THE EFFECT of INVASIVE SPECIES Lantana Camara on SOIL

Home , Ol Donyo Sabuk

THE EFFECT OF INVASIVE SPECIES Lantana camara ON SOIL

CHEMISTRY AT OL- DONYO SABUK NATIONAL PARK, KENYA.

MARYSTELLA NANGO’NI WEKHANYA (B.Ed Science)

Reg No: I56/CE/22253/2010

A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE

REQUIREMENTS FOR AWARD OF THE DEGREE OF MASTERS OF

SCIENCE (PLANT ECOLOGY) IN THE SCHOOL OF PURE AND APPLIED

SCIENCES KENYATTA UNIVERSITY

OCTOBER, 2016

DECLARATION

I declare that the work presented in this thesis is my own original work and has not been presented for award of a degree in any other University or any other award.

Marystella Nango’ni Wekhanya (B.Ed Science)

I56/CE/22253/2010

Department of Plant Sciences, Kenyatta University

Signature …………………………. Date ……………………………

We confirm that the work reported in this thesis was carried out by the candidate under our supervision as the supervisors:

Prof. Paul Kamau Mbugua

Department of Plant Sciences, Kenyatta University

Signature …………………………. Date ……………………………

Prof. John Kiogora Mworia

Department of Biological Sciences, Meru University of Science and Technology

Signature …………………………. Date ……………………………

iii

DEDICATION

This work is dedicated to the All Mighty God, my husband Cyrus Gakuo and children

Tony, Chris and Arthur who have been my inspiration.

ACKNOWLEDGEMENTS

Special thanks go to Kenya Wildlife Service Deputy Director, Biodiversity Research and Monitoring Department Mr Samuel M. Kasiki for granting me permission to conduct my research project at Ol-Donyo Sabuk National Park. The support and encouragement I received from Senior Warden, Oldonyo Sabuk National Park Mr.

Thomas Mailu. My sincere gratitude goes to my supervisors Prof. Paul Kamau

Mbugua and Prof. John Kiogora Mworia for their valuable time, comments, guidance and encouragement throughout this study. I greatly appreciate their technical input and intellectual support in carrying out detailed corrections while writing the research proposal and thesis. I acknowledge many individuals who helped me in numerous ways throughout my study period without them this work would not have been possible. I am particularly indebted to Mr. Wilson Korir- Assistant Director, Southern

Conservation Area of Kenya Forest Service for his invaluable support, Mr. Stephen

Gichobi- Chief Technician, Department Plant Sciences Kenyatta University, Mr. Philip

Mwang’ombe- Research Scientist, Ol-Donyo Sabuk National Park, Mr. Josephat

Wambua, Kenya Wildlife Service, for his moral support and encouragement to pursue the invasive plant species project. Finally, I am thankful to my many friends, Morris

Muthini, Olive Sande, Everlyne Kakai and Lydia Mugao whom we have worked together. My gratitude also goes to my family: my husband, Cyrus Gakuo and my children Tony Kinyanjui, Arthur Njonge and Chris Njuguna, for having to endure my long time of absence. Their love and support has made me accomplish this work. I will always be indebted to you.

TABLE OF CONTENTS DECLARATION ...... ii DEDICATION ...... iii ACKNOWLEDGEMENTS ...... iv LIST OF TABLE ...... vii LIST OF FIGURES ...... viii LIST OF PLATES ...... ix ABSTRACT ...... x

CHAPTER ONE: INTRODUCTION...... 1 1.1 Background Information ...... 1 1.2 Problem Statement and Justification ...... 4 1.3 Study Hypothesis ...... 5 1.4 Research Questions ...... 6 1.5 Objectives ...... 6 1.5.1 General Objective ...... 6 1.5.2 Specific Objectives ...... 6 1.6 Scope of the Study ...... 7 1.7 Significance of the Study...... 7

CHAPTER TWO: LITERATURE REVIEW ...... 8 2.1Definition of Invasive Species ...... 8 2.1.1Mechanisms of Invasion by Invasive Species ...... 9 2.2.1.1 Ecosystem-based Mechanisms ...... 10 2.2.1.2 Species-Based Mechanisms...... 11 2.2.2 Impacts of Invasive Species on Communities and Ecosystem ...... 14 2.2.2.1 Disturbance of Regime, Climate and Physical Habitat...... 15 2.2.2.2 Species Extinctions and Community Structure ...... 16 2.2.2.3 Change in Energy, Nutrient and Water Cycling ...... 16 2.3 Impacts of Invasive Species...... 17 2.4 Impacts of Invasive Plant Species on Soil Development ...... 18 2.5 General description and Ecology of Lantana camara ...... 20 2.5.1 Taxonomy ...... 20 2.5.2 Description ...... 20 vi

2.5.1 Habitat Description ...... 24 2.5.2 Reproduction ...... 26 2.6 The Effects of Lantana camara on Soil Chemistry...... 27

CHAPTER THREE: STUDY AREA AND METHODOLOGY ...... 30 3.1 Study Site – Ol Donyo Sabuk National Park...... 30 3.1.1 Topography...... 31 3.1.2 Soils ...... 31 3.1.3. Wildlife Communities ...... 32 3.1.3.1Animals...... 32 3.1.3.2Vegetation...... 32 3.1.3 Land Use ...... 34 3.2. Materials and Methods ...... 34 3.3 Laboratory Analysis of the Soil Samples ...... 38 3.4 Data Analysis...... 41

CHAPTER FOUR: RESULTS ...... 43 4.1 Soil Characteristics During the Wet Season ...... 43 4.2 Soil Characteristics During the Dry Season ...... 49 4.3 Effect of Seasons and Invasion on Soil Nutrients and pH ...... 55 4.4 Consistency of pH and soil nutrients in the Lantana invaded areas...... 64 4.5 Soil Texture Characteristics...... 69

