A New Approach to Diversity Indices u2013 Modeling and Mapping Plant Biodiversity of Nallihan.pdf

8/9/2019 A New Approach to Diversity Indices u2013 Modeling and Mapping Plant Biodiversity of Nallihan.pdf

1/24

-1

A new approach to diversity indices – modeling andmapping plant biodiversity of Nallihan(A3-Ankara/Turkey) forest ecosystem in frame of geographic information systems

HAKAN METE DOGAN1,* and MUSA DOGAN21GIS and RS Department, Central Research Institute for Field Crops (CRIFC), Gulderen Sok.

Ziraat Loj. No: 3, Yenimahalle, 06070 Ankara, Turkey; 2Biology Department, Middle East Technical

University, Ankara, Turkey; *Author for correspondence (e-mail: [email protected]; phone:

+90-312-327-01-50; fax; +90-312-315-14-66)

Received 10 February 2004; accepted in revised form 4 August 2004

Key words: Flora, Geographic information systems, Mapping, Modeling, Plant biodiversity, Plant

ecology, Remote sensing, Spatial analysis

Abstract. Modeling and mapping possibilities of Shannon–Wiener, Simpson, and number of

species (NS) indices were researched using geographic information systems (GIS) and remote

sensing (RS) tools in Nallihan forest ecosystem of Turkey. The relationships between the indices

and a number of independent variables such as topography, geology, soil, climate, normalized

difference vegetation index (NDVI), and land cover were investigated to understand relationships

between plant diversity and ecosystem. Georeferenced field data from the established 56 quadrats

(50 · 20 m) were used to calculate the indices. Principle component analysis (PCA) and multipleregression were employed for data reduction and model development, respectively. Three diversity

maps were produced using the developed models. Residual maps and logical interpretations in

ecological point of view were used to test the validity of the models. Elevation and climatic factors

formed the most important components that are effective determinants of plant species diversity,

but geological formations, soil, land cover and land-use characteristics also influenced plant

diversity. Considering the different responses of the models, Shannon–Wiener (SWI) and NS

models were found suitable for rare cover types, while Simpson (SIMP) model might be appro-

priate for single dominant land covers in the study area.

Introduction

Conservation Biology is an emerging discipline dedicated to the preservation of

endangered species and habitats. To develop effective protection strategies,experts need to understand the relationship between species and ecosystem.

Most importantly, they need to decide which areas are the most important to

protect. Consequently, mapping the areas with high plant biodiversity has a

priority for decision-makers. Effective management plans and actions can only

be achieved with this valuable spatial information. Ecologists define species

diversity on the basis of two factors: species richness and species evenness. The

number of species (NS) in the community is called species richness, while the

relative abundance of species is described as species evenness (Molles 1999).

Biodiversity and Conservation (2006) 15:855–878 Springer 2006

DOI 10.1007/s10531-004-2937-4


2/24

How environmental structure affect species diversity is one of the most fun-

damental subject of investigation about communities (Barbour et al. 1987;

Molles 1999). Although having been much criticized for the imperfect defini-

tion of the concept of diversity and sampling difficulties, diversity indices are

still widely used to evaluate, survey, and conserve ecosystems (Pielou 1966;

Barbour et al. 1987; Riitters et al. 1995; Mouillot and Leprêtre 1999). The most

popular indices that have been used to quantify landscape composition are

Shannon’s index, believed to emphasize the richness component of diversity,

and Simpson’s index, emphasizing the evenness component (Magurran 1988;

Nagendra 2002). Choosing appropriate methods and tools, it is believed that

these indices have a potential to map diversity. At this point, geographic

information systems (GIS) that has been recently recognized by conservationbiologists can be an appropriate and powerful tool in the spatial analyzes

performed in conservation biology (Kadmon 1997; Dogan 1998; Kress et al.

1998; Lenton et al. 2000; Dogan 2001). The spatial nature of the biological data

lets GIS to develop spatial models of which they might also be used as a

solution for predictive mapping (Franklin 1998; Gottfried et al. 1998). Where

as the monitoring results and mapping of earlier periods are considered as vital

information for such kind of GIS databases. This need is fulfilled generally by

aero-space remotely sensed data (Fjeldsa et al. 1997) in which at some regions

of the globe the data set can go back to early 1950s via aerial photographs to

recent via high resolution multispectral global coverages for the diversity

studies. Within this frame, the aim of this study is to create a new approach to

the conventional diversity (Shannon-Wiener, Simpson, NS) indices using GISand remote sensing (RS) tools. Consequently, in order to reach this goal plant

biodiversity of Nallihan forest ecosystem was modeled and mapped within the

frame of this new approach between the years 2001 and 2002.

Materials and methods

Study area

This study was conducted in Nallihan administrative district of Ankara

province in Turkey. According to the grid system based on two degrees of

latitude and longitude (Davis 1965–1988); the study area is located in the A3

grid square of Central Anatolia (Figure 1a). This location is within the Irano-Turanian phytogeographical region with some Mediterranean penetrations

(Davis 1971), and the records of Turkey’s Plant Database (TUBIVES 2003)

pointed out 119 family, 553 genera, and 1350 plant species in this square.

Recently, a new Acantholimon (Plumbaginaceae) species was published from

the area (Dogan and Akaydin 2002). The study area is specifically called

Erenler forest region, and covers 327.31 km2 (32731.29 ha) area. The general

topography of the study area is mountainous (Figure 1b). Generally, agricul-

tural lands are concentrated along the river basins, while forests dominate the

856


3/24

higher elevations. There are 28 settlements in the study area, and majority of

them are small villages. Major human effects on the forest can be seen in the

area like agriculture and urbanization. Moreover, a considerable part of this

area faces erosion. About 5.6% forest was degraded by natural or anthropo-

genic causes in the area.

According to Emberger classification system, the climate of the study area

showed ‘Semi-arid Upper Mediterranean Bioclimate’ characteristics with cold

winters (Akman and Daget 1971; Akman 1999). Basically, four climatic sea-

sons are recognized in the study area. Precipitation is mostly in the form of rain

Figure 1. Physiographic setting (a) and physical geography (b) of the study area (projection

systems of physiographic setting and physical geography maps were defined as geographic with

European datum (spheroid international 1909) and UTM with European datum (zone: 36, spheroid

international 1909), respectively).

