Breeding bird communities of the Talysh mountains (Azerbaijan) … · 2010-07-15 · Introduction...

I

Ernst-Moritz-Arndt-University Greifswald

The breeding bird communities of the Talish mountains (Azerbaijan)

and their response to forest degradation

Diploma thesis in the study programme Landscape Ecology and Nature Conservation

Michael Heiß

-June 2010-

Supported by DAAD Michael Succow Foundation

Supervised by Prof. Dr. Michael Succow Dr. Martin Flade

I

Front: Semi-collared Flycatcher (Ficedula semitorquata) in primeval forest. Siov, 27.04.2008. Contact: [email protected]

II

Contents

1 Introduction and research questions .....................................................................................- 1 -

2 Study site ..............................................................................................................................- 4 -

2.1 Location .......................................................................................................................- 4 -

2.2 Climate.........................................................................................................................- 5 -

2.3 Landscape types ...........................................................................................................- 6 -

3 Methods ..............................................................................................................................- 15 -

3.1 Bird sampling.............................................................................................................- 15 -

3.2 Habitat sampling ........................................................................................................- 17 -

3.3 Data analysis ..............................................................................................................- 17 -

3.3.1 Data treatment ...................................................................................................- 17 -

3.3.2 Species richness ................................................................................................- 18 -

3.3.3 Calculation of relative abundances ...................................................................- 18 -

3.3.4 Nesting guilds....................................................................................................- 19 -

3.3.5 Statistical analysis .............................................................................................- 19 -

4 Results ................................................................................................................................- 21 -

4.1 Species number and species richness.........................................................................- 21 -

4.2 Breeding bird communities........................................................................................- 24 -

4.3 Parameters influencing breeding bird communities...................................................- 32 -

4.4 Relative abundances of bird species ..........................................................................- 35 -

4.5 Response to forest degradation on species-level........................................................- 40 -

4.6 Response to forest degradation on guild-level ...........................................................- 42 -

5 Discussion...........................................................................................................................- 44 -

5.1 Breeding bird communities........................................................................................- 44 -

5.2 Parameters influencing bird communities..................................................................- 46 -

5.3 Species and species richness......................................................................................- 50 -

5.4 Conclusion and conservation implications ................................................................- 52 -

5.5 Scope and limitations.................................................................................................- 54 -

6 Summary.............................................................................................................................- 55 -

7 Zusammenfassung ..............................................................................................................- 56 -

8 Acknowledgements ............................................................................................................- 58 -

9 References ..........................................................................................................................- 59 -

10 Appendices.....................................................................................................................- 67 -

III

Figures Figure 1: Location of Azerbaijan and the study site………………………………………………4

Figure 2: Climate diagram of Lenkoran…………………………………………………………...5

Figure 3: Climate diagram of Ardebil……………………………………………………………..5

Figure 4: Schematic of the vertical and horizontal distribution of the landscape types…………...7

Figure 5: The Caspian lowland…………………………………………………………………….8

Figure 6: Drainage channel………………………………………………………………………...8

Figure 7: Widespread natural forest stage of the Caspian forest…………………………………10

Figure 8: Grazing is common in the park-like forest stage……………………………………….11

Figure 9: Intense grazing and lopping ……………………………………………………………11

Figure 10: Hay making in the montane meadow belt…………………………………………….12

Figure 11: Montane semi-desert of the Zuvand…………………………………………………..13

Figure 12: Thorny cushion-forming tragacanthic vegetation……………………………………..13

Figure 13: Like oases meander riparian forests through the montane semi-desert……………….14

Figure 14: Lush meadows of the riparian forest………………………………………………….14

Figure 15: Rocky habitats with cushion-forming plant species…………………………………..14

Figure 16: Rocky habitats above 2000 m a.s.l……………………………………………………14

Figure 17: Data distribution per date……………………………………………………………..15

Figure 18: Overview of the study site…………………………………………………………….22

Figure 19: Species-sampling effort relationship illustrated by the landscape types……………...23

Figure 20: The nine breeding bird communities of the Talish mountains………………………..24

Figure 21: Ordination of breeding bird communities…………………………………………….33

Figure 22: Ordination of breeding bird communities within the forest degradation stages………34

Figure 23: Boxplots of height and cover of each vegetation layer for every community………...36

Figure 24: Boxplots of the altitudinal distribution of breeding bird species……………………..37

Figure 25: Relative abundances values of the 15 most common bird species per landscape type

(forest degradation stages)………………………………………………………………………...38

Figure 26: Relative abundances values of the 15 most common bird species per landscape type

(outside the forest belt)…………………………………………………………………………….39

Figure 27: Negatively and strongly negatively response of selected forest bird species…………41

Figure 28: Positive response of selected forest bird species to forest degradation……………….42

Figure 29: Response of nesting guilds to forest degradation……………………………………..43

Figure 30: Breeding bird response to forest degradation…………………………………............47

IV

TablesTable 1: Totals of surveyed transect length and totals of found breeding bird species per landscape

types…………………………………………………………………………………………….…22

Table 2: Frequency table of the breeding bird communities……………………………………...25

AppendicesAnnex 1: Commented list of the observed bird species of the Talish mountains region…………67

Annex 2: Nesting guilds of selected bird species…………………………………………………74

Annex 3: Bird community table of the nine breeding bird communities………...……………….75

Annex 1: All indicator species of each community and every combination of communities…….76

Annex 5: Statistical analysis of the site parameters of the NMDS including all transects………..78

Annex 6: Statistical analysis of the site parameters of the NMDS including all forest transects...78

Annex 7: Relative abundance values (territory/km) of each bird species per landscape type……79

Annex 8: Relative abundance values (individuals/km) of each bird species per landscape type…81

4

Introduction The breeding bird communities of the Talish mountains (Azerbaijan) and their response to forest degradation

- 1 -

1 Introduction and research questions

Deciduous broadleaf forests of the northern hemisphere are one of the most threatened

ecosystems on earth. They occur in three major, disjunct expressions (western Eurasia,

eastern Asia and eastern North America). An increasing human population and their

increasing demands for food and biofuels cause a rapid decline and fragmentation of these

forests due to agriculture, livestock farming, fuel wood gathering and timber exploitation.

Large areas of deciduous broadleaf forests are already converted into coniferous wood-

lands or have been cleared for agriculture and pastures. Primeval deciduous broadleaf

forests, which never suffered a human impact, are nowadays rare and isolated. A primeval

forest, once cut, is irreversible extinct, it can never be reconstructed (Knapp 2005).

These forests survived in western Eurasia only in small isolated remnants, mostly in in-

accessible areas or within large forested areas, for example Białowieża Forest at the

Polish/Belarussian border (Tomiałojć & Wesołowski 2004), Šúr National Nature Reserve

in Slovakia (Korňan 2009), Colchic forests in Georgia and partly the Caspian forests in

Azerbaijan and Iran (Knapp 2005). The largest remnants of primeval deciduous broadleaf

forests on earth can be found in northern Iran and the adjacent Azerbaijan and is called the

Caspian or Hirkanian forest. This forest is regarded as tertiary relict forest with a remark-

able biodiversity including ca. 90 tree and 211 shrub species with a high proportion of

endemic species (Prilipko 1954, Knapp 2005, Seifollahian 2005). 100000 ha (hectare) of

pristine forest can be found within the 1.8 million ha sized forest belt at the northern

slopes of the Alborz mountains in Iran (Knapp 2005). Globally, the Caspian forest is one

of the most important deciduous broadleaf forests, due to its remarkably biodiversity,

primeval conditions and large extension. In Iran are 590000 ha of Caspian forest under the

protection of 1 national park, 5 nature reserves and 3 natural monuments besides further

forest reserves and wildlife refugees (Knapp 2005). Even outside the protected areas, the

Caspian forest in Iran appears naturally due to its sustainable use supported by laws for a

natural forestry, forestry plenter-use, different aged deciduous forest, indigenous tree

The breeding bird communities of the Talish mountains (Azerbaijan) and their response to forest degradation Introduction

- 2 -

species, resettlement of the wood population out of the forest, reduction of silvopastures

(Abdollahpour & Atui 2005, Mohadjer 2005, Saffari 2005).

Azerbaijan holds a total of about 100000 ha of the Caspian forest (Michael Succow

Stiftung 2009), but the current situation of the Caspian forest in Azerbaijan is different to

that in Iran. After the breakdown of the Soviet Union in 1991, no laws for protecting the

forest have been reintroduced followed by an absence of a regular forestry or management

plans. Timber exploration is for many inhabitants the only monetary income after Russian

kolkhozes, as main employer, disappeared. Thus, an ‘open access’ situation arose, which

led to a rapid forest degradation caused by overusing the forest (Noack & Hidayatov

2007, Michael Succow Stiftung 2009). Further reasons for the degradation are

silvopastures and tree lopping for livestock fodder and fuel wood gathering (Scharnweber

et al. 2007). Currently, only 17% are treated as pristine forest, whereas 44% show a heavy

degradation (Rietschel 2010). The size of the Hirkan National Park in Azerbaijan was en-

larged in 2008 to 40358 ha (MENR 2010).

No information is actually available about the impact of the ongoing degradation to the

fauna of the Caspian forest. Birds are useful bioindicators. Their ecological behaviour is

widely known and they respond fast to a changing environment (Flade 1994). The results

and understandings of the bird’s response to forest degradation can be used to derive re-

actions of the whole fauna of that ecosystem and subsequently guide further nature con-

servation assignments.

Since about 150 years, the Caspian forest of Azerbaijan is of ornithological interest. Its

scenically beauty and the uniqueness of its flora and fauna within the country and the

former Soviet Union led to several biological studies. Gustav Radde, a German in Russian

service, was one of the first who described in detail the avifauna of the region (Radde

1884, Radde 1886a, Radde 1886b) followed by studies of several Russian biologists. The

last intense researches of the avifauna were conducted in the 1970s (e.g. Agaeva 1972,

Agaeava & Mustafaev 1973, Loskot 1978, Agaeva 1979, Mustafaev & Agaeva 1981).

Nevertheless, there are many further gaps of knowledge according to the occurrence and

distribution of birds. Several species listed by Radde (1886b), for example Caspian Snow-

Introduction The breeding bird communities of the Talish mountains (Azerbaijan) and their response to forest degradation

- 3 -

cock (Tetraogallus caspius), Crimson-winged Finch (Rhodopechys sanguinea) or Radde’s

Accentor (Prunella ocularis), seemed to be disappeared in the last 130 years, as no recent

records exist. The Caspian Snowcock was found by Radde (1884, 1886a, 1886b) and

Patrikeev (2004) expected this species to be extirpated in the Talish mountains. The

occurrence of Shikras (Accipiter badius) in that region was unclear for a long time

(Patrikeev 2004, Gregory 2007), but could recently clarified as a regular breeding bird of

the Caspian lowland (Heiss & Gauger 2009). The occurrence of further species, like Pha-

sianus colchicus talischensis, Gypaetus barbatus, Apus affinus, Halcyon smyrnensis,

Picus canus, Phoenicurus erythrogaster or Tichodroma muraria, are up to day unclear.

Species responses to forest degradation, fragmentation and deforestation have garnered

much recent interest throughout all continents and all forest types (e.g. Edenius &

Elmberg 1996, Poulsen 2002, Sekercioglu 2002, Echeverria et al. 2007, Fuller et al.

2007b, Murakami et al. 2008). The impact of the degradation of the Caspian forest to the

fauna is unknown. No detailed studies are published yet.

Therefore, the aim of this study is to answer the following questions:

• What are the breeding bird communities of the Talish mountains and which

species do they contain?

• Which parameters are responsible for their species composition?

• How do the breeding birds respond to the degradation of the Caspian forest in

Azerbaijan?

• What are the derived nature conservation implications?

The breeding bird communities of the Talish mountains (Azerbaijan) and their response to forest degradation Study site

- 4 -

2 Study site

2.1 Location

The study site is located in the southernmost part of Azerbaijan at the border to Iran

(Figure 1). It includes the districts of Masally, Lenkoran, Astara and Lerik. The total area

covers 3960 km² (Statistical Committee Azerbaijan 2009).

Figure 1: Location of Azerbaijan and the study site. Note the dark green band in northern Iran and adjacent southern Azerbaijan indicating the distribution of the Caspian forest. Clearings of the nearby Caspian lowland are light green (image source: NASA 2004).

Study site The breeding bird communities of the Talish mountains (Azerbaijan) and their response to forest degradation

- 5 -

The altitude ranges from -24 m a.s.l. (metres above sea level) at the Caspian coast in the

East and 2492 m a.s.l. of the Kiumiurkei mountain at the Iranian border in the West

(Skvorov 1976a, Skvorov 1976b).

The study site contains several protected areas. The three largest are the Hirkan

National Park, which was established in 2004 and enlarged in April 2008 to 40358 ha

(MENR 2010), the Gizilaghaj state nature reserve at the Caspian coast, which covers

88360 ha and is an important area for wintering and breeding waterbirds (Patrikeev 2004,

Sultanov 2008, MENR 2010) and the 15000 ha sized Zuvand state nature sanctuary

(MENR 2010).

2.2 Climate

The study site can be subdivided into two differing climatic regions. The eastern slopes of

the Talish mountains down to the coastal lowland are characterised by a warm-temperate

and humid climate. The annual rainfalls exceed 1000 mm per year and peak in February

and October. The summer temperatures are warm and the winters are mild (Figure 2).

Winter means rarely drop below zero (Mammadov et al. 2007, Mühr 2007).

Figure 3: Climate diagram of Ardebil 50 km south of Zuvand upland based on data of IRIMO (2010).

Figure 2: Climate diagram of Lenkoran after Mühr (2007).


- 6 -

The Zuvand area is the second climatic region. Due to mountain ridges, which shield

humid air masses coming from the Caspian sea, the Zuvand upland is much drier and the

climatic situation is similar to the adjacent Iranian semi-desert. It has an annual mean

temperature of 4-8°C. The winters are cold and occasionally the temperature drops below

-20°C. The summers show a distinctive drought from June to October and the annual

rainfalls are between 200-400 mm per year (Mammadov et al. 2007, cf. Figure 5 in Knapp

2005). No meteorological data was available from the Zuvand region and thus the data of

Ardebil (Iran) with a similar climate was taken depicted in Figure 3.

2.3 Landscape types

The Talish mountains are subdivided into six landscape or habitat types (Figure 4). The

forest belt contains five degradation stages. Each type has its own characteristics:

Caspian lowland

From the shore of the Caspian sea to the foothills of the Talish mountains stretches the

Caspian lowland, a long but rather narrow stripe including the cities Masally, Lenkoran

and Astara (Figure 1, Figure 4). Once it was covered with subtropical broad-leaf forests

and impassable wetlands (Radde 1886a). To fight malaria and to gain land for agriculture,

forests had been cleared and a widespread system of drainage channels indicate the large

range of former wetlands (Patrikeev 2004). Nowadays, the lowland is largely covered

with fields, pastures and human settlements (Figure 5, Figure 6). The last remnants of the

former lowland forest can be found in a 91 ha sized part of the Hirkan National Park

called ‘Moscow forest’ (MENR 2010). The only further forest-like structures in the low-

land are afforestations of Quercus spec., orchards or cemeteries.

Costal habitats like lagoons, shores or large reedbeds are excluded from this study.


- 7 -

Figure 4: Schematic of the vertical and horizontal distribution of the landscape types based on Grossheim (1926), Knapp (2005) and recent Google Earth (2010) satellite images.


- 8 -

Forest belt and the degradation stages

The deciduous broadleaf forest of the Talish mountains ranges from the sea level up to

1800 m a.s.l. (MENR 2004) but the upper tree line is mostly reduced to lower altitudes

due to human activities. The number of tree species is high (ca. 90) and includes many

endemic or tertiary relict species like Gleditsia caspica, Parrotia persica, Quercus casta-

neifolia, Albizzia julibrissin, Buxus hyrcana, Ruscus hyrcana or Acer hyrcanum (MENR

2004, Hajiyev 2006). Details about forest types and tree species composition of the

Caspian forest are given in Knapp (2005).

Scharnweber et al. (2007) identified and described six degradations stages of the

Caspian forest in the Talish mountains, on which this study base:

Natural forest stage

This stage shows no signs of human activities. It is generally restricted to higher altitudes

above 500 m a.s.l., but occurs also elsewhere along steep slopes or apart from human

settlements and roads (Figure 7). The natural forest stage can be treated as a primeval

forest comparable to the Białowieża forest in Poland. These relicts can be distinguished

from other temperate deciduous or mixed forests by their large heights, multistorey profile

of stands, a diverse tree community and large amounts of dead wood (Wesołowski 2007).

I found this stage near the village Siov and the abandoned village Armudy within the

Hirkan National Park.

Figure 6: Drainage channel. Sirabil, 21.05.2008. Figure 5: The Caspian lowland is characterised by agriculture and settlements. Sirabil, 21.05.2008.


- 9 -

Slightly disturbed forest stage

The slightly disturbed forest stage is similar to the natural forest stage. The human impact

is low and often not obvious. Thus, this stage was in a few cases difficult to distinguish

from the natural forest stage. Hints for a slightly disturbed forest stage are paths, single

snags, partly missing dead wood and evidences for grazing like excrements or traces of

domestic animals or a reduced understorey. According to Scharnweber et al. (2007), are

there no differences to the natural forest stage concerning tree species composition and

stand structure. Old-growth trees are a common aspect.

Intermediate disturbed forest stage

Logging, lopping and grazing are clearly visible in this stage. This stage can be found near

villages (2-4 km) or along roads. Loose cattle or grazing flocks of sheep can regularly be

seen in the forest. The tree layer is still good developed, but with a less diverse stand

structure and some gaps in the canopy. The intermediate disturbed forest is more suffused

with light. This stage was common and widespread, especially along the main roads from

Lenkoran to Lerik and from Masally to Iardimli.

Park-like forest stage

This degradation stage is characterised by a high logging and grazing activity, which lead

to a park-like appearance (Figure 8). Trees are mostly cut down and only single old-

growth trees are left, which mostly have a chopped appearance due to lopping. Snags are

common. The foliage cover of the tree layer is low (ca. 5-10%). The herb layer is short

due to grazing. Dense forest fragments can occur. This stage can be found closer to

villages (1-2 km), on ridge tops or in plains. Large areas of this stage could be found e.g.

near Dashtatuk, Bilarsar, Tankivan or Günesli.

Shrubby woodland stage

The shrubby woodland stage is situated close to villages. It is the result of a frequently

used former forest. Large trees are almost completely removed and shrubs are dominating


- 10 -

(Figure 9). Due to lopping and pollarding, trees appear like small bushes, especially Car-

pinus betulus and Quercus castaneifolia, which rarely exceed heights of more than three

metres. Grazing intensity is very high. This stage can also be found at higher altitudes

along the upper tree line. Here, the occurrence is probably naturally triggered, as a growth

of trees is inhibited by the rougher climate (lower precipitation, lower mean annual temp-

erature, strong winds).

I neglected the described treeless bracken fern stage that is dominated by Pteridium

aquilinum. It is limited to a few small locations only and could not be adequately sampled.

The work of Rietschel (2010) confirmed the classification of Scharnweber et al. (2007).

Figure 7: Widespread natural forest stage of the Caspian forest. Siov, 26.04.2008.


- 11 -

Montane meadow belt

Montane meadows are distributed above the forest belt around the smaller towns Lerik,

Shingedulan and Iardimli. Due to deforestation, it has an open landscape character with

gentle slopes ranging from 700 to 1200 m a.s.l. (Figure 4). Typical are meadows, which

are used for hay making (Figure 10). Furthermore, agriculture and pastures are common.

Figure 8: Grazing is com-mon in the park-like forest stage. Lopped trees are visible in the background. Zunguliash, 28.05.2008.

Figure 9: Intense grazing and lopping is responsible for a shrubby appearance of the forest. Logging activity completely re-moved larger trees. Shinaband, 06.05.2008.


- 12 -

Figure 10: Hay making in the montane meadow belt. Muria, 05.06.2008.

Forest-like structures are rare, but along steep slopes the shrubby woodland stage is partly

developed. Otherwise, shrubs are scattered along field paths or roads. Radde (1886a)

described the area around Lerik 130 years ago also as an open landscape with agriculture.

Zuvand

South of Lerik, the villages Mistan, Kialvaz, Allar, Gosmalian and Shonadzohla comprise

the Zuvand upland (Figure 1). Here, three landscape types can be found:

Montane semi-desert

Gentle slopes barely covered with thorny cushion-forming tragacanthic vegetation (e.g.

Tragacantum, Acantholimon) feature the montane semi-desert (Atamov et al. 2006,

Hajiyev 2006, Michael Succow Stiftung 2009) (Figure 11, Figure 12). Tree growth is

prevented due to low annual precipitations and long summer droughts (Figure 3).


- 13 -

However, a stand of Juniper excelsa in a runnel near Diwagatsh may indicate the

distribution of a former Irano-Turanian steppe-forest which can be found in the adjacent

Iran (Zohary 1973). The altitudinal distribution of the montane semi-desert ranges from

1400 to 2500 m a.s.l. (Figure 4, Figure 1). Humans use the semi-desert mainly as pastures.

At a few locations, I observed transformations into cultivated land.