CHAPTER FIVE: DISCUSSION, CONCLUSIONS AND RECOMMENDATIONS ...... 71 5.1 Discussion ...... 71 5.2 Conclusion ...... 77 5.3 Recommendations ...... 78

REFERENCES ...... 79 APPENDICES...... 93 APPENDIX I: SOIL TYPES FOR THE FIVE STUDY SITES ...... 93

vii

LIST OF TABLE

Table 3.1: Study Sites and Some of the Native Plant Species...... 33

viii

LIST OF FIGURES

Figure 2.1: Lantana camara showing fruits and inflorescence. (IUCN, 2004) ...... 22 Figure 3.1: Ol Donyo Sabuk National Park; Vegetation Distribution (KWS, 1999) ... 30 Figure 3.2: Map of Ol-Donyo Sabuk National Park Control Blocks for L. camara (KWS, 1999)…………………………………………………………..………….…..36 Figure 3.3: Divided quadrant for random composite sampling ...... 37 Figure 3.4: Soil Textural Triangle ...... 41 Figure 4.1: Comparison of pH and soil nutrients during wet season ...... 47 Figure 4.2: Comparison of pH and soil nutrients during dry season ...... 52 Figure 4.3: Interactions between Lantana invaded, non-invaded, wet and dry seasons and sites ...... 59 Figure 4.4: Consistency of nutrients levels in invaded and non-invaded areas...... 66 Figure 4.5: Soil types percentage values for L. camara in invaded and non-invaded areas ...... 70

LIST OF PLATES Plate 3.1: Soil sampling using an auger...... 37 Plate 3.2: Collected soil samples (yellow labels for dry season and blue for wet season) ...... 38

ABSTRACT

Invasive species are a major ecological and management concern in natural ecosystems and pose a threat to many of Kenya’s protected areas. Invasive plant species compete and hybridize with native species often to the disadvantage of the native species. Invasion can lead to the phasing out of native species and loss of ecosystem services such as water filtration, soil stabilization, and pest control. They may also result in reduction of wildlife forage or death of animals when they feed on poisonous invasive plant species. Invasion of native plant habitats by invasive plants can drastically change soil chemical properties such as pH, mineral composition and mineral levels. The aim of this study was to establish whether the invasive plant Lantana camara L. alters the soil chemical properties at Ol-Donyo Sabuk National Park. The key objective was to evaluate the soil nutrient composition in areas invaded by Lantana camara L. and how these differ from areas without Lantana camara. Five study sites were selected by purposeful sampling out of the existing 10 blocks. Soil samples were collected randomly from L. camara invaded areas and similarly from adjacent areas free from L. camara. The soil samples were collected during the wet season (November-December, 2014) and during the dry season (January-March, 2015). The soil samples were analysed for the following nutrients and parameters: pH, potassium (K), calcium (Ca) magnesium (Mg), total nitrogen (N), phosphorous (P), total organic carbon, manganese (Mn), copper (Cu), iron (Fe), zinc (Zn), sodium (Na) and texture. The data obtained was analysed using Two-way ANOVA test to determine difference in nutrients composition in Lantana invaded and non-invaded areas. Three-way ANOVA test was also used to determine the interactions between wet and dry season, invaded and non-invaded areas and the study sites. A post-ANOVA test, Tukey's Honest Significant Difference was done to separate the means. The analysed results were presented in graphs and descriptive tables. Results from the study indicated Lantana invaded areas had an increase in pH value (invaded 6.88, non-invaded 6.30), P (invaded 20.76, non-invaded 18.81), N (invaded 0.36, non-invaded 0.18), Mn (invaded 1.03, non-invaded 0.84), Fe (invaded 24.97, non-invaded 17.72) and total organic carbon (invaded 1.73, non-invaded 1.72) compared to the patches with native plant species. During both wet and dry seasons Lantana invaded areas had an increase in pH value (dry-invaded 6.88, non-invaded 6.30; wet-invaded 6.48, non-invaded 6.30), P (dry-invaded 20.76, non-invaded 18.81; wet-invaded-21.11, non-invaded 18.81)) and Mn (dry-invaded 1.03, non-invaded 0.84; wet-invaded 0.94, non-invaded 0.77) compared to non-invaded areas. Most nutrient levels were found to be higher during the wet season compared to the dry season which could be attributed to high pH and accelerated Lantana biomass litter decomposition. High pH also makes P to be more available to plants that is why P was high in the L. camara invaded areas. The soil texture was almost the same in Lantana invaded and non-invaded areas. This study has revealed that Lantana camara remarkably changes the soil nutrient levels leading to changes in soil chemistry of invaded areas. This is in a bid to suit its survival to the detriment of the native plants. This study is hence vital for designing an effective eradication and preventive strategy of Lantana in Ol-Donyo Sabuk National Park and in other protected ecological habitats in Kenya.

CHAPTER ONE

INTRODUCTION

1.1 Background Information

Invasive plant species are among the world's greatest threats to biodiversity and native species in protected areas (Lwando, 2009). Invasive species may change the community structure through competition which is categorised into two; exploitation competition whereby there is indirect interactions in the use of resource and interference competition whereby invasive plants produce allelo-chemicals that affect the native plants (Callaway and Ridenour, 2004).