857


4/24

throughout the year except winters, and total number of snowy days does not

exceed 20 days. The main tree species of the study area are black pine (Pinus

nigra), juniper (Juniperus spp.), red pine (Pinus brutia), and oak (Quercus spp.).

According to the digital forest stand map, the study area can be generalized in

six categories as non forest (28.39%), oak forest (0.47%), erosion/stony

(5.60%), degraded forest (33.85%), Black Pine forest (31.47), and Red Pine

forest (0.21%).

Methods

A flowchart of the methodology is given in Figure 2. Digital geological and soilmaps of the study area were obtained from the General Directorate of Mineral

Research and Exploration (MTA) and the General Directorate of Rural Af-

fairs (KHGM), respectively. Topographical and forest stand maps were digi-

tized in UNIX Arc/Info 7.0.4 and PC Arc/Info 3.5 software (ESRI 1994; ESRI

1997). The LANDSAT-TM image, acquired on August 21st 2000, was utilized

to develop land cover and normalized difference vegetative index (NDVI) maps

in Erdas Imagine 8.5 software (ERDAS 1997). Supervised classification

method (maximum likelihood parametric rule), 4-5-3 band combination, and

statistical filtering (7 · 7) were used to develop a land cover map. The unsigned

8-bit NDVI model was utilized to establish NDVI classes. Arc/View 3.2 soft-

ware (ESRI 1996) and Inverse Distance Weighted (IDW) method were em-

ployed to produce the interpolated surfaces (grid maps) of climatic(temperature, precipitation, and potential evapotranspiration (PET)), and

additional soil (K2O, P2O5, organic matter, pH, salt, CaCO3, saturation and

texture) variables. To conduct spatial analysis, all developed maps were con-

verted to grid themes by using 30 · 30 m grid size in Arc/View 3.2. Universal

Transverse Mercator (UTM) projection system (spheroid international-1909,

datum: European-1950, zone: 36) was applied to all map data.

Georeferenced point data (791 points) were collected to classify LANDSAT-

TM image and to conduct accuracy assessment (Figure 3a). Land-use (FAO

1990) and formation classes (UNESCO 1973) were utilized to identify the main

land-use and vegetation types in the field. Detailed plant data for diversity

indices were collected from the established quadrats (Figure 3b). The number

of quadrats was determined as 56 considering the quadrat surveys of Magurran

(1981), and the quadrat sites were established according to stratified randomsampling design (McGrew and Monroe 1993). The size of each quadrat was

20 · 50 m following Grossman et al. (2003). Plant parameters collected from

each quadrat were (1) species component, (2) NS, (3) species cover (%), and (4)

species density (number of plant/m2). From the quadrats, soil samples were

also taken according to the certain soil sampling methods (Atesalp 1976; Ulgen

and Yurtsever 1995). On the map of study area, total 570 points were deter-

mined to aggregate climatic data (Figure 3c). The LOCCLIM software

(Grieser 2002) was employed with the digital elevation model (DEM) to

858


5/24

calculate the best estimates of focused climatic variables in each determined

point. Climatic variables were investigated in both ‘annual’ and ‘seasonal’

basis. The time period between May and September was taken as seasonal

because of low precipitation and high PET values within this period.

Total 752 plant specimens, pressed and dried following the rules and defi-

nitions explained by Davis and Heywood (1965), were identified in the

ANKARA Herbarium of Ankara University. The Davis’ Flora of Turkey and

the East Aegean Island Vol. 1–10 (Davis 1965–1988) were used as the mainreference throughout the herbarium studies. Species diversity indices were

calculated for each quadrat at the end of this work. The formulas

H ¢ = P

( pi loge pi ) and D =P

( pi 2) were employed to calculate Shannon–

Wiener and Simpson indices, respectively (Barbour et al. 1987; Molles 1999). In

both formulas, pi values indicate the proportional abundance of the i th species

in a quadrat. On the other hand, NS index has no formula, and it was deter-

mined by using the total species number in a quadrat.

Figure 2. The flowchart of the methodology (the rounded rectangles indicate the analyses and

processes, rectangles show output products).

859


6/24

Spatial analyses and model development were conducted in four steps

(Figure 2). Kaiser–Meyer–Olkin (KMO)–Bartlett tests were conducted to test

the suitability of the data for factor analysis. Then, principle component

Figure 3. Georeferenced point data for supervised classification (a), established quadrats (b), and

established point data to derive climatic variables (c).

860


7/24

analysis (PCA) with varimax rotation was applied for data reduction (SPSS

2001). Multiple regression, regressing a variable on a series of independent

variables (Sokal and Rohlf 1995), was chosen to formulate the relationships.

This was achieved by applying the linear regression with ‘enter’ method in

SPSS-11 software (SPSS 2001). Applying the models, species richness maps

were produced in Arc/View 3.2. The reliability of the maps was tested by

residual maps and ecological interpretations. Residuals were calculated by

using the observed and computed values of indices in each quadrat, and IDW

method was employed to map them. To evaluate different indices in the same

base, interpolating surfaces of the residuals were developed by using standard

deviation values of each index.

Results

Plant species

Total 239 species belonging to 45 families were determined in the study area.

According to Davis (1965–1988) and the records of Turkey’s Plant Database

(TUBIVES 2003); 14 species were detected as endemic for the study area. NS

recognized in each family is stated in Table 1. Leguminosae, Compositae,

Labiatae, Rosaceae, Cruciferae, and Gramineae families have more species

comparing to the others. The full list of identified species was given in

Appendix 1

Table 1. NS recognized in each family.

Family No. of species Family No. of species Family No. of species

Leguminosae 37 Ranunculaceae 3 Iridaceae 1

Compositae 34 Cistaceae 3 Acanthaceae 1

Labiatae 28 Papaveraceae 3 Anacardiaceae 1

Rosaceae 15 Fagaceae 3 Chenopodiaceae 1

Cruciferae 10 Santalaceae 2 Convolvulaceae 1

Graminae 10 Illecebraceae 2 Coryllaceae 1

Liliaceae 9 Rhamnaceae 2 Crassulaceae 1

Boraginaceae 8 Geraniaceae 2 Equisetaceae 1Scrophulariaceae 8 Linaceae 2 Euphorbiaceae 1

Caryophyllaceae 7 Berberidaceae 2 Globulariaceae 1

Umbelliferae 7 Cyperaceae 2 Guttiferae 1

Campanulaceae 5 Paeoniaceae 2 Malvaceae 1

Rubiaceae 5 Pinaceae 2 Orchidaceae 1

Cupressaceae 4 Valerianaceae 2 Polygalaceae 1

Plumbaginaceae 4 Dipsacaceae 1 Urticaceae 1

Number of determined plant species in this study were given with their families in this table. In this way, overall

results about the recognized species were summarized efficiently. Details about the determined species were also

given in Appendix 1.