Riparian forest

Embedded in the barren montane semi-desert along small rivulets, open park-like riparian

forests of willow species (Salix alba, S. purpurea, S. caprea) and Populus nigra can be

found (Michael Succow Stiftung 2009). Riparian forests range from 1200 to 1800 m a.s.l.

and are located for example around the villages Gosmalian, Mistan, Kialvaz and Govari.

They are rather narrow stripes and restricted to water containing valley bottoms (Figure

13). The inhabitants of the villages along the forest created a complex irrigation system to

water their lush and flower-rich meadows, fields and gardens (Figure 14). Large terraces

have been built and are widely used as orchards or almond plantations. The vegetation

structure of this type is similar to the open appearance of the park-like stage within the

forest belt.

Figure 11: Montane semi-desert of the Zuvand. Divagach, 08.05.2008.

Figure 12: Thorny cushion-forming tragacanthic vegetation. Mistan, 19.05.2008.


- 14 -

Rocky habitats

I found rocky outcrops at altitudes ranging from 1200-2500 m a.s.l., especially around the

villages Mistan, Kialakhan, Digia, Pirasora, Bizeir and Khozavi. Steep, inaccessible cliffs

and stony terrain is typical (Figure 15, Figure 16). At lower, more humid altitudes ranging

from 1200 to 1600 m a.s.l., this type it is often mixed with dense shrubs or pastures. In

drier, higher regions of the Zuvand, rocky habitats are often encircled by montane semi-

desert and include cushion-forming plant species.

Figure 14: Lush meadows and planted willow trees are a common aspect of the riparian forest. Shonadzhola, 04.05.2008.

Figure 15: Rocky habitats with cushion-forming plant species. Piresora, 30.05.2008.

Figure 13: Like oases meander riparian forests through the montane semi-desert. Kialakhan, 01.06.2008.

Figure 16: Rocky habitats above 2000 m a.s.l. Mistan, 17.05.2008.

Methods The breeding bird communities of the Talish mountains (Azerbaijan) and their response to forest degradation

- 15 -

3 Methods

3.1 Bird sampling

I surveyed birds from 03 April to 08 June in 2008. This short period was necessary to

guarantee the highest territorial activity of birds during the breeding season in spring.

After finishing the survey on 08 June 2008, further excursions took place until 15 July

2008 to investigate more places in the Talish region.

I visited the different landscape types equally during spring to avoid date-biased

results, but the higher altitudes (montane semi-desert, rocky habitats) were studied later

(Figure 17).

Figure 17: Data distribution per date. Each grey point represents a day at which transects were surveyed according to the landscape types.

I chose the line transect method, which was the most suitable method for this bird

survey. This method is a compromise between point count method and mapping-census.

The main advantage of the line transect method is the good ratio between less time effort

and gained data (Flade 1994, Bibby et al. 1995, Südbeck et al. 2005). To accomplish this

large area (3960 km²) in a short time span (03 April - 08 June), line transect method

The breeding bird communities of the Talish mountains (Azerbaijan) and their response to forest degradation Methods

- 16 -

would give the best results. Südbeck et al. (2005) recommend to survey each transect four

times during the breeding season. This recommendation was refused and each transect

was surveyed only once. This modification was necessary to achieve a widespread

overview of the large Talish region. Furthermore, recovering an already visited transect

would be difficult in remote regions without paths or the reliable accuracy of the GPS

device (Garmin GPS 60), especially in dense forests.

Methodological problems of transects without repeated counts are discussed in

Pëterhofs & Priednieks (1989) and Hilden & Järvinen (1989). Main concern according to

Hilden & Järvinen (1989) was that only 50% of the breeding pairs are detectable during

one count. This problem was reduce as I walked transects slowly (ca. 1-1.5 km/hour)

recommended in Südbeck et al. (2005), but faster in open landscapes, for example in

montane semi-deserts where bird are not abundant and easier to recognise. A too slow

pace could increase double countings of birds and was therefore avoided. Nevertheless,

like most bird survey methods, this method is biased against quiet, secretive and nocturnal

species (Bibby et al. 1995).

It was not necessary to consider different detectability among species because I do not

use statistical comparisons among species. A quiet and well-camouflaged treecreeper

(Certhia) is always more difficult to recognise than a loud singing thrush (Turdus). An

interspecific comparison would be inappropriate, but intraspecific comparisons of a single

thrush species and their abundance of different habitats are legitimately.

Surveying was done in the morning and evening when the diurnal bird activity peaked.

No counts took place under poor weather conditions like rain or strong winds (Flade 1994,

Bibby et al. 1995, Südbeck et al. 2005).

Each study site and transect route was selected following the determination of

landscape types (cf. chapter 2.3.). I avoided human settlings. The transect line was as

straight as possible, which was at a few locations difficult to manage because of difficult

topography, rivers or dense, thorny shrubs.


- 17 -

3.2 Habitat sampling

I took every 500 m a waypoint to characterise each transect line. This point describes a

500 m segment of the transect line in detail. Each waypoint consist of date and time,

landscape type, geographic coordinates (WGS 84), height above sea level, slope ex-

position, slope steepness, relief (plain, valley bottom, lower slope, middle slope, upper

slope, ridge top) and a characterisation of vegetation structure including average height

and cover of each vegetation layer (herb, shrub, lower tree and upper tree layer). For

describing a complete transect line, I used the data of the waypoints to determine transect

length, mean altitude, mean exposition, landscape type composition and vegetation struc-

ture, including mean height and cover of the vegetation layers.

Along each transect line all birds and their activity (e.g. singing, calling, feeding, flying

etc.) were continuously noted. I took the coordinates of every bird within the transects,

which consisted of transect metre and distance of the bird to the transect line.

3.3 Data analysis

I compiled the data with Microsoft Access 2003 and Microsoft Excel 2003. Processing the

data and statistical calculations were performed using the statistical software package R

version 2.9.2. (R Foundation for Statistical Computing 2009). For compiling maps the

geographic information system QGIS was used (Quantum GIS Development Team 2009).

Bird taxonomy follows the Clements Checklist to the Birds of the world 6.3.2 (Clements

2008)

3.3.1 Data treatment

The determination of “territories” was less constrained, because every transect was visited

only once. Väisänen (1989) categorised the breeding evidences for bird species, which I

modified in this study. I treated the following observations as a territory: a single bird

(male, female or juvenile), a pair, an occupied nest or a family. Birds breeding in colonies

(e.g. Delichon urbica, Apus apus, Merops apiaster) show no territorial behaviour and the


- 18 -

number of territories was estimated by counting the total number of adults dividing by two

or counting the nests if possible (Bibby et al. 1995). Some bird species (e.g. Hirundo

rustica, Merops apiaster, Sturnus vulagris) searched for food in the surrounding areas of

their breeding colonies, which were not directly touched by the transect line or maybe

overlooked. I treated these birds also as territorial because I expect their breeding colonies

to be within surveyed landscape type. I treated flying individuals as territories as long as

they show territorial behaviour or at least an association to the habitat. Migrating or high

flying birds, non-breeding flocks or birds apart from their typical nesting habitats (e.g.

Acrocephalus scirpaceus in park-like forest stage or Merops persicus in woody habitats)

or apart from their breeding ranges (Lanius isabellinus, Locustella fluviatilis) were listed

as no territories. Birds further than 200 metres from the zero transect line are not included

in the calculations to guarantee their relation to the landscape types.

3.3.2 Species richness

To determine the relative species richness of each landscape type, I applied a comparison

using log10-transformed values of species number and log10-transformed transect length

for each landscape type originating from Table 1. The log10-values on both axes are neces-

sary to fit a simple regression line. This linear regression analysis shows not a species-

area relationship sensu the theory of island biogeography. It reveals a species-sampling

effort relationship from which the relative species richness can be derived. However, this

analysis lacks generally of quality, as an area-proportionate sampling, which also lead to

time-based problems, could not be achieved (Mac Nally & Horrocks 2002) and results

should interpreted with caution.

3.3.3 Calculation of relative abundances

For comparison of each bird species and their distribution according to each landscape

type (not to each community), I calculated the relative abundances of every bird species

using this formula: A=T/L

A= relative abundance, T= number of territories, L= total transect length per landscape type (km)


- 19 -

To assure comparability of the relative abundances of species to each landscape type

the calculation was modified. According to the variable migratory behaviour of each bird

species, are long-distance migrants not included in the early transects of this study, as they

arrive later in spring. Using the total transect length per landscape type would under-

estimate their relative abundance. Hence, for all bird species the date of its first obser-

vation was used to sum the length of all kilometres per landscape (L) starting with this

day. Therefore, the length used for calculation for long-distance migrants is shorter than

for residents.

3.3.4 Nesting guilds

At the guild level, more coherent in the responses to forest degradation can be expected,

because species belong to the same guild have similar life history traits, dispersal ability,

and spatial distribution and may show similar responses to disturbance (Pearman 2002).

To analyse, which guilds are affected by forest degradation, I partitioned them into four

nesting guilds (ground nesters, shrub nesters, canopy nesters, cavity nesters) basing on

literature review (Glutz von Blotzheim & Bauer 1991, Glutz von Blotzheim & Bauer

1994, Flade 1994, Glutz von Blotzheim & Bauer 1998, Urquhart, & Bowley 2002,

Alström & Mild 2003, Patrikeev 2004, Andretzke et al. 2005, Kirwan et al. 2008) (Annex

2). If a bird species belongs to more than one guild, I used the most characteristical.

Cuculus canorus had to be excluded from this analysis, because it is unclear which nest-

ing guilds he parasitises most in the Talish mountains.

3.3.5 Statistical analysis

To classify breeding bird communities, I applied a hierarchical, agglomerative cluster

analysis using a dissimilarity matrix with Bray–Curtis measure (Bray & Curtis 1957,

Leyer & Wesche 2007) based on species abundance for each transect (territories per km)

and Ward’s minimum variance method as linking procedure (Ward 1963). As the Ward’s

minimum variance method is not compatible to dissimilarity matrix using semi-metric

Bray-Curtis measure (Legendre & Legendre 1998, McCune & Grace 2002, Leyer &


- 20 -

Wesche 2007), I took the square root of the dissimilarity matrix. The abundance data have

not been transformed and contains rare species.

Basing on the output of the cluster analysis, I then created a bird community table,

where the communities are arranged following an altitudinal gradient. Within the forest

belt, the order followed the intensity of degradation from natural forest stage to the

shrubby woodland stage. The socio-ecological species groups were clustered by

transposing the dissimilarity matrix to obtain a rough overview, and I finally arranged

them manually to their best fit with the help of my expert knowledge and further

observations besides the surveying time.

I also used NMDS (non-metric multidimensional scaling) to plot ordinations using the

function “metaMDS” of the package “vegan” of the statistical software package R. In

these NMDS are the transects arranged in scatter-plots regarding to their species compo-

sition and abundance (territories/km per transect) to illustrate similarities and dissimi-

larities between the transect and the gradients which influences their composition (Leyer

& Wesche 2007). NMDS base on Bray–Curtis measure (Bray & Curtis 1957) without the

square root of the dissimilarity matrix. I then used the stress-value to deter-mine the num-

ber of dimensions for the two NMDS (McCune and Grace 2002).

For identifying associations of species to communities or combinations of communi-

ties, the R package “indicspecies” (version 1.0) the function “multipatt” was applied using

1000 permutations. I rejected the indicator species analysis by Dufrene & Legendre

(1997) as it identifies species indicating only one community but not a combination of

communities.

In the following text, the code of the significance level is ‘*’ = p-value < 0.05, ‘**’ = p-

value < 0.01 and ‘***’ = p-value < 0.001.

Results The breeding bird communities of the Talish mountains (Azerbaijan) and their response to forest degradation

- 21 -

4 Results

I surveyed 178 km (kilometre) consisting of 94 transects (Figure 18). The length of

transects differed between 200 and 4500 metres. The average length was 1894 metres

(n=94). I excluded three transects from the cluster analysis and NMDS analysis, as their

length was below 500 m, which do not represent breeding bird communities adequately.

4.1 Species number and species richness

I found 10104 individuals of 197 bird species (Annex 1). I expect about 147 species of

them to breed in the Talish region (Table 1). (Further breeding birds of wetlands and

costal habitats are not included.)

From the faunistic aspect, the rediscovery of breeding Shikras (Accipiter badius) was a

surprising highlight, as no confirmed breeding records existed since 1933. This is the only

breeding site within the Western Palaearctic (Heiss & Gauger 2009). Further breeding

species are new to that region or rediscovered after more than a century like Prunella

ocularis, Rhodopechys sanguinea, Irania gutturalis or Bucanetes githagineus (Annex 1).

Table 1: Totals of surveyed transect length and totals of found breeding bird species per landscape types. Note that bird species can occur in more than one landscape type.

Landscape type Transect length [km] Species number

Caspian lowland 23.3 54

Natural forest stage 8.6 32

Slightly disturbed forest stage 12.1 40

Intermediate disturbed forest stage 26.1 46

Park-like stage 11.9 41

Shrubby woodland stage 20.4 59

Montane meadow belt 20.6 48

Riparian forest 15.7 44

Montane semi-desert 18.1 44

Rocky habitats 21.2 68

Total 178 147

The breeding bird communities of the Talish mountains (Azerbaijan) and their response to forest degradation Results

- 22 -

Figu

re 1

8: O

verv

iew

of t

he st

udy

site

. C

olou

red

lines

illu

stra

te th

e su

rvey

ed tr

anse

cts.

The

num

bers

are

the

trans

ect n

umbe

rs.


- 23 -

However, several species listed in Agaeva (1979) or Patrikeev (2004) could not be con-

firmed in this study like Poecile hyrcana, Hippolais languida, Emberiza buchanani,

Gypaetus barbatus, Apus affinus, Halcyon smyrnensis, Picus canus(!), Phoenicurus ery-

throgaster, Carpospiza brachydactyla or Tichodroma muraria.

The most species-rich landscape type are rocky habitats (Figure 19). Rather poor in

species are the less disturbed forest degradation stages (natural, slightly and intermediate

disturbed forest stages). According to this analysis, montane semi-desert and montane

meadow belt are also regarded as relatively poor in species. Interestingly, the heavily

degraded forest stages (park-like forest stage and shrubby woodland) show the highest

relative species richness.

Figure 19: Species-sampling effort relationship illustrated by the landscape types (Casp = Caspian lowland, Nat = Natural forest stage, Sli = Slightly disturbed forest stage, Int = Intermediate disturbed forest stage, Park = Park-like forest stage, Shr = Shrubby woodland stage, Mead = Montane meadow belt, Rip = Riparian forest, Mon = Montane semi-desert, Rock = Rocky habitats) and a fitted regression line (red line) indicating the same amount of relative species-richness. Landscape types above the regression line are rich in species and below are poor in species.


- 24 -

4.2 Breeding bird communities

Figure 20: The nine breeding bird communities of the Talish mountains represented by clusters. The cluster analysis bases on species abundance of each transect using Bray-Curtis measure (Bray & Curtis 1957) and Ward’s method (Ward 1963) as linking procedure. Cluster 6 consists of two merged clusters.

The cluster analysis identified ten breeding bird communities (Figure 20). I merged two

clusters (cluster 6) by a local reduction of the cut level, because in field both had a

shrubby appearance. Thus, I found nine breeding bird communities in the Talish moun-

tains.

The frequency table (Table 2) and the bird community table (Annex 3) show these 9

breeding bird communities and 27 socio-ecological species groups. Species of each socio-

ecological group reflect the same ecological demands and are similar distributed to the

communities. Due to the determination of clusters (communities) by breeding birds, some

transects are not within their assumed landscape type, for example transect 6 in

community 5 (Annex 3). This inaccuracy was neglected, because identifying habitats by

plants, as Schwarnweber et al. (2007) did and which I used for landscape type deter-

mination, and birds are different approaches.


- 25 -

Table 2: Frequency table of the breeding bird communities. This table is a simplification of the bird community table (Annex 3). Theoretically, a comparison of frequencies of transects with different lengths is not allowed due to species-area relationships. Therefore, the frequency values may give a slightly different impression than the bird community table does (for details see Annex 3). Community 1 2 3 4 5 6 7 8 9 Number of transects 9 10 13 10 11 12 5 10 11 Transect length [m] 2589±1033 2160±1309 1662±833 1980±1209 1636±609 2042±1336 2740±864 1980±756 1355±610 Mean altitude [m a.s.l.] -4±24 506±246 443±243 204±123 1060±290 1045±485 1482±144 1685±209 1846±233 Landscape composition: . . . . . . . . . Caspian lowland [%] 100 . . . . . . . . Natural forest stage [%] . 41 . . . . . . . Sligthly disturbed forest [%] . 46 10 . . . . . . Intermediate disturbed forest [%] . 13 56 38 18 . . . . Park-like stage [%] . . 23 35 . . . . . Shrubby woodland stage [%] . . 11 27 54 12 . . . Montane meadow belt [%] . . . . . 82 . . . Riparian forest [%] . . . . 11 . 100 . . Montane semi-desert [%] . . . . . 6 . 82 13 Rocky habitats [%] . . . . 17 . . 18 87 Vegetation structure: . . . . . . . . . Upper tree layer height [m] 4.9±3.0 25.1±2.1 20.1±2.5 19.2±4.2 10.7±7.2 1.9±0.7 15.3±2.0 - 3.0±0.0 Upper tree layer cover [%] 4.3±4.3 42.3±18.4 20.2±10.9 14.1±9.3 8.8±8.6 1.1±0.8 6.1±2.0 - 0.2±0.0 Lower tree layer height [m] 1.7±1.7 13.5±4.9 9.6±2.5 9.5±4.2 6.3±2.8 - 6.6±2.4 - - Lower tree layer cover [%] 1.7±1.7 22.2±12.6 21.3±14.0 24.8±21.8 13.2±10.2 - 6.8±2.2 - - Shrub layer height [m] 1.5±0.7 1.8±1.2 1.3±0.6 1.6±0.4 2.0±0.8 1.0±0.7 1.9±0.3 0.5±0.4 0.9±0.4 Shrub layer cover [%] 6.7±5.8 8.2±3.8 16.1±11.3 14.2±9.5 38.6±22.3 3.2±4.1 9.6±5.9 0.9±0.9 5.3±4.6 Herb layer height [m] 0.3±0.1 0.2±0.1 0.1±0.0 0.2±0.0 0.1±0.0 0.2±0.2 0.2±0.0 0.1±0.0 0.1±0.0 Herb layer cover [%] 79.5±8.2 39.2±19.6 30.2±12.4 20.7±11.6 37.4±22.2 59.5±17.8 61.3±7.1 13.8±10.0 24.0±12.8 Carduelis carduelis 67 10 50 90 36 38 100 40 45 Cuculus canorus 67 70 43 70 36 15 20 40 73 Parus major 56 80 86 100 100 31 100 . 18 Hippolais pallida 67 . . . . . . . . Acrocephalus schoenobaenus 67 . . . . . . . . Circus aeruginosus 56 . . . . . . . . Calandrella rufescens 33 . . . . . . . . Motacilla flava 44 . . . . . . . . Cercotrichas galactotes 22 . . . . . . . . Streptopelia decaocto 22 . . . . . . . . Merops persicus 22 . . . . . . . . Passer montanus 33 . . . . . . . . Remiz pendulinus 22 . . . . . . . . Acrocephalus arundinaceus 67 . . . . 8 . . . Francolinus francolinus 11 . . . . . . . . Acrocephalus scirpaceus 11 . . . . . . . . Ixobrychus minutus 11 . . . . . . . . Sylvia mystacea 11 . . . . . . . . Alcedo atthis 67 10 7 10 . . . . . Glareola pratincola 11 . . . . . . . . Hirundo rustica 100 . . . . 38 . 10 . Coracias garrulus 56 . . . . 8 . . . Coturnix coturnix 11 . . . . 23 . . . Melanocorypha calandra 11 . . . . 23 . . . Emberiza melanocephala 22 . . . 18 38 . . 9 Motacilla alba 78 . 7 10 27 54 100 10 9 Passer domesticus 100 . . 10 9 31 60 . . Merops apiaster 56 . . . . . 60 40 . Luscinia megarhynchos 100 20 . 100 45 38 80 10 18 Lanius collurio 44 10 . 80 27 46 80 30 73 Certhia familiaris . 30 . . . . . . . Dryocopus martius . 10 . . . . . . .