Once invasive plants become established, they change the soil chemistry and shift nutrient cycling in an ecosystem. This can have impacts on plant composition, diversity, and succession within a community and also cycling of critical elements like carbon and nitrogen on a larger, potentially even global scale (Rout et al., 2013). Both native and invasive plants form some relationships with bacteria and fungi in the soil that facilitate the extraction and conversion of elements to biologically usable forms

(Batten et al., 2006). Yet it is not well understood whether the changes in soil bio- geochemistry are due to an advantage that invasive plants acquire from interacting with their micro-biome or when alien species invade and colonise the area with other micro-organisms. Many plant species are host to a whole group of micro-organisms that not only live in plant cells, but also in the soil surrounding the plants' roots (Rout and Callaway, 2012). These micro-organisms often form close mutual relationships with their plant hosts. Some convert atmospheric nitrogen into usable forms that are then exchanged for carbon from the plant. Nitrogen in form of nitrates and nitrites are frequently limiting in soils, yet many invaded ecosystems have shown to have more 2

carbon and nitrogen in plant tissues and soils compared with systems dominated by native plants (Mack et al., 2000). Could the invasive species be responsible for the changes in soil nutrient concentrations? Could the nutrient changes be a strategy for enabling the continued growth and colonisation of the invader species?

Tourism is one of the key pillars of the Kenyan economy. It generates foreign income through wildlife-based activities like scenic views, Park entry fees, Park lodges and camp sites amongst others (Ngoru et al., 2007). However, wildlife proliferation and stability of this important sector is threatened by many factors including poaching, climate change, human encroachment and now more recently, invasive species.

Invasive species pose a serious threat to biodiversity and are now considered second to habitat destruction in driving global biodiversity loss (Batten et al., 2006).

Lantana camara L. is one of the most invasive plants and has been ranked as the highest impacting invasive species (Batianoff and Butler, 2003). It is top among the world’s 100 worst invasive alien species (GISP, 2003), because it possess great potential to escape cultivation and have deleterious effect on species richness (Islam,

2001). Lantana camara L. is a major weed in over 60 countries and is one of the 10 worst weeds worldwide (Sharma et al., 2005). In Australia, Lantana is a weed which has evoked a national outcry, having invaded at least 4 million hectares of predominantly coastal and sub-coastal ranges of eastern Australia (Parsons and

Cuthbert, 2001). In Kenya protected areas like Nairobi National Park, Oldonyo-Sabuk

National Park, Nakuru National Park, Lantana is threatening the habitats and biodiversity. 3

However, the impact of invasive species in protected areas, mostly National Parks and reserves in Kenya is currently poorly understood and the magnitude of the problem is not well appreciated (Simba et al, 2013). Invasive plant species have many impacts on plant communities through their direct and indirect effects on soil chemistry

(Ehrenfeld, 2003). As plant community composition changes, preferred wildlife habitats are destroyed hence destabilizing such protected areas. Ol-Donyo Sabuk is a protected National Park due to its unique mountain and fragile ecological diversity.

The Park is a home to several endemic plant and animal species which are facing pressure from the impacts of invasive species that include pricky berry (L. camara L.), kei apple (Dovyalis caffra L.), Sodom apple (Solanum incanum L.), castor oil (Ricinus communis L.) and Jimsonweed (Datura stramonium L.) among others (Wambua,

2012).

An invasive plant species like Lantana camara has invaded disturbed forest land and neglected pasture in much of its naturalized range. In some areas, competition by the shrub results in a reduction of biodiversity (Kumar and Rohatgi, 1999). Despite the establishment of a number of natural enemies of Lantana in exotic populations, control of its populations has been usually limited or failed (Day et al., 2003). In thick stands, the shrub increases costs in forest management by inhibiting access in stands for thinning and felling, competes through rapid reproduction and increases fire hazards (Graaff, 1986). Lantana leaves contain poisonous triterpines and lantadenes A and B that may cause death of horses, cattle, sheep, goats, and rabbits by failure of the liver and other organs (Munyua et al., 1990). Its green fruits also contain the poisons and have caused illness and death in children (Morton, 1994).

Studies show that Kenya has been invaded by 34 different species; eleven arthropods, ten micro-organisms, nine plant species and four vertebrates (Farrell et al., 1995;

Lyons, 1999). Some of these invasions have yielded significant consequences on socio-economic status. Notable examples of invasive plants include the water hyacinth

(Eichhornia crassipes Mart) (Hill et al., 1999) in Lake victoria, the water fern

(Salvinia molesta D.S Mitch.) in Lake Naivasha, Mathenge (Prosopis juliflora S.W) in the Rift Valley (Baringo); Tickberry (Lantana camara L) in Nairobi National Park;

Jimsonweed (Datura stramoniun L) grows more in disturbed habitats example construction sites; Long spine cactus (Opuntia exaltata L); Sweet prickly pear

(Opuntia ficus indica L); Drooping prickly pear (Opuntia vulgaris Miller) Opuntia species mostly in drier areas as hedges and boundaries (Naivasha); Wild garlic (Allium vineale L) Mt Elgon; Mexican marigold (Tagetes minuta L.) Morning glory (Ipomoea indica (L) Roth) in Kiambu County and Eucalyptus (Eucalyptus globules Labill)

Kenyan riparian zones (Kedera, 2005).

1.2 Problem Statement and Justification

Exotic plant invasions are increasing in frequency and severity which elicit a concern for a decrease in biodiversity and habitat loss. These species have altered ecosystem structure, functions and biodiversity by displacing indigenous plants which shifts the floristic and faunal populations in protected areas (Vitousek, 1990). These invasive plants modify the invaded community in several ways that include alteration of native species diversity, evenness and richness. Furthermore, they also cause major ecosystem-level changes on nutrient cycling; nutrient availability and soil chemistry subsequently modify ecosystemic functionality (Gordon, 1998; Ehrenfeld, 2003).

Lantana camara is the most invasive plant at Ol-Donyo Sabuk National Park 5

(Wambua, 2012). This species is also top ranking invasive species globally (GISP,

2003).

Invasion of Ol-Donyo Sabuk National Park by Lantana has led to reduction of pasture grass while their impenetrable thorny thickets impede access to desired plant species by herbivores. In addition, this invasive species has been found to drastically reduce above ground biomass by smothering native species thus impacting negatively on the abundance of wildlife forage and consequent loss of biodiversity (Wambua, 2012;

Brocque et al., 2013). Ol-Donyo Sabuk National Park has a natural montane landscape with a rich biodiversity that require protection from the invasive species for posterity.