861


8/24

Remote sensing data

Supervised classification obtained 92.16% overall accuracy with a Kappa

coefficient of 0.8828, and produced a reliable result. Moreover, NDVI map

delineated the areas where the plants dominated. Therefore, these two map

layers supplied valuable spatial information that can be effective on plant

species diversity. The grid maps of land cover and NDVI were given in Fig-

ure 4 with the original LANDSAT-TM (band 3) image.

Data reduction

Initial results of KMO measure of sampling adequacy indicated that factor

analysis would be an appropriate statistics for data reduction. The best solu-

tion was found after the second pass with the removal of (1) aspect, (2) slope,

(3) P2O5, (4) K2O, (5) salt, (6) erosion, and (7) seasonal maximum temperature.

After the removal of these seven variables the KMO measure of sampling

adequacy was increased for all indices (Table 2).

Total five factors were determined according to their Eigenvalues in the

second pass (Table 3). Consequently, stable models were produced. The sta-

bility of the models can be seen in the generic differentiation of the factors and

their responsible variables. For instance, the first component consists of ele-

vation and climatic variables for all indices that is reasonable because of the

clear relationship between elevation and climatic factors. Similarly, classes

derived from satellite images (NDVI and land cover classes) take part in thefifth component of all indices.

Modeling

The results of linear (multiple) regression were summarized in Table 4. The

Analysis of Variance (ANOVA) showed the acceptability of the models from a

statistical perspective, and the model summary reported the strength of the

relationship between the models and the dependent variables (Table 4). Large

values of the multiple correlation coefficient (R) indicated a strong relationship.

Table 2. Kaiser–Meyer–Olkin and Bartlett’s Test results for the 22 variables in second pass.

Second pass SWI index SIMP index Number of species

KMO measure of sampling adequacy 0.808 0.811 0.808

Bartlett’s test of sphericity Approx. v2 2502.219 2454.079 2426.494

df 210 210 210

Significance 0.000 0.000 0.000

This table summarized the KMO measure of sampling adequacy results after the second pass with

the removal of (1) aspect, (2) slope, (3) P2O5, (4) K2O, (5) salt, (6) erosion, and (7) seasonal

maximum temperature variables. Increasing KMO and Bartlett’s Test results (0.808 for Shannon–

Wiener, 0.811 for Simpson and 0.808 for NS) for the last 22 variables and the low significance levels

(0.00 for all indices) indicated the suitability of the for data reduction.

862


9/24

Moreover, the significance values of the F statistic are less than 0.05 in all models,

which means that the variation explained by the models is not due to chance. The

unstandardized coefficients were defined as the coefficients of the estimated

regression model, and they were used in the developed models (Table 5).

Figure 4. Original LANDSAT-TM (band 3) image (a), land cover map (b), and NDVI (8 bit)

classes (c) of the study area.

863


10/24

Mapping

The developed grid themes (complementary data set) and map calculator

functions of Arc/View 3.2 were employed throughout the application process

of the three models. With the power of GIS, mathematical operations

were easily conducted on grid themes. Consequently, species diversity maps of

focused indices were developed (Figure 5).

Discussion

The reliability of species diversity maps was questioned in two ways. These

were (1) mapping residuals to predict the locations where the models work

perfectly and (2) logical interpretations in ecological point of view.

Residual from regression is simply the difference between observed and

computed value (Berry and Marble 1968; McGrew and Monroe 1993), and is a

good indicator to show where the models work perfectly or imperfectly. In

general, low residual values indicate the robust models. Produced residual maps

were given in Figure 5. The percent area covered by each distinct residual class

indicated the credibility of three models. The less predictive areas for three

models covered small percentages (SWI: 7.82%, SIMP: 6.60% and NS: 7.84%),

while the strongly predictive areas contained significant parts (SWI: 64.85%,

SIMP: 68.12%, and NS: 67.57%). Moderately predictive areas were alsodetermined as approximately one fourth of the total area (SWI: 27.33%, SIMP:

25.28% and NS: 24.59%) for each index. Considering strongly and moderately

predictive areas together, it seems that each model runs very well in itself.

Overall results of the study indicated that Simpson model worked inversely

comparing the Shannon–Wiener and NS models (Figure 5). Low Simpson (0–

0.42) high Shannon–Wiener (2.70–3.59 and 3.60


11/24


12/24

T a b l e 4 .

T h e s t a t i s t i c s o f l i n e a r ( m u l t i p l e ) r e g r e s s i o n .

S h a n n o n – W i e n e r

A N O V A b

M o d e

l 1

S u m o f s q u a r e s

d f

M e a n s q u a r e

F

S i g n i fi c a n c e

R e g r e

s s i o n

2 9 . 1 7 3

2 0

1 . 4 5 9

1 7 . 0 5 5

0 . 0 0 0 a

R e s i d

u a l

2 . 9 9 3

3 5

0 . 0 8 6

T o t a l

3 2 . 1 6 7

5 5 M

o d e l s u m m a r y b

M o d e

l 1

R

R 2

A d j u s t e d R 2

S t a n d a r d e r r o r

0 . 9 5 2 a

0 . 9 0 7

0 . 8 5 4

0 . 2 9 2 4 5 2

S i m p s o n

A N O V A b

M o d e

l 1


d f

M e a n s q u a r e

F


R e g r e

s s i o n

2 . 6 6 9

2 0

0 . 1 3 3

5 . 0 2 7

0 . 0 0 0 a

R e s i d

u a l

0 . 9 2 9

3 5

0 . 0 2 7

T o t a l

3 . 5 9 8

5 5 M


M o d e

l 1

R

R 2

A d j u s t e d R 2


0 . 8 6 1 a

0 . 7 4 2

0 . 5 9 4

0 . 1 6 2 9 3 8

866


13/24

N u m b e r o f S P .