- 26 -

Columba oenas . 40 29 . 18 . . . . Pyrrhula pyrrhula . 50 14 . . . . . . Ficedula semitorquata . 50 21 . . . . . . Carduelis spinus . 60 29 20 . . . . . Coccothraustes coccothraustes . 70 93 60 9 . . . . Sitta europaea . 100 79 80 18 . . . . Ficedula parva . 100 50 60 9 . 20 . . Hippolais icterina . 50 14 90 . . . . . Dendrocopos minor . . 29 10 . . . . . Garrulus glandarius . 20 21 . 36 8 . . . Periparus ater . 100 93 60 45 . . . . Erithacus rubecula . 100 79 50 45 . . . . Aegithalos caudatus . 70 50 60 45 . . . . Turdus philomelos . 100 93 40 36 . . . . Dendrocopos major . 100 86 90 27 . . . . Phoenicurus phoenicurus . 10 21 10 36 8 80 . . Picus viridis . 70 50 40 36 . 80 . . Prunella modularis . 20 7 . 9 . 20 . . Motacilla cinerea . 20 36 30 9 . 20 . 9 Cinclus cinclus . . 7 10 . . 20 . . Phylloscopus nitidus 11 60 7 40 27 8 40 . . Troglodytes troglodytes . 100 79 40 27 . 20 . 9 Sylvia atricapilla . 100 64 80 36 . 60 . 18 Fringilla coelebs 11 100 93 100 73 15 100 10 9 Turdus merula . 100 93 100 100 38 100 . 45 Turdus viscivorus . 10 7 10 27 8 . . . Jynx torquilla . 10 21 30 9 15 . . . Muscicapa striata 22 60 21 70 . . 60 . . Carduelis chloris 22 70 79 70 36 . . . . Cyanistes caeruleus 33 100 93 60 64 . 100 . . Corvus cornix 100 . 14 70 36 62 80 10 . Streptopelia turtur 33 . 7 10 27 23 . . . Oriolus oriolus 33 . . 80 9 31 40 10 9 Pica pica 89 . . . 18 69 100 20 18 Lanius minor 33 . . . 9 23 80 10 9 Sylvia communis 33 . . . 36 31 80 . 27 Falco tinnunculus 11 . . . 9 . 60 20 55 Upupa epops 67 . 7 10 45 62 100 40 18 Emberiza calandra 89 . . . 64 85 100 30 27 Dendrocopos syriacus 22 . . . 9 . 80 . . Cettia cetti 11 . . . . . 100 . . Falco subbuteo 11 . . . . . 40 . . Acrocephalus palustris 11 10 . . . 8 40 . . Sturnus vulgaris 78 . . . . 54 100 50 . Carduelis cannabina . 10 . . 45 31 . 80 100 Emberiza cia . . . . 27 15 40 70 91 Lullula arborea . . . 30 73 62 60 100 64 Alectoris chukar . . . . . 15 40 100 64 Petronia petronia . . . . 9 8 60 70 91 Oenanthe oenanthe . . 7 . 9 8 . 30 45 Alauda arvensis . . . . 9 38 . 30 45 Anthus campestris . . . . 9 15 . 90 36 Melanocorypha bimaculata . . . . . 8 . 30 9 Eremophila alpestris . . . . . 15 . 90 55 Emberiza hortulana . . . . . 8 . 20 27 Sitta neumayer . . . . 18 . 20 60 82 Aquila chrysaetos . . . . . . . 10 9 Buteo rufinus . . . . . . . 10 9 Oenanthe finschii . . . . . . . 60 36 Oenanthe isabellina . . . . . . . 60 18 Sylvia curruca . . . . 18 . 60 . 55 Carpodacus erythrinus 11 20 7 . 36 . 60 . 18


- 27 -

Columba palumbus . . 7 . 18 . 60 . 9 Apus apus 67 . 7 . . 31 . . 27 Delichon urbicum 33 . . . . 8 . 20 27 Ptyonoprogne rupestris . . . . . . . . 18 Tachymarptis melba . . . . . . . . 27 Monticola saxatilis . . . . . . . 30 100 Monticola solitarius . . . . . . . . 18 Oenanthe hisp. melanoleuca . . . . . . . 10 18 Serinus pusillus . . . . . . . 20 64 Turdus torquatus . . . . . . . 10 45 Bucanetes githagineus . . . . . . . 10 18 Saxicola torquatus . . . . 9 8 . . 18 Phoenicurus ochruros . . . . . . . 10 91 Rhodopechys sanguineus . . . . . . . . 9 Irania gutturalis . . . . . . . . 9 Excluded species: . . . . . . . . . No grouping: . . . . . . . . . Saxicola rubetra . 10 . . . 8 . 10 . Lanius senator 11 . . . . 8 . . 9 Anthus spinoletta . . . . . . . 10 . Ardea cinerea 11 . . . . . . . . Prunella collaris . . . . . . . 10 . Anthus trivialis . . . . 36 23 20 . 9 Corvus corax . . 7 . 9 23 . 20 18 Sylvia nisoria . . . . 9 . . . 9 Excluded owls and raptors: . . . . . . . . . Buteo buteo . 20 29 10 9 23 . . 18 Aquila pomarina . . 7 . . 8 . . . Accipiter badius . . . 10 . . . . . Circaetus gallicus . . . . 9 . . . . Neophron percnopterus . . . . . . . . 9 Accipiter nisus . . . . . 8 . . . Strix aluco . 10 . . . . . . . Otus scops . . . 10 9 . 20 . . Breeding status unclear: . . . . . . . . . Phylloscopus sindianus . . . . 27 . . . . Crex crex . . . . . 8 . . .

Annex 4 provides the detailed results of the indicator species analysis. Significant species

associated to only one group based on the indicator species analysis, are marked with the

significance code within the following description of the nine breeding bird communities.

The landscape type composition is given in the header data of Table 2:

Community 1 (Caspian lowland)

I found this community exclusively in the Caspian lowland. Several species are restricted

to this community like typical steppe species (Hippolais pallida***, Cercotrichas

galactotes**, Merops persicus*, Francolinus francolinus, Sylvia mystacea) or species of

marshes and wetlands (Acrocephalus schoenobaenus***, Acrocephalus scirpaceus,

Acrocephalus arundinaceus***, Circus aeruginosus***, Remiz pendulinus*, Glareola


- 28 -

pratincola, Motacilla flava***, Ixobrychus minutus). Most of them occur along draining

channels. Species that generally prefer open landscapes can also be found here and in

higher altitudes above the forest belt, mainly in the montane meadow belt (Emberiza

melanocephala, Melanocorypha calandra, Emberiza calandra, Coturnix coturnix) or

riparian forests (Dendrocopos syriacus, Cettia cetti, Falco subbuteo, Merops apiaster),

where they belong to community 7 and 6 respectively.

As the lowland is widely covered with human settlings, village related birds occur in

high abundances (Passer domesticus***, Hirundo rustica***, Sturnus vulgaris, Apus

apus, Pica pica, Motacilla alba).

Community 2 (Natural forest)

Community 2 occurs along transects conducted in natural forest or slightly disturbed

forests (both together: 87%). Certhia familiaris** and Dryocopus martius appear

exclusively in this community. The social-ecological group containing Ficedula

semitorquata**, Pyrrhula pyrrhula*** and Columba oenas are restricted to community 2

and 3, but only the first two species indicate community 2, whereas Columba oenas is an

indicator species for the combination of community 2 and 3 (Annex 4).

Typical forest species like Periparus ater, Sitta europaea, Dendrocopos major,

Erithacus rubecula or Troglodytes troglodytes reach their highest abundances in this

community (Annex 3), but occur also elsewhere within the forest belt in other breeding

bird communities. Species of open woodland are scarce (Luscinia megarhynchos, Lanius

collurio, Lullula arborea).

Community 3 (Intermediate disturbed forest)

Due to the cluster analysis community 3 consists mainly of transects conducted in

intermediate disturbed forests (56%) according to the classification of Scharnweber et al.

(2007). It also presents bird species that are typical for forest habitats. The species

composition of community 3 is similar to community 2 but open woodland species occur

rarely (Streptopelia turtur, Corvis cornix). Lullula arborea is missing.


- 29 -

Community 4 (Strongly disturbed forest)

The proportion of transects made in park-like forests is relatively high (35%). Never-

theless, also transects of intermediate disturbed forests (38%) and shrubby woodland

(27%) had been classified due to the cluster analysis to community 4.

Community 4 contains typical forest species (Turdus merula, Fringilla coelebs, Sitta

europaea, Periparus ater), but with lower relative abundances. In comparison with

community 2 and 3, species of open habitats are more abundant (Lullula arborea, Corvus

cornix, Streptopelia turtur, Oriolus oriolus, Luscinia megarhynchos, Lanius collurio).

No indicator species are within community 3 and 4.

Community 5 (Shrubby woodland)

The species composition and the species abundances of this community indicate this

community as shrubby woodland. 54% of shrubby woodlands are included in this com-

munity. In comparison with community 2, 3 and 4, several forest species are missing in

community 5 (Pyrrhula pyrrhula, Ficedula semitorquata, Carduelis spinus) or are less

abundant (e.g. Sitta europaea, Ficedula parva, Dendrocopos major, Dendrocopos minor,

Coccothraustes coccothraustes, Turdus philomelos, Fringilla coelebs, Periparus ater, Eri-

thacus rubecula). Open woodland species (Corvus cornix, Streptopelia turtur) or

generally open landscapes preferring species (Emberiza melanocephala, Emberiza calan-

dra, Emberiza cia, Carduelis cannabina, Sylvia communis) are present and can be

common.

This degradation stage was mostly close to human settlements (Noack 2007,

Scharnweber et al. 2007). Therefore, bird species that are also related to human

settlements were present in small numbers (Upupa epops, Motacilla alba, Pica pica).

The only indicator species of this community is Phylloscopus sindianus**, but it is un-

clear whether this species is a breeding bird in that region (Patrikeev 2004).

Community 6 (Montane meadows)

I found this community mostly within the montane meadow belt (82%). It is closely re-

lated to community 1 (Figure 20). A few species are only distributed in these two com-


- 30 -

munities such as Coracias garrulus, Coturnix coturnix* and Melanocorypha calandra.

Community 6 lacks of typical forest species (Periparus ater, Ficedula parva, Sitta

europaea) but includes species of open woodland (Corvus cornix, Streptopelia turtur,

Lullula arborea). Emberiza melanocephala**, Lanius minor, Emberiza calandra,

Carduelis cannabina prefer shrubs in an open landscape within their breeding territories

and therefore also occur in community 6. As the montane meadow belt is dominated by

agriculture, which includes many villages, also species assigned to human settlements can

be found here (Hirundo rustica, Passer domesticus, Apus apus, Delichon urbicum).

Community 7 (Riparian forest)

The breeding bird community 7 reflects the birds of the riparian forest in 1200 to 1800 m

a.s.l. (Table 2). Typical species of the forest-interior do not occur in the riparian forests

(Sitta europaea, Ficedula parva, Dendrocopos major, Periparus ater, Erithacus

rubecula), except some widespread forest species (Fringilla coelebs, Turdus merula,

Cyanistes caeruleus). As the riparian forest is situated within the montane semi-desert,

some species, like Alectoris chukar and Petronia petronia, extend their breeding range

into this community, but playing a minor role. A socio-ecological group that contains

Dendrocopos syriacus***, Cettia cetti***, Falco subbuteo** and Acrocephalus

palustris**, is restricted to community 7 and 1, but their species indicate the riparian

forest community.

Furthermore, species of open landscapes (Lanius minor**, Sylvia communis***, Em-

beriza calandra, Falco tinnunculus) or human settlements (Passer domesticus, Upupa

epops) breed here. Sturnus vulgaris reaches here highest breeding densities. Further in-

dicator species are Phoenicurus phoenicurus*** and Columba palumbus**.

Community 8 (Montane semi-desert)

The breeding bird communities of community 8 and 9 are different to community 1-7.

The species composition changed almost completely and several species occur only in

these two communities. Community 8 includes typical montane semi-desert species

(Melanocorypha bimaculata, Oenanthe isabellina***, Anthus campestris***), but also


- 31 -

species that breed in lower altitudes of community 6 (montane meadow belt), like

Oenanthe oenanthe or Alauda arvensis, or in rocky habitats of community 9 (Oenanthe

finschii, Eremophila alpestris, Emberiza hortulana). Along rocky outcrops within the

montane semi-desert also typical rock associated species occur, but with generally low

abundances like Sitta neumayer, Monticola saxatilis and Oenanthe finschii.

Community 9 (Rocky habitats)

Rocky outcrops which I found at 1500 – 2300 m a.s.l feature community 9. It contains

rock-dependent species, which are exclusively related to community 9 (Ptyonoprogne

rupestris, Tachymarptis melba*, Monticola saxatilis***, Monticola solitarius, Oenanthe

hispanica melanoleuca, Bucanetes githagineus, Phoenicurus ochruros***, Rhodopechys

sanguineus) or exclusively related but depending on shrubs (Serinus pusillus***, Turdus

torquatus**, Saxicola torquatus, Irania gutturalis). I observed species of this community

also elsewhere, but predominantly in community 8 (Alectoris chukar, Petronia petronia,

Emberiza cia, Carduelis cannabina, Lullula arborea, Sitta neumayer, Oenanthe finschii,

Oenanthe oenanthe**). The only forest species occurring here are Turdus merula or rarely

Troglodytes troglodytes, Sylvia atricapilla and Parus major. This is the only community

where I found Apus apus and Delichon urbicum besides human settlings.

Indicator species for a combination of forest communities

Additionally, the analysis of indicator species identified species occurring significantly in

two or more communities. Especially, indicator species of the forest belt rarely appear

significantly in only one community (Annex 4).

For community 2 and 3, indicating a natural, slightly and intermediate disturbed forest,

significant indicator species are Turdus philomelos***, Troglodytes troglodytes***,

Erithacus rubecula*** and Columba oenas**. Species with a broader amplitude are

Dendrocopos major***, Sitta europaea***, Periparus ater***, Coccothraustes

coccothraustes***, Ficedula parva***, Carduelis chloris*** and Carduelis spinus**.

They are significantly related to all forest degradation stages except the shrubby woodland

stage (community 2+3+4). In contrast, Dendrocopos minor* is an indicator for breeding


- 32 -

bird communities preferring degraded forests (community 3+4). Aegithalos caudatus***

is the only species which is significantly related to all types of the forest belt (community

2+3+4+5), whereas Turdus merula***, Fringilla coelebs***, Sylvia atricapilla*** and

Picus viridis*** are additionally indicator species for the riparian forest (community

2+3+4+5+7).

This deeper insight into the species-community relationships reveals that several

species indicate a large variety of landscape and habitat types. Even a generic species can

be an indicator species like Turdus merula and Fringilla coelebs, as the most widespread

forest species (Table 2, Annex 3).

4.3 Parameters influencing breeding bird communities

The analysis of parameters is important to understand ecological communities. For this

purpose, I carried out a NMDS analysis. This ordination is used to describe relationships

between species composition patterns and the influencing parameters onto a two

dimensional space. Figure 21 illustrates the nine breeding bird communities along two

ordination axis. The altitudinal gradient is the strongest site parameter, which is under-

lined by an approach on species-level (Figure 24). The altitudinal gradient ranges from

bottom left to top right with community 8 and 9 at the end of this vector representing the

highest altitudes. NMDS1 axis follows the vegetation structure of tree and shrub layer.

From left to right a gradient concerning naturalness of forest appearance can be derived.

Open landscape communities are on the right and forested landscapes are on the left. A

gradient following NMDS2 axis reflects the herb layer. Community 1 (Caspian lowland)

contains the best-developed herb layer, followed by community 6 (montane meadows).


- 33 -

Figure 21: Ordination of breeding bird communities using a NMDS generated from abundances of all species of all transects after 20 random starts of a two-dimensional solution (final stress = 16.06, rmse = 0.002, max residual = 0.0072). Highly significant site parameters with a p-value < 0.001 are plotted as red vectors and other are black (for details, see Annex 5). Lengths of vectors indicate the significance of each parameter. Transect numbers are given in symbols. Parameter abbreviations: ALTI = altitude, DATE = date, EXPO = slope exposition (after Parker 1982), SLO = slope steepness (after Parker 1982), HH = Herb layer height, HC = Herb layer cover, SH = Shrub layer height, SC = Shrub layer cover, LTH = Lower tree layer height, LTC = Lower tree layer cover, UTH = Upper tree layer height, UTC = Upper tree layer cover. Landscape abbreviations: Casp = Caspian lowland, Nat = Natural forest, Int = Intermediate disturbed forest, Stro = Strongly disturbed forest, Shr = Shrubby woodland, Mead = Montane meadows, Rip = Riparian forest, Semi = Montane semi-desert, Rock = Rocky habitats.


- 34 -

Figure 22: Ordination of breeding bird communities within the forest degradation stages using a NMDS generated from abundances of all species of transects of community 2, 3, 4 and 5 after 100 random starts of a two-dimensional solution (final stress = 15.83, rmse = 0.00015, max residuals = 0.00054 after 81 tries). Highly significant site parameters with a p-value < 0.001 are plotted as red vectors and other are black (for details, see Annex 6). Lengths of vectors indicate the significance of each parameter. Transect numbers are given in symbols. Parameter abbreviations: ALTI = altitude, DATE = date, EXPO = slope exposition (after Parker 1982), SLO = slope steepness (after Parker 1982), HH = Herb layer height, HC = Herb layer cover, SH = Shrub layer height, SC = Shrub layer cover, LTH = Lower tree layer height, LTC = Lower tree layer cover, UTH = Upper tree layer height, UTC = Upper tree layer cover. Landscape abbreviations: Nat = Natural forest, Int = Intermediate disturbed forest, Stro = Strongly disturbed forest, Shr = Shrubby woodland.


- 35 -

To get a deeper insight into the forest degradation stages and their explanatory vari-

ables, only transects of community 2, 3, 4 and 5 have been calculated again in a further

NMDS (Figure 22). These transects were conducted within the forest belt (Figure 4). I

excluded transect 5 and 6 as they belong to other landscape types, whereas transect 73 was

treated as belonging to community 5, as it was securely made in the shrubby woodland

stage and only due to the cluster analysis considered as belonging to community 6.

NMDS1 axis reveals a degradation gradient (Figure 22). Vectors of the tree layer are

pointing towards transects performed in natural and slightly disturbed forests and vectors

of the shrub layer into the opposite direction towards transects of the shrubby woodland

communities. The altitudinal gradient also plays an important role, but is less significant

compared to the landscape-scaled approach in Figure 21.

Each bird species and therefore each bird community has a different ecological de-

mand. The differences of each community according to their preferred vegetation struc-

tures are given in Figure 23 and the header data of Table 2. Communities 2-5 and 7 re-

present forested landscapes. The height and cover of the upper tree layer decreases conti-

nuously from community 2 to 5, whereas the cover of the shrub layer increases. This gra-

dient also appears in Figure 22. The breeding bird communities clearly show a response to

the vegetation parameters resulting in different values. Interestingly, these results are

similar to Scharnweber et al. (2007), which base on vegetation analysis. For comparison,

communities besides the forest degradation stages are also included. Community 1, 6, 8

and 9 are different. They present open landscapes with a missing or reduced tree layer.

4.4 Relative abundances of bird species

Due to the heterogeneity of several transects according to their landscape composition,

especially within the forest belt, where transects consisted often of several degradation

stages, the calculation of relative abundances could not be performed basing on bird


- 36 -

Figure 23: Boxplots of height and cover of each vegetation layer for every breeding bird community.


- 37 -

Figure 24: Boxplots of the altitudinal distribution of breeding bird species basing on the num

ber of territories (n≥5).


- 38 -

Figure 25: Relative abundances values of the 15 most common bird species per landscape type (forest degradation stages).


- 39 -

Figure 26: Relative abundances values of the 15 most common bird species per landscape type (outside the forest belt).


- 40 -

communities. Thus the relative abundance values base on the landscape type deter-

mination (chapter 2.3). This was in field more precisely and therefore used for further

calculations, especially to compare the species response in different forest degradation

stages.

Figure 25 shows the relative abundances of the most common bird species within the

forest degradation stages (forest belt). The most abundant species in the lesser disturbed

forests are Periparus ater and Fringilla coelebs. However, a comparison between bird

species is theoretically not allowed, because of the different detectability among species.

This comparison is also biased towards eye-catching and loud-voiced birds.

A shift regarding species composition and relative abundance takes place in the park-

like forest stage (Figure 25). The abundance of typical forest species decline with ongoing

forest degradation (e.g. Periparus ater, Fringilla coelebs, Troglodytes troglodytes, Eri-

thacus rubecula), whereas the relative abundance of open woodland species increases.

Most increasing species are related to shrubs of open landscapes (Luscinia megarhynchos,

Emberiza melanocephala, Sylvia communis, Lanius collurio).

4.5 Response to forest degradation on species-level

I found 32 breeding bird species within the natural forest stage. The obtained species

number is the result of a total transect of 8.6 km (Table 1). 28 species are regarded as

‘forest species’. I then converted the relative abundance values of each species per forest

degradation stage into percent values to compare the impact of human activities within the

forest belt. The natural forest stage serves as a gauge that gives information about a

pristine forest without any human activities. I used the converted abundance values of the

natural forest stage as reference values constituting 100%.

Negatively affected or strongly negatively affected

Figure 27 shows a clear negative response of forest species to forest degradation. Most of

these species prefer the forest interior. A decline of about 50% according to an unaffected

population is visible in the intermediate disturbed forest stage. Some species, like


- 41 -

Fringilla coelebs, are also common in the park-like forest stage, but show a rapid decline

in the shrubby woodland stage. Only 6% or less of the bird population of strongly affected

species remain in the shrubby woodland stage. 65% of the 28 forest species show a neg-

ative or strongly negative response to forest degradation.

Figure 27: Negatively and strongly negatively response of selected forest bird species to forest degradation. Values are scaled relative to the natural forest stage (reference stage) constituting 100%.


- 42 -

Positively affected

Generally, open landscapes and shrub formations preferring species are positively affected

by forest degradation, e.g. Luscinia megarhynchos. Only a few species, treated as forest

species or at least ‘edge species’, show a positive trend (Figure 28). These species have

low abundances in natural forests and are common in degraded forests resulting in a vast

increase of a few species, e.g. Carduelis chloris. The treatment of some of those species

as forest species might be discussible. Only 21% of the 28 forest species show a positive

response to forest degradation. Most of them show a peak in the park-like forest stage, but

decrease in the shrubby woodland stage (Figure 28).

