diversity
in comparing with other land uses. Rangeland use has distributed in the central region of the study area. Despite the high density
agriculture and rangeland a lot of potentials there are two conversions this land as a sustainable ecosystem (forests, agriculture
integrated, and rangelands), that can be by applying as an appropriate method in management and control policies and converting
artificial land uses to the natural or semi natural land uses according to the advantages of such land uses in view of sustainability. We
conclude that fragmentation of natural land uses such forest and rangelands should be reducing and maintain large patches of natural
vegetation to sustainable land management in this region.
Key words: Metrics, FRAGSTATS, Iran, land use, landscape ecology
INTRODUCTION
Human development since its presence on this planet
was life on natural resources, and intervention in natural
ecosystems and the domestication of plants and animals on
the planet. Human economical activities are mainly in
landscape scale, so the landscape is appropriate spatial
scales for the study of environmental change is a result of
human activities. Eventually all human activities lead to
land use location and structure changes, so landscapes
provide a reflection of past human land use and as alive
and dynamic framework for sustainable land use are
imposed (Xiao and Zhong 1998). There are many
similarities between the structure of landscape, land use
structure, function of landscape, land use changes and
changes in landscape, but the primary goal of ecological
studies in all the land is landscape (Naveh and Liberman
1984). Development landscape ecology provides a suitable
theoretical basis to study land use changes. When the
connection between landscape ecology and sustainable
development has been created, principles ecological lands-
cape as the main concepts related to sustainable land use is
proposed. Obviously, in areas where land management
(land use) with the potential ecological imbalance, must
provide reasonable and practical solutions to resolve this
inconsistency, and the severity of the imbalance is greater,
in terms of management reform, has high priority. The first
step in planning is knowledge. Knowing the current
situation, awareness of process and forecasting the future is
base of planning in the landscape management. To
determine the status of land management and land use in
different areas and compare them together, to set
redundancy between optimal and the current situation, can
move towards sustainable development and appropriate
land. Landscape ecology is strongly associated with land
use, particularly which this relationship of aspects the
spatial patterns as is made of land use planning and
management it the foundation (Wu 2000). According to the
landscape ecology principles analysis of spatial and
structural characteristics of land use patches has key role in
the interpretation and modeling of spatial and chronically
variation of land uses (Wegener 1994; Herzog and Lausch
2001). Besides studying the structure according to
principles of ecology, quantitative metrics of landscape
appropriate suitable tools for quantifying the spatial
characteristics of the components of the landscape.
Concepts of landscape ecology can be expanded in
planning for land use management in order to reduce the
negative effects of human manipulations in land using
(Lausch and Herzog 2002).
As mentioned, landscapes are strongly changing. In
monitoring the areas and landscape changes in the position
and applications of various lands should be considered.
This information can be very useful for users, designers
B I O D I V E R S I T A S
15 (1): 53-59, April 2014
54
and managers in land use managing (Zhang and Ni 2001).
Landscapes metrics can consider as algorithms for
quantifying the spatial characteristics of patches, classes or
landscape whole patches metrics, Metrics with the
ecological principles of landscape and interpret it are as the
best way to compare the situation of landscape in different
lands. These metrics also can be considered as comparative
tools to analysis different landscape scenarios or obtain
knowledge about landscape land situation changes over
time. Meanwhile, such metrics can be used for the design
and find an accurate relationship between the structure and
function in landscapes (Botequilha-Leitão et al. 2006).
According to the given concepts, in this paper has tried to
work on landscape ecology, In order to quantify and
interpret the location and structure for sustainable land and
reduce them to be considered unstable, until damage and
easy loss (fragmentation) of these complex patterns to be
normal.
The most regions of Iran because of the ecological
richness and genetic resources have high ecological value,
but in Zagros region because of the of high biodiversity,
genetic conditions, cultural, and social climate of the region
and particularly on the specific structure of tribal life,
traditional economy based on agriculture and animal
husbandry and as a result dependence of livelihoods on
natural resources management, planning for sustainable
and appropriate utilization of these resources are necessary.
Therefore it was necessary for greater recognition and
status for different regions, the research done in this area.
Thus, the goal of this research was to analysis the
position of different land use from each other and try to
find how located them in the context of Landscape.