Despite a wealth of knowledge and many studies on the effects of invasive species, few studies have been done on their impacts on soil properties. This study therefore sought to understand the role played by this invasive plant species; - Lantana camara in altering soil chemistry in its bid to effectively colonise the invaded areas.

1.3 Study Hypothesis

1. Soil composition is the same in areas invaded by Lantana camara and those that

are not.

2. There is seasonal variation in soil nutrient composition in areas invaded by L.

camara and those that have native plants.

3. There is consistency of changes in soil nutrient properties across L. camara

invaded sites.

1.4 Research Questions

The study shall seek to answer the following questions: i) Does soil nutrient composition differ between areas invaded by Lantana camara

and those that are not? ii) Does soil nutrient composition differ with seasons in areas invaded by L. camara

and those that have native plants? iii) Is there any change in soil nutrient composition, across L. camara invaded sites?

1.5 Objectives

In order to understand the role played by invasive species in altering ecosystem structure, this study focused on how the invasive plant species alter the soil chemical properties which in turn, influence plant community composition and subsequently their ecosystem function.

1.5.1 General Objective

To determine the effect of invasive plant species on the soil chemistry at Oldonyo

Sabuk National Park.

1.5.2 Specific Objectives

i) To investigate whether soil composition is the same in areas invaded by

Lantana camara and those that are not.

ii) To investigate whether there is seasonal variation in soil nutrient composition

in areas invaded by L. camara and those that have native plants. 7

iii) To investigate consistency of changes in soil nutrient properties, if any, across

L. camara invaded sites.

1.6 Scope of the Study

The study is on invasive plant species found at Ol-Donyo Sabuk National Park and specifically L. camara which is the most abundant at the Park. The study shall limit itself to interplay between this invasive plant and soil nutrients.

1.7 Significance of the Study

Ol-Donyo Sabuk National Park is an isolated habitat patch which is surrounded by non- conservation land-uses resulting from privatisation of formally large communal land. It is therefore prone to invasion by alien plants which are common in agricultural farmlands. The Park has flora and fauna species that are threatened by the proliferation of invasive plant species. In order to design an effective eradication and preventive strategy for the invasive species in such an important ecological habitat, the relationship between soil chemistry and plant invasion needs to be well understood.

CHAPTER TWO

LITERATURE REVIEW

2.1 Definition of Invasive Species

Invasive species is a term that has generated debate within invasion biology terminology. The variation in its definition is mostly based on the location, interest or the discipline at play. A formal definition by Richardson et al., 2000, states that invasive plant species are naturalized plants that produce reproductive offspring often in large numbers, at considerable distances from parent plants and thus have the potential to spread over a considerable area. Invasive plant species are alien plant species also referred to as ‘non-native’, exotic, non-indigenous, which are introduced to an area, survive, and reproduce, and causes harm economically or environmentally within the area of introduction (Lockwood et al., 2007). They are also defined as widespread non–indigenous species that have adverse effects on the invaded habitat

(Colautti and MacIsaac, 2004). They are also termed as species that displace native species and have the ability to dominate an ecosystem, or a species that enters an ecosystem beyond its natural range and causes economic or environmental harm. Not all non-native/introduced, non-indigenous species are considered invasive because some have no negative effect and can be beneficial. A “non-indigenous” species is an organism (plant, animal and microbe) found living beyond its historic native range, which is usually taken as the area where it evolved to its present form (Lockwood et al., 2007).

An introduced or non-indigenous species becomes invasive when it becomes extremely abundant/dominant and causes economic or ecological harm. Such species are able to out-compete native species for resources such as nutrients, light, physical 9

space or water. Successful competition then allows the invader to quickly proliferate, become abundant and even expand its geographical range. Native species on the other hand are synonymous with indigenous species. They are species that occur naturally in an area, and have not been introduced by humans either intentionally or unintentionally (Lockwood et al., 2007). Invasive plant species often co-exist with native species for a while, then gradually out-competing them by adapting to the new location and growing in number and population density. They have traits or combinations that enable them to out-compete native species. The competition may be in form of growth rates and reproduction. Common invasive species traits include: fast growth, rapid reproduction, high dispersal ability, phenotypic plasticity (the ability to alter growth form to suit current conditions), tolerance of a wide range of environmental conditions (ecological competence), and prior successful invasions

(Ewell et al., 1999).

2.1.1 Mechanisms of Invasion by Invasive Species

Several mechanisms have been proposed to explain behaviour of invasive species

(Lockwood et al., 2007); these are species-based and ecosystem-based mechanisms. It is likely that a combination of several mechanisms cause an invasive situation to occur since not all introduced plants become invasive and in fact some of them become beneficial to the ecosystem. Species-based and ecosystems-based are among the mechanisms that when combined, establish invasiveness in a newly introduced species.

2.2.1.1 Ecosystem-based Mechanisms

The amount of available resources in the ecosystem and the extent to which these resources are utilised by living organisms determine the effects of additional new species on the ecosystem. In a stable ecosystem, equilibrium exists in the utilisation of available resources. These mechanisms describe a situation in which the ecosystems have been affected by disturbances that shifts the equilibrium of the ecosystem. When natural calamities occur in an ecosystem, such as forest fires, volcanic eruptions, flooding and earthquakes in an area, normal succession would favour certain native species. Invasive species mostly exploit disturbances of an ecosystem then colonize the area, for example large wildfires can sterilize soils at the same time adding a variety of nutrients (Davis et al., 2000). In the resulting free-for-all ecosystem, formerly well-established species lose their advantage, leaving more chance for the invasive species. In such cases plants species that can regenerate from their roots have an advantage. Non-natives with this ability can benefit from a low intensity fire that removes surface vegetation, leaving natives that rely on seeds for propagation to find their niches occupied when their seeds finally sprout (Brooks et al., 2004). This has an impact on the ecosystem by altering the composition of its organisms and utilization of available resources. Nitrogen and phosphorus nutrients are often the limiting factors in such situations.