Figure 28: Positive response of selected forest bird species to forest degradation. Values are scaled relative to the natural forest stage (reference stage) constituting 100%.

4.6 Response to forest degradation on guild-level

The guild concept is needed because it can help to identify the habitat characteristics that

determine the structure of bird communities and because it would be possible to under-

stand processes organising communities (Casenave et al. 2008).

In total, the natural and slightly disturbed forest stages reach their highest relative

abundance values within the forest belt. The shrubby woodland stage holds the lowest

values (Figure 29).


- 43 -

The dominating guild within all degradation stages are cavity breeders. This guild in-

cludes typical cavity breeders, like Dendrocopos major or Cyanistes caeruleus, and semi-

cavity breeders, like Certhia familiaris. Cavity breeders reach their highest relative abun-

dances in natural and slightly disturbed forests with 25.2 and 24.8 territories per km,

respectively. In contrast, only 6.5 territories per km of cavity breeders occur in the

shrubby woodland stage. Their proportion of about 40% is in all degradation stages more

or less equal, besides the shrubby woodland stage. Here, only 17% belong to the cavity-

breeding guild owing to the lack of cavities in this stage.

Canopy breeders are common in all degradation stages with 12.0 to 16.2 territories per

km, except for the shrubby woodland stage with only 7.1 territories per km. The highest

percentages can be found in intermediate disturbed forests and park with 32% each.

Shrub breeders dominate the shrubby woodland stage. Here, the highest proportion

(37%) and relative abundance (14.2 territories per km) can be found.

Birds that breed on the ground show a decline in relative abundance and proportion

concerning a degradation gradient from high values in natural forest stage and low values

in park-like stage. On the contrary, the values of relative abundance and proportion rise

again in the shrubby woodland stage.

Figure 29: Response of nesting guilds to forest degradation. (Nat = Natural forest stage, Sli = Slightly disturbed forest stage, Int = Intermediate disturbed forest stage, Park = Park-like forest stage, Shr = Shrubby woodland stage).

The breeding bird communities of the Talish mountains (Azerbaijan) and their response to forest degradation Discussion

- 44 -

5 Discussion

5.1 Breeding bird communities

The cluster analysis identified 9 breeding bird communities (Figure 20). Every community

is representing more or less one of the landscape types. Especially transects conducted

outside the forest belt show an apparent overlapping (header data of Table 2). Within the

forest belt, a differentiation of forest degradation stages by breeding bird communities is

imprecise owing to the problems of the survey method. For the cluster and NMDS

analysis each transect was treated as a single record, despite the differing transect length

and differing landscape type compositions. In a few cases, transects consisted of a mixture

of several degradation stages. For example, a 2000 m long transect contains 1500 m park-

like forest and 500 m shrubby woodland stage. The cluster algorithm may classifies this

transect to a community representing the park-like stage including 500 m of shrubby

woodland. A wrong determination of the landscape types in field could also cause blurred

results concerning the landscape composition of each community. Within the forest de-

gradation stages, it was sometimes difficult to separate the stages from each other. For

example, the natural forest stage was difficult to distinguish from the slightly disturbed

forest stage, as both are very similar. A further important point, regarding the landscape

composition of each breeding bird community, is the early date of surveying of a few

transects. The species composition of early transects (beginning of April) lacks most long-

distance migrants. These species contribute also to the classification of transects to a

community by the cluster analysis and if they are missing, the cluster analysis classifies

the transects only by the resident species. This problem could not be adequately solved by

beginning the survey when long-distance migrants completely arrived the Talish moun-

tains (end of May/beginning of June). Several residents already fed chicks and show a

reduced territorial behaviour, which leads to the opposite effect of ‘missing’ residents,

when the territorial activity of migrants peaks. The only way to solve this problem would

be a repeated surveying of each transect, but this was refused (chapter 2.6.). However, on

this landscape-scaled approach, open habitats fit to the open land breeding bird com-

munities. Within the forest belt with its degradation stages, it can only roughly be as-

Discussion The breeding bird communities of the Talish mountains (Azerbaijan) and their response to forest degradation

- 45 -

signed (header data of Table 2), but a seperation following a man-made degradation

gradient is clearly visible (Figure 21, 22).

Each community differs according to their species composition and relative abundance

values (Table 2, Annex 3). Interestingly, on this landscape-scaled approach, which se-

parated for example Caspian lowland from rocky habitats by breeding birds only, a dif-

ferentiation between forest communities (community 2, 3, 4 and 5) is evident. This is

remarkable as the clusters of the cluster analysis were cut at the same level, representing

an equal degree of similarity (Figure 20). However, the differences in bird species

composition between communities of the forest belt are low (Table 2, Annex 3). Within

the forest communities, indicator species occurred only in community 2 (natural forest):

Pyrrhula pyrrhula***, Certhia familiaris**, Ficedula semitorquata** (Annex 4). These

few indicator species are in other forest communities missing, which underlines the strong

similarity between these communities. As the cluster analysis separates clusters by species

and their abundances, hence the forest communities are mainly differentiated by the

relative abundance values. The differentiation by species inventory plays a minor role in

the forest communities. Several forest bird species, like Periparus ater, Troglodytes

troglodytes, Fringilla coelebs or Sitta europaea, reach their highest relative abundances in

community 2 representing natural forests (Table 2, Annex 3). In contrast, Tomiałojć &

Wesołowski (2004) found remarkably low abundance values in the primeval forest of

Białowieża (Poland) compared to data from man-transformed places (e.g. fragmented

forests of Western Europe). Species of community 2 are also present in communities

representing strong forest degradation, for example community 5 containing 54% of the

shrubby woodland stage. These differences of relative abundance values are responsible

for the separation of bird communities by the cluster analysis within the forest belt

(degradation stages), despite the low differences in species composition.

However, to give community 2, 3, 4 and 5 the status of a breeding bird community

might be discussable due to the similar species inventory. The separation of forest

breeding bird communities by species only is difficult or almost impossible. Here, the

main problem occurs for an practical use of this from the vegetation ecology derived

concept. Other problems are the migratory life history traits of birds. To evaluate the


- 46 -

quality of a landscape or a habitat by the development of a breeding bird community is

under the circumstance of a not sessile living bird inaccurate. The reason for the presence

or absence of a species might not be a loss of habitat quality on the breeding site, it can

also be hunting pressure on migration routes or quality loss on the wintering sites. Plants

are therefore more practical for the evaluation of habitat quality as they spend their whole

life on one location. Furthermore, the determination by their abundance values is not

practical in field, as it depends on a large amount of surveying time, which avoids a quick

identification of each forest community. Nevertheless, human activity created different

breeding bird communities within the forest belt, which are statistically verified in a

landscape-scaled approach.

5.2 Parameters influencing bird communities

The vegetation structure is the key parameter influencing the composition of breeding bird

communities in the Talish mountains (Figure 21-23). Heterogeneous vertical and

horizontal vegetation structures are an important factor influencing breeding bird

communities (abundance and diversity), because it offers a large number of ecological

niches for a broad variety of bird species. (Mac Arthur & Mac Arthur 1961, Mac Arthur

1968, Erdelen 1984, Tews et al. 2004, Kati & Sekercioglu 2006). Plant species diversity

of (temperate) forests is less important and has nothing to do with bird species diversity

(Mac Arthur 1961). Thus, I expect the high diversity of woody plant species in the Talish

mountains has no effect on the species diversity of bird communities, although it was not

tested in this study. For the formation of each breeding bird community is the develop-

ment of the tree layers, as the main characteristic of the vertical vegetation structure with

different tree heights and cover values, most important (header data of Table 2, Figure

23). In Figure 21 and Annex 5 are information given about the importance of each

vegetation parameter. Every vegetation parameter is highly significant in this large-scaled

approach. At a smaller-scaled approach, regarding forest communities (2, 3, 4 and 5) only,

height and cover of the upper tree layer, height of the lower tree layer and cover of the

shrub layer are among all vegetation parameters these with the highest


- 47 -

significance (p-value < 0.0001) (Figure 22, Annex 6). High values of tree height and tree

cover indicate the occurrence of forests, which is obviously crucial to the bird species

composition of each community. If trees, as main contributor to the vertical vegetation

structure, are missing, completely different bird communities are developed. The resulting

open land communities consist mainly of open land species like Alauda arvensis, Hirundo

rustica or Emberiza melanocephala. This separation from open land and forest com-

munities is in agreement with other studies (e.g. Kati & Sekercioglu 2006).

A removal of forests would cause a species turnover by extirpating of most forest bird

species and colonisation of open land species (Figure 21, 30). This is also observable in

tropical rain forests after intensive logging or clear-cutting (e.g. Thiollay 1999, Patten et

al. 2010). A possible recolonisation of forest species could never bring back the whole

assemblage due to the extinction of some bird species. The British avifauna for example,

lacks of several forest bird species, for which no geographical or biological reasons exist,

probably caused by a widespread removal of forests centauries ago (Fuller et al. 2007a,

Figure 30: Breeding bird response to forest degradation. Summarised schematic of the forest degradation stages.


- 48 -

Wesołowski 2007). Nevertheless, a complete removal of the Caspian forest in Azerbaijan

is unlikely, but the current forest degradation changes the vertical vegetation structure,

which affects most forest birds negatively (Figure 27).

The differences in the vertical vegetation structure lead to the differing appearance of

breeding bird communities, which roughly represent each degradation stage. As already

mentioned, each forest community is more differentiated by the species abundance then

the species composition. High trees and high foliage cover are typical for undisturbed

primeval forests (Wesolowski 2007). Trees provide cavities for hole nesters and their

crowns offer canopy breeders nesting opportunities. The older and larger the trees are the

more cavities are available and the higher is the abundance of cavity breeders in the

natural forest stage compared to other degradation stages (Enoksson et al. 1995, Poulsen

2002) (Figure 29). The high proportion of cavity breeders is typical for European old-

growth forests (e.g. Wesołowski & Tomiałojć 1997, Saniga & Saniga 2004, Korṅan

2009). Old and large trees have also voluminous crowns and can therefore harbour more

canopy breeders (Figure 29). In addition, also ground breeders achieve high values in

natural forests. A good developed, diverse and undisturbed herb and shrub layer enables

several species to breed on the ground. In contrast, within the park-like forest stage is the

herb layer short due to grazing of cattle and sheep. Here, breeding possibilities for ground

breeders are low and disturbances are frequent. Therefore, abundances of ground breeders

are lower in the park-like forest stage (Figure 29).

Compared to the natural forest stage with high and different aged trees and a multi-

storey structure, the vertical vegetation structure of each forest degradation stage is less

diverse and complex. Furthermore, natural forests are structurally enriched by the ubi-

quitous presents of freshly fallen trees with their rootwads, fallen decaying logs, snags,

various tree holes, temporary pools etc. (Tomiałojć & Wesołowski 2004). The diverse and

complex structures in natural forests form a unique system of ecological niches that can

support high species diversity, as it reduces the intraspecific and interspecific competition

for resources, like nesting sites or food (Tomiałojć & Wesołowski 2004). There is no

evidence for an interspecific competition caused by a shortage of nesting holes and food

limitations in primeval forests (Wesołowski & Tomiałojć 1997, Wesołowski 2003,


- 49 -

Wesołowski 2007). I regarded this assumption as the main reason why abundances of

forest birds reach their highest values in natural or slightly disturbed forest (community

2). The only nesting guild, which is positively affected by forest degradation, are shrub

breeders (Figure 29).

I expect a shortage of food and nesting sites in the degraded forests due to the observed

lower relative abundances of forest species. Furthermore, edge effects of fragmented

forests increase the risk of predation resulting in lower abundances within degraded

forests. Forest fragmentation is the replacement of large areas of native forest by other

ecosystems leaving isolated forest patches, with deleterious consequences for most of the

native forest biota (Saunders et al. 1991 cited in Murcia 1995). The maintenance of large,

structurally complex patches of native vegetation is particularly important in landscapes

where many species are area-sensitive and confined to native vegetation, and where

locations outside these patches are entirely uninhabitable by many native species (Fischer

et al. 2006).

Human activities are the main driver for the development of the breeding bird

communities within the forest belt of the Talish mountains, because they change the

vertical vegetation structure. A removal of large trees due to logging, a destruction of the

understorey and herb layer caused by fuel wood gathering or silvopastures reduce the

breeding possibilities and success of each forest bird, which generally lead to lower

abundances and a clear decline (Figure 25, 27, 29 and Annex 3).

Additionally, the vertical vegetation structure is also triggered by climatic conditions.

Tree growth is prevented or reduced in regions with a too dry and too cold climate. In the

Talish mountains no trees occurred in the dry montane-semi desert. In higher altitudes

above 1800 m a.s.l. (natural tree line), their growth is reduced due to the rougher climate.

Here, the forest becomes the appearance of natural shrubby woodlands. Furthermore,

fires, windthrows or landslides can locally completely deplete forests resulting in a

changed vertical vegetation structure. Figure 21 reveals that the altitude is an important

parameter influencing the breeding bird communities. This parameter was neglected, as

the vegetation structure depends on the altitude, but birds depend on vegetation structure

and not on the altitude.


- 50 -

5.3 Species and species richness

A high richness and diversity of species is for nature and bird conservation a main

concern (e.g. Ramírez-Albores 2007). As the calculation of diversity indices was improper

due to methodological reasons, only the species number of each landscape type was taken

into account. The most species-rich landscapes are rocky habitats of the Zuvand upland

(Figure 19). This landscapes type consists of a variety of different habitats. Rocks are a

common aspect and enable several rock-dependent bird species to breed here. In addition,

this type is often mixed with shrubs, pastures and montane semi-desert (tragacanthic ve-

getation), which offer further species breeding opportunities. Horizontal heterogeneity and

a diverse landscape are important factors for high species richness (Pino et al. 2000,

Sekercioglu 2002, Kati & Sekercioglu 2006). However, this fact should be interpreted

with caution under nature conservation issues, because within a large extended primeval

forest, horizontal heterogeneity can only be achieved due to fragmentation by grazing and

logging. Unsurprisingly, the most species-rich forest degradation stages are the shrubby

woodland stage and the park-like forest stage (Figure 19, Figure 30). Both are under

intense human utilisation pressure resulting in a diverse horizontal heterogeneity with

shrub formations, pastures, open and dense forests. This horizontal diversity enables

breeding opportunities for a broad variety of open land and forest-dependent bird species

in a relative small area (Table 1, Table 2). This consideration is in agreement with other

studies (e.g. Herold 2005). Ugalde-Lezama (2010) made the ‘theory of intermediate

disturbance’ (Connell 1978) responsible for his observations.

In contrast, less disturbed forests like natural or slightly disturbed forests are relatively

poor in species (Figure 19, Figure 30). Their horizontal structure is more uniform with

only a few natural forest clearings due to windthrows, fires or landslides. Open land spe-

cies are rare or missing, which reduces the total amount of species in these types.

Tomiałojć & Wesołowski (2004) stated that primeval forests are generally rich in

species, which might be the case for the Białowieża primeval forest, but the situation of

the Caspian forest is different. This forest is isolated and surrounded by deserts and tree-

less steppes. No connection to the boreal forest zone exists and therefore many boreal

species, which contribute to the high species number in Białowieża forest, are missing.


- 51 -

Especially some non-migratory species of the genera owls, tits, warbler, woodpeckers and

grouses do not occur. A general high richness in species is therefore not a feature of

primeval forests of the temperate zone, it is a feature of the transition zone of deciduous

broadleaf forest to boreal forest. Nevertheless, despite the lower total species richness of

natural forests, compared to the degradation stages, the diversity of forest-interior species

is higher (Figure 30, Annex 7).

Inspired by the island biogeography theory, Askins et al. (1987) stated that the number

of species and the density of forest birds is similar in forests of different sizes, but the

composition of the bird community shows consistent differences. Interior-edge birds are

equally common in large and small forests, while both the density and species richness of

forest-interior birds are higher in larger forests. In short, not only the worsening of the

habitat quality is responsible for the absence of some forest species, also the frag-

mentation and isolation of forest remnants lead to local extinctions, depending on varying

spatial scales, of forest-interior bird species.

A change in species composition per se can also trigger extinction cascades. The loss

of individual species is particularly likely to trigger extinction cascades when ‘keystone’

or ‘strongly interacting’ species are involved because they exert a disproportionate effect

on ecosystem function relative to their abundance (Chapin et al. 2000, Fischer &

Lindenmayer 2007). This means that each species in a (pristine) forest ecosystem fulfills

its ecological functions. For example, the loss of large predators due to hunting or habitat

loss may lead to increasing numbers of herbivores, which leads to a decreasing tree

recruitment. The absence of single bird species can also cause a reduced seed dispersal of

different tree species (Lindenmayer et al. 2000). Insectivorous bird species control pest

outbreaks, which reduces plant damages (Sekercioglu et al. 20004). Species diversity in-

fluences the resilience and resistance of ecosystems to environmental change (Chapin et

al. 2000).

However, species richness alone should not be used to guide conservation decisions.

Other factors such as endemism, rarity, habitat specialisation, complementary of sites and

biological organisation also need to be taken into account (Walther & Martin 2001). Bird

endemism of the Caspian forest exists mainly on the subspecies-level. Almost all of them


- 52 -

are more or less forest-dependent birds, like Phasianus colchicus talischensis, Den-

drocopos minor quadrifasciatus, Dendrocopos major poelzami, Garrulus glandarius

hyrcanus, Poecile hyrcana hyrcana, Periparus ater gaddi, Sitta europaea rubiginosa, Re-

gulus regulus hyrcanus, Erithacus rubecula hyrcanus (Agaeva 1979, Patrikeev 2004,

Clements 2008). Poecile hyrcana hyrcana was recently splitted from Poecile lugubris and

is probably the only endemic species, but prefers degraded forests (Loskot 1978). Forest-

dependent endemic species or subspecies are directly affected by deforestation. To protect

them is important to preserve the global biodiversity. Diversity at all organisational levels,

ranging from genetic diversity within populations to the diversity of ecosystems in land-

scapes, contributes to global biodiversity (Chapin et al. 2000). The protection of the

Caspian forest is therefore strongly linked with the protection of birds.

5.4 Conclusion and conservation implications

Bird species richness and species abundance patterns of the Caspian forest are triggered

by vertical and horizontal vegetation structures (Figure 30). A high heterogeneity of the

vertical vegetation structure causes a high abundance of forest species. A high horizontal

vegetation structure, caused by forest degradation, reduces the abundance of forest species

and enables the breeding of open land species, which increases the species richness.

Several forest-dependent species can still exist in strongly degraded park-like forests

(Figure 25). The impact of forest degradation to species composition is negligible in

slightly or intermediate disturbed forests. Both are able to accommodate a viable

population of forest-dependent bird species. Therefore, most important for the

conservation of forest species is the prevention of large-scaled natural forests with a

diverse vertical heterogeneity. A sustainable and natural forestry with unused areas, also

outside the Hirkan National Park, would protect the Caspian forest avifauna. Preservation

of a system of small reserves rather than a large reserve of the same total area would

probably result in the loss of area-sensitive species and increased rarity for other species

of forest-interior birds that are present, but relatively uncommon, in small forests (Askins

et al. 1987). Old-growth trees of these natural areas should also be prevented from log-


- 53 -

ging. Mature deciduous trees are of great importance to many resident birds (Enoksson et

al. 1995).

Moreover, primeval forests are ‘windows into the past’ through which one would gain

insights into the ecology of pristine plant and animal communities (Wesołowski 2003).

The data collected here may, besides enhancing our understanding of the fundamental

problems of community organisation, also serve as a gauge for the bird community studies

made in transformed woodland habitats (Wesołowski & Tomiałojć 1997). Nowadays,

when the natural habitats are exposed to a permanent negative impact resulting from hu-

man activities all over the world, the knowledge, gained here, is especially needed for

forest biodiversity conservation, sustainable use and ecological restoration.

Clear-cutting and further fragmentation should be avoided also outside the Hirkan

National Park. In particular, smaller remnant fragments were highly vulnerable to ongoing

disturbances as they were accessible for logging and clearance (Echeverria 2005 cited in

Echeverria et al. 2007). Additionally, extensive infrastructure developments of remote

areas, especially by the construction of roads, could cause further degradation and a fur-

ther loss of native biodiversity. Roads offer easy accessibility and forest degradation is

generally high along roadsides (Edenius 1996, Thiollay 1999). In addition, a touristic use

of the protected areas of the Caspian forest should be carefully managed, as it may ac-

company with ecological degradation (Liu et al. 2001, Krüger 2005). Inaccessibility is the

best protection for primeval forests (Joppa et al. 2008).

To preserve such large-scaled forested areas is crucial to the whole flora and fauna.

Birds, as a generally well studied taxonomic group (Flade 1994), may stand for the broad

variety of further taxonomic groups like mammals, amphibians, reptiles or insects. Their

reaction is completely unknown in that region, but their response can be derived from the

observed response of birds. Paralleling the bird results, I expect further local extinctions

of single species among these taxonomic groups accompanied by a loss of biodiversity

with ongoing degradation and fragmentation. As long as the use of the forest is sustain-

able and in balance with ecological processes, an extinction of species and a loss of bio-

diversity is unlikely.


