PHYTOSOCIOLOGICAL INDICATORS FOR ’S HABITAT USE AND OCCURRENCE IN

RABIA AFZA

1

DEPARTMENT OF BOTANY HAZARA UNIVERSITY MANSEHRA

2016

PHYTOSOCIOLOGICAL INDICATORS FOR PHEASANT’S HABITAT USE AND OCCURRENCE IN AYUBIA NATIONAL PARK

SUBMITTED BY RABIA AFZA PhD Scholar

RESEARCH SUPERVISOR PROF. DR. HABIB AHMAD Tamgha-e- Imtiaz Dean Faculty of Sciences Hazara University Mansehra

2

DEPARTMENT OF BOTANY HAZARA UNIVERSITY MANSEHRA 2016 HAZARA UNIVERSITY MANSEHRA

Department of Botany

PHYTOSOCIOLOGICAL INDICATORS FOR PHEASANT’S HABITAT USE AND OCCURRENCE IN AYUBIA NATIONAL PARK

By

Rabia Afza

3

This research study has been conducted and reported as partial fulfillment for the requirement of PhD degree in Botany awarded by Hazara University Mansehra

Wednesday 10, October 2015

4

Dedicated

To My Parents for their persistent support, inspiration and love

5

TABLE OF CONTENTS

INTRODUCTION...... 8 1.1 Biodiversity of Pakistan ...... 9 1.2 ...... 10 1.3 Study Area ...... 12 1.4 History of the Ayubia National Park ...... 13 1.5 Geomorphology of the Park ...... 13 1.6 Climate and Biogeography...... 15 1.7 Associated human communities with ANP ...... 18 1.8 Land tenure and resource ownership ...... 20 1.9 Park Management ...... 22 1.10 Biodiversity Threats to the Park ...... 22 1.10.1 Timber Harvesting ...... 22 1.10.2 Fuelwood Collection ...... 23 1.10.3 Grazing and Fodder Collection ...... 23 1.10.4 Non-timber Forest Products ...... 24 1.10.5 Hunting and Poaching ...... 26 1.10.6 Killing and Poisoning of Common Leopards ...... 27 1.10.7 Forest Fires...... 29 1.11 Literacy and Services ...... 29 1.12 Avian Fauna ...... 30 1.13 ...... 31 1.14 World Distribution ...... 32 1.14.1 Habitat Preferences ...... 33 1.14.2 Conservation status...... 34 1.15 Distribution of Pheasants in Pakistan ...... 35 1.15.1 The ...... 36 1.15.2 The ...... 37 1.16 Multivariant statistical tools ...... 39 1.16.1 Hierarchical Clustering ...... 39 1.16.2 Ordination or gradient Analysis ...... 40 1.17 Species Diversity Index ...... 40 1.17.1 Shanon-Wiener Index ...... 41 1.18 Habitat Suitability Modelling Techniques ...... 41 1.18.1 Maximum Entropy Model ...... 41 1.18.2 Ecological Niche Factor Analysis ...... 42 1.19 Habitat suitability mapping ...... 42 1.20 Objectives of the study ...... 43 2 MATERIAL AND METHODS ...... 44 2.1 Data collection...... 44 2.2 Vegetation Analysis ...... 45 2.3 Data Analysis ...... 47 2.3.1 Hierarchical Cluster Analysis ...... 47 2.3.2 Ordination ...... 48 2.3.3 Species diversity and plant communities...... 48 2.3.4 Species-Presence only Data ...... 49 2.4 Habitat Suitability Modelling Techniques ...... 50

6

2.4.1 Maximum Entropy Model ...... 50 2.4.2 Ecological Niche Factor Analysis ...... 50 2.4.3 Ecogeographical Data Analysis ...... 51 2.5 Remote Sensing (RS) and Geographic Information System (GIS) data generation 50 ...... 51 2.5.1 Predictive Habitat distribution map for Kalij and Koklass Pheasant ...... 53 2.5.1.1 Determination of Spatial Relationship of Species by using GIS tool ...... 55 3 RESULTS ...... 55 3.1 Hierarchical clustering ...... 56 3.1.1 Pinus wallichiana – Viola canescens - Vibernum mullaha community ...... 56 3.1.2 Abies pindrow – Vibernum grandiflorum – Dryopteris ramosa plant ...... 60 community ...... 60 3.1.3 Pinus wallichiana – Fragaria nubicola –Indigofera heterantha plant ...... 61 community ...... 61 3.1.4 Ecological specificity in plant species ...... 65 3.1.5 Plant-Species Diversity, Richness and Evenness ...... 65 3.2 Forests Inventory...... 67 3.3 Ordination ...... 68 3.4 Phytosociological correlation of Pheasants with ecological communities ...... 70 3.4.1 Nesting sites in plant community I: Pinus wallichiana – Viola canescens - ...... 70 Vibernum mullaha ...... 70 3.4.2 Nesting Places of Kalij Pheasant ...... 73 3.4.3 Nesting sites of Pheasants in plant community II: ...... 76 Abies pindrow – Vibernum grandiflorum – Dryopteris ramosa ...... 76 3.4.4 Nesting sites in plant Community III: Pinus wallichiana – Fragaria nubicola ...... 79 –Indigofera heterantha...... 79 3.4.5 Nest architecture / Nidification...... 84 3.4.5.3 Correlation of Eco variables with Koklass and Kalij ...... 86 3.5 Predictive Modelling Techniques ...... 86 3.5.1 Maximum Entropy Model (Maxent) ...... 87 3.5.2 Ecological Niche Factor Analysis (ENFA) ...... 90 3.6 Geographic -Information System (GIS) and Remote -Sensing (RS) Data ...... 95 Generation ...... 95 3.6.1 Probability of Plant Community occurrence in ANP ...... 95 3.6.2 Aspect Value ...... 98 3.6.3 Elevation ...... 99 3.6.4 Compound Topographic Index ...... 100 3.6.5 Terrain Ruggedness ...... 100 3.6.6 Soil Adjusted Vegetation Index ...... 100 3.6.7 Slope Length Factor ...... 101 3.6.8 Slope (radians)...... 101 3.6.9 Stream Power Index ...... 101 3.6.10 Topographic position index ...... 101 3.6.11 Terrain Ruggedness Index ...... 102 3.6.12 Topographic Wetness Index ...... 103 3.6.13 Non Differential Vegetation Index ...... 103 3.6.14 Tasselled Cap Index ...... 103 3.6.15 Band 7 and Band 4 ...... 107 3.6.16 Standard Normalized difference vegetation index ...... 107 3.7 Predictive habitat suitability places of Pheasant in Ayubia National Park ...... 107 4 DISCUSSION ...... 111 4.1 Vegetation Analysis ...... 113

7

4.1.1 Floristic composition of Ayubia National Park ...... 115 4.1.2 Forests Inventory ...... 116 4.1.3 Phytosociological analysis in Pakistan ...... 117 4.2 Phytoecological factors effecting habitat selection of Himalayan Pheasant ...... 119 4.3 Pheasants of Pakistan ...... 120 4.3.1 Habitat occurrence of Koklass Pheasant ...... 121 4.3.2 Habitat utilization of Kalij pheasant ...... 125 4.4 Environmental gradient analysis ...... 128 4.5 Ground information system (GIS) and Habitat suitability ...... 128 4.6 Habitat suitability modelling ...... 129 4.6.1 Maximum Entropy Modelling ...... 130 4.6.2 Ecological Niche Factor Analysis ...... 130 4.7 Habitat suitability (HS) Maps ...... 133 4.7.1 Koklass Pheasant...... 134 4.7.2 Kalij Pheasant ...... 134 4.8 CONCLUSION ...... 135 4.9 RECOMMENDATIONS ...... 137 5 REFERENCES ...... 139 6 ANNEXTURE...... 164

Chapter 1 . INTRODUCTION

Pakistan, a land of 796,095 km2 is basically a subtropical country. It lies between 24 - 37o ' N and 61 - 75o ' E (Ashiq et al., 2010). The country is stretched from 0 m at Karachi to 8611 m at K2 (Khan et al., 2013). The country is divided into 9 major phytoecological zones, largely arid, with three-fourths receiving an annual precipitation of less than 250 mm and 20% of it, less than 125 mm of rain annually (Afza, 2006). Only about 10% of the area in the Northern Himalayan and Hindukush mountain ranges receives a precipitation between 500 mm and 1500 mm. Broadly Pakistan is a forest deficient country with a natural forest cover of nearly 5.2% that corresponds to 4.57 million hectares (Jan, 1993; Ahmad & Mahmood, 1998; Shah, 2011; Shehzad et al., 2014). Official statistics measures coniferous forest in Northern Pakistan as covering 1.93 million hectares and are identified as a major source of timber, resin, medicine and wild fruits (Ahmed et al., 2006; Tariq et al., 2014). Generally hyper-arid, arid and semi-arid climatic conditions (Fig 1.1) prevail in most parts of the country (Afza, 2006; Tariq et al., 2014). Nearly 70% of the land area of the country is arid with bare land and sparse scrubs at riverbanks and deltas, however; riverine and mangrove forests are present (Arshad et al., 2012).

8

At the medium elevation ranging from 500 m to 1,000 m subtropical forests are there, whereas the temperate forest, forming the backbone of national economy in terms of ecosystem services, water recharge source and the main source of timber and non-timber forests products (NTFPs) occupies the altitudinal limits of 1,500 m to 3,000 m. The temperate forests contributes 50% share to the whole of the forest area of the country (Tariq et al., 2014).

Fig.1.1. Forests cover map of Pakistan (Source: FSMP, 1992).

1.1 Biodiversity of Pakistan Pakistan has an idiosyncratic biogeoclimatic conditions (Arshad et al., 2013; Ahmed et al., 2015). Its unique geology, inclusive latitudinal coverage and massive altitudinal range supports a remarkable number of the world’s ecological regions (Olson & Dinerstein, 2002; Ahmad et al., 2015; Khan et al., 2015). About 60% of the country consists of elevated plateaus and mountainous terrains, whereas the rest of the area is lowland having an elevation of less than 300m (Anonymous, 1990). Amongst the world’s highest cold and hottest low areas, several are situated in Pakistan (Roberts 1991). Although the area of the country is relatively small but it supports most important biomes of the world (Dasmann, 1972; Jepson & Whittaker 2002; Roberts, 1970). These ranges from

9 the sandy beaches, blue lagoons and the mangrove forests on the Arabian coastal areas in the south to the high mountain tops, picturesque valleys and the endless glaciers in the north, where the three mountain ranges of the world– the , Hindu Kush and Karakoram ranges meet (Owen & England, 1998; Owen, 1996). In between there are vast sandy deserts, isolated plateaus, scrub forests, the highly fertile and productive Indus basin, irrigated plains, riverine tracts, subtropical and temperate forests, alpine pastures and permanent snowfields (Khan et al., 2013). Based on such geo-climatic variations, numerous vegetation types have been identified in the country (Beg, 1975; Robert, 1991). The biodiversity of the country comprises a combination of Indo- Malayan (Indus plains located east of the Indus River) and Palearctic (Himalayas and Western mountain regions) elements, with some groups also containing forms from the Ethiopian region i.e. dry southwest along the coast of Makran and Thar Desert (Pilbeam et al., 1977; Ali & Qaiser, 1986; Ali, 2008; Ali, 2010; Shah, 2011). The country is represented by more than 6000 vascular plants, 195 mammals, 668 species of , 192 reptiles, 22 amphibians, 788 marine and 198 freshwater fishes, twenty thousand species of insects and terrestrial and freshwater invertebrates; and 700 species of marine invertebrates have been documented. Among these 6 mammals, 9 amphibians, 18 reptiles, 41 butterflies and 29 freshwater fish are endemic to Pakistan (Chaudhry et al., 2012; Qaimkhani, 2009). Similarly, 20 mammals, 25 birds, 6 reptiles are threatened with extinction (Qaimkhani, 2009; Ali et al., 2011; Subhani et al., 2012) but according to IUCN Red List (2010) of threatened species indicated 23 species of mammals, 26 species of birds, 10 species of reptiles, 33 species of fishes and 15 species of other invertebrates are threatened with the risk of extinction in the country.

1.2 Khyber Pakhtunkhwa The Khyber-Pakhtunkhwa (KP) Province wherein the study area included is an important province with respect to biodiversity of Pakistan. The province is and 74°7' E. The total ׳and 36°57' N. latitude and 69°16 ׳located between 31°4 area of the province is 74521 km2. Khyber-Pakhtunkhwa is bounded by to the north-west, Gilgit Baltistan province to the north-east, Azad

10

Kashmir to the east, Federally Administrated Tribal Areas (FATA) to the west and south, part of Baluchistan and Punjab to the south and Punjab and the capital – to the south-east (Fig. 1.1). The province is divided into 24 districts. The current population of the province is over 22 million (Annoymonous, 2010b). The province has diverse landscape with dry rocks and vast barren plains in the South, low hills in the middle and high mountains and green plains in the north. The lesser Himalayas form its western corner (Alam & Ali, 2010; Ali & Qaiser, 2010). The altitudinal variation ranges from 300 m in Dera Ismail Khan (Badshah et al., 2014) to 7,690 m at Tirichmir, located in the northern part of the province. The major rivers include Indus, , Swat, Panjgora, Kabul, Bara, Kurram, Tochi, Gomal and Zhob (Farooq, 2011; Afza et al., 2014). Khyber Pakhtunkhwa province is represented by diverse habitats. These diverse habitats support a wide variety of biota (Malik, 2004; Ahmad & Sumaira, 2007). The Khyber Pakhtunkhwa Province is represented by about 98 species of mammals, and 48 species of reptiles 456 species of birds (Anonymous, 2010a). The 3 endemic mammals of Pakistan i.e. Woolly flying squirrel (Eupetaurus cinereus), Vole (Hyperacrius wynnei) and Indus blind dolphin (Platanista indi) are also reported from this province. Similarly, 2 species of lizards i.e. Cryptodactylus chitralensis and C. mintoni are endemic to the province (Malik, 2004). The mountains of the province are considered as important centres of plant endemism as they contain up to 90% of endemic species (Alam & Ali, 2010; Amjad et al., 2014). Although Khyber Pakhtunkhwa has an impressive diversity of wildlife, it has already lost some wildlife species and the populations of carnivores and herbivores is in decline, many species are considered to be have become endangered and some are on the verge of extinction. The different factors that are responsible for depletion of wildlife include socioeconomics, politics, legal and financial and administrative constraints, lack of community’s interest, illegal trade, and shortage of skilled manpower (GOP & IUCN, 1992; Malik, 2004; IUCN, 1994; Afza, 2006; Turner et al., 2007; Clements et al., 2010).

11

1.3 Study Area The area selected for PhD research project was Ayubia National Park (ANP), a classic example of a traditional National Park model, termed as a ‘Golden Era National Park by Eagles (2009). The ownership of the land and resources of the Park is with the Government of Khyber Pakhtunkhwa. According to the IUCN criteria, the Park falls into Category V, under Protected Landscape 15 (Farooq, 2011). Various villages and hamlets surround the Park and the local population is dependent upon the Park resources for grazing of livestock and collection of firewood, fodder, medicinal plants and edible wild vegetables and mushrooms. Ayubia National Park was established in 1984 under the provisions of the Khyber Pakhtunkhwa Wildlife Act of 1975, for the fortification and perpetuation of its outstanding panorama, flora and fauna in the natural condition of the area. According to the Wildlife Act, the National Park is offered to the public as source or refreshment, education and exploration, subject to such restrictions as imposed by the Park management. Likewise, all the developmental works and forestry activities are carried out in the Park sustainably not to impair the Park objectives. According to the Wildlife Act (section 16 (4)) the following are forbidden in a National Park:

i. Shooting and trapping of any wild in a National Park or within 3 miles of its boundary radius ii. Any act which may disturb any animal or or doing any act which interferes with the breeding places like gun firing and poaching

iii. Wood cutting , tapping and burning

iv. Damaging and destruction of trees by lopping, pruning or collecting /removing any plant from the Park,

v. Land clearing / breaking for cultivation or for any other purpose, vi. Polluting water flow source in and through the National Park.

Though, the government may, for scientific ambition or endurance of the National Park, authorize the aforementioned activities by the Wildlife departmental acts (Wildlife Act, 1975).

12

1.4 History of the Ayubia National Park Ayubia National Park was established in 1984 without any consultation of the associated communities. It is quite clear that when the government allocates certain land without taking into confidence the local communities, the local communities either the restrictions or get into violent conflicts with the government (Peterson et al., 2010; Dickman, 2010). The natures of such conflicts are continuously happening every time in terms of biodiversity conservation aspect of ANP. It was originally established on 1,683 hectares (ha) of land but the area was increased to 3,312 ha in 1998. The Park is declared within the Reserved Forests of Galliat. During the summer season, the Park attracts thousands of visitors, due to its cool climate and a vision of pristine nature – forest, streams and wildlife (Hamilton and Hamilton, 2006).

1.5 Geomorphology of the Park Geographically, the Park falls in Lesser Himalayas lying between 34°1' to 34° 3.8’N latitude and 73° 22.8’to 73° 27.1'E longitude. The Park represents Western Himalayan Subalpine Conifer forests, which is a component of the Western Himalayan Temperate Forests Ecoregion (Mani, 1974). This Ecoregion is globally recognized due to its unique biodiversity features that include it in the list of Global 200 Ecoregions of International Significance preserving the typical features of Sino-Japanese region (Mani, 1974; Afza, 2006). The elevation of the Park ranges between about ±1,050 m at Lahur to ±3,027m at Miranjani top (Afza, 2006). Mean annual rainfall has been recorded above 1500mm (with heavy winter snow), whereas mean annual temperature is recorded as 21 °C (Afza, 2006; Saima et al., 2009). Dungagali is the Park headquarter, situated at a distance of 43 Km from while, 30 Km from Murree. The Park is easily accessible through a construct road from Abbottabad and Islamabad via Murree, which also marks it the western boundary of the Park.

13

Fig.1.2. Satellite image showing physical features and boundaries of ANP (Source: WWF-Pakistan).

The Park is bounded by Namli Maira and Phalkot by the Reserve Forests and Bakot Forest Compartment No.6, Kuzagali, Darwaza Reserve Forest compartment-3 (ii) and Khanispur in the south. While the Birote Reserved Forest and Lahurkus village lies in the east. Similarly Bagan Reserve Forest of Compartment-10, Kalabagh and Nathiagali Location, Kundla and Toheedabad lie in the western parts of the Ayubia National Park. Petrography of ANP shows that the permanent rocks of Ayubia National Park are fourty millions to one hundred and seventy millions years old. The rocks are sedimentary, mainly limestone but also altering Shale and Sandstone (Calkins et al., 1975; Latif, 1976). The contemporary geological structure has resultant due to eclectic folding, trimming and reduction in connection with regional crystal distortion

14 mounting from the Indian sub continental plate that is present beneath the Europium plate. The foremost rock forms are slate, limestone, and a metamorphic succession of phylite, schist and granites with an impartial depth of soil minerals, proficient to support an average quality of Pinus wallichiana, Abies pindrow, Taxus baccata, broad leaved and a variety of medicinal plants and agricultural crops in the contiguous reserved forests (Negi, 2000; Adnan, 2011; Amjid et al., 2014). The bedrock in the Ayubia National Park is common with the rest of Gullies area is sedimentary comprising of limestone, shale and sandstone belongs to Margalla Hills and ranging in age from Triassic to Eocene. Limestone is found in combination with marl and shale (grey color, fine, medium - grained) that are rarely gigantic. The marl is grey/ brownish grey while the shale is greenish brown to brown in color. Moreover, rocks of the Kawagarh creation is the example of the Cretaceous time period, Lockhart limestone, Patala is a development of Paleocene age, and while the Eocene time period formation are also visible in the area in the form of Chorgali type (Middlemiss, 1896; Hooker, 2003). The bedding of rocks is distinct with least signs of metamorphism. However, the beds have different orientation due to heavy bending and lining up with inter phase of clay of several colours (Middlemiss, 1896). Soil in the Park is shallow and loamy, while in grazing areas the soil is shallow due to exposed bedrocks which indicating high biotic pressure. The soil of forest has upper thicker layer of humus layer, which has make it fertile and moist. The soil is predominantly clayey and at some places blend with sand and gravel is recorded. In depth soil layers get changed from shallow to relatively deep color, depending upon the slope. In the valleys bottom and mountaintop the soil- layer is profound in having the capacity to support suitable vegetation (rich mineral content). The soil on steep- slopes is either very shallow or entirely absent with decreasing capacity to support vegetation.

1.6 Climate and Biogeography The topography of Ayubia National Park is rugged with precipitous slopes. The area is divided into spurs creating small side valleys, having almost all aspects giving great physical diversity to the area accounting for a variety of habitats

15 for almost all kinds of wildlife. The main axis of the mountainous range is north-north-east extending to the south and south-east forming two major mountainous ridges extending to the peaks of Miranjani in the northeast and Mushkpuri in the central area of the Park with altitudes of 2980 m and 2820 m, respectively. The altitudinal variation ranges from 1050-3027 m (Chughtai et al., 1989; Thomas et al., 2004; Afza, 2006). The climate of the area is moist temperate with extremely cold and snowy winter and highly pleasant summer spell. In winter months the temperature falls below freezing point. The Park receives a mean annual precipitation between 1065 – 1500 millimetre (mm) and snowfall of about 1-2.5 mm. Most of the rainfall is received during monsoon period in the months July to September, while October is the driest month. Rainfall, temperature and humidity conditions of the area are highly conductive for rich vegetation growth (Afza, 2006). The Park carries one of the best intact examples of the moist temperate forests in the country, with a wide diversity of vulnerable plant and animal species (Farooque, 2002). While the maximum area of Park characterizing moist temperate vegetation pattern along with indicating features of sub-alpine (meadows) and sub-tropical (Pine forests) region (Afza et al., 2004; Hamilton & Hamilton, 2006). According to earlier reports, the Park supports a lavish variability of characteristic local plants, wherein about seven hundred and fifty seven vascular plants are reported. Conifers are the dominate forests trees while broadleaved are rather uncommon (Afza, 2006) major tree species are Blue Pine (Pinus wallichiana A.B. Jacks.), Chir pine (Pinus roxburghii Sarg.), Yew (Taxus baccata L.), Silver Fir (Abies pindrow (Royle ex D. Don) Royle), Spruce (Picea smithiana (Wall.) Boiss.), Deodar (Cedrus deodara (Roxb. ex D.Don) G. Don)). While broad leaved e.g. Horse Chestnut (Aesculus indica (Wall. ex Cambess.) Hook.)), Akhrot (Juglans regia L.), Oak species (Quercus incana W. Bartram, Quercus baloot Griff and Quercus dilatata Lindl. ex A. DC.), Maple (Acer caesium Wall. ex Brandis), Poplar (Populus ciliata Wall. ex Royle), Bird Cherry (Prunus padus L.) are dominant trees.

16

The Park also supports a wide variety of medicinal plants, including the Aconitum violaceum Jacquem. ex Stapf., Adiantum venustum D. Done , Androsace hazarica R.R. Stewart ex Y.J. Nasir, Ajuga integrifolia Buch.-Ham., Ajuga parviflora Benth., Asparagus filicinus Buch.-Ham. ex D. Don, Bergenia ciliata (Haw.) Sternb., Geranium wallichianum D. Don ex Sweet, Paeonia emodi Royle, Swertia alata C.B. Clarke, Viola canescens Wall., Valeriana jatamansi Jones, Berberis Parkeriana C.K. Schneid. Skimmia laureola Franch. etc. (Afza, 2006). Other key plants include various species of Morel mushrooms (Morchella esculenta (L.) Pers.ex.Fr.) and wild vegetables i.e. Kunji (Dryopteris ramosa C. Hope) C. Chr.), Mushkani (Nepeta connata Royle ex Benth.), Mirchi (Solanum surattense Brum.f. d), Tandi (Dipsacus inermis Wall.) are preferably collected from the Park (Afza, 2006). Sub-Alpine Meadows/pastures are found on comparatively gentle and steep slopes around the two highest peaks of Miranjani and Mushkpuri that exceed tree line. The subtropical pine forest ecotype is present at lower altitudes in areas dominated by Chir Pine (Pinus roxburghii) with broad-leaved plant species at altitudes ranging from ±1,050 to ± 2,000 m (Bakot compartment/ Khun).

Fig.1.3. Monthly data precipitation received by ANP during study period (source: Pakistan meteorological department).

17

60 49.42 50

40 32.48 30

20 11.18 10 4.03 0.14 0.12 0 Conifer forest Shadow Mixed forest Conifer forest Pastures land Water/ Wet conifer forest with shrubs / grasses soil and grasses

Fig.1.4. Land cover classes (%) in Ayubia National Park (Source: Saeed, 2008).

The key wild of the Park (includes Common Leopard (Panthera pardus), Rhesus Monkey (Macaca mulatta), Yellow Throated Marten (Martes flavigula), Koklass Pheasant (Pucrasia macrolopha), White Crested Kalij pheasant (Lophura leucomelanos Latham), Orange Bullfinch (Pyrrhula aurantiaca), Murree Vole (Hyperacrius wynnei) etc. (Farooq, 2011). 1.7 Associated human communities with ANP The total population of villages surrounding ANP is about 50,000 people, living in 8,333 households (avg. house hold size = 6) and annual increase in population @ 3% (Afza, 2006). There are twelve main villages surrounding the Park (Fig.1.5) that are Mallachh, Pasala, Kundlla, Darwaza, Moorti (Kuzagali), Riyala, Lahurkus and Khun Khurd. Some of the villages are relatively bigger and are divided into smaller hamlets. The ethnic composition is mixed with two major ethnic groups that are Karalls and Abbasi. Beside that other ethnic groups are Awan, Rajput Syed, Gujar, Mughals, and Turks. None of the individual from the communities or outside own any piece of land within the Park area but own communal as well as individual lands outside the Park boundary (Aumeeruddy-Thomas et al., 2004; Adnan et al, 2005; Farooq, 2011). Mallach is one of the biggest villages (Fig.1.6) and is comprised of 6 diverse settlements in Kanisan, Pata, Saire, Janswanra, Kala Band and Sokha Kus. On the basis of population of residential people, Janswanra is the major community settlement, followed by Kala Band, Sokha Kus, Sairi, Pata and Kanisan. Most of

18 the registered violation cases are against the populaces of Kala Band and kanisan for lopping of fire wood collection in the Park. Since 2000, about 13 casualties of female have been recorded during tree lopping only from Mallach village especially, Kanisan (Ahmad & Afza, 2014). Darwaza is another village, which comprised Nakhar and Dara. This is a low pressure village as compared to Mallach and Pasala as the people are mostly educated and having business or government jobs Pasala comprises Kundla, Badyar and Toheedabad settlements. Toheedabad is the largest hamlet (populated) which followed by Badyar and Kundla. The utmost rate of biotic pressure for fire wood collection in the Park in recorded from Touheedabad. Moorti is the smallest hamlet around the Park. Similarly, Kuzagali is another small village around the Park. Khun Khurd is comprised of four different hamlets Bari, Seri, Mandri and Retri. Bari is much populated hamlet followed by Retri, Seri and Lundi or Mandri.

Fig.1.5. Location map of Ayubia National Park

19

Fig. 1.6. A view of human settlements around Ayubia National Park

Riala is a large village and is comprised of two settlements of Khanispur and Mominabad (Fig.1.6). In terms of human population, Mominabad is the largest settlement. Most of the violation cases were against the inhabitants of Mominabad while Khanispur hamlets have educated people and not going to the forests for timber extraction. Lahurkas, another village surrounding ANP (Fig.1.6) comprises settlements of Moohra, Terhatti, Chalotay, and Thura. Among these four settlements Moohra is the largest hamlet. High intensity of biotic pressure is recorded from Lahurkas.

1.8 Land tenure and resource ownership

Land tenure stands for the relationship, whether legally or customarily defined, among people, as individuals or groups; with respect to land (Walpole et al., 2009). This land tenure system was created by the colonial powers and was inherited by the country in 1947. The land tenure in forests is determined by history, local customs, laws, rules, government orders and manipulations of politically influential families (Khattak, 2002). However; in Pakistan the land tenure system is very complicated and is considered to be one of the major causes for the gradual disappearance of forest trees (Jan, 1993). The first land settlement in Hazara was carried out in 1872 to 1874, whereby the forested areas were divided into two broad categories i.e., Government, State or Reserved Forests and Private, Communal or Guzara Forests. Currently the major land tenure in Galliat area is shown in Fig. 1.4. Reserved Forests is a category of forest lands where all acts are prohibited unless permitted. Such forests are

20 properly demarcated with boundary pillars. In Galliat tract, these forests are located mainly on the ridges above the habitations. In these forests local communities have very limited rights however; concessions are occasionally granted to the local communities for grazing, collection of firewood and fodder. The Guzara Forests are forest areas which are left around the Reserved Forests to meet the basic needs of the local communities and are privately owned, but still permission is required from the Forest Department for cutting any timber tree (Hamilton & Hamilton, 2006). Since, as these are managed by the Forest Department, so they also charge 20% in administrative charges. These forests have almost been cleared of trees, due to permanent dependence of the local communities for grazing their domestic animals and obtaining timber, firewood, fodder etc. Critics believe that with the degradation of Guzara Forests, the biotic pressure on the surrounding Reserved Forests is increasing. Cantonment Forests were carved out of the Reserved Forests of Galliat for establishing military summer stations. Cantonment Forests are declared in five different locations including Khairagali, Changlagali, Ghoradhaka (Ayubia), Kalabagh and Baragali. The management of this category of forests also rests with the Forest Department and these are free from all kinds of private rights. Location Forests are also managed by the Forest Department, and these are also free from private rights. However, the forest is either in the use of civil administration or is rented to the private sector for building residential areas and other business centres. Location Forests are declared in four different locations including Kuzagali, Nathiagali, Dungagali and Thandiani.

Ayubia National Park is part of the Reserved Forests; however; the concessions otherwise granted in Reserved Forests were withdrawn from the area, when it was declared a National Park. Both the Forest and the Wildlife Departments are involved in Park administration. Critics believe that such dual management within the Park area adds to the complexity of the management and creates uncertainties about the relative boundaries of the authority of both the sister organizations (Aumeeruddy-Thomas et al., 2004; Hamilton & Hamilton, 2006; Farooq, 2011). It is further believed that such narrow, short-term bureaucratic management approaches invariably turn into reactive treatment of symptoms

21 rather than more effective and efficient, largescale, long-term strategic actions guided by adaptive management (Farooq, 2007; Brunckhorst, 2013).

1.9 Park Management Since 1984, the Park has been managed by the provincial Wildlife Department under the Wildlife Act, 1975 that is responsible for the conservation of the Park resources against the utilization, for enhancement of ecotourism and promotion of public awareness, extension, as well as different prospects for research and rehabilitation of endangered or rare wild species (flora and fauna). Since its creation, this Park has been under heavy social pressure to cater to the daily needs of the local communities for firewood, timber, fodder, medicinal plants, wild vegetables and grazing of livestock. The collectors of firewood and fodder are mostly women and children and every year several deaths are reported due to falling off trees in a bid to cut branches (Aumeeruddy-Thomas et al., 2004; Afza, 2006).

1.10 Biodiversity Threats to the Park Ayubia National Park is facing a number of threats , the major threats is wood cutting , fodder collection and other minor forest products as well as grazing their animals in the Park. The poorly regulated collection of various Park resources is leading to the overall degradation of the resource base, which seriously affects the regeneration and tree growth all these biotic pressures pose serious threats to the Park resources. These threats are described below:

1.10.1 Timber Harvesting Ayubia National Park is simultaneously managed by the Wildlife Department and the Forest Department. This dual management often creates policy issues, which ultimately affect the Park resources. Such clash of interest is clearly visible in ANP, where the Wildlife Department is concerned with the conservation of flora as well as fauna of the Park, but the Forest Department is more interested in the timber harvesting and revenue generation. The scholars of environmental governance argue that such overlaps of administrative functions, along with contradictions in conservation and developments goals,

22 often lead to serious management problems (Smith et al., 2003; Brooks et al., 2006; Rands et al., 2010; Khan et al., 2015). Consequently, the conservation goals are compromised due to lack of coordination among concerned organizations which have competing mandates (Brunckhorst, 2013). It is believed that such sector-based decision-making is partially responsible for the problem of biodiversity loss (Daily & Matson, 2008; Nelson et al., 2009).

1.10.2 Fuelwood Collection During the harsh winter, survival of human life is much dependent on energy requirements as compared to other needs (Khan et al., 2015). According to local communities they have nothing to burn, except firewood in severe winter period. The fuelwood requirements vary from community to community. According to the local communities, their firewood consumption during summer is half of the overall consumption during winter. The consumption of fuelwood is greater in the communities which are located at higher elevations. The fuelwood collectors are mostly female and an average of 4-5 death causality is recorded per year during fuelwood collection. Moreover, some of the hamlets of this community are located near the Park and road sides; whereas others are located at lower elevation and it takes 5 to 6 hours for locals to reach to the Park. Besides the local community pressure on the Park for fuel wood, hotels and restaurants located around the ANP are also extracting fuel from the Park by illegal ways (Afza et al., 2004). This indirectly indicates that the enforcement staffs are not so strict in controlling the consumptive uses of firewood, because of the poor monitoring associated with the location.

1.10.3 Grazing and Fodder Collection Grazing of domestic animals and collection of fodder is illegal within the Park. However, women of the neighbouring communities normally collect fodder from spring to autumn and use the fresh fodder to stall-feed the livestock (Hamilton & Hamilton, 2006). Fresh fodder is obtained either from lopping or through harvesting of fresh herbaceous growth, specifically grasses. During severe winters, the domestic animals are fed with crop residues or by grazing the animals in the unmanaged areas around the habitation. The local

23 communities and especially the women are involved in these illegal activities which ultimately result in serious conflicts between them and the Park staff. Such fodder collection cannot be controlled in the circumstances, when there is no formal channel for negotiations between the two key stakeholders (Afza, 2006; Hamilton & Hamilton, 2006; Farooq, 2011).

1.10.4 Non-timber Forest Products Besides the firewood and fodder collection, some other minor forest products are also regularly extracted from ANP. These include various mushrooms, medicinal plants, wild vegetables, wild flowers, etc. Such NTFPs are important sources livelihood for the local communities. During the summer months, children may be seen all along the main roads, selling the headbands made up of wild flowers to the tourists visiting the area (Afza, 2006). The details of different minor forest products collected in ANP are as follows:

1.10.4.1 Mushrooms collection Various types of mushrooms are extracted by the local communities from the National Park during the spring season. Different species of morels are from the genus Morchella and the most common species that is extracted from the Park is Morchella esculenta, locally called Ghuchi. Theses mushrooms are mostly available in moist, shady habitats within the Park, where the deadwood is available, and that is one of the key reasons why the mushrooms are only extracted from the Park, because the deadwood cannot be found outside the Park boundary. These mushrooms are considered as the highest priced non- timber forest products (NTFPs) in the region and, therefore, these mushrooms have many collectors (Hamilton and Hamilton, 2006). A total of 38% of the collectors of the mushrooms including women (19%), men (10%), boys (5%) and girls (4%) from the surrounding villages (Afza, 2006). The mushrooms are mostly exported to Europe and, according to Hamilton and Hamilton (2006), about 99% of the production of these mushrooms are exported, whereas the remaining 1% is locally consumed for medicinal purposes, e.g., analgesic, aphrodisiac and for the treatment of rheumatoid arthritis. But the human impact reduced the biodiversity of mountainous

24 ecosystem (Gibson et al., 2011; Quetier & Lavorel, 2011; Sutherland et al., 2014), particularly disturbed the population of pheasants during breeding and hatching season in Ayubia National Park.