A N O V A b

M o d e

l 1


d f

M e a n s q u a r e

F


R e g r e

s s i o n

1 1 9 8 . 6 9 5

2 0

5 9 . 9 3 5

2 . 0 2 6

0 . 0 3 3

R e s i d

u a l

1 0 3 5 . 4 3 0

3 5

2 9 . 5 8 4

T o t a l

2 2 3 4 . 1 2 5

5 5 M


M o d e l 1

R

R 2

A d j u s t e d R 2


0 . 7 3 2 d

0 . 5 3 7

0 . 2 7 2

5 . 4 3 9

T h e r e s u l t s o f l i n e a r ( m u l t i p l e )

r e g r e s s i o n p r o d u c e d f o r e a c h i n d e x w e r e u n

i fi e d i n t h i s t a b l e . T h e a n a l y s i s o f v a r i a n c e (

A N O V A ) s h o w e d t h e a c c e p t t a b i l i t y o f

t h e m o d e l s f r o m a s t a t i s t i c a l p e r s p e c t i v e , a n d t h e m o d e l s u m m a r y r e p o r t e d

t h e s t r e n g t h o f t h e r e l a t i o n s h i p b e t w e e n t h e m o d e l s a n d t h e d e p e n d e n t v a r i a b l e s .

L a r g e v a l u e s o f t h e m u l t i p l e c o

r r e l a t i o n c o e ffi c i e n t ( R ) i n d i c a t e d a s t r o n g r e l a t i o n s h i p . M o r e o v e r , t h e s i g n i fi c a n c e v a l u e s o f t h e F s t a t i s t i c a r e l e s s t h a n 0 . 0 5 i n

a l l m o d e l s , w h i c h m e a n s t h a t t

h e v a r i a t i o n e x p l a i n e d b y t h e m o d e l s i s n o t d u e t o c h a n c e .

N o t e : F o r a b b r e v i a t i o n s s e e T a b l e 3 .

a P r e d i c t o r s : ( C o n s t a n t ) , S P V S D

, G E O , P H , S L D P T , O R G M , N D V I , M I N

T A , T E X T R , S O I L G , C A C O 3 , P R C P S , S T

R , P E T A N , M E T S ,

P R C P A , M A X T A , M I N T S , E

L E V , P E T S E , M E T A .

b D e p e n d e n t v a r i a b l e : S W I .

c D e p e n d e n t v a r i a b l e : S i m p s o n .

d P r e d i c t o r s : ( C o n s t a n t ) , P E T S E , O R G M , N D V I , G E O , S L D P T , P H , S P V

S D , T E X T R , S O I L G , C A C O 3 , P R C P S , S T R , M E T S , P R C P A ,

M A X T A , M I N T A , P E T A N , M

I N T S , E L E V , M E T A .

e D e p e n d e n t v a r i a b l e : N u m b e r

o f s p e c i e s .

867


14/24

The relationships between the indices and elevation can be recognized when

the elevation (Figure 1) and diversity maps (Figure 5) were examined together.

A direct relationship between the elevation and indices was detected for

Shannon–Wiener and NS models. On the other hand, this relationship turned

an inverse character in Simpson model. Depending on these results, animportant question arises: which model has the capacity to delineate real sit-

uation in the field? In basic, there are two general concepts: (1) a monotonic

decrease in species richness with increasing elevation (Stevens 1992; Huston

1994; Rahbek 1995; Brown and Lomolino 1998) and (2) a peak in richness at

intermediate elevations (800–1400 m) exemplified by a hum-shaped distribu-

tion (McCoy 1990; Rahbek 1997; Fleishman et al. 1998). Considering the

elevation range (144–1740 m) of the study area, Simpson model might be

found reasonable within the first concept. On the other hand, Shannon–Wiener

Figure 5. Plant species diversity and residual maps of Shannon–Wiener (a), Simpson (b), and NS

(c) models.

868


15/24

T a b l e 5 .

D e v e l o p e d m o d e l s o f e a c h i n d e x ( S h a n n o n – W i e n e r , S i m p s o n , a n d n u m b e r o f s p e c i e s ) .

M

o d e l s

S h a n n o n – W i e n e r i n d e x =

2 2 . 2 9

6 +

( 0 . 0 0 8 * E L E V ) +

( 1 . 6 8 8 * M E T A ) +

( 0 . 9 4 4 * M I N T S ) +

( 0 . 0 9 7 * P R C P A ) +

( 0 . 2 2 6 * G E O ) +

( 0 . 0 7 1 * O R G M ) +

( 0 . 0 2 0 * C A C O 3 ) + ( 0 . 0 0 5 * S O I L G ) + ( 0 . 1 3 9 * S P V S D )

( 0 . 9 8 1 * M E T S )

( 1 . 0 6 8 * M A X T A )

( 0 . 2 8 9 * M I N T A )

( 0 . 1 4 3 * P R C P S )

( 0 . 1 6 8 * P E T A N )

( 0 . 0 3 2 * P E T S E )

( 0 . 0 0 4 * S T R )

( 0 . 0 0 8 * T E X T R )

( 0 . 3 7 8 * P H )

( 0 . 0 8 1 * S L D P T )

( 0 . 0 0 2 * N D V I )

S i m p s o n i n d e x =

1 0 . 4 2 4 + ( 0 . 4 3 7 * M I N T A ) + ( 0 . 0 8 7 * P R C P S ) + ( 0 . 0 2 8

* P E T S E ) + ( 0 . 0 7 9 * T E X T R ) + ( 0 . 0 0 4 *

O R G M ) + ( 0 . 0 3 0 * S L D P T ) + ( 0 . 2 6 9 *

P H ) +

( 0 . 0 0 1 * N D V I )

( 0 . 0 0

2 * E L E V )

( 0 . 0 0 9 * M E T A )

( 0 . 2 2 0 *

M E T S )

( 0 . 2 9 3 * M A X T A )

( 0 . 4 2 8 * M

I N T S )

( 0 . 0 6 4 * P R C P A )

( 0 . 0 2 7 *

P E T A N )

( 0 . 0 0 3 * S T R ) ( 0 . 0 6 4 * G E O )

( 0 . 0 0 1 * S O I L G )

( 0 . 0 1 1 * C A C O 3 )

( 0 . 0 5 3 * S P V S D )