Quantify the structure and characteristics of ecosystem as
an ecological analysis and understanding the different
ecosystems function(among different land use) are basis for
comparison how effects of human and environmental
changes in overtime on land use(Bell 1999).Also in this
research composition and spatial distribution of structural
elements in the landscape had studied, that their main role
is in improvement ecological functions of the study area,
and what corrective actions can cause improvements in
ecological structure and in result processes associated with
them will be discussed. In other word the main goal of this
research we are interested to quantify the information on
fragmentation of natural land uses in the Zagros Mountains
region in the west of Iran. The specific objectives were
calculating spatial pattern of land use patches including:
number of patches (NP), land area fraction by patches
(PLAND), patch density (PD) and Shannon's diversity
index (SHDI) in class and landscape levels
MATERIALS AND METHODS
Study area
This study was conducted in the Zagros Mountains
region in the west of Iran. Kurdistan province is located in
the northern Zagros Mountains in the west of Iran (latitude
range in: 34◦-, 44◦ 36, 30 N; longitude range: 31◦-45◦, 48,
16 E). Territory of Zagros Mountains region is now one of
the oldest sources of biodiversity and the one of the first
settlement place of the Iranian people (Razmara 2005).
According to meteorological statistics, Zagros Mountains
region has a semi-humid climate with very cold winters
and the Alpine and average annual rainfall is 512mm, that
volume equivalent to14 billion cubic meters annually it is
estimated (Anon. 2000). Based on vegetation maps
produced about 60% the total area of the Zagros Mountains
region have is vegetation cover forest and pasture. In this
province due to geographical environment diversity, and
uneven condition of weather conditions, land use different
types can be observed, but land use with cover forest and
pasture associated with species of Quercus persica,
Quercus libani, Quercus infectoria, Crataegus aronia,
Pistacia mutica, Amygdalus communis, Cotoneaster spp.
and also grass species such Salvia eremophila, Ferula
ovina, Achillea wilhelmsii, Vicia sativa and etc. with the
highest level and use the region and the species that of
plant specific species are considered to these areas
(Marvie-Mohadjer 2005). The region soil of type soil rocks
with a schist bed rock of type brown calcareous and in the
mountains and edges are Rendzine. Agriculture land use
focuses more on the west the province (Figure 1). In the
last decades construction and developing of new land uses
by human in towns and villages across the province led to
the have been a creation specific type of land use called the
forestry-agriculture-livestock, that need to multipurpose
managing lands in ‘agroforestry’ systems also use more
participate of local people in this area (Shamekhi 2007).
Procedures
This research had carried out in three scales including
patches, classes (zones) and Landscape (McGarigal and
Marks 1994). The bases of our study were support and
maintain more valuable patches and increase convergence
of sustainable patches.
In this study we used maps of land use created by the
Iran Organization of Forests, Range and Watershed
Management (IOFPWM) were prepared using the accurate
geostatistics for the total Iran in 2010, It should be
mentioned that this map is based on purposes the
researcher reclassified and six categories including: the
forest defined as land with tree crown cover or equivalent
crown canopy level of more than 5 percent and area of
more than 0.5 ha, pastures including rangelands and
pastures, agricultural lands, no coverage and bare land (e.g.
Arid and salty soil land), residential areas and Rivers (e.g.
Rivers and canebrake).
The approach used in this research was based on
landscape ecology, in this approach, the relationship
between structure and function of process-scale approach is
landscape had considered (Botequilha-Leitao and Ahren
2002). Using this approach and its method and techniques,
are known as series disruption of ecosystems related to
each other, one effective method is considered for
ecological analyzing the area (Ingegnoli 2002). Afterward
spatial structure and analysis of different metrics of patches
was quantified using FRAGSTATS 3.3software
(McGarigal and Marks 1994). FRAGSTATS is a Spatial
Pattern Analysis Program and complete set of landscape
metrics.
KARAMI et al. - Land uses patches in the Zagros Mountains, Iran
55
Figure 1A. Location of the study area in the west of Iran. B. Map of land use Kurdistan province (Forests, Range and Watershed
management Organization, I.R. Iran (2010) and reclassified by Karami (2011).
This program does not create restrictions on the scale and is
suitable software for spatial patterns and various metric
landscape patches analyzing that make up the territory in a
heterogeneous environment and is suitable in different
conditions. The FRAGSTATS ARC software is used to
demonstrate the linkage of the U.S. Forest Service’s
FRAGSTATS program to the Arc Info GIS (ESRI 1992)
which is updated version of the USDA Forest Service -
General Technical Report PNW-GTR-351. In this research
metrics have been calculated according to the purposes in
meter or hectare. For each input to FRAGSTATS software,
3 outputs file is created. The file named Patch, Class and
the Land as a text file and all files are visible. These tools
to analyze spatial patterns, especially in modeling habitat;
wildlife protection and forest management are applied.
Ability to describe a quality of landscape, a prerequisite is
for studying the function and change the landscape, and
different metric to achieve this goal; the ecology has been
extracted from landscape, for example the application
model provided by a Foreman landscape focuses on four
models for the sustainable planning (McGarigal and Mark
1995) including (i) Maintain large patches of natural
vegetation, (ii) maintenance of wide river corridors, (iii)
maintains the continuity of key species among large
patches and (iv) maintaining the heterogeneous parts of
natural human development in the region. In this study
before to data entering operation in FRAGSTATS
software, we needed to prepare the data in the IDRISI
software, because of raster format of data on Zagros
Mountains land use maps. In relation to some of the
landscape metrics that were used in this research, we had
defined metrics as shown in Table 1. Some metrics are the
standard metrics of FRAGSTATS but some of them such
as standard deviation of neighbor distance used in analysis
fragmentation of land use patches.