1.10.4.2 Medicinal Plants Collection Medicinal plants are important for ordinary people within Pakistan and about 50% of the population uses herbal medicines for treating their minor and in some case major diseases (Khan, et al., 2004; Waseem et al., 2005). As a result of such huge demands, there are over 25 large manufacturing companies in Pakistan which are involved in commercial production of herbal medicines (Khan et al., 2004; Qureshi et al., 2008; Ali & Qaiser, 2009). About 2000 medicinal plants species are known from Pakistan and among those 59 species are found in Ayubia National Park. Similarly, a number of medicinal

Fig.1.7. Biodiversity threats to Ayubia National Park plants are collected from the ANP (Aumeeruddy-Thomas et al., 2004; Afza, 2006; Hamilton & Hamilton, 2006; Tariq et al., 2014). The collectors are mostly women and children (Afza, 2006). However, according to the literature, the wastage is much more serious with regard to the medicinal plants due to improper drying (Aumeeruddy-Thomas et al., 2004; Hamilton & Hamilton, 2006). Moreover, due to intensive collection, a number of medicinal plants have gone extinct, besides ruining the habitat for the associated wildlife species

25

(Gibson et al., 2011). Some of the vital medicinal plants collected in ANP including Taxus wallichiana (Partal), Sinopodophyllum emodi (Bankakdri), Berberis lyceum (Sumbal), Bergenia ciliata (Bhut paye), Peonia emodi (mameekh) Valeriana jatamonsii (Mushk–e-bala) and Skimmia laureola (Naire) (Afza, 2006).

1.10.4.3 Wild Vegetables collection A number of wild vegetables grow in the Park. The neighbouring communities of the Park consume various kinds of wild vegetables mostly collected between April and the end of June. The most collected vegetables are, Teraxicum (hund), Dipsacus inermis Wall. (Tandi), Trifolium (Seengi), Dryopteris ramosa (Kunji), Nepeta laevigata, Rumex nepalensis (Hula), Malva salvestris (Panre), Morchella esculenta (Ghuchi), Solanum villosum Mill. (Soorangi) etc. (Afza, 2006).

1.10.5 Hunting and Poaching Today the native communities are in way involved in hunting of wildlife but before the creation of National Park or even earlier the people used to hunt (Bonnicksen & Stone, 1985; Tallis et al., 2008; Schulze & Mooney, 2012). During the last fourty to fifty years, four mammalian species including Selenarctos thibetanus (Black Bear), Muntiacus muntjak (Barking Deer), Moschus moschiferus (Musk Deer), Naemorhedus goral (Grey Goral) and one bird species, pheasant (Lophophorus impejanus), have been reported extinct from the area (Farooq, 2002). Similarly now Koklass and Kalij pheasants are under threats due to pouching (within the Park) and hunting in the territory of the Park. Dependant community also complained all over about the increase in the number of leopards and livestock losses due to predation. In instances there are reports of killing of leopards by the peoples as results of perdition on livestock and killing of Human beings. People killed 71 leopards either in self-defence or retaliation, and so far 29 humans were reported killed by leopards during the same period. This conflict between people and wildlife has recently been identified as a threat to wildlife and leopard’s habitat in and around the Park (Lodhi, 2007). It is anticipated that the Park still supports 38 threatened plant species, 23 rare and threatened butterfly species, the endemic Murree Vole

26

(Hyperacrius wynnei) and the Murree Hill Frog (Paa vicina) which was recently discovered in this area during specialized resource assessment studies of the Park (Farooq, 2002). Though no study has been done to determine the causes leading to these extinctions, direct persecution by human beings and habitat change are the main causes reported in the recently compiled management plan for the Park (Ahmad & Afza, 2014).

1.10.6 Killing and Poisoning of Common Leopards According to the data available with the Wildlife Department, WWF and survey from local communities, 30 individuals, have been killed and dozens other injured by the common leopards since the establishment of the National Park in 1984 till April 2015. The killed individuals were mostly women, girls and children who were probably easy prey for the man-eater leopards. Though, normally leopards attack livestock and pets of the local communities. Since the establishment of the Park (1984) till April 2015, the current study compiled about 873 cases of livestock depredation by leopards in the Park and the surrounding areas of during the current study but there is no compensation scheme for the damages inflicted by wildlife, specifically the common leopard, and the numbers of such incidences are also increasing, due to the increase in the number of these wildlife species. It is reported that during 2005, the common leopards of the ANP killed six women in the area, which created a lot of public outrage against the Wildlife Department and the man-eater leopards of the Park but for first time by efforts of WWF- Peshawar, the Government of KP compensation to the affectes (Lodhi, 2007; Farooq, 2011).

27

8 7 7 6 5 5 4 3 3 2 2 2 2 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0

Year

Fig.1.8. Number of people attacked by common Leopard in ANP, since the establishment of Park (modified source of Farooq, 2011)

1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014

Fig.1.9. Leopards killed by local communities in and around National Park (Modified source of Farooq, 2011).

The local communities consider leopards as a key threat to their life and livestock; in the absence of any formal compensation system the local communities do not miss any opportunity to kill the leopards of ANP. However, there is no data about the exact number of leopards killed in the area, but the current study through survey and data collected from wildlife department revealed that a total of 72 leopards were killed till April 2015 in and around the National Park.

28

1.10.7 Forest Fires Forest fire mostly breaks out during dry summer and autumn months. However, the fire problem is not as serious as in the subtropical Chir pine forest due to cooler climate of Moist Temperate Zone.

Fig. 1.10. Vegetation stand of mixed coniferous forest burnt in 2006 in ANP at 2621 m elevation.

The cause of fire may be accidental and deliberate (Fig. 1.10). Contrary to the conditions in sub-tropical Chir pine Zone fire incident are less common in the months of May and June. This is due to comparatively cooler temperature in these months and freshly sprouting grasses and herbs.

1.11 Literacy and Services Literacy rate is very low, especially among women, with an average literacy rate estimated is about 45.6% for both the sexes. There are 72 educational institutes surrounding present Ayubia National Park.. From the number of schools in the area it appears that education facilities in relation to population are adequate in comparison to other rural area of the country. Separate primary school for boys and girls are present in most villages of the area while high- level schools for boys and girls. About 43% of the people earn their living from labor, 6% are engaged in private business, while the share of agriculture is only 12%. About 8% are in services of different nature, 29% are engaged in miscellaneous jobs like transport etc. and only 2% work abroad (Afza, 2006). Most of the people from Khanspur, Mallachh and Pasala transmute to 29

Rawalpindi, Murree and Abottabad during the winter season for earning through business. Residents of the area mostly rely on agriculture, forest resources and tourism for their daily earnings. Most men have seasonal jobs in summer with average earnings ranging from Rs. 12000-20000 per month (Ahmad & Afza, 2014). There are two hospitals one each in Khanaspur and Nathiagali. Number of Government dispensaries is 3, one each in Toheedabad, Darwaza and Namli Maira. Number of private practioner/dispenser is three, two in Toheedabad and one in Riala. Medical parties for vaccination and inoculation of children keep visiting the area (Afza, 2006). All the villages at the periphery of the Park have piped water supply. The chief source of water supply is perennial springs. All the villages around Ayubia National Park have electric power supply available in almost all places. Apart from villages in Khanspur that rely mostly on Liquefied Petroleum Gas (LPG) and kerosene for cooking and heating, all other villages use firewood as the main source of fuel. The law and order situation in the area is very good and extremely friendly. Their economic dependence on tourism and important hill resorts, since the British times has made the people very friendly and peaceful (Ahmad & Afza, 2014).

1.12 Avian Fauna Birds are the unique and most fantastic creature of the earth. Birds have always been serving for their existence and aesthetic needs (Block & Brennan. 1993; Butchart et al., 2010; Barrientos et al., 2014). Their excessive exploitation leads many species with the risk of extinction (Butchart et al., 2010; Miller, 2010; Chatterjee, 2015). There are more than 9000 species of living birds, in which almost 150 having become extinct after the arrival of humans (Rodrigues et al., 2004; Acharya and Chettri, 2012). The earliest known species of birds (class Aves) is Proto Aves about 200 million years ago (Moreau, 1934; Kirby et al., 2008; Butchart et al., 2010). Pakistan hosts a striking number of species of resident and migratory birds. Among them many native species are relatively common as avian fauna of Pakistan is primarily Paleoarctic in nature, especially in the winter spell with an incursion of migrant/refugee bird species.

30

As a whole 668 bird species are documented from Pakistan, of these specific to Oriental (36.6%) and Palearctic or Holarctic (63.4%) affinities and less than 0.5% are studied as truly cosmopolitan or pan-sub-tropical (Grimmett et al., 2008). The intrusion of winter visiting birds from north breeding domain/lands, or summer breeding visitors composed of the Indus Plains and northern mountain to hotter southern latitudes (Ali, 2005; Chaudhry et al., 2012; Bibi & Ali, 2013). Among all the total 668 Pakistan’s avian fauna, about 30% of the birds visit Pakistan for one year times period as long distance refugees, about 43% are (Palearctic) species visiting the country only for breeding and 28% are routine winter visitors, which breed extra-limitally and mainly in trans- Himalayan northern regions (Ali & Ripley, 1987; Ali et al., 2011). About one third of the bird species in Pakistan use wetlands for food, shelter, and (or) breeding (Chaudhry et al., 2012). However, the birds that visit or breed in poorer quality habitats do not contribute to a sustainable population through the years (Ali et al., 2011). Ayubia National Park is a hotspot for Western Himalayan moist-temperate biodiversity but a very little is known regarding the biodiversity of the Park. Up till now no proper maps based on the geographical distribution and habitats of these species have not been developed. Hence scientific endeavour was important as PhD project to elaborate the vegetational profile of the Park and correlate its impact on the availability and distribution of Pheasants in the area.

1.13 Pheasants Pheasants are large, ground dwelling the wherever the males often with brightly coloured and inhabit diverse habitats in the tropical and temperate forests of Asia and Africa. Taxonomically, they represent the family of the order . Among 52 species of pheasants from 16 genera have been documented to date, 51 species survive and are present (Gatson, 1996; Krupnick & Kress, 2003). Fascinatingly, 50 of them are Asian in origin, the unique Congo (Afropavo congensis) restricted Virgin forests of central east Congo basin (McGowan et al., 1995; Johnsgard, 1999; Chatterjee, 2015).

31

1.14 World Distribution Pheasant species survive from sea level to 4,200 m and can be clearly separated in to two clusters; high elevation Pheasants, which inhabit the snowbound habitats of the Himalayas and the higher altitudinal ranges of Japan, and Taiwan. However; the low altitude species needs heat and safegaurd to endure winters with subzero temperatures (Beebe, 1936; Howman, 1993). These species are recorded from Indonesia, Malaysia and Philippines. Red Jungle Fowl is another important member of the pheasant’s family which inhabit from the middle to the upper limits of the altitudinal delineation (Howman, 1993). Pheasants have inhabited from Flores, which is in the east of Java, at about 8° South, through the equatorial forests of the Thai-Malay Peninsula, to northeastern China at about 50° north. Pheasants also occur all along the Himalayan chain, and extends as far east as Taiwan at 121° east, and Japan at 145° east (Anonymous, 1994). The Himalaya is the tallest and newest mountain range in the world, spreads over 2400 Km2 in bow like from northwest to southeast and covers 150 Km2 to 250 Km2 in girth. This group is Asian in their native spreading’s, excluding , which is endemic to the Democratic Republic of Congo of central Africa ( Sclater,1858; Howman 1993; IUCN 1994a; Holt et al., 2013).

32

Fig. 1.11. Wallace hotspot distribution map for pheasants.

1.14.1 Habitat Preferences Pheasants have a vast range of habitat preferences and most pheasants inhabit dense, wooded forests. These habitats range from lowland tropical rainforest and montane tropical forests to temperate coniferous forests. Examples of some of the pheasant species which are found in these forests are Mountain Peacock Pheasant and Western Horned . Some pheasant species prefer open habitats of sub-alpine region (i.e. blood pheasant), alpine pastures (i.e. Chinese Monal), and savannas or grass land (i.e. ) (Beebe, 1936; IUCN, 1998). These species are distributed in wide range of area and occupy variety of habitat types ranging from lowland tropical rain forest (e.g. Crested fire back Lophura ignita) to alpine meadows (e.g. Chinese Monal Lophophorus lhuysii). Koklass Pheasant (Pucrasia macrolopha), (Tragopan melanocephalus) have not been extensively studied in their natural habitat and same is the case of Kalij Pheasant (Lophora leucomelana) (Miller, 2010). Due to the charismatic features, their imperative ecological features in the high altitude ecosystem is indicator of the forests health, as key prey base for predatory birds and mammals and for adverse human impacts

33 on their ecosystems (Nawaz et al., 2000; Malik, 2004; Bhattacharya et al., 2007). They are interpreted as idiosyncratic avian family of the Himalaya (Kaul, 1989; Hussain & Sultana, 2013; Kaul, 2014). Their huge body size and luminous plumage have been the providing reasons for hefty hunting pressure resulting in many species being vanished from their native habitats and have delimited some of populations into fragmented pieces (Johnsgard, 1999). Pheasants are also known as game birds and associated with the social and religious values of Asians and European ( and Quails).

1.14.2 Conservation status Due to habitat fragmentation and disturbance, all pheasant species are susceptible in most of their native habitats. They are endangered, threatened or vulnerable. According to the IUCN Red- Data Book, more than one third of the total species of pheasants are legally reported with risk of extinction from their natural environments. As a whole 57% (29) species of the whole group are listed as critically endangered, threatened or vulnerable in the Mace-Lande threat category, provided by the IUCN (International Union for Conservation of Natural Resources) and SSC-Pheasant Specialist Group (McGowan & Garson, 1995; McGowan & Madge, 2010). The magnitude of threatened species is due to hunting and poaching is at the highest than in any other avian family, with a probability of extinction for many species within the next 100 years (McGowan & Madge, 2010). The appraisal and valuation survey of conservation status for pheasants and partridges in Southeast Asia revealed substantial decline in its population (Del Hoyo et al., 1994). The numbers of pheasants are continuously declining due to killing and eggs damages especially during breading season, generally because of their charismatic plumage and proteinaceous meat (Nan et al., 2004; Bhattacharya et al., 2007). All pheasant species in their natural habitat face a number of threats; generally related to habitat loss, poaching, and diseases (Bhattacharya & Sathyakumar, 2007; Miller, 2010; Sukmul et al., 2010). In Khyber Pakhtunkhwa, the following are some the problems which are common in almost all habitat types, for all pheasant species, and in all geographic areas (Zaman, 2008):

34 a) Habitat fragmentation and destruction. b) Hunting and poaching. c) Hybridization and genetic drift, as well as change in social and biological behavior in captivity. d) Predation, disturbance and diseases in the wild and in captivity.

These threats are getting more severe with the passage of time and this should lead to the development of a comprehensive management strategy with the help of all stake holders to initiate conservation awareness and population monitoring programs for pheasants in their natural habitat. The term pheasants is specifically referred to those individuals of subfamily in which the birds spectacle more . The existing knowledge on the ecological distribution, biology and social behaviour of pheasants is reliable with the exception of the ring - necked pheasant. Opportunities are available for exploration and learning for wildlife specialist (avian scientists) and biologist. The data recoded from trophy hunters concluded that there is no scientific explanation on the status and distribution of pheasants that provides management tools for conservationists. Though in Himalayan region extensive assessments and surveys have been undertaken in the current decades but most of them were focused on population status and distribution, and still there is need to study the population dynamics and habitat selection based on ecological niche models (Kamal et al., 2011; Singh et al., 2011; Saqib et al., 2013; Selvan et al., 2013).

1.15 Distribution of Pheasants in Pakistan Pheasants are important environmental indicators and among total number of recorded species of pheasants in the world, five of them are endemic to Pakistani Himalaya and Hindukush. Out of the 51 species of Pheasants, 20 (39%) are endemic to the Himalayan region, which include the genera of Lophura, Pucrasia, Ithaginis, Tragopan, Lophophorus, Catreus, Crossoptilon and Polypectron and of which five species are native to Pakistan that are only found in Khyber Pakhtunkhwa (Nawaz et al., 2000). Khyber Pakhtunkhwa (KP) harbours five species of pheasants viz. Koklass (Pucrasia macrolopha), Monal (Lophophorus impeyanus), Western Tragopan

35

(Tragopan melanocephalus), White Crested Kalij Pheasant (Lophora leucomelana), and Cheer Pheasant (Catreus wallichii), found mostly in remote areas of Hindukush Himalaya (Shafique, 1999; Zaman, 2008). Out of these five species, the Western Tragopan is one of the most magnificent pheasants in the world. Among the 5 species of Himalayan pheasants Koklass (Pucrasia macrolopha), White Crested Kalij (Lophora leucomalana), Monal (Lophophorus impeyanus), Western Horned Tragopan (Tragopan melanocephalus), and Cheer (Catreus wallichii) are native species of KPK forests. Monal pheasant is locally extinct from Ayubia National Park; the Tragopan melanocephalus is threatened; the population of Cheer and Kalij are declining; and Koklass is abundant but needs simple protection (Johnsgard 1999; Nawaz et al., 2000; Zaman, 2008).

1.15.1 The Koklass Pheasant Koklass (Pucrasia macrolopha) is a medium sized montane dimorphic pheasant of 11-14“ (inches) in size (Baker, 1928; Severinghaus et al., 1979). In both sexes the head is completely feathered with lanceolate body feathers. This species is generally associated with montane forest (conifers and woodland). The chestnut breast, black head, and white patches on the sides of the neck are distinctive, and the female also has a unique whitish-patch on neck side. Both male and female have slightly elongated (blackish to brownish) tails. The male's call in spring and summer is a lurid pok-pok .….. pokias, voiced mainly during morning and evening times (Khan & Shah, 1982; Nawaz et al., 2000). Koklass is a very shy bird therefore is very difficultly observing it in the wild (Severinghaus et al., 1979; Hussain & Sultana, 2013). Just a single species is recognized of Koklass.

36

Fig. 1.12. Distribution map of Koklass pheasant in Pakistan.

The normal altitudinal range of this species is ranging between 2200 to 2500 m while in exceptional cases they extended the suitable habitat below 2500 m (Gaston et al., 1998) because the amount of snowfall is much more in western Himalayas as compare eastern Himalaya. Therefore the bird has a choice to move down in snow fall (Lawin & Lawin, 1984; Roberts, 1991). In Koklass breeding season generally started from April to June (Baker 1930) but some time extend to July depends upon the melting of snow and other environmental conditions (MacArthur & MacArthur 1968; Zaman, 2008). The most common clutch size from 5 - 7 mentioned by Baker (1930) and a clutch size 6 was reported by Hownan (1979) in wild while in captivity the average clutch size 9-12 is normal. In Pakistan two subspecies are found in KPK and Azad Jammu (AJK): i).Western Koklass (Pucrasia macrolopha castanea ) ii).Kashmir Koklass (Pucrasia macrolopha biddulphi ) According to Malik (2004), Koklass is not a high value bird because its plumage less inclined to trap as compare to other pheasants but the only threat to their survival is poaching.

1.15.2 The Kalij Pheasant In Khyber Pakhtunkhwa, Kalij is found in the southern regions from Swat (Malakand division ) to Kaghan, Siran, Valleys to Ayubia National Park of

37

Hazara division (Nawaz et al., 2000; Zaman 2008) and also in subtropical, and lower spread of temperate conifer forests and sub humid forests (Bhagnotar, Ghari- Habibullah, Kawai , Shinkiari and Shogran) recorded by Mirza ( 1978). White Crested Kalij is also distributed in moist temperate forests in Azad Kashmir. According to Nawaz et al. (1991) at least 30 breeding pairs are existing at the lower spreads of ANP on southern aspects.

Fig. 1.13. Distribution map of Kalij in Pakistan

There are nine subspecies of Kalij Pheasants (Lophora leucomelana) worldwide and the White-crested Kalij Pheasant (Lophura leucomelanos hamiltonii) is the only one of them found in Pakistan. This species is native to the Himalayas in parts of Northern , as well some areas of Western and northern Pakistan (KP). According to Delacour (1977), Kalij is found in evergreen and deciduous forests up to 3,300 m elevation. The average weight of male Kalij ranges from 0.5 to 1 kg, whereas female ranges from 0.8 to 1.1 kg (Ali and Ripley 1987) while the subspecies (Lophora leucomelana hamiltoni) having an average weight of the adult male as 2.4 lbs and adult female has 2 lbs (Baker, 1930). The eggs colure of this species is similar color and shape to the eggs of domestic chickens, 48.7 x 37.3 mm average dimension and 37.4 g weight (Baker 1930). Kalij is a dimorphic bird species. The Adult males have a prolonged head with a black crest that give purplish blue colour in sunshine. Kalij has brownish or bluish beak (Robert, 1991). The adult female is smaller as compared to the male

38 with short tail. During threat or danger, a drumming sound is produced by wing-whirring, and the species' alarm call is a repeated whoop-keet-keet- keet (Baker, 1930). Kalij preferred termites and insects, herb, small seeds, and also snakes and lizards. Sometimes tuberous roots are digging out (Baker, 1930). They migrate in winter period to warm and snow free sites, travelling to a large distance (Miller, 2009; Gaston et al., 1981) and generally select a medium size tree for perching (Baker, 1930). The population explosion, intrusion, poaching, and habitat destruction, pheasants are facing a high threat of extinction because along with ecological importance of pheasants, they have also fascinating values that are mostly providing a reason for their decrease in population due to poaching in their intrinsic habitat (Nawaz et al., 2000). There is a lack of in-depth research, effective survey protocols, and database on the basis of which estimate of wild population cannot be ascertained. The present study was made to identify phytosociological indicators for pheasant’s habitat use and occurrence in Ayubia National Park which hosts two species of pheasants i.e. Kalij and Koklass, while Monal pheasant get extinct from the area. Field studies were conducted in 2012-2014 from April to December. To find out the phytosociological association of pheasants with vegetation types of Ayubia National Park, phytosociological and vegetation analysis were conducted and established an association of plant species with Pheasant’s nests.

1.16 Multivariant statistical tools To classify the vegetation of an area, different clustering methods and algorithms exist. There is no single way to decide which is best (Blashfield, 1976) i.e. depend on the choice of methods used (Kent, 2011). The results should therefore; be seen as ways of viewing the data, rather than showing real biological structure. One way reduce the subjective nature of results is to try a range of different methods and see if the results are similar for each (Milligan, 1979; Beals, 1984; Oksanen et al., 2007).

1.16.1 Hierarchical Clustering

Hierarchical clustering is widely used method for grouping a set of species in the studying area by assigning species to its own cluster and then algorithm

39 proceeds iteratively (Bray & Curtis, 1957). Formation of a single cluster comes in to existence when Joining of two closely related groups at each stage is possible therefore; at each step distances amongst clusters are recomputed by the update formula of Lance-Williams dissimilarity. A number of different clustering methods are available for analysing dissimilarities including Ward’s Method “hclust” to find out the compact spherical clusters (Ward’s 1963) among different sites, using the default setting in R-Software (Vegan package). The concept of characterization of the identified vegetation (ecological) communities was based on fidelity and constancy (Kent and Coker, 1992; Clarke, 1993; Bassille et al., 2008).

1.16.2 Ordination or gradient Analysis Gradient analysis or ordination is a complementary technique in multivariant analysis for clustering of analytical data. Ordination is one of the extensively used methods that make an effort to show the relationships between ecological communities and with environmental variables (Cormack, 1971). It is the grouping of species or samples along diverse gradients, and represents samples and species relationships in a decisive way in a lowdimensional space. There are various ordination techniques available e.g. NMDS (Nonmetric multidimensional scaling), PCA (Principle components analysis), RDA (Redundancy analysis), CA (Correspondence Analysis) and Bray-Curtis ordination. The NMDS ordination is performed in vegan package and correlation between fitted vectors and ordination values is calculated (Oksanen & Huttunen, 1989) by using the stress plot function of vegan (R-Software). The nonmetric multidimensional scaling (NMDS) ordination offered a graphic illustration of vegetation clustering (Podani, 2006) and is reported to be suitable when floristic (high turnover of species) and environmental gradients are inclusive (Kenkel & Orlóci, 1986).

1.17 Species Diversity Index

40

1.17.1 Shanon-Wiener Index There are large numbers of recommended indices of diversity to deliver vital statistics about infrequency and prevalence of species in a group and is hence a significant method for community study. For this purpose the most frequently used index is the Shanon-Wiener Index proposed by Shannon and Wiener in 1949 and represented by H', t o combine species richness with relative abundance. Species richness is a total sum of the number of the plant species in a quadrat or community. Evenness or equitability indices are used to standardise relative abundance of species within the sample plots and in the community. The higher value of E indicating the evenness, i.e. more even the species will be distribution in their inside the community.

1.18 Habitat Suitability Modelling Techniques

Species distribution modelling tools are becoming increasingly popular in ecology and are being widely used in many ecological applications (Phillips, 2007; Elith et al., 2009; Peterson et al., 2011). These models establish relationships between occurrence of species and biophysical environmental conditions of the study area (Philips et al., 2006; Warren & Seifert, 2011). A variety of species modelling approaches are existing for the prediction of conceivable or potential habitat for a studying species (Elith et al., 2011; Wisz et al., 2008). However, comparatively few predictive models have been used for rare and endangered plant species (Engler et al., 2004) and there are fewer examples of studies using small sampling units (Pearson et al., 2007).

1.18.1 Maximum Entropy Model Maximum species-distribution modelling techniques are sensitive to sample size (Elith and Leathwick; 2009) and therefore; may not accurately predict the habitat distribution patterns of species (Parker, 1999; Phillips & Dudík, 2008; Peterson, 2011). The maximum entropy model (MAXENT) is based on presence only and is the most unconstrained model (Khanum et al., 2013; Chetan et al., 2014; Isner, 2014) The Maxent calculation are based on presence only data to predict the distribution of a species by supporting the maximum entropy theory and, is used for estimating species distribution and occurrence

41 probability from closest to uniform. Maxent is influenced by a number of parameters (Phillips et al., 2006; Elith et al., 2011) where dependencies are explained by simple functions resulting from environmental variables, called features. In Maximum entropy, the complexity of dependencies is controlled by the choice of feature types, and by regularization (parameters), which prevent Maxent from matching the input data too closely (over fitting). The dataset is based on presence-only data resultant from overall data collection. The true distribution of a species is represented as a probability distribution p (non-negative value) over the set X of sites in the study area.

1.18.2 Ecological Niche Factor Analysis Species never live in isolation and individual species only succeed within conclusive ranges of environmental conditions based on ecological niche theory. Whenever more than one species live together, biotic interactions may considerably affect their niches. The presence of a grander competitor may prevent a species from occupying some part of its niche, leading to a truncated or even bimodal niche. The Exploratory analyses required a preliminary for modelling analyses as they lead to select the variables of interest to model the habitat e.g. the ecological-niche factor analysis (ENFA) described by Hirzel et al. (2001). The ENFA is factorial analysis and is based on directions in ecological space bases on marginality (one axes) and specialization (several axes) and is an effective method for searching species distribution based on the Hutchinson’s (1957) ecological niche concept. According to Hutchinson (1957) “a niche is a hyper volume in the multidimensional space of ecological variables wherever a species can with stand its presence”. Therefore; niche based modelling require to examine different ecological attributes and alterations in a species niche requisites (Hirzel et al., 2007).

1.19 Habitat suitability mapping The prophecy of species distribution from limited data is highly focused on the study to acquire consistent maps. Numerous statistical analyses have been recently developed to provide the probability of occurrence of wildlife species (habitat selection) using presence-only with the help of Eco geographical

42 information through instant increase in the application of spatially explicit habitat models elsewhere. The ecological niche factor analysis calculates habitat suitability maps which obliquely demonstrate species possible distribution with no absence data by comparing the species reaction to different environmental variables in the entire study area. According to Grinnell (1917) no two species regularly established the in a single fauna having the same niche relationship. Therefore this is a big challenge for a single species to co-occur and allot parallel ecological niche, which mostly results in exclusion of one of the co-occurring species. Therefore; in order to identify key environmental variables that determine a niche, is one of the most vital operations and relies on expert knowledge (Chahouki & Khalasi, 2012). Habitat modelling has been very effective for theoretical studies on species ecological niches as well as for defining and managing Protected Areas (PAs) and National Parks (Cains et al., 1969). Each species has a specialize niche where more than two species intermingling with each other’s niches (competition). Diminution in such interaction may occur due to seasonal variation, divergence in diet forms or because of specific habitat - variables spatial i.e. habitat and diet; and temporal or seasonal variables defines the ecological-niche of a species in a set environment (Hutchison, 1957; Boyce et al., 2002) along with populationdensity and the time -spent in a particular habitat and help in evaluating the conservation value of a niche (Braunisch & Sushant, 2008; Jiang et al., 2009). Lack of knowledge about the presence of species, their environmental and biological interactions and; protected areas management in majority of the cases becomes intricate. Therefore; there was a need to understand about the specific ecological niche of both species of pheasant in Ayubia National Park to identify their general habitat to address the conservation and management issues in future.

1.20 Objectives of the study

1. Analysis of habitat features and phytosociological communities in and around occurrence of Koklass and Kalij pheasants

43

2. Determine relationship of pheasant with phytosociological communities in Ayubia National Park as indicator of the floral and faunal health 3. Determine impact of deforestation on pheasant's population in Ayubia National Park 4. Evaluate the impact of anthropocentric pressures on Koklass and Kalij to determine various factors contributing to decline and the proportion of deforestation/impairment of phytosociology in it.

Chapter 2 2 MATERIAL AND METHODS

For getting insight of the biological resources, their potential and associated problems, Ayubia National Park was thoroughly analysed both for their available secondary information and practical scientific facts. The methodology used is summarized in this chapter.

2.1 Data collection For collecting social data, both individual and participatory approaches were used. Secondary information was collected through individual questionnaire survey while for group interviews; participatory approach was used by conducting semi structured interviews (Fig 2.1). The key stakeholders and target groups that directly and indirectly influence ANP was analyzed. Ayubia National Park is managed by both Wildlife and Forests Departments of Khyber Pakhtunkhwa province therefore; official of both departments and their expertise were used during research. The focus target group was female because they are going to the forests on daily basis for fuelwood and fodder collection. Men within the community were interviewed for collection of NTFPs such as Morchella and for pheasant’s hunting in the Park and adjacent areas.

44

Fig. 2.1. Interviewing local women and seasonal residents for their inputs in PhD research

2.2 Vegetation Analysis Field information was collected from spring 2012 to 2014, to cover the floristic diversity of the parts of ANP being frequently used by Koklass and Kalij Pheasant. The sample plots were positioned in the whole area of the Park (32km2) by using systematic methodology keeping in view the phytosociological and environmental correlation with pheasant’s nesting places in consideration. Fixed point sampling method was adopted and the size of each plot was 0.1 hectare (Fig.2.2) within circle having 17.84 m radius. The total number of plots taken were 130 and were equally divided in the Park area i.e. 1 plot within 20.5 ha (0.1 ha/plot).

17 .84 m radius / 0.1 ha area

Fig.2.2. Diagrammatic presentation of fixed point survey

45

For the phytosociology, 18 x 18 m2 plots were laid down to assess the number of trees under 0.1 ha area. However while line transect of 15 m, was laid for shrubby and herbaceous vegetation (Canfield,1941; Causton, 1988) were laid within 18 x 18 m2 circle of 0.1 ha plot for recording the DBH (Diameter at the breast height) data (Khan et al., 2015) . For equal distribution of plots and to avoid biasness, a grid was drawn on the known scaled map of the ANP. A map having scale 1:50000 was selected and a grid of 0.9 x 0.9 cm (20.5 ha/scale) drawn on it. Sample plots were taken in the centre of each grid.

Fig. 2.3. Data collection for vegetation analysis through line transects and DBH data collection in 18m circular plot.

46

Fig. 2.4. Different steps for plant collection and preservation.

The geographical position of every sampling plot was noted by applying Garmin-Global Positioning System (GPS) Receiver. Plant habits were noted within the sampling plots, the collected plants were identified from Herbaria at Department of Botany Hazara University, Mansehra; Pakistan Forests Institute (PFI), Peshawar, Pakistan; Museum of Natural History, Islamabad and the Department of Plant Sciences, Quaid-i-Azam University, Islamabad. The nomenclature presented by flora of Pakistan was used as standard.

2.3 Data Analysis Vegetation of Ayubia National Park, was analyzed with PC- ORD ver. 10 (RSoftware), a multivariate statistical software. Classification was performed by Bray-Curtis (1957) and Wards (1963) clustering methods.

2.3.1 Hierarchical Cluster Analysis Hierarchical Clustering is widely used method for clustering a set of species of a given area by assigning species to its own cluster and then algorithm proceeds iteratively (Bray & Curtis, 1957). Hierarchical clustering method was used to classify the plant communities in the study area that provides hierarchical similarity of sites. Agglomerative clustering algorithms start by treating each site as a cluster of 1. The closest two clusters are joined to form a new cluster. This method can use a matrix based on the Bray-Curtis and Warts Method. We used Ward’s Method “hclust” to find out the compact spherical clusters (Ward’s 1963) in among different sites, using the default setting in RSoftware (Vegan package). Cluster analysis was performed by Bray-Curtis (1957) and Wards (1963) methods, where the clustering criterion is based on applied distance matrix. A single cluster was formed by joining two closely related groups at each stage and at each step distances amongst clusters were recomputed by the update formula of Lance-Williams dissimilarity. The algorithm used in hclust was to order the subtree in to the closer clusters and Bray-Curtis dissimilarity was used to quantify the compositional dissimilarities

47 between the two different sites in ANP, based on counts at each site (Wards, 1963) and then arranged the sites into groups. Clusters formed from different sites of similar species composition provided a presentation in the form of a dendrogram. The dendrogram provided a summary of the similarity of the recorded species composition at various sites of ANP in a community form.

2.3.2 Ordination To support the Bray-Curtis hierarchal clustering analyses and to describe relationships between environmental variables and vegetation types, non- metric multidimensional scaling (NMDS) by using the vegan package (R- software), a PC- ORD Software for ordination of sampling plots was used. In the first analysis, all plots (160) were included but in the second analysis included plots belonging to Ward’s h-clusters (130). The NMDS procedure was also merged with default options of Bray-Curtis dissimilarity index. In order to evaluate the ordination a correlation between fitted vectors and ordination values was calculated by superimposing the fitted environmental vectors and centroids in to the envfit function of vegan. In order to further investigate the relationships between vegetation and environmental factors, we overlaid the selected variables on the NMDS ordination in the vegan package of R (Crawley, 2007). In order to assess the influence of environmental factors, the geographical locations of the sampling plots were used to extract the correlated values of ecogeographical variables in Arc GIS ver. 10.2 software (ESRI). Plant constancy was used in this analysis to name vegetation communities (ecological zone, which refers to the number of times each plant species is present in sampling plots belonging to a specific vegetation community. The results were explained in the form of three communities, which indicated an impressive interpretation of the over-all arrangements of the vegetation types in ANP.

