POPULATION STRUCTURE AND DYNAMICS OF DODONAEA

VISCOSA (Linn.) Jacq., IN

PAKISTAN

BY

SIRAJ AHMAD

DEPARTMENT OF BOTANY UNIVERSITY OF MALAKAND CHAKDARA, DIR (L) 2016

POPULATION STRUCTURE AND DYNAMICS OF DODONAEA

VISCOSA (Linn.) Jacq., IN MALAKAND DIVISION

PAKISTAN

BY

SIRAJ AHMAD

A thesis submitted to the Department of Botany University of Malakand for the partial fulfillment of the requirement for the degree of Doctor of Philosophy (Ph.D) in Botany

DEPARTMENT OF BOTANY UNIVERSITY OF MALAKAND CHAKDARA, DIR (L) 2016

DECLARATION

I declare that this work is original and I have not used other than the declared sources and have explicitly marked all materials which have been quoted either literally or by content from the used sources. I also declare that this work has so far neither been submitted to the

Department of Botany, University of Malakand, Pakistan, for obtaining the degree of Ph.D in

Botany.

The present study was carried out in the Laboratory of the Department of Botany, University of Malakand Chakdara (Lower Dir), Khyber Pakhtun Khwa (KPK), Pakistan.

______

Siraj Ahmad

Table of Contents LIST OF TABLES ...... iv LIST OF FIGURES ...... v ABSTRACT ...... vi CHAPTER 1:GENERAL INTRODUCTION ...... 1 1.1 Profile of the study area ...... 2 1.2 Agro-Ecosystem zoning of Malakand Division...... 3 1.3 Ecological Zonation of study area ...... 3 1.3.1 Sub-humid tropical zones ...... 3 1.3.2 Sub-tropical zone ...... 4 1.3.3 Humid temperate zone ...... 4 1.3.4 Cool temperate zone ...... 4 1.3.5 Cold temperate zone ...... 4 1.3.6 Alpine zone ...... 5 1.3.7 Cold desert zone ...... 5 1.4 History of study area ...... 6 1.5 Introduction to plant (Dodonaea viscosa)...... 10 Aims and Objectives ...... 19 CHAPTER 2: COMMUNITY STRUCTURE AND NATURAL DYNAMICS OF Dodonaea viscosa ...... 20 2.1 INTRODUCTION ...... 22 2.1.1 Community structure and natural dynamics of vegetation ...... 22 2.1.2 Multivariate analysis ...... 24 2.1.3 Background information on Multivariate analysis in Pakistan ...... 25 2.1.4 Study area ...... 26 2.2 MATERIALS AND METHOD ...... 30 2.2.1 Field Method ...... 30 2.2.2 Design of sampling points and collection of data ...... 30 2.2.3 Soil samples collection ...... 31 2.2.4 Soil Analysis ...... 32 2.2.5 Statistical Analysis ...... 32 2.3 RESULTS ...... 34 2.3.1 Floristic composition ...... 34 2.3.2 Classification of Vegetation Communities’ through Ward’s Cluster Analysis ...... 37 2.3.3 Dodonaea communities ...... 40

i

2.3.4 Environmental variables related to various plants communities derived through Ward’s cluster analysis ...... 42 2.3.5Correlation among environmental variables ...... 45 2.3.6 Vegetation-environmental relationship (Ordination) ...... 47 2.3.7 Correlation of Environmental variables with NMS ordination axis ...... 50 2.3.8 DCA ordination...... 51 2.3.9 Correlation of environmental variables along with DCA ordination axis ...... 55 2.3.10 Cover /ha ...... 61 2.3.11 Height size structure of Dodonaea ...... 64 2.3.12 Cover size classes of Dodonaea ...... 69 2.4 DISCUSSION ...... 72 CHAPTER 3: REGENERATION STATUS OF DODONAEAVISCOSA IN STUDY AREA ...... 79 Abstract ...... 80 3.1 INTRODUCTION ...... 80 3.2 MATERIALS AND METHODS ...... 83 3.2.1 Field survive ...... 83 3.2.2 Design of quadrat and collection of data...... 83 3.2.3 Collection of soil samples ...... 83 3.2.4 Laboratory procedure ...... 83 3.2.5 Statistical Analysis ...... 84 3.3 RESULTS ...... 85 3.3.1 Regeneration Potential of Dodonaea viscosa...... 85 3.3.2 Cross correlation of seedling and sapling with environmental variables ...... 85 3.3.3 Regression analysis ...... 88 3.4 DISCUSSION ...... 104 CHAPTER 4: THE EFFECT OF DIFFERENT SOIL AND SHADE REGIME ON GERMINATION AND GROWTH PATTERN OF DODONAEA VISCOSA ...... 106 Abstract ...... 107 4.1 Introduction ...... 108 4.2 Materials and Methods ...... 110 4.2.1 Study site ...... 110 4.2.2 Experimental manipulation ...... 111 4.3 RESULTS ...... 114 4.3.1 Soil and stand characteristics ...... 114

ii

4.3.2 Seed germination ...... 114 4.3.3 Seedling survival ...... 115 4.3.4 Seedling growth ...... 115 4.4 Discussions ...... 117 Conclusions ...... 120 Recommendations ...... 122 REFERENCES ...... 123

iii

LIST OF TABLES

Table 2. 1: The families and their representative plants species of the study area...... 34 Table 2. 2: IVI mean values of species in different group’s results from Ward’s Cluster Analysis...... 41 Table 2.3: Mean Values of Soil and environmental variables related to different plants Communities...... 44 Table 2.4: Cross correlation among environmental variables ...... 46 Table 2.5: Correlation of NMS ordination Axis with environmental variables...... 50 Table 2. 6: Correlation of environmental variables along with DCA ordination axis ...... 55 Table 2.7: Density/ha Mean±SE values of species in different communities ...... 59 Table 2. 8: Groups Density/ha Range of Species in different communities derived through Ward’s Cluster Analysis ...... 60 Table 2. 9: Mean Values of Cover/ha of Species in different Communities ...... 62 Table 2. 10: Cover/ha range of species in different communities Derived through Ward’s Cluster Analysis ...... 63 Table 2. 11: Height size classes of Dodonaea viscosa in different vegetation groups ...... 65 Table 2. 12: Cover size classes of Dodonaea viscosa in different Communities’ groups ...... 70

Table 3. 1: Density/ha mean values of Seedling, sapling in comparison to mature Dodonaea viscosa ...... 85 Table 3. 2: Inter-correlation among Seedling, Sapling and environmental variables ...... 87 Table 3. 3: Regression relationship of seedling, sapling /environmental factors ...... 89

Table 4.1: Percent seed germination and reproductive capacity (seed output) of Dodonaea viscose in four different soil types...... 115 Table 4.2: Average (±SE) values of different parameters of Dodonaea viscosa in four different soil types...... 116 Table 4.3: Average (±SE) fresh and dry biomass of Dodonaea viscose in four different soil types...... 116

iv

LIST OF FIGURES

Figure 2.1: The families of plants present (%) in area of study i.e., Malakand Division Pakistan ...... 36 Figure 2.2: Cluster Dendrogram derived through Ward’s Cluster Analysis showing different plants communities...... 38 Figure 2.3: Two way clusters Dendrogram representing the presence and absence of species in different sampling stands...... 39 Figure 2.4: NMS ordination shows the distribution of 28 stands in Different Groups on the two Axis of NMS plot...... 48 Figure 2.5: NMS ordination of vegetation data with total variance explained along axis 1 and axis 2...... 49 Figure 2. 6: DCA...... 54 Figure 2. 7: Height size of Dodonaea in Five Groups ...... 69 Figure 2. 8: Dodonaea viscosa Size Classes ...... 72

Figure 3. 1: Seedling /Sapling ...... 90 Figure 3. 2: Sapling /elevation ...... 90 Figure 3. 3: Seedling/Elevation...... 91 Figure 3. 4: Sapling /Slope angle ...... 91 Figure 3. 5: Seedling Slope angle ...... 92 Figure 3. 6: Sapling /Aspect...... 92 Figure 3. 7: Seedling /Aspect ...... 93 Figure 3. 8: Seedling /water ho capacity ...... 93 Figure 3. 9: Sapling /Water holding capacity ...... 94 Figure 3. 10: Seedling /pH ...... 94 Figure 3. 11: Sapling /pH ...... 95 Figure 3. 12: Seedling OM% ...... 95 Figure 3. 13: Sapling /%OM ...... 96 Figure 3. 14: Seedling /Lime% ...... 96 Figure 3. 15: Sapling Lime% ...... 97 Figure 3. 16: Seedling /Nitrogen ...... 97 Figure 3. 17: Sapling Nitrogen...... 98 Figure 3. 18: Seedling /Phosporus ...... 98 Figure 3. 19: Sapling /Phosporus ...... 99 Figure 3. 20: Seedling /Potasium ...... 99 Figure 3. 21: Sapling /Potassium ...... 100 Figure 3. 22: Sapling /Sand % ...... 100 Figure 3. 23: Seedling Sand% ...... 101 Figure 3. 24: Seedling Clay % ...... 101 Figure 3. 25: Sapling/Clay % ...... 102 Figure 3. 26: Seedling Silt% ...... 102 Figure 3. 27: Sapling /Silt % ...... 103

v

Abstract The regeneration status of Dodonaea viscosa communities was investigated in Malakand division. It lies in Hindukush 71.43°South to 73.85°North and 36.07°West to 36.40°East. Seedling and sapling data was collected using quadrat method. Various physical and chemical factors were measured. Density/ha of seedling, sapling and mature plants of Dodonaea viscosa were calculated. Pearson’s correlation and regression analysis were performed. In protected area the high density of seedling, sapling and mature plants observed respectively which show the normal regeneration, however overgrazing and anthropogenic activities delimits natural regeneration of Dodonaea. Highest density of seedling and sapling at altitude (1083) m observed in group III and then decrease gradually. Pearson’s correlation co-efficient showed a positive relationship between sapling and seedling densities (p > 0.001), sapling and organic matter (p > 0.05) seedling and organic matter, while a negative significant relationship was found between sapling and soil pH (p > 0.05), seedling and soil pH (p > 0.001), as well as sapling and elevation (P > 0.01). Regression analysis showed positive relationship between Seedling and Sapling (r =0.919 at P > 0.001), sapling /organic matte (r=0.457 at P > 0.05). The Seedling/Elevation showed a negative significant relationship (r = 0.525 at P > 0.01). The R-values of regression analysis of seedling/soil pH and sapling /pH was r = 0.529, r = 0.386 respectively, which shows a negative significant relationship of pH with seedling and sapling at the p-values P > 0.01, P > 0.05. Both soil and shade are the most essential factors for the establishment and growth of plants. The study investigated two stages of Dodonaea viscosa reproduction, seed germination and seedling growth on various screening effects of four combinations i.e. B1S1_seeds sowed in pure Garden soil; B2S2_seeds sowed in field soil substrate containing organic manure (Humus); B3S3_seeds sowed in original soil collected from the field (OS_ Original soil); B4S4_seeds sowed in undergrowth soil substrate and placed under shade of trees. The seeds were sowed according to their polarity in the bags, which were arranged according to a randomized complete block design (RCBD), with four treatments per block and four repeats. Each experimental unit included 30 plant pots, each of which contained three seeds. The results showed significantly different impacts on Dodonaea viscosa seed germination in the soil types with various textural, physiochemical compositions and the influence of shade. Various soils type and shade regimes from different sites of the study area have significant differences (p < 0.05) for germination percentage, leaf morphology, stem diameter, height, cover, root length and above ground biomass respectively Germination percentage was adversely influenced by shade. In the comparisons among original soil with same environmental factor, it was always showed that germination rate of seeds was higher in the order of Garden soil, Humus mixed soil, and soil placed in shade. It was concluded from the results that Dodonaea viscosa have variable response to different soil substrate and shade and well adopted to original soil with open sunshine. The present study aims to investigate the structure and natural dynamics of Dodonaea viscosa in Malakand division. For sampling a 10x10m quadrat was used and 28 stands were sampled. Various phytosociological attributes of the species were calculated. The IVI of 27 plants species was subjected to Cluster Analysis and Ordination which results into the formation of 5 plants communities dominated by Dodonaea viscosa. NMS ordination axes 2 was significantly correlated with slope, Potassium and clay % while axes 1 was only found in significant correlation with clay particles. DCA axes 1 was significantly correlated with slope, pH, and organic matter while, axes 2 was correlated significantly with water holding capacity, Nitrogen contents and organic matter. Difference among the density and cover/ha of Dodonaea viscosa and associated species in different communities is due to the difference in the values of environmental factors, highest density/ha of Dodonaea viscosa was recorded at

vi

high altitude, high slope angle, organic matter, water holding capacity, clay %, and Nitrogen content in the soil. Dodonaea viscosa mostly grows in calcareous soils with a texture loamy Sand to sandy loam with low phosphorus, marginal nitrogen and potassium contents. Dodonaea viscosa can grows at any aspect but it favors areas which receive full sunlight. The height and cover sized classed showed a multi7model pattern (reverse J7 shaped, L7shaped, Bell7 shaped almost bell shaped and Un7even). The difference in pattern of size classes is due to anthropogenic activities, competition, regeneration pattern, difference in the soil and micro climatic condition of the site.

vii

CHAPTER 1: GENERAL INTRODUCTION

1

Chapter 1: General Introduction

1.1. Profile of the study area

Malakand Division is the northern most administration unit of , Pakistan. It is located between 30° and 36° N and 71° &73° East, longitude and latitude respectively (AR.Bag. 1979). To its west & north is Afghanistan, towards its east is Gilgit Baltistan, river Indus form a natural boundary between Malakand & Hazara Division for a good part. It covers an area of 29,872 km, which is 40.1 % of the total area of North West Frontier Province (NWFP). The division comprises of five districts and one Agency, namely Swat, Buner, Dir, Shangla, Chitral and Malakand Agency. These are the areas with deep valleys and lofty peaks in the Himalayan and Hindukush Mountains. The region is comprised of a wide range of agro-ecological zones from the semi-arid sub-humid sub- tropical southern plains of Malakand Agency (500 meters) to the arid and clod temperate valley of Chitral district (2,500 meters). The total cultivated area is just 11.8 % of the total land and in most of the areas two cropping patterns are dominant.

In the district of Swat and Dir, November, December, January and February are the coldest months in Swat and March is also cold in Dir, while June, July, and August are the hottest months with maximum temperature not exceeding 40 C°, night are comparatively cool. High seasonal temperature allows the cultivation of most deciduous fruits and temperate types of vegetables.

A feature of monthly rain fall distribution is the double peak in early spring and late summer. Obviously the region is subject to monsoon rain when soil temperature is at highest. Apples are therefore, susceptible to crown rooting under such conditions. Also extreme fluctuations in soil moisture level during the growing season can favour the diseases. The rain fall pattern leads to alternate period of wet and dry. South of Mingora is basically too hot for apples at low altitudes, but suitable for plum, apricot, peach and cherry cultivation with varieties that mature before the monsoon. This area is not suitable for vegetable seed production.

2

1.2. Agro-Ecosystem zoning of Malakand Division. Given the difference temperature and rainfall regions, Malakand Division has been divided into four distinct agro-ecological zones. Malakand agency, Swat, Dir and Chitral districts for the purpose of possible fruits and vegetables. Higher altitude pastures and forests above the snow line have been life aside. The main agro-ecological zones are: Zone I: Hot and humid in summer, cold in winter, with forest occurring at night at December and January. This zone include lower Swat, lower Dir, lower Chitral district and Malakand agency. ZoneII: Warm and humid in summer with heavy frost and occasional light snow in certain areas during December and January. This zone is composed of middle Swat, Dir and Chitral.

Zone III: High altitude areas with temperate climate where summers are comparatively mild and heavy snow falls during winter only one crop is possible during the summer. This zone includes upper Swat, upper Dir and upper Chitral.

Zone IV: Lower part of the Malakand division. This is a unique pocket where the winter is frost free and the summer is hot and humid.

1.3. Ecological Zonation of study area Depending on various factors; climatic variability, floral diversity and agricultural Land use pattern, eight agro-ecological zones were identified by Ahmad and Ahmad (2004).

1.3.1 Sub-humid tropical zones

This is the lowest basin of the valley and is characterized by shorter winters and longer summers. Two crops are harvested here mainly rice in rotation with Wheat. Citrus is the most commonly grown fruit of the zone (Ahmad and Ahmad,2004). This zone contains the graveyard of pando baba (jola gram). Here the Indicator species are; Phoenix sylvestris, bauhinia vargata, and reptonia buxifolia.

3

1.3.2 Sub-tropical zone Most of the Swat comes under this category of classification. The Altitudinal variation is from 600푚to1000푚. This is the lower extreme and the summer is hot and humid (Ahmad and Ahmad2004). Bananas, guava and oranges are the tropical indicator fruits of this region, whilst rice and wheat are the commonly grown cereals. Acacia modesta and Olea ferruginea are the indicator species.

1.3.3 Humid temperate zone This zone starts from the 1000푚 altitude and reaches1500푚. The summer is hot and humid especially in the months of June and July. The apple is the commonest fruit, but maize rice, onion and lentils are also commonly grown. Pinus roxburghii and Quercus incana are the native indicator trees (Ahmad and Ahmad).

1.3.4 Cool temperate zone The altitude for this zone is from 1500푚 to 2000푚 and the typical examples are the sub- valleys e.g. miandam, malam jaba, cahil, mankial and lalkoh (Ahmad and Ahmad, 2004). The weather is very cold in winter and a considerable Snow fall is common occurrence although the summer is very short, a double Cropping system still prevails. The common crops are: potato, maize and wheat. Keeping livestock is common profession here, while the people are dependent on forest products to a great extent. The indicator species are Pinus wallichiana and Quercus dilatata (Ahmad and Ahmad, 2004)

1.3.5 Cold temperate zone This is very densely forested zone in the Swat valley, with an altitude range of 2000-2500m. The summer here is short and the snowfall is heavier in winter, remaining for up to four months. Potato and beans are grown mixed with maize. Livestock is the major part of the economy. This zone includes areas like Sulatanr, Mankial, Jaba and Ladoo. Abies pindrow and Picea smitiana are the indicator species (Ahmad and Ahmad 2004). Subalpine zone these are the areas covered with snow for more than five months and demarcate the tree line from2500 − 3500푚. These are the stomachs of the Livestock of Swat districts and are grazed by herds of sheep and goats in the summer months. Normally, agricultural is not a common practice inthese areas and the products are usually the medicinal herbs growing

4

here. Betula utilis and Quercus semecarpifolia are the indicator tree species of the zone (Ahmad and Ahmad, 2004).

1.3.6 Alpine zone The highest points of the Swat valley consist of glacial lakes and streams. The Altitude is 3500 – 4500푚 with extremely harsh weather, intense UV radiation and strong winds, all characteristics of the zone. Chukial, Basaro SAR in Chail and Soor karr in Mankial share the same zonal conditions. Medicinal plants are collected from the region in summer, from the limited resources available (Ahmad and Ahmad, 2004).

1.3.7 Cold desert zone These are the extreme high peaks of the swat valley and are covered by snow and glaciers all year long. The mountains range from 4500to 6000푚; the Falakser and mingo pass are the representative areas (Ahmad and Ahmad2004). Some threatened endemic fauna of the region is present, including snow Leopards and snow cocks, but no obvious macro flora is present. It is very obvious that the presence of these different ecological zones provide Micro-climates and ecological niches to a wide variety of flora and fauna, contributing to the complexity of biodiversity of the Swat valley. The swat valley, known as the Switzerland of the east, is situated in the Khyber Pakhtunkhwa (KPK) (former north west frontier) province of Pakistan and can be traced on the globe at 34° 34° to 35° 55°N and 72°08° to 72° 50° E (Hinwari et al., 2003b). Swat is bounded by Chitral and Ghizer on the north. Indus Kohistan and Shangle to the east, Buner and FATA Malakand Agency on The south and district Dir on the west (Anon, 1998). Swat remained as an independent state but was absorbed into Pakistan in1969. The total area is 5337 square kilometers from the village of landaakay in the south to the valley of Gabaral in the north (Shinwari et al., 2003b). In the center of the valley is the main river, the River Swat (“d’ swat sind” in the Pashto language). Swat is the most holiday hotspot for the whole of Pakistan and people come here all year along for the different attractions in summer to escape the burning heat of the center and south of Pakistan, and in winter to enjoy Pakistan’s only ski resort. Swat used to get foreign tourist in the last decade but because of the carelessness of the government towards the improvement of ecotourism industry and other disastrous events in the area, it is now considered un-safe for the outsiders.

5

1.4 History of study area

Malakand Division has a unique and rich history. Once the seat of Ancient Buddhist Civilization the landscape is now dominated by Pashtoon majority. The area includes one of the tallest mountain peaks to luxuriant forests; tallest of all is that of Hindukush Range. Historic ruins, founded at different places in the agency, indicate that this area was part of Gandhara civilization and Buddhist peoples lived here. The last Buddhist ruler, Raja Gira, seems to have ruled over there about nine hundred years ago. Sultan Mahmood of Ghazni, a Muslim ruler, came there from Afghanistan through Bajaur and defeated the Buddhist ruler, Raja gira.

Later, another Afghan ruler, Muhammad Ghauri, invaded the area and Islam began to spread there. The Yousafzai Pathan tribe came to inhabit this area is the wake of the invasion. About 400 years ago, successive Mughal rulers attempted in vain to capture this area. After the fall of the Mughals, Sikh rulers tried to conquer this area but we repulsed. The British had always looked at this area with covetous eyes but dared no venture to flirt with it openly. In1882, The British approached to the elders of Malakand Agency with the request to allow the passage of post to Chitral, which was then in the Administrative sphere of Gilgit. In1885, the Chitral Expedition necessitated the British intervention in this area. British officer and troops had been besieged in Chitral by Chitralis. To support their forces there, they needed a route to Chitral as the Gilgit-Chitral road, the only route at that time, was covered with snow and they had left with no option except to pass through Malakand Agency. The British therefore, laid siege of the Malakand pass. The people fought bravely and offered stubborn resistance to the enemy. The British artillery particularly proving more than a match for the old and rusty guns and swords of the natives. To fortify their position and ensure the safety of the strategically important Chitral road, they constructed two forts at Malakand and Chakdara with many piquet overhead the surrounding hills. One of them Churchill piquet, was named after Lt. Churchill who later on became the Prime Minister of Britain. Since then the British intervened in the politics of the area. A political Agent was stationed at Malakand to mediate between the British and the people of the Area.

Topography: Hindu Raj Range splits the total area into two tracts i.e. Chitral and rest of districts. District Dir has range branch of southern Hindukush. (Akram, J.1990), District Shangla has several mountain peaks of which the notable one include. can mainly be vaguely classified into two topographical regions, i.e. the valley basin, the plain region and the

6

mountainous region. Although there is No actual flat area in the valley, the low lands of the valley is considered Uniform low terrain. The plain area is very small and the average approximate Width of the valley is 6 km, while the total length of the valley from landakay to Gabral is 145푘푚 (Shinwari et al., 2003b). The plain area can be further subdivided into lower (kuz) and upper (bar). The widest open portion of the valley is between the barikot and . There are some subsidiaries Valleys are called “darey” which are narrow passages between mountains and they are basically the branch outs/off-shoots of the main valley. Swat is the area where the world’s great mountain ranges. Namely the Karakourum, Hindukush and Himalaya meet and give this area an immense strategic and geographic importance (Shinwari et al., 2003b). Some of these mountains have their peaks Covered in snow all year long. These mountains provide a natural barrier to the Monsoon and help with precipitation in the valley. There are a number of high Mountains such as: Falkser 5917푚, chokial 6174푚 and mankia l5589푚. People from time immemorial have travelled through these rugged mountains which have witnessed primitive civilizations.

Kischar Ghar (4464푚), Yakh Ghar (4179푚) & Badarsan (4000푚). (Younus ‘M 2001) Swat District shows a huge contrast in elevation i.e. from 600푚 to 6069푚. The highest peak in Swat District is that of Falakser which rises up to6096푚.

Soil: Malakand Region soils are of both transported and residual type. Thick vegetation cover in most of the parts is major contributing factor in connection with all soil forming processes. Mostly the foot hills and low lands have deep soil profile with large pore space due to large particle size. In forest floors the soils are quite rich in organic matter which is due to rapid humification of soils. Soil cover ranges from light brown to deep brown as one travel from lower to higher parts of the Malakand Division. Some portion of District Buner such as Juwar (salarzai) have dry soils as the area receive less periodic moisture in companion to rest of Malakand Division.

Climatology: The weather of the study area is not very harsh, but there is a considerable variation in the mean temperature of the lower and the upper parts of the valley Majority of the inhabitants are agriculturists. Region has some very fertile agriculture lands, as well as forests. Forest has luxuriant growth of gymnosperms and angiosperms. On higher altitudes in presence of periodic moistures bryophytes and pteridophytes show prominent presence. Major crops

7

include Maize, Wheat & Rice. Peaches of swat are matchless in terms of their taste value and yield. Apple and strawberries are also among the leading fruits produced by Malakand Region. Palai area produces finest of the citrus fruits. The Hottest month is normally June and the maximum temperature reaches 33 C° While the minimum 16 C° (Shinwari et al., 2003b). January is the coldest month and the mean maximum and minimum temperatures are 11C° and 2C°, Respectively Phyto geographically, most of the area comes under the sino-japanese region (Ali and Qaiser, 1986), where monsoon rain mostly occurs in the summer. The climate supports bi-crops culture in the lower areas while only one crop is grown in the upper, mountainous areas (Ahmad and Ahmad, 2004). Languages: Majority of the people in Malakand division speak Pashto language in the yousafzai dialect. In Chitral “khewar” is spoken while few setllers in upper Chitral also speak “Wakhi”. In upper parts of Malakand division Gujro & Kohestani are also spoken.