Every species of living things has a role to play in its native ecosystem, some species have many and varied roles while others are specialized. These roles played by organisms in an ecosystem are known as niches. Some invading species are able to take up niches that are not utilised by native species and they can also create niches that did not exist. 11

Ecosystem changes can also alter species’ distributions for example the edge effects whereby when part of an ecosystem is disturbed due to anthropogenic factors such as road construction, agriculture and human settlement. The area remaining between undisturbed and the disturbed habitat forms a distinct habitat, creating new winners and losers and possibly hosting species that would not thrive outside the boundary habitat.

2.2.1.2 Species-Based Mechanisms.

Invasive species seem to have specific characteristics or combinations of specific characteristics that allow them to out-compete native species. Sometime they just have the ability to grow and reproduce more rapidly than native species this includes early sexual maturity, high reproductive output, the ability to disperse young widely, tolerance of a broad range of environmental conditions and other times, it is more complex involving a number of traits and interactions. Studies indicate certain traits like; fast growth, rapid reproduction rate, high ability of dispersal, phenotypic plasticity (ability to alter growth form to suit current conditions), tolerance to a wide range of environmental conditions mark a species as potentially invasive. One study found that from a list of invasive and non-invasive species, 86% of the invasive species could be identified from the traits alone (Callaway and Ridenou, 2004).

Another study found invasive species tended only to have a small sub-set of the invasive traits and that many of these invasive traits were found in a non-invasive species as well as indicating that invasiveness involves complex interaction which is not easily categorized (Thebaud et al., 1996; Reichard and Hamilton, 1997; Kolar,

2001). 12

Typically introduced species must survive at low population densities before it becomes invasive in a new location. At low population densities, it can be difficult for the introduced species to reproduce and maintain itself in a new location, so the species might be transported to a location a number of times before it becomes fully established. Related patterns of human movement from one location to another, such as ships sailing to and from ports or cars driving up and down highways, allow species to have multiple opportunities for establishment (also known as a high propagate pressure). An introduced species might become invasive if it can out-compete native species for the resources, such as nutrients, light, physical space, water and food. If these species evolved under great competition or predation, the new environment may allow them to proliferate quickly. Ecosystems in which all available resources are being used to their fullest capacity by native species can be modelled as zero sum systems, where any gain for the invader is a loss for the native (Stohlgren et al., 1999).

However, the unilateral competitive superiority and extinction of native species with increased populations of the invader is not always the case. Invasive species often co- exists with native species for some time, and then gradually the superior competitive ability of an invasive species become apparent as its population grows larger and denser and it adapts to its new location. An invasive species might be able to use resources previously unavailable to native species such as deep-water sources accessed by long tap roots, for example barbed goatgrass (Aegilops triuncialis L) was introduced to California on serpentine soils, which have low water-retention, low nutrient levels, a high Magnesium/Calcium ratio, and possible heavy metal toxicity.

Plant populations on these soils tend to show low density, but goatgrass can form 13

dense stands on these soils, crowding out native species that have adapted poorly to serpentine soils (Huenneke et al., 1990).

Ecological facilitation is another mechanism by which some species can alter their environment using chemicals or manipulating vital growth factors allowing it to thrive, while making the environment less favourable to other species, which it outcompetes. One such facilitative mechanism is allelopathy, whereby chemicals are present in parts of the plant and upon their release to the surrounding; they interfere with germination of many species (Kumar et al., 2011). Allelopathy is a form of interference competition where a plant releases chemicals into its immediate environment that suppresses the germination and/or growth of neighbouring seedlings and which may limit their survival (Wardle et al., 1998; Bais et al., 2003). Allelopathy of L. camara may be the cause of its toxicity to other plant species and its ability to cause shifts in species distribution and composition when it invades an ecosystem (R.

Choyal and Sharma, 2011). Lantana is an aggressive invader of natural ecosystem

(Kumar et al, 2011) therefore the establishment and persistence of L. camara may be partly driven by allelopathy (Achhireddy and Singh, 1987; Mersie and Singh, 1987).

Aqueous leaf extract of L. camara produced inhibitory effect on growth of

Parthenium hysterophorus (Mishra and Singh, 2010) and caused significant inhibitory effect on germination, root and shoot elongation and development of lateral roots. The lack of resistance or tolerance among indigenous plants provides a competitive advantage to the invader (Callaways and Ridenou, 2004).

Soil fungi play a key role in ecosystems, influencing a large number of important processes including plant nutrient acquisition, carbon cycling and soil formation (Van 14

der Heidjen et al., 2008) but the fungi can be impacted by plant allelochemicals. Rose et al. (1983) found high concentrations of water soluble extracts of the litter of four shrub and three conifer species either stimulated ectomycorrhizal fungal growth, inhibited growth, or had no effect depending on both fungal and litter species. Tissue extracts of Lantana may inhibit a range of fungal species, including saprotrophic and plant pathogenic taxa, and reduce overall fungal diversity (Sharma and Raghubanshi,

2011). Contrastingly, treatment of soils with Lantana leaf extracts increased the number of soil fungal species present in pot trials, though it inhibited interactions between plant roots and endophytic fungal taxa (Shaukat and Siddiqui, 2001).

2.2.2 Impacts of Invasive Species on Communities and Ecosystem

Invasive species may adversely affect the habitats and ecosystems they invade economically, environmentally and ecologically. They modify the invaded community in several ways that include alteration of native species diversity, evenness and richness via reduction of light penetration to the ground level where herbs grow and competition for available resources. They can cause ecosystem-level changes on nutrient cycling and hydrology which modify ecosystem functionality (Gordon, 1998).