2.3.3 Species diversity and plant communities In order to define species diversity between ecological communities classified by the cluster analysis (Bray-Curtis, 1957 and Wards, 1963), the Shannon and Wiener species diversity index (H') was computed by using the following formula: 48

Diversity H pi ln pi ……………………….…… (Eq. 1)

Where pi is the proportion of individuals extracted as a proportion of total cover or abundance, where ln = natural logarithm. The species richness was calculated for all the three ecological communities for further use in analyses. The Shannon-evenness index (E1) was used to calculate the evenness factor of species diversity by following equation: E1 H / ln s ……………………………………..…. (Eq. 2)

Where H' is Shanon - Wiener index; while “s” is the total number of species in a community and “ln” is the natural logarithm of the total number of the species.

2.3.4 Species-Presence only Data We used species presence-only data that consisted of 39 plots having pheasant nesting places out of total 130 pots (0.1 ha/plot), observed in a time period from 2012 – 2014. In all sampling points, the GPS (Global Positioning System) location of nesting places was recorded with associated plant species and prevailing environmental variables. Other essential data was also documented e.g. day - time, direction of nest, nest architecture, clutch size, soil type and associated plant species. During hatching period (May–July) the chick ratio and their movement were also observed. The field data surveys were carried out in a total area of 32 km2. Among the total of 19 eco-variables applied for phytosociological analysis, 11 variables were selected to predict the habitat preferences of both species under the influence of different environmental conditions and were linked by Remote Sensing (RS) and GIS techniques (Kerr & Ostrovsky, 2003; Chefaoui et al., 2005). The key rational of this data-collection was to specifically evaluate the species– presence in Ayubia National Park and to estimate their habitat distribution in relation to the observed and selected eco-variables as well as and vegetation types (Dettmers & Bart, 1999; Li et al., 2011). The presence data from the field were collected from mid-April to July (breeding season), in addition to some casual observations in other parts of the year. Topographic sheet survey of Pakistan with scale 1:50000 were used to convert the GIS based layer containing species presence records. The information gathered from species-presence data

49 and the ecogeographical variable grids was analysed in R-Software (Vegan) and was mapped in Arc-GIS (10.2) resolution by applying the same projection and extent of the study area, in order to evade the problems of overlapping maps in additional data analysis and habitat-prediction (Lisón and Calvo, 2013).

2.4 Habitat Suitability Modelling Techniques One of our objectives was to predict the suitable habitat distribution of Koklass and Kalij pheasant under the influence of different environmental variables. We used species presence evidences in all classified plant communities, GIS (geographical information system) data, and environmental variable layers in MAXENT (Maximum Entropy Modelling) and ENFA (Ecological Niche Factor Analysis) modelling to predict potential and appropriate habitat for pheasant species.

2.4.1 Maximum Entropy Model To estimate the distribution of Pheasants in Ayubia National Park, Maxent modelling was chosen based on presence only data. We produce a model of p for Kalij and Koklass pheasant, as probability distribution (constraint’s functions of the eco variables), known as features. The occurrence records of pheasants were obtained from the total laid plots in the Park, one of the world’s biodiversity hotspots defined by Myers et al., (2000). We considered ninteen explainatory-variables as conceivable predictors of the pheasant’s nest distribution in the Park. These variables were selected to analyse their biological significance to plant species distributions and other habitat modelling studies (Bashir et al., 2014).

2.4.2 Ecological Niche Factor Analysis

The Exploratory analyses for pheasant’s specific habitat required a preliminary for modelling analyses as they lead to select the variables of interest to model the habitat. For this purpose we used the ecological-niche factor analysis (ENFA). ENFA, a factorial analysis calculated the marginality (one axes) and specialization (several axes) for Kalij and Koklass pheasant in the entire area

50

(32Km2). As the niche based modelling requires examining different ecological attributes and alterations therefore, we used Boyce Indics (B) for both Pheasant species.

2.4.3 Ecogeographical Data Analysis Information regarding 19 explainatory ecovariables was extracted both from the field and GIS based dataset and two environmental -descriptor categories i.e. topographic and habitat variables were comprised the analyses (Franklin, 1995; Chahouki & Azarniv, 2010; Gabriela et al., 2014). Th e t opographic variables encompassed of slope, elevation terrain ruggedness and aspect while the habitat-variables included land cover-classes (Sérgio et al., 2007; Marthews et al., 2015). The eco-variables were chosen on the bases of their consequences in habitat prediction of Koklass and Kalij Pheasant in the Park. The selected eco-variables were applied in determining the niche physiognomies and characteristics of Kalij and Koklass pheasants in ANP (Devictor et al., 2008). The over-all information comprised secondary-review of literature, interrogation with Forests and Wildlife Department, local communities (female and mushroom collectors) hunters, and elites of the area (seasonal visitors). 2.5 Remote Sensing (RS) and Geographic Information System (GIS) data generation

The GIS data computed for the present study, included topographic factors (Aspv, elevation, SLF, slope, TPI and CTI), plant communities and satellite based landcover areas (NDVI, Tasselled Cap indices and B7b4) information for the study area (Ali et al., 2013). The slope was considered in degrees and a GIS based slope - map was created to indicate the specific-niche requirements of the species, because slope-variable profoundly classifies species distribution in to a complex mountain ecosystem. In terms of physical geography, aspect mainly indicates the directions to which a mountain- slope faces. The consequential aspect of a slope may significantly influence on its native climate which eventually marks species distribution. The aspect is more or less equally distributed from 0° to 360° but we used aspect value (av) in current data analysis by transforming to “aspect value” by applying the following formula:

Aspect value (av) = (Cos ((asp-30)/180*ʌ) +1)/2………………….. (Eq. 3) 51

Elevation also considerably contributes i n the composition of plants and animal communities in a set environment. Digital Elevation Model (DEM) was derived from ASTER satellite images using ArcGIS software. Topographic position index (TPI) was used to measure the topographic slope positions in the study area for landform classifications vegetation analysis and ENFA. Relative topographic position i.e. terrain ruggedness a significant and notable variable in habitat selection (nesting) which assists them to evade from predator attack was measured. The ruggedness factor was measured through GIS tool by combining the heterogeneity of slope and aspect. Vector Ruggedness Measure (VRM) was used by applying normal three-dimensional dispersion of vectors (orthogonal) to planar facets on a landscape. A layer of VRM values by using an ArcView (10.2) script (ESRI) was produced by using the original values, which specified numbers ranges from 0 to 1 (high to low). A steady state wetness index, a function of both slope and upstream called compound topographic index (CTI), was considered to analyse the soil characteristics Soil-horizon depth, organic matter content, silt percentage. Wet topographic index (WTI) was calculated in ArcGIS (10.2) using Arc Hydrology tool. Ridges were delineated from the DEM (digital elevation model) and flow direction grid and flow accumulation grid were calculated for the watershed demarcation in the Park (Yang et al., 2008). Furthermore threshold-stream grid was anticipated from flow-accumulation grid and its direction by using Strahler method (stream order increases when streams of the same order intersect). The ridge topography was also defined by using flow accumulation value (equal to zero threshold values) to demarcate the ridges (Marthews et al., 2013). The vegetation variables NDVI, SNDVI, Tasselled Cap Index ( greenness, brightness and wetness) were derived from cloud free Landsat-TM data.

Non Differential Vegetation Index (NDVI) is particularly used to assess the presence and condition of vegetation in a particular area. NDVI calculations are based on the principle that actively growing green plants strongly absorb radiation in the visible region of the spectrum (the “PAR,” or “Photosynthetically Active Radiation”) while strongly reflecting radiation in the Near Infrared region. NDVI states the ratio between red and near-infrared

52 reflectance captured by satellite sensors was calculated by using the following equation (values ranges between -1.0 to + 1.0):

NDVI = RNIR – RRED/ RNIR + RRED………………. (Eq. 4) Band 4 – Band 3 and Band 4 + Band 3 Where RNIR = near-infra -red and RRED = red wavebands.

The NDVI values = 0.1 to 0.2, denotes grasslands, herbs and shrubs, while values = 0.2 to 1, indicates dense green vegetation (Li et al., 2011). The ArcMap raster calculator was used to calculate NDVI. The NDVI (< 0), a negative value indicates water and urban topographies while the values ranging from 0 to 0.1 denotes either snow land or soil and barren rocks. Tasselled Cap Transformation (TM) is one of the preferred methods for development of spectral information content of a Landsat-TM data. We used Tasselled Cap Index that created three band images; represent brightness, greenness, wetness for area under consideration (Niamir et al., 2011).

2.5.1 Predictive Habitat distribution map for Kalij and Koklass Pheasant

The ENFA results were further used in calculating the habitat suitability maps that obliquely demonstrated the pheasant’s distribution with no absence data by comparing the species reaction to different environmental variables in the entire study area (Jackknife test of MAXENT).

There are manifold statistical modelling tools and procedures for predicting the probability of species occurrences in an area in relation to environmental variables for the existence of species. We selected Maximum Entropy (Maxent) and ENFA models for our study that is based on presence-only data to train the model for data analysis. Both models needs the presence only data of a species and have the ability to robust the intricate relationship between the species and their surrounding environmental variables, including interactions between variables, unlike other models. Therefore these models are very suitable for studying the important predicting variables and for interpreting the response of the species to each variable (predictor). The Maxent model is specially designed to produce predictive maps of the study area with use of continual and specific environmental variables. All the default parameters of Maxent

53 were used for analysis, including convergence threshold, background points, regularization multiplier, replicates and replicated run type = cross validation; and feature type = Auto features.

The selected default parameters were based on Phillips et al. (2008) recommendations, who concluded that Maxent defaults are applicable to a wide range of presence-only datasets, prominently datasets with 11-13 environmental variables and > 100 presences. We used 11 environmental variables for 39 presence cells.

Presence only data Independent test

ENFA MAXENT

ENFA Maxent Probability Probability map map

Classification of suitable Data Extraction Habitat

Data spread sheet of independent Model validation with Boyce test indics and AUC (confusion matrix)

ENFA Habitat Maxent habitat Suitability map suitability map

Combine Maxent and ENFA habitat suitability map

Fig. 2.5. Steps applied for the analysis of habitat suitability of Koklass and Kalij pheasant in ANP

The entire dataset of pheasant presence plots and the EGVs were subsequently transformed in to IDRISI file format (Clark Labs, 2000) to perform the predictive modelling for both species. Ecological Niche Factor Analysis (ENFA) and MEXANT modelling and were computed in Biomapper

54

4© software to produce predictive maps of ANP. The data sets of ecological variables (EGVs) comprised the same resolution and magnitude for additional projection in ENFA modelling software. A land use map of ANP was created through NDVI tools and by applying classification techniques of Mathieu et al. (2007) and Jenson (2005), using widespread ground control data. The data comprised pure herbaceous layer (mainly grasses), mixed forests (coniferbroad leaved), pure coniferous forests, broad leaved forests (Quercus sp., Acer caesium, Prunus padus and Betula utilis), exposed rocks with bare ground and rocks with grass cover.

2.5.1.1 Determination of Spatial Relationship of Species by using GIS tool

In order to determine the special relationship between plant species and pheasants under the influence of different selected environmental variables, additional analyses were performed (Lenton et al., 2000). The habitat was classified in to suitable (adopt/accepted) and optimal (ideal/prime) habitat. They were grouped together as very highly suitable and unsuitable habitat1` (marginal). The GIS-based layers for both species were overlapped to clinch the suitable and unsuitable habitats and to compute the habitat suitability (suitable nesting places used Koklass and Kalij) map of the Park. A dditionally, a number of alternative measures were statistically tested by joining the map layers with low-suitable and highly-suitable habitats in order to assess the areas of separation among species distribution in relation to selected environmental variables.

Chapter 3 RESULTS 3 RESULTS

Results regarding the vegetation survey of ANP (3312 hector area) documented 250 plant species belonging to 216 genera of the 79 families (Annex I). Analysis

55 of the vegetational data through hierarchical clustering and ordination techniques is given bellow:

3.1 Hierarchical clustering

Our results regarding the hierarchical clustering is given in appendix I, which resolved the vegetation of pheasants associated of the Park (ANP ) in the form of three plant communities. The clustering results are portrayed in a dendrogram (Fig. 3.1) showing the level where clusters were joined together, and the sites within each cluster. The process can be imagined as proceeding over a range of distances from 0 to some maximum value. Gradually during the process, distance considered is increased. At the beginning of the process, the method looks for smallest distance in the distance matrix. At this step, the two sites that have this distance are placed in the same cluster. Next, larger distances are considered, these could either be distances between clusters that were formed earlier or distances with sites that were not included in any cluster yet. The clustering pattern of vegetation (Fig. 3.1) is based on 130 sampling plots of 250 plant species.

3.1.1 Pinus wallichiana – Viola canescens - Vibernum mullaha community

This community predominantly spreads at an elevation range from 1467 m to 2693 m (Fig. 3.2) and composed of 74 sampling plots, within the study area. This plant community is mostly situated on the steep and rugged mountain of National Park. This ecozone in the lower elevation is dominated by the indicator species of subtropical zone i.e. Pinus roxburgii, Olea ferrogena, Punica granatum and Zanthoxylum armatum.

56

Fig.3.1. A dendrogram showing distance matrix for vegetation data using Bray/Curtis method of classification

The higher elevation vegetation is represented by Abies pindrow (Royle ex D. Don) Royle, Pinus wallichiana A.B. Jacks., Cedrus deodara (Roxb. ex D.Don) G.Don, Taxus baccata and broad leaves i.e. Acer caesium Wall. ex Brandis, Prunus padus L., Juglans regia L., Cornus macrophylla Wall., Quercus dilatata Lindl. ex A.DC., Quercus incana Bartram and Parrotiopsis jacquemontiana (Decne.) Rehder. This plant community is composed of dense

57 forests and providing an important habitat for the wildlife especially pheasants. The characteristic and indicator species of plant community I are listed in table 3.4. The highest number of pheasant nests was recorded from this plant community which indicates their habitat preference zone of vegetation or hotspots. The most dominant herb species were Viola canescens Wall, Ainsliaea aptera DC. Nepeta connata Royle ex Benth., Dioscorea deltoidea Wall. ex Griseb., Iris hookerana Foster, Geranium wallichianum D. Don ex Sweet, Bergenia ciliata (Haw.) Sternb., Dryopteris ramosa (C. Hope) C. Chr., Adiantum venustum D.Don, Gentiana kurroo Royle, Swertia alata C.B. Clarke, Swertia paniculata Wall etc. (Fig 3.4).

Distribution Pattern 3500

3000

2500

2000 Series2 1500 Series1 1000

500

0 1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52 55 58 61 64 67 70 73 76

Fig. 3.2. Distribution pattern of plant community I in Ayubia National Park

The shrub layer of the community is represented by Viburnum mullaha Buch.Ham. ex D. Don, Skimmia laureola Franch., Lonicera quinquelocularis Hard., Jasminum humile L., Berberis kunawurensis Royle, B. Parkeriana C.K. Schneid.,

58

Fig. 3.3. Vegetation features of Pinus-Viola-Vibernum community: A view of view of shrub layer at Tajwal compartment (A), Jhandi of Kao forest (B) and mixed forest of Olea ferrogena at Bhagan 8- iii (C).

Fig. 3.4. Important medicinal plants of Plant community I. A-Swartia chirata, BPeonia emodi, C- Bergenia ciliata, D- Gallium aprine, E- Androsace foliosa and F- Viola canescens.

Indigofera heterantha Brandis, Daphne papyracea Wall. ex G. Don and Cotoneaster bacillaris Wall. ex Lindl. The dominant grass species of the community includes Bromus hordeaceus L. (prevailing grass), Bromus pectinatus Thunb., Digitaria sanguinalis (L.) Scop., and Poa angustifolia L. Due to thick bushy forests and steep mountains (Fig 3.3.), the pheasant probably select this as being more secure from their predators. Although dense forests in this ecological zone, faces anthropogenic pressure in the form of fuel wood and fodder collection from the adjoining population.

59

3.1.2 Abies pindrow – Vibernum grandiflorum – Dryopteris ramosa plant community

This plant community is recognised at an elevation of 1709 m to 2685 m (Fig 3.5 & 3.6), comprising of 22 sampling plots, wherein 199 plant species were recorded (Table 3.1) with species Shanon diversity (4.8) and evenness (0.9). The coniferous vegetation is represented by Pinus wallichiana, Abies pindrow, Cedrus deodara, and Picea smithiana. Taxus baccata is also frequently recorded. The broad leaves are dominated by Acer caesium, Rhamnus purpurea, Prunus padus, Quercus baloot and Pyrus pashia. Alnus nitida is also recorded from Chota Gali. The important grass species are Apluda mutica, Bromus hordeaceus, B. catharicus, Stipa caragana Trin., Phleum pratense L. , Poa pratensis L. While the herb layer was composed of Dryopteris ramosa, Fragaria nubicola (Lindl. ex Hook.f.)Lacaita, Adiantum venustum, Sinopodophyllum hexandrum (Royle) T. S. Ying, Hedera nepalensis K. Koch., Gentiana kurroo Royle, Bupleurum hamiltonii N.P. Balakr., Sorbaria tomentosa (Lindl.) Rehder and Asparagus filicinus Buch.-Ham. ex D. Don. Important shrubs are Rubus pedunculosus D.Don, Indigofera heterantha Brandis, Viburnum grandiflorum Wall. ex DC., V. mullaha Buch.-Ham. ex D. Don, Lonicera quinquelocularis Hard., Rosa macrophylla, Skimia laureola, Isodon coetsa (Buch.-Ham. ex D. Don) Kudô, Spiraea canescens D.Don and Jasminum humile L. etc. This ecological community mainly consisted of mixed coniferous forests, therefore; plays a significant role by providing habitat to different wildlife species, especially pheasants.

Fig.3.5. An overview of the vegetation type and plant community II in ANP

60

Distribution pattern of Plant community II 3000 Series2, 2500 2686.007328 2000

1500

1000

500

0 Series1, 22 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22

Fig. 3. 6. Distribution pattern of plant community II at ANP

A total of nine nests were recorded from this ecological zone. This ecological community covering the lower blocks of Bakot (1708 m) and Kao compartments (1977 m); extending to Darwaza (D2), Bakot (BK3-i; Bk4ii-iii, Bk5i ), Kao (K5i,ii, K4i) and Bagan (B3, B8ii-iii; B8i ) at more than 2700 m elevation range.

3.1.3 Pinus wallichiana – Fragaria nubicola –Indigofera heterantha plant community

This plant community is mainly found at an elevation of 1752 m - 3033 m (Fig. 3.6) within the study area (highest elevation range) and is mostly found in all aspects. This plant community consists of 34 plots and 176 plant species.

61

Fig. 3.7. Some of the important representatives of plant community II; A- Rubus fruticosus B- Cotoneaster bacillaris C- Jasminum humile D- Aconitum violaceum E- Gentiana Kuroo F- Anemone tetrasepala (endemic sp).

The dominant tree species are Pinus wallichiana, Abies pindrow, Taxus baccata and Cedrus deodara while Picea smithiana is not frequent and broad leaved are Acer caesium, Aesculus indica, and Prunus padus. The species diversity (4.8) and evenness (0.92) is presented. The two highest tops of Ayubia National Parks are covered on top by this ecological community i.e. Miranjani and Mushkpuri top (Fig. 3.9) of subalpine zone while the lower limits starts from Bakot (B2), spreading to Kao forests and Bagan (B8,iii, ii and B9).

62

Fig. 3. 8. Distribution pattern of Plant community III in ANP

On lower limits, Pinus roxburgii is recorded in two plots from Mominabad side while stunted growth of tree species with lush grasses at peaks was observed. On top broad leaved species like Betula utilis and Salix denticulata can be observed in association with Euphorbia wallichii and Bromus hordeaceus (Fig. 3.10). The dominant tree species are Pinus wallichiana, Abies pindrow, Taxus baccata, Cedrus deodara, while Picea smithiana is less frequent. Among the broad leaved tree species Acer caesium, Aesculus indica, Prunus padus, Salix denticulata Andersson are common. The dominant herbaceous flora is covered by Persicaria amplexicaulis (D.Don) Ronse Decr. Syn: Bistorta amplexicaulis, Fragaria nubicola, Viola canescens Wall. Achillea millefolium L., Euphorbia wallichii Hook.f. Leucanthemum vulgare (Vaill.) Lam and Ranunculus muricatus L. Besides these plants Leontopodium brachyactis Gand., Rumex hastatus D. Don., Galium elegans Wall. ex Roxb., Valeriana jatamansi Jones, Veronica laxa Benth., Trifolium repens L., Nepeta erecta (Royle ex Benth.) Benth., N. connata Royle ex Benth., Origanum vulgare L., Potentilla nepalensis Hook., Staphylea emodi L., Iris hookerana Foster and Cotoneaster bacillaris Wall. ex Lindl were also recorded from this plant community.

63

Fig. 3.9. A view of Miranjani top (A) and Mushkpuri top (B), highest peaks of the Park (3033 & 3000 m, respectively) with representation of Sub-alpine flora of ANP

Fig. 3.10. Some plant specimens of plant community III. A- Euphorbia wallichii B- Indigofera heterantha, C- Iris hookerana, D- Persicaria amplexicaulis E- Fragaria nubicola and F- Potentilla nepalensis.

This plant community served as an important summer habitat for the wild animal and birds. A total of seven nests were recorded from this zone. This ecozone is severely disturbed by the contiguous communities in by free grazing and fuel wood stocking. The dominant shrubs are Vibernum grandiflorum, Indigofera heterantha, Lonicera quinquelocularis, Isodon coetsa and Desmodium elegans DC. Grasses are Bromus hordeaceus, Agrostis stolonifera L., Bothriochloa 64

bladhii (Retz.) S.T. Blake, Bromus porphyranthos Cope and Digitaria sanguinalis (L.) Scop.

3.1.4 Ecological specificity in plant species The results showed that each classified plant community has some specific plant species that are not found in the other two communities (Table 3.4). This specialization is due to ecological distribution done by NMDS ordination (Fig. 3.12 & 3.13) that distributed the plants according to their response to different environmental variables of the Park. The results of the present study revealed that community I comprises of 238 plant species that were documented from 74 sampled plots, while community II composed of 22 ecological plots having 199 plant species. Similarly, plant community III, consists of 34 plots and comprises a total of 176 species. There are only sixteen species which occurs only in plant community I, four plant species are specific to community II and three plant species are found only in community III (Table No. 3.1).

3.1.5 Plant-Species Diversity, Richness and Evenness Plant communities revealed variation in terms of species-richness, diversity and evenness (Table 3.1). The highest value of species richness (Fig. 3.10 A) is attained in plant community I (236), followed by plant community-III (coniferous zone) and declined to 177 in ecological community II. The Evenness (H value) of different ecological communities (Table 3.1) showed the maximum value of H (0.925) by ecological community III, followed by ecological community II (0.914) and I (0.896). The Shannon Diversity Index (H') of the three ecological communities revealed more or less similar trend to that obtained in the case of species-richness. As richness increases, diversity also increased (Fig. 3.15 A) Table 3.1: The Shannon-Wiener diversity (H') and Shannon-Weiner evenness (E1) indics values are presented for comparative purposes (Community I = PVV; Community II = AVD; Community III = PFI)

Code Plant Indicator species Shanon Shanon Species Community Diversity Evenness Richness Index Index (H’) (E1)

65

PVV Pinu.wal - Pinus wallichiana – 4.898 0.896 236 Viol.can – Viola canescens – Vibu.mull Viburnum mullaha

AVD Abie.pin - Abies pindrow - 4.732 0.914 177 Vibu.gra Viburnum grandiflorum - - Dryopteris remosa Dyop.ste PFI Pinu.wal – Pinus wallichiana – 4.895 0.925 199 Frag.nub – Fragaria nubicola- Indg.het Indigofera heterantha

250 236 199 0.924 200 177 Community III 4.895 150 Community II 0.9 14 4.732 100 Community I 0.8 96 50 4.897

0 0 2 4 6 Community I Community Community E H' II III

A B Fig. 3.11. Species richness in different ecological communities of the study area is shown in A, while Shanon diversity index with species evenness values are presented in B.

The highest H' value with 4.898 was recorded in ecological community I, and chased by ecological community III (4.895) and community II (4.732) respectively.

Abbreviation Botanical Name Family Occurrence Code of Species Alth.ros Alcea rosa Malvaceae 0.03

Anis.ind Anisomeles indica Lamiaceae 0.07 Aste.fal Aster falcifolius Asteraceae 0.04 Bide.chi Bidens pilosa L. Asteraceae 0.08 Datu.str Datura stramonium L. Solanaceae 0.01

66

Lesp.jun Lespedeza juncea var. Fabaceae 0.07 sericea

Loni.web Lonicera webbiana Caprifoliaceae 0.04 Olea.fer Olea feruginea Oleaceae 0.01 Rhem.aus Rheum austral Polygonaceae 0.01 Salv.nub Salvia nubicola Lamiaceae 0.01 Samb.wig Sambacus wightiana Sambucaceae 0.03 Sarc.sal Sarcococca saligna Buxaceae 0.03 Sola.sur Solanum surattense Solanaceae 0.04 Verb.off Verbena officinalis Verbenaceae 0.01

Zizi.oxy Ziziphus oxyphylla Rhamnaceae 0.08 Cala.hyd Calamintha hydaspidis Lamiaceae 0.18 Care.fol Carex foliosa Cyperaceae 0.18 Clin.vul Clinopodium vulgare Lamiaceae 0.18

Euon.ham Euonymus hamiltonianus Celastraceae 0.09 Frax.exc Fraxinus excelsior Oleaceae 0.06 Lotu.cor Lotus corniculatus var. Fabaceae 0.03 corniculatus

Table 3.2: Specific plant species in plant communities of Ayubia National Park

3.2 Forests Inventory The results revealed different BDH (diameter at breast height) classes of forests tress are available in ANP. The stock table data indicated a total of 226 trees per hectare of the Park and the total area of Ayubia National hosts 749801 trees (Table 3.3.), the first ever to calculate the tree species in Park through proper fixed point survey techniques (before that the survey was based on estimation only). The conifers are the dominant class covering 94.4% vegetation, followed by Taxus baccata i.e. 1% while the broad leaved are 4.6%. The dominant conifer tree species in the Park is Pinus wallichiana (59.7%). The percentage of Pinus roxbughii is negligible in the Park.

Table 3.3: Composition and number of tree species in Ayubia National Park S. No. Species Botanical name Number Percentage (%)

67

1 Kail Pinus wallichiana 447567 59.7 2 Fir Abies pindrow 240160 32.0 3 Deodar Cedrus deodara 19296 2.6 4 Spruce Picea smithiana 815 0.1 5 Taxes Taxus baccata 7331 1.0 6 B. Leaved - 34632 4.6 Total 749801 100

3.3 Ordination Ordination mapping was carried out by using Nonmetric Multidimensional Scaling (NMDS). The distribution of the sampling plots dataset along with NMDS ordination axes is shown in Fig 3.12. In order to associate the plant classification and ordination results of ecological communities, the resultant hierarchical clustering were superimposed on NMDS ordination. The horizontal axis of the diagram reflects the variation in topography. The vertical NMDS axes showing considerable variation in geographic and climatic factors. The Wards method of clustering and NMDS ordination method classify the vegetation of Ayubia National Park into of three ecological communities on the basis of presenceonly data set. The multiple regression analysis shown that the NMDS1 axis is explaining the elvation (0.967) , ruggredness (0.8513) and TWI (0.625) based variations. This axis additionally explain that the ecological communities located at height shows high positive values (CIII), while the axis also shown the positive distribution of some plots from community I (Fig 3.3). For second axis i.e. NMDS2, there are servaral factors responsible for the distibution of plant communities alog the axis, that are Topographic wetness index (0.7), ASPV (0.59), rugardness (0.524), CTI (0.15) and SPI (0.06). There is an overlap of community I and II on this axis.

68

Fig. 3.12. NMDS ordination of plant communities in Ayubia National Park

Fig. 3.13. Response of plant communities to eco- variables through NMDS ordination The distribution pattern of sampling polts follwing the same pattern as explained by hierachial clustering method based on Bray-Curtis distance materix of classification. The results of the test revealed that the role of elevation was highly significant (Pr(>r)=0.001) in distribution and classification of plant communities, while the SLF and SPI were significant with value (Pr(>r)=0.025) and (Pr(>r)=0.012) respectively but EVs like ASPV and slope values were non-significant i.e. (Pr(>r)= 0.056) and (Pr(>r)= 0.172) are

69 respectively, indicating their less contribution in the distribution of plant communities in Ayubia National Park.

3.4 Phytosociological correlation of Pheasants with ecological communities The cumulative sum of all present species occurrence was compared to the number of sampling plots described their extent of prevalence (commonness) or uniqueness (rareness) inside the plant communities in the Park (Table 3.5).

3.4.1 Nesting sites in plant community I: Pinus wallichiana – Viola canescens - Vibernum mullaha This community is located at an altitudinal range from 1467 m to 2693 m in ANP and comprises steep slopes, highly rugged areas (Fig.3.4). The vegetation type is composed of thick forests. The forest thickness increases with elevation, and distance from road sides. We observed a total of 22 nests of pheasants within this ecological community at different elevation (Table 3.5).

3.4.1.1 Nesting places of Koklass Pheasant There an upward progress in nests site selection in both species. The maximum number of nesting places for Koklass were recorded between 2034m to 2674 m (13 nests) and the rest were recorded above the 2700 m elevation range.

Table 3.4: Non- parametric test for ecogeographical variables of ANP Environmental NMDS1 NMDS2 r2 Pr(>r) Significance Variables level

CTI -0.98852 0.15107 0.0118 0.484

Rugged 0.85136 0.52458 0.0095 0.534

SLF -0.99974 -0.02292 0.0633 0.025 *

SPI -0.99803 0.06269 0.0697 0.012 *

TWI 0.62546 0.78026 0.0034 0.815

ASPV -0.80544 0.59268 0.0443 0.056 .

70

Slope -0.73643 -0.67652 0.0281 0.172

Elevation 0.96763 -0.25239 0.2527 0.001 *** Signif. codes: 0 ‘***’ 0.001 ‘**’ 0.01 ‘*’ 0.05 ‘.’ 0.1‘’1

Table 3.5: Preferred ecological communities with nesting data of Koklass and Kalij pheasant Community Code No. of No. of No. of Elevation Code plots plant Pheasant’s nest Range(m) species Koklass Kalij

PVV I 74 238 17 5 1467 2693

Pink II 22 199 6 3 1709 2685

Green III 33 176 5 2 1752 3033

3.4.1.2 Associated plant species In order to analyse the association of plants with the nesting habit of pheasants, we take in consideration the associated plants selected for nesting sites by both pheasants in the Park and established as correlation based on the constancy of plant species with pheasant (Table 3.6). Our results revealed conifers were the most constant tree species in all nest presence plots were Pinus wallichiana (94.1%) followed by Abies pindrow (88.2%). While the broad leaved found were Acer caesium (53%) and Prunus padus (53%) followed by Rhamnus purpurea (47%) and Cornus macrophylla (29.4%).

71

Fig.3.14. Box and whiskers representation of environmental factors significant in classification of ecological communities.

72

Nesting places of Koklass 3000

2500

2000

1500 Series2 Series1 1000

500

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Fig. 3.15. Nesting places of Koklass in Plant Community I

The dominant shrubs were Vibernum mullah followed by (82.3%), followed by Vibernum grandiflorum (47%) and Rosa chinensis (23.5%). Among herbaceous flora, Fragaria nubicola (82.3%) was found as the prevailing herb in the presence plots followed by Dryopteris stewartii (76%), Adiantum venustum (70.5%), Valeriana jatamonsii (64.7%) and Thalictrum cultratum (58.8%). The dominant shrub was Bromus hordeaceus (70.5%).

3.4.2 Nesting Places of Kalij Pheasant The nesting places of Kalij were recorded from 1546.5m to 2441m in ecological community I. A total of five presence (nests) plots were recorded along with their associated plant species (Table 3.7). Each nest was constructed on sloppy places on loose soil and surrounded herbs and shrubs and in most cases hosted by broad leaved tree, rather than a conifer.

3.4.2.1 Associated plant species The highly associated plants in this case were broad leaved i.e. Acer caesium (80%) and not any coniferous tree. Table 3.6: Synoptic table showing the constant plants in the presence plots of Koklass Pheasant in community I

73

Associated species Nesting places

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Pinus wallichiana + + + + + + + - + + + + + + + + + 94.1 Abies pindrow + + + + + - + + + - + + + + + + + 88.2 Fragaria nubicola - - + + + + + + + + + + + + + - + 82.3 Vibernum mullaha - + + - + + + + + + + + - + + + + 82.3 Dryopteris ramosa - + + + + + + + + - + + + - - + + 76.4 Adiantum venustum + + + + + + + + - - + + + - - - + 70.5 Verbascum thapsus - + + + + + + + - - - + + - + + + 70.5 Bromus hordeaceus + - + + + - + + - + + + + + - - + 70.5 Valeriana jatamonsii - + + + + - + + - - - + + - + + + 64.7 Thalictrum cultratum + + + + + + - + + - - - - + - - + 58.8 Acer caesium + + + + + - - + + - - - - + - - + 52.9 Geranium wallichiana - + - - + - - - - - + + + - + + + 47.0 Rhamnus purpurea + + - + + - - - + + - - - - - + + 47.0 Taxus baccata + + - + + - + - + - + + - - - - - 47.0 Viburnum grandiflorum - + + - + + - - + + - + - - + - - 47.0 Viola canescens + + - + ------+ + + - - + 41.1 Prunus padus + - + + - - + - + + + - - + + - - 52.9 Persicaria amplexicaulis - - + + + - - - + - + - - - + - + 41.1 Rosa chinensis - + - - - + - - + ------+ - 23.5 Cornus macrophylla - + + + - - - - + - + - - - - 29.5 Quercus dilatata ------+ - - - + - - - - 11.76

The other braod leaved recorded were Parrotiopsis jacquemontiana and Quercus incana with 40% constancy. The associated conifers were Abies pindrow (60%) and Pinus wallichiana (60%), while Taxus baccata (60%) was also found with the same constancy. The dominant shrubs were Indigofera heterantha, Rubus pedunculosus, Vibernum grandiflorum and V. mullaha with 60% constancy and Lonicera quinquelocularis with 40% constancy. The herbaceous flora is dominated by Adiantum venustum, Dryopteris ramosa, Pteris acanthoneura, Sinopodophyllum haxandrum and Thalictrum cultratum with 60% constancy. Among grass flora Dactylis glomerata was dominant with 40% constancy.

74

Fig.3.16. Koklass siting in an under rock nest on a clutch of 8 eggs in Bagan 8 at 2556.6m elevation

Fig. 3.17. Phytosociological association of Koklass pheasant in ANP of different clutch sizes 1, 8, 8, 6 , 4, 10 in A, B, C, D, E, F respectively.

75

Nesting places of Kalij Pheasant 3000 2500 2379.73632 2434.736 2239.872 2000 2142.288 1694.496 1500 Series1 Series2 1000 500

0 1 2 3 4 5 1 2 3 4 5

Fig.3.18: Distribution of nesting places of Kalij in plant community I

3.4.3 Nesting sites of Pheasants in plant community II: Abies pindrow – Vibernum grandiflorum – Dryopteris ramosa

This plant community is spread over from 1467 m to 2693 m elevation. Plant community II, provides nine nesting sites for Pheasants at different elevation range (Table 3.7).