Geology:

Malakand region exhibits a complex geological make up. In valleys the glacial are visible. General out look of the rocks suggest their origin in quaternary to carboniferous periods (A.R.Baig, 1974). Rocks in Chitral area are slates, schists, shales, limestones etc. those of Kalam, swat region are quartz rich in mica. These are mainly of sedimentary type. In Kohistan selt the rocks are medium grime and contain dcorites. Vegetation:

Major portion of the area is a natural forest which is subjected to tremendous biotic pressure mainly due to population blast and urbanization. Southern Chitral parts are forests while towards north of Chitral there forest vanish completely. One the reason behind disappearing forests in agriculture extension. One restricted to foot hills, now agriculture extends as high as 2400푚 in cedrous deodara zone, with wheat as major crop. This huge biotic pressure is coupled with grazing pressure. In general vegetation of Malakand Region can be summed up as follow. 1. In Buner area, up to 250 − 600푚, Tropical dry decidum forsts are found forming a narrow strip here mean annual rain fall in 600 − 1100푚푚. Mainly during summer. Such forest is mainly composed of decidum plants such as Lannea wodier, Pistacia integerrima, Acacia catechu etc.

8

2. In lower swat adjoining Buner area and southern parts of Dir, dry subtropical broad leaved forests are seen up to450 − 900푚. Here mean annual rain fall ranges between 250 − 750푚푚. Most common plants of their vegetation type are Olia cuspidate, Acacia modesta, Gymnosporia spinosa, etc. In Timergara area a transition is seen and there is thick growth of Dodonaea-Olea-Acaccia commonly in shrubs. Dodonaea viscosa dominants all. 3. Between 900 − 1600푚 sub-tropical pine forests are found. Which sometimes ascend up to 2100푚. In south of Swat & Dir there forest recive up to 750 − 1250푚푚 of annual mean rainfall Major plant form in this type include Pinus roxbergi, Quercus iricana, Rhododendron arboveccm, Zizyphus oxiyphyla, Rubus fruiticosus etc Dodonaea viscosa is a major shrub in their vegetation as well. 4. In lower swat between 1500 − 3300푚 where precipitation ranges between 625 − 750푚푚, moist temperate forests are seen. Forests mainly consist of conifers. Such as Pinus wallichiana, Abies pindrow, Picea smithiana. Some parts are covered by Oak forests as well. With most common species being Querum incana. 5. Dry temperate forests form a zone between 1500 − 3300푚. These forests are seen in Chitral, Dir &Swat. Annual mean rainfall on these altitude ranges between 250 − 725푚푚. Pistaccia mutica, Quercus ilex, Cedrous deodara, Pinus gerardiana, Juniperous polyeapos are represented plant types of Dry temperate forests. 6. Sub alpine forests: This characteristic vegetation type is observed between 3300 − 3900푚. Forest of this type includes Betula utilis and Abies spectabilis with patches of Pinus wallichiana, shruby under growth of Vibernum and salix is also seen. 7. Alpine shrub is found between 3600 − 3900푚. There region receive heavy snow fall. Juniperous community, Juniperous squamala are common to this zone. 8. Shangla is located in the Khyber Pakhtunkhwa province of Pakistan. The district headquarter is located at Alpurai. It was previously sub division of swat district, but was upgraded to status of a district on July 10, 1995 by chief minister Aftab Khan Sherpao. The total area of the district is 1,586 square kilometer. Shangla comprises to sub division Alpurai and Puran tehsil. There are four sub tehsil , , and Makhuzai.Shangla has the lowest human development index in the province and second lowest in the country. There are 28 union counsil Alpura tehsil sub division of nineteen union council and Puran tehsil consists of 9 union council.

9

Shangla is located in Himalayan mountainous ranges and is a unique for its diverse flora.About 50% of the area is covered by forests and are declared as protected forest by the government. The forest of the shangle anot add only the beauty of the area but also play important role in the district economy. The local people use forest wood for their daily life and also in some area of the district. The people use forests for commercial purposes due to which forest facing high threat of vanishing form some areas. 9. Normally the chir pine Pinus roxberghii are found above the 1220 − 1829푚 elevation on will drained ridges with southerly aspects, but in certain areas the lower limit descends to 1067푚 and the upper limit on cooler aspect is reduced to 1677푚 elevation. These forests are present at the lower part of Chakesar, Besham, Martong, Makhozai and Puran areas. The crops consist of pure chir but kail also appears in the upper parts.The growth of chir pine in Chakesar and in cooler aspects is vigorous while Martong and Makhozai areas it is stunted and malformed.(Younas,M.2001). 10. The dry sub-tropical broad leaved forests or scrub forests are found below the chir zone at the height of 1067푚 − 1220푚 depending upon the aspect. They are found in Puran, Martong, Chakesar and Besham areas of the lower indusKoshistan. The common species are kau (olea cuspidate), palosa (Acacia modesta), kangar (Pistacia itegerrima), Biakar (Adhatoda vissica), Saantha (Dodonaea viscosa), and it is the most free dominant species of the grave yard and other scattered species. Regeneration of olive is almost absent. The area near the villages is less dense due to browsing and damage by inhabitants.

1.5.Introduction to plant (Dodonaea viscosa) Family sapindaceae consist of 150 genera and about 20000 species (flora of Pakistan; Abdulrahman 2013). Among the 150 known genera of sapindaceae only Dodonaea genus is consist of sixty species which are widely distributed in the warmer area of the world, among these one of the most important species is Dodonaea viscosa (L) which is evergreen woody perennial shrub ( Khan et al.,2013).

Systematic position of Dodonaea viscosa

Kingdom : plantae Division : Magnaliophyta Class : Magnoliopsida

10

Order : sapindales Family : Sapindaceae Genus : Dodonaea Species : D.viscosa

Description of plant parts

Stem

Dodonaea viscosa is a multi-stem shrub or a small tree reach to a height of about seven meter. The plant may be dioecious or monoecious. The stem bark of Dodonaea viscosa is blackis in colour while, the bark colour of the branches is radish brown or blackish (flora of Pakistan; Rajamanickam et al., 2010).

Leaves

The leaves are simple, alternate and exstipulate; petiole is short (2.5푚푚) or some time absent. Leaf blade is oblaceolate ellitcal. Margin of the leaves are entire, glabrous at both surface and cover with a sticky glandular exudate in young stages (flora of Pakistan).

Inflorescence

Inflorescence of Dodonaea viscosa is panicle, terminal in position and bear at the tip of braches. The flower of Dodonaea viscosa are bisexual some time unisexual and greenish yellow or white in color, 8 − 15푚푚 long pedicel,2 − 2.5푚푚 long sepals 3 − 4 in number, petals are totallyabsent,(7 − 9) stamens with short filament and oblong anther while, absent in female flowers. Ovary is superior and oblong, flattend (2 − 3 celled). Style has two to three lobes (Abdulrahman 2013; flora of Pakistan).

Fruit and seed

The color of fruit of Dodonaea viscosa is brown or purplish or white in outlook and is a papery capsule (15 – 23 푚푚 × 18 – 25 푚푚) bearing two to three wings dehiscent by splitting anlong 2 – 3 central septa and each cell two seeded. The color of seed is black and 3푚푚 in diameter, sub-globules and more or less compressed (Abdul Rahman 2013; flora of Pakistan)

Distribution

11

It is believed that Dodonaea viscosa is a native species of Australia, now it is spread throughout the world (Rajamanickam et al., 2010). This species is distributed in the tropical region of Arizona, Florida, Mexico, Africa, New Zealand, Puerto Rico, India (West et al., 1948) as well as in Pakistan and so many other countries (Abdulrahman 2013).

Medicinal uses

Traditionally the whole plant of Dodonaea viscosa is used for the treatment of skin diseases of head and face, the leaves are also used for treatment of fever, sore throat, malaria, angina, arthritis and so many other diseases (Abdulrahman 2013). The leaves of Dodonaea viscosa are also used for the gastrointestinal disorder (diarrhea), swelling as well as used as antispasmodic agent (Rojas et al 1996). Crib et al., (1981) reported that the leaves and roots of Dodonaea viscosa are used for the treatment of headache and toothache. The flower also used as a tonic (Wagnerr et al., 1987) the wood of Dodonaea viscosa is also used as a fuel in many countries (Jain et al., 1999). Beside this Dodonaea viscosa is also used as a fuel for so many other biological activities such as wound healing (Habbu et al., 2007), antioxidant (Teffo et al., 2010), analgesic (Anilreddy, 2009), antidiabetic (Veerapur et al., 2010), Hepato protective (Ahmad et al., 2012), antifungal, and anti-bacterial activities (Pirzada et al., 2010; Nasrullah et al., 2012; khurram et al., 2009., Mothana et al., 2010)

12

Isolated Male plants due to anthropogenic activities

Agricultural Extension in research area

13

Isolated Female plant in eroded soil

Agricultural Extension, construction activates

14

Over grazing in research area

Invasio plants eucalyptus plantation in Dodonaea community

15

Dodonaea near residential area

Control from grazing

16

Monitored site

Unprotected site

17

Author in the field

18

Aims and Objectives

The main objectives of the present investigation were to evaluate the influence of environmental variables on vegetation cover with special emphasis on Dodonaea viscosa forests. To realize this goal, a number of specific objectives were formulated as below:

1. To evaluate the structure of Dodonaea viscosa forests in different areas of Malakand division.

2. To identify the main indicators of environmental variables in relation to vegetation cover through multivariate techniques.

3. To investigate the natural regeneration of the species in their original habitats and on trial basis to predict the present and future status of the species.

19

CHAPTER 2: COMMUNITY STRUCTURE AND NATURAL DYNAMICS OF Dodonaea viscosa

20

CHAPTER 2: COMMUNITY STRUCTURE AND NATURAL DYNAMICS OF Dodonaea viscosa

Abstract The aim of the present study was to explore the structure and natural dynamics of Dodonaea viscosa in Malakand division. For data collection a 10 × 10푚 quadrat was used and 28 stands were sampled. Frequency, relative frequency, density, relative density, density/ha and cover/ha of all the species were calculated. The data of Environmental variables was also collected. The IVI data of 27 plants species was subjected to Cluster analysis and ordination (NMS and DCA) which results into the formation of 5 distinct plants communities dominated by Dodonaea viscosa but the co-dominant species in community I, II was indegopera in III Berbers in IV Aillanthus and in that of V was Gymnosporia. NMS ordination axes 2 was significantly correlated with slope, potassium and clay % while axes 1was only found in significant correlation with clay particles. DCA axes 1 was significantly correlated with slope, pH, and organic matter while, axes 2 was correlated significantly with soil water holding capacity, Nitrogen contents and organic matter. Difference among the density and cover/ha of Dodonaea viscosa and other associated species in different communities is due to the difference in the values of environmental variables which were found in significant correlation with DCA, and NMS ordination axes. highest density/ha of Dodonaea viscosa was recorded in community situated at high altitude, high slope angle, organic matter, water holding capacity clay%, and Nitrogen content in the soil were high. Dodonaea viscosa mostly grows in calcareous soils with a texture Loamy Sand to sandy loam with low phosphorus and marginal nitrogen and potassium contents. Dodonaea viscosa can grows at any aspect but it favors areas which receive full sunlight. The height and cover sized classed showed a multi- model pattern such as reverse J- shaped, L-shaped, Bell-shaped almost bell shaped and Un- even pattern. The difference in pattern in the size classes is due to anthropogenic activities, inter-specific and intra-specific competition, pattern of regeneration, difference in the soil and micro climatic condition of the site.

21

2.1. INTRODUCTION

2.1.1. Community structure and natural dynamics of vegetation In order to know about the structure and natural dynamics of vegetation, it is necessary to classify the communities on the basis of structural and functional parameters because, it has practical importance for the maintenance of ecosystem (Grime, 2001). However, lack of information on the dynamics and diversity of the ecosystems they support hinder formulation and implementation of sound management plan (Betru, 2005). Ecologists have tried to understand patterns and processes in a system by classifying them into manageable uniform units (Shariatullah 2013). Many ecologists agree that plants interact in a community with diffused boundaries (Whittaker 1967, Odum, 1971). Sustainability of forest resources is of utmost importance in the life of local communities as well as prevailing environmental conditions. Quantitative species data helps in understanding the status, degree of usefulness and intensity of the anthropogenic pressure on each species (Cronin & Pandya, 2009). Thus phytosociological information about each individual tree species is essential for understanding their ecology and establishing conservation management policies for these under pressure forests (Kharakwal, 2009). Plants have played a critical role in maintaing human health and civilizing, the quality of human life for thousands of years (Dhankhar et al., 2011). Round about 29% of the land in the world is covered by forest and about 321, 2012 plants species has been reported up till now (Wahab 2011). The total area of Pakistan is about 7,96,096 square kilometer and has only little percent covered by forests (world (Arthur, 2009; FAO, 2009; IUCN 2010; Wahab 2011). This area comprised of about 6,000 higher plants species (Shinwari, 2010). In Pakistan about 4.8% of the total area is covered by forests (Ahmed et al., 2010; Khan, 2012) in which one-third is categorized as productive forests vegetation while, the remaining is categorize as protective forests (Sethi, 2001). These forests provide fodder and shelter for livestock, fuels-wood folklore medicines for local community (Nabi et al 2015). Due to the influence of a number of interacting factors such as climatic change topography of the area edaphic factor and anthropogenic disturbance change the composition of natural vegetation from area to area with the passage of time (Wahab 2011). As it is a widely accepted phenomenon that forests production and conservation greatly improve our environment and water resources, however their unplanned utilization greatly damages the land and water resources and strangle its aesthetic values, leading to total deforestation of the ecological landscape (Jan 2011).Due to anthropogenic disturbance The

22

forests of Malakand division are under severe threat (Wahab 2011; Rahman 2013; Nabi et al 2015) likewise other areas of the country, little conservatory measures from forest department have been made in Malakand division (Wahab 2011). As a result of increasing population, unavailability of alternative source for fuel purposes, over grazing and fodder demand for livestock not only disturb the communities’ structure but also reduced the diversity of plants species of the area (Meijard et al., 2005; Asner et al., 2006; Timilsina et al., 2007; Gairola et al., 2008). Due to overexploitation the annual deforestation rate of Pakistan is raised to 3% and more than 25% forests has been lost in the last three decades as reported by Shaheen & Shinwari, (2012) and Ahmad et al (2010).

Beside the anthropogenic disturbance such as overexploitation, grazing of animals , and demand for fuel purposes the communities’ structure and diversity pattern is mostly influenced by ecological factors such as altitudinal gradient, slope aspect annual rainfall (Whittaker and Niering 1975, Shariat 2013; Khan 2012; Rahman 2013) of the area (Meijard et al., 2005; Asner et al., 2006). The detailed understanding of the relationship between vegetation communities and environmental factors of an ecosystem is a vital branch of environmental science (Tavili & Jafari 2009). Among environmental factors greatly affect the distribution and richness of species (Schuster and Diekmann 2005) one of the most important factors is altitude of the area (Merganic et al. 2004; Khan et al 2013). Rahbek (1995) reported that mid-altitude support more species as compared to low and high altitude. As the elevation of the area increase the species diversity and rich also increase (Brown 1988, Stevens 1992; Lomolino 2001). Beside the elevation plant diversity also depend upon the latitude and size of the area selected for sampling of vegetation communities (Rosenzweig 1995; Currie and Paquin 1987; Shariatullah 2013). The diversity and richness also depend upon the aspect of the site because aspect regulate temperature and moisture availability (Panthi et al 2007; Tavili & Jafari 2009) and the north facing aspect get maximum amount of moisture as compared to south facing aspect (Vetaas 2002). The physical and chemical properties of the soil such as texture, nutrient availability (organic and inorganic) lime contents and pH etc. also greatly affect the distribution of vegetation (Khan et al 2010; Zare et al 2011; Khan and Bibi 2013; Shariatullah 2013; Rahman 2013).

23

2.1.2. Multivariate analysis

Multivariate analyses are the statistical analysis techniques widely use for the analysis of vegetation communities (Khaznadar et al. 2009; Khan and Hussain 2013). Through these techniques one can handle large species and environmental data sets by reducing the number of dimensions throughout the use of redundancy of information (McGarigal et al., 2000) as well as used for a huge data set (Gauche, 1982a). Multivariate analysis gives an apparent picture of the relation of environmental factors with the vegetation data and reduced the complexity in analysis (McCune & Mefford 1999; Anderson et al 2006). There are mainly two groups of multivariate analysis namely classification (Cluster Analysis) in which sample and species score is used for the grouping of sampling plots based on similarities and differences in the data set (Sokal, 1973) and ordination which are used for the formation of scatter diagram or arrangement of sampling plots along ordination axes (Palmer, 2005). There are different ordinations techniques generally used for the investigation of ecological data sets such as Two-Way Indicator Species Analysis (TWINSPAN), Principal Component Analysis (PCA), Bray-Curtis Ordination, Detrended Correspondence Analysis (DCA) and Non-metric Multidimensional Scaling (NMS) (Whittaker1967; Orloci 1978; Gauch 1982). Among these ordination techniques Detrended correspondence analysis (DCA) is an eigenanalysis ordination technique based on reciprocal averaging (Hill 1973) and widely used for the analysis of vegetation communities (Okland and Eilertsen 1996; Exner et al. 2002; Panthi et al 2007). DCA ordination not only tells us about the pattern in complex data set (Ter-Braak 1986) but, also geared to species and sample (McCune & Grace 2002). DCA ordination did not require environmental data but only species data is used to presume the gradients (Sagers and Lyon, 1997). In DCA ordination analysis the distribution of species and sample are not inhibit by ecological variables because it is an indirect gradient analysis and focus on the analysis of pattern of species distribution (Sagers and Lyon, 1997). DCA ordination produce the sample and species ordination at the same time that’s why it is more advanced than all other ordination techniques (McCune & Grace 2002). DCA ordination has the ability to removes unnecessary outliers and separates them before analysis (McCune & Grace 2002) as well as gives us the most explainable ordination result of the data set (Hill & Gauch 1980). Beside DCA ordination another technique which is mostly used in the field of ecology is Non-numeric Multidimensional Scaling (NMS). It is usually suitable for the analysis of ecological data set which is no normal, arbitrary. NMS ordination is

24

fundamentally different in design and interpretation from all other ordination techniques (Kruskal and Wish 1978; Clarke 1993; McCune and Grace (2002). This method is not only used as an ordination technique, but as a method for the assessment of dimensionality of the data set (McCune & Grace 2002). NMS ordination based on the rank distance and tries to linearize the relation between environmental and ecological distance, letting off zero truncation problem (Beals 1984). It gives outcome for that data in which there is lesser amount of variation (Gauch 1982). This method is also apply for the identification of similarity in ecological data set as well as tries to minimize space portrayal of associations (Gauch 1982).

2.1.3. Background information on Multivariate analysis in Pakistan The application of multivariate techniques such as classification and ordination for the analysis of ecological data were started in Europe and North American countries large ago. While in These techniques are recently introduced in Pakistan and for the first time used by Shaukat and Qadir (1971). After that, Ahmed and Qadir (1976) Used multivariate techniques for the analysis of the vegetation of Chilton Baluchistan and northern areas. Shaukat et al (1980) applied multivariate techniques for the analysis of vegetation of Gadap (Sind). Peer et al (2001) used CCA ordination and Two-way Indicator Species Analysis for the vegetation analysis of Hindukush Mountains was classified into four groups. Hussain et al (1993) used multivariate techniques for the analysis of dry deciduous forests near Swabi and Mardan. The vegetation of industrial area of Punjab was documented by Maria et al (2004) using TWINSPAN (two-way indicator species analysis). Enright et al., (2005) conducted ecological survey on vegetation of Kirthar National Park and made their ordination and classification with environmental and edaphic gradients.Riffat and Hussain (2007) conducted quantitative survey on the vegetation of the foot hills (Islamabad) while studying the effect of Broussonetia papyrifera on native plants species .he used CA, DCA, PCA and CCA ordination for classification. Jabeen and Ahmed (2009) studied floristic composition of vegetation in Ayub Nation Park using TWINSPAN for classification and DCA and CCA for ordination and recognized two plant communities. Wazir et al (2008) reported the vegetation of Chapursan using cluster analysis and DCA ordination. Ahmed et al (2009) documented the communities’ structure of Oleaferruginea forests in District Dir (Lower) by the application of Ward’s cluster analysis and ordination. Ahmed (2009) used TWINSPAN and DCA ordination for the communities’ analysis of herbaceous plants species in Margalla Hills National Park and reported four distinct communities. Siddiqui et al (2009) conducted phytosociological study on Pinus roxburghii

25

and associated tree species from lesser Himalayan and Hindukush Range of Pakistan. Ahmed et al (2011) conduct a detailed survive on vegetation of Cedrus deodara forests in relation to environmental variables in Hindu Kush and Himalayan ranges of Pakistan. He used DCA ordination and Ward’s cluster analysis for classification. Shaheen et al (2011) explored the structure diversity and dynamics of the vegetation of Bagh (Kashmir). The structure diversity and endemic richness of Karambar Lake vegetation of Chitral district was explored by Shaheen and Shinwari (2012) using DCA ordination and TWINSPAN analysis. Khan et al (2011) reported community structure of Monotheca buxifolia and associated vegetation of District Dir. Saeed and Ahmad (2011) documented the relation of road side vegetation of Lahore-Islamabad motorway and edaphic factors. He used CCA for the ordination of vegetation and soils variable of the study area. Wahab (2011) used multivariate techniques of the exploration of structure and dynamics of Pine tree species in Dir. Rahman (2013) used Ward’s cluster analysis, NMS and DCA ordination and conducted phytosociological study on Seriphidium brevifolium in Dir lower and reported six plants communities. Khan et al (2013) used DCA ordination for the analysis of vegetation environmental relationship in district Chitral. Ali (2013) studied Phytosociology, Structure and Dynamics of Pinus roxburghii associations from Northern Pakistan by the application of advance multivariate techniques (NMS ordination) and reported three plants communities. Shariatullah (2013) reported the relation of environmental variables and Justicia adhatoda communities in Malakand division. He applied DCA, NMS and PCA ordination for ordering the data and Ward’s cluster analysis for classification. Khan et al (2014) conducted phytosociological survive on the structure and dynamics of Pinus roxburghii from northern area of Pakistan applied ward’s cluster analysis and NMS ordination and reported three plants communities.

As in the light of above mentions information the data is conducted mostly on overall vegetation of the particular area or either on single tree species but less attention is given to shrubs and no information is available on the detailed structure and natural dynamics of Dodonaea viscosa dominated communities in Malakand division. Therefore an attempt is made to conduct detailed study on the structure and natural dynamics of Dodonaea viscosa dominated areas in relation to environmental and edaphic variables in Malakand division.

2.1.4. Study area

The forest of Malakand division falls in sub-tropical zone characterized by Sub-tropical Scrub Forest at the lower elevation and Sub-tropical broad leaves ever green Forest at the

26

higher altitudes, having 4631 hectare forest area. These forests not only provide fodder and shelter for livestock, but also local community with medicinal plants, fuel-wood and non- timber forest produce like olives and Chir gum. People of the area also fulfill their daily fuel wood and timber requirements from these forests. Owing to diversified habitat, this Division has great potential for wildlife conservation (Anon 2012).

Site description

The area of Malakand division is situated in the northern side of Khyber Pakhtunkhwa and southern aspect of Gilgit Baltistan. It is located in the Hindukush range of Pakistan and positioned at latitude 3530 N and 72 00 E longitude at altitude between 700 − 7000푚 . It included Chitral, Dir Lower Dir Upper, District swat, Buner, Shangla (Khan, 2011). Its protected area comprised of 952 푘푚2 (Shariatullah 2013) the forest of Malakand division lies in the sub-tropical zone and is comprised of Sub-tropical Scrub Forest at the lower altitude while Sub-tropical broad leaves ever green Forest are positioned on the higher altitudes(Nabi et al 2015). Out of the total area of Malakand division about 4631 hectare area is covered by forests and is comprised of highly diverse vegetation (Nabi et al 2015; Wahab 2011, Khan 2011; Shariatullah 2013) which fulfill the need of the local inhabitant for timber, fuels woods (Rahman 2013; Shariatullah 2013). At the high altitudes of different districts of Malakand Division such as Dir upper, Swat, Shangla and Buner the forest are comprised of mostly comprised of conifer species such as Cedrus deodara, Picea smithiana, Abies pindrow, Pinus gerardiana, Pinus wallichiana and Juniperus excelsa (Saddozai 1998; Wahab, 2011) At the lower elevation Pinus roxburghii (Khan, 2012) beside this Several broad leaved species like Monotheca buxifolia, Punica granatum, Olea ferruginea, Acacia modesta, Ficus palmata are mostly found (Khan et al., 2010) Quercus representing three species i.e. Quercus baloot, Quercus incana and Quercus dilitata as reported by Habibullah (2010). Lower areas below the pine zone there is rich forest of Dodonaea viscosa forming communities with Indigofera gerardiana, Otostegia limbata, Plectranthus rugosus, Periploca aphylla, Micromeria biflora, Myrtusand Buddleja etc. species (Yousifzai et al., 2012). The characteristics of Buner, Shangla and Swat district are almost the similar like that of the adjoining districts in terms of natural vegetation, topography, and soil and other ecological aspects. On the other hand lies at a strategically important position as it acts as a gateway to Swat, Dir, Chitral and Bajaur (Beg and Khan, 1974). It is surrounded by a series of mountains that were overgrown with different kinds of trees in the past though

27

they have a barren look today (Shariatullah). Malakand Pass connects Mardan to Swat and Dir is located near Dargai (Barkatullah and Ibrar, 2011).