Data analysis
We analyzed data in two different levels: (i) metrics
analysis in class level to calculate land use Changes in
similar land use patches and (ii) metrics analysis at the
landscape level to calculate land use total patches in the
entire study area.
Class analysis
At this level there are 6 classes included Forest,
Rangelands, Agricultural area, bare land, Residential area
and Rivers which metrics were calculated for each class.
The metrics used in this level included: The mean shape
index (MSI), number of patches in class level (NP) which
used to calculate the number of patches in each class.
Largest patch indexes (LPI), largest patches in each class
were calculated. Patch density (PD), Density of patches per
class per unit area was calculated. Edge density (ED) and
Total edge (TE).
Also the length strips around each land use and
complexity of it calculated. Class area (CA), the area
covered by each class had calculated. Characteristic of the
Euclidean distance between patches include (ENN_MN,
B
A
B I O D I V E R S I T A S
15 (1): 53-59, April 2014
56
ENN_AM, ENN_MD,
ENN_RA, ENN_SD and
ENN_CV). Metric related
Statistics including mean,
median, range, standard
deviation and coefficient of
variation had calculated
between patches. We should
be mentioned McGarigal and
Marks (1994) note that the
value of the Fractal dimension
as computed in FRAGSTATS
is dependent upon the patch
size and the units used
(Rogers 1993).
Landscape analysis
Metrics analysis at
landscape level was done with
calculating metrics of total
land use patches. For analysis
Metric of at the landscape
level, in addition, using
metrics that mentioned in
class level for all patches;
Simpson and Shannon
diversity index (SHDI, SIDI,
MSIDI, SHEI, SIEI and
MSIEI)of the metrics were
calculated in order to diversity
and uniformity analysis of
land uses types. The Shannon
index of diversity (SHDI, see
Shannon and Weaver 1949) is
the most widely used index
for diversity evaluation. This
index, ranging in theory from
0 to infinity, estimates the
average uncertainty in
predicting which land cover
type a randomly selected sub-
unit of the landscape will
belong to. Also widely used,
the Simpson index (SIDI, see
Simpson 1949), Producing
values from 0 to 1, Simpson’s
index defines the probability
that two equalized subs-units
of the landscape, selected at
random, belong to different
cover types.
Both these indices of
diversity combine evaluations
of richness and evenness. The
landscape metrics are shown
in Table 1.
They increase under
situations where the number of land cover types (landscape
richness) increases, or the equitability of distribution of
land amongst the various cover types (landscape evenness)
increases, or both (Nagendra 2002).
Table 1. Description of metrics used in landscape and class level analysis.
Character Formula and
abbreviations
Description
Unit
Range
ofchanges
Landscape and class level analysis
nNP
Number of patches
-
NP>0
100
1
A
a
j
ij
a
i
PPLAND
Area fraction with patches
%
0PLAND100
10010000
A
n
PD
i
Number of patches per 100
ha
#
PD>0
10000
1
A
m
k
ik
ED
e
Edge density, wherei is the
land use classes andk is for
edge length
Meter per
hectare
ED>0
Landscape level analysis
)ln(
i
m
i
i
PPSHDI
1
Shannon's diversity index
-
SHDI≥1
2
1
1 Pi
s
i
SIDI
Simpson's diversity index
-
0 ≤SIDI≤ 1
100
max
A
a
ij
LPI
Largest patch index
Percentage
(%)
0LPI100
i
i
e
e
LSI
min
Landscape shape index
-
LSI1
m
h
MNN
m
i
i
1
Mean nearest-neighbor
distance
(mu)
MNN>0
m
MNNh
NNSD
m
i
i
1
2
Nearest-neighbor standard
deviation
(mu)
NNSD>0
m
K
iK
eTE
1
Total edge
(mu)
TE ≥ 0
m
m
i
i
P
i
P
SHEI
ln
1
ln
Shannon's evenness index
-
0 ≤ SHEI ≤ 1
m
m
i
SIEI
i
p
1
1
1
1
2
Simpson's evenness index
-
0 ≤ SIEI ≤ 1
000,10
1
1
n
j
ij
aCA
Class area
(ha)
CA > 0
KARAMI et al. - Land uses patches in the Zagros Mountains, Iran
57
RESULTS AND DISCUSSION
According to the analysis of land
use patches in the Zagros Mountains
region (Figure 1B), agricultural
patches observed more in the east of
Kurdistan province, forest patches in
the west and rangelands in the central
part as north-south strips are
distributed. The map of land use
distribution illustrated in Figure 1B,
which variety of land uses and
patches are detectable.