3.4.3.1 Nesting places of Koklass Pheasant The nesting places of Koklass pheasant spread over altitudinal ranges of 2031 m to 2686 m height (Fig. 3.21). The nests were mostly found in wet places as well as in north facing direction of this plant community. A total of six nests were recorded from plant community II in association with different plant species.

3.4.3.2 Plant association with Koklass nest in Plant community II In plant community II, the most dominant and consistent tree was Pinus wallichiana with 100% consistency, followed by Abies pindrow (83%) while among broad leaved Aesculus indica was dominant (66%) followed by, Acer caesium (50%) and Prunus padus (33%). Among the shrub layer dominant species were Skimmia laureola (66%), followed by Viburnum grandiflorum (50%),

76

Fig. 3.19. Kalij Pheasant siting on nest in Kao forest of ANP

Fig. 3.20. Phytosociological association of Kalij Pheasant in ANP with different clutch sizes 6, 3, 7, 12 hatch eggs 4, 12 in A, B, C, D, E, F respectively.

Table: 3.7. Synoptic table showing the constant plants in the presence plots of Kalij Pheasant in community I

77

Nest Constancy (%) Associated species 1 2 3 4 5 Acer caesium + + - + + 80 Adiantum venustum + + + + - 80 Dryopteris ramosa + + + + - 80 Thalictrum cultratum + + - + + 80 Pteris acanthoneura + - + + + 80 Indigofera heterantha - - + + + 60 Isodon coestsa + - - + + 60 Abies pindrow - + + + - 60 Viburnum mullaha + + - + - 60 Viburnum grandiflorum - - + + + 60 Rubus pedunclosus - + + + - 60 Sinopodophyllum hexandrum + + - - + 60 Pinus wallichiana + + - - + 60 Strobilanthus urticifolia - - + + + 60 Taxus baccata - - + + + 60 Valeriana jatamonsii + - + - + 60 Verbascum thapsus + - + - + 60 Lonicera quinquelocularis + - - - + 40 Parrotiopsis jacquemontiana - - + + - 40 Quercus incana + - - + - 40 Dactylis glomerata - - - 40

3000

2500

2000

1500 Elevation (m) Nest 1000

500

0 1 2 3 4 5 6

Fig. 3.21. Nesting places of Koklass in plant community II

78

Indigofera heterantha, Rubus pedunculosus, Rosa torantifolia and Sorbus cuspidata with 33% consistency. The herb layer was dominated by Fragaria nubicola and Hedra nepalensis (83%), followed by Geum elatum (66%), and Arabidopsis himalaica, Adiantum venustum, Dryopteris ramosa, Senecio nudicaulis, Solanum surattense, Podophyllum hexandrum, Geranium wallichii, Valeriana jatamonsii, Verbascum thapsus with 50% consistency while Rubia cordifolia was 33%. Among grass species, Bromus hordeaceus dominated this plant community with 66% consistency.

3.4.3.3 Nesting places of Kalij Pheasant In plant community II, a total of three nests of Kalij were recorded. The nesting places were spreading over 1977 m to 2318 m elevation in ANP (3.22). All the three nests were constructed on steep slopes and moist, shady places.

3.4.3.4 Associated plants with Kalij nests In plant community II the nests were 100% associated with Acer caesium and Pinus wallichiana while Parrotiopsis jacquemontiana, Prunus padus and Abies pindrow with 66% consistency. So this plant community provided a mix situation of phytosociological correlation both with conifers and broad leaved. The shrub layer was dominated by Viburnum grandiflora and V. mullaha with 100% consistency followed by Rubus pedunclosus and Rosa chinensis with 66% consistency. The herb layer was dominated by Adiantum venustum, Dryopteris ramosa and Thalictrum cultratum with 100% consistency followed by Geum elatam, Sinopodophyllum hexandrum, Polygonatum multiflorum and Senecio analogus with 66% consistency while Fragaria was noted less frequent (33%) in this community of plants (Table 3.9).

3.4.4 Nesting sites in plant Community III: Pinus wallichiana – Fragaria nubicola –Indigofera heterantha

This plant community is mainly spread at an elevation of 1520-3033 m that comprising 34 sampling plots and 176 plant species. A total of six nests of Koklass were recorded from this plant community at different sites ranging from 2283 m to 2792 m elevation range (Fig. 3.23).

79

Table 3.8: Associated plants of Koklass in community II Associated species 1 2 3 4 5 6 Constancy (%) Pinus wallichiana + + + + + + 100 Abies pindrow - + + + + + 83.33333 Fragaria nubicola + + + - + + 83.33333 Hedra nepalensis + + + + + - 83.33333 Aesculus indica - + - + + + 66.66667 Bromus hordeaceus - + - + + + 66.66667

Geum elatum + + + - + - 66.66667 Skimmia laureola + + + + - - 66.66667 Acer caesium + + - + - - 50 Arabidopsis himalaica - + + - - + 50 Adiantum venustum + - + - - + 50 Dryopteris ramosa - + - - + + 50 Senecio nudicaulis + + + - - - 50 Solanum surattense - + + - - + 50 Podophyllum hexandrum + + - - - + 50 Geranium wallichii + - - + - + 50 Valeriana jatamonsii - + - + - + 50 Verbascum thapsus - + - + - + 50

Viburnum grandiflorum + + + - - - 50 Clematis corodata - - - + - + 33.33333

Dactylis glomerata - - + - - + 33.33333

Indigofera heterantha - - - + + - 33.33333

Prunus padus - + - + - - 33.33333 Pteris acanthoneura - - + + - - 33.33333 Rosa torantifolia - - + - - - 33.33333 Rubia cordifolia + + - - - - 33.33333 Rubus pedunculosus + + - - - - 33.33333 Sorbus cuspidata + + - - - - 33.33333 Stipa jacquemontii + + - - - - 33.33333

80

Nesting places of Kalij 2500

2000

1500 Elevation (m) 1000 Nest

500

0 1 2 3

Fig. 3. 22. Nesting places of Kalij in plant community II

Table 3.9: Associated plants with Kalij nest in Plant community Nest Constancy (%) Associated plant species 1 2 3 Acer caesium 1 1 1 100 Adiantum venustum 1 1 1 100 Dryopteris ramosa 1 1 1 100 Hedra nepalensis 1 1 1 100 Pinus wallichiana 1 1 1 100 Thalictrum cultratum 1 1 1 100 Viburnum grandiflora 1 1 1 100 Viburnum mullaha 1 1 1 100 Abies pindrow 1 1 0 66.66667 Parrotiopsis jacquemontiana 1 0 1 66.66667 Geum elatam 1 1 0 66.66667

Podophyllum hexandrum 1 1 0 66.66667 Polygonatum multiflorum 0 1 1 66.66667 Prunus padus 1 0 1 66.66667 Rosa chinensis 0 1 1 66.66667 Senecio analogus 1 1 0 66.66667 Rubus pedunclosus 1 1 0 66.66667 Fragaria nubicola 1 0 0 33.33333

81

Nesting sites of Koklass 3000

2500

2000

1500 Elevation (m) Nest 1000

500

0 1 2 3 4 5

Fig.3. 23. Distribution of nesting places of Koklass in plant community III

3.4.4.1 Associated plants with Koklass nest In plant community III, the nests were associated with conifers i.e. Pinus wallichiana, Abies pindrow and Taxus baccata with 60% consistency. While the broad leaved were dominated by Aesculus indica (60%) followed by Quercus dilatata, Cornus macrophylla and Quercus incana with only 20% consistency (Table. 3.10). The shrub layer was dominated by Viburnum mullaha (80%) followed by Indigofera heterantha (40%). The herb layer was dominated by Adiantum venustum, Dryopteris ramosa, Androsace foliosa, Persicaria amplexicaulis, and Fragaria nubicola with 60% consistency and; Impatiens edgwertii, Impatiens and Rananculus muricatus with 40% consistency. The grass layer was represented by Poa pratensis 60% consistency and Bromus hordeaceus of 40% consistency.

3.4.4.2 Associated plants with Kalij Nest in Plant community III In plant community III, a total of two Pheasant nests were recorded between elevation ranges from 1796 m to 2432.8 m. As compared to the rest of plant Table. 3. 10: Associated plants in community III for Koklass Pheasant

Nest

Associated Plant species N1 N2 N3 N4 N5

82

Abies pindrow + + + - - 60 Adiantum venustum - - - - + 60 Aesculus indica, + - + - + 60 Androsace foliosa + - + - + 60 Bistorta amplexicaulis - - - - + 60 Fragaria nubicola + + - - + 60 Geranium wallichii - + + - - 60 Pinus wallichiana - + - + + 60 Poa pratensis - + - + + 60 Taxus baccata + + + - - 60 Viburnum mullaha + + - + + 80 Dryopteris ramosa + - - - + 40 Impatiens edgwertii - + - - + 40 Indigofera heterantha - + - - + 40 Rananculus muricatus - - + - + 40 Bromus hordeaceus - - - + + 40 Quercus dilatata - - - + + 20 Cornus macrophylla - + - - - 20 Quercus leucotrichophora - + - - - 20

Nesting places of Kalij

3000

2500

2000 Series1 1500 Series2 1000

500 0 Nest Elevation (m) Fig.3.24. Distribution range of Kalij in plant community III communities in ANP, community III seems to be less significant in providing suitable habitat for Kalij Pheasant.

Table 3.11: Associated plant species with Kalij nests Associated plant species Nest

83

Constancy 1 2 (%) Abies pindrow + + 100 Acer caesium + + 100 Androsace rotundifolia + + 100 Bromus hordeaceus + + 100 Geum elatum + + 100 Pinus wallichiana + + 100 Quercus leucotrichophora + + 100 Taxus baccata + + 100 Thalictrum cultratum + + 100 Taxus baccata + + 100 Wulfenia amherstiana + + 100 Hedra nepalensis + - 50 Bupleurum tune - + 50 Cedrus deodara + - 50 Chrysopogon gryllus - + 50

Desmodium elegans + - 50 Dryopteris ramosa + - 50 Berberis Parkeriana - + 50 Persicaria amplexicaulis - + 50 Viburnum grandiflorum + - 50 Quercus baloot + - 50

The shrub layer was dominated by Desmodium elegans, Berberis Parkeriana and Viburnum grandiflorum with 50% consistency. The herb layer was dominated by Androsace rotundifolia, Geum elatum and Thalictrum cultratum with 100% consistency, followed by Hedra nepalensis, Bupleurum tune, Dryopteris ramosa, and Persicaria amplexicaulis with 50% consistency.

3.4.5 Nest architecture / Nidification Nidification or nest architecture means that how birds construct their nests by using different material. We observe different plants and plant material used by Koklass and Kalij for nest construction. In the course of three years’ time period we observed a total of 39 nests in Ayubia National Park. The

84 temperature recorded at all nesting sites were between 12 °C to 17 °C depends on day time.

3.4.5.1 Nidification of Koklass pheasant

The process of nest construction started from early March depends upon the melting of snow the direction of nest was not fixed, we found nests in southern- east and north -west sides but observed that they need sunlight during breeding and hatching period. Normally we observed koklass in nest at morning (0700hr) and noon (1300hr). Nest were made inside or below the trees, pole crops or dead coniferous trees from conifer needles and protected by herbs and shrubs. The major herbs recorded around nest were Adiantum venustum, Dryopteris ramosa, Fragaria nubicola, Geranium wallichianum, Hedra nepalensis, Verbascum thapsus Viola canescens, Podophyllum emodi, Primula and Persicaria amplexicaulis. While the important shrubs are Skimmia laureola, Vibernum grandiflorum, V. mullaha, Rubus and Rhamnus purpurea depends upon the altitudinal variation of nesting places. All the nests were constructed on rugged, sloppy terrains. The average circumference of Koklass nest was 9” x 8”.

3.4.5.2 Nidification in Kalij pheasant

If we compare the nesting art of Kalij and Koklass pheasant, the Koklass nest was precisely constructed as compare to Kalij. Kalij is elusive in nest construction that is merely made from few needles. The avg. nest circumferences of Kalij pheasant was 10 x 6.5 inches. The highly associated plant was Adiantum venustum and covered by Fragaria nubicola, viola canescens Vibernum grandiflorum, Desmodium sp., Spirea, Lonicera or Sarcococca.

85

Fig. 3.25. Nest architecture of Koklass pheasant in ANP

Fig. 3.26. Nest architecture of Kalij pheasant

3.4.5.3 Correlation of Eco variables with Koklass and Kalij In ANP, Kalij pheasant respond in a different way to the particular environmental variables (Fig. 3.4). The current distributions of Kalij and Koklass were largely affected by the topographic variables along with specific vegetation types. The distribution pattern was strongly predicted by TWI gradients and slop for Kalij while Koklass shows higher response to elevation and less response to other gradients i.e. CTI, Ruggedness, aspv and Twi. The distribution pattern was strongly predicted by TWI gradients and slop for Kalij while Koklass shows higher response to elevation and less response to other gradients i.e. CTI, Ruggedness, ASPV and TWI.

3.5 Predictive Modelling Techniques

86

3.5.1 Maximum Entropy Model (Maxent) In order to find out the occurrence probabilities of three communities, vegetation types were delimited by cluster analysis, mapped through Maxent modelling. The Maxent Model for the both species performed better than random, with average test of AUC values ranging from 0.75 (Kalij) to 0.87 (Koklass). The current distributions of Kalij and Koklass were largely affected by different ecovariables. This model predicted that different environmental variables strongly affected the distribution of Koklass and Kalij Pheasant inside the Park under the influence of a particular vegetation type. The Maxent model for Koklass performed very well with an average AUC value of 0. 87 (Fig. 3. 38). The current suitable habitat of Koklass Pheasant was predicted from an elevation range 2032 m to 2724 m in the Park. The top environmental variable for Koklass was elevation with 0.30 regularized training gains. The Maxent model for Kalij performed with an average AUC value of 0.75 (Fig. 3.39) on the basis of which the current suitable habitat for Kalij Pheasant was predicted from an altitudinal range of 1634 m to 2441 m in the Park.

While for Kalij pheasant the top environmental variable the regularized training gain for TWI (topographic wetness index) and slope were 0.18 and 0.002 respectively (Fig. 3.38). Both of these variables were the top two predictors in the Maxent model for Kalij pheasants.

87

Fig. 3.27. Jackknife test for Koklass pheasant in ANP

Fig. 3.28. AUC Sensitivity and Specificity values for Koklass

88

Fig. 3.29. Jackknife of regularized training gain for Kalij pheasant

Fig. 3. 30. AUC Sensitivity and Specificity values for Kalij pheasant

89

3.5.2 Ecological Niche Factor Analysis (ENFA)

The ENFA shows different results for Koklass and Kalij. The Boyce indices showed a decline (0.93 – 0.76) with the decrease in the number of presence plots (nest sites).

Table 3.12: Parameters used and extracted in ENFA model for Koklass and Kalij Pheasant

Species No. of Marginality Specialization Level of No. of nest EGV’s presence Factor Factor ( SF) Tolerance plots (MF) (I/S)

Koklass 29 11 0.408 1.38 0.725

0.664 Kalij 10 11 3.4 0

Total 39

(M, varying generally between 0 and 1), specialisation (S, indicating some degree of specialisation when superior to 1) and the level of Tolerance (showing the inverse of specialisation) on a scale of 0 – 1.

Both species attained a variable marginality coefficient from 0.408 (Koklass) and 0.664 (Kalij) respectively while the specialisation is more than one for both species, showing certain degree of exclusive conditions in the study area for both species of pheasants being highly specialized. However, the tolerance value (1/S) also showed variability 0.19 Koklass to 0 for Kalij. Thus indicating specialized habitat requirements in the case of Koklass to than Kalij. A brief of all the feasible parameters that were selected in the models in calculating habitat- suitability models for the two species were presented in following.

3.5.2.1 Ecological Niche Factor Analysis of Koklass Pheasant in Ayubia National Park

The entire number of presence-plots for the nests of Koklass Pheasant was 30. The model for Koklass pheasants was well fitted as indicated by ContinuousBoyce Index (Mean ± SD) that is represented by B= 0.13 ± 1 (which is the theoretical maximum 1) with a monotonically cumulative line p/e curve (Fig. 3.40). This indicates that p/e increases as suitability increases the results

90

(Fig.3.36) determined that Koklass Pheasant showed an overall marginality of 0. 408 with specialization of 1. 38 and tolerance coefficient of 0.727. The tolerance coefficient (1/S) of 0.727 (closer to 1) denoted that the species has a wider niche (favourable habitat) in the Park. This evidently described that the species requirements are to some extent different from the average habitat conditions existing in the study area. The specialisation factor revealed certain distinct habitat prerequisites of Koklass but still this species occupied a broader niche (Fig. 3.36). The marginality axis elucidated a high significant correlation with elevation by Koklass nesting (coefficient of 0.626) within the classified plant communities, and less significant relationship with relatively steep slope (0.555).

Table 3.13: Eigen values for Koklass pheasant Eigen values

S.No. Value Explained Cum. Explained Specialization Specialisation 1 2.727 0.13 0.13

2 5.099 0.244 0.374

3 3.806 0.182 0.556

4 2.359 0.113 0.668

5 1.549 0.074 0.742

6 1.328 0.063 0.806

7 1.239 0.059 0.865

8 0.988 0.047 0.912

9 0.825 0.039 0.951

10 0.589 0.028 0.98

11 0.427 0.02 1

Table 3.14: Relevant axes (with eigenvalues shown in %) and the related ecogeographical variables for Koklass in the study area. EVI’s 1 2 3 4 5 6 7 8 9 10 11 aspv1_Q -0.177 0.264 -0.058 -0.092 0.329 0.092 -0.384 -0.538 0.299 -0.101 -0.022 T

91 b7b41 -0.065 0.44 -0.4 0.166 -0.402 -0.01 -0.55 0.217 0.16 0.025 0.129 c1_ST 0.264 -0.123 -0.042 -0.234 -0.51 -0.069 -0.153 -0.385 0.214 0.409 0.303 c2_ST 0.161 0.354 -0.259 0.018 0.254 0.238 0.182 0.038 -0.153 0.188 0.637 c3_ST 0.557 -0.418 -0.165 0.251 0.075 0.1 0.098 -0.405 0.246 -0.64 -0.103 cti1 0.22 0.482 -0.179 -0.582 -0.17 -0.425 0.226 -0.072 0.148 -0.354 0.125 elev1 0.626 0.351 0.388 -0.036 0.204 0.109 -0.114 0.398 0.135 0.266 -0.094 ndvi1 0.172 0.008 -0.53 -0.486 -0.076 -0.035 -0.552 0.355 -0.057 0.25 -0.289 slf1 -0.057 -0.164 0.517 -0.022 -0.401 0.714 -0.252 0.022 -0.292 0.067 -0.125 slope1 -0.173 0.05 -0.089 -0.403 0.354 -0.397 -0.04 0.239 0.525 -0.331 0.555 tpi1 -0.234 0.19 -0.043 -0.328 -0.192 0.244 0.224 0.051 0.593 -0.072 -0.21 The positive and negative values are relevant only for the first marginality axis coefficient defining the avoidance (-) and preference (+), while the absolute values were taken into account in the remaining axes.

Expl.Spec. Cum.Expl.Specialisation 0.3 1.2

0.25 1

0.2 0.8

0.15 0.6

0.1 0.4

0.05 0.2

0 0 1 2 3 4 5 6 7 8 9 10 11 Eigen values (Axis)

Fig. 3.31. p/e value for Koklass pheasant

Koklass proved restrictions to various landcover classes during different spells of the year (avoided snow covered regions, agriculture, human population and bare rocks). The specialisation axes described that Koklass confines their habitat to steep slopes (0.555) in order to escape predators and preferred high elevations (0.626). The two axes also confirmed the presence of Koklass in different aspects, however; shows a diversion towards southern and eastern aspects (0.299 – 0.329). Therefore; the effect of aspect is not so important for pheasants distribution because avoided in most cases it was denied (-0.177). A robust correlation with escape terrain (0.58) was indicated by the test. The values of slope length factor (SLF), indicated variation from medium to high

92

(0.517 to 0.714) that presented its role as a fundamental factor t o impelled Koklass Pheasant for the selection of nesting places in a specific region. The topographic index (TPI) is also its importance (0.593) for Koklass distribution. However; Compound terrain index (CTI) is preferred in some cases (0.482) by Koklass while avoided in most of the cases (-0.425). The response to b7b4 was shown with a coefficient value of 0.44. The marginality axis explained significantly high correlation with plant community II (0.637), followed by Community III (0.557) and less response to Community 1 (0.409).

3.5.2.2 The Ecological factor analysis of Kalij Pheasant in Ayubia National Park

The total number of presence plots found for the Kalij Pheasant in the Park was 10. The low number of presence plots was most probably due to change in habitat range (low elevation bird) because of deforestation, habitat fragmentation and human population explosion. These factors may be compel Kalij towards high elevation rather than from to their globally known habitat range. The p/e curve (Fig. 3.41) showed a variable trend in defining the habitat suitability values.

The effective global value of marginality with 0.664 and specialization of 3.4 revealed that Kalij Pheasants are highly restricted in their nesting habitat in correlation with ecological and environmental conditions of Ayubia National Park (Table 3.16). The ENFA model for Kalij Pheasant also performed well, as presented by Continuous Boyce Index (Mean ± SD), where B= 0.35± 1 (which is the theoretical maximum 1) with cumulative increasing line denoted in the form of p/e curve (Fig. 3.1 b).

Table 3.15: Eigen values for Kalij Pheasant (Explained and Cum. Explained specialization) E igen values

S. No. Value Explained Cum. Explained Specialization Specialisation

1 4.49185E+15 0.358 0.358

2 5.91175E+15 0.472 0.83

93

3 2.13121E+15 0.17 1

4 17.936 0 1

5 6.763 0 1

6 5.724 0 1

7 1.784 0 1

8 1.293 0 1

9 0.625 0 1

10 0.397 0 1

11 0.132 0 1

A low tolerance- coefficient (I/S) = 0.00 (0) value for Kalij Pheasants, explaining the presence of more specialized niche (narrow niche) for nesting sites under the existing environmental circumstances (Fig. 3. 37). The marginality factorial axis showed a strong relationship for Kalij observations in the presence plots, where they preferred steep slopes (coefficients of 0.51). The association of Kalij Pheasant with b7b4 was positively shown in 20% nesting sites with coefficient value of 0.388 while in 80% nesting sites, this factor was avoided ( -0.257). The role of aspect was preferred in few case (0.559) but in most cases it is avoided (-0.52 to - 0.559). However; Kalij showed positive correlation with compound terrain index (CTI) with coefficient value of 0.359 but a negative response in few sites (-0.265) was also noticed. The marginality axis explained significantly high correlation of Kalij with community III (0.359), followed by Community I (0.295) and very less response to Community II (0.026).

94

Expl.Spec. Cum.Expl.Specialisation 0.5 1.2 0.45 0.4 1 0.35 0.8 0.3 0.25 0.6 0.2 0.15 0.4 0.1 0.2 0.05 0 0 1 2 3 4 5 6 7 8 9 10 11 12 Eigen values (Axis)

Fig . 3. 32. p/e curve for Kalij p heasant

The effect of TPI on Kalij distribution is more or less preferred ( 0.2) in few cases but in majority of nesting sites, this ecological factor is avoided by this bird (– 0.197). Same correlation is shown by Kalij with slope length factor (SLF) that was preferred in few nesting sites with (0.26) while in majority of nesting sites this factor shows negative correlation with a coefficient value (-0.571).

3.6 Geographic -Information System (GIS) and Remote -Sensing (RS)

Data Generation

The Geographic information (GIS) data and remote sensing data was calculated in correlation with 19 ecovariables, selected for the present study. The selected ecovariables included topographic factors (Aspv, Elevation, SLF, Slope, TPI and CTI), vegetation / plant communities based variables and landcover based (NDVI and B7b4).

3.6.1 Probability of Plant Community occurrence in ANP

Probability maps of plant communities were produced to predict their occurrence in ANP. Therefore a generalized map was produced with the help of probability maps of three plant communities determined by clustering method.

95

Table 3. 16: Relevant axes (with eigenvalues shown in %) and the related ecogeographical variables for Kalij in the study area. Variables 1 2 3 4 5 6 7 8 9 10 11

aspv1_QT -0.134 0.388 0.052 -0.559 -0.559 0.559 -0.559 -0.559 -0.559 -0.559 -0.559

b7b41 0.252 0.324 -0.344 -0.257 -0.257 0.257 -0.257 -0.257 -0.257 -0.257 -0.257

c1_ST -0.222 -0.1 -0.185 0.295 0.295 -0.295 0.295 0.295 0.295 0.295 0.295

c2_ST -0.393 -0.618 -0.073 0.026 0.026 -0.026 0.026 0.026 0.026 0.026 0.026

c3_ST -0.082 0.359 -0.255 0.183 0.183 -0.183 0.183 0.183 0.183 0.183 0.183

cti1 0.071 0.053 0.321 0.265 0.265 -0.265 0.265 0.265 0.265 0.265 0.265

elev1 -0.298 0.221 0.191 -0.454 -0.454 0.454 -0.454 -0.454 -0.454 -0.454 -0.454

ndvi1 -0.649 0.39 -0.088 -0.341 -0.341 0.341 -0.341 -0.341 -0.341 -0.341 -0.341

slf1 0.268 0.135 -0.571 -0.256 -0.256 0.256 -0.256 -0.256 -0.256 -0.256 -0.256

slope1 0.319 0.037 0.51 0.059 0.059 -0.059 0.059 0.059 0.059 0.059 0.059

tpi1 -0.144 -0.014 0.202 -0.197 -0.197 0.197 -0.197 -0.197 -0.197 -0.197 -0.197

A clear NDVI map was developed that markedly show the distribution pattern of the three classified plant communities in ANP from the lower limits to the topmost boundaries of the Park (3.38). According to probably map of plant communities distribution in ANP, the plant community I is covering Kao (K1 to K5 ii), Darwaza (II), Tajwal and Bagan (B8, i, ii, iii) compartments and touching the Bakot compartment boundaries of ANP (Fig 3.43). This plant community is facing extreme pressure due to heavy collection of fuel wood and fodder by the adjacent residential people of Mallach, Pasala, Lahurkus and Kun. While plant community II has starting boundaries from Bakot (BK1i) with a spread to Kao forest including compartments K1ii (Khun and Khanispur side), and Bagan (B8ii, iii). This zone included Kao forests from Khanispur side, vegetation above and below the Pipeline tract; west of Muskpuri top and upper limits near to Miranjani top (Fig. 3.39). Table 3.17: Correlation half matrix of the variables used in habitat suitability modelling of the Kalij and Koklass pheasant in three plant communities. The highly correlated variables were excluded from the analysis

Sample size : 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 27709 aspv b7/b4 Com.1 Com.2 Com.3 cti1 elev1 gvi1ndvi1 rough1 savi1 sbi1 slf1slope1 spi1 tpi1 tri1 twi1 wet1

96

1 Aspect value

2 B7/b4 - 0.18 3 Community 1 - - (Probability of 0.34 0.08 occurrence) 4 Community 2 - - 0.10 (Probability of 0.18 0.15 occurrence) 5 Community 3 - 0.10 0.37 0.46 (Probability of 0.53 occurrence) 6 Compound 0.04 - - 0.09 - Topographic 0.03 0.21 0.18 Index 7 Elevation - 0.05 0.41 0.12 0.36 - 0.22 0.19 8 Tasselled Cap - 0.02 0.08 0.32 0.57 - 0.04 Greenness 0.39 0.04 9 Normalized - - 0.06 0.43 0.39 0.03 - 0.72 Difference 0.14 0.51 0.03 Vegetation Index 10 Terrain 0.00 - 0.01 ------Roughness 0.03 0.25 0.12 0.26 0.14 0.07 0.12 11 Soil Adjusted - 0.13 0.05 0.34 0.60 - 0.01 0.99 0.68 - Vegetation 0.40 0.03 0.09 Index 12 Tasseled Cap - 0.48 0.00 0.19 0.54 - 0.01 0.87 0.34 - 0.91 Soil Brightness 0.43 0.06 0.05 13 Slope Length 0.07 - - - - 0.44 - - - 0.39 - - Factor 0.08 0.23 0.11 0.27 0.41 0.12 0.06 0.13 0.12 14 Slope 0.01 - 0.00 ------0.85 - - 0.46 (radians) 0.07 0.24 0.14 0.29 0.17 0.10 0.12 0.13 0.10 15 Stream 0.07 - - - - 0.57 - - - 0.23 - - 0.95 0.29 Power Index 0.10 0.26 0.02 0.27 0.42 0.10 0.02 0.11 0.13 16 Topographic - 0.06 0.08 - 0.12 - 0.13 0.05 0.00 - 0.06 0.08 - - - Position Index 0.03 0.02 0.32 0.02 0.34 0.06 0.39 17 Terrain 0.01 - 0.00 ------0.96 - - 0.41 0.85 0.25 - Ruggedness 0.04 0.26 0.13 0.24 0.15 0.08 0.12 0.11 0.07 0.02 Index 18 Topographic 0.01 0.06 - 0.19 0.02 0.41 0.03 0.09 0.12 - 0.11 0.08 - - - - - Wetness Index 0.10 0.75 0.23 0.84 0.03 0.07 0.74 19 Tasseled Cap 0.38 - 0.03 - - 0.05 - - - 0.04 - - 0.11 0.09 0.12 - 0.05 - Wetness 0.70 0.07 0.43 0.02 0.68 0.10 0.75 0.94 0.08 0.07

97

Communities

3.0

2.5

2.0

1.5

1.0 0 2 3 km

73.38 73.40 73.42 73.44

Fig. 3.38. Probability map for occurrence of three plant communities in ANP

Fig. 3.34. GIS maps for distribution of plant communities in ANP.

Similarly, plant community III comprising the pastures of Miranjani and Mushkpuri top (subalpine zone) while the lower limits starts from Bakot (Bk2), spreading to Kao Forests and Bagan compartments (B8-3-2 and B9). Only from lower limits, Pinus roxburghii was recorded (Fig. 3.39).

3.6.2 Aspect Value

The term aspect is generally used to the directions of mountain slope and significantly affects the species distribution in the area. As the aspect is almost homogenously spread from 0° to 360° therefore; in current analysis , I applied

98 the aspect (av), which indicated insignificant (0 value) relationship for habitat distribution of both species Pheasant in the study area (Fig. 3.44).

3.6.3 Elevation Elevation considerably influences the eco-distribution of flora and fauna within a given set of environmental conditions. The Digital Elevation Model (DEM) derived from ASTER satellite images revealed different results in distribution of plant communities and the presence of nesting sites at high and low elevation zones of the Park (3.40).

99

Fig.3.35. GIS maps showing Aspect Value (A), Elevation (B), Compound topographic index (C) and Soil Adjusted Vegetation Index.

3.6.4 Compound Topographic Index

Compound Topographic Index (CTI) is a state of perpetual wetness index and is a function for upstream and slope. Compound Topographic Index is highly correlated with several soil attributes. Plant Community III (0.09) shows positive correlation with CTI while community I (-0.21) and community II

(-018) shows negative correlation with CTI (Fig. 3.40).

3.6.5 Terrain Ruggedness

Terrain, a significant variable for niche preferences and nesting sites distribution in case of both species of pheasants, because it helps them to evade from predators during the time of predation’s attack. The values of ruggedness were measured in Arc-GIS by incorporating the heterogeneity of slope and aspect together. A layer of VRM values using ESRI (ArcView script) was produced, which show-up a number of ranges from 0 (high) to 1 (low). The obtained values were used to produce the terrain based map of the Park (Fig.3. 41).

3.6.6 Soil Adjusted Vegetation Index Soil Adjusted Vegetation Index (SAVI) is used for areas where vegetative cover is low < 40% and the soil surface is exposed, the reflectance of light in the red and near-infrared spectra can influence vegetation index values. The SAVI is structured similar to the NDVI but an additional factor of soil brightness is taken as correction factor. The output of SAVI is a new image layer with values ranging from -1 to 1. The lower value, the lower the cover of green vegetation. The results of SAVI revealed that vegetation cover in healthy form at height and in areas middle of the forests that away from disturbance sides and roads. The barren lands were clearly indicated by map near to the road sides especially in Khanispur and Khun hamlets.

100

3.6.7 Slope Length Factor

A probability map of slope length factor (SLF) was computed to describe the combined effects of slope length i.e., flow length and to predict the erosion rate in ANP. The computed map clearly show the eroded and stream flow the par and defined the distance from the point of origin of overland flow to the point where the slope decreases sufficiently for deposition to occur or to the point where runoff enters a defined channel (wet or dry).

3.6.8 Slope (radians) The slope is always measured in degrees and the output slope grid contained values. A slope map was produced for modelling and analysis of specific niche necessities of the species, as slope variable visibly marks speciesdistribution in an intricate mountainous- ecosystem.

3.6.9 Stream Power Index Stream power index (SI) map was computed to analyze the potential flow water and erosion in ANP. An SI map was produced for modelling of specific niche of Pheasants and their correlation with source of water. When the s catchment area of slope gradient increases, the amount of water contributed by upslope areas and the velocity of water flow increase, which increase the stream power index and risk of erosion. Stream power index controls potential erosive power of overland flows, thickness of soil horizons, organic matter, pH, silt and sand content, plant cover distribution

3.6.10 Topographic position index Topographic position index (TPI) is used to measure the topographic slope positions in the study area for landform classifications vegetation analysis.

101

3.6.11 Terrain Ruggedness Index

Relative topographic position is terrain ruggedness metric and a local elevation index to identifying landscape patterns and boundaries that may correspond with rock type, dominant geomorphic process, soil characteristics and vegetation.

102

Fig.3.36. ESRI maps of terrain ruggedness (A), Slop length factor (B), Slop (C) and Stream Power index in ANP. The datasets of all eco-geographical variables have the same resolution, magnitude and projection for further uses in the Ecological Niche Factor Analysis (ENFA) modelling.

3.6.12 Topographic Wetness Index

Topographic Wetness Index (TWI) was calculated by Arc- Hydrological tool in Arc-GIS (10.2). The computed NDVI map clearly demarcated the wet area in ANP (Fig. 3.37). The Ridges were defined from the DEM and direction of flow- grid and flow accumulation grid was calculated from watershed delineation process. The threshold- stream grid was ascertained from flow accumulation grid and its order by applying Strahler method according to which the stream order increases when streams of the same order cross together. The ridge features were delineated by applying the threshold values (flow accumulation value = zero), represents the ridges in the whole Park area. Finally a map was produced showing the streams and ridges in Arc-GIS.

3.6.13 Non Differential Vegetation Index Non Differential Vegetation Index (NDVI), a vegetation index is specially used as proxy to assess the vegetation and landcover area in the Park. We calculated NDVI of the three plant communities and developed NDVI maps to assess the vegetation status in the Park in each plant community. The results indicated vegetation lost and disturbance from read sides and accessible areas in all the adjoining boundaries of Ayubia National Park especially from Pasala, Khun, Riala and Mominabad helmets. The NDVI also indicates the presence of green vegetation in the nesting hotspot sites for both species of Pheasants.