Population of the area

According to (DCR 1994) the population of Malakand Division is 596 people per sq.km while its total population was recorded as 567,000 in 2004 − 05 in which pashton population is dominant in the overall division. Out of the total area of Malakand Division about 456600 hectors area is used for the agriculture purposes and it is considered as the major source of income of the inhabitant of the area (Shariatullah 2013).

Geography

The soil of Malakand division is mostly loamy having high content of moist. The cultivated area is irrigated by river swat flowing from swat through Kohistan and joining with river Kabal near the capital city Peshawar. The main income generation sources of Malakand division are Dargi and Khas Malakand however there is other many other suitable places which need the attention of government for the construction of hydro-power points in these sites. The mountains of Malakand division are rich with mineral resources such as chromites iron, china clay and fuller earth etc. beside these minerals there is possibility for so many other minerals but due to the poverty of the inhabitants the exploration and exploitation of these minerals is not only difficult but impossible for the peoples of the area and required the attention of .the peoples of other districts (Rahman 2013). If investors from the other districts and provinces diverted their attention towards mineral wealth they can find and get vast mineral treasures.

Climate

The climate is varying with the elevation in different districts of Malakand division. However, major stands dominated by Justicia adhatoda were sampled in district Dir and adjoining areas of district Swat. Therefore, the climatic data of Dir Meteorological station is presented here. The climate of the investigated area is broadly described as continental, which is hot during summer and cold in winter (Wahab et al., 2008). The climate varies considerably between the north and south due to difference in altitude (Khan et al., 2010, 2011; Khan 2012). The area has four distend season’s i.e. summer, winter, spring and autumn. The summer season is moderate and warm; June and July are hot months. In June the

28

mean maximum and minimum temperature has been recorded as 32.52 Co and 15.67Co respectively. The winter season is cold and severe and the temperature abruptly falls from November and onwards. December, January and February are very cold months and the temperature generally falls below freezing point (Khan, 2012). The mean maximum and minimum temperature in the month of January has been recorded as 11.22Co and – 2.39Co respectively. The precipitation is received throughout the year in the form of rain, snow and hail. The maximum rainfall has been recorded in March is 242.22푚푚 while the relative humidity is quit high throughout the year. Climate data of district Dir metrological satiation was recorded by Pakistan metrological department Karachi.

29

2.2. MATERIALS AND METHOD

2.2.1: Field Method

The area of different districts of Malakand division were search in the years (2010 − 14) for Dodonaea viscosa communities. As the species Dodonaea viscosa mostly grows in the foothills of Malakand division, those areas were select for sampling of Dodonaea viscosa communities where the species was found dominantly and the sites were least disturbed. The data was also taken from graveyards because the Muslim’s graveyards are almost protected in Malakand divisions (Rahman 2013). Different quantitative methods were applied for sampling Dodonaea communities following standard procedure of Mueller-Dombois and Ellenberg (1974), Cottam and Curtis (1956). For sampling of the Dodonaea viscosa communities’ Quadrat method was applied.

2.2.2: Design of sampling points and collection of data

In order to collect the data on phytosociological aspect of the target species and other associated plants species, standard procedure of Cottam and Curtis (1956) and detailed by Ahmad and Shaukat (2012). A 10 × 10 meter size points were select for sampling vegetation data of Dodonaea viscosa communities. 10Quadrats were taken randomly at each sampling sites where Dodonaea viscosa communities were spread up to 1 hector area following Cox (1990) method (shariatullah 2013; Rahman 2013). Height and cover of all the species were measured in cm and their numbers of individuals were noted. Inside the sampling points (Quadrat) data of seedling and sapling of Dodonaea viscosa was collected following the procedure used by Hussain (1984). Elevation of the sampling stand was measured in meter obtained through GPS (global positioning system) and aspect of the sampled stands was determined through magnetic compass while, clinometer was used for the measurement of slope angle following (Khan2012; Khan2013; Shariatullah 2013; Rahman 2013). The plants specimens were collected and identified with the help of flora of Pakistan (Nasir & Ali 1972; Ali & Qaisar 1995 − 2007).

30

2.2.3: Soil samples collection One kg Soil samples were collected in polythene bags from four different places in each stand up to a depth of 30푐푚. The polythene bags were labeled and taken to the agricultural research center Takhtaband Mingora Swat for analysis.

II: Laboratory Methods Vegetation data analysis The collected data of vegetation communities was entered into micro soft office Excel 2010 for the analysis of physiological attributes like frequency, relative frequency, density, relative density, cover, relative cover, Importance values (IVI), density/ha and cover/ha were calculated for each species using the formulas given below..

Number of quadrats in which a species occured Frequency (F1)  100 Total number of quadrats

Frequency of a species in all quadrats Relative Frequency (F3)  100 Frequency of all species in all quadrats

Number of individual s of a species in all quadrats Density (D1)  Number of quadrats taken

Number of individual s of a species in all quadrats Relative Density (D3)  100 Number of individual s of all species in all quadrats

Cover of a species in a quadrat Cover (C1)  Number of individual of a species in a quadrat

Sum of cover of a species in quadrat Relative Cover (C3)  100 Sum of cover of all species in all quadrat

RD + RC + RF Importance value index (IVI)  3 Density of a species Density/ha (D2)  10,000 Area of quadrat  no of Quadrats

Cover of a species Cover/ha (C2)  10,000 Area of quadrat  no of Quadrats

31

2.2.4 Soil Analysis

The soil samples collected from different sampling sites were thoroughly mixed properly labeled and was taken to the agriculture research Centre Takhtaband (Swat) for analysis of physical and chemical parameters. Soil was analyzed for ten parameters such as soil water holding capacity, organic matter, soil pH, lime contents, soil texture (silt, sand and clay particles) and inorganic nutrients such as N.P.K. Soil water holding capacity was determined following the method of Harding and Ross (1964). 1: 5Soil water suspensions were used for the determination of pH of the soil samples following the procedure used by Black (1965). For the calculation of silt, sand and clay) % hydrometer method was used following (Bouyoucos 1936). The % age of organic matter of the soil was determined following Walkley (1947), while acid base neutralizationmethod was used for lime contents of the soil (Rahman et al 2012). The inorganic nutrients of the soil such as N.P.K were also determined. Bingham (1994) method was used for the determination of phosphorus determination while (Sultan-pur & Schwab 1977) was followed for the determination of the remaining two in organic nutrients and were extracted M.No. 3, and AB-DTPA was used for the extraction of N and K from Basic soil (Sultan-pur & Schwab 1977)

2.2.5. Statistical Analysis PC-Ord (Window Version 5.10) statistical software was used for the statistical analysis of the vegetation communities Ward’s cluster analysis was used for the exposition of underlined groups structures and major gradients to overcome in the composition of vegetation data (McCune & Medford 2005). Data of vegetation communities was classified through Wards clustering procedure using Pc-Ord software. The vegetation data (IVI) of all the species and the data of environmental variables such as elevation, slope angle and aspect of the sampled stands were subject to pc-ord software for cluster analysis and Ordination. The aspect of the sampling stands were taken in numerical codes following (Palmer (2005) Khan (2011).The species and stands which make noise and obscure the structure of communities groups were first removed from the data set (McCune et al., 2000Timilsina, 2009).The data of environmental variables was first subjected to DCA and NMS ordination. Then the vegetation data in correlation to environmental variable was subjected to Cluster analysis which results to the formation five vegetation communities at fifty % information of the species (McCune and Mefford, 1999). The mean+se of all the species was calculated using Microsoft office 2003 for all communities which were results from Cluster analysis. The data

32

of environmental variables of the sampling stands was correlated with the ordination axis of DCA and NMS.

The height and cover of the target species (Dodonaea viscosa) was classified in different classes on per hector basis (Shariatullah 2012). For the height of Dodonaea 40푐푚 class interval was taken while for cover size classes a 100푐푚 class interval was taken following Ahmed and Shaukat, (2012). For the representation of pattern of individuals in communities groups a polynomial curve was fit to the entire diagram (Shariatullah 2012, Rahman 2013). In NMS ordination the score of axes was taken at low stress after 100 time run (Gupta et al 2008; Khan et al 2011; Khan2012). Pearson product movement correlation and regression analysis were applied for the interpretation of vegetation data.

The height and cover of individuals of Dodonaea viscosa was classified into different size classes on hector basis using Microsoft Office Excel 2003 following Ahmed and Shaukat (2012). The height of individuals was classified at a regular interval of 24 푐푚 height, while the cover classes were made at 40 푐푚 class interval. The pattern of Dodonaea viscosa plants in different groups (resulted from Ward’s cluster analysis) was represented while fitting a polynomial curve to the entire diagram. The cross correlation among the environmental variable was also perform following Shariatullah (2012). For the investigation of regeneration status of Dodonaea in different communities the density/ha of seedling and sapling was calculated and compared with the density/ha of mature plants of Dodonaea viscosa. The density/ha values of seedling and sapling was used for the inter correlation and regression analysis (Rahman 2013).

33

2.3. RESULTS

2.3.1. Floristic composition

Floristically the twenty-eight sampling sites were dominated by 27 plants species. These species belong to twenty one families and twenty six genera. The results of floristic composition shows that out of twenty-one families only six families shared two (for each) these dominant families in term of their species are family Asclepidaceae (Periphloca aphyla), family Asteraceae (Artimisia scoparia, Xanthium strumarium), family Lamiaceae (Plectranthus rugosus, Otostegia limbata), Oleaceae (Olea cuspidate, Daphne mucronata), Papilonaceae (Indigofera gerardiana, Acacia modesta) and family Rosaceae (Cotoneaster microphyllum, Rubus fruticosus). The remaining thirteen families are nonspecific and each family has contributed single species to the floristic composition of the sampling area. Except family Asclepidaceae every one of the remaining families is represented by a single genus. The families and their representative species are represented in the table 2.1 as well as in the figure2.1.

Table ퟐ. 1: The families and their representative plants species of the study area.

s/no Family name Species Habit N of species % species

1 Acanthaceae Justicia adhatoda 1 3.704 2 Almaceae Celtis. astralis 1 3.704 3 Asclepidaceae Periphloca phyla 2 7.407 Periphloca. aphylla 4 Asteraceae Artisimia. scoparia 2 7.407 Xanthium. stramarium 5 Berberidaceae Berberis lycium 1 3.704 6 Budlejaceae Plectranthus. 1 3.704 rugosus 7 Celustraceae Gymnosporia 1 3.704 8 Euphorbiaceae Millilotus 1 3.704 philiphinsis 9 Fagaceae Quercus dilate 1 3.704 10 LamiaceaeF Plectranthus. 2 7.407 rugosus Otostegia limbata 11 Moraceae Ficus. carica 1 3.704 12 Myrsinaceae Myrsine Africana 1 3.704 13 Myrtaceae Eucalyptus 1 3.704

34

lanceolatus 14 Oleaceae Olea cuspidate 2 7.407 D. mucronata 15 Papilonaceae I. gerardiana 2 7.407 Accacia modesta 16 Rosaceae C. microphyllum 2 7.407 Rubus fruticosus 17 Rutaceae Cotoniaster 1 3.704 microphylus 18 Sapindaceae Dodonaea viscose 1 3.704 19 Simarubaceae Aillanthus altissima 1 3.704 20 Trichocomaceae Asparigus gracilis 1 3.704 21 Verbenaceae Vitex negundo 1 3.704 Total 27 100

35

Myrtaceae Sapindaceae 4% 4%

Papilonaceae Simarubaceae Rosaceae Oleaceae 7% 4% 7% 7% Rutaceae 4% Trichocomaceae 4% Myrsinaceae 4% Moraceae Verbenaceae 4% 4% Acanthaceae 4%

Lamiaceae 7% Almaceae Fagaceae 4% 4% Budlejaceae Euphorbiaceae Asteraceae 4% 4% Berberidaceae Celustraceae 7% 4% 4% Asclepidaceae 7%

Figure 2.1: The families of plants present (%) in area of study i.e. Malakand Division Pakistan

36

2.3.2. Classification of Vegetation Communities’ through Ward’s Cluster Analysis

Figure 2.2 represent Dendrogram results from Ward’s cluster analysis showing different groups of vegetation, the five groups are separated at fifty % remaining information of the species individuals. On the basis of stands group I is the largest group include 9 stands present at the top of Dendrogram. Figure 2.3 represent the two way cluster Dendrogram of the vegetation data in Dodonaea viscosa dominated communities. On one side it represent stands while on the other side represent species distributed in different groups at 50% remaining information of the species. The white dots represent the absence while the black dots represent the presence of species distributed in the study area. The two ways cluster Dendrogram shows that Indegopera gerardiana, Olea cuspidate, P. rugosus and Gymnosporia are the species mostly repeating withDodonaea viscosa in the sampling sites while, species such as Asparigus gracilis, Plactrenthus rugosus, Quercus dilatata, Periphloca aphylla, Artemisia scoparia, Cotoniaster microphylus, Myrsine Africana, P. aphylla, Asparigus gracilis, Plactrenthus rugosus, Ailanthus altissima and Xanthium stramarium are rarely found in the communities of Dodonaea viscosa. Based on importance values the details of different communities groups are summarized below.

37

Figure 2.2: Cluster Dendrogram derived through Ward’s Cluster Analysis showing different plants communities.

38

Figure 2.3: Two way clusters Dendrogram representing the presence and absence of species in different sampling stands.

39

2.3.3. Dodonaea communities

Group I: Dodonaea-indegopera Community

It is the largest group consists of 9 stands and 20 species in which Dodonaea viscosa is the dominant species followed by indegopera gerardiana and Gymnosporia royliana. The IVI mean value of Dodonaea viscosa is 29.4 ± 4.6, Indigofera gerardiana13.7 ± 1.9 and Gymnosporia8.7 ± 2.1%. Among the remaining 17 species of this group the ivi mean value of P. rugosus is 8.3 ± 1.1%, Olea cuspidate7.1 ± 1.5%, Daphne mucronata3.4 ± 1.4, Justacia adhatoda4.2 ± 1.7%,Cotoneaster microphyllum 5.8 ± 2.0%, Mallotus philiphinsis 2.8 ± 1.9% and Accacia modesta2.9 ± 1.5% as depicted in table (2.2). The ivi mean value of Berberis lycium and Otostegia limbata is same as 2.3 ± 1.2 % while, that of Ficus carica is 1.3±1.0 and Cotoniaster microphylus1.0 ± 0.7%. Artemisia scoparia, Periphloca Aphylla, Eucalyptus lanceolatus, Vitex negundo, and Myrsine africana and C. astralis are other species of group I and their IVI mean values ranged from 0.7 ± 0.7 to 0.1 ± 0.1% as shown in table (2.2).

Group II: Dodonaea-Indegopera Community

This group consists of 6 stands and 17 species. Similar to group I the leading dominant species of this community is Dodonaea viscosa with mean ivi (33.00 ± 0.91)% and co- dominant species is Indigofera gerardiana(11.50 ± 2.90)%. Among the remaining species of this group the ivi of Otostegia limbata is 10.50 ± 0.65, Berberis lycium 7.75 ± 0.85% and Cotoneaster microphyllum7.00 ± 2.42%. Among the remaining 12 species the ivi of C. astralis is 6.25 ± 2.25%, Justicia adhatoda 6.00 ± 0.41%, Eucalyptus lanceolatus 5.25 ± 1.93%, Gymnosporia 3.00 ± 2.38%, Asparigus gracilis 2.50 ± 2.18%, Plactrenthus rugosus 2.00 ± 1.68%, Olea cuspidate 1.50 ± 0.65% and Vitex negundo1.25 ± 0.95%.The ivi of the remaining four species of this group namely P. rugosus, Daphne mucronata, A. scoparia, and P. aphylla is less than 1% (Table 2.2).

Group III: Dodonaea-Berberis Community

Group 3 consist of three stands and six species. The dominant species of this community is Dodonaea viscosa while the co-dominant species is Berberis lycium. The ivi mean value of Dodonaea viscosa in this group is 45.67 ± 0.67% which is the largest value as compared to other groups. The mean ivi of Berberis lycium is 13.67 ± 2.73% followed by P. rugosus(13.33 ± 2.33)% and Accacia modesta(12.33 ± 2.33)% (Table 2.2).

40

Group IV: Dodonaea-Aillanthus Community

Group IV consist of four stands and10 species in which Dodonaea viscosa is dominant species sharing 25.75 ± 2.46% ivi while the co-dominant species is Aillanthus altissima whose ivi is 12.25 ± 1.65%. Among the other species of this group the ivi of Justicia adhatoda is 10.00 ± 1.08%, Daphne mucronata12.25 ± 1.65%, Olea cuspidate9.25 ± 0.48%Mallotus philiphinsis8.00 ± 1.22%, P. rugosus7.00 ± 0.41%. Ficus carica, Indigofera gerardiana and Xanthium stramarium are the remaining species present in group IV in association with Dodonaea viscosa. The result also shows that Aillanthus altissima and Xanthium stramarium are the species which are only found in the sampling stands of group IV (2.2).

Group V: Dodonaea-Gymnosporia Community

Group V is also dominated by Dodonaea viscosa but the ivi mean value of Dodonaea viscosa (23.67 ± 4.78)% is low as compared to other groups. Out of 10 species of group V Gymnosporia is the co-dominant species sharing 14.67 ± 2.38% ivi mean value with group V followed by P. rugosus(14.00 ± 2.35)%, Justicia adhatoda(13.83 ± 1.05)%, Otostegia limbata(13.50 ± 1.38)% and Rubus fruticosus(5.83 ± 2.63)% mean ivi values. Indigofera gerardiana, Eucalyptus lanceolatus Olea cuspidate and Accacia modesta are the remaining four species belong to six stands of group V and their ivi mean values are (4.67 ± 2.16,3.83 ± 1.72, 3.50 ± 1.63 and 2.50 ± 1.20)% respectively as shown in table (2.2).

Table 2. 2: IVI mean values of species in different group’s results from Ward’s Cluster Analysis. Specie 9 Group I Group II Group III Group IV Group V s Code 20 Mean±SE Mean±SE Mean±SE Mean±SE Mean±SE Dv Dodonaea viscosa 29.4±4.6 33.00±0.91 45.67±0.67 25.75±2.46 23.67±4.78 Bl Berberis lycium 2.3±1.2 7.75±0.85 13.67±2.73 - - Ig Indigofera 13.7±1.9 11.50±2.90 9.00±4.93 3.75±1.65 4.67±2.16 gerardiana Oc Olea cuspidate 7.1±1.5 1.50±0.65 - 9.25±0.48 3.50±1.63 Pr P. rugosus 8.3±1.1 0.75±0.75 13.33±2.33 7.00±0.41 14.00±2.35 Cm Cotoneaster 5.8±2.0 7.00±2.42 - - - microphyllum Dm Daphne 3.4±1.4 0.75±0.75 - 9.50±0.87 - mucronata Gy Gymnosporia 8.7±2.1 3.00±2.38 - - 14.67±2.38 Pp Periphloca phyla 0.1±0.1 - - - - OL Otostegia limbata 2.3±1.2 10.50±0.65 - - 13.50±1.38

41

Ja Justicia adhatoda 4.2±1.7 6.00±0.41 - 10.00±1.08 13.83±1.05 As Arthmesia. 0.1±0.1 0.75±0.75 - - - Scoparia Mm Cotoniaster 1.0±0.7 - - - - microphyla El Eucalyptus 0.2±0.2 5.25±1.93 - - 3.83±1.72 lanceolatus Rf Rubus fruticosus - - - - 5.83±2.63 Am Accacia modesta 2.9±1.5 - 12.33±2.33 - 2.50±1.20 Ca Cyltis. astralis 0.7±0.7 6.25±2.25 - - - Pa Periploca. - 0.25±0.25 - - - Aphylla Vn Vitex negundo 0.2±0.2 1.25±0.95 - - - Fc Ficus. Carica 1.3±1.0 - - 4.00±1.41 - Ag Asparigus gracilis - 2.50±2.18 - - - Ma Myrsine Africana 0.6±0.6 - - - - Plac Plactrenthus - 2.00±1.68 - - - rugosus Qd Quercus dilatata - - 6.00±1.53 - - Mp Mallitus 2.8±1.9 - - 8.00±1.22 - philiphinsis Aa Aillanthus - - - 12.25±1.65 - altissima Xs Xanthium - - - 0.50±2.40 - stramarium

2.3.4 Environmental variables related to various plants communities derived through Ward’s cluster analysis The mean values of environmental variables of the sampling sites are represented in table

(2.3). The results of environmental variables shows that the sampling stands of Dodonaea communities of group I, and III and V are situated at high altitude with a mean value of

1011.9 ± 101.4푚, 1083.7 ± 232.4, 1172.8 ± 155.0 while vegetation of community II is situated at low altitude. The slope angles of sampling sites of group III is large, (53.3 ± 6.7)ᵒ while that of group IV is small (28.5 ± 2.2)ᵒ than all other communities groups. The slope angle of community I, II and V is 49ᵒ, 50.8ᵒ and 41.5ᵒ (ASPECT)

3 Slope 49.0±4.4 50.8±4.0 53.3±6.7 28.5±2.2 41.5±5.4 4 Aspect 1.3±0.3 1.5±0.5 3.0±0.0 2.0±0.4 2.5±0.3

The water holding capacity of the soil belongs to the vegetation of group III is high 11.0 ±

4.4/10𝑔푚 followed by group IV and group I but low of groups II and V. based on pH mean

42

values the soil belongs to the Dodonaea viscosa dominated communities of groups (I, II, III and V) was acidic while, that of group IV was normal soil with a mean value 7.0 ± 0.2. The soil organic matter of all sampling sites was marginal but the mean value of organic matter in group V was smaller and that of Group III was greater than all the vegetation groups (Table

3.2). Lime % content of the sampling soils of group I is14.0±1.4 which is high than all other groups followed by group V, III and IV while low in community II the mean % values of the soil lime contents show that all Dodonaea viscosa communities belong to calcareous soil.

Total available nitrogen contents mean value was (1.3 ± 0.1) in community II which is the lowest value but high mean value of total nitrogen was found in group III (2.8 ± 0.5) followed by Group IV and I (2.6 ± 0.5, 2.2 ± 0.3) respectively which show that nitrogen content of all the communities groups is marginal except group II in which it is low. Among the other two inorganic nutrients in the soil of Dodonaea dominated communities Phosphors contents (mg kg-1) the mean value in the soil of community I is 2.4 ± 1.2푚𝑔 푘𝑔−1, in II it is

2.0 ± 0.2푚𝑔 푘𝑔−1 while in the soil of group (III, IV, and V) is (2.8 ± 0.5, 2.6 ± 0.5 and

1.7 ± 0.2)푚𝑔 푘𝑔−1 respectively which shows that the phosphors contents are low in the soil off all sampling sites but marginal in the soil of group V.K (mg kg-1) contents is 114.2 ±

23.2 in the soil in group I which is low than group IV but high than the soils K contents of other communities. The mean values of K (푚𝑔 푘𝑔−1) soil contents are marginal in all groups while adequate in amount in the soil of community group IV. Among the physical parameters

% of Silt particles is high than Sand (64.0 ± 3.7)% Clay (5.5 ± 0.9)% in community group

I, which shows that the soil is loamy in texture. Among the soil of other groups the % mean values of sand particles is high than silt and clay Particles.

43

Table 2.3: Mean Values of Soil and environmental variables related to different plants Communities.