- 54 -

5.5 Scope and limitations

Some limitation of the present work should be considered. The sampling design was made

for a landscape-scaled approach. Other methods, like point count method, might be more

suitable for a smaller-scaled approach dealing with forest degradation stages only. Here, a

detailed record of the vegetation with more parameters, especially concerning vegetation

structure, could deliver better results.

Most of the surveyed bird species were small passerines, which generally have smaller

territories. They are probably not much affected by forest degradation, as their demand for

large-scaled habitat structures is less complex. This study could not implement any

suggestions about rare or nocturnal species. Most of them a large sized (e.g. Ciconia

nigra, Aquila pomarina, Circaetus gallicus) with a need for large territories within the

forest belt. These species are not adequately censused in this study and a further scientific

research should also focus on them, for example to proof whether existing protected areas

are sufficiently large to maintain viable populations.

Monitoring species, which indicate natural or slightly disturbed forests such as

Pyrrhula pyrrhula, Ficedula semitorquata and Certhia familiaris, or groups of birds e.g.

woodpeckers (Drever & Martin 2010) is an efficient method to monitor the ecological

state of forest communities and it gives a more direct insight into bird habitat quality. The

paucity of consistent and complete knowledge about species biology or ecological

processes, however, leads to believe that adequately identified indicator species can be

useful for management, conservation, and restoration of natural and seminatural

ecosystems, especially for large-scale projects (Bani et al. 2006).

Summary The breeding bird communities of the Talish mountains (Azerbaijan) and their response to forest degradation

- 55 -

6 Summary

This study analyses the breeding bird communities and their response to forest

degradation in the Talish mountains of Azerbaijan. Primeval deciduous broadleaf forests

belong globally to the most threatened ecosystems. Only small, isolated remnants

survived in Europe. The Caspian forest is the largest primeval deciduous broadleaf forest

and situated in northern Iran and the adjacent Azerbaijan. In Azerbaijan, the Hirkan

National Park protects 40358 ha. Outside, the forest is exposed to livestock grazing and

logging resulting in forest degradation and fragmentation.

I conducted this study in the breeding season from April to June 2008 using the line

transect method. I then used the obtained data to link breeding bird communities with

environmental parameters. For this purpose I calculated abundance values per transect for

cluster analysis and ordination analysis (NMDS). Furthermore, I compared five forest

degradation stages (natural forest, slightly disturbed forest, intermediate disturbed forest,

park-like forest and shrubby woodland) according to relative abundances of bird species.

From a faunistic aspect, the breeding records of Radde’s Accentor (Prunella ocularis),

Crimson-winged Finch (Rhodopechys sanguinea), White-throated Robin (Irania guttu-

ralis), Trumpeter Finch (Bucanetes githagineus) and Shikra (Accipiter badius) are of

national importance, as they are newly discovered breeding birds or rediscovered after

many decades.

The cluster analysis revealed nine breeding bird communities. They are arranged

mainly along an altitudinal gradient ranging from the Caspian lowland to montane semi-

deserts. Four of the communities are within the forest belt and a result of forest

degradation. 65% of the forest-dependent bird species are negatively affected by forest

degradation. Several species, like Eurasian Bullfinch (Pyrrhula pyrrhula), Semi-collared

Flycatcher (Ficedula semitorquata) and Eurasian Treecreeper (Certhia familiaris), are ex-

pected to become extinct with ongoing degradation. Nine endemic subspecies are

threatened.

The breeding bird communities of the Talish mountains (Azerbaijan) and their response to forest degradation Summary

- 56 -

The differences between natural forests and only slightly disturbed forests are low

regarding species composition and relative abundance. With further degradation, the

relative abundance and the number of forest-dependent species decreases, but the total

species number increases. The strong vertical vegetation heterogeneity of natural forests

with high and old-growth trees, multistorey profiles and a large amount of dead wood

cause a high abundance of forest-dependent species. Natural forests are horizontally

homogenous landscapes resulting in a lower total species number. In contrast, degraded

forests have a reduced vertical heterogeneity, which reduces the abundance of forest

species. The strong horizontal heterogeneity of degraded forests including shrub

formations, different-sized forest fragments and open lands (pastures) enables a broad

variety of bird species breeding opportunities resulting in high species richness.

A slightly use of the Caspian forest does not have a serious impact on the forest

avifauna. Hence, a natural and sustainable forestry, conserving a rich vertical vegetation

structure, would protect the forest avifauna and prevent a loss of global biodiversity.

7 Zusammenfassung

Primäre Laubwälder gehören global zu den am stärksten bedrohten Ökosystemen. In

Europa finden sich nur noch kleine, isolierte Reste primärer Laubwälder. Das weltweit

größte zusammenhänge Primärwaldgebiet, der Kaspische Wald, befindet sich im Norden

des Irans und im angrenzenden Aserbaidschan. In Aserbaidschan werden 40358 ha im

Hirkan National Park geschützt. Außerhalb des Nationalparks führen Viehwirtschaft und

intensive Holznutzung zu einer Degradierung und Fragmentierung des Waldes. Die

vorliegende Arbeit untersucht die Brutvogelgemeinschaften des Talisch Gebirges in

Aserbaidschan. Zudem wurde deren Verhalten unter dem Aspekt der fortschreitenden

Walddegradierung betrachtet.

Im Frühjahr 2008 führte ich die Geländeuntersuchung mittels der Transektmethode

durch. Anhand der gewonnenen Daten wurden die errechneten Brutvogelgemeinschaften

mit Umweltparametern korreliert. Dazu wurde die berechnete Abundanz pro Transekt für

Summary The breeding bird communities of the Talish mountains (Azerbaijan) and their response to forest degradation

- 57 -

die Clusteranalyse und Ordination (NMDS) benutzt. Des Weiteren verglich ich fünf

Degradierungsstadien des Waldes (Naturwald, leicht gestörter Wald, mittel gestörter

Wald, parkartiger Wald and Gebüschwald).

Unter faunistischen Gesichtspunkten sind Brutnachweise bzw. Brutzeitfeststellungen

von Steinbraunelle (Prunella ocularis), Rotflügelgimpel (Rhodopechys sanguinea),

Weißkehlsänger (Irania gutturalis), Wüstengimpel (Bucanetes githagineus) und Schikra

(Accipiter badius) von nationaler Bedeutung, da diese Arten neu entdeckt bzw. nach

vielen Jahrzehnten wiederentdeckt wurden.

Die Clusteranalyse ergab neun Brutvogelgemeinschaften. Diese sind hauptsächlich

entlang eines Höhengradienten angeordnet, welcher vom Kaspischen Tiefland bis in die

Gebirgshalbwüste verläuft. Vier der Gemeinschaften befinden sich im Waldgürtel und

sind das Ergebnis der Walddegradierung. 65% der Waldvogelarten sind von der

Walddegradierung negativ betroffen. Mehrere Arten, wie z.B. Gimpel (Pyrrhula

pyrrhula), Halbringschnäpper (Ficedula semitorquata) und Waldbaumläufer (Certhia

familiaris), werden mit fortschreitender Degradierung im Talisch Gebirge aussterben.

Neun endemische Unterarten sind bedroht.

Die Unterschiede zwischen Natur(Primär-)wäldern und nur leicht gestörten Wäldern

sind gering bezogen auf das Arteninventar und deren relativen Abundanz. Mit

zunehmender Degradierung verringert sich die Abundanz und Artenzahl von

Waldvogelarten bei einer Zunahme der Gesamtartenzahl. Die starke Heterogenität der

vertikalen Bestandsstruktur von Naturwäldern, mit hohen und alten Bäumen, vielen

Vegetationsschichten und einem hohen Anteil an Totholz, verursachen eine hohe

Abundanz von Waldvögeln. Aber ihre räumliche (horizontale) Homogenität verursacht

eine vergleichsweise geringe Gesamtartenzahl. Im Gegensatz zu Naturwäldern haben

degradierte Wälder eine verringerte Heterogenität der vertikalen Bestandsstruktur, die mit

einer Reduzierung der Abundanz von Waldvögeln einhergeht. Die starke räumliche

Heterogenität von degradierten Wäldern (z.B. durch Gebüsche, lichte und geschlossene

Waldfragmente, Viehweiden), ermöglicht einer großen Anzahl von Vogelarten

Brutmöglichkeiten, die eine hohe Gesamtartenzahl bewirken.

The breeding bird communities of the Talish mountains (Azerbaijan) and their response to forest degradation Summary

- 58 -

Eine geringfügige Nutzung der Kaspischen Wälder hat keinen schwerwiegenden

Einfluss auf die Waldavifauna. Daher würde eine naturnahe und nachhaltige

Forstwirtschaft, welche eine abwechslungsreiche vertikale Bestandsstruktur über weite

Flächen bewahrt, dem Schutz der Waldavifauna dienen und damit eine globale

Verarmung an Biodiversität verhindern.

8 Acknowledgements

I am grateful to Prof. Dr. Michael Succow and Dr. Martin Flade for the supervision of my

diploma thesis. The Ministry of Ecology and Natural Resources of Azerbaijan Republic

kindly granted the necessary access to the Hirkan National Park. I thank Kai Gauger for

the indispensable support during the fieldwork. I also wish to acknowledge Jan Peper for

useful critics and the brilliant introduction to multivariate statistics. Furthermore, I thank

Nigar Agaeva, Jonathan Etzold, Benjamin Herold, André Jankowski, Dr. Vladimir M.

Loskot, Prof. Dr. Michael Manthey, Dr. Hartmut Müller, Jan Peters, Tobias Scharnweber

and Sebastian Schmidt for providing literature and useful comments. The people of the

Talish mountains, especially Novrus Hüseynov and his family, Akif Aliyev and

Babakhan, also contribute with their support and overwhelming hospitality to this work.

This study could not be accomplished without the financial support of the DAAD

(German Academic Exchange Service) and the help of the Michael Succow Foundation.

References The breeding bird communities of the Talish mountains (Azerbaijan) and their response to forest degradation

- 59 -

9 References

Abdollahpour, M., & Assadi Atui, A. R. (2005). Entwicklung der geregelten

Forstwirtschaft im Iran. In: Nosrati, K., Marvie Mohadjer, R., Bode, W. and Knapp, H.D., Schutz der Biologischen Vielfalt und integriertes Management der Kaspischen Wälder (Nordiran). Bundesamt für Naturschutz. Bonn, 99-120.

Agaeva, T. A. (1979). [Avifauna of the Talish mountains]. Dissertation. Azerb. Univ.

Baku. Agaeva, T. A. (1972). [Short notes on some birds of Talish Mountains] (report 1). Proc.

Azerb. Univ., boil. ser., no. 2, 54-57. Agaeva, T. A. & Mustafaev, G. T.(1973). [Short notes on some birds of Talish

Mountains] (report 2). Proc. Azerb. Univ., boil. ser., no. 3, 51-54. Alström, P. & Mild, K. (2003). Pipits & Wagtails of Europe, Asia and North America.

Christopher Helm. Andretzke, H., Schikore, T. & Schröder, K. (2005). Artsteckbriefe. In: Südbeck, P.,

Andretzke, H., Fischer, S., Gedeon, K., Schikore, T. Schröder, K. & Sudfeldt, C. (eds.): Methodenstandards zur Erfassung der Brutvögel Deutschlands. Radolfzell. 135-695.

Askins, R. A., Philbrick, M. J. & Sugeno, D. S. (1987). Relationship Between the

Regional Abundance of forest and the Composition of Forest Bird Communities. Biological Conservation 39: 129-152.

Atamov, V. V., Cabbarov, M. & Gurbanov, E. (2006). The Pytosociological

Characteristics of Ecosystems of Mountain of Talysh Region of Azerbaijan. Asian Journal of Plant Sciences 5 (5): 889-904.

Bani, L., Massimino, D., Bottoni, L. & Massa, R. (2006). A Multiscale Method for

Selecting Indicator Species and Priority Conservation Areas: a Case Study for Broadleaved Forests in Lombardy, Italy. Conservation Biology, vol. 20, no. 2: 512-526.

Bibby, C. J., N. D. Burgess & D. A. Hill (1995): Methoden der Feldornithologie.

Radebeul, 270 S. Bray, J. R. & Curtis, J. T. (1957). An Ordination of the Upland Forest Communities of

Southern Wisconsin. Ecological Monographs, Vol. 27, No. 4, 326-349.

The breeding bird communities of the Talish mountains (Azerbaijan) and their response to forest degradation References

- 60 -

Casenave, J. L., Cueto, V. R. & Marone, L. (2008). Seasonal dynamics of guild structure in a bird assemblage of the central Monte desert. Basic and Applied Ecology 9: 78-90.

Chapin, F. S., Zavaleta, E. S., Eviner, V. T., Naylor, R. L., Vitousek, P. M., Reynolds, H.

L., Hooper, D. U., Lavorel, S., Sala, O. E., Hobbie, S. E., Mack, M. C. & Diaz. S. (2000). Consequences of changing biodiversity. Nature, vol. 405: 234-242.

Clements, J. F. (2008). Birds of the World: A Checklist – 6 th Edition. Ibis Publishing,

Vista, CA. Connell, J. H. (1978). Diversity in Tropical Rain Forests and Coral Reefs. Science, New

Series, vol. 199, no. 4335: 1302-1310. Drever, M. C. & Martin, K. (2010). Response of woodpeckers to changes in forest health

and harvest: Implications for conservation of avian biodiversity. Forest Ecology and Management 259: 958-966.

Dufrene, M. and Legendre, P. (1997). Species assemblages and indicator species: the need

for a flexible asymmetrical approach. Ecol. Monogr. 67: 345-366. Echeverria, C., Newton, A. C., Lara, A., Benayas, J. M. R. & Coomes, D. A. (2007).

Impacts of forest fragemtation on species composition and forest structure in the temperate landscape of southern Chile. Global Ecol. and Biogeogr. 16: 426-439.

Edenius, L. & Elmberg, J. (1996). Landscape level effects of modern forestry on bird

communities in North Swedish boreal forests. Landscape Ecology. vol. 11 no. 6: 325-338.

Enoksson, B., Angelstam, P. & Larsson, K. (1995). Deciduous forest and resident birds:

the problem of fragmentation within a coniferous forest landscape. Landscape Ecology vol. 10 no. 5: 267-275.

Erdelen, M. (1984). Bird communities and vegetation structure: 1. Correlations and

comparisons of simple and diversity indices. Oecologia 61: 277-284. Fischer, J. Lindenmayer, D. B. & Manning, A. D. (2006). Biodiversity, ecosystem

function, and resilience: ten guiding principles for commodity production landscapes. Front Ecol. Environ. 4 (2): 80-86.

Fischer, J. & Lindenmayer, D. B. (2007). Landscape modification and habitat

fragmentation: a synthesis. Global Ecol. Biogeogr. 16: 265-280. Flade, M. (1994). Die Brutvogelgemeinschaften Mittel- und Norddeutschlands. IHW-

Verlag, Eching.


- 61 -

Fuller, R. J., Gaston, K. J. & Quine, C. P. (2007a). Living on the edge: Brithish and Irish woodland birds in a European context. Ibis 149 (Suppl.2): 53-63.

Fuller, R. J., Smith, K. W., Grice, P. V., Currie, F. A. & Quine, C. P. (2007b). Habitat

change and woodland birds in Britain: implications for management and future research. Ibis 149 (Suppl. 2): 261-268.

Glutz von Blotzheim, U. & Bauer, K. (eds.) (1991): Handbuch der Vögel Mitteleuropas,

vol. 12, Aula Verlag, Wiesbaden. Glutz von Blotzheim, U. & Bauer, K. (eds.) (1994): Handbuch der Vögel Mitteleuropas,

vol. 9, Aula Verlag, Wiesbaden. Glutz von Blotzheim, U. & Bauer, K. (eds.) (1998): Handbuch der Vögel Mitteleuropas,

vol. 11, Aula Verlag, Wiesbaden. Google Earth (2010). Satellite Images. http://earth.google.de . Accessed: April 2010. Gregory, G. (2007). Shikra in the Western Palearctic. Birding World 20: 159–161. Grossheim, A. A. (1926). [Flora of the Talish mountains]. Isdanie Narkomzema Az. SSR. Hajiyev, V. (2006). Biodiversity of the flora of the Hirkan Biosphere Park. In: Aliyev, I.,

Engels, B., Hedden-Dunkhorst, B., Heinrichs, A. K. & Okada, C. (Eds.), Protection and Utilisation of Biological Resources for a Sustainable Development, with Special Reference to Azerbaijan - Proceedings of a NATO CCMS supported Seminar. Bundesamt für Naturschutz. Bonn, 16-19.

Heiss, M. & Gauger, K. (2009). The rediscovery of breeding Shikras Accipiter badius in

the Western Palaearctic. Sandgrouse 32: 134–137. Herold, B. (2005). The Avian Communities and their Habitat Associations of the Walnut

Fruit Forests of Southern Kyrgyzstan. Diploma thesis. Ernst-Moritz-Arndt-University. Greifswald.

Hilden, O. & Järvinen, A (1989). Efficiency of the line-transect method in mountain birch

forest. Ann. Zool. Fennici 26: 185-190. Islamic Republic of Iran Meteorological Organization (IRIMO). Climatic Statistics.

http://www.irimo.ir/. Accessed: 15.04.2010. IUCN Red List (2010). IUCN Red List of Threatened Species. http://www.iucnredlist.org/

Accessed: 15.05.2010.


- 62 -

Joppa, L. N., Loarie, S. R. & Pimm, S. L. (2008). On the protection of “protected areas”. PNAS, vol. 105, no. 18: 6673-6678.

Kati, V. I. & Sekercioglu, C. H. (2006). Diversity, ecological structure, and conservation

of the landbird community of Dadia reserve, Greece. Diversity Distrib. 12: 620-629. Kirwan, G. M., Boyla, K. A., Castell, P., Demirci, B., Özen, M., Welch, H. & Marlow, T.

(2008). The Birds of Turkey - The Distribution, Taxonomy and Breeding of Turkish Birds. Christopher Helm.

Knapp, H. D. (2005). Die globale Bedeutung der kaspischen Wälder. In: Nosrati, K.,

Marvie Mohadjer, R., Bode, W. and Knapp, H.D., Schutz der Biologischen Vielfalt und integriertes Management der Kaspischen Wälder (Nordiran). Bundesamt für Naturschutz. Bonn: 45-70.

Korňan, M. (2009). Structure of the breeding bird assemblage of a primeval alder

swamp in the Šúr National Nature Reserve. Biologia 64/1: 165—179, 2009. Krüger, O. (2005). The role of ecotourism in conservation: panacea or Pandora’s box?

Biodiversity and Conservation 14: 579-600. Legendre, P. & Legendre, L. (1998). Numerical Ecology – Second English Edition.

Elsevier B. V. Leyer, I. & Wesche, K. (2007). Multivariate Statistik in der Ökologie - Eine Einführung.

Springer-Verlag Berlin Heidelberg. Lindenmayer, D. B., Margules, C. R. & Botkin, D. B. (2000). Indicators of Biodiversity

for Ecologically Sustainable Forest Management. Conservation Biology, vol. 14, no. 4: 941-950.

Liu, J., Linderman, M., Ouyang, Z., An, L., Yang, J. & Zhang, H. (2001). Ecological

Degradation in Protected Areas: The Case of Wolong Nature Reserve for Giant Pandas. Science 292. 98-101.

Loskot, V. M. (1978). [The Hyrkanian Tit, Parus hyrcanus Sarudny et Loundon]. Proc. of

Zool. Inst. of Acad. Sci. of the USSR, vol. 76: 46-60. Mac Arthur, R. H (1968). The theory of the niche. In R. C. Lewontin, ed., Population

Biology and Evolution: 159-176. Mac Arthur, R. H., & Mac Arthur, J. W. (1961). On Bird Species Diversity. Ecology 42:

594-598.


- 63 -

Mammadov, G.S., M.P. Babaev, Z.R. Movsumov, C.M. Ceferova & V.H. Hesenov (2007). Torpaq Atlasi (Soil Atlas, in Azeri and Russian). Baki Kartografiya Fabriki.

McCune, B. & Grace, J. B. (2002). Analysis of Ecological Communities. MjM Software

Design. Michael Succow Stiftung (2009). Potential Analysis for Further Nature Conservation in

Azerbaijan – A Spatial and Political Investment Strategy. Geozon Science Media. Ministry of Ecology and Natural Resources of Azerbaijan Republic (MENR) (2004).

Nomination of the "Hirkan forests" of Azerbaijan as UNESCO World Heritage Site. Baku: unpublished manuscript.

Ministry of Ecology and Natural Resources of Azerbaijan Republic (MENR) (2010).

Biodiversity. http://www.eco.gov.az. Accessed: 30.04.2010. Mohadjer, R. M. (2005). Prinzipien der naturnahen Waldwirtschaft. In: Nosrati, K.,

Marvie Mohadjer, R., Bode, W. and Knapp, H.D., Schutz der Biologischen Vielfalt und integriertes Management der Kaspischen Wälder (Nordiran). Bundesamt für Naturschutz. Bonn, 121-127.

Mühr, B. (2007). Das Klima in Lenkoran.

http://www.klimadiagramme.de/Asien/lenkoran.html. Accessed: 15.04.2010. Murakami, M., Hirao, T., Iwamoto, J. & Oguma, H. (2008). Effects of windthrow

disturbance on a forest bird community depend on spatial scale. Basic and Applied Ecology 9: 762-770.