N u m b e r o f s p e c i e s i n d e x =

2 4 4 . 8 0 4 + ( 0 . 1 3 6 * E L E V ) + ( 0 . 3 4 0 * O R G M

) + ( 0 . 1 7 5 * C A C O 3 ) + ( 4 . 9 3 5 * T E X T R ) + ( 1 1 . 5 6 5 * S O I L G ) + ( 1 . 0 9 9 * G E O ) +

( 0 . 0 2 6 * N D V I ) + ( 0 . 3 6 5 * S P V S D ) + ( 2 3 . 0 8 9 * M E T A ) + ( 4 . 7 6 6 * M E T S ) + ( 3 . 8 0 4 * M A X T A ) + ( 6 . 6 9 7 * M I N T A ) + ( 1 . 1 5 2 * P R C P S ) + ( 3 . 9 3 2 * P E T S E )

( 5 . 6 4 2 * P H )

( 0 . 3 9 5 * S T R )

( 0 . 4 0 2 * S L D P T )

( 2 . 0 7 9 * M I N T S )

( 0 . 2 0 4 * P R C P A )

( 1 0 . 7 7 0 * P E T A N )

T h i s t a b l e s t a t e s t h e m o d e l s ( r e g r e s s i o n e q u a t i o n s ) a c c o r d i n g t o t h e r e s u l t s

o f m u l t i p l e r e g r e s s i o n . I n t h e e q u a t i o n s , t h e u n d e r s t a n d a r d i z e d c o e ffi c i e n t s a r e t h e

c o e ffi c i e n t s o f t h e e s t i m a t e d r e g r e s s i o n m o d e l . E a c h m o d e l a l s o h a s a c o n s t a n t v a l u e s u c h a s ; 2 2 . 2 9 6 f o r S h a n n o n – W i e n e r , 1 0 . 4 2 4 f o r S i m p s o n , a n d 2 4 4 . 8 0 4 f o r

n u m b e r o f s p e c i e s .

N o t e : F o r a b b r e v i a t i o n s s e e T a b

l e 3 .

869


16/24

and NS models could be found acceptable according to the second concept. So,

the question remains as to which diversity index reflected the reality. According

to The Ecological Society of America Committee on Land Use (Dale et al.

2000); Shannon’s index of diversity has greater sensitivity to rare cover types

and it needs to be given greater importance during interpretation. However,

Simpson’s index of diversity might be preferred in landscapes where a single

dominant land cover type is of interest. Therefore, the appropriateness of the

models depends on the aims what the decision-makers seek. Shannon–Wiener

and NS Models might be useful to detect the areas where rare and endangered

species in focus. On the other hand, Simpson model could be best fit to

determine the areas where dominant species in point of concentration.

Conclusion

In this study, we tested the modeling and mapping capabilities of some diversity

indices by using a new approach. The forest ecosystem was handled as a whole,

and the relationship between the plant biodiversity and the factors effective on

ecosystem were investigated. The complementary data about topography,

geology, soil, forest, climate, land cover, and NDVI supplied very important

information, and played the backbone role at the spatial analysis and modeling

stages. The importance of quantitative field data was also emphasized. The

results showed that plant diversity can be modeled by using index values and

complementary data set. Both GIS and RS are important tools at the analysisand visualizing (mapping) stages. According to the results; both Shannon–

Wiener and NS models could be successful to reveal the richness aspect of

species diversity, while Simpson model might be acceptable to delineate the

evenness aspect indicating single dominant land cover types. Although this

study suggested an applicable method, it is implied that researchers should be

cautious to select appropriate index according to their aims.

Acknowledgements

The authors wish to thank the following individuals for their contributions in

various parts of this research study: Vedat Toprak, Lutfi Suzen, Unal Sorman,

and Zuhal Akyurek from Middle East Technical University (METU); OsmanKetenoglu from Ankara University (AU); Ali Mermer, Ediz Unal, Tuncay

Porsuk, Oztekin Urla, and Hakan Yildiz from the GIS and RS Department of

Central Research Institute for Field Crops (CRIFC-GIS and RS); Murat

Cetiner and Irfan Artuc from the Nallihan Forest Management District

(NFMD). Thanks are also due to METU Research Fund for making financial

assistance, Soil and Fertilizer Research Institute for analyzing soil samples, and

the Keeper of the Ankara (ANK) Herbarium for making the herbarium

facilities available.

870


17/24

Appendix 1 List of identified species in the study area (endemics were stated in bold and marked

with are asterisk (*))