S/N Environmental Group I Group II Group III Group IV Group V o variable 1 Mean±SE Mean±SE Mean±SE Mean±S Mean±SE E 2 Elevation 1011.9±10 461.2±124 1083.7±23 842.5±56 1172.8±15 1.4 .6 2.4 .3 5.0 3 Slope 49.0±4.4 50.8±4.0 53.3±6.7 28.5±2.2 41.5±5.4 4 Aspect 1.3±0.3 1.5±0.5 3.0±0.0 2.0±0.4 2.5±0.3 5 water hc/10gm 4.2±0.6 3.2±0.3 11.0±4.4 5.8±2.1 2.5±0.5 6 pH 6.4±0.4 4.6±0.7 6.2±0.1 7.0±0.2 6.2±0.3 7 %OM 1.4±0.2 1.0±0.0 1.5±0.3 1.4±0.3 0.9±0.1 8 %Lime 14.0±1.4 10.9±1.2 13.2±2.2 12.3±1.9 13.4±2.0 9 Tot.N (𝑔 푘𝑔−1) 2.3±0.3 2.0±0.2 2.8±0.5 2.6±0.5 1.7±0.2 10 P (푚𝑔 푘𝑔−1) 2.4±1.2 0.6±0.1 1.9±0.6 1.9±1.4 2.9±1.5 11 K (푚𝑔 푘𝑔−1) 114.2±23.2 63.7±5.3 94.0±8.7 164.5±37 110.0±37.6 .8 12 Sand % 64.0±3.7 70.0±2.4 55.7±4.9 62.8±8.4 65.1±5.6 13 Clay % 5.5±0.9 4.0±0.5 8.3±3.6 8.5±2.2 7.7±1.4 14 Silt% 25.6±3.5 13.0±1.1 36.0±2.1 28.7±6.4 27.6±5.0

44

2.3.5 Correlation among environmental variables Table (2.4) shows the cross correlation between different parameters of the sampling stands such as edaphic, physiographic, topographic and environmental factors. The results shows that there is a strong significant relationship between soil pH and elevation, (푟 = 0.425, 푝 > 0.05), phosphorus and elevation (푟 = .599, 푝 > 0.001), potassium and elevation (푟 = 0.394, 푝 > 0.05) as well as silt particles and elevation of the sampling stands at the given probability level (푟 = 0.517, 푝 > 0.01). The r value of cross correlation shows that there is a negative significant relationship between slope angle of the sampling sites and phosphorus (푟 = −0.408, 푝 > 0.05), slope angle and potassium contents of the soil (푟 = −0.544, 푝 > 0.01) while aspect of the sampling sites was found in positive significant relationship with clay particles (푟 = 0.465, 푝 > 0.01). A significant relationship is also found between pH and lime content (푟 = 0.513, 푝 > 0.01), pH, clay % (푟 = 0.418 푝 > 0.01) and pH and silt % of the soil at (푟 = 0.55, 푝 > 0.01) probability level. A significant relation was also found between lime % and clay particles and lime % with silt content of the soil (푟 = 0.640, 0.641, 푃 > 0.001), while a negative significant relation was found between lime % and sand particles of the soil with (푟 = −0.684, 푝 > 0.001). Among the remaining factors a positive significant relationship was found among organic matter and nitrogen contents of the soil (푟 = 0.854, 푃 > 0.001). Similarly organic matter were also found in significant relationship with potassium (푟 = 0.397, 푝 > 0.05) as shown in the table (3.3). The results also show that there is a significant relationship between nitrogen and potassium (푟 = 0.456, 푝 > 0.01) as well potassium and phosphorus (푟 = 0.673, 푝 > 0.001). Clay particles are found in negative significant relations with sand (푟 = −0.673, 푝 > 0.001) and in positive significant relation with silt contents of the soil (푟 = 631, 푝 > 0.001).

45

Table 2.4: Cross correlation among environmental variables

El Sl As Whc/10gm pH %OM %Lime N (g kg-1) P (mg kg-1) K (mg kg-1) Sa% Cl % Si % E 1 Sl -0.226 1 As 0.234 -0.039 1 W hc/10gm 0.077 -0.048 0.12 1 Ph 0.425* -0.031 0.27 0.126 1 %OM 0.06 -0.047 -0.177 0.221 0.057 1 %Li 0.236 0.121 0.358 -0.168 0.513** -0.015 1 N (g kg-1) 0.013 -0.183 -0.129 0.269 0.063 0.854*** -0.076 1 P (mg kg-1) 0.599*** -0.408* -0.269 0.083 0.052 0.195 -0.207 0.249 1 K (mg kg-1) 0.394* -0.544** -0.196 0.036 0.182 0.397* -0.108 0.456** 0.734*** 1 Sa % -0.175 0.145 -0.165 0.208 -0.31 -0.246 -0.684*** -0.192 -0.06 -0.144 1 Cl % 0.182 -0.256 0.465** -0.154 0.418** 0.029 0.640*** 0.043 -0.084 0.2 -0.673*** 1 Si% 0.517** -0.252 0.367 -0.003 0.553** 0.243 0.641*** 0.291 0.269 0.359 -0.811 0.631*** 1 Key: El=Elevation, Sl=Slope, As= Aspect, Whc/10g=water hc/10gm, pH.% OM=Organic matter, % Li=%Lime, N(g kg-1)=Tot.N (g kg-1), Phosphors=P (mg kg-1), K (mg kg-1), =Potassium(mg kg-1), Sa= Sand %, Cl=Clay %, Si=Silt %.

46

2.3.6. Vegetation-environmental relationship (Ordination)

Non-Metric Multidimensional Scaling (NMS ordination)

NMS ordination plot shows the distribution of Dodonaea viscosa dominated vegetation communities. Based on mean importance values of species the groups of vegetation data were derived through cluster analysis (CA) are also superimposed on the axis of NMS plot (Axis 1, 2). In NMS plot the groups of Dodonaea viscosa dominated communities shows an almost clock wise rotation between the two axis. Group I which consists of 9 and stands and 20 plants species is present in the middle of NMS plot. In this group the ivi of Dodonaea viscosa is third highest and elevation mean value of the group is second highest. Group II which is present at low altitude and comprised of six stands and 17 species is present at the right side and in the bottom towards axis 1.the ivi mean value of Dodonaea viscosa in group II is the second highest (33%) as compared to other groups. Group III comprised of three stands six species and is present in the extreme left side of NMS plot toward axis 2. The elevation mean value of this group is second highest while, the ivi of Dodonaea viscosa (45%) is high as compared to the ivi of Dodonaea viscosa in all other communities groups. Group IV comprised of four stands and 10 species and is present in the extreme upper part between the two axis of NMS plot (Fig: 2.3, 2.4) the ivi of Dodonaea viscosa in group iv is lower than the ivi of Dodonaea viscosa in all other communities except group V. the elevation mean value of group iv is greater than group II while smaller than the elevation of all other groups. The 10 species and six stands of Group V in which the ivi mean value of Dodonaea viscosa is medium (23%) and elevation is high is present in the extreme right side of axis 1.

47

Figure 2.4: NMS ordination shows the distribution of 28 stands in Different Groups on the two Axis of NMS plot.

48

Figure 2.5: NMS ordination of vegetation data with total variance explained along axis 1 and axis 2.

49

2.3.7. Correlation of Environmental variables with NMS ordination axis Table (2.5) represents the co-relations of different environmental variables with the ordination axis 1 and axis 2 of NMS plot. Among the studied 13 environmental factors only one factor (% Clay particles) showed significant relationship with the NMS ordination axis 1 (푟 = 0.494, p-value> 0.01) , while the NMS ordination axis 2 showed significant relationship with slope 푟 = 0.457, 푝 > 0.01), Potassium (푟 = 0.346, 푝 > 0.01) (and % clay particles (푟 = 0.494, 푝 > 0.01). the calculated 푟 values of the remaining factors is less than the tabulated 푟 value of regression table at the given degree of freedom. Therefore we can say that there is not found any significant relationship of the remaining factors with the NMS ordination axis 1 and axis 2.

Table 2.5: Correlation of NMS ordination Axis with environmental variables.

Correlation Axis I Axis 2 E. Factors R value Remarks R value Remarks Elevation 0.132 0.231 No sig Slope 0.217 0.457 Sig Aspect 0.193 0.230 water ℎ푐/10𝑔푚 0.286 0.327 pH 0.026 0.365 %OM 0.088 0.365 %Lime 0.234 0.239 Tot.N (𝑔 푘𝑔−1) 0.083 0.133 P (푚𝑔 푘𝑔−1) 0.088 0.200 K (푚𝑔 푘𝑔−1) 0.079 0.346 Sig Sand % 0.286 0.117 Clay % 0.494 Sig 0.343 Sig Silt% 0.203 0.264

50

2.3.8. DCA ordination Figures 2.6 represent the DCA ordination of vegetation communities along axis 1, 2, 2, 3 and 1, 3. As depicted in figure (3.5), the communities are isolated from each other and the rotation of vegetation communities groups is clock wise between axis 1 and axis 2. The vegetation community of group I is present in the center of DCA ordination plot between axis 1 and axis 2. Group II is located in the extreme left side overlapping towards axis 2. Group III is located in the upper part toward axis 2 while, group IV is situated in the extreme right side of ordination (DCA) toward axis 21. The vegetation communities of group V are located in the bottom of DCA ordination plot in the right side. The rotation of Dodonaea viscosa dominated communities along ordination axis 1.2 and 2, 3 shows an overlapping between some sampling sites in the vegetation groups derived through Ward’s cluster analysis. In the DCA ordination plot between axis 1 and axis 3 some stands of community group I are overlapped with group III. Group I is situated in the center, group II is located in the extreme left side, while groups III, V are located in the upper portion in the left side of the ordination axis 1, 3. Vegetation communities of group IV are located in the extreme right side of DCA Ordination plot. The ordination of different groups is irregular between ax1s 1,3,2,3. The DCA plot between axis 2,3 shows that the sampling stands of group V is located in the upper portion towards left side while group III is present in the extreme right side, these two groups are isolated while the remaining three groups which overlapping each other are present in the middle of DCA ordination axis 2 and axis 3.This ordination of vegetation communities is due to the correlation of environmental variables which are found in significant relationship with DCA ordination axis as depicted.

51

52

53

Figure 2.6: DCA

54

2.3.9. Correlation of environmental variables along with DCA ordination axis The correlation of DCA ordination axis with environmental variables is summarized in table (2.6). The results showed that among 13 studied parameters of the sampling stands only 3 parameters such as slope angle of the site (푟 = 0.394 푝 > 0.05, soil pH (푟 = 0.540푝 > 0.01) and organic matter (푟 = 0.382 푝 > 0.05) were found in significant relationship with DCA ordination axis 1. Similarly DCA ordination axis 2 was correlated significantly with three environmental factors such as soil water holding capacity (푟 = 0.495푝 > 0.01), organic matter (푟 = 0.472, 푝 > 0.05), and nitrogen contents of the soil(푟 = 0.415 푝 > 0.05). The results also show that DCA ordination axis 3 is not significantly correlated with any one of the 13 studied environmental factors of the sampling stands. As it is clear from the results that the distribution of vegetation communities is due to those parameters which are significantly correlated with ordination axis while, the other factors are moving along.

Table 2.6: Correlation of environmental variables along with DCA ordination axis DCA ordination Axis 1 Axis 2 Axis 3 E. Factors R value Remarks R value Remarks R value Remarks Elevation 0.138 0.18 0.22 Slope 0.394 Sig(푝 > 0.05) 0.351 0.226 Aspect 0.103 0.007 0.242 water hc/10gm 0.202 0.495 Sig (푝 > 0.01) 0.022 pH 0.54 Sig (푝 > 0.01) 0.063 0.353 %OM 0.382 Sig(푝 > 0.05) 0.472 Sig(푝 > 0.05) 0.153 %Lime 0.172 0.036 0.022 Tot.N (𝑔 푘𝑔−1) 0.321 0.415 Sig(푝 > 0.05) 0.008 P (푚𝑔 푘𝑔−1) 0.068 0.203 0.221 K (푚𝑔 푘𝑔−1) 0.414 0.118 0.051 Sand % 0.164 0.176 0.1 Clay % 0.305 0.065 0.265 Silt% 0.348 0.084 0.242

2.3.10: Density/ha Density/ha mean values and individual range of Dodonaea viscose along with associated vegetation are summarized in table (2.7 − 2.8).

In group I the density of the dominant species Dodonaea viscosa is 928.67 ± 128.96 mean individuals/ha with a range of 186 − 1292 individuals/ha in different communities.

The co-dominant species in term of density/ha is Indegofera gerardiana (404.42 ± 77.55)

55

followed by P. rugosus (204.98 ± 29.85) and Gymnosporia (186.32 ± 61.14). The individual range/ha of these species is 157 − 866, 0 − 288 and 0 − 580 respectively as given in table (2.8). Among the other species of community group I the mean density/ha of

Olea cuspidate is 141.71 ± 34.77, Cotoneaster microphyllum126.34 ± 46.23, Justicia adhatoda 80.97 ± 37.86, Daphne mucronata 80.34 ± 35.71, Otostegia limbata72.67 ±

44.08Mallotus philiphinsis 64.68 ± 44.05, Acacia modesta63.04 ± 33.81 and Berberis lycium is 46.03 ± 24.63 individuals/ha. The individual range/ha of these species is 0 − 286,

0 − 359, 0 − 307, 0 − 274 and 0 − 392 individuals/ha (Table 2.8). The mean density/ha of the remaining species such as F. carica is 34.51 ± 27.04C. astralis (26.77 ± 26.77)

Eucalyptus lanceolatus (12.80 ± 8.21) and Myrsine Africana12.38 ± 12.38 individuals. In terms of density/ha Cotoniaster microphylus, Vitex negundo, Periphloca aphylla, Eucalyptus lanceolatus, Myrsine Africana and A. scoparia are the remaining species rarely present in the community of group I in association with Dodonaea viscosa. The mean density/ha of these species is given in table (2.7) while their range of density/h is given in table (2.8).

Among the 18 species of group II the dominant species is Dodonaea viscose with mean density/ha 864.85 ± 60.61followed by Otostegia limbata (363.89 ± 50.32),

Indigofera gerardiana (231.39 ± 91.51), Cotoneaster microphyllum (192.43 ± 40.97),

Berberis lycium (168.85 ± 26.85) and Eucalyptus lanceolatus (138.96 ± 35.48). The density/ha range of the dominant species is 654 − 1033 individuals /ha in different communities types. The individual range of density/ha of the co-dominant species Otostegia limbata is 175-529/ha, Indigofera gerardiana(0 − 500), Cotoneaster microphyllum (0 −

281), Berberis lycium (97 − 281) and Eucalyptus lanceolatus0 − 245 individuals/ha (Table

3.4. 퐶. astralis, Justicia adhatoda, Gymnosporia, Vitex negundo,F. carica, Olea cuspidate,

Asparigus gracilis, P. aphylla, Plactrenthus rugosus, Daphne mucronata, P. rugosus and A. scoparia are other species in association with Dodonaea community. Their mean density/ha

56

ranged from (96.92 ± 46.65 to 11.11 ± 11.11)/ha while, their individual range of density/ha is given in table (2.8).

Based on number of species, community group III is the smallest community comprised of six species in which Dodonaea viscosa is the dominant species with 1524.39 ±

26.46 mean density/ha and 1493 − 1577 range of density/ha. Table (2.8) also show that the density/ha of Dodonaea is more in community III as compared to all other communities.

Acacia modesta is the co-dominant species sharing 256.12 ± 59.34 mean density with a range of 183 − 374 individuals/ha. Among the remaining four species of this community, the mean density/ha of P. rugosus is 233.66 ± 62.76,Indigofera gerardiana 219.02 ±

109.61, Berberis lycium 193.60 ± 5.52and Quercus dilatata is 73.22 ± 12.46/ha. The individual density/ha range of these species is 115 − 329, 0 − 337,183 − 202 and 58 − 98 respectively as depicted in table (2.8).

In group IV the leading dominant species is Dodonaea viscosa with 605.10 ± 95.64 mean density/ha with a range of 472 − 888 individuals. As it is clear from the results that the mean density/ha values of community group IV and Group V is low as compared to other communities. The density of the dominant species is same in community IV and V. In terms of mean density/ha the co-dominant species of community IV is Aillanthus altissimasharing

312.92 ± 53.17 mean density with in a range of 167 − 397individuals/ha. Among the remaining eight species community IV the mean density/ha of Xanthium stramarium is

295.06±90.04 followed by Justicia adhatoda (277.76 ± 44.43), Olea cuspidate (224.47 ±

64.09), P. rugosus (222.03 ± 37.32), Daphne mucronata (214.61 ± 39.31) and Mallotus philiphinsis (201.73 ± 8.72) (Table 3.6). The individual range/ha of these species is 129 −

461,174 − 391,86 − 387, 125 − 306, 129 − 319 and 184 − 220 individuals respectively.

57

The mean density/ha of the remaining two species is 88.95 ± 40.16, 57.39 ± 22.74 while their individual/ha range is 0 − 186 as shown in table (2.8).

Out of six stands and 10 species of community V the mean density/ha of the leading dominant species, Dodonaea viscosa is same as in community group V but, the difference was found in range of density /ha which is 200 − 888individuals in different sampling stands of this group. The co-dominant species in association with Dodonaea viscosa is

Gymnosporia with 469.97 ± 84.28 density/ha and 278 − 807 range of individuals/ha.

Among the remaining species of group V the mean density/ha of Otostegia limbata is

395.38 ± 37.48, Justicia adhatoda300.08 ± 26.59. P. rugosus225.34 ± 49.41, Rubus fruticosus186.85 ± 87.72 and Indigofera gerardiana is 101.33 ± 50.01while, their individuals /ha in different stands are324 − 575,195 − 377,89 − 373,0 − 486 and 0 −

284respectively. Acacia modesta, Olea cuspidate, Eucalyptus lanceolatus are the remaining rarely present species of community V. their density/ha mean values and range of density/ha are given in tables (2.7, 2.8).

58

Table 2.7: Density/haMean±SE values of species in different communities Species Group I Group II Group III Group IV Group V Code Mean±SE Mean±SE Mean±SE Mean±SE Mean±SE Dv 928.67±128.96 864.85±60.61 1524.39±26.46 605.10±95.64 605.57±105.95 Bl 46.03±24.63 168.85±26.85 193.60±5.52 - - Ig 404.42±77.55 231.39±91.51 219.02±109.61 88.95±40.16 101.33±50.01 Oc 141.71±34.77 47.76±27.95 - 224.47±64.09 85.15±38.53 Pr 204.98±29.85 14.37±14.37 233.66±62.76 222.03±37.32 225.34±49.41 Cm 126.34±46.23 192.43±40.97 - - - Dm 80.34±35.71 23.55±16.67 - 214.61±39.31 - Gy 186.32±61.14 76.39±42.29 - - 469.97±84.28 Pp 2.22±2.22 - - - 0 OL 72.67±44.08 363.89±50.32 - - 395.38±37.48 Ja 80.97±37.86 78.49±16.87 - 277.76±44.43 300.08±26.59 As 1.48±1.48 11.11±11.11 - - - Mm 5.56±5.56 - - - - El 12.80±8.21 138.96±35.48 - - 57.26±26.54 Rf - - - - 186.85±87.72 Am 63.04±33.81 - 256.12±59.34 - 73.08±38.53 Ca 26.77±26.77 96.92±46.65 - - - Pa - 38.86±33.74 - - - Vn 4.13±4.13 73.83±36.46 - - - Fc 34.51±27.04 5.97±5.97 - 57.39±22.74 - Ag - 46.74±36.98 - - - Ma 12.38±12.38 - - - - Plac - 25.66±23.90 - - - Qd - - 73.22±12.46 - - Mp 64.68±44.05 - - 201.73±8.72 - Aa - - - 312.92±53.17 - Xs - - - 295.06±90.04 -

59

Table 2. 8: Groups Density/Ha Range of Species in different communities derived through Ward’s Cluster Analysis Species Group I Group II Group III Group IV Group V Code Range Range Range Range Range Dv 186-1292 654-1033 1493-1577 472-888 200-888 Bl 0-192 97-281 183-202 - - Ig 157-866 0-500 0-337 0-186 0-284 Oc 0-286 0-179 - 86-387 0-196 Pr 0-288 0-86 115-329 125-306 89-373 Cm 0-359 0-281 - - - Dm 0-274 0-100 - 129-319 - Gy 0-580 0-225 - - 278-807 Pp 0-20 - - - - OL 0-392 175-529 - - 324-575 Ja 0-307 0-119 - 174-391 195-377 As 0-13 0-67 - - - Mm 0-50 0 - - - El 0-74 0-245 - - 0-142 Rf - - - - 0-486 Am 0-256 - 183-374 - 0-216 Ca 0-241 0-263 - - - Pa - 0-206 - - - Vn 0-37 0-215 - - - Fc 0-246 0-36 - 0-111 - Ag - 0-226 - - - Ma 0-111 - - - - Plac - 0-145 - - - Qd - - 58-98 - - Mp 0-354 - - 184-220 - Aa - - - 167-397 - Xs - - - 129-461 -

60

2.3.10: Cover /ha The cover/ha of Dodonaea viscosa and associated species in different plants communities was taken in cm and are given in table (2.9 − 2.10). The detail of cover/ha of these species is summarized below. In group I the cover/ha of Dodonaea viscosa is 1195.24 ± 111.01푐푚with a range of

730 − 1620cm individual cover /ha. The results shows that cover /ha of Dodonaea in group I is more than all other communities. In terms of cover/ha the co-dominant species is

Indigofera gerardiana(241.87 ± 40.74푐푚/ℎ푎, 82 − 434 range) followed by Olea cuspidate, Gymnosporia, Cotoneaster microphyllum, P. rugosus. The mean cover/ha of these species ranged 127.74 to 182.46 while, their individual cover/ha is 0 − 475푐푚/ℎ푎 (Table

2.9). The cover/ha mean values of the remaining species such as Justicia adhatoda, Mallotus philiphinsis, Myrsine, Africana, Daphne mucronata, Acacia modesta, Cotoniaster microphylus, Berberis lycium, F. carica, Vitex negundo, C. astralis, Eucalyptus lanceolatus,

Otostegia limbata, Periphloca aphylla and A. scoparia ranged from 0.11 ± 0.11푐푚/ℎ푎 to

97.50 ± 41.98푐푚/ℎ푎 while their individual cover /ha in different communities is 0 −

530푐푚/ℎ푎.

The mean cover/haof Dodonaea viscosa in group II is 1182.75 ± 59.67푐푚 which is less than group I but more than all other communities.In group III it is 1034.68 ±

43.22푐푚/ℎ푎 while in group IV and V the mean cover/ha is,(906.75 ± 153.34, 662 ±

296.45) cm which is less cower/ha values as compared to other communities (Table 3.8).

The individual cover range/ha of Dodonaea viscosa in group II-V is, (914 − 1312, 958 −

1108, 666 − 1315, 249 − 2102) cm respectively (Table 3.9). The co-dominant species in community I is Indigofera gerardiana in community II Otostegia limbata while in group III,

IV and V the co-dominant species are Indigofera gerardiana, Aillanthus altissima and P. rugosus with mean values of cover/ha (241.87 ± 40.74, 186.94 ± 24.67, 410.55 ± 205.28,

268.68 ± 78.06, 545 ± 170.56)푐푚 respectively as summarized in table (2.9). The individual

61

range/ha of the co-dominant species of community I is (82 − 434, 106 − 260, 0 −

617,136 − 492 and 77 − 1116) cm/ha. The cover/ha mean and range/ha of the remaining species is given in table (2.9 − 2.10).

Table 2. 9: Mean Values of Cover/ha of Species in different Communities Species Group I Group II Group III Group IV Group V Code Mean±SE Mean±SE Mean±SE Mean±SE Mean±SE Dv 1195.24±111.01 1182.75±59.67 1034.68±43.22 906.75±153.34 662±296.45 Bl 36.80±22.04 152.43±27.92 262.66±32.44 - - Ig 241.87±40.74 181.75±82.87 410.55±205.28 63.42±32.81 83±37.32 Oc 182.46±51.10 55.67±34.96 - 152.19±34.44 72±33.06 Pr 127.74±29.48 7.30±7.30 334.87±79.09 139.38±32.14 545±170.56 Cm 144.17±53.81 160.79±36.51 - - - Dm 55.70±24.28 16.39±10.48 - 249.56±86.38 - Gy 171.19±52.94 52.59±29.26 - - 251±57.02 Pp 1.57±1.57 - - - - OL 10.96±5.28 186.94±24.67 - - 207±66.17 Ja 97.50±41.98 84.11±18.09 - 202.36±34.02 413±110.48 As 0.11±0.11 5.67±5.67 - - - Mm 4.02±4.02 - - - - El 19.70±15.51 150.53±34.54 - - 124±59.20 Rf - - - - 103±52.34 Am 51.16±30.87 - 275.52±79.65 - 40±19.57 Ca 19.85±19.85 91.60±42.28 - - - Pa - 27.32±24.65 - - - Vn 3.85±3.85 97.57±48.24 - - - Fc 34.78±27.07 5.40±5.40 - 94.19±45.86 - Ag - 30.16±25.53 - - - Ma 9.69±9.69 - - - - Plac - 10.98±10.09 - - - Qd - - 181.72±60.91 - - Mp 91.71±64.10 - - 213.92±66.21 - Aa - - - 268.68±78.06 - Xs - - - 209.55±53.25 -

62

Table 2.10: Cover/ha range of species in different communities Derived through Ward’s Cluster Analysis Species Group I Group II Group III Group IV Group V Code Range Range Range Range Range Dv 730-1620 914-1312 958-1108 666-1315 249-2102 Bl 0-194 68-255 214-324 0 - Ig 82-434 0-482 0-617 0-14 0-185 Oc 0-475 0-221 - 72-232 0-173 Pr 0-252 0-44 205-478 91-232 77-1116 Cm 0-398 0-247 - - - Dm 0-193 0-55 - 44-431 - Gy 0-460 0-144 - - 98-451 Pp 0-14 - - - - OL 0-48 106-260 - 41-490 Ja 0-293 0-124 - 144-300 88-880 As 0-1 0-34 - - - Mm 0-36 - - - - El 0-142 0-58 - - 0-302 Rf - - - - 0-278 Am 0-252 - 165-430 - 0-116 Ca 0-179 0-227 - - - Pa - 0-150 - - - Vn 0-35 0-268 - - - Fc 0-248 0-32 - 0-204 - Ag - 0-156 - - - Ma 0-87 - - - - Plac - 0-61 - - - Qd - - 117-303 - - Mp 0-537 - - 42-336 - Aa - - - 136-492 - Xs - - - 88-346 -

63

2.3.11. Height size structure of Dodonaea

Table (2.11) represent the group’s density/ha of Dodonaea viscosa in different size classes at forty class interval. The frequency of classes shows that more number of individuals fall in lower size classes while, the frequency of individuals in upper size classes is low as well as there is found a gap in the height classes such as in group I, II and IV. The result also shows that the maximum height of Dodonaea viscosa is found in the range of 411 − 450 in group

V. In all the communities’ groups the frequency of 3rd class is high as compared to other classes and the height of individuals in this class ranged from 131 − 170푐푚. in the communities groups the variable number of individuals in different height classes of

Dodonaea viscosa represents multimodal shapes. Communities Group 1, III and V represent reverse J-shaped or L-shape structures. In group I maximum individual’s/ha were present in class no. 3 (followed by class 4 and 5 while class 9 is totally missed. In the height of maximum individual ranged from 91 − 130푐푚, 131 − 170푐푚 and 171 −

210푐푚respectively as shown in table (2.11). Similarly in group III the maximum individuals

/ha were found in the height class 3rd followed by 4th and 2ndrespectively while in community

V the height of maximum individuals ranged from 91 − 130푐푚 (class 3) followed by class 2 whose height was in the range of 51 − 90푐푚. The results shows that in all these three groups there is found a gap in the height size classes, which may be due to the anthropogenic disturbance which was observed during the field survey of different sampling sites. The remaining two groups (II, IV) of the Dodonaea viscosa dominated vegetation derived through

Ward’s cluster analysis represent almost bell-shaped structure. In these groups there is not found any gape in the height size classes though the highest number of individuals fall in the middle classes. Ought of nine classes of group II, only three percent individuals/ha of

Dodonaea viscosa attained 331 − 370푐푚 height. The maximum heights of Dodonaea

64

viscosa in community Group V was found in the range of 411 − 450푐푚 and 0.4% individuals/ha was found in this class.