Murcia, C. (1995), Edge effects in fragmented forests: implication of conservation. Tree

vol. 10 no. 2: 58-62. Mustafaev, G. T. & Agaeva, T. A. (1981). [The birds of the Talish Mounains at

Azerbaijanian border]. Azerb. gos.107-115. Baku. Mac Nally, R. & Horrocks, G. (2002). Proportionate spatial sampling and equal-time

sampling of mobile animals: A dilemma for inferring areal dependence. Austral Ecology 27: 405-415.

NASA (2004). The Caucasus-Caspian region (30.09.2004). http://visibleearth.nasa.gov.

Accessed: 12.02.2010. Noack, F. & Hidayatov, A. (2007). The Economics of Forest Depletion in Southern

Azerbaijan. Archiv für Naturschutz und Landschaftsforschung , 2 (46), 109-131.


- 64 -

Parker, A. J. (1982). The topographic relative moisture index: an approach to soil-moisture assessment in mountain terrain. Phys. Geogr. 3(2):160-168.

Patrikeev, M. (2004). The Birds of Azerbaijan. Pensoft, Sofia/Moscow. Patten, M. A., de Silva, H. G. & Smith-Patten, B. D. (2010). Long-term changes in the

bird community of Palenque, Chiapas, in response to rainforest loss. Biodivers Conserv 19: 21-36.

Pearman, P. B. (2002). The scale of community structure: Habitat variation and avian

guilds in tropical forest understory. Ecological Monographs, 72, 19–39. Pëterhofs, E. & Priednieks, J. (1989). Problems of applying the line-transect method

without repeated counts when the breeding season is long. Ann. Zool. Fennici 26: 181-184.

Pino, J., Rodá, F., Ribas, J., Pons, X. (2000). Landscape structure and bird species

richness: implications for conservation in rural areas between natural parks. Landscape and Urban Planning 49: 35-48.

Poulsen, O. B. (2002). Avian richness and abundance in temperate Danish forests: tree

variables important to birds and their conservation. Biodiversity and Conservation 11: 1551-1566.

Prilipko, L., (1954). Lesnaja Rastitelnost Aserbaidschana. Baku: Isdatelstwo Akademi

Nauk Aserbaidschanskoi SSR.

Quantum GIS Development Team (2009). Quantum GIS Geographic Information System. Open Source Geospatial Foundation Project. http://qgis.osgeo.org

R Foundation for Statistical Computing (2009). R: A Language and Environment for

Statistical Computing. Vienna, Austria. http://www.r-project.org Radde, G. (1884). Ornis Caucasica. Kassel: Theodor Fischer. Radde, G. (1886a). Reisen an der persisch-russischen Grenze -Talysch und seine

Bewohner. Leipzig: F.A.Brockhaus. Radde, G. (1886b). Die Flora und Fauna des südwestlichen Caspi-Gebietes –

Wissenschaftliche Beträge zu den Reisen an der Persich-Russischen Grenze. Leipzig: F.A.Brockhaus.

Ramírez-Albores,J.E.(2007). Bird diversity and conservation of Alto Balsas

(Southwestern Puebla), Mexico. Rev. Biol. Trop. Vol. 55 (1): 287-300.


- 65 -

Rietschel, M. (2010). Multitemporale Untersuchungen Hyrkanischer Wälder mittels Change Detection auf der Basis von Landsat-Daten. Diploma thesis. Ernst-Moritz-Arndt-University. Greifswald.

Saffari, N. G. (2005). Landnutzung, ländliche Entwicklung und Gefährdung der

Kaspischen Wälder. In: Nosrati, K., Marvie Mohadjer, R., Bode, W. and Knapp, H.D., Schutz der Biologischen Vielfalt und integriertes Management der Kaspischen Wälder (Nordiran). Bundesamt für Naturschutz. Bonn, 89-97.

Saniga, M. & Saniga, M. (2004). Influence of forest stand structure on the occurence of

bird community in Skalná Alpa National Nature Reserve in the Vel’ká Fatra Mts. (West Carpathians). J. For. Sci. 50: 219-234.

Scharnweber, T., Rietschel, M., & Manthey, M. (2007). Degradation stages of the

Hyrcanian forests in southern Azerbaijan. Archiv für Naturschutz und Landschaftsforschung , Vol.46 ( Heft 2), 133-156.

Seifollahian, M., Ebrahimi Rastaghi, M. & Hedayati, M.A. (2005). Die Bedeutung der

Pflanzenarten der Kaspischen Wälder. In: Nosrati, K., Marvie Mohadjer, R., Bode, W. and Knapp, H.D., Schutz der Biologischen Vielfalt und integriertes Management der Kaspischen Wälder (Nordiran). Bundesamt für Naturschutz. Bonn, 71–80.

Sekercioglu, C. H. (2002). Effects of forestry practices on vegetation structure and bird

community of Kibale National Park, Uganda. Biological Conservation 107: 229-240.

Sekercioglu, C. H., Daily, G. C., Ehrlich, P. R. (2004). Ecosystem consequences of bird

declines. PNAS, vol.101, no.52: 18042-18047. Skvorzov, G. A. (1976a). Topographical Map 1:100 000 J-39-37 Lerik. General Staff of

Soviet Military. Skvorzov, G. A. (1976b). Topographical Map 1:100 000 J-39-38 Lenkoran. General Staff

of Soviet Military. Südbeck, P., Andretzke, H., Fischer, S., Gedeon, K., Schikore, T. Schröder, K. &

Sudfeldt, C. (2005): Methodenstandards zur Erfassung der Brutvögel Deutschlands. Radolfzell.

Sultanov, E. (2008). Recent Waterbird Counts in Gyzylagach, the Azerbaijan Republic’s

Most Important Ramsar Site, with Comments on Conservation. Podoces, 3(1/2): 31-38.

Tews, J., Brose,U. ,Grimm, V., Tielbörger, K., Wichmann, M. C., Schwager, M. &

Jeltsch, F. (2004). Animal species diversity driven by habitat


- 66 -

heterogeneity/diversity: the importance of keystone structures. J. Biogeogr. 31: 79-92.

The State Statistical Committee of the Republic of Azerbaijan (2009). Demographic

indicators. http://www.azstat.org. Accessed: 03.03.2010. Thiollay, M-J. (1999). Responses of an avian community to rain forest degradation.

Biodiversity and Conservation 8: 513-534. Tomiałojć, L. & Wesołowski, T. (2004). Diversity of the Białowieża Forest avifauna in

space and time. J. Ornithol. 145: 81–92. Ugalde-Lezama, S., Alcántara-Carbajal, J. L., Valdez-Hernández, J. I., Ramírez-Valverde,

G., Velázquez-Mendoza, J. & Tarángo-Arámbula, L. A. (2010). Richness, abundance and diversity of birds in the temperate forest with different conditions of perturbance. Agrociencia 44: 159-169.

Urquhart, E. & Bowley, A. (2002). Stonechats - A Guide to the Genus Saxicola.

Christopher Helm. Väisänen, R. A. (1989). Renewal of methodology in the second bird atla of Finland, 1986-

1989. Ann. Zool. Fennici 26: 167-172. Walther, B. A. & Martin, J-L. (2001). Species richness estimation of bird communities:

how to control for sampling effort? Ibis 143: 413-419. Ward, J. H. (1963). Hierarchical Grouping to Optimize an Objective Function. American

Statistical Ass.58 (301), 236-244. Wesołowski, T. & Tomiałojć, L. (1997). Breeding bird dynamics in a primeval temperate

forest: long-term trends in Białowieża National Park (Poland). Ecography 20: 432–453.

Wesołowski, T. (2003). Bird community dynamics in a primaeval forest – is interspecific

competition important? Ornis Hungarica 12-13: 51-62. Wesołowski, T. (2007). Primeval conditions – what can we learn from them? Ibis 149

(Suppl. 2), 64–77. Zohary, M. (1973). Geobotanical Foundations of the Middle East. Stuttgart: G.Fischer.

Appendices The breeding bird communities of the Talish mountains (Azerbaijan) and their response to forest degradation

- 67 -

10 Appendices

Annex 1: Commented list of the observed bird species of the Talish mountains region including breeding status (b = breeding, n = non-breeding), based on own observations and interpretations, and IUCN Red List (2010) category (EN = Endangered, VU = Vulnerable, NT = Near Threatened, Least Concern not mentioned).

No. Species Comment Status IUCN1 Cygnus cygnus 1 immature bird on 20 May near Narimanabad. n 2 Cygnus olor Up to 85 individuals in wetlands near Liman. n 3 Anas querquedula 1 bird near Narimanabad on 20 May n

4 Alectoris chukar Common at rock outcrops in Zuvand upland. 1 nest with 20 eggs found near Kialakhan on 2 June.

b

5 Tetraogallus caspius On 9 July a calling bird was heard in rocks near Khozavi. b

6 Francolinus francolinus

Only 2 displaying males observed on 5 May near Tazakend (Caspian lowland).

b

7 Perdix perdix Rare. A displaying male heard on 3 April near Aliabad (montane meadow belt). 2 birds in montane semi-desert near Nalabin on 24 June.

b

8 Coturnix coturnix Singing males were common in the montane meadow belt and the Caspian lowland.

b

9 Phasianus colchicus

No observation. Only a plucking of a female or juvenile found in shrubs at steep slope near Lerik on 14 April.

b

10 Podiceps cristatus A colony of 49 breeding pairs in wetlands near Liman on 20 May. 2 chicks also present.

b

11 Phalacrocorax carbo 13 birds in wetlands near Liman on 20 May. n

12 Phalacrocorax pygmaeus

Some observations around Liman, Narimanabad and Lenkoran river.

(b) NT

13 Ardea cinerea Single birds throughout the Caspian coast and lowland. (b) 14 Ardea purpurea Found in lowland, especially in wetlands near Liman. (b) 15 Ardea alba 1 adult in costal lagoons near Narimanabad. (b) 16 Egretta garzetta Common in wetlands of the Caspian lowland. (b) 17 Ardeola ralloides Up to 30 birds on 20 May in wetlands near Liman. (b)

18 Bubulcus ibis 2 observations in the Caspian lowland in May. On 6 May 5 birds foraging in montane meadow near Aliabad.

(b)

19 Nycticorax nycticorax Regularly seen in the lowland, especially in wetland near Liman. (b)

20 Ixobrychus minutus

Occurred in the lowland. 4 migrants on 15 May in riparian forest near Gosmalijan.

b

21 Plegadis falcinellus Up to 200 birds in wetlands near Liman on 2 July. (b)

22 Ciconia nigra Occasionally seen in the Caspian lowland and the forest belt. b 23 Ciconia ciconia Breeds in villages in the lowland, but is rather rare. b 24 Pernis apivorus A few migrants seen from 7 to 14 May. (b) 25 Milvus migrans 1 bird on 10 May near Ashagy Bilnia and 1 bird on 27 May in (b)

The breeding bird communities of the Talish mountains (Azerbaijan) and their response to forest degradation Appendices

- 68 -

Moscow forest.

26 Neophron percnopterus Only a few times seen near Lerik and in Zuvand. (b) EN

27 Gyps fulvus Only observed in Zuvand. 1 near Divagach on 8 May and 16 near Mistan on 15 May.

n

28 Circaetus gallicus Regularly seen at altitudes below 1200 m a.s.l. b

29 Circus aeruginosus Breeds in the Caspian lowland. b

30 Circus macrourus Migrant. n NT

31 Aegypius monachus 1 individual on 3 April near Aliabad. n NT

32 Accipiter badius 9 territories were found in the Caspian lowland around Masally and Lenkoran (Heiss & Gauger 2008).

b

33 Accipiter nisus Nine observations from April to July . b 34 Accipiter gentilis Twice seen in April in the lower foothills. (n) 35 Buteo buteo Common in the forest and montane meadow belt. b

36 Buteo rufinus Regularly seen in Zuvand upland with territorial behavior of a pair west of Pirasora

b

37 Aquila pomarina Displaying individuals in forest near Günesli and montane meadows near Ashagy Bilnia. Regularly seen within the forest belt.

b

38 Aquila nipalensis Several migrating birds seen from 3 April to 20 Mai. n

39 Aquila chrysaetos Regularly seen in Zuvand. 1 abandoned nest found in rocks near Shonadzhola.

b

40 Aquila pennata Few observations within the forest belt. b

41 Falco naumanni A small colony with 6 breeding pairs was found on 19 April in a building in Khialakhan.

b VU

42 Falco tinnunculus Breeds in the riparian forest and in rocky habitats. Only 1 observation in the Caspian lowland on 20 May near Boladi.

b

43 Falco subbuteo Nests in the riparian forest, but was also observed in the Caspian lowland (near Tazakend, Moscow forest).

b

44 Falco biarmicus An adult seen on 6 July on rocks near Pirasora. n 45 Falco peregrinus 1 individual on 9 July west of Khozavi. (n)

46 Crex crex One singing male on 10 May in monane meadows near Ashagy Bilnia.

(b) NT

47 Gallinula chloropus

An adult with a juvenile observed at a back water of the Lenkoran river in the lowland.

b

48 Charadrius dubius

Was seen at costal lagoons near Narimanabad and Lenkoran river.

(b)

49 Himantopus himantopus 10 birds in wetlands near Liman on 10 July. b

50 Limosa limosa Migrant. 70 birds on 2 July and 30 birds on 10 July in wetlands near Liman.

n NT

51 Tringa totanus 10 birds on 2 July in wetlands near Liman. n

52 Tringa ochropus Migrant. Observed along rivulets of the riparian forest and Lenkoran river near Vel.

n

53 Actitis hypoleucos

Two displaying individuals on 22 April at Lake Xanbulan are maybe migrants. Resting birds were found in the beginning of July at several locations in the Caspian lowland (costal lagoons near Narimanabad, wetlands near Liman, Lenkoran river).

(b)

54 Arenaria 1 at a costal lagoon near Narimanabad on 20 and 24 May. n


- 69 -

interpres

55 Philomachus pugnax 1 bird in wetlands near Liman on 10 July. n

56 Glareola pratincola

Occurred only in the lowland. 2 birds near Sirabil on 21 May and 4 birds at costal lagoons near Narimanabad on 20 May.

(b)

57 Larus cachinnans Regularly at the Caspian coast. b

58 Chroicocephalus ridibundus Regularly at costal lagoons and Lenkoran river (lowland). (n)

59 Chroicocephalus genei 30 individuals on 20 May near Liman. (n)

60 Thalasseus sandvicensis 10 birds on 20 May at the Caspian coast near Narimanabad. n

61 Sterna hirundo A few observations in wetlands near Liman and the Caspian coast.

n

62 Sternula albifrons A few observations in wetlands near Liman and the Caspian coast.

n

63 Chlidonias hybrida Common, with up to 2500 individuals in wetlands near Liman. b

64 Chlidonias leucopterus

Common in wetlands near Liman. 1 migrating flock with 150 individuals was seen on 2 May at 2300 m a.s.l. near Mistan.

(b)

65 Stercorarius parasiticus An adult seen at the Caspian coast near Narimanabad. n

66 Columba livia A wild bird was seen on 1 June near Lialiakeran (Zuvand). b 67 Columba oenas Common in the forest belt. b

68 Columba palumbus

Common in the riparian forest. Only 3 observations within the forest belt with 1 singing male near Günesli and 1 singing male near Tankivan.

b

69 Streptopelia turtur

A few singing males observerd in the montane meadow belt and Caspian lowland.

b

70 Streptopelia decaocto Occasionally seen in the Caspian lowland, e.g. Masally. b

71 Streptopelia senegalensis 1 bird at the Lenkoran river near Vel on 10 July. b

72 Cuculus canorus Common in the Talish mountains at all altitudes. b

73 Otus scops Several singing males with the montane meadow belt and the riparian forest. Only 1 singing male in the lowland near Hirkan village on 3 July.

b

74 Bubo bubo 3 fledged juveniles found on 4 and 5 July in a rocky cliff east of Mistan.

b

75 Strix aluco Almost everywhere heard at night within the forest belt. b

76 Athene noctua Seen in montane meadow belt near Bilaband and the riparian forest.

b

77 Caprimulgus europaeus

First seen on 22 April near lake Xanbulan. Singing males were found from the Caspian lowland to rocky habitats throughout all habitat types. 2 chicks were found on 11 July near Vel (lowland).

b

78 Tachymarptis melba

Regularly seen along rocky cliffs e.g. Pirasora, Nalabin, Mistan and Khozavi.

b

79 Apus apus Breeds in rocky habitats and towns throughout all altitudes. b

80 Alcedo atthis Occurred common along channels of the Caspian lowland. A few observations were made along rivers within the forest belt.

b

81 Merops persicus 1 colony with about 100 pairs found north of Kumbashi (Caspian lowland).

b


- 70 -

82 Merops apiaster

During migration common throughout the Talish mountains. Breeds only at sandy cliffs of the montane semi-desert and along rivers of the lowand. Only small colonies were found with 2-3 pairs each.

b

83 Coracias garrulus Displaying individuals were found in the Caspian lowland and rarely in the montane meadow belt.

b NT

84 Upupa epops Regularly seen in villages throughout the Talish mountains, especially in riparian forest, montane meadow belt and Caspian lowland.

b

85 Jynx torquilla Several singing males within the forest belt, especially the park-like degradation stage.

b

86 Dendrocopos minor

Rare breeder of the forest belt. Occurred in the lowland only in the Moscow forest.

b

87 Dendrocopos major Common in the forest belt. b

88 Dendrocopos syriacus

Common in the riparian forest, but was also observed in the Caspian lowland near Kumbashi and Tazakend.

b

89 Dryocopus martius

1 singing male on 27 April in primeval forest near Siov at 800 m a.s.l.

b

90 Picus viridis Common in the forest belt and riparian forest. b

91 Lanius collurio Common wherever shrubs are present, especially in the montane meadow belt, Caspian lowland, riparian forest and shrubby woodland stage.

b

92 Lanius isabellinus Migrant. 1 individual on 9 May near Ambu and 1 individual on 18 May near Shinaband.

n

93 Lanius minor Common in the riparian forest, but occurred also in the montane meadow belt and Caspian lowland.

b

94 Lanius senator Only a few single birds in rocky habitats, riparian forest, montane meadow belt and Caspian lowland.

b

95 Garrulus glandarius Regularly seen in the forest belt. b

96 Pica pica Common breeder of Caspian lowland, montane meadow belt and riparian forest.

b

97 Pyrrhocorax pyrrhocorax A pair was several times observed in rocks west of Pirasora. b

98 Corvus frugilegus Common around Kialvas, where it breeds in a colony in riparian forest.

b

99 Corvus cornix Common breeder in open habitats, e.g. Caspian lowland, montane meadow belt, riparian forest.

b

100 Corvus corax Regularly seen in the forest belt and rocky habitats, where it breeds.

b

101 Melanocorypha calandra

Occurred in the montane meadow belt near Aliaband and Muria and in the Caspian lowland west of Masally.

b

102 Melanocorypha bimaculata Regularly in montane semi-desert above 1500 m a.s.l. b

103 Calandrella rufescens

Locally common in the Caspian lowland around Sirabil, Sarchuvar and west of Masally.

b

104 Galerida cristata Only 1 bird seen west of Masally. b 105 Lullula arborea Common in the montane meadow belt and Zuvand. b

106 Alauda arvensis Common in the montane meadow belt around Lerik and montane semi-deserts and subalpine meadows e.g. around Mistan, Orand and Pirasora. Did not occur below 1000 m.a.s.l.

b


- 71 -

107 Eremophila alpestris

Breeds in the montane semi-desert. The first fledged juvenile was observed on 15 May near Mistan.

b

108 Ripparia ripparia Common migrant. No breeding colonies found. b

109 Ptyonoprogne rupestris Regularly seen at steep, rocky cliffs. 2 nests found near Nalabin. b

110 Hirundo rustica Breeds in villages and is common in the Caspian lowland. b 111 Delichon urbicum Breeds in rocky habitats and towns throughout all altitudes. b 112 Periparus ater Common breeder of the forest belt. b 113 Parus major Common throughout the Talish mountains. b

114 Cyanistes caeruleus Common within the forest belt and riparian forest. b

115 Remiz pendulinus Occurred in the Caspian lowland. b

116 Aegithalos caudatus Regularly seen in the forest belt. b

117 Sitta europaea Common in the forest belt. b 118 Sitta neumayer Common in rocky habitats. b

119 Certhia familiaris Only 4 singing males observed. 2 near Siov in natural forest and 2 near Sifiakeran in slightly disturbed forest.

b

120 Troglodytes troglodytes

Common in the forest belt. A rare breeder in riparian forest with a nest found near Mistan.

b

121 Cinclus cinclus Breeds along rivers throughout the Talish mountains except the Caspian lowland.

b

122 Cettia cetti

This species was distributed in two regions. It was common in riparian forest above 1200 m a.s.l., e.g. near Gosmalian, Shonadzhola and Mistan and occurred also in the Caspian lowland near Boladi and Liman along channels.

b

123 Locustella naevia Rare. 1 singing male on 28 April near Sim at a forest clearing was maybe a migrant. Furthermore, 1 singing male on 5 June near Bilaband in the montane meadow belt.

n

124 Locustella fluviatilis

Migrant. On 16 to 17 May a singing male in riparian forest near Gosmalian. On 19 May a singing male near Kumbashi. 2 singing males near Boladi on 20 May.

n

125 Acrocephalus schoenobaenus

Was treated as a breeding bird, but is probably only a migrant. Singing males were regularly seen along channels of the lowlands, but no observation was done after 1 June.

b

126 Acrocephalus scirpaceus

Due to the hidden behavior and the similar song with Hippolais pallida maybe overlooked. Only 3 observations. A singing male on 27 May in a park-like forest stage near Piran and a singing male in the riparian forest near Mistan on 1 June are migrants. A further singing male on 20 May near Boladi (Caspian lowland) occupied probably a territory.

b

127 Acrocephalus dumetorum

1 singing male on 19 May in bushes near Kumbashi. This species is not listed in Patrikeev (2004), but obviously a regular migrant in autumn (Gauger 2005). This is the first spring record to Azerbaijan.

n

128 Acrocephalus palustris

Was treated as a breeding bird, but is probably only a migrant. Was seen mostly in the riparian forest, but occurred rarely also in the montane meadow belt and the lowland from 11 May to 1 June.