No. Species name Family

1 Alhagi pseudalhagi (Bieb.) Desv. Leguminosae

2 Anthyllis vulneraria L. subsp. boissieri (Sag.) Bornm. Leguminosae

3 Astragalus angustifolius Lam. subsp. angustifolius Leguminosae

4 Astragalus densifolius Lam. Leguminosae

5 Astragalus glycyphyllos L. subsp. glycyphylloides (DC.)

Matthews

Leguminosae

6 Astragalus lycius Boiss. Leguminosae

7 Astragalus macrocephalus Willd. subsp. Macrocephalus Leguminosae

8 Astragalus microcephalus Willd. Leguminosae

9 Astragalus micropterus Fischer Leguminosae

10 Astragalus squalidus Boiss. & Noe ¨ * Leguminosae

11 Astragalus trichostigma Bunge * Leguminosae

12 Chamaecytisus pygmaeus (Willd.) Rothm. Leguminosae

13 Cicer pinnatifidum Jaub. & Spach Leguminosae

14 Conorilla varia L. subsp. Varia Leguminosae

15 Dorycnium pentaphyllum Scop. subsp. anatolicum

(Boiss.) Gams

Leguminosae

16 Hedysarum varium Willd. Leguminosae

17 Lathyrus aureus (Stev.) Brandza Leguminosae

18 Lotus aegaeus (Gris.) Boiss. Leguminosae

19 Lotus corniculatus L. var. corniculatus Leguminosae

20 Lotus corniculatus L. var. tenuifolius L. Leguminosae

21 Medicago polymorpha L. var. vulgaris (Benth.) Shinners Leguminosae

22 Medicago sativa L. subsp. Sativa Leguminosae23 Medicago varia Martyn Leguminosae

24 Melilotus alba Desr. Leguminosae

25 Melilotus officinalis (L.) Desr. Leguminosae

26 Onobrychis argyrea Boiss. Subsp. argyrea Leguminosae

27 Onobrychis armena Boiss. & Huet. Leguminosae

28 Onobrychis hypargyrea Boiss. Leguminosae

29 Ononis adenotricha Boiss. var. adenotricha Leguminosae

30 Ononis spinosa L. subsp. Leiosperma (Boiss.) S ˇ irj. Leguminosae

31 Pisum sativum L. subsp. Elatius var. elatius Leguminosae

32 Trifolium arvense L. var. arvense Leguminosae

33 Trifolium barbulatum (Freyn & Sint.) Zoh.* Leguminosae

34 Trifolium repens L. var. repens Leguminosae

35 Vicia cracca L. subsp. Stenophylla Vel. Leguminosae

36 Vicia grandiflora Scop. var. grandiflora Leguminosae

37 Vicia narborensis L. var. narborensis Leguminosae38 Achillea biebersteinii Afan. Compositae

39 Achillea setacea Waldst. & Kit. Compositae

40 Acroptilon repens (L.) DC. Compositae

41 Anthemis tinctoria L. var. discoidea (All.) DC. Compositae

42 Cardopodium corymbosum (L.) Pers. Compositae

43 Carlina corymbosa L. Compositae

44 Centaurea deprassa Bieb. Compositae

45 Centaurea solstitialis L. subsp. solstitialis Compositae

46 Centaurea triumfettii All. Compositae

871


18/24

Appendix 1. (Continued )


47 Centaurea urvillei DC. subsp. Urvillei * Compositae

48 Centaurea virgata Lam. Compositae

49 Chardinia orientalis (L.) O. Kuntze Compositae

50 Chondrilla juncea L. var. juncea Compositae

51 Cichorium intybus L. Compositae

52 Cirsium arvense (L.) Scop. subsp. vestitum Compositae

53 Cirsium hypoleucum DC. Compositae

54 Crepis sancta (L.) Babcock Compositae

55 Doronicum orientale Hoffm. Compositae

56 Echinops ritro L. Compositae

57 Inula oculus-christi L. Compositae

58 Lactuca serriola L. Compositae

59 Leontodon asperrimus (Willd.) J. Ball. Compositae

60 Petasites hybridus (L.) Gaertner Compositae

61 Pilosella echioides (Lumn.) C.H. & F.W.Schultz subsp.

procera (Fries) Sell & West

Compositae

62 Pilosella hoppeana (Schultes) C. H. & F.W. Schultz

subsp. testimonialis (NP.) Sell &West

Compositae

63 Scorzonera cana (C.A.Meyer) Hoffm. Compositae

64 Scorzonera laciniata L. Compositae

65 Senecio vernalis Waldst. & Kit. Compositae

66 Sonchus asper L. Hill subsp. glaucescens (Jordan) Ball. Compositae

67 Tanacetum poteriifolium (Ledeb.) Compositae

68 Tanacetum vulgare L. Compositae

69 Taraxacum seronitum (Waldst. & Kit.) Poiret in Lam. Compositae

70 Tragopogon latifolius Boiss. var. angustifolius Boiss. Compositae

71 Xeranthemum annuum L. Compositae

72 Acinos rotundifolius Pers. Labiatae

73 Ajuga chamaepitys (L.) Schreber, subsp. chia (Schreber)

Arcangeli, var. chia

Labiatae

74 Lamium macradon Boiss. & Huet Labiatae

75 Marrubium parviflorum Fisch. & Mey. subsp. oligodon

(Boiss.) Seybold *

Labiatae

76 Mentha spicata L. subsp. tomentosa (Briq.) Harley Labiatae

77 Nepeta nuda L. subsp. albiflora (Boiss.) Gams Labiatae

78 Phlomis armeniaca Willd. * Labiatae

79 Phlomis nissolii L. Labiatae

80 Prunella vulgaris L. Labiatae

81 Salvia aethiopis L. Labiatae

82 Salvia hypargeia Fisch. & Mey. Labiatae

83 Salvia sclarea L. Labiatae

84 Salvia tomentosa Miller (Syn: S. grandiflora Etl.) Labiatae

85 Salvia verticillata L. subsp. amasiaca (Freyn & Bornm.)

Bornm.