3.6.14 Tasselled Cap Index

In present study we used Tasseled Cap index (Greenness, brightness and wetness) to transformed data to analyze the vegetation types and niche identification. The three TM bands produced an image which shows the greenness (measure of vegetation cover) and inter-relationship of soil and canopy moisture in Ayubia National Park. Tasselled cap index provided unit

103 within the range of -1 to 1. The Brightness (B1) reflectance or brightness was used to discern the soil exposure level between -1 to 1 to show the minimum bared soil surface.

The wetness index of tasselled cap was used as distinction between nearinfrared (visible) and shortwave-infrared reflectance. This index is providing a measure of soil wetness of moisture along with vegetation density. The values of this part of tasselled cap are also ranging between -1 to 1 (driest to wettest soil or low to high vegetation moisture- content that was highlighted in the map. The index tasselled cap- greenness shows differences between near infra-red and visible light and provided as an index for vegetation cover in the form of plant density and lifeform etc. this index is parallel to NonDifferential Vegetation Index but additionally used instead to two using, six ETM+ bands. The value of greenness also ranges between -1 to 1 (lowest density to highest vegetation density) which clearly. The Tasselled Cap index transformed the data of vegetation types in to three TM bands images where band was 1 for brightness (measure of soil); band 2 is for greenness (measure of vegetation) and band 3 is for wetness (interrelationship of soil and canopy moisture).

3.6.14.1 Tasselled Cap Soil Brightness (B1)

To measure the soil brightness, B1 map was produced. The map clearly indicate the soil brighten colour indicate the degraded land in the Park

104

Fig.3.37. GIS maps showing Topographic position Index (A), Terrain roughness (B), Topographic wetness index (C) and Normalized differential vegetation index (D).

105

Fig. 3. 38. GIS maps showing Tasselled cap greenness (A), tasselled cap brightness (B), SDNVI (C) and B7/4 (D) of Ayubia National Park.

106

3.6.14.2 Tasselled Cap greenness

To measure the soil greenness map was produced. The map clearly indicates the green colour indicated the greenness and dark green colour indicates the thick forests in ANP.

3.6.15 Band 7 and Band 4

Band 7 and Band 4 (B7/b4) were used to determine the vegetation types. The B7/b4 shows insignificant relationship (-0.18) to determine the vegetation distribution pattern.

3.6.16 Standard Normalized difference vegetation index

Standard Normalized difference vegetation index (SNDVI) data derived from visible and near-infrared data acquired by the MODIS was calculated and map was produced Which indicated that vegetation on tops elevation in the Park (Miranjani and Mushkpuri), green pastures and stunted tree forms (Fig. 3.47).

3.7 Predictive habitat suitability places of Pheasant in Ayubia National Park To Predicted the suitable nesting sites of Koklass and Kalij in Ayubia National Park, habitat suitability classes were divided in to three probability classes were divided to three classes, covering an area (ha) of > 50, > 60 and > 70 were calculated (Fig. 3. 46 ). The model explained the unsuitable habitats are denoted (P/E<1) and suitable habitats (P/E>1). Below the dashed line (P/E = 1) of the model, fewer presences are predicted for both species than expected by chance and vice versa Fig (3. 47 A & B). The three indices that were used for assessing the habitat suitability classes in the model, indicates high suitability for Kalij and Koklass (Table 3.19) by showing values of Absolute Validation Index (AVI), the Contrast Validation Index (CVI) and Boyce continuous index.

107

Habitat suitability Fig.3.39. Expected (p/e) ratio curve (Mean±SD), threshold between suitable (p/e>1) and unsuitable areas (p/e<1) for Koklass (left) and Kalij (right) models. Thick solid line = mean of the Boyce continuous curve, dotted lines = standard deviation. Dashed line (P/e = 1); below this line the model predicts fewer presences than expected by chance and vice versa.

An AVI of 1 (best value) could be obtained by a model predicting presence everywhere. A model predicting presence everywhere would get a CVI of 0 (the CVI is always lower than the AVI) as proposed by Boyce et al, 2002. All the presence-only evaluation (presence/absence indices as well) shows sensitivity to the study area. The AVI for both species signified that the model predicted high habitat suitability for those sites where the nests were recorded. A mean CVI (indicating the extent of suitability o f a map differing from a purely random-model) of 0.136± 0.351for Koklass and 0.297±0.317 for Kalij indicated that the maps are (the model is better than chance). A low value of CVI means that the model used the eco- geographical variables had intricately delineated the specific habitat for the species as compared to the overall habitat available in the reference area as the study area (closely fitting the ecological prerequisites of the species). Comparatively, the Continuous Boyce Index was preferred over the AVI and CVI. We followed the strategy proposed by Hirzel and Arlettaz (2003) and Reutter et al. (2003) for both pheasant species. The results revealed that Koklass ( 0.042 ±0.549) and Kalij have high value (0.17± 0.607) of AVI, which defining an effective predictive power. The low

108

Fig.3.40. Habitat suitability bins for Koklass and Kalij pheasant level of standard deviations attached to the Boyce Index for individual species showed its high robustness. The high mean values of indices indicate a high consistency with evaluation data sets. The lower the standard deviation (SD), the more robust is the prediction and vice versa.

109

Table 3.19: A few model evaluation indices drawn from Biomapper© 4, for the habitat suitability maps of Koklass and Kalij, computed with k-fold cross validations following Huberty’s rule. Species Absolute Contrast Validation Boyce Index (B) Validation Index Index (CVI) (AVI)1 Koklass 0.4±0.378 0.351 ± 0.136 0.549±0.042 (Mean ±SD) Kalij 0.418 ± 0.4 0.317±0.297 0.607 ± 0.17 (Mean ±SD) AVI varies from 0 to 1. CVI varies from 0 to AVI. Boyce’s Index varies from -1 to + 1, with 0 indicating a random model

Kalij Koklas 800.00 22.85 % 700.00 600.00 15.11 % 500.00 13.59 % 400.00 300.00 6.54 % 6.54 % 200.00 100.00 0.47 % 0.00 >50 >60 >70 Predicted distribution (probability)

Fig. 3.41. Predicted area (hectare) suitable for nesting of two pheasants in ANP. The figures in parenthesis show the area in terms of proportion of total Park area.

The positive values of Boyce index predicted the habitat suitability maps for Kalij and Koklass Pheasants in the Park (Fig 3.51). The results revealed that habitat class >50 is more suitable for both species with occurrence probability of 22.85% for Koklass and 6.54% for Kalij Pheasant, while the rest of two classes has shown less suitability for both species of Pheasant in ANP.

110

Fig. 3.42. Habitat Suitability map for Kalij (A) and Koklass (B) Pheasant (More aptly it should be nesting sites).

Chapter 4 DISCUSSION 4 DISCUSSION The climate of Himalayan region is distinct from the western and southern Asian climatic zone with rain shadow valleys in the inner side. This region is consisting of the world’s youngest and tallest mountains, with biodiversity hotspots (Khan et al. 2013) therefore; the region is marked as a distinct province of biodiversity hotspot. The natural vegetation pattern indicates the effect of precipitation received in the form of monsoon rains (Mueller and Ellenberg, 2013). Ayubia National Park is located in the West of Himalayan province; receives most of the precipitation in the form of snow during the months of December to February which characterised the vegetation as moist temperate and having variable degree of coniferous species including Pinus wallichiana, Pinus

111 roxburghii, Abies pindrow, and Cedrus deodara. Taxus wallichiana is also very common tree in the area. The common broad leaves are Acer caesium, Prunus padus, Aesculus Indica and Quercus Species. The forests fluctuating from canopy to open lush grassy meadows with a high variety of ferns and perennial herbs; lichens on the trees trunks and an assortment of mosses. The Forests type of Ayubia National Park is composed of conifers (49.2 percent) and shadow conifers (32.2 percent); conifer forest, shrubs and grasses (4.03 percent) broadleaved mix forests (11.18 percent), pastures and grasses (0.14 percent). Conifer forest is a prominent vegetation type of ANP, covering an area of 1679 ha. Blue Pine, Deodar and Fir have been fall as a single Land cover class. Shadow conifer forest covers an area of 1103 ha (approximately). Mixed Forest class represents a mixture of conifer forest and broadleaved forest. This class represents the transition zone of forest that covers 380 ha. Coniferous forest, shrubs and grasses class represents a mixing of shrubs and grasses with sparse conifer forest. Total area covered by this class is 137 ha. Pasture lands / grasses cover an approximate area of 4.67 ha. Water / wet soil class covers an area of 3.98 ha approximately. According to Hussain and Afza (2004); there were a total of 23 natural springs in the Park in which six are dried up and 17 are alive so no permanent water body exists; only seasonal drainages are found. The vegetation of the study area was first described by Champion et al. (1965) and Beg (1975) before the inception of Ayubia National Park in 1984. After the establishment of the Park the area was described by a number of ecologists i.e. Hussain and Ilahi (1991); analysed the vegetation classes of the area; Ahmed et al. (2006); Saima et al. (2009); Adnan (2011) and Ahmad (2012) analysed the floristic composition of the area by using multivariant approaches. Rabia et al. (2004); and Nadeem et al. (2013) described the ethnobotanical aspect of ANP. In 2006, Rabia conducted a research study on linkages between forests conservation and medicinal plants of ANP. Recently Yasmeen et al. (2011), explore the antibacterial plants from the Park. Similarly Tariq et al. (2014) highlighted the NTFPs in ANP. It is obvious that animals, birds and plants have different ecological niches and no two similar species can occupy the same niche (Hutchinson, 1957) due to different food and other ecological requirements which lead to habitat diversity

112 of these species on the earth. According to Elith et al. (2010) some of the new methods have better predictive accuracy than the primary established ones for presence only techniques because they can fit more complex models from smaller datasets and using explicit regularization mechanisms to prevent model complexity from increasing beyond what is supported by the empirical data (Hirzel et al., 2006) . Animals and birds use the same habitat for different requirements with a specific season (Miller, 1942; Julliard et al., 2006; Devictor et al., 2007; Alejūnas et al., 2010 and Jolli et al., 2011). With seasonal variation animals migrate up and down or to other places (Adams, 1908; Hildon, 1965), particularly avian fauna migrate down in snowfall and harsh winter (Lack, 1933; David, 1937; Ahmad and Rabia, 2014). Therefore, the identification of habitat use by avian fauna is vital for the analysis of the phytosociological relationship of bird’s niche (Rotenberry et al., 1980; Wiens et al., 1987; Orians et al., 1991). According to Adam (1908) the avian habitat occupancy is based on plants communities of the area. This statement is supported by various ecologists (Lack, 1933; Moreau, 1934; Zaniewski et al., 2001, Hussain & sultana, 2013; Ahmad & Afza, 2014). The prime rationale of this research study was to identify overall ecological distribution pattern of vegetation (plant communities) in ANP in order to assess suitable habitat for Kalij and Koklass pheasants under different environmental variables.

4.1 Vegetation Analysis Timely and accurate information of forest types and areas at the regional scale is needed for natural resource management, carbon cycle studies and modelling of biogeochemistry, hydrology and climate (Beaumont et al., 2007; Joshi, 2011; Negi, 2011). Remote sensing (satellite-based) approach provides an option to accurate and timely information for biodiversity analysis of an area (Saeed, 2008). The tangible usage of biological reserves needs essential information from the grass root regarding the current available resources, exclusively supported by scientific facts and figures. Modern expansion in statistical programmes has abetted to regain baseline evidences of the ecological clusters (different sets) of an area (Elith et al., 2011). These methods

113 are appropriate and are gradually adopted in Pakistan for the categorization and ordination (origination) of vegetation classes. Numerous analyses are presently conducted in different phytogeographical zones of Pakistan. Synecology is one of the most exciting and interesting branch of modern environmentrics, classically it was limited to analyse the effects of environmental factors in parallel on a number of species. Therefore, different multivariate approaches are developed to analyse community data, due to their apt ecological community data analysis, by decreasing the module of the dataset from many (often highly correlated) variables, into a limited uncorrelated variables that are not traditionally measured but only explicated. A great number of ordination algorithms have been defined by the ecologists of whom the geometric projective methods were developed (Legendre & Legendre, 2012). The present computed advances and multivariate tests have played a grander role in identification of the core environmental gradients which effecting the species clustering. Classification is the assortment of species and sample units in to sets or groups, and ordination is the arrangement or ordering of individual or sample units along gradients (Kindt, 2014). We used one of the software PCORD ver. 10 (R- software) programmes to apply nonmetric multi-dimensional scaling (NMDS) approach for ordination purpose and Bray-Curtis method for clustering of sampling plots and associated species into significant clusters driven by definite environmental attributes. The ordination, using NMDS is selected to set vegetation plots in relation to one another based on their resemblance species composition. NMDS indicated persistent and complete presentation of vegetation distribution in ANP on both axes. Kendall (1971); Kenkel & Orlóci, (1986) recommended NMDS scaling as superior test in comparison to Matric Scaling due its less presumption base and use of only rank order. The present study delineated three ecological habitats on the basis of species richness and frequency of occurrence, in a scientifically laid- out sampling plots (0.1 ha/plot) by characterizing the prime vegetation types in the absolute elevation ranges of the Park. Among the 250 plant species recorded from the Park were placed in 79 families and 216 genera. The most frequent plant families were; poaceae with 27 species, followed by Asteraceae (25), Rosaceae

114 and Lamiaceae (16); Rananculaceae (14), and others like Pinaceae, Polygonaceae (8), Apiaceae and Caprifoliaceae (6), Primulaceae (5), Taxaceae (1) and Ulmaceae (1) etc. The identified plant communities included palatable – grasses, herbs, shrubs, braod leaved and coniferous trees, shows that ANP sustained a good dwelling places for Koklass and Kalij Pheasants (Annex II). The classification of plant communities was established using Bray-Curtis and Ward’s methods. The ordination using NMDS (Nonmetric Multidimensional Scaling and horizontal axis of NMDS revealed variation in topography while the vertical NMDS axes showed considerable variation in geographic and climatic factors (Fig. 3.12 & 3.13).

4.1.1 Floristic composition of Ayubia National Park Cluster analysis classified the entire sampling plots into three major plant communities that including: (i) Pinus wallichiana – Viola canescens - Vibernum mullaha community, predominantly spreading at an elevation range from 1467 m to 2693 m and composed of 74 sampling plots, (ii) Abies pindrow – Vibernum grandiflorum – Dryopteris ramosa. community, spreading over at an elevation of 1709 to 2685m and comprising 22 sampling plots, and (iii) Pinus wallichiana – Fragaria nubicola –Indigofera heterantha community covering an elevation range of 1752 to 3033m (highest elevation range) and was mostly found in all aspects. This ecological community was composed of 34 plots. Similar results have been reported by Daniels 1991; Peer et al. 2001; Everitt et al. 2001; Dasti et al. 2007; Peer et al. 2007; Evett et al. 2013. The authors have described the importance of altitude as an environmental factor affecting plant species association. However; our results showed that elevation and slope are the two major environmental factors in determining the differences between the identified plant communities that are providing habitat for Kalij and Koklass pheasants in different seasons in the Park. The identified plant communities of the region were different from the previous studies conducted in ANP by different from time to time e.g. Champion et al. (1965) and Beg (1975), described only in relation to rocky slopes and elevation, before the establishment of National Park in 1984. Parts of the vegetation of the Park has also been described by a number of ecologists in to different

115 vegetation classes; for example Chughtai et al. (1989); Hussain and Ilahi (1991); Ahmed et al. (2006), Saima et al. (2009); Adnan et al. (2010); Adnan (2011); Ahmad (2012), analyzed the floristic composition of plants of the area by using different types of approaches and response of species to different eco-variables. The linkages between forests conservation and medicinal plants were described by Afza et al. (2004) and Afza, (2006). Similarly Gillani et al. (2006) and Nadeem et al. (2013) described the ethnobotany of the study area while Yasmeen et al. (2011) described the antibacterial plants of ANP. Tariq et al. (2014) described the non-timber forests products of the Park. The previous studies reported on the vegetation of the Park are based only on a selected portion. Saima et al. (2009) conducted phytosociological study in the area by using 18 km long line transect from south to north and recorded 142 plant species of from 180 sampling stands. DCA (Detrended Correspondence Analysis) multivariant analysis classified the sampling stands in to 5 clusters or plant communities. Similarly Ahmad (2012) analysed the floristic composition of ANP and recorded 59 plant species from 32 families by using quadrate method. For clustering of flora, CCA (Canonical correspondence analysis) technique was used that classify the plants in to two biplots under selected environmental gradient. Adnan (2011) analysed the abundance of medicinal plants in degraded and non- degraded forests area in the Park and used DCA and TWINSPAN tools. Raja et al. (2014) recorded 44 plant species belonging to 30 families from Park and used DCA technique for clustering, that established four types of plant associations from 27 sampling plots based on less wooded forests and high wooded forests up to 2310 m height. But the present study attempted to not only analyse the vegetation types by applying the cluster analysis and NMDS tools but also correlated the phytosociology of the Park with habitat of Himalayan Pheasants to classify the vegetation communities.

4.1.2 Forests Inventory Evaluation of vegetation types of an ecosystem is a basic requirement for conservation and management of an area (Ewald, 2003; Aho, et al., 2008; Caceres et al., 2012; Bass & Mayers; 2013) particularly in areas like mountain ecosystem comprising overlapping complex of environmental conditions and

116 vegetation types (Naidoo et al., 2008; Ahmed et al., 2011; Khan, et al., 2013). The Himalayan mountainous has a vast diversity of flora and fauna, particularly bird’s species. Similarly; each plant community of forests stand has unique assemblage pattern, environmental conditions and specific soil type that are playing vital role in providing micro and macro habitats for species along with other environmental services (Burrascano et al., 2013; Joshi, 2011; Semwal et al., 2004; Schild et al., 2008). The forests inventory of the Park not only presented a broader picture of tress composition in ANP but also helped in the estimated figure of total number of trees in the Park and available volume table of tree stock (Table 3.3). The study indicated that there are a total of 749801 trees in Ayubia National Park (226 trees / hectare) while the figure mentioned in the compartment history forms file of Ayubia National Park for 1993-94 to 2013-14 of Forest department, the total estimated trees in ANP are 58656. Our study revealed that declaration of the Park has not only banned the excessive extraction of timbers from the Park boundary but also the pole crops get significant increment to become trees in the past 30 to 40 years that resulting an increase productivity.

4.1.3 Phytosociological analysis in Pakistan In the field of ecology the analysis of multivariate data is becoming increasingly important in ecological research (Kindt, 2014), biodiversity or environmental impacts studies of different habitats like terrestrial systems, mangroves and other aquatic systems (Blashfield, 1976; Bloom, 1981; Causton, 1988; Chahouki, 2013). There is often a need to test hypotheses concerning the effects of experimental factors on whole assemblages of species at once, especially in ecological studies (Boitani and Fuller, 2000; Bolker et al., 2009). Statistical approaches (hierarchical clustering and ordination) not only make the data precise but also determined the similar and dissimilar clusters (Krushkal 1964b; Dixon, 2003). Ecovariables like aspect, slope and elevation extracted from plot location data, were useful in further defining the floristic composition and distribution (Zhang et al., 2010). The same techniques were used in Pakistan by different scientists to analyse the vegetation types of similar regions for example; Siddique et al. (2010), investigated the environmental variables

117

(complex) of Pine forests in the temperate region of Pakistan by applying TWINSPAN and DCA while clustering was done by Wards method of classification. Their results revealed that the Pines layer is related to elevation gradient while the understory is affected by elevation, soil PH, aspect and canopy cover. Shaheen et al. (2012) studies the phytosociological aspect moist temperate forests of western Himalayan region in Bagh district (Kashmir) at 13 sites, classified in 180 sampling plots and 5 plant communities. The study revealed positive correlation with environmental gradients while CCA established a negative correlation with altitudinal gradient (slope and aspect) in respect to diversity and richness. Reduction in forests structure and regeneration status was concluded. Arshad (2011) applied the same technique to analyze the vegetation of Chital Gol National Park, and divide the vegetation of the area in to four plant communities. To establish the relationship of plant communities with environmental gradients, TWINSPAN and DCA were applied. Similarly; Khan et al. (2013) conducted vegetation analysis to find out plant communities with different environmental gradients in Naran region of western Himalaya, by using the cluster analysis and ordination method. The cluster analysis grouped the plant communities in to five major classes. The use of DCA and CCA techniques indicates the distribution of plants with different environmental gradients. In Pakistan Saqib et al. (2006) and Ahmad et al. (2012) highlighted species response to different environmental variables. The current study established the relationship of plant distribution in the Park with 19 environmental gradients. Similar approaches of using multivariate analyses were applied in mountain protected areas of Pakistan for vegetation classification and understanding the ecological communities in response to different environmental and topographic variables. Khan, et al (2013) conducted study on Phytoclimatic gradient of vegetation in habitat specificity in the high elevation. Their study revealed that phanerophytes (tree species) are widely distributed on northern aspect while shrubs dominate the southern aspect. They further concluded that woody plants are dominant at 2450-2800 m (lower altitude) and hemicryptophytes and

118 cryptophytes (3900-4400 m), gradual change of vegetation type from moist cool temperate to dry cold subalpine and alpine vegetation in the upper elevations. Saqib et al. (2013) assessed the vegetation communities of Palas valley and classified the vegetation in to eight broader plant communities. The conclusion of the study was that spatial distribution of plant communities were strongly correlated with aspect, elevation and heat load indices. Khan et al. (2011) assessed the vegetation of Naran valley by using indicator species analysis (ISA) and two way cluster analysis (TWCA) that recognized five major plant communities from mesic cool temperate type to cold dry subalpine vegetation and concluded that soil depth, aspect and altitude were the strongest environmental variables in determining the vegetation type that is also confirmed by the current study. Durrani et al. (2010) conducted a study on ecological characterization of the flora of Aghberg rangeland, Baluchistan; Wazir et al. (2008) conducted study on vegetation types of Hapursan valley (Alpine meadows) in Northern Areas of Pakistan. Recently Farrukh et al. (2015) conducted a study on the floral diversity and ecological characterization of Mustuj valley Chitral (Hindukush range). They recorded 571 plant species including Juniprus semiglobosa, a new plant entry for flora of Pakistan and recommended further flora exploration and conservation of plant resources. Still, effective surveillance tools are needed to identify the impact of different environmental variables in describing the vegetation types in Ayubia National Park.

4.2 Phytoecological factors effecting habitat selection of Himalayan Pheasant The requirement based concept of the ecological niche was proposed by Hutchinson (1957) who defines it as a function that links the fitness of individuals to their environment. According to Hutchinson a realized niche is a subset of the fundamental niche in which a species was forced to occupy because of the association with other species considered it as a hyper volume of the multidimensional area control by environmental factors, inside which the populations of the species be able to endure. Therefore, the facts about species distribution in correlation to its instant environment are critical for identification of several aspects of their ecology (habitat selection) to assess the

119 anthropogenic pressure and to apply conservation measures (Lisón & Calvo, 2013). Habitat is a templet for ecological and evolutionary processes (Adams, 1908). Every single bird species entails a specific vegetation cover with complex environmental conditions (Habib & Afza, 2014). Himalayan region is the youngest complex and diverse mountain range; having unique diversity of flora and fauna (Zeitler, 1985; Martens et al., 2011). Some 876 bird species are recorded from Pakistan in which majority are found in the Himalayan belt, including five species of Pheasants. Therefore; this is essential to understand the habitat relationship with birds in a given area for future conservation and management of the species and to protect them from the risk of extinction. The availability of pheasants in its natural habitat indicates the forests health. According to Fuller and Garson (2000); and Miller (2010), pheasants are bio indicators of human disturbance and habitat destruction. Himalayan region is studied from time to time e.g. Robert, 1970; Kaul 1989; Laxman et al. 2008; Hussain et al. 2013; Shah et al. 2013 and Awan et al. 2014 along with seasonal surveys based on vocalization (counting number of calls) or flashing techniques to assess the population and ecological distribution in different areas were conducted by Gatson et al. (1980) and Nawaz et al. (2000) or on vocalization surveys by Miller (2009) and Miller (2010). The habitat utilization of cheer Pheasant was studied by Jolli et al. (2011) in great Himalayan National Park conservation area and established a strong correlation with vegetation (grassland) and distance from disturbance at an altitudinal range of about 1200 m to 3000 m. Garson et al. (1992) conducted studies on the ecology and conservation of the Cheer pheasant in the Kumaon Himalaya.

4.3 Pheasants of Pakistan In Pakistan and AJK, five Pheasant species have been recorded including Western Tragopan, Cheer Pheasant, Monal, Kalij and Koklass pheasant but the well-studied species in this region is Cheer Pheasant and Western Tragopan. In 1905, Rattray conducted research on nests of Pheasant in Murree hills and Galliat and published his research in Bombay natural history society. Similarly Saqib et al. (2013) conducted a regional assessment study on the landcover dynamics of Western Tragopan occurrence in Pakistan (Palas Valley). A study similar to Jolli et al. (2011) study was conducted by Awan et al. (2004) on habitat 120 utilization of cheer pheasant in Jhelum valley Muzaffarabad, AJK, Pakistan. A phytosociological analysis was carried out at five selected sites (calling sites) and to find out the associated plants including Pinus wallichiana but the shrub cover shows affinity with cheer Pheasant population, not the canopy cover of large trees. Khan and Awan (2006) carried out research on population status of cheer pheasant in District Bagh, AJK and Pakistan. The data was collected in the Phalla Game Reserve and a total of 49 adults of cheer Pheasants were estimated by calling method and documented the presence of different Pheasants species. Ayubia National Park, the moist temperate Himalayan region is hosting two species of Pheasants i.e. Koklass and Kalij. Since the inception of the Park in 1984, no ecological research study has been conducted on Pheasants at microhabitatlevel or their correlation with plants. Pheasant is an interesting galliforms but very little is known about their habitat occurrence, distribution, and phytosociology in ANP. Their natural is under severe threat due to many factors including deforestation, fire, poaching and grazing. The current study established the phytosociological correlation of two pheasant species i.e. Koklass and Kalij present in the Park. Kalij and Koklass are different species of pheasant and therefore, having separate ecological niche requirements. There are many factors that lead both species to select habitat in the forest of ANP for breeding and nonbreeding stages of life. Our results indicate that habitat selection and niche occupancy of the studied bird species were based on different phytoecological and edaphic factors. With in the Park a strong correlation was established among different plant species by Kalij and Koklass Pheasant. The major difference was the association of Kalij with broad leaves and comparatively low elevation for nesting while Koklass need high elevation and coniferous forest for nesting.

4.3.1 Habitat occurrence of Koklass Pheasant In 1928, Baker conducted research on the food biology of Koklass and concluded this species as highly vegetarian and comparatively less insectivorous because this species is grainy grasses, fern and moss (maidenhair). This statement was also supported by Howman (1979) who concluded that Koklass is a herbivorous species (eating green plants). There is

121 a seasonal downward movement up 1000 m is common in this species throughout the year, depends upon the amount of snow fall that is also confirmed by the current study. Hussain et al. (2001) investigate selected aspects of the ecology (microhabitat) of the Koklass and the Kalij Pheasant in Himalayan region (Kumaon area). They found that the determining factor for the distribution of Kalij and Koklass is altitude that is also supported by present study as major distribution factor for Kalij and Koklass in Ayubia National Park. Hussain et al. (2001); Hussain and Sultana (2013), where the authors established strong association of Koklass with high elevation, diverse and well developed ground cover, and sensitivity to degradation as compare to Kalij pheasant. The current study goes parallel to the findings for most nest presence plots but in few cases nest was found in close vicinity to main roads, walking tracts or other source of disturbance. Shah et al. (2014) recoded the presence of Koklass at three selected sites of Kalam valley, KPK, Pakistan at an elevation of 2590 and 2895 m. Although, Koklass is a very shy bird and very difficultly observe in the wild (Hussain & Sultana, 2013; Severinghaus et al. 1979) but we observed the bird during breading stage in nest at different elevation and from 2600 m at Bhagan- 9 compartment and at Jhandi of Kao forests of ANP, we also get chance to capture photographs for first time in wild from Ayubia National Park (Fig. 3.16 & Fig. 3.19). Miller (2010) conducted survey on Pheasants of Great Himalayan National Park, Himachal Pradesh, India and concluded that there is need to study Koklass Pheasant in their natural habitat range from 2100 m to 3300 m in temperate forests (broad leaved bamboo), oak forests (subalpine) and coniferous forests. There is a seasonal downward movement up to 1000 meter is common for this species throughout the year, depending upon the amount of snow fall that exposed them to high risk of hunting.

4.3.1.1 Phytosociological association The occurrence of Koklass is reported from different forests types i.e. coniferous forest at higher elevation, deciduous coniferous at mid-elevation and at an elevation range between 1000 – 3000m (Baker., 1930; Garston et al., 1981; Nan et al., 2004). According to Gaston et al. (1981), Koklass is found at an elevation

122 range of 2000 to 3700 m, with dominant associated tree species as Pinus roxburgii in lower zones and Pinus wallichiana, Quercus leucotrichophora), and Rhododendron arboreum at higher elevations of (3000 m to 3050 m) along with Betula utilis, Abies pindrow, Picea smithiana. The two subspecies of pheasant are relatively common in Pakistan and the present study identified highly associated plants for Koklass as conifers. The most consistent tree was Pinus wallichiana (89.2%) followed by Abies pindrow (82%), Acer caesium (43%) and Prunus padus while Quercus dilatata (10%) and Q. leucotrichophora (3.5%) showed weak association. The shrub layer was dominated by Vibernum mullaha (64%) followed by Rosa chinensis (40%), Viburnum grandiflora (39.2%), Skimmia laureola (37%) and Rubus pedunculosus (17%) as an important layer in majority of nesting sites, which contradict the findings of Hussain et al. (2001), where the authors mentioned that tree and shrub layers are insignificant in habitat distribution of Koklass pheasant. Although the current study is also supporting the finding of Gaston et al. (1981) for habitat selection of Koklass pheasant in the Park.

4.3.1.2 Nidification During the three years of survey period, twenty nine nests of Koklass and ten nests of Kalij pheasant were recorded in Ayubia National Park. According to the best of my knowledge this is the first attempt to record the nests in such number from an area along with nidification and clutch size of both species. Shah et al. (2014) recorded only two nests of Koklass Pheasant from Kalam valley, KPK in which one nest was empty while the second one has ten eggs. The observed nest was under the Oak tree in bushy habitat at an elevation of 8620 ft while as per the current study findings no nest was found under the Oak tree but under different coniferous trees and two nests were found under Cornus macrophylla. According to Baker (1930), Ramesh et al. (1999) and Shah et al. (2014), Koklass construct an underground nest composed of dry leaves of oak and different shrubs along with pine needles at bushy and sloppy places. Our study also supports this statement but according to our findings 99% of the nest material used is conifer needles rather than oak twigs. Shah et al. (2014) described the South-West direction for Koklass nest; however we observed nests of Koklass in all directions on slopes, exposed to sunlight. The author further established 123 the nesting places with plants like Quercus, Cedrus deodara, and Pinus wallichiana while according to our findings the highly correlated plants are conifers and Acer caesium (Endangered sp), not of any oak species.

4.3.1.3 Breeding season and clutch size In Koklass breeding season generally starts from April to June (Baker 1930) and often extend to July depending upon the melting of snow and other environmental conditions (Zaman, 2006; MacArthur & MacArthur 1961) but in Ayubia National Park the current study reported that the breeding season started by mid-March till late June. The normal clutch size is 5-7 (Baker, 1930) but according to Howman (1993), clutch size 6 is normal in wild and 9 - 12 eggs in case of aviaries. The current study recorded a normal clutch size of 10 in wild (Fig 3.17 F). Shah et al. 2014, observed a nest of Koklass with clutch size of 10 from Kalam valley at elevation 2627 m to 2836 m.

4.3.1.4 Distribution range Nan et al. (2004) carried out study in Xiaoshennongjia Mountains in the northern of Hubei Province, China on four species of Pheasants including Koklass. They concluded the habitat range of Koklass varies between 1000 to 4000m in various forests types i.e. coniferous forest at higher elevation and deciduous coniferous at mid-elevation but in ANP, we recorded 29 nesting sites at elevation range between 2032m to 2723 m as breading habitat, almost similar elevation range as recorded by Gaston et al. (1981). According to the authors, normal altitudinal range of Koklass range between 2200 to 2500 m while in exceptional cases they extended the suitable habitat (Gaston et al. 1981), and this is influenced by the amount of precipitation in the form of snowfall (Roberts, 1991; Wang et al. 2013). In Pakistan, Koklass is mainly found in temperate coniferous forests (Fig.1.13) of Hazara division mainly in Galliat, and in Kohistan (Hazara, Dir, Indus and Chitral). Similarly, some parts of Murree hills, Gilgit and Azad Kashmir sustain significant population of Koklass (Roberts 1991; Nawaz, et al. 1991).

124

4.3.1.5 Conservation status According to Mirza, (1977) one of the main objectives for the establishment of ANP was to provide promising habitat and protection to Koklass pheasant because the prime habitat of this species is vulnerable or endangered (Mirza, 1980; Gaston 1981a). Malik (2004) described that the only threat to this species is poaching and needs simple protection, the finding of current study is also supporting the same idea because hunting of this species is not possible at ANP and most likely the threat is poaching by female (grass and fuelwood collectors) and male (mushrooms collectors) from the Park area. During survey 20% of the observed Koklass nests (with eggs) were found empty after our next visit to the nest, with clear symptoms of grass cutting and tree cuttings. According to IUCN Redlist, Koklass is fall in least concern (Ic) category and widespread in its distribution range.

4.3.2 Habitat utilization of Kalij pheasant Iqbal (1993) studied the habitat use pattern of Kalij at two sites in the Indian Himalayas. Similarly, Philip and Gale (2007) studied the response of Kalij Pheasant in relation to climate change in Kao Yei National Park, Thailand. The authors concluded that climate change especially; rise of temperature has changed the cited habitat of Kalij from lowland to high elevation. They further suggested for detailed study on microhabitat of Kalij Pheasant. Their statement of habitat shift from low elevation to higher elevation is also supported by the current study in most of the cases (Fig. 3.32). According to Baker (1930), Ali and Ripley (1978), Nawaz et al., (2000), Hussain and Sultana (2013), the major floral composition for Kalij habitat in subtropical zone is cheer pine (Pinus roxburghii) with associated herbs and shrubs, while at elevation between 2000m to 3000m in the moist temperate zone, Quercus incana (Oak) and Pinus wallichiana (blue pine) will be the prevailing plants however, our findings indicted the dominant and consistent trees for both Kalij habitat were Acer caesium (90%), followed by Pinus wallichiana (80%), Abies pindrow (70%), Quercus incana (40%) and Taxus baccata

125

(33%) in moist temperate zone and Parrotiopsis jacquemontiana (37%) while Quercus baloota (10%) in subtropical zones. We found very weak association with of Kalij with Pinus roxburghii (2% consistency), recorded from the presence plot in case of Kalij in ANP. Sathyakumar et al. (1993) studied the density estimate and habitat use by Kalij and Monal pheasants in the Kedarnath Wildlife Sanctuary. Selven et al. (2013) assessed the abundance, habitat utilization and other activities of Kalij pheasant along the Red Jungle Fowl and Grey Peacock in low land tropical forest of Arunachal Pradesh (Eastern Himalaya), India. They found that Kalij is common in Eastern Himalaya in low canopy cover, high tree density, high cover of shrubs and low grass-cover which support the findings of Sathyakumar et al. (1993) on habitat preference of Kalij in Western Himalaya and authors concluded that about 80% of detected species included Alianthus integrefolia, Dillenia indica, Soxylum sp, and Paederia scandens. But in Ayubia National Park, the results were contradictory, among shrub density Viburnum grandiflorum, Viburnum mullaha, Indigofera heterantha, and Lonicera quinquelocularis Hard., Lonicera webbiana Wall. ex DC, were dominant while the highly associated trees species were broad leaved (Acer caesium), followed by conifers (Pinus wallichiana, Abies pindrow) along with dense grass cover. Gaston et al. (1983) carried out surveys in Himachal Pradesh and reported seven pheasant species and the proximate threat factors for the reported species. The micro and macro habitat use by Kalij pheasant is highly affected by different factors including distance from disturbance point and this contradictory to Gaston et al. (1981) in many cases where the Kalij pheasant was positively associated with source of disturbance including human, livestock and roads etc. the current study found only one nest of Kalij pheasant near human habitation.