(b)

129 Acrocephalus arundinaceus

Common along channels and reedbeds in the lowland. 1 territory was found near Aliabad with a singing male from 6 May to 5 June.

b


- 72 -

130 Hippolais pallida Common in the Caspian lowland. First observed on 5 May. b

131 Hippolais icterina Common in lower altitudes of the forest belt (0-500 m a.s.l.). Singing males observed from 23 April to 21 June.

b

132 Phylloscopus trochilus Several singing migrants from mid-April to mid-May. n

133 Phylloscopus collybita

Maybe only a migrant. Several singing males heard at the begining of April. However, 4 singing males heard near Shinaband on 5 June.

n

134 Phylloscopus sindianus

Some singing males found in shrubs at higher altitudes above 1000 m a.s.l., but no sure signs of breeding.

(b)

135 Phylloscopus nitidus Common in the forest belt. b

136 Sylvia atricapilla Common in the forest belt. b

137 Sylvia borin Migrant. 1 bird on 16 May near Gosmalian, 1 bird on 17 May near Mistan and 1 bird on 18 May near Pirasora.

n

138 Sylvia communis Common in riparian forest. b

139 Sylvia curruca Several singing males in rocky habitats with bushes and in dense shrubs above 1000 m a.s.l.

b

140 Sylvia nisoria Probably an uncommun breeder in dense shrubs of higher altitudes. Observed near Shinaband, Mistan and Khosavi.

b

141 Sylvia mystacea Breeds rarely at a few locations in the Caspian lowland. Occurred as migrant also elsewhere, e.g. Zuvand.

b

142 Muscicapa striata Common in the forest belt and riparian forest. b

143 Ficedula hypoleuca

A male bird was observed on 4 April in riparian forest near Divagach.

n

144 Ficedula semitorquata Regularly seen in less degraded forests, especially around Siov. b NT

145 Ficedula parva Common breeder of the forest belt. Was at 3 April already present.

b

146 Erithacus rubecula Common in the forest belt. b

147 Luscinia megarhynchos

Common in montane meadow belt, riparian forest, shrubby woodland stage and Caspian lowland.

b

148 Irania gutturalis

Breeds in rocky habitats with bushes above 2000 m a.s.l. in Zuvand upland. 2 singing males and a female found on 30 May west of Orand. A total of 4 pairs found near Mistan. 2 families observed on 4 July with 3 and 2 juveniles each.

b

149 Cercotrichas galactotes

On 21 May 3 birds near Garibljar and on 24 May 1 singing male near Boladi.

b

150 Phoenicurus ochruros Common in rocky habitats. b

151 Phoenicurus phoenicurus Regularly in park-like forest stage and riparian forest. b

152 Saxicola rubetra Rare. Observed e.g. at forest clearings near Siov or near Mistan and Pirasora in Zuvand.

b

153 Saxicola torquatus Breeds in rocky habitats of Zuvand upland. b

154 Oenanthe oenanthe

Locally common in rocky habitats of Zuvand, e.g. near Pirasora. Also a rare breeder of the montane meadow belt.

b

155 Oenanthe finschii Common at rocky outcrops of Zuvand, especially near Pirasora, Kialvas and Kialakhan.

b

156 Oenanthe Rare breeder in rocky habitats of Zuvand. b


- 73 -

hispanica melanoleuca

157 Oenanthe isabellina Common breeder of montane semi-desert. b

158 Monticola saxatilis Common in rocky habitats. b

159 Monticola solitarius

This species was only regularly seen in rocks between Shinaband and Nalabin. A pair was observed on 22 June near Kialakhan (K. Gauger pers. comm.)

b

160 Turdus torquatus Locally common in rocky habitats above 1500 m a.s.l., e.g. near Shinaband, Mistan, Nalabin, Kialakhan, Orand.

b

161 Turdus merula Common. b

162 Turdus philomelos Common in the forest belt. b

163 Turdus viscivorus Regularly seen in the forest belt, especially at the lower and upper treeline and the park-like forest stage. A family with 3 juveniles were found near Mamedoba at the lower tree line.

b

164 Oriolus oriolus Occurred in forest habitats. b

165 Sturnus vulgaris Common breeder of the riparian forest. Occurred also in montane meadow belt and Caspian lowland.

b

166 Pastor roseus Several migrating flocks seen in the beginning of May. Occurred also elsewhere, but disappeared at the end of May.

n

167 Prunella collaris Rare breeder in rocky habitats of altitudes above 2000 m a.s.l., e.g. Mistan, Pirasora.

b

168 Prunella ocularis

The specimen of the species was shot in June 1880 at the Kyziurdi mountain (Radde 1886a) and since then never again seen in the region (Patrikeev 2004). Dr. H. Müller rediscovered this species on 17 May in a rocky cliff with Juniperus spec. northeast of Mistan, where it was also seen on 22 June and 5 July by K. Gauger, J.Etzold and the author.

b

169 Prunella modularis

Rarely seen in the forest belt. Occurred also in riparian forest near Mistan.

b

170 Motacilla alba Regularly seen in villages, Caspian lowland and riparian forest. b 171 Motacilla citreola A bird on migration on 9 May near Dshangemiran. n

172 Motacilla flava feldegg

A locally common breeder of the lowland, especially around Boladi, Sirabil and Sarchuvar.

b

- Motacilla f. flava A pair was observed on 17 May near Pirasora. n 173 Motacilla cinerea Regularly seen along rivers of the forest belt. b

174 Anthus campestris Several territories found in Zuvand in 1400-2300 m.a.s.l in montane semi-desert.

b

175 Anthus trivialis Occurred in April in migrating flocks of about 5 birds. Singing males were observed since May e.g. around Dshangemiran, Shonadzhola or Pirasora.

b

176 Anthus pratensis 1 migrating bird near Aliabad on 4 April. n 177 Anthus cervinus 1 migrating bird near Ashagy Bilnia on 10 May. n

178 Anthus spinoletta 8 singing males and a food carrying bird was seen at 2000-2300 m.a.s.l. on a sub-alpine meadow between Razgov and Mistan.

b

179 Emberiza cia Common in rocky habitats. b

180 Emberiza hortulana

Common in rocky habitats around 2000 m a.s.l. Occurred during migration also elsewhere.

b

181 Emberiza melanocephala

Abundant in the montane meadow belt and also present in drier areas of the Caspian lowland. Arrived on 24 April from its

b


- 74 -

wintering grounds.

182 Emberiza calandra

Common in the montane meadow belt and Caspian lowland. Occurred also along the riparian forest.

b

183 Fringilla coelebs Common and widespread species of the forest habitats. Occurred also in the lowlands along afforstations.

b

184 Carpodacus erythrinus

Regularly seen in the riparian forest above 1400 m a.s.l., but occurred also elsewhere.

b

185 Carduelis chloris Common below 1000 m a.s.l., above a rare sight. b 186 Carduelis spinus Regularly seen within the forest belt. b

187 Carduelis carduelis Common in the whole Talish mountains. b

188 Carduelis cannabina

Occurred above 1200 m a.s.l. near Lerik and was common in Zuvand upland.

b

189 Serinus pusillus Locally common in rocky habitats with a few shrubs, e.g. near Nalabin, Khozavi, Orand.

b

190 Pyrrhula pyrrhula Only a few observations within the forest belt. b

191 Coccothraustes coccothraustes Common in the forest belt. b

192 Rhodopechys sanguineus

1 pair at rocky cliffs northeast of Mistan and another pair, including a singing male, on a rocky outcrop north of Pirasora on 15 and 29 May, respectively.

b

193 Bucanetes githagineus

Fist seen on 8 May in Zuvand upland. Occurred regularly with up to 4 individuals near Divagach, Pirasora and Kialakhan. A fledged juvenile was seen in July near Divagach (C. Völlm pers. comm.).

b

194 Passer domesticus Common in villages, especially in the Caspian lowland with a large flock of 500 individuals near Hirkan village on 2 July.

b

195 Passer hispaniolensis 3 birds near Divagach on 22 June (K. Gauger pers. comm.). (b)

196 Passer montanus Only a total of 18 birds seen in the Caspian lowland. b 197 Petronia petronia Locally common breeder in rocks, e.g. west of Orand, Khozavi. b

Annex 2: Nesting guilds of selected bird species based on literature review (Flade 1994, Andretzke et al. 2005, Glutz von Blotzheim & Bauer 1991, Glutz von Blotzheim & Bauer 1994, Glutz von Blotzheim & Bauer 1998, Urquhart, & Bowley 2002, Kirwan et al. 2008, Alström & Mild 2003, Patrikeev 2004).

Bird species Nesting guild Bird species Nesting guild

Aegithalos caudatus canopy Lanius minor shrub

Alauda arvensis ground Lanius senator shrub

Alcedo atthis cavity Lullula arborea ground

Alectoris chukar ground Luscinia megarhynchos ground

Anthus campestris ground Melanocorypha bimaculata ground

Anthus trivialis ground Melanocorypha calandra ground

Aquila pennata canopy Motacilla alba cavity

Aquila pomarina canopy Motacilla cinerea cavity

Athene noctua cavity Motacilla flava ground

Buteo buteo canopy Muscicapa striata cavity

Caprimulgus europaeus ground Oenanthe finschii ground

Carduelis cannabina shrub Oenanthe hi.melanoleuca ground


- 75 -

Carduelis carduelis canopy Oenanthe isabellina ground

Carduelis chloris canopy Oenanthe oenanthe ground

Carduelis spinus canopy Oriolus oriolus canopy

Carpodacus erythrinus shrub Otus scops cavity

Certhia familiaris cavity Parus major cavity

Ciconia nigra canopy Passer domesticus cavity

Cinclus cinclus cavity Passer montanus cavity

Circaetus gallicus canopy Periparus ater cavity

Coccothraustes coccothraustes canopy Petronia petronia cavity

Columba oenas cavity Phoenicurus ochruros cavity

Columba palumbus canopy Phylloscopus collybita ground

Coracias garrulous cavity Phylloscopus nitidus ground

Corvus corax canopy Phylloscopus sindianus ground

Corvus cornix canopy Phylloscopus trochilus ground

Cuculus canorus x Pica pica canopy

Cyanistes caeruleus cavity Picus viridis cavity

Dendrocopos major cavity Prunella modularis shrub

Dendrocopos minor cavity Pyrrhula pyrrhula canopy

Dendrocopos syriacus cavity Remiz pendulinus canopy

Dryocopus martius cavity Saxicola rubetra ground

Emberiza calandra ground Saxicola torquatus ground

Emberiza cia ground Sitta europaea cavity

Emberiza melanocephala shrub Streptopelia turtur shrub

Erithacus rubecula ground Strix aluco cavity

Falco subbuteo canopy Sturnus vulgaris cavity

Falco tinnunculus cavity Sylvia atricapilla shrub

Ficedula hypoleuca cavity Sylvia communis shrub

Ficedula parva cavity Sylvia curruca shrub

Ficedula semitorquata cavity Sylvia nisoria shrub

Fringilla coelebs canopy Troglodytes troglodytes ground

Garrulus glandarius canopy Turdus merula shrub

Hippolais icterina shrub Turdus philomelos canopy

Hippolais pallida shrub Turdus torquatus shrub

Hirundo rustica cavity Turdus viscivorus canopy

Jynx torquilla cavity Upupa epops cavity

Lanius collurio shrub

Annex 3: Bird community table of the nine breeding bird communities including parameter values and species abundance per transect (territory/km).

Please see attatched extra sheet!


- 76 -

Annex 4: All indicator species of each community and every combination of communities. Community Species Stat p-value

1 Hirundo rustica 0.963 0.000999***

1 Passer domesticus 0.937 0.000999***

1 Acrocephalus schoenobaenus 0.816 0.000999***

1 Hippolais pallida 0.816 0.000999***

1 Acrocephalus arundinaceus 0.802 0.000999***

1 Alcedo atthis 0.749 0.000999***

1 Circus aeruginosus 0.745 0.000999***

1 Coracias garrulus 0.723 0.000999***

1 Motacilla flava 0.667 0.000999***

1 Calandrella rufescens 0.577 0.003996**

1 Passer montanus 0.577 0.002997**

1 Cercotrichas galactotes 0.471 0.00999**

1 Merops persicus 0.471 0.012987*

1 Remiz pendulinus 0.471 0.012987*

1 Streptopelia decaocto 0.471 0.00999**

2 Pyrrhula pyrrhula 0.639 0.000999***

2 Ficedula semitorquata 0.622 0.003996**

2 Certhia familiaris 0.548 0.007992**

5 Phylloscopus sindianus 0.522 0.00999**

6 Emberiza melanocephala 0.576 0.006**

6 Coturnix coturnix 0.478 0.038*

7 Cettia cetti 0.983 0.000999***

7 Dendrocopos syriacus 0.802 0.000999***

7 Phoenicurus phoenicurus 0.707 0.000999***

7 Sylvia communis 0.672 0.000999***

7 Lanius minor 0.615 0.003996**

7 Columba palumbus 0.591 0.002997**

7 Acrocephalus palustris 0.585 0.003996**

7 Falco subbuteo 0.567 0.005994**

8 Anthus campestris 0.727 0.000999***

8 Oenanthe isabellina 0.7 0.000999***

9 Monticola saxatilis 0.93 0.000999***

9 Phoenicurus ochruros 0.907 0.000999***

9 Serinus pusillus 0.768 0.000999***

9 Turdus torquatus 0.637 0.001998**

9 Oenanthe oenanthe 0.608 0.006993**

9 Tachymarptis melba 0.522 0.015984*

1+7 Sturnus vulgaris 0.853 0.000999***

1+9 Apus apus 0.634 0.007**

2+3 Turdus philomelos 0.903 0.000999***

2+3 Troglodytes troglodytes 0.884 0.000999***


- 77 -

2+3 Erithacus rubecula 0.865 0.000999***

2+3 Columba oenas 0.562 0.006993**

2+4 Hippolais icterina 0.816 0.000999***

3+4 Dendrocopos minor 0.466 0.039*

5+7 Carpodacus erythrinus 0.539 0.028*

7+9 Sylvia curruca 0.686 0.000999***

7+9 Falco tinnunculus 0.633 0.002997**

8+9 Carduelis cannabina 0.831 0.000999***

8+9 Alectoris chukar 0.831 0.000999***

8+9 Emberiza cia 0.83 0.000999***

8+9 Eremophila alpestris 0.792 0.000999***

8+9 Sitta neumayer 0.782 0.000999***

8+9 Oenanthe finschii 0.69 0.000999***

1+4+7 Oriolus oriolus 0.672 0.000999***

1+5+6 Streptopelia turtur 0.505 0.025*

1+6+7 Emberiza calandra 0.917 0.000999***

1+6+7 Pica pica 0.856 0.000999***

1+6+7 Motacilla alba 0.762 0.000999***

1+7+8 Merops apiaster 0.707 0.000999***

2+3+4 Dendrocopos major 0.947 0.000999***

2+3+4 Sitta europaea 0.931 0.000999***

2+3+4 Periparus ater 0.914 0.000999***

2+3+4 Coccothraustes coccothraustes 0.871 0.000999***

2+3+4 Ficedula parva 0.824 0.000999***

2+3+4 Carduelis chloris 0.792 0.000999***

2+3+4 Carduelis spinus 0.603 0.004995**

2+3+5 Garrulus glandarius 0.507 0.02*

2+3+7 Cyanistes caeruleus 0.914 0.000999***

2+4+7 Muscicapa striata 0.74 0.000999***

5+6+7 Anthus trivialis 0.499 0.023*

6+8+9 Alauda arvensis 0.62 0.011*

7+8+9 Petronia petronia 0.868 0.000999***

1+3+4+7 Carduelis carduelis 0.769 0.003**

1+4+5+7 Luscinia megarhynchos 0.846 0.000999***

1+4+6+7 Corvus cornix 0.812 0.000999***

2+3+4+5 Aegithalos caudatus 0.754 0.000999***

2+4+5+7 Phylloscopus nitidus 0.639 0.000999***

4+6+7+9 Lanius collurio 0.749 0.000999***

1+5+6+7+8 Upupa epops 0.724 0.000999***

2+3+4+5+7 Turdus merula 0.974 0.000999***

2+3+4+5+7 Fringilla coelebs 0.96 0.000999***

2+3+4+5+7 Sylvia atricapilla 0.825 0.000999***

2+3+4+5+7 Picus viridis 0.728 0.000999***


- 78 -

5+6+7+8+9 Lullula arborea 0.84 0.000999***

1+2+3+4+5+7 Parus major 0.922 0.000999***

Annex 5: Statistical analysis of the site parameters of the NMDS including all transects (Figure 21). ALTI = altitude, DATE = date, EXPO = slope exposition (after Parker 1982), SLO= slope steepness (after Parker 1982), HH = Herb layer height, HC = Herb layer cover, SH = Shrub layer height, SC = Shrub layer cover, LTH = Lower tree layer height, LTC = Lower tree layer cover, UTH = Upper tree layer height, UTC = Upper tree layer cover.

Site parameters r² Pr(>r)

ALTI 0.7225 0.0001***

SLOP 0.4568 0.0001***

UTH 0.7977 0.0001***

UTC 0.5126 0.0001***

LTH 0.6683 0.0001***

LTC 0.4452 0.0001***

SH 0.2757 0.0001***

HC 0.4185 0.0001***

SC 0.1629 0.0004***

HH 0.1579 0.0006***

DATE 0.0979 0.0109*

EXPO 0.0543 0.09119

Annex 6: Statistical analysis of the site parameters of the NMDS including all forest transects (Figure 22). ALTI = altitude, DATE = date, EXPO = slope exposition (after Parker 1982), SLO= slope steepness (after Parker 1982), HH = Herb layer height, HC = Herb layer cover, SH = Shrub layer height, SC = Shrub layer cover, LTH = Lower tree layer height, LTC = Lower tree layer cover, UTH = Upper tree layer height, UTC = Upper tree layer cover.

Site parameters r² Pr(>r)

UTH 0.7847 0.0001***

UTC 0.4377 0.0001***

LTH 0.5234 0.0001***

SC 0.3934 0.0001***

ALTI 0.3213 0.0007***

LTC 0.2197 0.005199**

DATE 0.1995 0.010799*

HH 0.1323 0.055594

SH 0.1348 0.058194

HC 0.1127 0.082692

EXPO 0.0654 0.258474

SLO 0.0414 0.431857


- 79 -

Annex 7: Relative abundance values (territory/km) of each bird species per landscape type.