Table 2. 11: Height size classes of Dodonaea viscosa in different vegetation groups

S/no. Group1 Sum of F* (D/ha)* % Individual/ha* 1 11-50 66 59 6 2 51-90 109 98 11 3 91-130 407 366 39 4 131-170 158 142 15 5 171-210 126 113 12 6 211-250 51 46 5 7 251-290 62 56 6 8 291-330 52 47 5 9 331-370 - - - 10 371-410 1 1 0 Sum 1032 928 100 S/no. Group2 Height Sum of F* (D/ha)* % individual/ha* 1 11-50 27 36.3 4.2 2 51-90 103 138.3 16.0 3 91-130 180 241.7 28.0 4 131-170 105 141.0 16.3 5 171-210 86 115.5 13.4 6 211-250 35 47.0 5.4 7 251-290 51 68.5 7.9 8 291-330 38 51.0 5.9 9 331-370 19 25.5 3.0 Sum 644 864.9 100 Group 3 S/no. Height Sum of F* (D/ha)* % individual/ha* 1 11-50 36 110.6 7.3 2 51-90 71 218.2 14.3 3 91-130 157 482.5 31.7 4 131-170 76 233.6 15.3 5 171-210 61 187.5 12.3 6 211-250 33 101.4 6.7 7 251-290 31 95.3 6.3 8 291-330 23 70.7 4.6 9 331-370 - - - 10 371-410 8 24.6 1.6 Sum 496 1524.4 100

65

S/no. G4 height Glasses Sum of F* (D/ha)* % individual/ha* 1 11-50 34 38.5 6.4 2 51-90 47 53.3 8.8 3 91-130 200 226.6 37.5 4 131-170 105 119.0 19.7 5 171-210 67 75.9 12.5 6 211-250 29 32.9 5.4 7 251-290 36 40.8 6.7 8 291-330 14 15.9 2.6 9 331-370 2 2.3 0.4 Sum 534 605.1 100 S/no. G 5 Height Classes Sum of F* (D/ha)* % individual/ha* 1 11-50 37 39.3 6.5 2 51-90 92 97.7 16.1 3 91-130 182 193.4 31.9 4 131-170 79 83.9 13.9 5 171-210 86 91.4 15.1 6 211-250 29 30.8 5.1 7 251-290 27 28.7 4.7 8 291-330 26 27.6 4.6 9 331-370 - - - 10 371-410 10 10.6 1.8 11 411-450 2 2.1 0.4 Sum 570 605.6 100

66

400 Height size of dodonea in G I 350

300

250

200

150 Density/ha

100

50

0

Classes

300.0 height size classes of dodonea in G II

250.0

200.0

150.0

100.0 Density/ha

50.0

0.0

Classes

67

600.0 height size of dodonea in G 3

500.0

400.0

300.0 Density/ha

200.0

100.0

0.0

Classes

height size of dodonea in G IV 250.0

200.0

150.0

Density /ha Density 100.0

50.0

0.0

Classes

68

250.0 height size of dodonea in Group V

200.0

150.0

Density/ha 100.0

50.0

0.0

classes

Figure 2.7: Height size of Dodonaea in Five Groups

2.3.12. Cover size classes of Dodonaea Table 2.12 representing cover size of Dodonaea viscosa in the five communities groups results from Ward’s cluster analysis. The cover size classes were made on per/ha basis and the cover size of Dodonaea viscosa was classified into different size classes at a regular interval of 100푐푚. The structure of the five communities are represented in the figures (1-5). The results indicating that the cover/ha of individuals of Dodonaea viscosa is high in the middle size classes in communities group I, 2 and 3 and representing bell-shape structure. Out of the four classes of group III communities the maximum cover/ha is found in 2nd class (51%) followed by class 2 (25%) and 4th (21%). In group four the cover is classified in to five classes in which maximum the cover is found in 3rd class followed by class 4th and five. The results indicate that the cover/ha was high in upper three classes. As indicated in the figures the polynomial curve of community group III and IV represent a unique pattern due to the large difference in the middle and lower size classes. These two communities’ groups (III, IV) representing un-even shaped structure.

69

Table 2.12: Cover size classes of Dodonaea viscosa in different Communities’ groups

Group I Classes Cover Cover/ha %Cover/ha 1 Upto-100 12086 56.91171 4.76153 2 101-200 53264 250.8146 20.98445 3 201-300 68533 322.7147 26.99999 4 301-400 38602 181.7728 15.20806 5 401-500 30181 142.1192 11.89043 6 501-600 35984 169.4449 14.17664 7 601-700 15176 71.46219 5.978899 Sum 253826 1195.24 100 Group II Classes Cover Cover/ha %Cover/ha 1 Upto-100 2743 17.04 1.44 2 101-200 28702 178.29 15.07 3 201-300 61079 379.41 32.08 4 301-400 55512 344.83 29.16 5 401-500 20519 127.46 10.78 6 501-600 13757 85.46 7.23 7 601-700 8091 50.26 4.25 Sum 190403 1182.75 100 Group III Classes Cover Cover/ha % Cover/ha 1 Upto-100 709 13.58522 1.312987 2 101-200 13978 267.8338 25.88566 3 201-300 27583 528.5205 51.08058 4 301-400 11729 224.7405 21.72077 Sum 53999 1034.68 100 Group IV Classes Cover cover/ha % cover 1 Upto-100 289 51.06211 5.631333 2 101-200 870 153.7164 16.95246 3 201-300 2013 355.6679 39.22447 4 301-400 1028 181.6327 20.03118 5 401-500 932 164.6709 18.16056 Sum 5132 906.75 100 Group V Classes Cover Cover/ha % Cover/ha 1 Upto-100 670 13.21516 2.124623 2 101-200 8570 169.0357 27.17615 3 201-300 10106 199.3319 32.04693 4 301-400 3283 64.75427 10.41065 5 401-500 5406 106.6286 17.14286 6 01-600 3500 69.03441 11.09878 Sum 31535 622 100

70

Group I

350 300 250 200

150 Cover/ha 100 50 0

Classes

Group II 400.00 350.00 300.00 250.00 200.00 150.00

Cover/ha 100.00 50.00 0.00

Classes

Group III

600

500

400

Cover Cover 300

200

100

0

Size Classes

71

Group IV 400

350

300

250

200 Cover Cover 150

100

50

0

Size Classes

Group V 250

200

150

Cover Cover 100

50

0

size classes

Figure 2. 8: Dodonaea viscosa Size Classes 2.4. DISCUSSION The area of Malakand division was search for Dodonaea viscosa communities in order to know about the structure and natural dynamics of the target species. Based on classification Ward’s cluster analysis Dodonaea viscosa dominated area of Malakand division was classified into five distinct plants communities. All the communities were dominated by Dodonaea viscosa while, the co-dominant species were different in different communities. The co-dominant species of community I, II were Indigofera gerardiana, in community III Berberis lycium, in community IV Aillanthus altissima while in community V the co- dominant species was Gymnosporia. A total of 27 plants species are recorded in twenty-eight sampling sites belong to twenty one families and twenty six genera. Out of twenty-one

72

families only six families shared two (for each) species. These dominant families are Asclepidaceae, Asteraceae, Lamiaceae, Oleaceae, Papilonaceae and family Rosaceae. The remaining thirteen families are monospecific. Except family Asclepidaceae all the remaining families are represented by a single genus. Similar study was conducted by Shariatullah (2013) and reported 34 species while sampling Justicia adhatoda communities in Malakand division. Rahman (2013) reported 73 species belong to 34 families while studying Seriphidium brevifolium communities in Dir Lower. Dodonaea viscosa was reported as a co- dominant species in association with Seriphidiumbrevifolium and Justicia adhatoda (Shariatullah 2013; Rahman 2013). The ivi mean values 28 stands and 27 species found in the Dodonaea viscosa dominated communities was subjected to multivariate analysis (Ward’s cluster analysis). The vegetation data was separated into five distinct plants communities such as Dodonaea-indegopera Community (I), Dodonaea-Indegopera Community (II), Dodonaea-Berberis Community (III), Dodonaea-Aillanthus Community IV, Dodonaea- Gymnosporia Community (V). Similar study was conducted by Ahmad et al (2011) for the communities’ description of Deodar forest in Himalayan range of Pakistan and reported seven plants communities. Kavgaci et al (2013) applied classification and ordination techniques and reported four plants communities of Pinus nigra dominated forests in Turkey. The same procedure was applied by Rahman (2013) and reported six plants communities. Khan et al (2011) reported five plants communities while studying the regeneration potential and structure of Monotheca buxifolia communities in district Dir lower using multivariate analysis and ordination techniques. Some species such as Indegopera gerardiana, Olea cuspidate, P. rugosus and Gymnosporia are mostly repeated with Dodonaea viscosa in the sampling sites, the reason may be the same environmental condition and nutritional need as reported by the other authors (Ali et al 2007; Ahmed et al (2009; Shariatullah 2013; Khan and Bibi 2013; Rahman 2013). Some species such as Asparigus gracilis, Plactrenthus rugosus, Quercus dilatata, Periphloca aphylla, Artemisia scoparia, Cotoniaster microphylus, Myrsine africana, P. aphylla, Asparigus gracilis, Plactrenthus rugosus, Ailanthus altissima and Xanthium stramarium are rarely found in the communities of Dodonaea viscosa. These speciesmay be loss in the future if proper care was not taken (Rahman 2013) because, these species are locally used by the inhabitant (khan et al 2011) of the area for different purposes such as medicine fuels, shelters and fodders (Ahmad et al 2009; Shariatullah 2013).

In term of species composition, Community I is the richest as compared to other communities. This community comprised of 20 species. In this community the IVI mean

73

value of Dodonaea viscosa is high as compared to group IV and V but less than II and III. This community is situated at high altitude but, low than Group III. Slope angle is steep for all communities except group IV which is situated at Moderate slope (Khan et al 2013). Beside the co-dominant species (indegopera gerardiana), some other important species of community I are, Gymnosporia, P. rugosus Daphne mucronata, Justacia adhatoda, Mallotus philiphinsis, Accacia modesta, Olea cuspidate, Berberis lycium Otostegia limbata, Vitex negundoand C. astralis etc. Beside these M. africana,C. microphyllum, F. carica and P. phyla were only found in stands of group I. Community II is the second richest comprising of 17 plants species. In this community the ivi of dominant species Dodonaea viscosa was greater than all other communities except community III (45%), but altitude was low (461푚) as compared to other such groups. The co dominant species in this community was also indegopera gerardiana but difference was found in the IVI mean value. Group III consist of three stands and six species. The dominant species of this community is Dodonaea viscosa while the co-dominant species is Berberis lycium.The ivi mean value of Dodonaea viscosa in this group was the largest as compared to other groups. It is because Dodonaea viscosa establish community best in the areas where other vegetation is disturbed or not present. Our finding is strongly supported by Bekele (2000) who reported that Dodonaea colonize barren areas and suggested that it could be used at early stages of restoration of natural forest. P. rugosus, Accacia modesta, Indigofera gerardiana and Quercus dilatata are found in association with the dominant species. Group IV comprised of 10 species in which Dodonaea viscosa is dominant while the co-dominant species is Aillanthus altissima. The other species of this group are Justicia adhatoda, Daphne mucronata, Olea cuspidate, Mallotus philiphinsis, P. rugosus, Ficus carica, Indigofera gerardiana and Xanthium stramarium. The result also shows that Aillanthus altissima and Xanthium stramarium are the species which are only found in the sampling stands of group IV. Group V is also dominated by Dodonaea viscosa but the ivi mean value of Dodonaea viscosa is low as compared to other groups. Out of 10 species of group V Gymnosporia is the co-dominant species while other species which are found in association with Dodonaea are P. rugosus, Justicia adhatoda, Otostegia limbata, Rubus fruticosus, Indigofera gerardiana, Eucalyptus lanceolatus, Olea cuspidate and Accacia modesta. The difference in the species composition and ivi of Dodonaea is because of the environmental variables and soil constituents in which difference is found in different communities. Much has been discussed about the relationship of species richness with environmental variables and soil properties of the area (Schuster and Diekmann 2005; Currie and Paquin 1987; Stevens 1992, Merganic et al. 2004; Khan et al 2011; Khan et al 2013;

74

Shariatullah 2013). The inter correlation was performed among different parameters in order to know about their relationship. A strong positive significant relationship of elevation was found with soil pH phosphorus, potassium and silt particles of the sampling stands. There is a negative significant relationship between slope angle with phosphorus, potassium while, positive significant relationship with clay particles. PH was found in significant relationship with lime content, clay % and silt % of the soil. Organic matter was found in significant relation with nitrogen potassium, nitrogen and potassium, potassium and phosphorus, lime content with silt, while negative relation was found between lime and sand particle as well as between sand and clay particles of the soil. Similar study was conducted by Khan (2012) and reported positive significant relation between altitude and potassium, slope and aspect, potassium and sodium, water holding capacity with nitrogen. There was not found any significant relation of organic matter with all the studied 15 parameters. Zhang and Zhang (2011) reported positive significant relation of soil organic matter with elevation, Nitrogen, phosphorus, and reported that the degradation of organic matter, as both Nitrogen and phosphorus comes from the same source (organic matter) because organic matter is the most important variable in the soil (Tang and Ohsawa 1997; Martins et al., 1999) while the role of aspect is not clear on the nutrient of soil but affect the pattern of vegetation of communities. In the present study the groups derived through Wards cluster analysis were also superimposed on the NMS and DCA ordination plots. Greig-Smith (1983) reported that cluster analysis and ordination are complimentary to each other though fundamentally used for different purposes. In the present study 13 different variable were correlated with NMS ordination axis I, II in which only one factor (% Clay particles) showed significant relationship with the NMS ordination axis 1. Axis 2 of NMS ordination showed strongly significant correlation with slope, Potassium and % clay particles. Rahman (2013) found positive significant correlation of altitude and soil organic matter with NMS axis 1, while altitude, slope, silt, sand, lime and potassium were found in positive significant correlation with axis 2 of NMS ordination. Similarly DCA ordination axes were found in correlation with altitude, slope, aspect, silt, sand, organic matter, lime, phosphorus (P) and potassium (K). Same procedure was applied by Khan (2012) while studying the community structure of Quercus baloot in District Dir lower and did not found any significant correlation of environmental factors with DCA ordination axes and argued that the soil moisture, soil nutrients, rainfall, past disturbances, mass effects and chance factor are responsible for the distribution of forest communities. In the present study slope angle of the site, pHand organic matter were found in significant relationship with DCA ordination axis 1. DCA ordination

75

axis 2 was correlated significantly with three environmental factors such as soil water holding capacity, organic matter and nitrogen contents of the soil. The results also show that DCA ordination axis 3 is not significantly correlated with any factor. Similar study was conducted by Khan et al (2011) and reported positive significant relationship of Ordination axis with elevation, salinity, conductivity, Na, Nitrogen and soil. Our result is in correlation with Hill and Gauch (1980), Gauch and Whittaker (1981), and McCune and Mefford (2005) who stated that DCA is only capable of yielding one basic gradient associated with the vegetation. As it is clear from the results that the distribution of vegetation communities is due to those parameters which are significantly correlated with ordination axis while, the other factors moving along. Taking into consideration the density/ha and cover/ha, a large difference is observed among the density and cover/ha of Dodonaea viscosa and other associated species. the difference is because of the variation in the values of environmental variables such as altitude, pH, Organic matter, clay % slope angle, soil water holding capacity and Potassium content of the soil as these parameters were found in significant correlation with DCA, and NMS ordination axes. The highest density/ha of Dodonaea viscosa was recorded in community group III as this community is situated at high altitude, high slope angle, organic matter, water holding capacity clay%, and Nitrogen content in the soil were high as compared to other communities while potassium content were medium in soil and the soil was acidic in nature. Our finding shows that the density/ha, cover/ha and distribution of Dodonaea viscosa and associated plants species is controlled by the above mentioned parameters which are found in significant correlation with the DCA and NMS ordination axes. Similar study was conducted by Ahmed et al (2009) studying the community structure of Olea ferruginea in Dir Lower and stated that the density decrease with increase in elevation. Ahmad et al (2006) reported that density of vegetation depend upon environmental factors and found the highest stand density at northern facing aspect in Cedrus dominated forests. Similarly Khan et al (2013) and Titshall et al (2000) reported that slope and altitude are the main environmental variables which control the distribution of vegetation density. Our finding is in correlation with Rahman (2013) reported that beside altitude aspect also play important roles in the distribution of vegetation communities. El-Sheikh and Yousaf (1981) reported that beside the other factors soil moisture play a key role in the in the distribution of plants communities. Tavili & Jafari (2009) reported that nutrient status, EC, soil texture, slope andaspect are the most important factors that correlated strongly with the distribution of ecological communities. Our results is also strongly supported by Zare et al. (2011) who described the

76

importance of slope, aspect elevation, soil texture, available nitrogen, potassium, organic matter, lime and soil moisture in the distribution of the vegetation. In community III only five species were found in association with Dodonaea viscosa which shows that the density of the target species increase when other vegetations are degraded (). The results of soil analysis also shows that Dodonaea viscosa mostly grows in calcareous soils as about 95% communities of Dodonaea viscosa belong to calcareous soils with a texture Loamy Sand to sandy loam with low phosphorus and marginal nitrogen and potassium contents. All the Dodonaea viscosa communities are reported from rocky soil and windy areas. Our results are also supported by Edward & Gilman (1999) who reported that Dodonaea viscosa tolerates sandy or rocky soils, salt spray, windy areas and drought conditions. The communities were reported from any aspect as north, south, east and west which shows that Dodonaea viscosa can grows at any aspect but it favors areas which receive full sunlight and is often cultivated in loamy or sandy soils. The height and cover sized classes were made based on density/ha and cover/ha in order to know about the structure pattern. The height and cover sized classed showed a multi- model pattern such as reverse J-shaped, L-shaped, Bell-shaped almost bell shaped and Un- even pattern. The reverse J-shaped and L-shaped represent that more individuals are present in lower sized classes while the bell-shaped and almost Bell-shaped indicating the high density in upper sized classes the high density in lower classes is due to two reason (i) the communities are recently introduced to the area or (ii) strong regeneration potential of Dodonaea while the density in middle sized classes shows bell shaped pattern which is due to the weak regeneration or anthropogenic disturbance due to which less number of individuals remains to produced seeds as the people cutting the plants for different purposes. The un- even pattern is because of the anthropogenic disturbance as observed during field survive. Similar study was conducted by (Rahman 2013) for Seriphidium brevifolium in District Dir lower and reported irregular, almost bell-shaped and bell-shaped pattern. That the high density in the lower is because of strong regeneration potential of the species while (Shariatullah 2013) while the less density in the higher classes may the growth rate as the seedling and juvenile plants grow more rapidly as compared to mature plants (Hitimana et al 2004). The pattern in the size classes can be modified due to anthropogenic activities inter- specific and intra-specific competition, pattern of regeneration, difference in the soil and climatic condition of different areas (Brunig 1983; Denslow 1995). (Faridah-hanum et al 2012). The density in different size classes tells us about the resource utilization of the site by

77

the species. There were also found some gapes in sized classes of Dodonaea viscosa which is because of the over exploitation as reported by (Shariatullah 2013; Rahman 2013).

78

CHAPTER 3: REGENERATION STATUS OF DODONAEA VISCOSA IN STUDY AREA

79

CHAPTER 3: REGENERATION STATUS OF DODONAEA VISCOSA IN STUDY AREA

Abstract

The present study the regeneration status of Dodonaea viscosa communities was investigated in Malakand division. Seedling and sapling data was collected in 28 sampling stands using quadrat method. Aspect, elevation and slope angle of the sampling site were measured. Soil samples were collected from each stand. Density/ha of seedling, sapling and mature plants of Dodonaea viscosa were calculated. Pearson’s correlation and regression analysis were performed. In all the communities seedling density was high than saplings as well as sapling than mature plants which representing the normal regeneration although grazing of animals, using for fuels and anthropogenic disturbance has delimited the natural regeneration of Dodonaea viscosa. The seedling and sapling density of Dodonaea viscosa was high in the communities’ group III indicating that density of seedling and sapling of Dodonaea viscosa increase up to certain height (1083)푚 and then decrease gradualy. Pearson’s correlation co- efficient showed a strong positive relationship between sapling and seedling densities (푝 > 0.001), sapling and organic matter (푝 > 0.05) and seedling and organic matter, while a negative significant relationship was found between sapling and soil pH (푝 > 0.05), seedling and soil pH (푝 > 0.001), as well as sapling and elevation (푃 > 0.01). Regression analysis of Dodonaea viscosa indicating a strong positive significant relationship between Seedling and Sapling (푟 = 0.919 at 푃 > 0.001), sapling/organic matter (푟 = 0.457 at 푃 > 0.05). The Seedling/Elevation showed a negative significant relationship (푟 = 0.525 at 푃 > 0.01). The R-values of regression analysis of seedling/soil pH and sapling /pH was 푟 = 0.529, 푟 = 0.386 respectively, which shows a negative significant relationship of pH with seedling and sapling at the p-values 푃 > 0.01, 푃 > 0.05. It means that among the studied parameters anthropogenic disturbance, soil pH, organic matter and elevation mostly affect the regeneration of Dodonaea viscosa.

3.1. INTRODUCTION Plant regeneration can be defined as the surviving ability of seedling and sapling to replace the mature plants that die is known as regeneration (NCRN 2012). It has been observed that beside destruction of habitat the high density of livestock mostly reduce the regeneration of seedling and sapling of forest vegetation and the diversity of plants communities (Tilghman 1989; Frelich and Lorimer 1985; McCormick et al 1993). The rate of shrub recovery depends

80

on several factors among which the major factor is the mode of reproduction of the species (Sampson 1944). Regeneration of most species of shrubs plants is through re-sprouting from the underground parts and through seedling from the seeds stored in the soil (Keeley 1977). Very little literature is available on the factors which promote the regeneration of seedling (keeley & sterling 1981). The factors which mostly influence the regeneration of seedling of shrubby species are latitude, elevation, topography, slope, aspect, and the pattern of weather condition of an area (Hanse 1971). It is also experimentally showed that natural regeneration of seedling and sapling depends on existing mature vegetation likely to produce seeds and generate new forests (Colombo 2005). The competition of vegetation species for nutrients, space, humidity, sunlight herbivores, nature and physical properties of soil (shallow or rocky soil), climate, topography and anthropogenic activities also limit the regeneration of seedling and sapling (Colombo 2005; Noor & Khatoon 2013).

The stage of seedling recruitment is the first step for the determination of future stand structures, condition of habitat as well as vulnerability to disturbance (Christopher et al 2005). Cutting for fuels, local and commercial purposes, grazing of animals and expansion of cultivated field are the main causes of disturbance of vegetation communities (Noor & Khatoon 2013).