Metric analysis of class level
Metric analysis of class level with
calculating percent of land use
metrics, patch density, average of
patches size, the largest size of
patches, metric of landscape shape
and the average landscape patches
had done. As shown in Table 2, the
density of agriculture patches (PD) in
this region is higher than other land
uses, and the forest, rangeland users,
and the land without cover (bare
land) were the next of orders,
respectively. According to Table 2,
and the number and distribution of
patches of urban and residential areas
in the landscape level, the close
relationship and intimately between
nature and humans there are in level
and should be considered in planning
sustainable land management.
According to Table 2, the largest
land use area and the greatest number
of patches, respectively, recorded in
agricultural land, Rangelands and
forest. Bare lands including no
covered area have minimum area and
river lands showed lowest number
have in this level compared to the
other land uses. The number of
residential areas patches (urban
patches) in this level had shown in
Table 2, according to this metric and
area of this province, high relation
between human community’s
presence and change in nature cane
revealed.
Analysis of the total land area
fraction (Table 2, PLAND) and total
edge (Table 2, TE) metrics among land uses in the Zagros
Mountains region showed that according to the percentage
cover metrics, agricultural lands, rangeland and forested
area in this region had covered the largest area,
respectively.
As we illustrated in Figure 3, statistical characteristics
of areas related metric showed the main area of land use
patches in agricultural, pasture and forest land uses through
the province was higher in comparison with other land
uses. The range of mean area variation in rangelands and
forest land use patches was more than the mean of other
land uses. This fact means the patches of rangelands and
forestry had composed from the changing area and human
based disturbances and interference in this land use was
considerably over the time.
Table 2. Patches structure related metrics in the class level
Zagros Mountains region,map 1 : 250,000 scale
metrics in the level of class
Land use types
(Map unit)
CA (ha)
NP
(n)
PD
(m)
LPI
(%)
ED
(m)
TE (m)
LSI
PLAND(%)
Forest
373432.12
223
0.01
6.24
3.96
7309130
14.91
12.68
Pastures
1243724
919
0.034
24.36
10.41
24260814
21.68
41.28
Agricultural land
1268342
1560
0.038
32.26
10.35
31503691
20.44
45.87
Bareland
4789.58
38
0.0003
0.01
0.01
129204.7
1.67
0.01
Residential
6169.49
30
0.0007
0.06
0.04
204448.5
2.05
0.1
Rivers
8382
13
0.0007
0.04
0.04
255599.6
2.28
0.06
Table 3. Mean (± S.E.), weighted mean, Median and Range of Euclidean distance the
nearest neighbor metric in class level in the Zagros Mountains region in the west of Iran
Land uses
Mean(km)
Weighted
mean (km)
Median
Range
Forest
1.83 ±2.8
0.45
0.50
14.65
Rangelands
0.51 ±0.62
0.12
0.26
3.95
Agricultural area
0.43 ±086
0.10
0.20
7.78
Other land uses
a
120.49
120.49
120.49
0
Total, allland uses
123.26 ±10.4
121.16
Note:
a
Other land uses including residential areas, bare lands and rivers.
Table 4. Statistical characteristic of metrics of land uses patches in landscape level in the
Zagros Mountains region Map with 1:1000000 scale
Symbol
FRAGSTATS metrics
FRAGSTATS
statistics
TE
Total edge(mu)
63762887.13
PR
Patch Richness (No unit)
6
PRD
Patch Richness Density (Meter In 100hectares)
0.002
RPR
Relative Patch Richness (%)
100
CA
Class area (ha)
2904839
PN
Patch number (No unit)
2783
PD
Patch density (Meter In 100hectares)
0.08
LPI
Largest patch index (%)
32.26
ED
Edge density (Meterper hectare)
13
LSI
Landscape shape index (No unit)
17.38
AREA_MN
Mean patch area (mu)
67.35
AREA_AM
Mean weighted patch area (mu)
50776.53
AREA_MD
Median patch area (mu)
1117.72
AREA_RA
Range in patch area (mu)
92317.42
AREA_SD
Standard deviation in patch area (mu)
7450.16
AREA_CV
Coefficient of variation in patch area (mu)
666.54
ENN_MN
Mean patch distance (mu)
1549.36
ENN_AM
Mean distance -weighted patch (mu)
221.14
ENN_MD
Median patch distance (mu)
249.93
ENN_RA
Range in patch distance (mu)
99913.08
ENN_SD
Standard deviation in patch distance (mu)
8910.04
ENN_CV
Coefficient of variation in patch distance (mu)
575.07