They may disrupt by dominating a region, wilderness areas, protected areas, particular habitats, or wild land - urban land interface, through loss of natural controls (such as predators or herbivores).

Invasive impacts are often classified as economic, environmental, or social in nature

(Charles and Dukes, 2007). Economic impacts are those of direct consequence to humans leading to monetary losses. Social impacts focus predominantly on human health and safety but can also cover quality of life and other aspects of social structure. Environmental impacts are those that affect ecosystem structure and 15

function, often referring to loss of biodiversity or unique habitats. These impacts affect the delivery of food, freshwater as well as water purification, pollination, natural pest control, disease regulation, soil fertility, nutrient and water cycling. The impacts of invasive species can be categorised into three: species extinctions, change in energy nutrient, water cycling and disturbance of regime, climate and physical habitat.

2.2.2.1 Disturbance of Regime, Climate and Physical Habitat.

Several invasive species alter disturbance regimes (including fire, erosion, and flooding) or act as agents of disturbance themselves particularly in soil disturbance

(Mack, 1998). Fire enhancement can occur when grasses invade shrub lands and increase fire frequency, extent or intensity whereas fire suppression is more likely to occur when trees invade grassland and decrease fuel load and fire spread (Mack,

1998).

Invasive species may also alter the physical habitat. Invaders are capable of outcompeting natives and taking over habitats. In addition, they make the habitat less suitable for other species. Invasive plants may decrease the suitability of soil for other species by secreting salts for example tamarix plant (Tamarix ramosissima) (Zavaleta,

2000) the ice plant Mesembryanthemum crystallinum, (Vivrette and Muller, 1977), by acidifying the soil, or by releasing novel chemical compounds, as in allelopathy

(Callaway and Ridenour, 2004).

2.2.2.2 Species Extinctions and Community Structure

The effects of invasive species on native bio-diversity and community structure are well known, but few studies have examined the mechanisms that lead to these effects

(Levine et al., 2003). Invasive species may alter community structure through exploitative competition (indirect interactions such as resource use), and interference competition (direct interactions such as allelopathy in plants; Callaway and Ridenour

2004). This can change the abundance of species with fundamental traits that influence ecosystem processes (Chapin, 2000).

2.2.2.3 Change in Energy, Nutrient and Water Cycling

Invasive species’ also manifest their impacts at the ecosystem level through the alteration of natural cycles. Energy flows can be altered by changes in trophic level interactions, food webs and fundamental species. For example, the herbivore Pomacea canaliculata Lamarck (golden apple snail) dramatically decreased aquatic plant populations in wetlands in Southeast Asia. This in turn has led to the dominance of planktonic algae, high nutrient levels, high phytoplankton biomass, and turbid waters, with implications for water quality and purification (Carlsson, 2004). Productivity can be altered by invasive species that use resources more efficiently or eliminate a fundamental species (Dukes and Mooney, 2004). Since primary productivity is itself an ecosystem service, this shift could be detrimental to humans.

Changes in decomposition rates might occur if an invasive species altered litter chemistry which can affect nutrient cycling as well. Nutrient cycling can also be altered by invasive plants that fix nitrogen, leach chemicals that inhibit nitrogen fixation, release compounds that alter nutrient availability or retention, including 17

nitrogen and phosphorus, and modify top-soil erosion (Dukes and Mooney 2004).

There are several previous studies done which can best explain this mechanism. These include the introduction of Myrica faya Aiton (fire tree) in Hawaii, New Zealand and

Australia, and Acacia mearnsii (black wattle) in South Africa (Levine, 2003).

Ehrenfeld (2003) has shown that invasive plant impacts on nutrient cycling can vary in magnitude and direction across both invader types and sites, indicating that patterns are not universal. Alteration of nutrient cycling has additional implications for maintenance of soil fertility and primary production.

Invasive plant species have been shown to alter hydrological cycles by changing evapo-transpiration rates and timing, runoff, and water table levels. The impacts are greatest when the invaders differ from natives in traits such as transpiration rate, leaf area index, photosynthetic tissue biomass, rooting depth, and phenology (Levine,

2003). There are studies showing some invasive plants use more water than native plants, and thus decreasing the water supply for humans. These include Tamarisk spp.

(salt cedar) in riparian zones of the south western United States, and pines in the Cape region of South Africa (Levine, 2003).

2.3 Impacts of Invasive Species

Exotic species invasion is amongst the key global scale problems experienced by natural ecosystems and is also considered as the second largest threat to global biodiversity (Drake et al., 2003). The International Union for Conservation of Nature,

(IUCN) 2004, states that the impacts of alien invasive species are immense, insidious, and usually irreversible. A large number of extinctions have been attributed to invasive species. The ecological cost is the irreversible loss of native species and 18

ecosystems. Species that appear in new environments may fail to survive but often they thrive, become dominant and slowly replace native species. In fact, native species are likely to be unprepared to defend themselves against the invaders. This process, together with habitat destruction, has been a major cause of extinction of native species throughout the world in the past few hundred years (Callaway and Reinhart,

2006). Although in the past many of these losses have gone unrecorded, today, there is an increasing realisation of the ecological costs of biological invasion in terms of irretrievable loss of native biodiversity (Kourtev et al., 2002).

Many previous studies have found that native species grew better in their native soil than in ‘invaded’ soil (Simba et al, 2013; Sharma and Raghubanshi, 2011). This pattern generally indicates that exotic species can markedly affect soil in a way that is distinct from native species within a community (Ehrenfeld, 2003). This is consistent with several hypotheses, including the degraded mutualist hypothesis (Stinson, et al.,

2006; Kulmatiski et al., 2008; Vogelsang and Bever, 2009; Bever et al., 2010) where exotic species are thought to suppress native communities. For example, Batten et al.