Labiatae

86 Salvia viridis L. Labiatae

87 Scutellaria orientalis L. subsp. macrostegia (Hausskn. ex

Bornm.) Edmondson

Labiatae

88 Sideriris montana L. subsp. montana Labiatae

89 Sideritis galatica Bornm. Labiatae

872


19/24



90 Stachys annua (L.) L. subsp. ammophila (Boiss. & Bl.)

Samuelss

Labiatae

91 Stachys annua (L.) L. subsp. a nnua var. annua * Labiatae

92 Stachys cretica L. subsp. anatolica Rech. fil. * Labiatae

93 Teucrium chamaedrys L. subsp. chamaedrys Labiatae

94 Teucrium parviflorum Schreber Labiatae

95 Teucrium polium L. Labiatae

96 Thymus leucostomus Hausskn. &Velen. var. leucostomus Labiatae

97 Thymus longicaulis C. Presl subsp. longicaulis Labiatae

98 Thymus sipyleus Boiss. subsp. sipyleus Labiatae

99 Ziziphora capitata L. Labiatae

100 Cotoneaster nummularia Fisch. & Mey. Rosaceae

101 Crataegus monogyna Jacq. subsp. monogyna Rosaceae

102 Crataegus orientalis Pallas ex Bieb. var. orientalis Rosaceae

103 Crataegus tanacetifolia (Lam.) Pers. * Rosaceae

104 Potentilla recta L. Rosaceae

105 Prunus avium (L.) L. Rosaceae

106 Prunus divaricata Ledeb. subsp. divaricata Rosaceae

107 Prunus spinosa L. subsp. dasyphylla (Schur) Domin Rosaceae

108 Pyracantha coccinea Roemer Rosaceae

109 Pyrus elaeagnifolia Pallas subsp. elaeagnifolia Rosaceae

110 Rosa canina L. Rosaceae

111 Rubus ideaus L. Rosaceae

112 Rubus sanctus Schreber Rosaceae

113 Sanguisorba minor Scop. subsp. muricata (Spach) Briq. Rosaceae

114 Sorbus umbellata (Desf.) Fritsch var. umbellata Rosaceae

115 Alyssum desertorum Stapf. var. desertorum Cruciferae

116 Alyssum murale Waldst. & Kit. var. murale Cruciferae

117 Alyssum sibiricum Willd. Cruciferae

118 Arabis nova Vill. Cruciferae

119 Barbera plantaginea DC. Cruciferae

120 Cardaria draba (L.) Desv. subsp. draba Cruciferae

121 Erysimum crassipes Fisch. & Mey. Cruciferae

122 Iberis taurica DC. Cruciferae

123 Thlaspi perfoliatum L. Cruciferae

124 Turritis glabra L. Cruciferae

125 Agropyron cristatum (L.) Geartner, subsp: pectinatum

(Bieb.) Tzvelev, var: pectinatum

Gramineae

126 Aegilops umbellulata Zhuk. Gramineae

127 Brachypodium sylvaticum (Hudson) P. Beauv Gramineae

128 Dactylis glomerata L. subsp. glomerata Gramineae

129 Festuca airoides Lam. Gramineae

130 Festuca anatolica Markgr.-Dannenb. subsp. anatolica Gramineae

131 Festuca ilgazensis Markgr.-Dannenb. Gramineae

132 Poa bulbosa L. Gramineae

133 Stipa bromoides (L.) Do ¨ rfler Gramineae

134 Stipa lessingiana Trin. & Rupr. Gramineae

135 Allium scorodoprasum L. subsp. rotundum (L.) Stearn Liliaceae

136 Gagea granatellii (Parl.) Parl. Liliaceae

137 Muscari armeniacum Leichtlin ex Baker Liliaceae

873


20/24



138 Muscari longipes Boiss. Liliaceae

139 Muscari neglectum Guss. Liliaceae

140 Muscari tenuiflorum Tausch Liliaceae

141 Ornithogalum oligophyllum E.D.Clarke. Liliaceae

142 Ornithogalum fimbriatum Willd. Liliaceae

143 Ornithogalum umbellatum L. Liliaceae

144 Adonis flammea Jacq. Ranunculaceae

145 Ranunculus argyreus Boiss. Ranunculaceae

146 Ranunculus ficaria L. subs. ficariiformis Rouy & Fouc. Ranunculaceae

147 Dianthus anatolicus Boiss. Caryophyllaceae

148 Dianthus ancyrensis Hausskn. & Bornm. * Caryophyllaceae

149 Dianthus zonatus Fenzl var. zonatus Caryophyllaceae

150 Herniaria glabra L. Caryophyllaceae

151 Minuartia hirsuta (Bieb.) Hand. & Mazz. Caryophyllaceae

152 Saponaria glutinosa Bieb. Caryophyllaceae

153 Silene supina Bieb. subsp. pruinosa (Boiss) Chowdh Caryophyllaceae

154 Astrodaucus orientalis (L.) Drude Umbelliferae

155 Coriandrum sativum L. Umbelliferae

156 Falcaria vulgaris Bernh. Umbelliferae

157 Laser trilobum (L.) Borkh. Umbelliferae

158 Malabaila secacul Banks & Sol. Umbelliferae

159 Turgenia latifolia L. Hoffm. Umbelliferae

160 Zosima absinthifolia (Vent.) Link Umbelliferae

161 Alkanna orientalis (L.) Boiss. var. orientalis Boraginaceae

162 Anchusa leptophylla Roemer & Schultes subsp. lepto-

phylla

Boraginaceae

163 Cerinthe minor L. subsp. minor Boraginaceae

164 Lithospermum officinale L. Boraginaceae

165 Onosma aucheranum DC. Boraginaceae

166 Onosma bornmuelleri Hausskn. Boraginaceae

167 Onosma isauricum Boiss. & Heldr. * Boraginaceae

168 Onosma tauricum Pallas ex Willd. var. tauricum Boraginaceae

169 Digitalis ferruginea L. subsp. ferruginea Scrophulariaceae

170 Digitalis orientalis Lam. Scrophulariaceae

171 Scrophularia scopolii [Hoppe ex] Pers. var. scopolii Scrophulariaceae

172 Verbascum cherianthifolium Boiss var. cheiranthifolium.* Scrophulariaceae

173 Verbascum glomeratum Boiss Scrophulariaceae

174 Veronica chamaedrys L. Scrophulariaceae

175 Veronica multifida L. Scrophulariaceae

176 Veronica pectinata L. var. pectinata Scrophulariaceae

177 Asyneuma limonifolium (L.) Janchen subsp. pestalozzae

(Boiss.) Damboldt.

Campanulaceae

178 Asyneuma rigidum (Willd.) Grossh. subsp. rigidum Campanulaceae

179 Campanula glomerata L. Campanulaceae

180 Campanula persicifolia L. Campanulaceae

181 Legousia speculum-veneris (L.) Chaix Campanulaceae

182 Asperula stricta Boiss. subsp. latibracteata (Boiss.) Eh-

rend.