The recruitment of seedling is the first and critical process followed by regeneration. The sequential pattern of seedling recruitment positioned the stage for the subsequent developmental pattern of the vegetation (Christopher et al 2005). Many species regenerate after particular type (fire or tree-fall) or size gap initiating disturbance while other establish beneath intact forest canopies (Taylor & Halpern 1991). Where forests are composed of diverse assemblages of tree species with distinct life histories or responses to disturbance, patterns of species, abundance may be used to reconstruct disturbance history (Oliver & Stephens 1977; Foster 1988). In some forests, however, plant species have multiple modes of regeneration and may established abundantly after various types or sizes of disturbances (Stewart 1986). In these systems, the population structure of forest patches, rather than their composition, better reflects the influence of disturbance on stand development.

Dodonaea viscosa L. is a perennial evergreen shrub belong to sapindaceae family (Nasir & Ali 1972). Though, it is Australian in origin but also distributed in the tropical (Rani et al 2009), subtropical ant temperate region of the world (Little & Skolmen 1989; Prakash et al 2012). The species is mostly grow in sandy or rocky, windy area in the drought habitat and

81

mostly favor east facing slope having sandy and loamy soil (Rani et al 2009). Usually Dodonaea viscosa regenerates through seeds though its propagation is also practiced successfully through cutting of twigs. The seeds are drought tolerant and remain viable for a long time and germinate successfully after rainfall. It is also investigated that for the survival of seedling early rainfall is more necessary (Gilman 1999).

Dodonaea viscosa is an important medicinal shrub use for the treatment of so many diseases locally as well as in pharmaceutical industries for the preparation of medicines. Leaves and stem of Dodonaea viscosa is used for the treatment of sore throats, fever, and seeds are use for malarial cure, as well as leaves are alone also use for irritation aches(Rojas et al 1996). Beside this it is also use for the treatment of ulcer, diarrhea, constipation (Cribb et al 1981; Little & Skolmen 1989; Wagner et al 1987) and so many other diseases (Ghisalberti 1998; Siddiqui 1998; Wagner et al 2005; Getie et al 2000).

Much has been discussed about the regeneration potential of seedling and sapling of different plants species (Bace et al 2011; Christopher et al 2005 Natalie et al 2011; Bekele 2000; Khan 2011; Rahman 2013), but little is known about the natural regeneration of seedling and sapling of Dodonaea viscosa in Malakand division. In the present study an attempt was made to investigate the natural regeneration of seedling and sapling of Dodonaea viscosa community’s in Malakand division.

82

3.2 MATERIALS AND METHODS

3.2.1. Field survive In the years (2010-14) the area of Malakand division was survived for the seedling and sapling of Dodonaea viscosa communities in order to know about the natural regeneration and future trends of the target species in their natural habitat. This species is mostly grow in the foothills of Malakand Division (Khan et al 2011; Shariatullah 2012; Rahman 2013) therefore those area were selected where the species was found dominant and the site was least disturbed.

3.2.2. Design of quadrat and collection of data.

Quadrate method was applied for the sampling of seedling and sapling of Dodonaea viscosa following Cox (1990). At each sampling site a total of ten quadrat were placed randomly and the size of quadrat was selected as 10 × 10푚 for sampling. Inside each quadrat the number of seedling and sapling along with Juvenile plants of Dodonaea viscosa was counted following the procedure used by Hussain (1984). Elevation of the sampling stand was measured in meter obtained through GPS (global positioning system) and aspect of the sampled stands was determined through magnetic compass while, clinometer was used for the measurement of slope angle following (Khan 2012; Khan et al 2013; Shariatullah 2013; Rahman 2013).

3.2.3. Collection of soil samples For the collection of soil samples polythene bags were used. 1kg Soil samples were collected at each sampling stand up to a depth of 30푐푚 from four different places and mixed to form a composite sample. The bags were labeled and taken to the agricultural research center Takhta band Mingora Swat for further analysis.

3.2.4. Laboratory procedure

Seedling and sapling data analysis

For the investigation of regeneration status of Dodonaea viscosa in different communities the density and density/ha of seedling and sapling was calculated and compared with the density/ha of juvenile plants of Dodonaea viscosa.The data of seedling and sapling of Dodonaea viscosa was entered into Microsoft Office Excel 2003 for the analysis of density, and density/ha using the formulas given below.

83

Number of individual s of a species in all quadrats Density (D1)  Number of quadrats taken

Density of a species Density/ha (D2)  10,000 Area of quadrat  no of Quadrats

Soil Analysis

Soil was analyzed for physical and chemical parameters such as soil water holding capacity, organic matter, soil pH, lime contents, soil texture (silt, sand and clay particles) and inorganic nutrients such as N.P.K in agriculture research Centre Takhtaband (swat). 1:5 soil water suspensions were used for the determination of pH of the soil samples following the procedure used by Black (1965). Silt, sand and clay)% were analyzed through hydrometer following (Bouyoucos 1936) while, the % age of organic matter was determined following Walkley (1947). Bingham (1994) method was used for the determination of phosphorus. (Sultan-pur & Schwab 1977) was followed for the determination Nitrogen and potassium, M.No. 3, and AB-DTPA was used for the extraction of N and K from Basic soil (Sultan-pur & Schwab 1977). Soil water holding capacity was determined following (Harding and Ross, 1964), while acid base neutralization method was used for lime contents of the soil (Rahman et al 2012).

3.2.5. Statistical Analysis

Pearson product movement correlation and regression analysis were applied for the interpretation of Soil, environmental variables, with seedling and sapling density/h of Dodonaea viscosa following (Khan et al 2011; Rahman 2013).

84

3.3. RESULTS

3.3.1. Regeneration Potential of Dodonaea viscosa The density/ha mean values of seedling, sapling and mature plants of Dodonaea viscosa in different groups is given in table (3.1). As it is clear from the results that the highest mean value of seedling density/ha, (3454±941) is found in group II followed by group III (2808 ± 145)/ha and group I (2439±773)/ha while, less number of seedling/ha were present in communities of group V. the sapling density/ha of Dodonaea viscosa was 2061±568 in group I, 2271±694 in group II, 2392 ± 83,and 1208 ± 414 density/ha in group (III, IV and V) respectively which indicating that the sapling density was high in communities of group III. The mature plants of Dodonaea were also more in group III as compared to other groups. A large difference is observed among the density/ha of seedling, sapling and mature plants of Dodonaea viscosa with in the communities groups. In all groups the density/ha of seedling is high followed by sapling and mature plants which indicating the normal regeneration potential but the anthropogenic disturbance and demand for fuels purposes reduce the population of Dodonaea viscosa.

Table 3. 1: Density/ha mean values of Seedling, sapling in comparison to mature Dodonaea viscosa

D/ha Group I Group II Group III Group IV Group V Total Parameter Mean±SE Mean±SE Mean±SE Mean±SE Mean±SE

Seedling 2439±773 3454±941 2808±145 2744±136 1296±389 12741

Sapling 2061±568 2271±694 2392±83 2363±131 1208±414 10295 Mature 928.67±12 864.85±6 1524.39±2 605.10±95. 605.57±10 4348.6 plants 9 0 6 6 5 27384.5 Total 5428.67 6409.85 6724.39 5712.1 3109.6 8

3.3.2. Cross correlation of seedling and sapling with environmental variables Table (3.2) represented the Pearson’s products movement correlation co-efficient of sapling and seedling densities /ha of Dodonaea viscosa with environmental and soil variables. A

85

strong positive inter-relation is found between sapling and seedling densities (푝 > 0.001), sapling and organic matter of the soil (푝 > 0.05) while a negative significant relationship is present between sapling and soil pH at the probability level 푝 > 0.05. There was not found any significant relationship of the remaining ten parameters with sapling of Dodonaea viscosa. A negative significant relationship was found between seedling density and elevation of the sampling sites at the probability level (푃 > 0.01). Similarly soil pH was also found in negative significant relation with seedling density at the given probability level (푝 > 0.001). There was found a positive significant relationship between soil organic matter seedling while, not found any significant relationship between seedling density and the remaining studied parameters. The inter correlation between the remaining parameters is represented in table (3.2).

86

Table 3.2: Inter-correlation among Seedling, Sapling and environmental variables

Sapling Seedling El Sl As Whc/10g pH %OM %Lime N(g/kg) P(mg/ kg) K(mg/ kg) Sa% Cl % Si % Sapling 1 Seedling 0.910*** 1 El -0.371 -0.525** 1 Sl -0.105 -0.016 -0.226 1 As -0.077 -0.225 0.234 -0.039 1 Whc/10g 0.010 0.021 0.077 -0.048 0.12 1 pH -0.386* -0.529*** 0.425* -0.031 0.27 0.126 1 %OM 0.458* 0.380* 0.06 -0.047 -0.177 0.221 0.057 1 %Lime -0.016 -0.175 0.236 0.121 0.358 -0.168 0.513** -0.015 1 N (g/ kg) 0.242 0.134 0.013 -0.183 -0.129 0.269 0.063 0.854*** -0.076 1 P(mg/ kg) -0.214 -0.230 0.599*** -0.408* -0.269 0.083 0.052 0.195 -0.207 0.249 1 K (mg/ kg) -0.006 -0.134 0.394* -0.544** -0.196 0.036 0.182 0.397* -0.108 0.456* 0.734*** 1 Sa% -0.234 -0.095 -0.175 0.145 -0.165 0.208 -0.31 -0.246 -0.684*** -0.192 -0.06 -0.144 1 Cl % 0.107 -0.090 0.182 -0.256 0.465* -0.154 0.418* 0.029 0.640*** 0.043 -0.084 0.2 -0.673*** 1 Si % -0.059 -0.265 0.517** -0.252 0.367 -0.003 0.553** 0.243 0.641*** 0.291 0.269 0.359 -0.811 0.631*** 1

Key: Sapling=sapling density/ha, seedling=seedling density/ha, El=Elevation, Sl=Slope, As= Aspect, Whc/10g=water hc/10gm, pH. %OM=Organic matter, % Li=%Lime, N(g kg-1)=Tot.N (g kg-1), Phosphors=P (mg kg-1), K (mg kg-1), =Potassium (mg kg-1), Sa= Sand %, Cl=Clay %, Si=Silt %.

87

3.3.3. Regression analysis Regression analysis (Table 3.3) was performed in order to know about the environmental variables which mostly affect the growth of seedling and sapling of Dodonaea viscosa. The results indicating that a strong positive significant relationship was observed between

Seedling and Sapling (푦 = 1.2369푥 + 18.633, 푟 = 0.919 at 푃 > 0.001, sapling/organic matter of the soil (푦 = 1397.8푥 + 271.07), 푟 = 0.457, 푃 > 0.05 probability level. The

Seedling/Elevation showed a negative significant relationship (푦 = −2.4454푥 + 4724.5),

푟 = 0.525푃 > 0.01. The R-values of regression analysis of seedling/soil pH and sapling /pH was 푟 = 0.529, 푟 = 0.386 respectively, which shows a negative significant relationship of pH with seedling and sapling at the probability values 푃 > 0.01, 푃 > 0.05. It means that among the studied thirteen parameters soil pH, organic matter of the soil and elevation are the most important factors which mostly affect the regeneration of Dodonaea viscosa. The regression relationship of seedling and sapling versus various soils and other environmental variables are also represented in figures (1 − 27).

88

Table 3. 3: Regression relationship of seedling, sapling /environmental factors

Parameters Equation R-values P-value Remark s Seedling /Sapling 푦 = 1.2369푥 + 18.633 푅 = 0.919 P>0.001 Sig Sapling /elevation 푦 = −1.2724푥 + 3162 푅 = .371 Seedling/Elevation 푦 = −2.4454푥 + 4724.5 푅 = .525 P>0.01 Sig Sapling /Slope angle 푦 = −10.558푥 + 2480.3 푅 = 0.104 Seedling / Slope angle 푦 = −2.1447푥 + 2591.8 푅 = 0.014 Sapling /Aspect 푦 = −94.843푥 + 2184.7 푅 = 0.076 Seedling /Aspect 푦 = −377.02푥 + 3221.7 푅 = 0.225 Seedling /water ho capacity 푦 = 10.232푥 + 2447.9 푅 = 0.021 Sapling / Water holding capacity 푦 = 3.6585푥 + 1985.1 푅 = 0.011 Seedling /pH 푦 = −727.43푥 + 6880 푅 = 0.529 P>0.01 Sig Sapling /pH 푦 = −390.21푥 + 4354.2 푅 = 0.386 P>0.05 Sig Seedling / %OM 푦 = 1555.6푥 + 568.51 푅 = 0.377 P>0.05 Sig Sapling /%OM 푦 = 1397.8푥 + 271.07 푅 = 0.457 P>0.05 Sig Seedling /Lime% 푦 = −80.198푥 + 3528.1 푅 = 0.174 Sapling Lime% 푦 = −5.3175푥 + 2070.3 푅 = 0.014 Seedling /Nitrogen 푦 = 321.29푥 + 1789.8 푅 = 0.133 Sapling Nitrogen 푦 = 426.64푥 + 1065.8 푅 = 0.242 Seedling /Phosphorus 푦 = −148.02푥 + 2787.8 푅 = 0.230 Sapling /Phosphorus 푦 = −101.35푥 + 2202.5 푅 = 0.214 Seedling /Potassium 푦 = −3.5681푥 + 2878.2 푅 = 0.133 Sapling /Potassium 푦 = −0.1231푥 + 2015 푅 = 0.006 Sapling /Sand % 푦 = −27.502푥 + 3774.6 푅 = 0.234 Seedling /Sand% 푦 = −15.216푥 + 3475.5 푅 = 0.095 Seedling Clay % 푦 = −46.004푥 + 2789.4 푅 = 0.09 Sapling/Clay % 푦 = 39.983푥 + 1745.6 푅 = 0.106 Seedling /Silt% 푦 = −42.207푥 + 3544.2 푅 = 0.264 Sapling /Silt % 푦 = −6.9199푥 + 2173.9 푅 = 0.059

89

Seedling /Sapling y = 1.2369x + 18.633 9000 R = 0.919, P>0.001 8000 7000 6000 5000

Sapling 4000 3000 2000 1000 0 0 1000 2000 3000 4000 5000 6000 7000

Seedling

Figure 3. 1: Seedling /Sapling

Sapling /elevation

y = -1.2724x + 3162 R = .3713, P>0.01 7000

6000

5000

4000

3000 Sapling

2000

1000

0 0 200 400 600 800 1000 1200 1400 1600 1800 2000

Elevation

Figure 3. 2: Sapling /elevation

90

Seedling/Elevation

y = -2.4454x + 4724.5 9000 R = .5251 8000 7000 6000 5000

Seedling 4000 3000 2000 1000 0 0 500 1000 1500 2000

Elevation

Figure 3. 3: Seedling/Elevation

Sapling /Slope angle

7000 y = -10.558x + 2480.3 R = 0.1048 6000

5000

4000

3000 Sapling

2000

1000

0 0 10 20 30 40 50 60 70

Slope angle

Figure 3. 4: Sapling /Slope angle

91

Seedling Slope angle

y = -2.1447x + 2591.8 9000 R = 0.0141 8000 7000 6000 5000

4000 Seedling 3000 2000 1000 0 0 10 20 30 40 50 60 70

Slope Angle

Figure 3.5: Seedling Slope angle

Sapling /Aspect

y = -94.843x + 2184.7 7000 R=0.0768 6000

5000

4000

3000

2000

1000

0 0 1 2 3 4 5

Sapling

Figure 3.6:Sapling /Aspect

92

Seedling /Aspect

y = -377.02x + 3221.7 9000 R =0.225

8000

7000

6000

5000

4000 Seedling 3000

2000

1000

0 0 1 2 3 4 5

Aspect

Figure 3.7: Seedling /Aspect

Seedling /water ho capacity

y = 10.232x + 2447.9 9000 R =0.02 8000

7000

6000

5000

seedling 4000

3000

2000

1000

0 0 5 10 15 20

water ho capacity

Figure 3.8: Seedling /water ho capacity

93

Sapling /Water holding capacity

y = 3.6585x + 1985.1 7000 R =0.01 6000

5000

4000

Sapling 3000

2000

1000

0 0 5 10 15 20

Water holdin capacity

Figure 3.9: Sapling/Water holding capacity

Seedling /pH

y = -727.43x + 6880 9000 R =0.529 8000 7000 6000 5000

seedling 4000 3000 2000 1000 0 0 2 4 6 8 10

pH

Figure 3.10: Seedling /pH

94

Sapling /pH

y = -390.21x + 4354.2 7000 R =0.386

6000

5000

4000

Sapling 3000

2000

1000

0 0 2 4 6 8 10

pH

Figure 3.11: Sapling /pH

Seedling OM% y = 1555.6x + 568.51 9000 R =0.3766 8000

7000

6000

5000

4000 seedling 3000

2000

1000

0 0 0.5 1 1.5 2 2.5

% OM

Figure 3.12: Seedling OM%

95

Sapling /%OM

y = 1397.8x + 271.07 7000 R =0.4576

6000

5000

4000

Sapling 3000

2000

1000

0 0 0.5 1 1.5 2 2.5

% OM

Figure 3.13:Sapling /%OM

Seedling /Lime%

y = -80.198x + 3528.1 9000 R =0.1746 8000

7000

6000

5000

Seedling 4000

3000

2000

1000

0 0 5 10 15 20 25

Lime %

Figure 3.14: Seedling/Lime%

96

Sapling Lime%

y = -5.3175x + 2070.3 7000 R =0.0141

6000

5000

4000

Sapling 3000

2000

1000

0 0 5 10 15 20 25

Lime %

Figure 3.15: Sapling Lime%

Seedling /Nitrogen y = 321.29x + 1789.8 9000 R =0.1337 8000 7000 6000 5000

4000 Seedling 3000 2000 1000 0 0 1 2 3 4 5

Nitrogen

Figure 3.16: Seedling /Nitrogen

97

Sapling Nitrogen

y = 426.64x + 1065.8 7000 R =0.2416

6000

5000

4000

Sapling 3000

2000

1000

0 0 1 2 3 4 5

Nitrogen

Figure 3.17: Sapling Nitrogen

Seedling /Phosporus

y = -148.02x + 2787.8 9000 R =0.2302 8000 7000 6000 5000

4000 seedling seedling 3000 2000 1000 0 0 2 4 6 8 10 12

Phosporus

Figure 3.18: Seedling /Phosporus

98

Sapling /Phosporus

y = -101.35x + 2202.5 7000 R =0.2142

6000

5000

4000

Sapling 3000

2000

1000

0 0 2 4 6 8 10 12

Phosporus

Figure 3.19: Sapling/Phosphorus

Seedling /Potasium

9000 y = -3.5681x + 2878.2 R =0.1337 8000 7000 6000 5000

4000 Seedling 3000 2000 1000 0 0 50 100 150 200 250 300

Potasium

Figure 3.20: Seedling /Potassium

99

Sapling /Potassium

y = -0.1231x + 2015 7000 R = 0.006

6000

5000

4000

Sapling 3000

2000

1000

0 0 50 100 150 200 250 300

Potassium

Figure 3.21: Sapling /Potassium

Sapling /Sand %

y = -27.502x + 3774.6 7000 R² = 0.0549

6000

5000

4000

Sapling 3000

2000

1000

0 0 20 40 60 80 100

Sand %

Figure 3.22: Sapling /Sand %

100

Seedling Sand%

y = -15.216x + 3475.5 9000 R =0.0953

8000

7000

6000

5000

4000 Seedling

3000

2000

1000

0 0 20 40 60 80 100

Sand %

Figure 3.23: Seedling Sand%

Seedling Clay %

y = -46.004x + 2789.4 R =0.09 9000

8000

7000

6000

5000

4000 seedling 3000

2000

1000

0 0 5 10 15 20

Clay %

Figure 3.24: Seedling Clay %

101

Sapling/Clay %

y = 39.983x + 1745.6 R =0.1067 7000

6000

5000

4000

Sapling 3000

2000

1000

0 0 5 10 15 20

Clay %

Figure 3.25: Sapling/Clay %

Seedling Silt%

y = -42.207x + 3544.2 9000 R = 0.2647 8000

7000

6000

5000

4000 Seedling

3000

2000

1000

0 0 10 20 30 40 50

Silt %

Figure 3.26: Seedling Silt%

102

Sapling /Silt %

7000 y = -6.9199x + 2173.9 R =0.591 6000

5000

4000

Sapling 3000

2000

1000

0 0 10 20 30 40 50

Silt%

Figure 3.27: Sapling /Silt %

103

3.4. DISCUSSION The Dodonaea viscosa dominated area of malakand division was surviewed in order to know about the natural regeneration status of seedling and sapling of the target species (Dodonaea viscosa) in different communities. Though all the sampling stands contain seedling and sapling of Dodonaea viscosa but a large difference is observed between the seedling and sapling densitiey in different stands. The difference in the seedling, sapling and mature plants of Dodonaea viscosa is due to the difference in the parameters such as soil pH, soil water holding capasity, organic matter, Nitrogen, Phosphorus ,Potassium contents, altitude, aspect and slope angle of the sampling sites. Our finding is in correlation with Hanse (1971) stated that latitude, elevation, topography, slope, aspect, and the pattern of weather condition of an area mostly influence the seedling regeneration of shrubby species. though in all the sampled sites, seedling density of Dodonaea viscosa was high than its sapling, sapling than mature plants density. Similar study was conducted by Rahman (2013) recorded high sapling density than seedling and argued the abnormal regeneration of the plant species. Khan et al (2011) also reported less density of seedling as compared to choped stems of Monotheca buxifolia and associated tree species which is also the abnormal regeneration as reported by West et al (1981). Our finding is not correlated with them because in our finding the seedling and sapling density in relation to mature plans of Dodonaea indicating the good and normal regeneration of seedling and sapling. Our result is strongly supported by Khan et al (1987) and Manoj et al (2008) who reported that the regeneration of a species depend upon the seedling and sapling density and good regeneration will be that in which seedling density is more than sapling and that of the sapling is greater than mature plant’s density. The result also indicated that although the regeneration is normal while anthropogenic disturbance, grazing of livestock and demand for fuels purposes has greatly affected the density of seedling and sapling of Dodonaea viscosa communities.The seedling and sapling density of Dodonaea was high in the communitie’s of group III which is situated at high altitued as compared to the other communities groups obtained through Ward’s cluster analysis this community was highly disturbed and comprised of only six species. Beside the environmental variable topography and soil contents it is also reported that Dodonaea viscosa can establish itself on degraded land (Bekele 2000). The stands of this group of Dodonaea viscosa communities were situated at (1083) m elevation, slope angle (53o) was high as compared to other groups which shows that the density of seedling and sapling of Dodonaea viscosa increase upto certain height (1083) m and then decrease gradualy. Similar study was

104

also conducted by Rahman (2013) to study the regeneration potential of Seriphidium brevifolia (Wall ex. DC)and Khan et al (2011) for the regenration capasity of Monothica buxifolia in district Dir (Lower). The soil water holding capasity, organic matter and nitrogen contents were high than all the remaining group. Our result is supported by Singh (1986) who stated that the colloidal nature of organic matter consequently increase the water holding capacity of soil. phospherus and potassium contents were mediumand and pH was low as compare to other groups. It is also reported that Seedling generally preferred less content of potassium in soil (Rahman 2013). The results obtained through Pearson’s correlation co- efficient showed that there was strong positive relationship between sapling and seedling densities (푝 > 0.001), sapling and organic matter (푝 > 0.05) and seedling and organic matter of the soil, while a negative significant relationship was found between sapling and soil pH (푝 > 0.05), seedling and soil pH (푝 > 0.001), as well as sapling and elevation (푃 > 0.01) while the remaining factors don’t shows significant relation with seedling and sapling density of Dodonaea viscosa. Grubb et al. (1963), reported that in an undisturbed area density is closely related to slope of the sampling site. Barnes et al (1997) reported that slope, aspect, and soil characteristics are the factors which determining the structure of vegetation. Khan et al (2011) and Shariatullah (2013) documented that low slope, high elevation promote the organic matter of the soil and lime content which finally have an effect on all the Eco physiological process of a species. Bekele (2000) reported that seasonal rainfall is a dominant factor which regulating and establishment, recruitment, survival and growth at seedling stage.

105

CHAPTER 4: THE EFFECT OF DIFFERENT SOIL AND SHADE REGIME ON GERMINATION AND GROWTH PATTERN OF DODONAEA VISCOSA

106

CHAPTER 4: THE EFFECT OF DIFFERENT SOIL AND SHADE REGIME ON GERMINATION AND GROWTH PATTERN OF Dodonaea viscosa

Abstract

Both of soil and shade are the most essential factors for the establishment and growth of plants. This study firstly investigated the reproduction of two stages of Dodonaea viscosa, seed germination and seedling growth. We focused on the various screening effects of four combinations i.e. B1S1-seeds sowed in pure Garden soil; B2S2-seeds sowed in field soil substrate containing organic manure (Humus); B3S3-seeds sowed in original soil collected from the field (OS-Original soil); B4S4-seeds sowed in undergrowth soil substrate and placed under shade of trees. The seeds were sowed according to their polarity in the bags, which were arranged according to a randomized complete block design (RCBD), with four treatments per block and four repeats. Each experimental unit included 30 plant pots, each of which contained a seeds. The test was conducted from 9 July 2012 − 4July2013. The results showed significantly different impacts on Dodonaea viscosa seed germination in the soil types with various textural, physiochemical compositions and the influence of shade. Various type soils and soil under shade regimes from different sites of the study area have significant differences (푝 < 0.05) for germination percentage, leaf morphology, stem diameter, height, cover, root length and above ground biomass respectively in the four experiments. Among the Dodonaea seeds collected from open areas and uphill slope there was wide variation in seed characteristics and germination patterns ranged from 11.11% to 53.33%, and the lowest germination was observed in shade pots. Germination percentage was adversely influenced by shade, where germination percentages were below 11% of the total 90 seeds sown. In the comparisons among original soil (OS) with the same environmental factor, it was alwaysshowed that the germination rate of seeds was higher inthe order of on Garden soil, Humus mixed soil, and soil placed in shade. It was concluded from the results that Dodonaea viscosa have variable response to different soil substrate and shade regimes and well adopted to original soil with open sunshine.