4.3.2.1 Breeding season and clutch size Kalij breeding season normally start from March till June (Baker 1930) but during the present study one nest was recorded on 6 July with 5 eggs that hatched on 13 July, supporting the findings of Sharma and Saklani (1993a) who monitored different populations of the Kalij in Garhwal Himalaya and

126 determined the breading time period is (March to July months) in Garhwal, India. The nests are on ground and covered by dense plant undergrowth or protected inside a cavern covered by vegetation (Baker 1930). Sometimes, the females lay another clutch if its nest is damaged by some external factors (poaching or predator attack) (Bump & Bohl, 1961) and the current study also recorded similar patterns at Jalsi where only on egg was recorded during the entire breeding season in year 2012. According to Baker (1930) the clutch size is normally 4 to 10 eggs per season in the wild. According to Baker (1930) and Howman (1993) the clutch size generally ranges from 9 - 15 eggs in case of aviaries but interestingly a clutch size of 12 was recorded at Kao forest near Mominabad in wild from a single nest in the present study (Fig. 3.20 F)

4.3.2.2 Distribution Range Kalij pheasants inhabit a wide range of habitats and elevation ranging from 400 m - 3300 m but the present study recorded nesting places from 1699 m to 2441 m elevation range in Ayubia National Park.

4.3.2.3 Conservation status Due to the elusive and shy nature of Kalij pheasant, hunting is difficult (Bump and Bohl 1961). In Pakistan, Kalij pheasant has a very limited habitat and is vulnerable; therefore, its population is reducing at an alarmingly rate (Nawaz et al., 2000). The current study also recorded only 10 nesting places from ANP, spanning over 32 km2 area. According to IUCN Redlist, the species is of least concerned (Ic) in the world but our study concluded that this species is facing a server threat of poaching reducing its population alarmingly in and hunting around the Park, although a very fair clutch size (12) was recorded. According to Nawaz et al. (2000), about 30 pairs are available in ANP but within a time period of three years we were able to record only 10 nesting sites. The individual and group interviews of target groups (hunters) that are either the local community members or seasonal visitors but residing in the summer capital of KPK, it is evident that both species, particularly Kalij is very frequently hunted as game bird more commonly in winter season when they come down for food at lower altitude.

127

4.4 Environmental gradient analysis The environmental niche of the species is the assimilation of all interactions correlated to ecological dynamics with most importantly the topographies, climatic factors, inter and intra specific struggle and predation. The ecological communities of the current study were defined through classification (Bray-Curtis & Wards methods) and the ordination (NMDS) techniques and screened out all general vegetation zones, inhabited by Kalij and Koklass pheasant in the Park. The collected data in the form of sampling plots and presence of nesting sites was very efficient to established relationship with environmental gradients in the Park, a tool for providing details (quantitative data) of specific ecological communities in the Park. The programme was used with default setting were adequate in explaining the variation of the vegetation and predicting habitat suitability classes in the Park for Kalij and Koklass pheasant. The classified plant communities not only provided habitat to Himalayan pheasants but also to common (personal observation) and a diverse number of fauna due to unique floral characteristics of the Park.

4.5 Ground information system (GIS) and Habitat suitability

The GIS techniques presented management solutions to the Park authorities for assessing the competition between the species and animal behaviour at the landscape level. It is easier to predict suitability maps for marginal species than for most widespread and abundant species through purely methodological reasons (Hirzel et al. 2006, Braunisch et al. 2008). This is due to the fact that the habitat conditions available in the study area mostly match species requirements, which showed a wider choice of habitat parameters by the studied species. GIS based evaluation indicated that Koklass mostly occupied the northern and south eastern aspects, whereas the Kalij favoured southern and south eastern aspects of the Park area. In addition, the choice of suitable habitats by over the unsuitable and marginal habitats is largely because of the difference in slope, distance to escape terrain and ruggedness.

128

4.6 Habitat suitability modelling

Prediction and mapping of potential suitable habitat is crucial for monitoring and restoration of specie’s declining native populations (Hirzel, et al., 2001), their in-situe and ex-situe conservation and management of native habitat is imperative (Abram, 1986; Moskt, 1991; Elith et al. 2009). Sometimes essential data regarding species distribution pattern and status are inadequate and difficult for used the cluster making approaches for habitat modelling (Kinnaird et al. 2003). Therefore, accurate modelling of geographic distributions of species is fundamental to various applications in ecology and conservation. Predictive models of species geographic distributions are important for a variety of applications in ecology and conservation (Zaniewski et al., 2002; Hirzel et al., 2002). To predict the habitat suitability, two suitability models i.e. Maxent and ENFA based on presence only data and ecovariables were applied in the current study. The observed presence only was used to assess the utilization of the space by the pheasants and plant species i.e. the ecological niche. The modelling of given habitat has helped increase in identification of species distribution pattern and their ecological correlation (Bogosian et al., 2012). Modelling of species distribution is useful from different aspects such as site selection for restoration programme, defining corridors for wildlife movement, ecological impacts assessments, and restoration of species conservation and management of wildlife species. Multivariate statistical tools are very popular for demonstrating the niche-geometry and identification of species relationship (habitat) in the natural world (Pianka, 1981; Arellano et al., 2014). These techniques have been extensively used in delimiting the species specific niches within a set environment based on habitat, topography and other variables (Buckland & Elston, 1993; Hirzel et al., 2002; Hirzel et al., 2006). The increasing availability of large quantities of presence data, encourage the assessment methods which do not require data on species absence, since absence data is missing in those large datasets and is hard to collect. Therefore, the presence only modelling methods require a set of known occurrences together with predictor variables such as edaphic, climatic, topographic, biogeographic and remotely sensed variables (Bradfield & Kenkel, 1987 ). Many techniques are available varying from climatic envelopes to multivariate

129 regression splines and boosted regression trees (Anderson and Gonzalez, 2011) but for particular applications very different predictions are obtained by different modelling methods (Broton et al. 2004; Calenge & Bassille, 2008).

4.6.1 Maximum Entropy Modelling Maximum Entropy Modelling (Maxent) based on presence only data performed well for analysing the environmental factors responsible for conclusive habitat separation of Kalij and Koklass pheasant. The Maxent model for both species performed better than random, with average test of AUC values ranging from 0.75 (Kalij) to 0.87 (Koklass). The model predicted that the distribution of Kalij and Koklass pheasant is highly influenced by different environmental variables which lead its distribution along with the specific vegetation type. The Jackknife regularized training gain for Koklass revealed high affinity of this species with elevation (0.30) and less relation with ASPV (0.03), CTI (0.02) and ruggedness (0.01), while the rest of ecovariables, Koklass shows no response (Fig. 3.36 & 3.37). Similarly Jackknife regularized training gain for Kalij indicated very high association with TWI and to some extent with slope (Fig 3.38 & Fig. 3.39) which clearly indicated that environmental factors like elevation, aspect, TWI and CTI influences the vegetation composition and habitat of Pheasants (Mani, 1974; Miller, 2010; Sathyakumar, 2011; Bashir et al., 2014) and turned out to be a key factor governing the distribution of these pheasant species over a specific habitat space (Hussain & Sultana, 2013). The unique position of each pheasant species in a distinct ecological space visualized the concept of niche diversification. Probability of pheasant’s occurrence with in the three communities was mapped through Maxent modelling. 4.6.2 Ecological Niche Factor Analysis The Exploratory analyses required modelling analyses as they lead to select the variables of interest to model the habitat e.g. the ecologicalniche factor analysis (ENFA) described by Hirzel et al. (2001). The ENFA is a factorial analysis based on directions in the ecological space bases on marginality (one axes) and

130 specialization (several axes). Niche based modelling require to examine different ecological attributes and alterations in species niche requisites. The ecological niche factor analysis calculates habitat suitability maps which obliquely demonstrate species possible distribution with no absence data compared species response to different environmental variables in the entire study area. According to Grinnell (1917), no two species regularly established in a single fauna having the same niche relationship. Therefore, this is a big challenge for a single species to co-occur in other parallel ecological niche, which mostly results in the exclusion of one of the co-occurring species. In order to identifying the key environmental variables that determine the niche, the most vital operations rely on expert knowledge (Chahouki & Khalasi, 2012). Habitat modelling has been very effective for theoretical studies on species ecological niches as well as for defining and managing Protected Areas (PA) and National Parks (Cains et al., 1969). Each species has a specialized niche where more than two species intermingling with each other’s niches (competition). Diminution in such interaction may occur due to seasonal variation, divergence in diet forms or because o f specific habitat - variables Spatial i.e. habitat and diet; and temporal or seasonal variables defines the ecological-niche of a species in a set environment (Hutchison, 1957; Boyce et al., 2002) along with population-density and the time -spent in a particular habitat and help in evaluating the conservation value of a niche (Braunisch & Suchant, 2008; Jiang et al., 2009). As the absence data is complex to evaluate (decreasing the consistency of the data) therefore, a predictive models was used to owing false absences which gives a choice to scientists to prefer work on presence only statistics helping to predict suitable distribution of species where favourable environmental conditions exists to retain a species. The observed presences are used to assess the utilization of the space by the wildlife and plant species i.e. the ecological niche. The modelling of a given habitat has helped in identification of species distribution pattern and their ecological correlation (Bogosian et al., 2012). A few research studies based on ENFA have been carried out, worldwide on bird’s distribution. In Pakistan, avian fauna especially the pheasants are not studied in terms of their habitat occurrence and their ecological niche preferences (Zaman, 2008). On pheasants of Pakistan, no

131 detailed study was conducted to analyse and predict their ecological niche in correlation with specific vegetation type. Only the counts data are available for different surveys are available but scientific research on the population dynamics are found especially on Kalij and Koklass Pheasant are present with reference to Pakistan. However, one study was conducted recently in Palas valley, KPK, Pakistan on Western Tragopan using multivariate analysis by Saqib (2013). The current study focused on ecological niche factor analysis (ENFA) for two pheasant species i.e. Koklass and Kalij of the Park in relation to biophysical variables (aspect, slope, elevation, distance to escape terrain, ruggedness and land cover classes) and vegetation type to describe their habitat. Among pheasants, Monal is extinct from the Park (also extinct in most of their historic ranges) mostly due to excessive hunting, predator risk, poaching, habitat - loss and land- clearing. According to the IUCN Red List 2012, the population of Kalij pheasant is still classified under endangered category. The Pheasant are shy birds and live in a specific habitat and mostly in isolation. These species select resources for food but their ecological niche requirements are different for summer and winter spells. Ecological niche specialization and selection of habitat of a species in a present set of ecological environment is a matter of interest for ecologiests from decades (MacArthur & Levins, 1967; Connell, 1970; Abrams, 1986). According to Hutchinson’s niche theory (1957), the same ecological niche will never be occupied two different species for a long time that was also confirmed by the current study. Koklass pheasant occupy the higher elevation in the Park as compared to Kalij pheasant that selected lower elevation for dwelling nests. Habitat selection and separation of similar species achieve the purpose of interspecific competition (Hirzel et al., 2001; Pianka, 1981) and often leads to coexistence of multispecies concept (Pulliam, 2000; Tokeshi, 2009). Therefore, the current study not only established the phytosociological association of two pheasant species i.e. Kalij and Koklass in ANP but also determined the effect of different environmental variables as an important habitat distribution factors. The current study analysed the ecological niche of Kalij and Koklass pheasant by thorough recording of 39 nesting places in the Park. The multivariate analysis highlighted that Koklass preferred higher elevations for nesting and

132 are separated from each other. However, Kalij preferred comparatively steeper slopes, rugged areas and escape terrain and comparatively low. The ENFA and Maxent modelling performed well in predicting habitat suitability maps of Kalij and Koklass in the entire area of ANP. To the best of my knowledge, it is the first attempt to model habitat suitability of pheasant species not only in Ayubia National Park but in its entire range of habitat in Pakistan. The model of ENFA gave generous habitat- suitability predictions (Nawaz et al. 2014) for both species of Pheasants in the Park, based on presence only data with absolutely variables extent. The models were highly fitted in both cases with high Boyce Indices value B= 0.042 ± 0.549 for Koklass Pheasant (table 3.19) and B= 0.17±0.607 for Kalij Pheasant (Table. 3.19). Evaluation indices for Kalij and Koklass recommended that the ENFA model although deal with variation in the presence only plots, predicted precisely the suitable habitats in the entire range of the Park and make it simpler especially for marginal species rather than for most prevalent and common species (Hirzel et al 2006). This is due to the fact that the habitat conditions available in the study area mostly match species requirements, which showed a wider choice of habitat parameters by the studied species. The observed accuracy in these habitat suitability models also illustrated that a higher number of presence locations does not affect model precision and reliability (Sattler, 2007; Zaniewski et al 2002). The ENFA model is focused on the respective ecogeographical variables (EGVs) and coupled with their associated values at each presence location, indicating that high values of Boyce Index is not only connected to the number of presence areas but also more linked with habitat suitability due to pre-defined ecogeographical variables. A total of three habitat suitability classes were delineated based on Hirzel et al. (2006) in nest presence sites. The data was collected over a period of three years during breeding and hatching season that specifies the use of habitat by the species in order to precisely predict the distribution and identify their habitat suitability classes.

4.7 Habitat suitability (HS) Maps

The habitat suitability maps predicted the distribution of suitable habitats of the entire Park area for the Koklass and Kalij pheasant (Fig. 3.42). By following

133

Hirzel et al., (2006), three habitat suitability classes were delimited based on field observation (nesting places) in three years’ time period. Koklass is more widely spread in ANP as compared to Kalij pheasant. The resultant habitat suitability maps (Fig. 3.41) identified three habitat classes based on the species preferred niches within the study area. According to HS models an over-all 22.8% of the Park area is suitable for Koklass and 6.54% is suitable for Kalij (Fig. 3.41). This management aspect of the research is desires to be discussed with the Park management and the local inhabitants of the area for effective management of pheasants in the protected areas of Pakistan in general and ANP in particular. The observed accuracy in these habitat suitability models are also illustrated that higher number of presence locations do not affect model precision and reliability (Boyce et al 2002; Segurado et al 2008; Hirzel et al., 2008). This method performed well in the presence of a range of (high to low) data, in the case of Ayubia National Park. As both species utilized different habitat types (Fig.3.42), this led to significant results for future management of the protected areas and as indicator of forests health.

4.7.1 Koklass Pheasant

Results revealed that Koklass has a wider niche breadth and preferred conditions which are not significantly different as present in the study area. However, specialisation factors showed some restrictions over the preference to elevation, slope, vegetation type and ruggedness but the habitat choices are static and matched up with the available conditions in the study area.

4.7.2 Kalij Pheasant Kalij Pheasant showed different results with high degree of specialization in habitat selection mostly because of the use of very steep slopes, low elevation range as compared to Koklass pheasant, close to escape terrain, rugged region and topographic wetness (WTI) in addition to specific vegetation type. Miller (2010) also emphasized on Kalij preferred habitats particularly, distance from escape terrain and steeper slopes. Ecological niche preferences and the inverse relationships of habitats between the individual species indicated that the suitable habitat types for Koklass and

134

Kalij but unsuitable for other Pheasant species cited in different localities of the KPK. When unsuitable and suitable habitats of both species were compared, we found that elevation, slope, ruggedness and TWI and distance to escape terrain are the major predictors in differentiating between the suitable and unsuitable habitats of these species along specific vegetation communities.

4.8 CONCLUSION Since the establishment of Ayubia National Park in 1984, the Wildlife Department reports and observations of the experienced professionals revealed that the population of Koklass and Kalij pheasants were abundant; Monal another pheasant that had been extinct from the Park area due to hunting and habitat degradation. But nothing like any authentic scientific data was available anywhere else regarding pheasant’s population and their ecology in ANP. Sum of 749801 trees of conifers and broad leaved are recorded from ANP by the present study. Our study was focused on the niche characterization for Koklass and Kalij pheasants in ANP in relation to vegetation types wherein pheasant nests were identified in ANP. About 90% of nests were found under trees e.g. Pinus wallichiana, Abies pindrow, Acer caesium, Cornus macrophylla and Parrotiopsis jacquemontiana; 3% in dead trees of Pinus wallichiana and Taxus baccata, 5% under shrubs like Viburnum mullaha , Viburnum grandiflorum, Rhamnus purpurea and Celtis australis while 2% under rocks. All the nests were covered by specific plants. The common predators for pheasants breeding and hatching season are monkeys, jackal, fox and crows. Along with predators, poaching was observed in majority of nesting places. Based on nesting data, the result revealed that the population of Koklass pheasant is relatively stable but Kalij pheasant is declining. Our year’s long experience with native people, vegetation and Galliformes of the Pak; and the data gathered thereby concludes that:

i. Ayubia National Park with diverse flora and fauna hosts only two galliforms species i.e. Koklass and Kalij. ii. The Park is represented by an integrated unit of 32 km2 are mostly subtropical, moist temperate and a little bit subalpine zone

135

iii. The life of the Koklass and Kalij pheasant in Ayubia National Park is affected by 250 plant species belonging to 79 families. iv. Three broader plant communities were recorded in the Park based upon species dominance.

v. The main determining factors for establishment of plant communities were elevation followed by aspect slope of the Park area.

vi. A baseline for further scientific observation with regards to pheasant’s nesting; habitat selection and indicative phytosociology for Western Himalayas have been established that will a helpful tool for the natural conservation particularly in protected Areas management. vii. The Park area was not uniform in terms of vegetation distribution e.g. plant community I, was most diverse in terms of plant species richness and diversity followed by plant community II and III.

viii. The Maxent and ENFA models generated meaningful habitat prediction in the Park and can be used in other such studies.

ix. Remarkable differences in the use of identical habitat features for Kalij and Koklass pheasant were recorded.

x. Koklass showed wider niche in the study area as compared to Kalij Pheasant, which is very specific in habitat selection and preferred TWI and slope for nesting.

xi. Koklass Pheasant showed preferences to high elevation, rough terrain and aspect values for nesting.

xii. The habitat suitability model determined nearly 22.8% area of ANP (in 32km2) is suitable for Koklass and 6.54% for Kalij Pheasant. xiii. The Koklass was mostly associated with conifers, whereas Kalij pheasant preferred broad leaved species in the nesting places. xiv. A clutch size 10 was recorded in Koklass and 12 in Kalij pheasant in the Park. xv. In two plant communities Kalij nest was associated with Acer caesium that is globally as threatened tree species according to IUCN Redlist.

136

xvi. Evidence of climate change that is provided by the two species through their recorded dispersal, nesting places and other behaviour in the Park.

4.9 RECOMMENDATIONS Keeping in view our practical experience with the Park and results presented has drafted recommendations for future endeavours with reference to Ayubia National Park:

I. Specific boundaries, GIS based, within the borders of ANP, that are being used by the pheasant in different seasons of the year, for breeding, foraging etc needs clear demarcation and prediction accordingly.

II. The ecological communities and environmental factors (edaphic and climatic) that determine the distribution of pheasants should be studied in other areas for sustainable development of the species.

III. The distribution and population of different predators that happens as serious management threats for the two species in their natural habitats in general and ANP in particular.

IV. Nature of incentives that is either missing from the management of ANP or otherwise is poorly understood needs to be resolved properly.

V. Inter and intra species competition for food, shelter and nesting sites needs to be studied thoroughly.

VI. The indigenous knowledge that may help in the management of pheasants should be properly documented and made readily available to the public.

VII. A comprehensive management plan of the Park including all the stakeholders during all steps of management is required.

VIII. The species recently escaped or extinct from the Park viz Monal pheasant needs re-introductory recovery for restoring ecosystem of the Park.

137

IX. Comparative assessment of the micro-climate and habitat characteristics of nesting sites of both species at present locations and at global niche level should be analyzed.

X. Response of predators to the nesting preference of pheasants should be thoroughly documented to reduce the risk of population depletion of both species.

XI. Annual losses to the population of pheasants because of the combined factors of predators and poaching by community must be documented and checked accordingly.

138

5 REFERENCES

Abrams, P. A. 1986. Character displacement and niche shift analysed using consumer resource models of competition. Theo. Pop. Biol., 29(1): 107160. Acharya, B.K., and B. Chettri. 2012. Effect of Climate Change on Birds, Herpetofauna and Butterflies In Sikkim Himalaya: A Preliminary Investigation. Climate Change in Sikkim Patterns, Impacts and Initiatives. Info and Public Relations Department, Govt. of Sikkim, Gangtok. Adams, C. C. 1908. The ecological succession of birds. The Auk, 109-153. Adnan, M. 2011. Diversity and abundance of medicinal plants among different forest-use types of the Pakistani Himalaya. Doctoral Dissertation, Georg August University, Göttingen. Adnan, M. and D. Hölscher, 2010. Medicinal plant abundance in degraded and reforested sites in Northwest Pakistan. Mount. Res and Deve., 30(1): 25-32. Adnan, S.M., H. Ahmad, R. Afza, S.K. Hussain and M. Waseem. 2005. Introduction of offseason vegetables for improved livelihood and conservation of plant resources in Galliat. Proceedings of the national workshop February 22-24, 2005 on Conservation linked to livelihood opportunities – case studies, WWF-People and Plants Pakistan, WWFP, Peshawar. Afza, R. 2006. Linkages between Forest Conservation and Medicinal Plants of Ayubia National Park: Prospective and Constrains. M. Phil Dissertation, Department of Plant Sciences, Quaid-e-Azam University, Islamabad. Afza, R., A. A. Khan, Z.K. Shinwari and T. Watanabe. 2004. Medicinal plants of Ayubia National Park: prospective and constraints. In International Symposium on Medicinal Plants: Linkages beyond national boundaries, Proceedings. PP: 153-161 Afza, R., A. Ali and J. Khan. 2014. Medicinal Plants of Hazarnao (Malakand Agency) and their conservation Status. In 23◦ SILAE Congress, 7-12 September, 2014, Marsala Italy. Afza, R., Z. W. Shah and S.K. Hussain. 2014. Regeneration status of Pinus gerardiana in Bohtuli, Chitral (Pakistan): Issues and Durable solution. In

139

North America Congress for Conservation Biology (NACCB), Missoula city, USA. Ahmad, H. and R. Afza. 2014. Nesting places of Koklass and Kalij Pheasants, associated plants and habitat losses in Western Himalayas. In North America Congress for Conservation Biology (NACCB), Missoula city, USA. Ahmad, S. S. and S. Javed. 2007. Exploring the economic value of underutilized plant species in Ayubia National Park. Pak. J. Bot., 39(5): 1435-1442. Ahmad, S. S. 2012. Species response to environmental variables in Ayubia National Park, Pakistan using multivariate analysis. Pak. J. Bot., 44(4): 1225-1228. Ahmed, J. and F. Mahmood. 1998. Changing perspectives on forest policy. Policy that works for forests and people series (1): ISSN 1028-8228. Ahmed, K., M. Shehbaz, A. Qasim and W. Long. 2015. The linkages between deforestation, energy and growth for environmental degradation in Pakistan. Ecol. Indic., 49: 95-103. Ahmed, M., T. Husain, A. H. Sheikh, S. S. Hussain, and M. F. Siddiqui. 2006. Phytosociology and structure of Himalayan forests from different climatic zones of Pakistan. Pak. J. Bot., 38(2): 361. Ahmed, M., M.F. Siddiqui, A. Ahmed, A.M. Sarangzai, T. Abbas and M.I Hussain. 2015. Present state and multivariate analysis of a few juniper forests of Baluchistan, Pakistan. Pak. J. Bot., 47(1): 135-140. Ahmed, M., S. S. Shaukat and M. F. Siddiqui. 2011. A multivariate analysis of the vegetation of Cedrus deodara forests in Hindu Kush and Himalayan ranges of Pakistan: evaluating the structure and dynamics. Turk. J. Bot., 35(4): 419-438. Aho, K., D. W. Roberts and T. Weaver. 2008. Using geometric and non geometric internal evaluators to compare eight vegetation classification methods. J. Veg. Scien., 19(4): 549-562. Alam, J. and S. I. Ali. 2010. Conservation Status of Androsace Russellii Y. Nasir: A Critically endangered species in Gilgit District, Pakistan. Pak. J. Bot, 42(3):1381-1393.

140

Alam, J. and S. I. Ali. 2010. Contribution to the red list of the plants of Pakistan. Pak. J. Bot., 42(5): 2967-2971. Ali, H. and M. Qaiser. 2010. Contribution to the Red List of Pakistan: A case study of Gaillonia chitralensis (Rubiaceae). Pak. J. Bot., 42, 205-212. Ali, S. I. 2008. Significance of Flora with special reference to Pakistan. Pak. J. Bot., 40(3): 967-971. Ali, S. I. and M. Qaiser. 2011. Flora of Pakistan 194-218 (1993). Department of Botany, University of Karachi. Ali, S. I., and M. Qaiser. 1986. A phytogeographical analysis of the phanerogams of Pakistan and Kashmir. Proceedings of the Royal Society of Edinburgh, Section B. Biol. Sci., 89: 89-101. Ali, S. I., and M. Qaiser. 2012. Flora of Pakistan. Department of Botany, University of Karachi. Ali, Z., Bibi, F., Shelly, S. Y., Qazi, A., and Khan, A. M. 2011.Comparative avian faunal diversity of Jiwani Coastal Wetlands and Taunsa barrage Wildlife Sanctuary, Pak. J. Ani. & P. Sci., 21(2): 381-387. Ali, S. and S. D. Ripley. 1987. Hand book of the Birds of India and Pakistan. New Delhi: Oxford University Press. Amjad, M. S., M. Arshad and S. K. Chaudhari. 2014. Structural diversity, its components and regenerating capacity of lesser Himalayan forests vegetation of Nikyal valley District Kotli (AK), Pakistan. A. Pac. j. trop medi., 7, S454-S460. Anderson, R.P. and I. Gonzalez. 2011. Species-specific tuning increases robustness to sampling bias in models of species distributions: an implementation with Maxent. Ecol. Model., 222(15): 2796-2811. Arellano, G., L. Cayola, I. Loza, V. Torrez and M.J.Macía. 2014. Commonness patterns and the size of the species pool along a tropical elevational gradient: insights using a new quantitative tool. Ecog., 37(6): 536-543. Arshad, M., F.M. Qamer, R. Saleem and R.N. Malik. 2012. Prediction of Kashmir markhor habitat suitability in Chitral Gol National Park, Pakistan. J. Biod., 13(2): 78-87.

141

Arshad, M., N. Mehmood, H. Muqadas, J. Chaudhry, I. Mustafa, M. R. Khan and H. Ahmed. 2014. Avifauna Studies in Co-Relation with Alteration in Climatic Patterns and Hydrology of Uchalli Lake, Punjab, Pakistan. Pak. J. Zool, 46(2):503-515. Arshad, M., R.N. Malik and Z. Saqib. 2013. Assessing potential habitats of Kashmir markhor in Chitral Gol National Park, Khyber Pakhtunkhwa, Pakistan. Pak. J. Bot., 45(S1): 561-570. Ashiq, M. W., Zhao, C., Ni, J., and Akhtar, M. 2010. GIS-based high-resolution spatial interpolation of precipitation in mountain–plain areas of Upper Pakistan for regional climate change impact studies. Theor and Appl. Climat., 99(3-4): 239-253. Aumeeruddy, T.Y., Z.K. Shinwari, A. Ayaz and A.A. Khan. 2004. Ethnobotany and management of fodder and fuel wood at Ayubia National Park, Northwest Frontier Province, Pakistan (People and plants working paper). Int. Plan. Conser. Un., WWF-UK. Austin, M.P. 1987. Models for the analysis of species' response to environmental gradients. Veg., 69(1-3): 35-45. Awan, M. N., Ali, H. and Lee, D.C. 2014. Population survey and conservation assessment of the globally threatened cheer pheasant (Catreus wallichi) in Jhelum Valley, Azad Kashmir, Pakistan. Zool. Res. 35(4): 338-345. Baker, E. C. S. 1930. Game Birds of India, Burma and Ceylon, Vol. 3. London: John Bale and Sons. Barrientos, R., L. Kvist, F. Barbosa, F. Valera and E. Moreno. 2014. Refugia, colonization and diversification of an arid adapted bird: coincident patterns between genetic data and ecological niche modelling. Mol. Ecol., 23(2): 390-407. Bashir, T., T. Bhattacharya, K. Poudyal, S. Sathyakumar and Q. Qureshi. 2014. Integrating aspects of ecology and predictive modelling: implications for the conservation of the leopard cat (Prionailurus bengalensis) in the Eastern Himalaya. Act. Therio., 59(1): 35-47. Basille, M., C., E. Calenge, R. Marboutin, Andersen and J.M. Gaillard. 2008. Assessing habitat selection using multivariate statistics: Some refinements of the ecological-niche factor analysis. Ecol. Mod.., 211(1): 233-240. 142

Bass, S. and J. Mayers. 2013. Policy that works for forests and people: real prospects for governance and livelihoods. Routledge. Beals, E.W. 1984. Bray-Curtis ordination: an effective strategy for analysis of multivariate ecological data. Advances in Adv. Ecol. Res., 14(1): 55. Beaumont, L.J., A.J. Pitman, M. Poulsen and L. Hughes. 2007. Where will species go? Incorporating new advances in climate modelling into projections of species distributions. Glob.Chang. Biol., 13(7): 1368-1385. Beebe, W. 1936. Pheasants: their lives and homes (Vol. 1). Doubleday.

Beebe, W. C. 1918–1922.A Monograph of the Pheasants, 4 vols. London: Witherby. Beebe, W. C. 1926. Pheasants: Their Lives and Homes, Vol. 2. Beg, A.R. 1975. Wild habitats of Pakistan. Pak. F. Inst. Bull, 5: 1-57. Bhattacharya, T., and S. Sathyakumar. 2007. An ornithological survey of Chenab Valley, Chamoli district, Uttaranchal, including notes on pheasants. Indi.Birds, 3(4): 138-145. Bibi, F. and Z. Ali. 2013. Measurement of diversity indices of avian communities at Taunsa Barrage Wildlife Sanctuary, Pakistan. J. Agri. Ani. Sci., 23(2): 2013. Bibi, Y., Nisa, S., Chaudhary, F. M., and Zia, M. 2011. Antibacterial activity of some selected medicinal plants of Pakistan. BMC compl. and altern. medi, 11(1): .52. Blashfield, R.K. 1976. Mixture model tests of cluster analysis: Accuracy of four agglomerative hierarchical methods. Phys. Bull.., 83(3): 377. Block, M.W. and L.A. Brennan. 1993. The habitat concept in ornithology. C. Orni. 1987(11): 35-91. Bloom, S.A. 1981. Similarity indices in community studies: Potential Pitfalls. Mari. Eco. 5: 12128 (Progress Series) Bogosian III, V., E.C. Hellgren, M.W. Sears and R.W. Moody. 2012. High resolution niche models via a correlative approach: Comparing and combining correlative and process-based information. Ecol. Model., 237: 63-73. Boitani, L. and T. K. Fuller. 2000. Research Techniques in Animal Ecology. Colum. Uni. Pre., New York. 442 pp.

143

Bolker, B.M., M.E. Brooks, C.J. Clark, S.W. Geange, J.R. Poulsen, M.H. Stevens, and J.S. White. 2009. Generalized linear mixed models: a practical guide for ecology and evolution. T. Ecol. Evol., 24: 127-135. Bonnicksen, T. M. and E. C. Stone. 1985. Restoring naturalness to National Parks. Env. Man., 9(6): 479-485. Boyce, M.S., P.R. Vernier, S.E. Nielsen and F.K. Schmiegelow. 2002. Evaluating resource selection functions. Ecol. M.., 157(2): 281-300. Bradfield, G. E. and N. C. Kenkel. 1987. Nonlinear ordination using flexible shortest path adjustment of ecological distances. Eco. 68:750-3 Braunisch, V. and R. Suchant. 2008. Using ecological forest site mapping for long-term habitat suitability assessments in wildlife conservation— demonstrated for capercaillie (Tetraourogallus). For. Ecol. Man., 256(5): 1209-1221. Bray, J. R. and J. T. Curtis. 1957. An ordination of the upland forest communities of southern Wisconsin. Ecol. mon., 27(4): 325-349. Brooks, T. M., R. A. Mittermeier, G. A. da Fonseca, J. Gerlach, M. Hoffmann, J. F. Lamoreux and A.S. Rodrigues. 2006. Global biodiversity conservation priorities. Scie., 313(5783): 58-61. Brooks, T. M., R. A. Mittermeier, C. G. Mittermeier, G.A.B. Da Fonseca, A.B. Rylands, W. R. Konstant. 2002. Habitat loss and extinction in the hotspots of biodiversity. Cons. Boil. 16 (4): 909-923. Brotons, L., W. Thuiller, M.B. Araújo and A.H. Hirzel. 2004. Presence absence versus presence only modelling methods for predicting bird habitat suitability. Ecog., 27(4): 437-448. Brunckhorst, D. J. 2013. Bioregional planning: resource management beyond the new millennium. Routledge. Buckland, S.T. and D. A. Elston.1993. Empirical models for the spatial distribution of wildlife. J. Appl. Ecol., 30: 478-495. Bump G. and W. H. Bohl. 1961. Red jungle fowl and kalij pheasants (No. 62). US Fish and Wildlife Service. Burrascano, S., W. S. Keeton, F. M. Sabatini and C. Blasi. 2013. Commonality and variability in the structural attributes of moist temperate oldgrowth forests: A global review. F. Ecol. and Manag., 291, 458-479.