Rubiaceae

183 Cruciata taurica (Pallas ex Willd.) Ehrend. Rubiaceae

184 Galium incanum Sm. subsp. elatius (Boiss.) Ehrend. Rubiaceae

874


21/24



185 Galium palustre L. Rubiaceae

186 Galium verum subsp. verum Rubiaceae

187 Cistus laurifolius L. Cistaceae

188 Fumana aciphylla Boiss. Cistaceae

189 Helianthemum nummularium (L.) Miller. subsp. ovatum

(Viv.) Schinz & Thell

Cistaceae

190 Juniperus communis L. subsp. nana Cupressaceae

191 Juniperus excelsa Bieb. Cupressaceae

192 Juniperus foetidissima Willd. Cupressaceae

193 Juniperus oxycedrus L. subsp. oxycedrus Cupressaceae

194 Scabiosa argentea L. Dipsacaceae

195 Quercus cerris L. var. cerris Fagaceae

196 Quercus pubescens Willd. Fagaceae

197 Quercus robur L. subsp. robur Fagaceae

198 Osyris alba L. Santalaceae

199 Thesium billardieri Boiss Santalaceae

200 Paronychia dudleyi Chaudhri Illecebraceae

201 Paronychia kurdica Boiss. subsp. kurdica var. kurdica Illecebraceae

202 Iris orientalis Miller. Iridaceae

203 Acanthus hirsutus Boiss. Acanthaceae

204 Paliurus spina-christi Miller Rhamnaceae

205 Rhamnus thymifolius Bornm. * Rhamnaceae

206 Geranium robertianum L. Geraniaceae

207 Geranium tuberosum L. subsp. tuberosum Geraniaceae

208 Linum hirsitum L. subsp. anatolicum (Boiss) Hayek * Linaceae

209 Linum tenuifolium L. Linaceae

210 Fumaria cilicica Hausskn. Papaveraceae

211 Hypecoum procumbens L. Papaveraceae

212 Papaver commutatum Fisch & Mey Papaveraceae

213 Rhus coriaria L. Anacardiaceae

214 Berberis crataegina DC. Berberidaceae

215 Berberis vulgaris L. Berberidaceae

216 Salsola ruthenica Iljin Chenopodiaceae

217 Convolvulus arvensis L. Convolvulaceae

218 Corylus avellana L. var. avellana Coryllaceae

219 Sempervivum armenum Boiss. & Huet. var. armenum Crassulaceae

220 Carex flacca Schreber subsp. serrulata (Biv.) Greuter Cyperaceae

221 Carex ovalis Good. Cyperaceae

222 Equisetum palustre L. Equisetaceae

223 Euphorbia macroclada Boiss. Euphorbiaceae

224 Globularia trichosanta Fisch. & Mey. Globulariaceae

225 Hypericum perforatum L. Guttiferae

226 Malva neglecta Wallr. Malvaceae

227 Cephalanthera rubra (L.) L.C.M. Richard Orchidaceae

228 Paeonia mascula subsp. mascula Paeoniaceae

229 Paeonia peregrina Paeoniaceae

230 Pinus brutia Pinaceae

231 Pinus nigra subsp. pallasiana Pinaceae

232 Acantholimon acerosum (Willd.) Boiss Plumbaginaceae

233 Acantholimon glumaceum (Jaub. & Spach) Boiss. Plumbaginaceae

875


22/24

References

Akman Y. and Daget Ph. 1971. Quelques aspects synoptiques des climats de la Turquie. Extrait du

Bulletin de la Socie ´ te ´ Languedocienne de Ge ´ ographie 5/3: 269–300.

Akman Y. 1999. Climate and Bioclimate (Bioclimate Methods and Turkey’s Climate). Kariyer

Press, Ankara, Turkey.

Atesalp M. 1976. Soil and water sampling for analysis. Soil and Fertilizer Research Institute,

General No: 68, Farmer Pub. No: 3.

Barbour M.G., Burk J.H. and Pitts W.D. 1987. Terrestrial Plant Ecology. Benjamin Cumming

Publishing Company, Inc., California.

Berry B.J.L. and Marble D.F. 1968. Maps of Residuals from Regression. Spatial Analysis. A

Reader in Statistical Geography. Prentice Hall, Englewood Cliffs, NJ, USA, p. 7.

Brown J.H. and Lomolino M.V. 1998. Biogeography, 2nd ed. Sinauer, Sunderland, Mass.

Dale V.H., Brown S., Haeuber R.A., Hobbs N.T., Huntly N., Naiman R.J., Riebsame W.E.,

Turner M.G. and Valone J.J. 2000. Ecological principles and guidelines for managing the use of

land. Ecol. Appl. 10: 639–670.

Davis P.H. 1965–1988. Flora of Turkey and the East Aegean Islands. Edinburgh University Press,

Edinburgh.

Davis P.H. 1971. Distribution patterns in Anatolia with particular reference to endemism. In:

Davis Harper and Hedge (eds), Plant Life of South-West Asia. Botanical Society of Edinburgh,

Great Britain.

Davis P.H. and Heywood V.H. 1965. Principles of Angiosperm Taxonomy. Von Nostrand, New

York.

Dogan H.M. 1998. Visualizing spatio-temporal mesquite variation on desert grassland under dif-

ferent grazing management applications. Master thesis, Graduate School of New Mexico State

University, Master of Applied Geography, New Mexico, Las Cruces, USA.

Dogan H.M. 2001. The applications of geographic information systems and remote sensing in

agriculture. Global Space Activities and Potential in Turkey, 1st International Space

Symposium, The General Commandership of Air Force Proceedings, Ankara, Turkey. pp.

III-265–271.

Dogan M. and Akaydin G. 2002. A new species of Acantholimon Boiss. (Plumbaginaceae) from

Ankara, Turkey. Bot. J. Linn. Soc. 140: 433–448.

ERDAS 1997. ERDAS Field Guide, 4th ed. ERDAS Inc., Atlanta, Georgia.

ESRI 1994. PC Arc/Info, Command References. Redlands, CA.

ESRI 1996. Arc/View spatial analysis, Advanced Spatial Analysis Using Raster and Vector Data.

San Diego, CA.

ESRI 1997. Understanding GIS, The Arc/Info method, version 7.1 for Unix and Windows NT.

John Viley & Sons Inc., New York.

FAO 1990. Guidelines for soil description. Food and Agriculture Organization (FAO), Rome,

Italy.

Appendix 1 Continued


234 Acantholimon reflexifolium Bokhari Plumbaginaceae

235 Plumbago europaea L. Plumbaginaceae

236 Polygala anatolica Boiss. & Heldr. Polygalaceae

237 Urtica dioica L. Urticaceae

238 Valeriana alliariifolia Adams Valerianaceae

239 Valerianella vesicaria (L.) Moench Valerianaceae

876


23/24


24/24

Riitters K.H., O’Neill R.V., Hunsaker C.T., Wickham J.D., Yankee D.H., Timmins S.P., Jones

K.B. and Jackson B.L. 1995. A factor analysis of landscape pattern and structure metrics.

Landscape Ecol. 10: 23–39.

Riitters K.H., Wickham J.D., Vogelmann J.E. and Jones K.B. 2000. National land-cover pattern

data. Ecology 81: 604.

Sokal R.R. and Rohlf F.J. 1995. Biometry the Principles and Practice of Statistics in Biological

Research, 3rd ed. W.H. Freeman and Company, New York, pp. 609.

SPSS. 2001. SPSS 11.0 for Windows. SPSS Inc., Chicago.

Stevens G.C. 1992. The elevational gradient in altitudinal range: an extension of Rapoport’s lati-

tudinal rule to altitude. Am. Nat. 140: 893–911.

TUBIVES 2003. Turkish plants data service. The Scientific and Technical Research Council of

Turkey (TUBı ´TAK, http://www.tubitak.gov.tr/tubives).

Ulgen N. and Yurtsever N. 1995. Fertilizers and fertilizing guide of Turkey. Soil and Fertilizer

Research Institute No: 209, Technical Pub.No: 66.UNESCO 1973. International Classification and Mapping of Vegetation. Ecology and Conserva-

tion, Unesco, Paris.

878

Documents

A New Approach to Diversity Indices u2013 Modeling and Mapping Plant Biodiversity of Nallihan.pdf