(2006) found that a native species, Lasthenia californica L, was negatively affected by soil conditioned by an invasive grass (Aegilops triuncialis L). The ‘home’ advantage for native species is also consistent with the novel weapons hypothesis (Rout et al.,

2013) that exotic species may produce allelopathic or anti-microbial root exudates.

2.4 Impacts of Invasive Plant Species on Soil Development

Invasive plants have a varied range of impacts on plant communities through their direct and indirect effects on soil chemistry. For example, they modify the soil properties through root and litter exudates that affect soil structure and mobilize 19

and/or chelate nutrients as well as microbial activities (Batten et al., 2006). Invasion of native plant habitats by invasive plants can drastically change soil properties such as pH, inorganic nitrogen content, mineralization rates and nitrification rates. Some invasive species may gain a competitive advantage through the release of compounds or combinations of compounds that are unique to the invaded community. Invasive species that invade native plant populations can change the flow of nutrients leading to variations in soil microbial load and/or composition for example fungal to bacterial dominance is expected to decrease as nutrient inputs into the soil increase (Van der

Heijden et al., 2008). Plants have been shown to structure microbial communities through influences on soil nutrient availability and differential root exudates

(Grayston et al., 1996; Westover et al., 1997).

Invasive plant species change soil nutrients where the most notable is total nitrogen found largely in organic reservoirs and is transformed to its different states by microbial immobilization and mineralization (Nannipieri et al., 2003). In soils, bacteria and fungi populations often perform similar functions of decomposition and nutrient cycling. These soil microbial communities may be affected by the invasion of non-native plant species (Kourtev et al., 2002). Potential impacts on microbial communities from invasive species often occur because non-native species differ in plant morphology, phenology, and leaf litter chemical composition compared to co- occurring native plants (Ehrenfeld, 2003). Invasive plant species commonly escape inhibitive regulation by natural microbes residing in roots and at the interface between roots and soils (Kulmatiski et al., 2008), in fact they often benefit substantially from mutualisms with soil microbes in the areas they invade (Marler et al., 1999; Callaway and Reinhart, 2006). 20

2.5 General description and Ecology of Lantana camara

2.5.1 Taxonomy

Scientific name: Lantana camara L.

Synonyms: Camara vulgaris L, Lantana scabrida L.

Common names: Sleeper weed, Lantana, Wild sage and Prickly berry

Taxonomic position:

Kingdom: Plantae

Division: Magnoliophyta.

Class: Magnoliopsida.

Order: Lamiales.

Family: Verbenaceae.

Genus: Lantana

Species: camara (Source: Walton, 2006).

2.5.2 Description

L. camara is a medium-sized perennial aromatic woody shrub, 2-5 m tall Lantana forms dense mono-specific thickets, 1-4 m high and approximately 1-4 m in diameter, although some of the thickets smother nearby trees and reach up-to the height of 8-15 m (Sharma et al., 2005b). It has arching quadrangular stems in cross-section with pithy centers and sometimes has prickles or spines. As Lantana stems mature, they become rounded and turn grey or brown. Frequently, multiple stems arise from ground level and aged stems can be up to 15cm in diameter. The leaves are simple and oppositely arranged along the stem. They have leaf stalks (petioles) that are 5 to 30 21

mm long and a crenate or serrated (toothed) leaf margin and oval or broadly lance- shaped, 2-12 cm in length, and 2 to 6 cm broad, having a rough surface and a yellow- green to green colour as shown in figure 2.1. Their size and shape depends on the type of Lantana and the availability of light and moisture. Identification of Lantana types by flower colour can be difficult, as the colours of the inner buds as well as the inner and outer flowers must be considered. The flat-topped inflorescence may be yellow, orange, white, pale violet, pink, pink edged red or red. Flowers are small, multicoloured; in stalked, dense, flat-topped inflorescences (clusters of 20–40 individual flowers) are about 2.5 cm in diameter. Tightly packed angular flower buds open from the outside towards the centre of the inflorescence as they mature. The dense flower clusters consist of numerous small tubular flowers (9-14 mm long and 4

- 10 mm across). Single-seeded hard green fruit, of about 5-7 mm wide and 1.5mm long, grow in clusters and ripen to shiny black or purple fleshy berries (similar in appearance to a blackberry). Flowering and fruiting throughout the year with a peak during the first two months of the rainy season. The plant has a shallow root system made up of short taproot with lateral roots branching out to form a mat.

Figure 2.1: Lantana camara showing fruits and inflorescence. (IUCN, 2004)

Lantana camara is one of the many invasive species that severely affects the properties of the ecosystems in which it inhabits. It is native to Mexico, Central

America, tropical South America and the Caribbean with geographical expansion between 35°N and 35°S (Bharath et al., 2012). Locations within which L. camara is naturalised include Africa, Australia, India, south-eastern Asia and many oceanic islands with warm climates. It is invasive in large parts of Kenya for example the replacement of native pastures by Lantana is threatening the habitat of the sable antelope (Hippotragus niger Harris) at Shimba hills National reserve, Tanzania and

Uganda (Lyons and Miller, 1999). This species has been widely promoted as an ornamental since the early 1800s and it is widely naturalized throughout the

Neotropics. It has become very widespread in Australia, India and South Africa, infesting millions of hectares of land (Bhagwat et al., 2012). It is a significant weed of which there are some 650 varieties in over 60 countries, it is established and expanding in many regions of the world, often as a result of clearing of forest for 23

timber or agriculture, it impacts severely on agriculture as well as on natural ecosystems. The plants can grow individually in clumps or as dense thickets, crowding out more desirable species in disturbed native forests. It can become the dominant understory species disrupting succession and decreasing biodiversity. At small sites like at south-east Queensland in Australia infestations by Lantana has been so persistent that they have completely stalled the regeneration of rain forest for three decades (Day et al., 2003).