Species Caspian lowland

Natural forest

Slightly disturbed forest

Intermediate disturbed forest

Park-like forest

Shrubby woodland

Montane meadow belt

Montane semi-desert

Riparian forest

Rocky habitats

Accipiter badius . . . 0.13 . . . . . . Accipiter nisus . . . . . . 0.05 . . . Acrocephalus arundinaceus 0.99 . . . . . 0.07 . . . Acrocephalus palustris 0.05 . 0.44 . . . 0.14 . 1.13 . Acrocephalus schoenobaenus 1.73 . . . . . . . . . Acrocephalus scirpaceus 0.06 . . . . . . . . . Aegithalos caudatus . 0.58 0.83 0.38 0.08 0.83 . . . . Alauda arvensis . . . . . 0.05 1.26 1.49 . 0.80 Alcedo atthis 0.69 . 0.23 0.11 . 0.11 . . . . Alectoris chukar . . . . . 0.21 . 0.99 0.22 1.43 Anthus campestris . . . . . 0.05 0.05 0.94 . 0.38 Anthus spinoletta . . . . . . . . . 0.60 Anthus trivialis . . . . . 0.34 0.24 . 0.06 0.05 Apus apus 2.36 . . 0.14 . . 0.43 . . 1.48 Aquila chrysaetos . . . . . . . . . 0.11 Aquila pomarina . 0.12 . 0.04 . . 0.15 . . . Ardea cinerea 0.06 . . . . . . . . . Bucanetes githagineus . . . . . . . 0.18 . 0.26 Buteo buteo . 0.23 0.17 0.08 0.09 0.16 0.36 . . 0.09 Buteo rufinus . . . . . . . . . 0.11 Calandrella rufescens 3.05 . . . . . . . . . Carduelis cannabina . . 0.08 . . 0.53 0.05 1.44 . 1.60 Carduelis carduelis 0.73 . 0.08 0.50 2.02 0.59 0.39 0.06 1.97 0.57 Carduelis chloris 0.17 0.12 0.66 0.96 2.35 0.54 . . . 0.05 Carduelis spinus . 0.58 0.25 0.23 0.42 0.05 . . . . Carpodacus erythrinus 0.04 0.12 0.08 0.04 0.09 0.56 . . 0.58 0.28 Cercotrichas galactotes 0.21 . . . . . . . . . Certhia familiaris . . 0.25 . . . . . . . Cettia cetti 0.13 . . . . . . . 2.55 . Cinclus cinclus . . . 0.08 . . . . 0.29 . Circaetus gallicus . . . . . 0.13 . . . . Circus aeruginosus 0.30 . . . . . . . . . Coccothraustes coccothraustes . 0.12 0.50 0.80 1.18 0.29 . . . . Columba oenas . 0.23 0.33 0.08 0.34 0.25 . . . . Columba palumbus . . . 0.04 0.08 0.05 . 0.06 0.57 0.05 Coracias garrulus 0.43 . . . . . 0.07 . . . Corvus corax . . . 0.08 . . 0.15 0.06 . 0.14 Corvus cornix 1.42 . . 0.11 0.84 0.44 1.07 0.11 0.89 0.09 Coturnix coturnix 0.04 . . . . . 0.98 . . . Crex crex . . . . . . 0.09 . . . Cuculus canorus 0.69 0.58 0.58 0.66 1.38 0.40 0.11 0.28 0.15 0.66 Cyanistes caeruleus 0.17 0.70 1.40 1.61 1.85 0.64 . . 1.53 . Delichon urbicum 0.47 . . . . . 0.76 0.22 . 1.04 Dendrocopos major . 3.84 3.47 2.11 2.94 0.44 . . . . Dendrocopos minor . . . 0.15 0.17 . . . . . Dendrocopos syriacus 0.13 . . . . . . . 1.72 . Dryocopus martius . 0.19 . . . . . . . . Emberiza calandra 3.95 . . . . 0.49 7.23 0.66 1.78 0.33 Emberiza cia . . . . . 0.28 . 1.22 0.29 3.92 Emberiza hortulana . . . . . . . 0.28 . 0.16 Emberiza melanocephala 0.34 . . . . 2.43 6.11 . . 0.11 Eremophila alpestris . . . . . . . 2.49 . 0.82 Erithacus rubecula . 5.70 4.30 1.95 0.50 0.34 . . . . Falco subbuteo 0.04 . . . . . . . 0.16 . Falco tinnunculus 0.04 . . . . . . 0.17 0.32 0.38 Ficedula parva . 3.60 3.14 1.76 1.26 0.05 0.05 . 0.19 . Ficedula semitorquata . 0.47 0.41 0.12 0.09 . . . . . Francolinus francolinus 0.04 . . . . . . . . .


- 80 -

Fringilla coelebs 0.04 7.21 9.50 7.39 6.30 2.70 0.15 0.06 2.87 0.14 Garrulus glandarius . 0.35 0.08 0.08 0.17 0.34 0.05 . . . Glareola pratincola 0.11 . . . . . . . . . Hippolais icterina . 0.35 1.41 1.21 1.10 0.50 . . . . Hippolais pallida 2.48 . . . . . . . . . Hirundo rustica 4.81 . . . . . 0.63 0.06 . . Irania gutturalis . . . . . . . . . 0.13 Ixobrychus minutus 0.05 . . . . . . . . . Jynx torquilla . . 0.11 . 1.10 0.11 0.11 . . . Lanius collurio 0.34 . 0.12 0.13 2.09 1.43 1.54 0.27 2.13 0.66 Lanius minor 0.21 . . . . 0.29 0.13 0.21 0.82 0.05 Lanius senator 0.04 . . . . . 0.07 . . 0.07 Lullula arborea . . . . 0.25 0.59 0.73 2.10 0.25 0.66 Luscinia megarhynchos 3.86 0.12 0.11 1.69 2.86 6.79 1.35 0.06 3.87 0.60 Melanocorypha bimaculata . . . . . . . 0.41 . 0.07 Melanocorypha calandra 0.15 . . . . . 0.73 . . . Merops apiaster 0.52 . . . . . . 0.50 0.22 0.11 Merops persicus 1.42 . . . . . . . . . Monticola saxatilis . . . . . . . 0.44 . 1.89 Monticola solitarius . . . . . . . . . 0.44 Motacilla alba 0.90 . . 0.08 . 0.10 0.63 0.06 1.15 0.24 Motacilla cinerea . 0.23 0.41 0.54 . 0.20 . . 0.13 0.09 Motacilla flava 1.12 . . . . . . . . . Muscicapa striata 0.13 0.35 1.18 0.81 1.65 . . . 1.23 . Neophron percnopterus . . . . . . . . . 0.05 Oenanthe finschii . . . . . . . 0.61 . 1.32 Oenanthe hispanica melanoleuca . . . . . . . 0.06 . 0.27 Oenanthe isabellina . . . . . . . 1.27 . 0.05 Oenanthe oenanthe . . . . 0.08 . 0.10 0.28 . 0.71 Oriolus oriolus 0.47 . . 0.80 1.45 0.21 0.40 0.27 0.49 0.05 Otus scops . . . 0.04 . 0.05 . . 0.13 . Parus major 0.90 0.81 1.49 1.76 3.61 3.33 0.29 . 3.44 0.24 Passer domesticus 8.97 . . 0.04 . 0.11 0.43 0.06 0.88 0.14 Passer montanus 0.23 . . . . . . . . . Periparus ater . 8.72 7.11 4.18 3.19 0.39 . . . . Petronia petronia . . . . . . 0.11 0.83 0.44 2.26 Phoenicurus ochruros . . . . . . . 0.21 . 2.80 Phoenicurus phoenicurus . . 0.08 0.04 0.83 0.05 0.10 . 1.02 0.05 Phylloscopus nitidus 0.04 2.79 2.39 0.44 0.28 1.07 0.05 . 1.46 . Phylloscopus sindianus . . . 0.06 . 0.19 . . . . Pica pica 2.06 . . . . 0.10 0.87 0.28 1.59 0.09 Picus viridis . 0.58 0.50 0.34 0.34 0.25 . 0.11 0.57 . Prunella collaris . . . . . . . . . 0.07 Prunella modularis . 0.47 . . . 0.28 . . 0.22 . Ptyonoprogne rupestris . . . . . . . . . 0.22 Pyrrhocorax pyrrhocorax . . . . . . . . . 0.08 Pyrrhula pyrrhula . 1.05 0.08 . 0.08 0.05 . . . . Remiz pendulinus 0.23 . . . . . . . . . Rhodopechys sanguineus . . . . . . . . . 0.07 Saxicola rubetra . . 0.12 . . . 0.07 . . 0.05 Saxicola torquatus . . . . . . 0.05 0.11 . 0.19 Serinus pusillus . . . . . . . 0.06 . 0.90 Sitta europaea . 5.47 4.63 2.30 2.52 0.10 . . . . Sitta neumayer . . . . . 0.05 . 0.33 0.13 3.07 Streptopelia decaocto 0.13 . . . . . . . . . Streptopelia turtur 0.21 . . . 0.18 0.45 0.27 . . . Strix aluco . . 0.08 . . . . . . . Sturnus vulgaris 2.40 . . . . . 1.02 0.61 7.20 0.19 Sylvia atricapilla . 2.91 1.65 1.46 1.18 1.81 . . 0.51 0.09 Sylvia communis 0.13 . . . . 1.43 0.94 0.07 3.52 0.40 Sylvia curruca . . . . . 0.24 . . 0.51 0.49 Sylvia mystacea 0.04 . . . . . . . . . Sylvia nisoria . . . . . 0.20 . . . 0.07 Tachymarptis melba . . . . . . . . . 0.14 Troglodytes troglodytes . 5.93 5.04 2.30 0.76 0.15 . . 0.06 0.05


- 81 -

Turdus merula . 3.95 4.79 3.22 5.29 4.02 0.19 0.11 2.36 0.85 Turdus philomelos . 1.28 2.56 1.76 0.76 0.39 . . . . Turdus torquatus . . . . . . . 0.21 . 0.77 Turdus viscivorus . . 0.11 0.05 0.37 0.23 0.05 . . . Upupa epops 0.39 . . . 0.25 0.25 0.92 0.55 1.59 0.33

Annex 8: Relative abundance values (individuals/km) of each bird species per landscape type (including migrants and other non-breeders).

Species Caspian lowland

Natural forest

Slightly disturbed forest

Intermediate disturbed forest

Park-like forest

Shrubby woodland

Montane meadow belt

Montane semi-desert

Riparian forest

Rocky habitats

Accipiter badius . . . 0.25 . . . . . . Accipiter nisus 0.04 . . . . . 0.05 . . . Acrocephalus arundinaceus 1.03 . . . . . 0.07 . . . Acrocephalus palustris 0.05 . 0.44 . . . 0.14 . 1.13 . Acrocephalus schoenobaenus 1.77 . . . . . . . 0.13 . Acrocephalus scirpaceus 0.06 . . . 0.14 . . . 0.37 . Actitis hypoleucos . . . . . . . . 0.08 . Aegithalos caudatus . 1.51 1.16 0.73 0.17 2.11 . . . . Alauda arvensis . . . . . 0.05 1.31 1.49 . 0.80 Alcedo atthis 0.69 . 0.23 0.11 . 0.11 . . . . Alectoris chukar . . . . . 0.29 . 1.16 0.22 1.65 Anthus campestris . . . . . 0.05 0.05 1.05 . 0.38 Anthus cervinus . . . . . . 0.09 . . . Anthus pratensis . . . . . . 0.05 . . . Anthus spinoletta . . . . . . . . . 0.60 Anthus trivialis . . . . . 0.64 0.24 . 0.06 0.05 Apus apus 4.08 . . 0.29 3.67 0.68 0.76 1.82 0.51 3.24 Aquila chrysaetos . . . . . . . . . 0.11 Aquila nipalensis . . . . . . 0.05 . 0.06 . Aquila pennata . . . . . 0.11 . . . 0.05 Aquila pomarina . 0.12 . 0.04 . . 0.19 . . . Ardea cinerea 0.06 . . . . . . . . . Ardea purpurea 0.23 . . . . . . . . . Bubulcus ibis 0.04 . . . . . . . . . Bucanetes githagineus . . . . . . . 0.35 . 0.40 Buteo buteo . 0.23 0.17 0.08 0.09 0.21 0.72 . . 0.09 Buteo buteo vulpinus . . . . . . . . . 1.37 Buteo rufinus . . . . . . . . . 0.11 Calandrella rufescens 3.26 . . . . . . . . . Carduelis cannabina . . 0.08 . . 0.85 0.05 2.10 . 2.12 Carduelis carduelis 1.55 . 0.08 0.54 2.18 0.88 0.53 0.33 2.36 1.27 Carduelis chloris 0.17 0.12 0.83 1.07 2.35 0.64 . . . 0.09 Carduelis spinus . 1.28 2.81 0.88 2.27 0.15 . . . . Carpodacus erythrinus 0.04 0.12 0.08 0.04 0.09 0.56 . . 0.66 0.28 Cercotrichas galactotes 0.28 . . . . . . . . . Certhia familiaris . . 0.25 . . . . . . . Cettia cetti 0.13 . . . . . . . 2.55 . Chlidonias hybrida 0.23 . . . . . . . . . Ciconia ciconia 0.13 . . . . . . . . . Ciconia nigra 0.04 . . . . 0.05 0.05 . . . Cinclus cinclus . . . 0.16 . . . . 0.29 . Circaetus gallicus . . . . . 0.25 . . . . Circus aeruginosus 0.30 . . . . . . . . . Circus macrourus . . . . . . 0.07 . . . Coccothraustes coccothraustes . 0.12 0.58 1.00 2.02 0.34 . . . . Columba oenas 0.47 0.23 0.33 0.08 0.42 0.34 0.73 0.22 . 0.09 Columba palumbus . . . 0.04 0.08 0.05 . 0.06 0.57 0.05 Coracias garrulus 0.43 . . . . . 0.13 . . . Corvus corax . . . 0.08 . . 0.24 0.06 . 0.19 Corvus cornix 1.89 . . 0.15 1.01 0.59 1.65 0.11 1.08 0.09 Corvus frugilegus . . . . . . 0.13 . . 2.91


- 82 -

Coturnix coturnix 0.04 . . . . . 0.98 . . . Crex crex . . . . . . 0.09 . . . Cuculus canorus 0.69 0.70 0.58 0.70 1.65 0.45 0.11 0.28 0.15 0.71 Cyanistes caeruleus 0.52 0.81 1.57 1.99 2.27 0.88 . . 2.10 . Delichon urbicum 0.69 . . . . . 1.47 0.28 0.15 1.98 Dendrocopos major . 3.95 3.72 2.11 2.94 0.44 . . . . Dendrocopos minor . . . 0.23 0.17 . . . . . Dendrocopos syriacus 0.13 . . . . . . . 1.78 . Dryocopus martius . 0.19 . . . . . . . . Egretta garzetta 0.20 . . . . . . . . . Emberiza calandra 3.99 . . . . 0.49 7.48 0.72 1.85 0.33 Emberiza cia . . . . . 0.40 . 1.49 0.29 4.29 Emberiza hortulana . . 0.12 . . 0.07 . 0.28 . 0.22 Emberiza melanocephala 0.34 . . . . 2.50 6.38 . . 0.11 Eremophila alpestris . . . . . . . 3.20 . 0.88 Erithacus rubecula . 5.70 4.46 1.99 0.50 0.34 . . . . Falco subbuteo 0.09 . . . . . . . 0.16 . Falco tinnunculus 0.04 . . . . . . 0.17 0.45 0.47 Ficedula hypoleuca . . . . . . . . 0.06 . Ficedula parva . 3.72 3.55 1.80 1.26 0.05 0.05 . 0.19 . Ficedula semitorquata . 0.47 0.41 0.12 0.09 . . . . . Francolinus francolinus 0.04 . . . . . . . . . Fringilla coelebs 0.04 7.44 9.83 7.55 6.39 2.94 0.15 0.06 3.06 0.19 Garrulus glandarius . 0.35 0.08 0.11 0.17 0.44 0.05 . . . Glareola pratincola 0.11 . . . . . . . . . Hippolais icterina . 0.35 1.41 1.28 1.10 0.50 . . . . Hippolais pallida 2.48 . . . . . . . . . Hirundo rustica 8.33 . . 0.11 0.76 0.25 1.17 0.61 2.61 . Irania gutturalis . . . . . . . . . 0.20 Ixobrychus minutus 0.05 . . . . . . . . . Jynx torquilla . . 0.11 . 1.10 0.11 0.11 0.06 . . Lanius collurio 0.34 . 0.12 0.20 2.64 1.64 1.81 0.41 2.62 0.77 Lanius isabellinus . . . . . . 0.07 . . . Lanius minor 0.30 . . . . 0.36 0.20 0.27 0.98 0.05 Lanius senator 0.04 . . . . . 0.07 . . 0.07 Locustella fluviatilis 0.10 . . . . . . . . . Locustella naevia . 0.20 . . . . . . . . Lullula arborea . . . . 0.25 0.59 0.73 2.27 0.25 0.71 Luscinia megarhynchos 3.91 0.12 0.11 1.69 2.86 6.79 1.35 0.06 4.16 0.60 Melanocorypha bimaculata . . . . . . . 0.68 . 0.07 Melanocorypha calandra 0.15 . . . . . 0.83 . . . Merops apiaster 2.36 0.58 . 0.13 1.10 1.38 0.88 0.94 2.12 0.11 Merops persicus 3.22 . . . . . 2.55 . . . Milvus migrans . . . . . . 0.09 . . . Monticola saxatilis . . . . . . . 0.44 . 2.08 Monticola solitarius . . . . . . . . . 0.44 Motacilla alba 1.16 . . 0.08 . 0.10 0.78 0.06 1.46 0.24 Motacilla cinerea . 0.23 0.41 0.57 . 0.34 . . 0.13 0.09 Motacilla citreola . . . . . . 0.07 . . . Motacilla flava 1.33 . . . . . 0.33 0.06 . . Muscicapa striata 0.17 0.35 1.18 0.87 1.87 . . 0.07 1.23 . Neophron percnopterus . . . . . 0.06 . 0.06 . 0.05 Nycticorax nycticorax 0.21 . . . . . . . . . Oenanthe finschii . . . . . . . 0.61 . 1.54 Oenanthe hispanica melanoleuca . . . . . . . 0.06 . 0.33 Oenanthe isabellina . . . . . . . 1.49 . 0.05 Oenanthe oenanthe . . . . 0.08 . 0.10 0.39 . 0.80 Oriolus oriolus 0.60 . . 0.80 1.74 0.21 0.40 0.27 0.49 0.05 Otus scops . . . 0.04 . 0.05 . . 0.13 . Parus major 1.03 0.81 1.65 2.18 4.03 3.77 0.39 . 4.52 0.24 Passer domesticus 17.21 . . 0.04 . 0.23 1.63 0.06 1.09 0.14 Passer montanus 0.34 . . . . . . . . . Pastor roseus 3.22 . . . . . 0.40 3.42 . 3.30 Periparus ater . 9.77 7.85 4.44 3.53 0.54 . . . . Pernis apivorus . . . . . . 0.14 0.07 . .


- 83 -

Petronia petronia . . . . . . 0.11 0.94 0.51 2.64 Phoenicurus ochruros . . . . . . . 0.21 . 2.91 Phoenicurus phoenicurus . . 0.08 0.04 0.83 0.05 0.10 . 1.15 0.05 Phylloscopus collybita . . 0.25 0.15 0.17 1.32 . . 0.64 . Phylloscopus nitidus 0.04 2.79 2.39 0.44 0.28 1.07 0.05 . 1.53 . Phylloscopus sindianus . . . 0.06 . 0.19 . . . . Phylloscopus trochilus 0.82 0.58 0.45 0.52 0.33 0.13 0.59 . 1.24 0.11 Pica pica 3.69 . . . . 0.20 1.36 0.39 2.36 0.09 Picus viridis . 0.70 0.50 0.34 0.34 0.25 . 0.11 0.57 . Prunella collaris . . . . . . . . . 0.07 Prunella modularis . 0.47 . . . 0.28 . . 0.22 . Ptyonoprogne rupestris . . . . . . . . . 0.38 Pyrrhocorax pyrrhocorax . . . . . . . . . 0.17 Pyrrhula pyrrhula . 1.05 0.08 . 0.08 0.05 . . . . Remiz pendulinus 0.34 . . . . . . . . . Rhodopechys sanguineus . . . . . . . . . 0.13 Ripparia ripparia 2.10 . . . . . 0.13 . 2.05 . Saxicola rubetra . . 0.12 . . . 0.07 . . 0.05 Saxicola torquatus . . . . . . 0.10 0.11 . 0.24 Serinus pusillus . . . . . . . 0.06 . 1.13 Sitta europaea . 5.81 5.62 2.57 2.86 0.10 . . . . Sitta neumayer . . . . . 0.05 . 0.33 0.13 3.44 Streptopelia decaocto 0.17 . . . . . . . . . Streptopelia turtur 0.30 . . . 0.18 0.51 0.60 . . . Strix aluco . . 0.08 . . . . . . . Sturnus vulgaris 12.10 . . . . . 18.79 0.94 13.38 1.46 Sylvia atricapilla . 2.91 1.65 1.46 1.18 1.86 . . 0.51 0.09 Sylvia borin . . . . . . . . 0.13 . Sylvia communis 0.13 . . . . 1.43 0.94 0.07 3.69 0.40 Sylvia curruca . . . . . 0.24 . . 0.51 0.49 Sylvia mystacea 0.04 . . . . . . 0.06 . . Sylvia nisoria . . . . . 0.20 . . . 0.07 Tachymarptis melba . . . . . 1.59 . . . 0.19 Tringa ochropus . . . . . . . . 0.06 . Troglodytes troglodytes . 5.93 5.12 2.30 0.76 0.15 . . 0.06 0.05 Turdus merula . 4.07 4.96 3.26 5.38 4.12 0.19 0.11 2.42 0.94 Turdus philomelos . 1.28 2.56 1.76 0.76 0.39 . . 0.06 . Turdus torquatus . . . . . . . 0.21 . 0.82 Turdus viscivorus . . 0.11 0.05 0.55 0.23 0.05 . . . Upupa epops 0.43 . . . 0.25 0.25 1.07 0.66 1.59 0.33


- 84 -

Erklärung zur Diplomarbeit

Hiermit erkläre ich, die vorliegende Diplomarbeit mit dem Thema: „The breeding bird communities of the Talish mountains (Azerbaijan) and their response to forest degradation“ selbstständig verfasst und keine anderen Hilfsmittel als die angegebenen verwendet zu haben. Aus anderen Werken in Wortlaut oder Sinngehalt entnommene Inhalte sind durch Quellenverweis, auch für Sekundärliteratur, kenntlich gemacht.

Greifswald, den 30.06.2010

Michael Heiß

Annex 3: Bird community table of the nine breeding bird communities including parameter values and species abundance per transect (territory/km).

To Diploma Thesis ‘The breeding bird communities of the Talish mountains (Azerbaijan) and their response to forest degradation’ by Michael Heiß

Breeding bird communities of the Talysh mountains (Azerbaijan) … · 2010-07-15 · Introduction...

Documents

Transcript of Breeding bird communities of the Talysh mountains (Azerbaijan) … · 2010-07-15 · Introduction...