Key words: Dodonaea viscosa, soil types, seed germination, shade regimes; Hindukush range

107

4.1. Introduction

Dodonaea viscosa Jacq., belong to Sapindaceae is a dominant shrub species in the sub- tropical dry temperate forests located in the northern Hindukush and Himalayan ranges mountains in Pakistan (Champion et al. 1965; Sheikh et al. 1994). It is a long lived woody perennial spreading or erect shrub or tree up to about 5 to 8 푚 (Walsh and Entwisle 1996). Dodonaea viscosa is very effective in sand dune fixation and controlling coastal erosion since its roots are excellent soil binders (Khan et al. 2010; Khan et al. 2011).

The species is also used to reclaim marshes, grown as an ornamental plant for its shiny foliage and pink-red winged fruit and Poles are useful in fencing (Personal observations). The timber is hard and durable and generally used for roof thatching materials of houses in northern areas of Pakistan (Ali et al. 2007). Studies have shown that roots are used in the preparation of medicinal oil, which is used to treat rheumatism whereas, leaves are also used in the treatment of rheumatism and bone fracture.

The geographic range of the species is very wide, occurs in all regions of Victoria and across all mainland Australian states, the Pacific Islands, Asia, Africa and the Americas (Nasir and Ali 1972). Dodonaea viscosa ssp. cuneata is typically found in dry forest communities, grasslands and grassy woodlands and some riparian ecosystems between an altitudinal range of 400푚 to 1800 푚. However, the information regarding its fragmentation and population density is yet not known though there has been a trial into establishment and germination (see Semple and Koen 1996).

The sparse and variable precipitation in arid and sub-tropical regions is believed to exert strong control over the life histories, physiological characteristics, and species composition of their biota (Champion et al. 1965; Khan et al. 2011).

A seed will go through many tests (e.g. predation, freezing, drought, physical damage and water stress) and stages (e.g. pre-dispersal, dispersal, germination, seedling and sapling) until being a survived sapling (Khan 2011; Wahab 2011). Although seed dispersal is a key process in plant population dynamics (Harper 1997), the post-dispersed environment has a powerful screening effect on the survival probability of seed and its subsequent periods.

108

As an outstanding germplasm resource in Inner Hindukush region, it is being seriously threatened by the deterioration of the sub-tropical environment (Ahmed et al. 2011). Dodonaea viscosa cannot be successfully regenerated by seeds in natural conditions, but it also could regenerate by a large production of roots. It’s extremely urgent for us to solve the problem of seed regeneration failure because of the lower diversity leading by long-term seed dormancy in the soil due various environmental factors.

Knowledge of seed movement and fates of Dodonaea viscosa are essential for ecosystem restoration and conservation efforts and for the control of alien species in all biomes (Khan et al. 2010), especially in the extreme arid regions. The fates of seed are affected both by abiotic and biotic factors as well in the sub-tropical dry and moist areas of located in Hindukush range of Pakistan. Some previous researches have figured out that mother shrub/trees of Dodonaea viscosa can produce hundreds to thousands of seeds every year, however, many seedlings hardly being found in the forest.

Instead, this species usually occurred as open areas, scree slope and forming a belt along the channels and rivers but quickly died in a large area (Personal Observations), only few ensued the least water demand for seeds and early seedlings and other recruitment limiting factors, such as soil surface, the site where seeds not only fall and germinate but also be stored. The desertification in the Dodonaea viscosa region has led soil contents, types and moisture changing and then it is shown that various Dodonaea viscosa forests primarily originated from seeds live on different soil substrates. However, there are few researches on the adaption mechanism of this species in the aspect of reproduction.

We assumed that the constantly changing and threatening environment was the key reason for sexual regeneration failure based on the fact of large production of vigorous seeds per year and temporal heterogeneity of seed rain. Besides, considering that plants exhibit a variety of behaviors in response to environmental stimuli (Karban 2008), this paper focused on the influence of two important natural factors (i.e., common soil substrate and shade regimes) on the seed germination and seedling survival in Dodonaea viscosa by manipulating their soil gradients or soil types and shade governance. Our goal was to figure out why Dodonaea viscosa failed to regenerate naturally by seeds to some extent, and then to provide some useful measures of management for Dodonaea viscosa shrub-land forest restoration and conservation in subtropical dry temperate areas of Hindukush range of Pakistan.

109

4.2 Materials and Methods

4.2.1. Study site

This experiment was conducted at the Botanical Garden Post Graduate Jahanzeb college Saidu Sharif Swat spanning between longitude 60표 22′ 40′′ north; latitude 20표37′08′′ east and altitude 13푚), which lies 20푘푚 east of River Swat. The Botanic Garden is located at altitude of 700푚 above sea level, which is suitable range for sowing seeds of Dodonaea viscosa because it’s populations at this altitude are stable in Hindukush and Himalayan regions of northern Pakistan (Fig. 1). Soils of the region are of both transported and residual type (Ali 2012) and mostly the foot hills and low lands have deep soil profile with large pore space due to large particle size (Khan and Bibi 2013). In forest floors the soils are quite rich in organic matter which is due to rapid humification and ranges from light brown to deep brown (Beg 1984).

The ground water is very close to the surface along the small tributaries and river Swat. The study area lying in the temperate zone in northern mountainous ranges and the weather is affected by the climatic factors, latitude, longitude and monsoon. In summer the area comes under the influence of monsoon and cyclonic current in the winter from Mediterranean Sea (Ahmed 2012).

The climate of area is generally continental with an average rainfall of 50 inches per year where, December to February is the winter rains that remain continuous to one or two weeks usually with snowfall. March to May is regarded as spring rains and after one dry month (June) the summer rains begin and end in September (Sher et al. 2009). The mean annual temperature is 8.2°C, with themean monthly maximum and minimum airtemperatures are 40°C and – 26°C, respectively. January is the coldest month (2°C to −2 °C) while July is the hottest month in which temperature rises to more than 32 °C in lowlands. The sunshine is 2075 hours per year accordingto 1980-2014 climate data.The mean annual temperature is 8.2°C, with themean monthly maximum and minimum air temperatures are 40°C and – 26°C, respectively. Thevegetation is characterized by broad leaved and coniferousforest (Ahmed et al. 2006). Q. baloot, Q. incana and O. ferruginea are common broad leaved whereas, P. roxburghii, P. wallichiana, C. deodara; A. pindrow and P. smithiana are common conifer species in the study area (Champion et al. 1965).

110

4.2.2. Experimental manipulation

Seed collection and selection

Seeds for germination trails were collected from four different districts of Malakand division at different altitudinal ranges in July and August in three consecutive years i.e. 2011, 2012 and 2013. Seeds were collected from at least 50 healthy Dodonaea viscosa plants with various diameters, height and crown cover from natural forest stands. The seeds were brought to the laboratory in paper bags and stored at approximately 25-30 °C temperature and 50-60% humidity until used. Prior to experimentation, seeds were sieved and exposed to sunlight for ventilation and later on sorted and damaged seeds were removed in order to remove undue effect in determining viability. On average 20 % of seeds were found with damaged embryo and were removed from the experiment. The selected seeds were treated with fungicide (Thiram, 80%) before the germination test following Avci and Kaya (2013).

For all field experiments, caryopses remained within the dispersal unit so as to best understand the germination ecology of these seeds in their natural state (Baskin and Baskin 2001). A subset of the sorted caryopses was placed on moist sand petri dishes with sealed para film (Baskin and Baskin 2001) and allowed to germinate in order to verify the percent of viable seeds. For all laboratory germination trails, 5 replicates of 10 caryopses were used. For these germination trials, petri dishes were placed under neutral shade for 10 days. Initial germination trails showed that Dodonaea viscosa seeds germinate rapidly, within 3-5 days, seeds were monitored for 3 weeks and no additional germination was noted beyond 10 days (Personal observation).

Soil samples collection and preparation

At the same time, soils were dug-out at a depth of 30 cm at three randomly selected areas within Dodonaea viscosa shrub-land following Khan et al. 2013 (Fig. 1; Table 1). At least 100 푘𝑔 soil were collected from individual site and pooled to form a homogenized composite sample following Khan et al. (2014). Before sowing seeds 90 perforated black polythene bags of 30 푐푚 × 199 푐푚2 size were filled with two types of soil.

The seed thus obtained were sowed out in 30 푐푚 × 199 푐푚2 perforated black polythene bags. Two types of substrate were used to fill the bags. The topsoil at the planting sites was mixed with organic manure (2/3 soil to 1/3 organic manure). The second substrate was

111

made from soil taken from the forest undergrowth at the research station in Botanical Garden Postgraduate Jahanzeb College Swat, which is the substrate generally used for young nursery plants. After the bags were filled, they were placed in open flat surface and in shade of trees.

Watering was conducted 2-3 times per week in order to keep the substrate moist. The seeds were sowed according to their polarity in the bags, which were arranged according to a randomized complete block design (RCBD), with four treatments per block and four repeats. Each experimental unit included 30 plant pots, each of which contained a seeds. The test was conducted from 9 July 2010 - 4 July 2011. The various treatments were as follows: B1S1: seeds sowed in field soil substrate containing organic manure; B1S2: seeds sowed in undergrowth soil substrate; B2S1: seeds sowed in soil substrate containing organic manure; B2S2: seeds sowed in undergrowth soil substrate and placed under shade of trees.

Observations

Every week for 2 years, the number of seeds geminated in each type of substrate was recorded, together with the number of leaves and morphology of the first leaf as soon as it developed. A final count was taken after three months for each type of substrate. Three months after sowing, the plants height was measured, based on a sample of 10 plants per treatment between the point of sprouting and the terminal bud, one year after planting out.

4.4.2.3 Statistics indexes and analysis

The average values for the different parameters studied were calculated. After the whole test, we calculated the germination percentage (G) and germination index (GI, an index of germination speed) according to Liu et al. (2011). We also calculated the root to shoot ratio, the absolute growth rate, the absolute elongation rate of shoot and the absolute elongation rate of root by the following equations: R/S = DWR / DWS, where R/S is the root to shoot ratio, DWR the dry weight of root biomass and DWS the dry weight of shoot biomass. AGRS = W / Dt, where AGRS is the absolute growth rate of seedlings and W the dry weight of seedlings in day t.

AERS = H / Dt, where AERS is the absolute elongation rate of shoot and H the height of seedlings in day t. AERR = ln (L / Dt) where AERR is the absolute elongation rate of root and L the length of main root in day t. SPSS 17.0 for The significance or insignificance of the

112

differences observed was evaluated by means of a variance analysis (ANOVA) and the Newman-Keuls test for the classification of averages.

Windows was used in one-way and two-way ANOVA with multiple comparisons, using least significant difference tests (LSD). The variance analysis model used was the 3-factor crossed model: a random factor (block) and two fixed factors: substrate and cutting type. The variance analysis was conducted with SAS software (version 9.1).

All the soils were chosen as three kinds of substrates for seed germination and seedling growth, and each was homogenized, sterilized (134°C) and sieved (5 mm). Soil chemical parameters are shown in Table 1. Moreover, 3 gradients of soil water content (i.e. 10%, 15% and 20%) were also set. The water contents were controlled twice a day (i.e. 9:00 AM and 9:00 PM) by weight ratio. To know the complete germination rate, we chose the pure water (i.e. distilled water, hereafter, CG) as an additional germination substrate. In the germination experiment, there were 5 replicates for each group and 30 seeds in each replicate.

The seeds were observed in a 4-hour-interval and the germinated ones were immediately taken into the other containers (3 seeds evenly in one container) with the same conditions for the subsequent growth experiment. The term of “germination” means that we observed a seed radical breaks its coat. Then, we counted the survived seedlings every day. After the 45thday from the treatment of the whole experiment, we randomly chose 5 seedlings from each group and recorded their heights, leaf area and taproot lengths. We also recorded the fresh and dry weights of above- and below-ground biomass by analytical balance (MS104S, Mettler- Toledo International Inc., USA).

In total, 1500 seeds (30seeds × 5 replicates × 3 soil substrates × 3 water gradients + 30seeds × 5 replicates × 1 pure water substrate) were used for the whole experiment. The containers for seed germination test and examining seedling growth were all the plastic bowls (150푚푚 in diameter and 100푚푚 in height) with full of different substrates placed in a room with a constant temperature of 25°C and a 16: 8 퐿: 퐷 light cycle with incandescent lumps.

113

4.3. RESULTS

4.3.1. Soil and stand characteristics

The distribution, geo-physical and ecological characteristics of the four sampling site for the seed collection of Dodonaea viscosa in Malakand division is shown in Table 4.1. The species is distributed between an attitudinal ranged from 500 푚 1800 푚 and generally show their dominance below Pinus zone forming an eco-tone. The range in the annual mean temperature was −3, – 38 °C, and the range in the mean precipitation surplus 3 to −145 푚푚 (Table 4.1). Subsoil texture and subsoil pH ranged from coarse sand to loam and from acid (4.6 ± 0.7) to neutral (7.0 ± 0.2) pH.

Water holding capacity (WHC) was ranged from 5.2 ± 0.6 to 11 ± 4.4 per 10 gram of soil while organic matter varied between 1.4 ± 0.4 to 2.3 ± 0.8% in the soil. The average lime content was in the ranged of 10.9 ± 1.2 to 14 ± 1.4% in the four sampling sites. Total nitrogen pools to soil depth 50 cm ranged from 2.0 ± 0.5 to 2.9 ± 0.5 g/kg-1, whereas Potassium (K2+) and Phosphorus (P2+) pools varied by factor 10-20 and about factor 50 for Phosphorus. Among the four sites C: N ratios in 0 − 15 푐푚 mineral soils were generally low (7 to 20) probably owing to the former agricultural use.

4.3.2. Seed germination

Significantly different impacts on Dodonaea viscosa seed germination was observed in the soil types with various textural, physiochemical compositions and the influence of shade (Table 4.2). Various type soils and soil under shade regimes from different sites of the study area have significant differences (푝 < 0.05) for germination percentage, leaf morphology, stem diameter, height, cover, root length and above ground biomass (Table 4.2) respectively in the four experimental trials using CRBD (Complete Randomized Block Design).

Among the Dodonaea viscosa seeds collected from open areas and uphill slope there was wide variation in seed characteristics and germination patterns. The germination percentage of the taxa ranged from 11.11% to 53.33%, and the lowest germination was observed in shade pots. Germination percentage was adversely influenced by shade, where germination percentages were below 11% of the total 90 seeds sown.

114

Taking the germination rate (G) and the germination index (GI)into account, there were differences in the terms of nutrient contents of the same substrate and substrates with the same water content (Fig. 4.2, 4.3).Table 4.2 shows that, to all the four soil types, the germination rates of Dodonaea viscosa went larger with the increasing soil contents in the original soil, and the curves became sharp from 11%for both the Humus mixed soil and the Garden soil that is, 43% to 52% respectively. In the comparisons among original soil (OS) with the same environmental factor, it was always showed that the germination rate of seeds was higher in the order of on Garden soil, Humus mixed soil, and soil placed in shade.

Table 4. 1: Percent seed germination and reproductive capacity (seed output) of Dodonaea viscose in four different soil types.

Soil type SG % RC (SO) % R.C M. Height G. Soil 48 53 39.5 53.33 21.06 29.4±1.74 H. Soil 38 42 39.5 42.22 16.67 28.2±2.43 O. Soil 62 69 39.5 68.88 27.20 31.5±2.97 S. Soil 10 11 39.5 11.11 4.38

Note: G. Soil = Garden Soil; H. Soil = Soil with Humus; O. Soil = original Soil; S. Soil = Shade Soil SG = Seed Germination; % = Percentage; RC (SO) = Reproductive Capacity (Seed Output); M. H = Mean Height

4.3.3 Seedling survival

As shown in Table 4.2 and4.3, different types of soil and treatment had diverse screening effects on seedling survival. Few seedling of Dodonaea viscosa were survived in original soil under shade stress but couldn’t pass through the ‘20-day-barrier’ at all. The reproductive capacity (RC) of seeds were highest for original soil (27.20%) followed by Garden (21.06%) and humus mixed soil. A somewhat, similar pattern was observed for seedling survival in the three soil types except original soil under shade. However, the seedling survival percentages were always higher in the condition of open pots exposed to light.

4.3.4 Seedling growth

Because the seedlings could not survive over 20 days in some conditions (Table 4.2), it was only necessary to test the growth indexes of seedlings on Garden, Humus mixed and original soil and with different composition of soils. A two way ANOVA was conducted to determine whether soil and water influenced the indexes of seedling growth (Table 4.3). The values of

115

each index served as the response variable, with soil type, water content, and their interaction serving as fixed effects.

Almost all the indexes showed that seedlings on RB significantly grew better than those on SM (Table 4.3). As to soil types, the P values of all the indexes except R/S were less than 0.01, which means that soil types have a significant effect on the seedling growth; however, as to the water content and the interaction between soil type and water content, only the P values of TL were less than 0.01. Besides, it was worth being mentioned that the taproot of seedlings at the 15% level could elongate longer and faster than those at the 20% level on the same substrates (Table 4.3).

Table 4. 2: Average (±SE) values of different parameters of Dodonaea viscosa in four different soil types.

Soil RL RC RFW RDW CC SFW SDW type Mean ±SE G. Soil 17±2.11 17.2±2.32 5.96±0.80 2.56±0.39 29.9±4.32 22.6±2.90 12.5±1.64 H. Soil 15.35±3.12 14.17±2.9 16.03±3.13 9.96±2.00 23.8±5.24 38.6±7.46 9.96±2.00 O. Soil 14.52±2.21 16.2±2.6 4.94±0.31 3.82±0.24 19.76±3.01 19.88±0.43 8.47±0.38 S. Soil

Note: RL = Root Length; RC = Root cover; RFW = Root Fresh Weight RDW = Root Dry Weight; CC = Canopy cover; SFW = Shoot Fresh Weight SDW = Shoot dry weight

Table 4. 3: Average (±SE) fresh and dry biomass of Dodonaea viscose in four different soil types.

Soil type NOL LFW LDW LL LW ASH

G. Soil 55.16±6.82 3.53±0.56 5.78±0.15 1.4±0.07 1.69±0.30 H. Soil 139.2±26.81 9.53±1.90 4.9±0.27 1.2±0.09 2.78±0.57 O. Soil 49.47±1.14 1.51±0.08 5.9±0.12 1.5±0.03 0.56±0.07 S. Soil

Note: RL = NOL = Number of Leaves LFW = Leave Fresh weight; LDW = leaves Dry weight LL = Leaves Length; LW = Leaves Width; Ash = Ash

116

4.4. Discussions

This study showed that the original soil was the most suitable soil substrate in the natural condition with sufficient amount of organic matter for Dodonaea viscosa seed germination and seedling growth. Original field soil always consists with lower salty content and higher nutrients than garden soil and Soil mixed with humus. Wu et al. (2007) have proved that lower concentration of NaCl solution has little effect on the total seed germination rate; and when the concentration reaches 90mmol/L, the rate will obviously decrease.

Besides, there were another two reasons for higher germination rate in Original field Soil (OFS): (1) because of its more organic constituents and higher compaction, Original field soil retention was greater than those of Soil of garden and Soil mixed with humus oils, for example, the seedlings growing on the original soil with 20% moisture cannot survive over 20 days, let alone with lower moisture in natural conditions ; (2) many arid plants and weeds, such as Amaranthus viridus, Euphorbia helioscopia etc., usually exude allele-chemicals into soil, inhibiting other plants’ seed germination or growth. Apart from these weeds some species are exotic that adversely affect the germination and growth and development of native species. One of these in the northern Pakistan is Parthenium hysterophorus that effect the germination of various native species. Zhang et al. (2005) figured out that low moisture, high salt content and allelopathy chemicals, which were the characteristics of S soil and SM soil, had negative effect on germination. Therefore, we argued that soil type in nature conditions could significantly influence the seed germination and seedling growth of Dodonaea viscosa, which is consistent with the results of two-way ANOVA.

Although higher soil moisture in original soil with high content of organic matter could effectively promote the seedling growth on the whole, less moisture availed the root elongation and AERR instead of the shoot height, leaf largeness or biomass accumulation of Dodonaea viscosa seedling. However, Peters (1957) found that at a soil suction (i.e. the relative vapor pressure of the soil moisture) of 1/3 bar and a bulk density of 1.25g·cm-2, corn roots elongated faster in a soil mixture with a gravimetric water content of 27% than in one with a water content of 8%, which was different with our results in Dodonaea viscosa. The actual reason for this may be the fact of different plant with different characteristics but still need further study.

117

All the values in the experimental group of ‘20%, RB’ showed that this condition could guarantee the higher seed germination rate and seedling survival percentage. This condition made most seeds germinate faster and longer and seedlings grow stronger, which may be the results of lower salty content, higher nutrients and less allelo-chemicals as above.

Many studies have focused on Dodonaea viscosa growth and its reliance on groundwater (Gries et al. 2003; Rüger et al. 2005; Liu et al. 2007; Fu et al. 2010), but still few on its seed and seedling stage. In natural conditions, though there are favorable conditions for Dodonaea viscosa seed germination during their long lives (Zhang et al. 2005), and the channeling has changed the natural hydrologic regimes and processes leading to the failure of regeneration of Dodonaea viscosa (Cao et al. 2009b).

However, we argued that the suitable habitat for its seed germination and seedling growth really existed in the nature and found that many branches of Swat and Dir River were not cemented except the main channel. The areas found being covered by many seed-originated seedlings are usually low and flat flood plains, which can store river water to keep high moisture and low salinity. But there were few seedlings in the other areas, such as in the sandy areas or in the main Dodonaea viscosa forests. That fact was just coincident with our experimental results.

Seed dispersal process was generally divided into two periods: Phase I, the movement of germinable seeds from the plant to a surface and Phase II, the secondary horizontal and vertical movements of seeds (Chambers and MacMahon 1994). As to the anemophilous Dodonaea viscosa seeds, wind is the key factor in Phase I. Considering the adhesive effect of abundant vegetation in the forests on seeds, we thought river bank and the other shallow sites with sufficient water were the most suitable ones for seed germination among all of the surfaces (e.g. river, soil and plants).

That may be the reason why Dodonaea viscosa forests are always called ‘riparian’. After quick germination in the water, seeds adhered to the flooded river bank as ‘belt’ where were suitable for seedling growth. Although Cao et al. (2009b) revealed that the supplemental flows were sufficient during the seed rain period of July 14th to August 28th; we still found that the water could not reach the Nature Reserve in that period every year. It was proved that the most steady water supplement was during the two periods of March to April and September to October every year, but unstable in the seed rain period (Chen, 2010), which

118

leaded to the discontinuous suitable habitats for seedlings in the preliminary key years and might be the main reason for the failure of Dodonaea viscosa reproduction.

Due to the huge economic effect of the hot and developing tourism of Dodonaea viscosa forest from September to October, the precious and rare river flow was usually supplied to create beautiful landscape in autumn and more impossible to be got from July to August in this extreme arid region, which accelerated the vegetation deterioration. It was reported that there was a ‘biennial fruiting variation’ phenomenon among many plants (Li et al. 1998; Borgardt and Nixon 2003; Hirayama et al. 2008).

Based on the assumption that the whole community gets the equal energy and nutrients every year, some thought that there might also be a biennial seed production in Dodonaea viscosa which severely affected the seed quality and then the subsequent growth stage. However, the high seed germination rate of our experiments in recent years proved this idea was not true (Zhang et al. 2005; Wu et al. 2007; Liu et al. 2011). Furthermore, light condition may be another limiting factor for Dodonaea viscosa regeneration (Zhang et al. 2005; Liu 2011). Therefore, we did a simple additional experiment about the effect of lightness on seed germination. It is obvious that light significantly lowered the seed vitality. In nature, the extremely high illumination intensity there made the ground temperature rise to over 70°C with rare vegetation, definitely leading to the death of weak seeds and seedlings.

Along the river bank, the conditions of vegetation were various and only the suitable shading condition provided by the abundant plants could pretend them from being burnt by sun. In a word, the water supply amount and time were the keys for the successful sexual regeneration of Dodonaea viscosa It is badly in need of the reasonable water adjustment and the foundation of river bank vegetation to protect the Dodonaea viscosa forest in Hindukush range Oasis.

Through our experiment connecting two essential environmental factors with seed and seedling periods, we provided some useful suggestions for artificial seedling cultivation of Dodonaea viscosa : 1) the harvested seeds should be germinated in the original field soil which can guarantee the highest germination rate; 2) the seeds should be kept in the water for 1-2 days before germination; 3) seed should not be sow in original, garden and original soil mixed with humus under shade because, the seeds remain dormant in shade regimes though there will be the availability of nutrients in enough quantity in the soils.