144

Butchart, S. H., Walpole, M., Collen, B., Van Strien, A., Scharlemann, J. P., Almond, R. E. and Watson, R. 2010. Global biodiversity: indicators of recent declines. Sci., 328(5982): 1164-1168. Cain, S. A., , C. M. Cottam, I. N. Gabrielson and T. L. Kimball. 1969. Wildlife management in the National Parks. US Department of the Interior, National Park Service, pp. 29-45. Calkins, J. A., T. W. Offield, S. K. Abdullah and S. T. Ali. 1975. Geology of the southern Himalaya in Hazara, Pakistan, and adjacent areas (No. 716C). US Govt. Print. Off. Canfield, R. 1941. Application of the Line Interception Method in Sampling Range Vegetation. J. F..39: 388–394. Carr, G. M. and M. Zimsky. 2009. Tracking progress toward the 2010 biodiversity target and beyond. Sci., 325(5947):1503-1504. Chaghtai, S.M., S. Z. Shah and J. Shah. 1989. Temporal Changes in the Vegetation of Miranjani, Top Gali, Hazara, NWFP, Pakistan. Chahouki, M.A.Z. 2013. Classification and Ordination Methods as a Tool for Analyzing of Plant Communities. INTECH Open Access Publisher. Chahouki, Z.A.M. and A.L. Khalasi. 2012. Predicting potential distributions of Zygophyllum eurypterum by three modelling techniques (ENFA, ANN and logistic) in North East of Semnan, Iran. R. Manag. Agrofor., 33(2): 123- 128. Chahouki, Z.M.A. and K.A.L.H. Azarniv. 2010. Environmental factors affecting distribution of vegetation communities in Iranian Range lands. Veg. j., 23(2): 1-15. Champion, S.H., S.K. Seth and G.M. Khattak. 1965. Forest types of Pakistan. Chatterjee, S. 2015. The rise of birds: 225 million years of evolution. JHU Press. pp 34

Chaudhry, M. J. I., M. Arshad and G. Akbar. 2012. Some Observations on Threatened and Near Threatened avifauna of Pakistan. Rec. Zool. Surv. Pak., 21, 65-72. Chefaoui, R.M. J. Hortal, J.M. Lobo. 2005. Potential distribution modelling, niche characterization and conservation status assessment using GIS tools: a case study of Iberian Coprisspecies. Biol. Cons., 122: 327-338.

145

Chetan, N., K. Praveen and G.K. Vasudeva. 2014. Delineating ecological boundaries of Hanuman langur species complex in peninsular India using Maxent modelling approach. PloSone., 9(2). Clark Labs, 2000. Idrisi 32.02. GIS software package, Clark University. Clarke, K.R. 1993. Non-parametric multivariate analyses of changes in community structure. Aus. J. Ecol., 18: 117-117. Clements, T., A. John, K. Nielsen, D. An, S. Tan, and E. J. Milner-Gulland, 2010. Payments for biodiversity conservation in the context of weak institutions: Comparison of three programs from Cambodia. Ecol. Econ., 69(6): 1283-1291. Collar, N. J., M. J. Crosby, A. J. Stattersfield and Birdlife International. 1994. Birds to watch 2: the world list of threatened birds: the official source for birds on the IUCN red list. B.,L. Int. Crawley, M.J. 2007. Statistics an Introduction using R. John Wiley and Sons Ltd. The Atrium, UK. 327pp. Daily, G. C. and P. A. Matson. 2008. Ecosystem services: From theory to implementation. In Proceedings of the national academy of sciences, 105 (28): 9455-9456. Dasmann, R. F. 1972. Towards a system for classifying natural regions of the world and their representation by National Parks and reserves. Biol. Cons., 4(4): 247-255. Dasti, A.A., S. Saima, A.M. Attiq-ur-Rahman and S.A Malik. 2007. Botanical composition and multivariate analysis of vegetation on the Pothowar Plateau, Pakistan. J. Bot. Res. Inst. Tex., 1(1): 557-568. De Cáceres, M., and S. K. Wiser. 2012. towards consistency in vegetation classification. J. Veg. Sci., 23(2): 387-393. Del Hoyo, J., A. Y. Elliott, J. Sargatal. 1994. Handbook of the birds of the world. New World vultures to guineafowl. Lynx Edicions, Barcelona

Delacour, J. 1977. Pheasants of the World. Spur Publications. Dettmers, R. and J. Bart. 1999. A GIS modelling method applied to predicting forest songbird habitat. Ecol. Appl., 9(1): 152-163. Dickman, A. J. 2010. Complexities of conflict: the importance of considering social factors for effectively resolving human–wildlife conflict. Ani.

146

Cons., 13(5): 458-466. Dixon, P. (2003). VEGAN, a package of R functions for community ecology. J. Veg. Sci., 14(6): 927-930. Eagles, P. F. 2009. Governance of recreation and tourism partnerships in Parks and protected areas. J. Sust. Touri., 17(2): 231-248. Elith, J. and J. R. Leathwick. 2009. Species distribution models: ecological explanation and prediction across space and time. Annual Review of Eco., Evo., and Syst., 40(1): 677. Elith, J., M. Kearney and S. Phillips. 2010. The art of modelling range shifting species. Meth. in ecol. and evol., 1(4): 330-342. Elith, J., S.J. Phillips, T. Hastie, M. Dudík, Y.E. Chee and C.J. Yates. 2011. A statistical explanation of Maxent for ecologists. Divers. Distrib., 17(1): 43- 57. Elphick, C., and J. B. Dunning, (Eds.) 2001. The Sibley guide to bird life & behaviour. New York: Alfred A. Knopf.

Engler, R., A. Guisan and L. Rechsteiner. 2004. An improved approach for predicting the distribution of rare and endangered species from occurrence and pseudo absence data. J. Appl. Ecol., 41(2): 263-274. Enright, N. J., B.P. Miller and R. Akhter. 2005. Desert vegetation and vegetation- environment relationships in Kirthar National Park, Sindh, Pakistan. J. A. Env., 61(3): 397-418. Everitt, B. S., S. Landau and M. Leese. 2001. Cluster Analysis, 4th Edition, Oxford University Press, Inc., New York; Arnold, London. ISBN 0340761199 Evett, R.R., A. Dawson and J. W. Bartolome. 2013. Estimating vegetation reference conditions by combining historical source analysis and soil phytolith analysis at Pepperwood Preserve, northern California Coast Ranges, USA. Rest. Ecol., 21(4): 464-473. Ewald, J. 2003. A critique for phytosociology. J. Veg. Sci., 14(2): 291-296. FAO. 2001. State of the World’s Forests. FAO Rome. Farooque, M. 2002. Management plan of Ayubia National Park 2002–2007. National Resource Conservation Project, Galiat, Abb., Pak.

147

Farooque, M. 2007. Management Plan of Ayubia National Park. Khyber Pakhtunkhwa Wildlife Department, Peshawar. pp 18-20 Farooque, M. 2011. Transforming theory and practice of environmental governance. A case study of Ayubia National Park, Pak. Doctoral Dissertation, University of Waterloo. Franklin, J. 1995. Predictive vegetation mapping: geographic modelling of bio spatial patterns in relation to environmental gradients. Pro. Phys. geo., 19(4): 474-499. Franklin, J. 2009. Mapping Species Distributions: Spatial Inference and Prediction: Cambridge.

Fuller, R.A. 2000. Pheasants: status survey and conservation action plan 20002004 (51). IUCN. Gabriel, A and M.J. Macía. 2013. Local and regional dominance of woody plants along an elevational gradient in a tropical montane forest of northwestern Bolivia. P. Ecol. Gabriela Z, H. Tuomisto, M. M. Jones, J. Prado, F.O.G. Figueiredo, G.M. Moulatlet, F. R. C. Costa, C.A. Quesada and T. Emilio. 2014. Predicting environmental gradients with fern species composition in Brazilian Amazonia. J. Veg. Sci., 25(5): 1195-1207.

Garson, P. J., L. Young and R. Kaul. 1992. Ecology and Conservation of Cheer Pheasant Catreus wallichii: Studies in the Wild and the Progress of a Reintroduction Project. Biol. Con., 59: 23–35. Gaston, A. J., P. J. Garson, and M. L. Jr Hunter. 1981. Present Distribution and Status of Pheasants in Himachal Pradesh, Western Himalayas. J. W. Pheas. Asso., 6: 10–30. Gaston, K.J. 1996. Species-range-size distributions: Patterns, mechanism sand implications. T. Ecol. Evol., 11: 197-201. Gaston, K.J., T.M. Blackburn and R.D. Gregory. 1998. Interspecific differences in intraspecific abundance range size relationships of British breeding birds. Ecog., 21(2): 149-158. Gibson, L., T. M. Lee, L. P. Koh, B. W. Brook, T. A. Gardner, J. Barlow and N. S. Sodhi. 2011. Primary forests are irreplaceable for sustaining tropical biodiversity. Nat., 478(7369), 378-381. 148

GoP and IUCN 1992. The Pakistan national conservation strategy. Pakistan: GOP, Urban Affairs Division and IUCN Pakistan. Grimmett, R., Roberts, T. J., Inskipp, T. and C. Byers. 2008. Birds of Pakistan. Guisan, A. and N. E. Zimmermann. 2000. Predictive habitat distribution model sin ecology. Ecol. Mod., 135: 147–186. Hamilton, A. and P. Hamilton. 2006. Plant conservation: An ecosystem approach. People and plants conservation series. Ear. Sca., UK. Har-Peled, S. and , A. Kushal. 2005. Smaller corsets for k-median and k-means clustering. In Proceedings of the twenty-first annual symposium on Computational geometry (pp. 126-134).ACM. Hirzel, A. H., J. Hausser and N. Perrin. 2007. Biomapper 4.0.Lab.Of Cons.Biol., Department of Ecology and Evolution, University of Lausanne, Switzerland.URL: http://www2.unil.ch/biomapper. Hirzel, A.H. and R. Arlettaz. 2003. Modelling habitat suitability for complex species distributions by environmental distance geometric mean. Env. Man., 32: 614–623. Hirzel, A.H., B. Posse, P.A. Oggier, Y. Crettenand, C. Glenz and R. Arlettaz. 2004. Ecological requirements of reintroduced species and the implications for release policy: the case of the bearded vulture. J. Appl. Ecol., 41(6): 1103-1116. Hirzel, A.H., G. Le Lay, V. Helfer, C. Randin and A. Guisan. 2006. Evaluating the ability of habitat suitability models to predict species presences. Ecol. mod., 199(2): 142-152. Hirzel, A. H., J. Hausser, D. Chessel and N. Perrin. 2002. Ecological niche factor analysis: how to compute habitat suitability maps without absence data. Ecol., (83): 2027-2036. Hirzel, A.H., V. Helfer and F. Metral. 2001. Assessing habitat-suitability models with a virtual species. Ecol. Mod., 145(2): 111-121. Holt, B. G., J. P. Lessard, M. K. Borregaard, S. A. Fritz, M. B. Araújo, D. Dimitrov and C. Rahbek. 2013. An update of Wallace’s zoogeographic regions of the world. Sci., 339 (6115): 74-78. Hooge, P.N. and B. Eichenlaub. 2000. Animal movement extension to Arcview.ver. 2.0. Alaska Science Center – Biological Science Office, U.S.

149

Geo. Sur., Ancho., AK, USA. Hooker, J. D. 2003. Himalayan J. II (Vol. 9).Psycho. P. Howman, K. 1993. Pheasants of the world: their breeding and management. H.H. Publ., Ltd. Hume, A. O., and C. H. T. Marshall. 1879. The Game Birds of India, Burma and Ceylon. Calcutta: Hume and Marshall. Hussain, F. and I. Ilahi. 1991. Ecology and Vegetation of Lesser Himalayas Pakistan. Bot. dept. Uni. of.Pesh., 187. Hussain, F., M. Ahmed, G. Shaheen and M. J. Durrant. 1994. Phytosociology of the vanishing tropical deciduous forest in district Swabi. Pakistan. II. Ordination. Pak. J. Bot, 26(1): 149-160. Hussain, M. S. and A. Sultana. 2013. Ecological separation of habitat variables among five rare pheasant species of the Himalayas, Ind. Zool. Ecol., 23(2): 97-103. Hussain, M. S., J. A. Khan and R. Kaul. 2001. Aspects of ecology and conservation of Kalij Lophura leucomelana and Koklass Pucrasia macrolopha in the Kumaon Himalaya, India. Trop. Ecol., 42(1):59-68. Hussain, S.K. and R. Afza. 2004. Watershed analysis of Ayubia National Park. A Research report, WWF, Pakistan. Hutchinson, G. E. 1957. Concluding remarks. Cold Spring Harbour Symposia on Quanti. Bio., 22: 415-427. Hutchinson, G. E. 1961. The paradox of the plankton. Am. Natur., 137-145. Iqbal, S.M. 1993. Patterns of Habitat Use by White-crested Kalij at Two Sites in the Indian Himalaya. In Pheasants in Asia, edited by D. Jenkins, 154. Reading: PA. Isner, A. 2014. Habitat Suitability Modeling of Mexican Spotted Owl (Strix Occidentalis Lucida) in Gila National Forest, New Mexico. Doctoral Dissertation, University of Southern California. IUCN. 2010. The IUCN red list of threatened species. Retrieved on 10 February,2015,fromhttp://www.iucnredlist.org/documents/summary statistics/2010 IUCN. 1990. IUCN directory of South Asian protected areas. IUCN, Cambridge, U.K.

150

IUCN. 1994. Guidelines for protected area management categories. Gland, Switzerland, and Cambridge, U.K. IUCN. 1996. People in Charge. Special issue on co-management of natural resources of World Conservation, no. 2. IUCN. 2007. IUCN Red List of Threatened Species. IUCN. 2008. IUCN Red List Categories. www.iucnredlist.org. Jacobs, L. L. 1977. Geology and palaeontology of Neogene strata of Pakistan. Jan, A.U., 1993. Forest Policy: Administration and Management in Pakistan. Jepson, P. and R. J. Whittaker. 2002. Ecoregions in context: a critique with special reference to Indonesia. Cons. Biol., 16(1): 42-57. Jiang, W.X., X.C. Fu, T. Tang and Q.H. Cai. 2009. Community structure and niche of macro invertebrates in the Xiangxi River in Hubei, China. Chin. J. Appl. Env. Biol., 15(3): 337-341. Johnsgard, P.A. 1999. The Pheasants of the World: Biology and natural history. Smithsonian Institution Press. Washington, D.C. 398 pp. Jolli, V., A. Srivastav and S. Thakur. 2011. Patch Occupancy for Cheer Pheasant Catreus wallichii in the Great Himalayan National Park Conservation Area .Int. J. Gallifor. Cons., 2:74–81. Jolli, V., and M. Pandit. 2011. Influence of Human Disturbance on the Abundance of Himalayan Pheasant (Aves, Galliformes) in the Temperate Forest of Western Himalaya, India. Vest. Zool., 45(6): e-40. Joshi, G., and Negi, G. C. 2011. Quantification and valuation of forest ecosystem services in the western Himalayan region of India. Int.J. Biod. Sci, E. Serv. & Man., 7(1): 2-11. Joshi, K.K., D. Bhatt and A. Thapliyal. 2012. Avian diversity and its association with vegetation structure in different elevational zones of Nainital district (Western Himalayan) of Uttarakhand. Int. J. Bio. Cons., 4(364): e376. Khan. J., R. Afza, S. U. Din. 2003. Trade of Folk Medicine in Peshawar City, NWFP-Pakistan. Workshop proceeding on Conservation and Sustainable Uses of Medicinal and Aromatic Plants in Pakistan, WWFPakistan.

151

Kamal K.J and D. Bhatt. 2011. Birds of Three Different Forest (Sal, Pine and Oak) Habitats in Nainital District (Western Himalaya) Of Uttarakhand, India. Nat.Sci, 9(7): http://www.sciencepub.net/nature Kaul, R. 2014. Ecology of the cheer pheasant Catreus wallichii in Kumaon Himalaya. Kenkel, N.C. and L. Orlóci. 1986. Applying metric and nonmetric multidimensional scaling to ecological studies: some new results. Eco., 919-928. Kent, M. 2011. Vegetation description and data analysis: a practical approach. John Wiley and Sons. Kent, M. and P. Coker. 1992. Vegetation Description and Analysis: A Practical Approach. Belhaven Press, London. Kerr, J.T. and M. Ostrovsky. 2003. from space to species: ecological applications for remote sensing. T. in Ecol. & Evo., 18(6): 299-305. Khan, F. H. and Faridullah, S. M. 2014. Ordination and Classification of Vegetation in Semi-Arid Area of Pakistan. G. J. Sci. F. Res., 14(2). Khan, N. 2012. A community analysis of Quercus baloot Griff, forest. District Dir, Upper Pakistan. Afr. J. P. Sci., 6(1): 21-31. Khan, N., S.S. Shaukat, M. Ahmed and M.F. Siddiqui. 2013. Vegetationenvironment relationships in the forests of Chitral district Hindukush range of Pakistan. J. For. Res., 24(2): 205-216. Khan, S. M., S. E. Page, H. Ahmad and D. M. Harper. 2013. Sustainable utilization and conservation of plant biodiversity in montane ecosystems: the western Himalayas as a case study. An. of bot., mct 125. Khan, S. M., S. Page, H. Ahmad, D. M. Shaheen and D. M. Harper. 2012. Vegetation dynamics in the Western Himalayas, diversity indices and climate change. Sci., Tech. & Dev, 31(3): 232-243. Khan, S. M., S. Page, H. Ahmad and D. Harper. 2013. Identifying plant species and communities across environmental gradients in the Western Himalayas: Method development and conservation use. Ecol. Inf., 14, 99- 103. Khan, S. M., Page, S., H. Ahmad, Z. Ullah, H. Shaheen, M. Ahmad, and D. M. Harper. 2013. Phyto-climatic gradient of vegetation and habitat

152

specificity in the high elevation Western Himalayas. Pak J. Bot., 45: 223230. Khan, W.M. and I.H. Shah 1982. Population dynamics of Koklass pheasant (Pucrasia macrolopha) in Malkandi Forest, Pakistan. Pheasants in Asia., 1982, 40-43. Khan, Z., M. Akbar, S.M.A. Rizvi, A. Hussain, I. Ali, M. Ali, S. Hyder. 2015. Assessment of deforestation using Diameter size classes distribution of trees in Ganji Valley Himalayan Range of Pakistan. Int. J. 3 (6): 76-86. Khanum, R., A. S. Mumtaz and S. Kumar. 2013. Predicting impacts of climate change on medicinal asclepiads of Pakistan using Maxent modelling. Act. Oecolo., 49: 23-31. Khanum, R. and Gilani, S. A. 2005. Conservational status of plant seedlings in Ayubia National Park, Pakistan. L. On J. Ecol., 8(1): 51-60. Khattak, G.M. 2002. Land tenure and resource ownership in Pakistan: Proceedings of the national workshop (September 2-4, 2002). Prepared by WWF-Pakistan under People and Plants Project. Kindt, R. 2014. Biodiversity R: GUI for biodiversity, suitability and community ecology analysis. Kinnaird, M.F., E.W. Sanderson, T.G. O'Brien, H.T. Wibisono and G. Woolmer. 2003. Deforestation trends in a tropical landscape and implications for endangered large mammals. Cons. Biol., 17: 245–257. Kirby, J. S., A. J. Stattersfield, S. H. Butchart, M. I. Evans, R. F. Grimmett, V. R. Jones and I. Newton. 2008. Key conservation issues for migratory landand water bird species on the world's major flyways. B. Con. Int., 18(1): S49. Krupnick, G. A. and W. J. Kress. 2003. Hotspots and ecoregions: a test of conservation priorities using taxonomic data. Biod. & Cons, 12(11):22372253.

Kruskal, J. B. 1964. Multidimensional scaling by optimizing goodness of fit to a nonmetric hypothesis. Psychomet., 29(1): 1-27.

153

Kruskal, J.B. 1964b. Nonmetric multidimensional scaling: a numerical method. Psychomet. 29:115-29 Lack, D. 1933. Habitat Selection in Birds, with Special Reference to the Effect of Afforestation on the Brekland Avifauna. J. Ani. Ecol., 2: 239– 262. Latif, M. A. 1976. Stratigraphy and micropaleontology of the Galis Group of Hazara, Pakistan. Geol. Bull. Punj. Uni., 13, 1-63. Legendre, P. and L. Legendre. 1998. Numerical Ecology, Second edition: Development in Environmental Modelling. Elsev. Sci., NewYork, USA. Legendre, P. and L.F. Legendre. 2012. Numerical ecology (Vol. 24). Elsevier. Lenton, S. M., J. E. Fa and V. J. Perezdel. 2000. A simple non parametric GIS model for predicting species distribution: endemic birds in Bio koIsland, West Africa.Bio.Cons., 9: 869–885. Lewin, V. and G. Lewin. 1984. The Kalij pheasant, a newly established game bird on the island of Hawaii. The Wil. Bull., 634-646. Lewis, C. 1996. Managing conflicts in protected areas. IUCN, Gland, Switzerland, and Cambridge, UK. Li, G., D. Lu, E. Moran and S. Hetrick. 2011. Land-cover classification in a moist tropical region of Brazil with Land sat Thematic Mapper imagery. Int. j. remo. Sen., 32(23): 8207-8230. Lisón, F. and J. F. Calvo. 2013.Ecological niche modelling of three pipistrelle bat species in semiarid Mediterranean landscapes. A. Oecol., 47: 68-73. Lodhi, A. 2007. Conservation of leopards in Ayubia National Park, Pakistan. Mater Dissertation. The University of Montana Missoula.

MacArthur, R. and R. Levins. 1967. The limiting similarity, convergence, and divergence of coexisting species. A. natur, 377-385. MacArthur, R. H. 1968. The Theory of Niche. Population Biology and Evolution. Syracuse, NY: Syracuse Univ. Press. Malik, M. M. 2004. Wildlife conservation through the management of available human resources in NWFP, Pakistan. Doctoral Dissertation. The University of Montana, USA. Malik, R. N., and S. Z. Husain. 2008. Linking remote sensing and ecological vegetation communities: a multivariate approach. Pak. J. Bot, 40(1): 337349.

154

Mallon, D. 1991. Biodiversity guide to Pakistan. IUCN and World Conservation Monitoring Center. Cambridge, UK. Mani, A. 1981.The climate of the Himalaya. The Himalaya: Aspects of Change. Oxf. Uni. P., Delhi, 115. Mani, M. S. 1974. Biogeography of the Himalaya. In Ecology and Biogeography in India .Spri. Neth., 664-681 Martens, J., D. T. Tietze and M. Päckert, 2011. Phylogeny, biodiversity, and species limits of passerine birds in the Sino-Himalayan region–a critical review. Ornitho. Monog., 70(1): 64-94. Marthews, T. R., S. J. Dadson, B. Lehner, S. Abele, and N. Gedney. 2015. Highresolution global topographic index values for use in large-scale hydrological modelling. Hydrol. Earth Syst. Sci., 19(1): 91-104. Mayers, J. and S. Bass. 2004. Policy that works for forests and people: real prospects for governance and livelihoods. Policy that works for forests and people: real prospects for governance and livelihoods. McGowan, P. J. K. and P. J. Garson, 1995. Status Survey and Conservation Action Plan 1995-1999. IUCN. Gland, Switzerland. UK. 116 pp (24). McGowan, P. and S. Madge. 2010. Pheasants, Partridges & : Including buttonquails, sandgrouse and allies. Bloomsbury Publishing. Middlemiss, C.S. 1896. The geology of Hazara and the black mountains. Memoirs of the Geological Survey of India 26: 88–177. Miller, A. H. 1942. Habitat selection among higher vertebrates and its relation to intraspecific variation. Amer. Natu., 25-35. Miller, J. R. 2009. New vocalizations by the Koklass pheasant Pucrasia macrolopha. Miller, J.R.B. 2010. Survey of Western Tragopan, Koklass Pheasant, and Populations in the Great Himalayan National Park, Himachal Pradesh, India. Ind. Bird., 6(3): 60–65. Milligan, G.W. 1979. Ultra-metric Hierarchical Clustering Algorithms. Psychomet. 44(3): 343–346. Mirza, Z. B. 1979. Satus of the pheasants in Pakistan. Pheas. in Asi., 27-28. Mirza, Z.B., A. Aleem and M. Asghar. 1978. Pheasant surveys in Pakistan. J. Bomb. Nat. Hist. Soc., 75: 292-296.

155

Moreau, R. E. (1934). A contribution to tropical African bird-ecology. The J. Ani. Ecol., 41-69.NewYork, NY: Doubleday, Page. Moskt, C. 1991. Multivariate plexus concept in the study of complex ecological data: an application of the analysis of bird-habitat relationships. Conenoses 6:79-89. Mueller-Dombois, D., and H. Ellenberg. 2013. Aims and methods of vegetation ecology. Geog.Revi. Muhammad, I., R. Afza and S. Yousaf . 2006. Ethno pharmacological survey of plants used for the treatment of stomach, diabetes, and ophthalmic diseases in Sudhan Gali, Kashmir, Pakistan. 云南植物研究, 28(5): 535-

542.

Murtagh, F. and P. Legendre. 2013. Ward's hierarchical agglomerative clustering method: which algorithms implement Ward's criterion? J. Classifi., (in press). Myers, N., R.A. Mittermeier, C. G. Mittermeier, G. A. Da Fonseca and J. Kent. 2000. Biodiversity hotspots for conservation priorities. Natur, 403(6772): 853-858. N.C. kenkel and L. Orloci 1986: Applying metric and nonmetric multidimentional scaling to ecological studies: some new results by Ecological society of America Nadeem, M., Shinwari, Z. K., and Qaiser, M. 2013. Screening of folk remedies by genus Artemisia based on ethnomedicinal surveys and traditional knowledge of native communities of Pakistan. Pak. J. Bot, 45(S1): 111117. Naidoo, R., A. Balmford, R. Costanza, B. Fisher, R.E. Green, B. Lehner and T.H. Ricketts. 2008. Global mapping of ecosystem services and conservation priorities. Proc. Nat. Acad. Sci., 105(28): 9495-9500. Nan, W., Z. Zhengwang, Z. Guangmei and P.J. Mcgowan. 2004. Relative density and habitat use of four pheasant species in Xiaoshennongjia Mountains, Hubei Province, China. Bird Cons. Int., 14(01): 43-54. Nawaz, M.A., J. Martin and J.E. Swenson. 2014. Identifying key habitats to conserve the threatened brown bear in the Himalaya. Biol. Con., 170: 198- 206.

156

Nawaz, R., P. J. Garson and M. Malik. 2000. Monitoring pheasant populations in montane forest: some lessons learnt from the Pakistan Galliformes project. In Proc. 2nd Int. Gallifo. Symp, W. Pheas. Assoc., Reading. 196-209. Negi, S. S. 2000. Himalayan forests and forestry. Indus Publishing. Nelson, E., G. Mendoza, J. Regetz, S. Polasky, H. Tallis, D. Cameron and M. Shaw. 2009. Modelling multiple ecosystem services, biodiversity conservation, commodity production, and trade-offs at landscape scales. Front. in Ecol. & the Env., 7(1): 4-11. Oksanen, J., and P. Huttunen. 1989. Finding a common ordination for several data sets by individual differences scaling. Veget. 83:137-45 Oksanen, J., R. Kindt, P. Legendre, B. O. Hara, M.H.H. Stevens, M.J. Oksanen and M.A.S.S. Suggests. 2007. The vegan package. Com. ecol. Packa, 631637. Olson, D. M., and E. Dinerstein. 1998. The Global 200: a representation approach to conserving the Earth’s most biologically valuable ecoregions. Cons. Biol., 12(3): 502-515. Olson, D.M. and E. Dinerstein. 2002. The Global 200: Priority ecoregions for global conservation. Ann. Missouri Bot. Gard., 199-224. Orlóci, L. 2013. Multivariate analysis in vegetation research. Springer. Owen, L. A. 1996. Quaternary lacustrine deposits in a high energy semi-arid mountain environment, Karakoram Mountains, northern Pakistan. J. Quatern. Sci., 11(6): 461-483. Owen, L. A. and J. England (1998).Observations on rock glaciers in the Himalayas and Karakoram Mountains of northern Pakistan and India. Geomor., 26(1): 199-213. Parker, V. 1999. The use of logistic regression in modelling the distributions of bird species in Swaziland, S. Afr. J. Zool., 34: 39-47. Pearson, R.G., C.J. Raxworthy, M. Nakamura and A. Townsend Peterson. 2007. Predicting species distributions from small numbers of occurrence records: a test case using cryptic geckos in Madagascar. J. Biogeogra., 34(1): 102-117.

157

Peer, T., J.P. Gruber, A. Millingard and F. Hussain. 2007. Phyto-sociology, structure and diversity of the steppes vegetation in the mountains of Northern Pakistan. Phytocoenolo, 37(1): 1-65. Peer, T., A. Millinger, J. P. Gruber, and F. Hussain. 2001. Vegetation and altitudinal zonation in relation to the impact of grazing in the steppe lands of the Hindu Kush Range (N-Pakistan). Phytocoe., 477-498. Peterson, A.T. (2011). Ecological niche conservatism: A time structured review of evidence. J. of Biog., 38(5): 817-827. Peterson, M. N., Birckhead, J. L., Leong, K., Peterson, M. J., & Peterson, T. R. 2010. Rearticulating the myth of human–wildlife conflict. Cons. Lett., 3(2):74-82. Phillips, S. J., R. P. Anderson and R. E. Schapire. 2006. Maximum entropy modelling of species geographic distributions. Ecol. Model., 190(3):231259. Phillips, S. J. 2008. Transferability, sample selection bias and background data in presence only modelling: a response to Peterson et al. 2007. Ecog., 31(2): 272-278. Phillips, S.J. and M. Dudík. 2008. Modelling of species distributions with Maxent: new extensions and a comprehensive evaluation. Ecog., 31(2): 161-175. Phillips, S.J., R.P. Anderson and R.E. Schapire. 2006. Maximum entropy modelling of species geographic distributions. Ecol. Model., 190(3): 231259.

Pianka, E. R. 1981. Competition and niche theory. Theor. Ecol. prin & appli, 167- 196. Podani, J. 2006. Braun-Banquet’s legacy and data analysis in vegetation science. J. Veg. Sci., 17(1): 113-117. Qaimkhani, A.M. 2009. Pakistan. Fourth National Report, Ministry of Environment, Government of Pakistan, Islamabad. Qazi, I.A. 1994. Pakistan: country and forests. Handbook of Forestry, 253. Quétier, F. and S. Lavorel. 2011. Assessing ecological equivalence in biodiversity offset schemes: key issues and solutions. Biol. Cons., 144(12): 2991-2999.

158

Rands, M. R., Adams, W. M., Bennun, L., Butchart, S. H., Clements, A., Coomes, D., and Vira, B. 2010. Biodiversity conservation: challenges beyond 2010. Sci., 329(5997): 1298-1303.

Rattray, R.H. 1905: Birds nesting in the Murree hills and gullies. J. B. Nat. Hist. Soc., 16(4): 657-663. Robert, T. J. 1991. The birds of Pakistan. Vol. 1. Regional studies and NonPasseriformes. Oxford University Press, Oxford New York, Karachi, 598. Roberts, T. J. 1970. A note on the pheasants of West Pakistan. Pak. J. Fores, 20(4): 319-326. Rodrigues, A. S., Akcakaya, H. R., Andelman, S. J., Bakarr, M. I., Boitani, L., Brooks, T. M. and Yan, X. 2004. Global gap analysis: priority regions for expanding the global protected-area network. Bio Sci., 54(12): 10921100. Round, P.D and G.A. Gale. 2008. Changes in the status of Lophura pheasants in KhaoYai National Park, Thailand: a response to warming climate? Biotrop., 40: 225–230. Saeed, U. 2008. Boundary delineation of Ayubia National Park. Boundary demarcation and renotification of protected areas project. GIS Laboratory, WWF-Pakistan, Lahore. Saima, S., Dasti, A. A., Hussain, F., Wazir, S. M., and Malik, S. A. 2009. Floristic compositions along an 18-km long transect in Ayubia National Park district Abbottabad, Pakistan. Pak. J. Bot., 41(5):2115-2127. Saqib, Z., Malik, R. N., and Von Wehrden, H. (2013). Land cover drnaics in relation to Western Tragopan occurrence in Pakistan: A regional assessment. Pak. J. Bot., 45(S1):551-559. Saqib, Z., Malik, R. N., Shinwari, M. I., and Shinwari, Z. K. 2011. Species richness, ethnobotanical species richness and human settlements along a Himalayan altitudinal gradient: Prioritizing Plant Conservation in Palas Valley, Pakistan. Pak. J. Bot., 43, 129-133. Saqib, Z., R.N. Malik and H. Von Wehrden. 2013. Landcover dynamics in relation to western tragopan occurrence in Pakistan: a regional assessment. Pak. J. Bot., 45(S1): 551-559. Sathyakumar, S. and R. Kaul. 2007. Pheasants. Galliformes of India. ENVIS Bulletin: Wild & Prot. Are, 10(1): 41.

159

Sathyakumar, S., S. N. Prasad, G.S. Rawat and A. J.T. Johnsingh. 1993. Ecology of Kalij and Monal in Kedarnath Wildlife Sanctuary. In Pheasants in Asia 1993, edited by D. Jenkins, 83–90. Reading: WPA. Schild, A. 2008. ICIMOD's position on climate change and mountain systems: the case of the Hindu Kush-Himalayas. Mount. Res & Deve., 28(3): 328331. Schulze, E. D., and Mooney, H. A. (Eds.). 2012. Biodiversity and ecosystem function. Springer Science and Business Media. Sclater, P. L. 1858. On the general geographical distribution of the members of the class Aves. J. of the Proceedings of the Linnean Society of London. Zool., 2(7): 130-136.

Semwal, R., Nautiyal, S., Sen, K. K., Rana, U., Maikhuri, R. K., Rao, K. S., and Saxena, K. G. 2004. Patterns and ecological implications of agricultural land-use changes: a case study from central Himalaya, India. Agr., Ecos. & Env., 102(1): 81-92. Sérgio, C., R. Figueira, D. Draper, R. Menezes and A. J. Sousa. 2007. Modelling bryophyte distribution based on ecological information for extent of occurrence assessment. Biol. Con., 135(3): 341-351. Severinghaus, S. R. 1979. Observations on the Ecology and Behaviour of the Koklass Pheasant in Pakistan. J. W. Pheas. Asso., 4: 52–69. Shafique, T. 1999. Nesting Record of White crested Kalij. W. Pheas Asso. News 60: 17. Shah, A., M. S. Khan, M. S. Awan, R. A. Minhas, N. Akhtar, K. Saeed. 2013. Spatial distribution and nidification record of Himalayan Monal and Koklass Pheasants in Kalam district Swat, KP, Pakistan. Shah, S.A. 2011. Conservation of endangered species in Khyber Pakhtunkhwa. J. Ani & P. Sci., 21(2): 400-404. Shaheen, H., Z. Ullah, S.M. Khan, and D.M. Harper. 2012. Species composition and community structure of western Himalayan moist temperate forests in Kashmir. For. Ecol. Manage., 278: 138-145. Shanon, C. E., and Weaver, W. 1949. The mathematical theory of communication. Urbana, Illinois: University of Illinois. 125 pp.