119

Conclusions

1. Present study reveal that Dodonaea viscosa is the most dominant shrub of foothills of Malakand division. Various administrative units of Malakand division of Swat, Dir, Shangla and Buner has diverse topographic edaphic and climatic condition but the favorable condition for growth of Dodonaea viscosa observed on calcareous soil with a loamy sand soil. 2. Dodonaea viscosa can grow at any aspect but it favour the area which receives full sunlight. Dodonaea viscosa was flushing well up to 1200 푚. 3. The different patterns of regeneration J-shaped, L-shaped, Bell-shaped is due anthropogenic activities, interspecific and intra specific competition, difference in the soil and climatic condition of the area also effect the regeneration of the specie. 4. The seedling density is higher in all community then sapling and sapling have higher density then mature plants which represent normal regeneration, but due to over grazing of animals, using for fuel and anthropogenic activities disturbed the natural regeneration of Dodonaea viscosa. 5. Study shows seedling and sapling density was in the communities group III which indicate the density of seedling and sapling increases up to a height of 1100m elevation and then decrease gradually. 6. A strong positive relationship found between seedling and sapling densities (푟 = 0.919 at 푃 > 0.001), sapling and organic matter sapling /organic matter (푟 = 0.457 at 푃 > 0.05). While a negative significant relationship was found between sapling and soil pH (푃 > 0.05), seedling and soil pH (0.001), also sapling and elevation (푃 > 0.001). It means that among the studied parameters anthropogenic disturbance, soil pH, organic matter and elevation mostly affect the regeneration of Dodona viscosa. 7. Both of soil and shade are the most essential factors for the establishment and growth of Dodonaea viscosa. 8. The original soil showed significantly different impacts on Dodonaea viscosa seed germination in the soil types with various textural, physiochemical compositions and the influence of shade. Germination percentage was adversely influenced by shade, where germination percentages were below 11% of the total 90 seeds sown. In the comparisons among original soil (OS) with the same environmental factor, it was

120

always showed that the germination rate of seeds was higher in the order of on Garden soil, Humus mixed soil, and soil placed in shade. It was concluded from the results that Dodonaea viscosa have variable response to different soil substrate and shade regimes and well adopted to original soil with open sunshine. 9. Due to high fuel wood quality manageable size of shoots easily chopped and easily

burned increase its fuel value.

121

Recommendations

1. Sustainable and timely chopping of shoots will be helpful in the maintenance and developing of Dodonaea forest. 2. The study area must be given a complete protection to restore the original vegetation, at least five year complete protection against any biotech interference for stabling the mature community of Dodonea. 3. The natural regenerating seedling and sapling are more beneficial which should be protected from over grazing of cattles, sheeps and goats to avoid tempting and grazed by animals. 4. The local inhabitants may be provided the propagules of fast growing fuel wood species like Robinia, Morus, Platinus, Alianthus and Populus to cope the urgent need of the local for their fuel consumption. 5. The Government should take interest in the area and should give subsidy for fossils fuels like Gas and oil as alternative source of fuel rather than Dodonaea viscosa. 6. Before apply means and methods to improve the forest, the current forest condition should preserve. 7. Awareness in local community should be required through environmental education and sustainable use management of such plant resources like Dodonaea viscosa which is multipurpose used i.e. ornamental, medicinal, construction of building, baskets, sticks, soil binder, as a hedge and insect repellent. 8. The study area may be converting to any conservatory status for such as national parks, game reserve, sanctuary or any other. 9. To achieve all this rights of the local people must be taken into consideration.

122

REFERENCES

123

AbdulrahmaAn (2013). An update phyto-pharmacological review on medicinal plant of Saudi Arabia - Dodonaea viscosa Linn.American Journal of research Communication, 2013, 1(12):519-531.

Abdulrahman K. Al-Asmari, Ketab E. Al Otaibi, Saud Al-Omani, M tanwir athar. An update phyto-pharmacological revies on medicinal plant of Saudi Arabia-Dodonaea viscosa Linn. American journal of research communication.

Ahma, M, Mahmood, Q, Gulzar, K, Akhtar, MS, Saleem, M, Qadir, MI (2012) Antihyperlipidemic and hepatoprotective activity of Dodonaea viscosa leaves extracts in alloxan-induced diabetic rabbits (Oryctolagus cuniculus). Pakistan veternity Journal 32:20 – 54.

Ahmad and Ahmad (2004) Dendrochronogical potential of a few tree species from Himalayan of Pakistan.J.pure and applied science, 11(2):65-67.

Ahmed M, Hussain T, Sheikh A H, Hussain S S, Siddiqui M F (2006) Phytosociology and structure of Himalayan forest from different climatic zone of Pakistan. Pakistan Journal of Botany, 38(2): 361-383. Ahmed M, Khan N, Wahab M, Hamza S, Siddiqui M F, Nazim K, Khan M U (2009) Vegetation structure of Olea ferruginea Royle. Forests of of Pakistan. Pakistan Journal of Botany, 41(6): 2683-2695. Ahmed M, Shaukat S S, Siddiqui F M (2011) A Multivariate analysis of the Vegetation of Cedrus deodara forests in Hindukush and Himalayan ranges of Pakistan:Evaluating the structure and dynamics. Turkish Journal of Botany 35, 419-438. Akram, M. (1972). The medicinal plants of Dir state Pakistan.pakistan journal of scientific Research, 24,107-113.

Ali A, Ahmad N, Habib G, Amjed M (2007) Study the vegetation capacity and physical characteristics of protected and open rangeland in District Dir (Lower). Sarhad Journal of Agriculture 23, 1. Ali kavgaci, orhan sevgi2, hüseyin bariş tecimen, osman yalçin yilmaz3, serdar carus4, türker dündar5 (2013) classification and ordination of pinus nigra dominated forests at alacam mountains (nw anatolia-turkey) eurasian journal of forest science (2013) 1(1): 38-50.

124

Ali Noor* And Surayya Khatoon (2013) Analysis of vegetation pattern and soil characterstics of Astore valley Gilgit- Baltistan. Pak. J. Bot., 45(5): 1663-1667, 2013. Ali, S. I. and M.Qaiser. (1986). Phyto-geographical analysis of phanerogrames of paksitan and Kashmir. Proc. Royal Society of Edinburgh, 89B:89-101.

Bace, R., Svoboda, M. & Janda, P. 2011. Density and height structure of seedlings in subalpine spruce forests of Central Europe: logs vs. stumps as a favourable substrate. Silva Fennica 45(5): 1065–1078.

Beg A.R (1979) Some more plant Communities and the future of Dry oak forest zone in swat valley . pak J Forestary,34:25:35

BEKELE T (2000) Plant Population Dynamics of Dodonaea angustifolia and Olea europaea ssp. cuspidata in Dry Afromontane Forests of Ethiopia. ACTA UNIVERSITATIS UPSALIENSIS UPPSALA 2000.

Bingham F T (1949) Soil Tests for Phosphate. California: Agriculture, 3(8): 11- 14.

Black C A (1965) Method of Soil Analysis, Part 2, Chemical and Microbiological Properties, American Society of Agronomy, Inc, Publisher, Madison, Wisconsin USA.

Borgardt, S.J. and Nixon, K.C. (2003) A comparative flower and fruit anatomical study of Quercus acutissima, a biennial-fruiting oak from the Cerris group (Fagaceae). American Journal of Botany, 90: 1567–1584.

Bouyoucos G J (1936) Direction for making mechanical analysis of soil by the hydrometer method. Soil Science, 42(3): 225-229.

Cao, D.C., Li, J.W., Chen, W.Q. and Li, J.Q. (2009b)Temporal heterogeneity of Populus euphratica seed rain in Ejina Oasis, China. Eurasian Journal of Forest Research, 12: 1–84.

Cao, D.C., Li, J.W., Chen, W.Q., Peng, C. and Li, J.Q. (2009a) Development and growth of root suckers of Populus euphratica in different forest gaps in Ejina Oasis. Acta Ecologica Sinica, 29: 1954–1961.

125

Chambers, J.C. and MacMahon, J.A. (1994) A day in the life of a seed: movements and fates of seeds and their implications for natural and managed systems. Annual Review of Ecology and Systematics, 25: 263–292.

Chen, W.Q. (2010) Study on the landscape pattern, dynamic and driving force in Ejina Oasis. Peng et al. Eurasian J. For. Res. 15-1(2012).

Chesson, P., Gebauer, R.L.E., Schwinning, S., Huntly, N., Wiegand K., Ernest M.S.K., Sher A., Novoplansky, A. and Weltzin J.F. (2004) Resource pulses, species interactions, and diversity maintenance in arid and semi-arid environments. Oecologia, 141: 236– 253.

Christopher R. Keyes2 and Douglas A. Maguire (2005) Positive Seedling-Shrub Relationships in Natural Regeneration of Ponderosa Pine. USDA Forest Service Gen. Tech. Rep. PSW-GTR-198. 2005.

Colombo S.j (2005) “A thin Green Line” A Symposium on the State-of-the-Art

Cox G W (1990) Laboratory Manual of General Ecology, 6th edition, William C. Brown Publishers, Dubuque, 1-272.

Cribb AB and CribbJW,Wild medicine in Australia,Collins and Sydney,1981,p228.

Cribb, AB, Cribb, JW (1981) Wild medicinal in Australia. Collins; Sydney in Reforestation. Applied Research and Development • Ontario Ministry Of Natural Resources Paper 160.

E. F Gilman (1999) University of Florida, 1999, Fact sheet FPS-181.

El-Sheikh, A. M. and Yousef, M. M. (1981). Halophytic and xerophytic vegetation near Al-Kharg springs. Journal of Colloid Science, 12(1), 5-12. Frelich L.E., and Lorimer C.G. 1985. Current and predicted long-term effects of deer browsing in hemlock forests in Michigan, USA. Biological Conservation 34:99-120.

Fu, A.H., Chen, Y.N. and Li, W.H. (2010) Analysis on the change of water potential of Populus euphratica Oliv. and P. Russkii Jabl under different irrigation volumes in temperate desert zone. Chinese Science Bulletin, 55 (10): 965–972.

126

Gauch H G Jr, Whittaker R H (1981) Hierarchical classification of community data. Journal of Ecology 69, 135-152.

Getie M G Rietz R and Neubert R H H Distribution of quercetin, kaempferol and isorhamnetin in some Ethiopian medicinal plants used for the treatment of dermatological disorders. EthiopianPharmacy .J, 2000, Vol 18, p25-34.

Greig-smith P (1983) Quantitative Plant Ecology. 3rd ed. Oxford: Blackwell scientific, pp: 1- 359. Gries, D., Zeng, F., Foetzki, A., Arndt, S.K., Bruelheide, H., Thomas, F.M., Zhang, X. and Runge, M. (2003) Growth and water relations of Tamarix ramosissima and Populus euphratica on Taklamakan desert dunes in relation to depth to a permanent water table. Plant, Cell and Environment, 26: 725–736.

Hanes T.L. 1971 succession after fire in the chaparral of southern California. Eceol. Monogr. 41: 27-52.

Harding, D.E., and Ross, D,J. 1964. Some factors in low tempaerature stiorage influencing the mineralizable nitrogen of soils. Journal of the Science of food and agriculture, 15: 829-834.

Harper, J.L. (1977) The Population Biology of Plants. Academic Press, New York.

Hill M O, Gauch H G (1980) Detrended Correspondence Analysis: An Improved Ordination Technique. Vegetatio 42, 47–58. Hirayama, D., Nanami, S., Itoh, A. and Yamakura, T. (2008) Individual resource allocation to vegetative growth and reproduction in subgenus Cyclobalabopsis (Quercus, Fagaceae) trees. Ecological Research, 23: 451–458. Hukin, D., Cochard, H., Dreyer, E., Thiec, D.L. and Boget-Triboulot, M.B. (2005) Cavitation

Hussain S S (1984) Pakistan Manual of Plant Ecology. National Book Foundation, Islamabad, pp: 1-242.

Jin-Tun Zhang1 and Feng Zhang (2011) Ecological relations between forest communities and environmental variables in the Lishan Mountain Nature Reserve, China African Journal of Agricultural Research Vol. 6(2), pp. 248-259, 18 January.

127

John E. Keeley and Sterling C. Kelley (1981) Post fire regeneration of southern California Chaparral. American journal of botany 68(4):524-530. 1981.

Karban R. (2008) Plant behavior and communication. Ecological Letters, 11: 727–739.

Keeley j.e. (1977) fire dependent reproductive strategies in arctostaphylos and canthus. In H.A. Mooney and C.E. Conrad [eds], proceeding of the symposium of the environmental consequences of fire and fuel management in Mediterranean ecosystem, p. 391-396.

Khan N (2012) A community analysis of Quercus baloot Girafft forest of District Dir Upper, Pakistan. African Journal of Plant Science, 5(12): 21-31.

Khan N, Ahmed M, Shaukat S S, Wahab M, Siddiqui M F (2011) Structure, diversity, and regeneration potential of Monotheca buxifolia (Falc.) A. DC. Dominated forests of Lower Dir District, Pakistan. Frontier of Agriculture China, 5(1): 106-121.

Khan N, Bibi S (2013) Phyto-ecological and ethno-botanical study of intercropping weeds in Research, 2(2): 5 – 17.

Khan N, Shaukat S S, Ahmed M, Siddiqui M F (2013) Vegetation-environment relationshipsin the forests of Chitral district Hindukush range of Pakistan. Journal of ForestryResearch, 24(2): 205−216.

Kürschner H. (2004) Phytosociological studies in the Alashan Gobi–A contribution to the flora and vegetation of Inner Mongolia (NW China). Phytocoenologia, 34: 169–224.

Li, R., Werger, M.J.A., During, H.J. and Zhong, Z.C. (1998) Biennial variation in production of new shoots in groves of the giant bamboo Phyllostachys pubescens in Sichuan, China. Plant Ecology, 135: 103–112.

Lindholm, C. (1979). Contemporary Politics in a triable society: Swat District, NWFP, Pakistan. Asian Survy, 19(5), 485-505.

Little EL and Skolmen RG (1989) Common forest trees of Hawaii(native and introduced).Agriculture hand book .679.U.S. Department of Agriculture, Washington, 1989,DC.321P TESFAYE BEKELE (2000).

128

Little EL and Skolmen RG, Common forest trees of Hawaii (native and introduced). Agriculture hand book .679.U.S. Department of Agriculture, Washington DC, 1989, p321.

Liu, P.X., Peng, J.F. and Chen, F.H. (2007) Hydrological response of Populus euphratica Olve. Radial growth in Ejinaa Banner, Inner Mongolia. Journal of Integrative Plant Biology, 49: 150–156.

Liu, Q.W. (2011) The effects of factors on seed germination and seedling growth of Populus euphratica in Ejina Oasis. Dissertation, Beijing Forestry University, pp. 36–39.

Liu, Q.W., Hao, P., Li, J.W., Tan, Z.G., Amugulang and Luo, J.Y. (2011) Effect of stand age and individual growth on seed germination of Populus euphratica. Forestry Studies in China, 13: 183–188.

M.Sandhya Rani*, Prof.Rao.S.Pippalla*, Prof Krishna Mohan (2009) Dodonaea viscosa linn.-an overview. Jprhc | july 2009 | vol. 1 | no.1 | 97-112.

Martins D, Odd E, Eli F, Jonas EL, Erik A (1999). Beech forest communities in the Nordic countries- A multivariate analysis. – Plant Ecol. 140: 203-220.

McCormick L.H., J.W. Groninger, K.A. Penrod, and T.E. Ristau. 1993. Deer exclusion effects on understory development following partial cutting in a Pennsylvania Oak Stand. In Proceedings of the 9th Central Hardwood Forest Conference. USDA For. Serv. Gen. Tec. Rep. NC-161.

McCune B, M J Mefford (2005) Multivariate Analysis of Ecological Data (PCORD Version 5.10.MjMSoftwareGleneden Beach, Oregon: United State of America. Mishra, Y. (1972). The hindu sahis of Afghanistan and the Punjab, A.D. 865 – 1026 ( Pattna; sm. Sushila Devi, 89).

Moinuddin Ahmed, kanwal nazim, muhammad faheem siddiqui, muhammad wahab, nasrullah khan, muhammad uzair khan and syed sadaruddin hussain (2010) community description of deodar forests from himalayan range of pakistan pak. J. Bot., 42(5): 3091-3102, 2010.

129

Monda, Y., Miki, N. and Yoshikawa, K. (2008) Stand structure and regeneration of Populus euphratica forest in the lower reaches of the Heihe River, NW China. Landscape and Ecological Engineering, 4: 115–124.

N.K. UDAYA PRAKASH *, C.R. SELVI 2. V. SASIKALA2., S. DHANALAKSHMI AND S. BHUVANESWARI UDAYA PRAKASH2 (2012) PHYTOCHEMISTRY AND BIO-EFFICACY OF A WEED, DODONAEAVISCOSA. International Journal of Pharmacy and Pharmaceutical Science. Vol 4, Issue 2, 2012

Nasrullah Khan (2012) A community analysis of Quercus baloot Griff, forest District Dir, Upper Pakistan. African Journal of Plant Science Vol. 6(1), pp. 21-31, 6 January.

Nasrullah KHAN , Moinuddin AHMED, Syed Shahid SHAUKAT, Muhammad WAHAB, Muhammad Faheem SIDDIQUI Structure, diversity, and regeneration potential of Monotheca buxifolia (Falc.) A. DC. Dominated forests of Lower Dir District, Pakistan. Front. Agric. China 2011, 5(1): 106–121

Nasrullah Khan • Syed Shahid Shaukat • Moinuddin Ahmed (2013) Vegetation-environment relationships in the forests of Chitral district Hindukush range of Pakistan Journal of Forestry Research 24(2): 205−216.

Natalie L. Cleavitt, Timothy J. Fahey, and John J. Battles Regeneration ecology of sugar maple (Acer saccharum): seedling survival in relation to nutrition, site factors, and damage by insects and pathogens.Can. J. For. Res. 41: 235–244 (2011).

NCRN (2012) National Capital Region Network, Inventory & Monitoring program: http://science.nature.nps.gov/im/units/ncrn/index.cfm September 2012.

Oliver, C. D.; Stephens, E. P. Reconstruction of a mixed species forest in Central New England. Ecology 1977, 58, 562-572.

Peters, D.B. (1957) Water uptake of corn roots as influenced by soil moisture content and soil moisture tension. Soil Science Society of America Journal, 21: 481–484.

130

Pizada, AJ, Shaikh, W, Usmanghaani, K, Mohiuddin, E (2010) Antifungal activity of Dodonaea viscosa Jacq extract on pathogenic fungi isolated from super ficial skin infection.Pak.J.pharm. Sci 23(3): 337-340

Plant Population Dynamics of Dodonaea angustifolia and Olea europaea ssp. cuspidata in Dry Afromontane Forests of Ethiopia. ACTA UNIVERSITATIS UPSALIENSIS UPPSALA 2000.

Rahman H U, Nabi G, Ali I, Tahir T, Ahmed M (2012) Growth and yield of Kinnow (Citrus reticulata Blanco) and soil physical properties as affected by orchard floor management practices in Punjab, Pakistan. Soil and Environment 31, 163-170.

Rajma anick, V, Rajasekaran, A, Anandarajagopal, K, Sidharan, D, Selvakumar, K, Rathinaraj, BS (2010) Anti- diarrheal activity of Dodonaea viscosa root extracts. Int. Jpharm. Bio sci 1(4): 182-185.

Rey, P.J. and Alcántara, J.M. (2000) Recruitment dynamics of a fleshy-fruited plant (Olea europaea): connecting patterns of seed dispersal to seedling establishment. Journal of Ecology, 88: 622–633.

Rojas AS, Cruz H, Ponce-Monter, and Mata R Smooth muscle relaxing compounds from Dodonaea viscosa Planta medica. 1996, Vol 62, p154-159.

Rüger, N., Schlüter, M. and Matthies, M. (2005) A fuzzy habitat suitability index for Populus euphratica in the Northern Amudarya delta (Uzbekistan). Ecological Modelling, 184: 313–328.

Shariatullah (2013) AN Ecological assessment of Justicia adhatoda L. in Malakand Division. M.Phil. Thesis Department of Botany University of Malakand Pakistan.

Shinwari, Z.K and S.S. Gilani (2003). Sustainable harvest of Medicinal plants at Bulshbar Nullah, Astore, Northern Pakistan. Journal of Ethnopharmocology 84, 289-298.

SiddiquiA A Chemical and pharmacological evaluation of Dodonaea viscosa Asian J. ofChemistry, 1998, Vol 10, p14-16.

131

Singh, A.P. 1986. Seasonal fluctuation of organic matter with relation to moisture retention characteristics and availability of water in salt effected soil (India) Acta Botanica Indica, 14: 73-76.

Soltan-pur P N, Schwab A P (1977) Use of ammonium bi-carbonate-DTPA soil test to evaluate element availability and toxicity. Common Soil Plant Analysis 16, 323-338.

Stewart, G. H. (1986) Population dynamics of a montane Conifer forest, Western Cascade Range, Oregon, USA. Ecology, 60, 937-945.

Sultani-I-Rome, (1999). Merger of the swat state with Pakistan causes and effects, MARC Ocacasional papers, (14 April 1999), the University of Geneva.

Sultani-I-Rome, (2005). Forestry in the principal state of swat and kalam (North West Pakistan): Ahistorical prospective on norms and practices, in: Individual project working paper, Vol.6. Swat. Pakistan.

Sultani-I-Rome, (2006). Administrative system of the princely state of swat. Journal of the Research Society of Pakistan. Lahore 43(2).

Tang CQ, Ohsawa M (1997). Zonal transition of evergreen, deciduous, and coniferous forests along the altitudinal gradient on a humid subtropical mountain, Mt. Emei, Sichuan, China. - Plant Ecol., 133(1): 63-78.

Tavili A, Jafari M (2009) Interrelations between plants and environmental variables. International Journal of Environmental Research 3, 239-246. Taylor, A. H.; Halpern, C.B. (1991) the structure and dynamics of Abies magnifica forests in the Southern Cascade Range, USA. J. Vegetation Sci. 1991, 2: 189-200.

Tesfaye Bekele 2000.Plant Population Dynamics of Dodonaea angustifolia and Olea europaea ssp.cuspidata in Dry Afromontane Forests of Ethiopia. Acta Universitatis Upsaliensis. Comprehensive Summaries of Uppsala Dissertations from the Faculty of Science and Technology 559. 47 pp. Uppsala. ISBN 91-554-4781-3.

Tilghman N.G. 1989. Impacts of white-tailed deer on forest regeneration in northwestern Pennsylvania. Journal of Wildlife Management 53:424-532.

132

Titshall L W, Connor O T G, Morris C D (2000) Effect of long-term exclusion of fire and herbivory on the soils and vegetation of sour grassland. African Journal of Range and Forage Science 17, 70–80. Vulnerability in roots and shoots: dose Populus euphratica Olive., a poplar from arid areas of Central Asia, differ from other poplar species? Journal of Experimental Botany, 56: 2003–2010.

Wagner H, In Handbook of Medicinal plants(Ed, Yaniv, Z. B., U.) (2005) Haworth press, New York.

Wagner H, Ludwig C, Grotjahn L, Khan MSY, Biologically Active Saponins from Dodonaea-Viscosa. Phytochemistry (Oxford) (1987), Vol 26, p697-702.

Wagner, H, Ludwig, C, Grotjahn, L, Khan, MSY (1987) biologically active saponins from Dodonaea viscosa. Phytochemistry 26(3): 697 – 701.

Wahab et al.2008 population dynamics and denderochrionlogical potential of pine tree species of district Dir Pakistan. Ph.D dissertation Department of Botany Fedra; University Karachi Pakistan.

Walkley A (1947) A Critical Examination of a Rapid Method for Determination of Organic Carbon in Soils - Effect of Variations in Digestion Conditions and of Inorganic Soil Constituents. Soil Sciences 63, 251-257.

West D C, Shugart H H, Rammey J W (1981) Population structure of forests over largearea. Forest Science 27, 701-710.

West et al. (1972) Land policy in Buganda, Cambridge University press, Cambridge UK.

Wu, F.P., Li, J.Q., Li, J.W., Wang, Y. and Yang, X.Q. (2007) Effect of salinity and different water content levels of three substratum on the seed germination of Populus euphratica Oliv. Journal of Modern Forestry, 1: 44–48.

Zare S, Jafari M, Tavili A, Abbasi H, Rostampour M (2011) Relationship between environmental factors and plant distribution in arid and semiarid area (Case study: Shahryar Rangelands, Iran). American-Eurasian Journal of Agriculture.

133

Zare S, Jafari M, Tavili A, Abbasi H, Rostampour M (2011) Relationship between environmental factors and plant distribution in arid and semiarid area (Case study: Shahryar Rangelands, Iran). American-Eurasian Journal of Agriculture and Environmental Sciences 10, 97-105. Zhang, Y.B., Li, J.W., Zhang, H., Li, J.Q., Wu, F.P. and Zhao, J. (2005) Analysis on the factors cause the failure of Populus euphratica sexual regeneration in nature. Science Technology and Engineering, 5: 467–472.

134