160

Sharma, D. K., and A. C. Saklani. 1993a. Biological Approach to the Conservation of Himalayan Pheasants. In Himalayan Biodiversity, edited by U. Dhar, 459–468. Nain- ital: Gyanodaya Prakashan. Sharma, D. K., and A. C. Saklani. 1993b. Reproductive Ecology of Whitecrested Kaleej in Garhwal Himalaya. In Pheasants in Asia 1992, edited by D. Jenkins, 156. Reading: WPA. Shehzad, W., M. A. Nawaz, F. Pompanon, E. Coissac, T. Riaz, S. A. Shah, and P. Taberlet 2015. Forest without prey: livestock sustain a leopard Panthera pardus population in Pakistan. Oryx, 49(02): 248-253. Siddiqui, M. F., M. Ahmed, S. S. Shaukatand, N. Khan. 2010. Advanced multivariate techniques to investigate vegetation-environmental complex of Pine forests of moist temperate areas of Pakistan. Pak. J. Bot., S. issu., 42, 267-293. Siddiqui, M. F., M. Ahmed, S. S. Shaukat and M. Ajab. 2011. Soil and foliar nutrients concentration of conifer species in the communities of moist temperate areas of southern Himalayan and Hindukush region of Pakistan. F. J. Biol, 1(1): 91-101. Singh, R. S., P. K. Rahlan, and S. P. Singh. 1987. Phytosociological and Population Structure of Mixed Oak Conifer Forest in a Part of Kumaon Himalaya. Envi. Ecol., 5: 475–487. Smith, R. J., R. D. Muir, M. J. Walpole, A. Balmford and N. Leader-Williams. 2003. Governance and the loss of biodiversity. Natu., 426(6962): 67-70. Smith, S. A., and M. J. Donoghue. 2010. Informing large-scale biogeography with niche models in Lonicera (Caprifoliaceae, Dipsacales) subgenus Caprifolium. Syst. Biol., 590, 322-341. Store, R. and J. Jokimäki. 2003. A GIS-based multi-scale approach to habitat suitability modelling. Ecol. Model., 169(1): 1-15. Subhani, A., M. S. Awan, M. Anwar, U. Ali, and N. I. Dar. 2010. Population status and distribution pattern of red jungle fowl (Gallus gallusmurghi) in Deva Vatala National Park, Azad Jammu & Kashmir, Pakistan: A pioneer study. Pak. J. Zool., 42(6): 701-706. Sukumal, N. and T. Savini, 2009. Altitudinal differences in habitat use by Lophura diardi and Lophura

161

nycthemera in KhaoYai National Park, Thailand. Int. J. Gallif. Cons., 1: 18–22. Sutherland, W. J., R. Aveling, T. M. Brooks, M. Clout, L. V. Dicks, L. Fellman, and A. R. Watkinson. 2014. A horizon scan of global conservation issues for 2014. T. ecol. & evolution, 29(1): 15-22. Tallis, H., P. Kareiva, M. Marvier and A. Chang, 2008. An ecosystem services framework to support both practical conservation and economic development. Proceedings of the National Academy of Sciences, 105(28): 9457-9464. Tariq, A., M. Adnan, N.M. Salam, H. Fouad, K. Hussain, R. Ullah and A. Ullah. 2014. Richness and Cover of Non-timber Economic Plants along Altitude in Temperate Himalayan Forest-Use Types. The Sci. W. J., 2014. Tavili, A. and M. Jafari. 2009. Interrelations between plants and environmental variables. Thomas, G. B. 2004. Landscape ecology and ecosystem management. FOR-76 UK, University of Kentucky, College of Agriculture, Lexington, and Kentucky State University, Frankfort. Walpole, M., R. E. Almond, C. Besançon, , S. H. Butchart, D. CampbellLendrum, G. M. Carr and M. Zimsky. 2009. Tracking progress toward the 2010 biodiversity target and beyond. Sci., 325(5947):1503-1504. Wang, W., B. D. McKay, C. Dai, N. Zhao, R. Zhang, Y. Qu, and F. Lei. 2013. Glacial expansion and diversification of an East Asian montane bird, the green backed tit (Parusmonticolus). J. Biogeo., 40(6): 1156-1169. Ward, A.E. 1923. Small game shooting in Kashmir and the adjacent hill provinces. J. B. Nat. Hist. Soc., 29: 653-658. Ward, J. H. 1963. Hierarchical Grouping to Optimize an Objective Function. J. Amer. Statist. Asso., 58: 236–244. Warren, D.L. and S.N. Seifert. 2011. Ecological niche modelling in Maxent: the importance of model complexity and the performance of model selection criteria. Ecol. App., 21(2): 335-342. Waseem, M., Shah, M. A. U., Qureshi, R. A., Muhammad, I., Afza, R., and Yousaf, S. 2005. Ethno pharmacological survey of plants used for the

162

treatment of stomach, diabetes, and ophthalmic diseases in Sudhan Gali, Kashmir, Pakistan. Acta Bot. Yunnan., 28(5): 535-542. Wazir, S.M., A.A. Dasti, S. Saima, J. Shah and F. Hussain. 2008. Multivariate analysis of vegetation of Chapursan Valley: An alpine meadow in Pakistan. Pak. J. Bot., 40 (2): 615-626. Wisz, M. S., R. J. Hijmans, J. Li, A.T. Peterson, C.H. Graham and A. Guisan. 2008. Effects of sample size on the performance of species distribution models. Diver. & Distri., 14(5): 763-773. Yang, J., P. Reichert, K. C. Abbaspour, J. Xia, and H. Yang. 2008.Comparing uncertainty analysis techniques for a SWAT application to the Chaohe Basin in China. J. Hydro., 358(1):1-23. Zahid, N. 2007. Vegetation structure and diversity of moist temperate rangelands of Pir-Chinasi hills, District Muzaffarabad. Doctoral Dissertation.PMAS-Arid Agriculture University, Rawalpindi. Zaman, I. 2008. Conservation of Pheasants in North West Frontier Province (NWFP). Pakistan, Master of Science Dissertation in Wildlife Biology. The Univ. of Mont. Miss., MT., USA. Zaniewski, A. E., A. Lehmann and J. M. Overton. 2002. Predicting species spatial distributions using presence-only data: a case study of native New Zealand ferns. Ecol. mod., 157(2):261-280. Zeitler, P. K. 1985. Cooling history of the NW Himalaya, Pakistan. Tect., 4(1):127-151. Zhou, T., P. Wang and F. Han. 1995. A preliminary study on the ecology of Koklass pheasant (Pucrasia macrolopha) in Guanshan Forests. Zool. Res., Dong wuxueyanjiu" bianjiwei yuan hui bianji., 17(1): 52-58.

163

6 ANNEXTURE

Annexure: Synoptic table of communities resulting from Bray-Curtis classification showing the frequencies of individual species, number of sampling plots (Plant Community I = PVV; Community II = AVD; Community III = PFI)

Frequency of occurrence of Species PVV AVD PFI

1 210 Abel.tri Abelia triflora R.Br. ex Wall. Caprifoliaceae 0.05 0.05 0.00

2 5, 35, Abie.pin Abies pindrow (Royle ex D. Pinaceae 0.72 0.86 0.68 192, Don) Royle 310 Syn: Abies webbiana var. pindrow (Royle ex D. Don) Brandis 3 26, Acer.cae Acer caesium Wall. ex Aceraceae 0.38 0.82 0.56 107, Brandis 333 Syn: Acer molle Pax 4 357, Achi.mil Achillea millefolium L. Compositae 0.01 0.14 0.12 148 5 78, Acon.vio Aconitum violaceum Ranunculaceae 0.05 0.05 0.03 161 Jacquem. ex Stap f. 6 11, Colu.lon Coluria longifolia Maxim. Rosaceae 0.24 0.36 0.21 58, Syn: Geum elatum 274 var. humile (Royle) Hook.f. 7 189 Anap.bus Anaphalis busua Compositae 0.01 0.00 0.21 (Buch.Ham.) DC. Syn: Anaphalis arenosa DC.

8 95 Acon.mol Aconogonon molle (D. Don) Polygonaceae 0.03 0.05 0.09 H. Hara Syn: Polygonum molle D. Don, Prodr. 9 15, Adia.ven Adiantum venustum D. Don Pteridaceae 0.69 0.45 0.12 332 10 99 Aegi.cyli Aegilops cylindrica Host. Poaceae 0.11 0.27 0.12

11 56 Aesc.ind Aesculus indica (Wall. ex Sapindaceae 0.35 0.05 0.38 Cambess.) Hook. Syn: Pavia indica Wall. ex Cambess. 12 10, Ager.hou Ageratum albidum (DC.) Compositae 0.04 0.18 0.00 201 Hemsl Syn: Ageratum albidum var. albidum 13 72 Agro.pil Agrostis pilosula Trin. Poaceae 0.08 0.05 0.12

164

14 109 Agro.sto Agrostis stolonifera L. Poaceae 0.07 0.09 0.18

15 38 Agro.vin Agrostis vinealis Schreb. Poaceae 0.08 0.09 0.09

16 301 Aila.alt Ailanthus altissima (Mill Simaroubaceae 0.01 0.00 0.03 Swingle 17 302 Ains.apt Ainsliaea aptera DC. Compositae 0.05 0.00 0.03

18 97 Ajug.int Ajuga integrifolia Buch.- Lamiaceae 0.08 0.00 0.24 Ham. Syn: Ajuga bracteosa Wall. ex Benth. 19 25 Ajug.par Ajuga parviflora Benth. Lamiaceae 0.08 0.09 0.00

20 89 Alnu.nit Alnus nitida (Spach) Endl. Betulaceae 0.01 0.05 0.03 Syn:Clethropsis nitida Spach.

21 191 Alce.ros Alcea rosea L. Malvaceae 0.03 0.00 0.00 Syn: Althea rosa L. 22 33. Anap.bus Anaphalis busua Compositae 0.04 0.00 0.06 142 (Buch.Ham.) DC. 23 88,34 Andr.foli Androsace foliosa Duby Primulaceae 0.08 0.14 0.09 2 24 13, Andr.haz Androsace hazarica R.R. Primulaceae 0.08 0.05 0.06 354, Stewart ex Y.J. Nasir 184 25 46, Andr.rot Androsace rotundifolia Primulaceae 0.19 0.14 0.09 213 Hardw.

26 3, 119 Anem.tet Anemone tetraflora Primulaceae 0.05 0.05 0.06

27 87, Anis.ind Anisomeles indica (L.) Lamiaceae 0.07 0.00 0.00 254, Kuntze 348 28 152, Ante.fil Antenoron filiforme (Thunb.) Polygonaceae 0.03 0.05 0.00 190 Roberty & Vautier 29 22, Aplu.mut Apluda mutica L. Poaceae 0.05 0.55 0.03 264 30 283, Aqui.fra Aquilegia fragrans Benth. Rananculaceae 0.07 0.55 0.00 225 31 20, Aqui.pub Aquilegia pubiflora Wall. ex Ranunculaceae 0.18 0.00 0.18 284 Royle Syn: Aquilegia pubiflora var. hazarica Qureshi & Chaudhri 32 235 Aren.ser Arenaria serpyllifolia Bourg. Caryophyllaceae 0.00 0.05 0.06 ex Willk. & Lange 33 66, Aris.jac Arisaema jacquemontii Araceae 0.19 0.09 0.12 279 Blume. 34 250, Aris.uti Arisaema utile Hook. f. ex Araceae 0.05 0.05 0.06 303 Schott 35 111, Arte.inc Artemisia incisa Pamp. Compositae 0.14 0.09 0.12 270

165

36 82 Arte.rox Artemisia roxburghiana Wall. Compositae 0.01 0.14 0.03 ex Besser 37 117 Arum.jac Arum jacquemontii Blume Araceae 0.08 0.05 0.06 38 129 Aspa.fil Asparagus filicinus Asparagaceae 0.14 0.14 0.09 Buch.Ham. ex D. Don

39 36 Aste.fal Aster falcifolius Compositae 0.04 0.00 0.00 HandelMazzetti 101 Aple.mur Apluda muricata L Poaceae 0.13 0.18 0.09 40 91 Atro.acu Atropa acuminata Royle ex Solanaceae 0.01 0.09 0.06 Lindl. 41 143, Berb.kun Berberis kunawurensis Royle Berberidacea 0.22 0.18 0.06 305 42 45,85, Berb.par Berberis Parkeriana C.K. Berberidaceae 0.03 0.00 0.06 140 Schneid. 43 21,44, Berg.cil Bergenia ciliata (Haw.) Saxifragaceae 0.03 0.18 0.03 69 Sternb. 44 51,18 Berg.str Bergenia stracheyii (Hook. f. Saxifragaceae 0.03 0.00 0.12 2, 202 & Thomson) Engl. 45 98, 153 Betu.uti Betula utilis D. Don Betulaceae 0.01 0.05 0.09

46 322 Bide.chi Bidens pilosa L. Compositae 0.08 0.00 0.00 47 351 Both.bla Bothriochloa bladhii (Retz.) Poaceae 0.07 0.05 0.15 S.T. Blake 48 265 Both.isc Bothriochloa ischaemum (L.) Poaceae 0.07 0.05 0.03 Keng 49 30 Brom.cat Bromus catharticus Vahl. Poaceae 0.01 0.14 0.09

50 319 Brom.hor Bromus hordeaceus L. Poaceae 0.54 0.27 0.41 51 330 Brom.pec Bromus pectinatus Thunb. Poaceae 0.04 0.05 0.00

52 36 Brom.por Bromus porphyranthos Cope Poaceae 0.15 0.00 0.15

53 23, 245 Bupl.ham Bupleurum hamiltonii N.P.Ba Apiaceae 0.04 0.14 0.09 lakr. 54 63, 255 Call.pim Callianthemum pimpinelloides Ranunculaceae 0.08 0.05 0.06 (D.Don ex Royle) Hook.f. & Thomson 55 352 Calt.alb Caltha palustris var. alba Rananculaceae 0.03 0.00 0.03 (Cambess.) Hook.f. & Thomson 56 226, Cann.sat Cannabis sativa L. Cannabaceae 0.03 0.00 0.03 325 57 50 Care.fil Carex filicina Nees Cyperaceae 0.16 0.05 0.06

58 127 Care.wal Carex wallichiana Spreng. Cyperaceae 0.00 0.18 0.00 59 55 Cedr.deo Cedrus deodara (Roxb. ex Pinaceae 0.16 0.41 0.24 D.Don) G.Don 60 77 Celt.aus Celtis australis subsp. caucasi Cannabaceae 0.03 0.09 0.00 ca (Willd.) C. C. Towns.

166

61 14 Cera.fon Cerastium fontanum Baumg Caryophyllaceae 0.03 0.00 0.03 62 139 Chry.gry Chrysopogon gryllus (L.) Poaceae 0.09 0.05 0.09 Trin. 63 235 Cich.int Cichorium intybus L. Compositae 0.05 0.00 0.06 64 51, 341 Clem.buc Clematis buchananiana DC. Ranunculaceae 0.05 0.00 0.09

65 108 Clem.cor Clematis catesbyana Pursh Ranunculaceae 0.15 0.09 0.18

66 121 Clem.gra Clematis graveolens Lindl. Ranunculaceae 0.15 0.00 0.06

67 48 Clem.mon Clematis montana Buch.Ham. Ranunculaceae 0.03 0.05 0.00 ex DC. 68 151, Clin.vul Clinopodium vulgare L. Lamiaceae 0.00 0.18 0.00 162 69 115 Clin.hyd Clinopodium hydaspidis (Fal Lamiaceae 0.00 0.18 0.00 ex Benth.) Kuntze 70 40 Coni.mac Conium maculatum L. Apiaceae 0.01 0.05 0.03

71 7, 320 Corn.mac Cornus macrophylla Wall. Cornaceae 0.24 0.27 0.18 72 140, Coto.bac Cotoneaster bacillaris Wall. Rosaceae 0.07 0.09 0.06 141 ex Lindl. 73 131 Cruc.him Crucihimalaya himalaica (Ed Brassicaeae 0.32 0.23 0.03 gew.) Al-Shehbaz, O'Kane & R.A. Price Syn: Arabidopsis brevicaulis (Jafri) Jafri 74 171 Cymb.dis Cymbopogon distans Poaceae 0.01 0.00 0.03 (Nees ex Steud.) W. Watson 75 60 Cyno.lan Cynoglossum lanceolatum Boraginaceae 0.15 0.18 0.06 Forssk. 76 65 Cype.cyp Cyperus cyperoides (L.) Cyperaceae 0.03 0.00 0.12 Kuntze 77 81 Dact.glo Dactylis glomerata L. Poaceae 0.15 0.05 0.03 78 355 Dact.hat Dactylorhiza hatagirea (D. Orchidaceae 0.00 0.05 0.09 Don) Soó 79 137 Daph.pap Daphne papyracea Wall. ex Thymelaeaceae 0.01 0.18 0.00 G. Don 80 120 Datu.str Datura stramonium L. Solanaceae 0.01 0.00 0.00

81 70 Delp.ves Delphinium vestitum Wall. Ranunculaceae 0.11 0.05 0.09 ex Royle 82 324 Desm.ele Desmodium elegans DC. Leguminosae 0.12 0.18 0.12

83 236 Digi.san Digitaria sanguinalis (L.) Poaceae 0.11 0.00 0.15 Scop. 84 118 Dios.del Dioscorea deltoidea Wall. ex Dioscoreaceae 0.05 0.09 0.06 Griseb. 85 52 Dios.lot Diospyros lotus L. Ebenaceae 0.08 0.05 0.00

167

86 92 Dips.ine Dipsacus inermis Wall. Caprifoliaceae 0.07 0.00 0.06

87 105 Drab.ore Draba oreades Schrenk Brassicaceae 0.04 0.00 0.15 88 136 Dryo.ste Dryopteris ramosa Dryopteridaceae 0.72 0.73 0.21 (C. Hope) C. Chr. 89 71 Duch.ind Duchesnea indica (Jacks.) Rosaceae 0.03 0.05 0.03 Focke 90 6, 110 Echi.cor Echinops cornigerus DC. Compositae 0.03 0.09 0.00 91 17, Epip.hel Epipactis helleborine (L.) Orchidaceae 0.08 0.05 0.06 337 Crantz 92 52 Epip.per Epipactis persica (Soó) Orchidaceae 0.03 0.05 0.00 Hausskn. ex Nannf. 93 181 Erig.roy Erigeron roylei DC. Compositae 0.01 0.00 0.03 94 314 Euon.fim Euonymus fimbriatus Wall. Celastraceae 0.00 0.09 0.00

95 125 Euon.ham Euonymus hamiltonianus Celastraceae 0.05 0.00 0.03 Wall. 96 34, Euph.hel Euphorbia helioscopia L. Euphorbiaceae 0.14 0.00 0.00 349 97 128 Euph.wal Euphorbia wallichii Hook. f. Euphorbiaceae 0.16 0.14 0.18 98 85 Ficu.pal Ficus palmata Forssk. Moraceae 0.01 0.05 0.03

99 61, Frag.nub Fragaria nubicola (Lindl. ex Rosaceae 0.78 0.64 0.59 246 Hook.f.) Lacaita 100 358 Frax.exc Fraxinus excelsior L. Oleaceae 0.00 0.00 0.06

101 275 Fuma.ind Fumaria indica Papaveraceae 0.03 0.05 0.00 (Hausskn.) Pugsley 102 217, Gali.ele Galium elegans Rubiaceae 0.27 0.23 0.35 340 Wall. ex Roxb. 103 49 Gali. Sub Galium subfalcatum Nazim. Rubiaceae 0.08 0.00 0.09 & Ehrend. 104 138 Gent.kur Gentiana kurroo Royle Gentianaceae 0.03 0.09 0.15 105 227, Gera.luc Geranium lucidum L. Geraniaceae 0.04 0.05 0.06 356 106 63, Gera.wal Geranium wallichianum D. Geraniaceae 0.42 0.23 0.12 256 Don ex Sweet 107 90 Gerb.gos Gerbera gossypina (Royle) Compositae 0.11 0.18 0.00 Beauverd 108 102 Hede.nep Hedera nepalensis K. Koch. Araliaceae 0.5 0.45 0.12 109 135 Hera.can Heracleum candicans Wall. Apiaceae 0.04 0.00 0.12 ex DC. 110 16 Hypr.per Hypricum perfuratum L. Hypericaceae 0.04 0.00 0.12

111 80 Impa.bic Impatiens bicolor Royle Balsaminaceae 0.32 0.18 0.06

112 19 Indi.het Indigofera heterantha Brandis Papilionaceae 0.39 0.27 0.29 113 176, Iris.hoo Iris hookerana Foster Iridaceae 0.01 0.05 0.15 177

168

114 27 Isod.coe Isodon coetsa (Buch.-Ham. Lamiaceae 0.2 0.05 0.21 ex D. Don) Kudô 115 126 Isod.rug Isodon rugosus (Wall. ex Lamiaceae 0.19 0.05 0.03 Benth.) Codd 116 196 Jasm.hum Jasminum humile L. Oleaceae 0.11 0.09 0.09

117 206 Jugl.reg Juglans regia L. Juglandaceae 0.2 0.05 0.12 118 261, Leon.bra Leontopodium Compositae 0.14 0.09 0.32 280 brachyactis Gand. 119 79 Leon.car Leonurus cardiaca L. Lamiaceae 0.01 0.09 0.03

120 358 Lepi.vir Lepidium virginicum L. Brassicaceae 0.11 0.14 0.24

121 271 Lept.cor Leptopus cordifolius Decne. Phyllanthaceae 0.05 0.05 0.00

122 9, 329 Lesp.jun Lespedeza juncea (L.f.) Pers Fabaceae 0.07 0.00 0.00

123 214 Leuc.vul Leucanthemum vulgare Compositae 0.07 0.23 0.32 (Vaill.) Lam. Syn: Chrysanthemum vulgare

(Lam.) Gaterau

124 237 Ligu.jac Ligularia jacquemontiana (De Compositae 0.12 0.09 0.00 cne.) M.A.Rau 125 112 Loli.mul Lolium multiflorum Lam. Poaceae 0.09 0.05 0.18

126 166, Loni.qua Lonicera quinquelocularis Caprifoliaceae 0.24 0.36 0.26 194 Hard. 127 251, Loni.web Lonicera webbiana Wall. ex Caprifoliaceae 0.04 0.00 0.00 278 DC. 128 37 Lotu.cor Lotus corniculatus L. Fabaceae 0.00 0.00 0.03

129 68 Lysi.che Lysimachia Primulaceae 0.14 0.00 0.06 chenopodioides Watt. ex Hook. f. 130 134 Malv.neg Malva neglecta Wallr. Malvaceae 0.07 0.05 0.03

131 221, Malv.ver Malva verticillata L. Malvaceae 0.00 0.05 0.03 343 132 241 Marr.vul Marrubium vulgare L. Lamiaceae 0.05 0.00 0.03

133 335 Matr.rec Matricaria chamomilla L. Compositae 0.01 0.00 0.03

134 222, Medi.lac Medicago laciniata (L.) Mill. Leguminosae 0.04 0.05 0.06 259, 360 135 188, Ment.lon Mentha longifolia (L.) L. Lamiaceae 0.04 0.09 0.03 242 136 84, Morc.esc Morchella esculanta (L.) Helvaveliacea 0.01 0.05 0.00 339 Pers.ex.Fr. 137 197 Mori.per Morina persica L. Caprifoliaceae 0.04 0.05 0.12

169

138 100 Moru.nig Morus nigra L. Moraceae 0.05 0.05 0.03 139 157, Nepe.con Nepeta connata Royle ex Lamiaceae 0.01 0.05 0.03 207 Benth. 140 179 Nepe.ere Nepeta erecta (Royle ex Lamiaceae 0.07 0.05 0.12 Benth.) Benth. 141 208 Oeno.ros Oenothera rosea L'Hér. ex Onagraceae 0.05 0.05 0.00 Aiton 142 167 Olea.fer Olea ferruginea Wall. ex Oleaceae 0.01 0.00 0.00 Aitch. 143 204 Onos.his Onosma hispida var. Boraginaceae 0.01 0.05 0.03 kashmirica (I.M. Johnst.) I.M. Johnst. 144 47, Opli.und Oplismenus undulatifolius (A Poaceae 0.03 0.00 0.00 359 rd.) Roem. & Schult.

145 147, Orig.vul Origanum vulgare L. Lamiaceae 0.12 0.00 0.21 203 146 124 Orth.sec Orthilia secunda (L.) House Ericaceae 0.01 0.00 0.03

147 185 Oxal.cor Oxalis corniculata L. Oxalidaceae 0.11 0.23 0.00

148 186, Paeo.emo Paeonia emodi Royle Paeoniaceae 0.03 0.09 0.00 358 149 75 Parn.lax Parnassia laxmannii Pall. ex Celastraceae 0.03 0.09 0.03

Schult.

150 174, Parr.jac Parrotiopsis jacquemontiana Hamamelidaceae 0.05 0.05 0.03 239 (Decne.) Rehder 151 122 Pers.amp Persicaria amplexicaulis (D.D Polygonaceae 0.45 0.27 0.56 on) Ronse Decr. Syn: Bistorta amplexicaulis (D.Don) Greene 152 57 Phle.him Phleum himalaicum Mez. Poaceae 0.03 0.00 0.03 153 114 Phle.pra Phleum pratense L. Poaceae 0.03 0.09 0.03

154 110, Phlo.bra Phlomoides bracteosa (Royle Lamiaceae 0.00 0.09 0.06 156 exBenth.) Kamelin & Makhm. 155 67, Phra.aus Phragmites australis (Cav.) Poaceae 0.04 0.05 0.06 149 Trin. ex Steud. 156 231, Phyt.aci Phytolacca acinosa Roxb. Phytolacaceae 0.05 0.05 0.03 243 157 116 Pice.smi Picea smithiana (Wall.) Pinaceae 0.03 0.05 0.06 Boiss. 158 31, 53 Pinu.rox Pinus roxburghii Sarg. Pinaceae 0.07 0.05 0.00

159 1, 54, Pinu.wal Pinus wallichiana A.B. Jacks. Pinaceae 0.88 0.86 0.88 323 160 42, 43 Pipt.aeq Piptatherum aequiglume Poaceae 0.16 0.05 0.06 (Duthie ex Hook. f.) Roshev. 170

161 8, 258 Plan.lan Plantago lanceolata L. Plantaginaceae 0.12 0.05 0.24

162 74 Plan.maj Plantago major L. Plantaginaceae 0.08 0.05 0.06

163 311 Pleu.sty Pleurospermum stylosum C.B. Apiaceae 0.03 0.05 0.06 Clarke 164 64,17 Poa.pra Poa pratensis L. Poaceae 0.09 0.18 0.09 5,212, 347 165 12, Poa.pra.1 Poa angustifolia L. Syn: Poaceae 0.08 0.14 0.18 345 Poa pratensis subsp. angustifolia (L.) Lej. 166 173, Sino.hex Sinopodophyllum hexandrum Podophylaceae 0.16 0.64 0.12 211 (Royle) T.S.Ying Syn:Podophyllum hexandrum Royle 167 132 Poly.mul Polygonatum multiflorum (L.) Asparagaceae 0.18 0.09 0.09 All. 168 85, poly.ver Polygonatum Asparagaceae 0.09 0.14 0.12 178 verticillatum (L.) All. 169 144, popu.cil Populus ciliata Wall. ex Salicaceae 0.05 0.14 0.00 199 Royle 170 155, Pote.nep Potentilla nepalensis Hook. Rosaceae 0.01 0.00 0.24 200 171 94 Prim.den Primula denticulata Sm. Primulaceae 0.04 0.05 0.12

172 145 Prun.vul Prunella vulgaris L. Lamiaceae 0.08 0.00 0.15 173 116 Prun.pad Prunus padus L. Rosaceae 0.27 0.36 0.29

174 18, Pter.cau Pteridium caudatum (L.) Dennstaedtiaceae 0.05 0.00 0.15 106,1 Maxon 30 Syn: Pteris caudata L 175 164, Pter.cre Pteris cretica L. Pteridaceae 0.01 0.09 0.03 209 176 183,3 Pter.aca Pteris acanthoneura Alston Pteridaceae 0.36 0.05 0.03 44 177 165 Puni.gra Punica granatum L. Lythraceae 0.03 0.00 0.06

178 104 Pyru.pas Pyrus pashia Buch.-Ham. ex Rosaceae 0.1 0.05 0.00 D.Don 179 193, Quer.dil Quercus dilatata Lindl. ex Fagaceae 0.24 0.18 0.09 207 A.DC. 180 169, Quer.bal Quercus baloot Griff. Fagaceae 0.04 0.05 0.00 305 181 73, 96 Quer.inc Quercus incana Bartram Fagaceae 0.11 0.09 0.15

182 150, Ranu.mun Ranunculus munroanus J.R. Rananculaceae 0.04 0.05 0.06 187 Drumm. ex Dunn 183 146, Ranu.mur Ranunculus muricatus L. Rananculaceae 0.18 0.09 0.29 195 184 243 Rhem.pur Rhamnus purpurea Edgew. Rhamnaceae 0.19 0.68 0.03

171

185 234, Rhem.aus Rheum australe D.Don Polygonaceae 0.01 0.00 0.00 321 186 263, Rhus. Suc Rhus succedanea L. Anacardiaceae 0.05 0.00 0.03 300 187 133 Robi.pse Robinia pseudoacacia L. Leguminosae 0.00 0.00 0.03

188 230, Rosa.mos Rosa moschata Herrm. Rosaceae 0.08 0.05 0.12 273 Syn: R. brunonii Lindl 189 244 Rosa.chi Rosa chinensis Jacq. Rosaceae 0.03 0.00 0.15

190 215, Rosa.mac Rosa macrophylla Lindl. Rosaceae 0.05 0.18 0.06 288 191 83 Rosa.mul Rosa multiflora Thunb. Rosaceae 0.08 0.00 0.03

192 216, Rosa.can Rosa canina L. Rosaceae 0.05 0.09 0.09 229 193 331, Rosa.web Rosa webbiana Wall. ex Rosaceae 0.07 0.18 0.12 283 Royle 194 22, 41 Rubi.him Rubia himalayensis Klotzsch Rubiaceae 0.07 0.09 0.03 195 240, Rubu.vul Rubus vulgaris Weihe & Rosaceae 0.12 0.68 0.00 306 Nees 196 247 Rubu.ped Rubus pedunculosus D.Don Rosaceae 0.04 0.09 0.00 197 76 Rume.ace Rumex acetosa L. Polygonaceae 0.08 0.14 0.09

198 317 Rume.has Rumex hastatus D. Don. Polygonaceae 0.18 0.09 0.32

199 218 Rume.nep Rumex nepalensis Spreng. Polygonaceae 0.05 0.00 0.15 200 266 Sagi.tri Sagittaria trifolia L. Alismataceae 0.03 0.00 0.06 201 158, Sali.den Salix denticulata Andersson Salicaceae 0.08 0.05 0.18 159 202 168 Sali.alb Salix alba L. Salicaceae 0.03 0.00 0.09 203 249 Salv.nub Salvia nubicola Wall. ex Lamiaceae 0.01 0.00 0.00

Sweet

204 253 Samb.wig Sambucus adnata Wall. Ex Sambucaceae 0.03 0.00 0.00 DC. 205 262 Sarc.sal Sarcococca pruniformis Lindl. Buxaceae 0.03 0.00 0.00

206 228 Sass.het Saussurea costus (Falc.) Compositae 0.01 0.00 0.03 Lipsch. 207 276 Saur.ven Sauromatum venosum (Drya Araceae 0.04 0.05 0.03 nd. ex Aiton) Kunth 208 238, Scro.cal Scrophularia calycina Benth. Compositae 0.14 0.05 0.09 257 209 86 Sedu.ewe Sedum ewersii Ledeb. Crassulaceae 0.04 0.05 0.03 210 28 Seli.wal Selinum wallichianum (DC.) Apiaceae 0.03 0.36 0.06 Raizada & H.O. Saxena 211 272 Sene.ana Senecio analogus DC. Compositae 0.03 0.36 0.12

172

212 282 Sene.nud Senecio nudicaulis Compositae 0.05 0.05 0.06 Buch.Ham. ex D. Don 213 233, Serr.pal Serratula pallida DC. Compositae 0.00 0.05 0.12 321 214 267, Seta.pum Setaria pumila (Poir.) Roem. Poaceae 0.08 0.05 0.06 285 & Schult. 215 93 Sisy.iri Sisymbrium irio L. Crucifererae 0.01 0.05 0.12 216 260, Skim.lau Skimmia laureola Franch. Solanaceae 0.32 0.23 0.03 286 217 334 Sola.sur Solanum surattense Brum.f. Solanaceae 0.04 0.00 0.00

218 375 Sola.vil Solanum villosum Mill Solanaceae 0.03 0.00 0.03 219 180, Soli.vir Solidago virgaurea L. Compositae 0.00 0.14 0.03 321 220 316 Sorb.tom Sorbaria tomentosa (Lindl.) Rosaceae 0.2 0.45 0.03 Rehder 221 313 Sorb.cus Sorbus cuspidata (Spach) Rosaceae 0.03 0.09 0.00 Hedl. 222 268, Spir.can Spiraea canescens D.Don Rosaceae 0.18 0.05 0.03 316 223 277, Stap.emo Staphylea emodi L. Staphyleaceae 0.18 0.05 0.24 326 224 309 Stip.car Stipa caragana Trin. Poaceae 0.08 0.23 0.06

225 223, Stip.jac Stipa jacquemontii Jaub. & Poaceae 0.01 0.05 0.00 307 Spach. 226 272, Stro.urt Strobilanthes urticifolia Wall. Acanthaceae 0.2 0.32 0.06 315 ex Kuntze 227 29 Swer.ala Swertia alata C.B. Clarke Gentianaceae 0.12 0.00 0.03

228 39 Swer.pan Swertia paniculata Wall. Gentianaceae 0.03 0.00 0.03

229 113 Tara.cam Taraxacum campylodes G.E.H Compositae 0.07 0.00 0.03 aglund 230 4, Taxu.bac Taxus baccata L. Taxaceae 0.31 0.73 0.5 170, 336, 359 231 2, 338 Thal.cul Thalictrum cultratum Wall. Rananculaceae 0.01 0.00 0.15 232 328 Tori.jap Torilis japonica (Houtt.) DC. Apiaceae 0.04 0.05 0.00

233 360 Trif.rep Trifolium repens L. Fabaceae 0.11 0.00 0.26

234 327 Tril.gov Trillium govanianum Wall. Melanthiaceae 0.03 0.00 0.06 ex D.Don 235 336 Tuss.far Tussilago farfara L. Compositae 0.01 0.14 0.06

236 318 Ulmu.wal Ulmus wallichiana Planch. Ulmaceae 0.08 0.00 0.12 237 300 Urti.dio Urtica dioica L. Urticaceae 0.3 0.45 0.29

173

238 346 Vale.jat Valeriana jatamansi Jones Caprifoliaceae 0.38 0.45 0.29

239 32 Verb.tha Verbascum thapsus L. Scrophulariaceae 0.14 0.18 0.03

240 103 Verb.off Verbena officinalis L. Verbenaceae 0.001 0.00 0.00

241 289 vero.lax Veronica laxa Benth. Scrophulariaceae 0.24 0.45 0.29 242 219, Vibu.gra Viburnum grandiflorum Caprifoliaceae 0.72 0.68 0.44 308 Wall. ex DC. 243 34, Vibu.mul Viburnum mullaha Caprifoliaceae 0.08 0.05 0.03 287 Buch.Ham. ex D. Don 244 123 Viol.can Viola canescens Wall. Violaceae 0.81 0.55 0.44

245 280 Part.sem Parthenocissus semicordata Vitaceae 0.04 0.05 0.00 (Wall.)Planch. 246 288 Wood.fru Woodfordia fruticosa (L.) Lythraceae 0.04 00.00 0.03 Kurz 247 312 Wul.amh Wulfeniopsis amherstiana Scrophulariaceae 0.08 0.14 0.12 (Benth.) D.Y. Hong 248 350 Zant.arm Zanthoxylum armatum DC. Rutaceae 0.01 0.00 0.03 249 220 Zeux.str Zeuxine strateumatica (L.) Orchidaceae 0.01 0.00 0.03 Schltr. 250 281 Zizi.oxy Ziziphus oxyphylla Edgew. Rhamnaceae 0.08 0.00 0.00

174