Due to its high growth rate, Lantana proliferates luxuriantly resulting in changes in species composition and soil properties. The growth architecture of Lantana is such that it prevents light infiltration to the ground. Resulting in marked heterogeneity in terms of irradiance beneath the Lantana bush and affects species diversity beneath its canopy. Light availability on the forest floor has been recognized as a key factor that influences intrinsic traits of inhabiting species (Jones et al., 1994; Walters, 1996). The dense cover created by vertical stratification of lantana may reduce the intensity or duration of light under its canopy thus decreases the herbaceous cover. This could be due to the creation of a photosynthetically inactive light regime at ground level

(Turton, 1992). Below certain thresholds, however, light limitation alone can prevent herbaceous species survival regardless of other resource levels (Tilman, 1982). It is likely that herbs are influenced by the amount of light that reaches the forest floor, and this may be probably one of the mechanisms responsible for the decline of herbaceous vegetation. Sharma and Raghubanshi (2007) advocated that the growth architecture pattern of lantana is such that it prevents the light penetration to the forest floor, leading to the decline of tree seedlings and possibly the herb flora.

Invasive plant species like Lantana camara L. possess the capability of creating niche by trapping leaf-litter. This trapping of litter is also dependent on Lantana cover, the denser the Lantana cover the greater the trapping potential. So, more organic matter accumulates up with increasing Lantana cover. Deposition of litter due to wind also affects the herbaceous vegetation (Everham and Brokaw, 1996). Accumulation of litter beneath the Lantana canopies builds up soil organic matter. Accumulation of soil

Nitrogen closely follows that of soil organic matter because, on average 99% of the

Nitrogen in terrestrial ecosystem is organically bound (Rosswall, 1976). Raghubanshi

(1992) reported strong positive relation between total nitrogen content and organic

Carbon content of soil in the dry deciduous forest ecosystem. According to Rawat et al., (1994) superiority in Nitrogen extraction from the soil along with an efficient re- translocation of Nitrogen from the senescing leaves enables Lantana to perform better as an invasive species. Several studies have shown that soil nutrient levels play an important role in determining community invisibility (Shea and Chesson, 2002;

Callaway and Reinhart, 2006). This self-perpetuating changed microhabitat could probably provide L. camara with increased resource leading to its successful proliferation. The allelopathy characteristics of Lantana species enable it to survive secondary succession and become monospecific thickets. For example, allelopathic effects resulting in either no growth or reduced growth close to Lantana camara have been demonstrated in Morrenia odorata L. (milkweed vine) (Achhireddy and Singh,

1987).

2.5.1 Habitat Description

Lantana has been the focus of biological control attempts for a century yet still poses major problems in many regions. The diverse and broad geographic distribution of

Lantana is a reflection of its wide ecological tolerance. It occurs in diverse habitats 25

and on a variety of soil types, soil range of mostly sandy to clay loam, water range from semi-arid to normal. Lantana generally grows best in open, un-shaded conditions such as wastelands, the edges of rain forests, on beachfronts, in agricultural areas, grasslands, riparian zones, scrub/shrub lands, urban areas, wetlands and forests recovering from fire or logging (Thakur et al., 1992). L. camara is an intolerant pioneer that colonizes disturbed areas like roadsides, railway tracks and canal banks, along fence lines and in pastures and parklands, in plantations, forest edges and gaps.

Lantana is also now seen invading native vegetation in woodlands and savannas notably in protected areas (Englberger, 2009). Where wet sclerophyll forests and rainforests have been disturbed through logging, gaps are created; this allows Lantana encroach on the forests.

Lantana grows under a wide range of climatic conditions, it grows within a pH range of between 4.5 to 8.5; it is intolerant of frequent or prolonged freezing temperature.

Some varieties can withstand minor frosts, provided these are infrequent. Prolonged freezing temperatures kill aerial woody branches and cause defoliation (Graaff, 1986).

This invasive plant is found at altitudes less than 2000m above sea level and prefers un-shaded habitats and can tolerate some shade. It does not grow at temperatures below 5oC (Cilliers, 1983). In Australia, the inland limit of its geographical range coincides with the 750mm isohyet in southern Queensland and the 1250mm isohyet in the north (Harley, 1973), with infestations being restricted to creek lines in drier areas.

In South Africa it is found in areas with a mean annual surface temperature greater than 12.5°C. Lantana does not invade intact rain forests, but is found on their margins where natural forests have been disturbed through logging creating gaps that enable

Lantana to encroach in the gaps. Further logging aggravates the condition and allows 26

Lantana to spread or become thicker in its growth. It cannot survive under dense, intact canopies of taller native forest species. The plant is susceptible to frosts and low temperatures, saline soils, boggy or hydromorphic soils, low rainfall, coralline soils with poor water-holding capacities and high incidence of tropical hurricanes

2.5.2 Reproduction

Lantana camara blooms almost continuously under favourable conditions. Somatic chromosome numbers of 33, 44, and 55 were recorded in India, the latter tetraploid being the most common (Sinha and Sharma, 1995). Approximately half the flowers form clusters of single-seeded berries. A single plant can produce up to 12000 fruits each year (Wellington and Amel, 2004). The fruits are blue-black when ripe and contain one seed each. Insects such as butterflies, moths, bees and thrips pollinate the flower clusters, self-pollination is not common. Seed dispersal can be biotic or abiotic; biotic is by frugivorous birds and to a lesser degree by foxes and other vertebrate foragers (Heleno et al., 2013a); they spread the seeds in their droppings. In Hong

Kong, L. camara is dispersed by 15 species of native birds (Corlett, 1998). Hawaii, it is mainly dispersed by exotics such as the Indian myna (Acridotheres tristis L)

(Atkinson and Atkinson, 2000; Heleno et al., 2013a). Some mammals also eat and disperse Lantana camara seeds; in the Galapagos Islands, it is one of the most dispersed alien plants, being mainly dispersed by two lizard species, and to a minor extent by the birds, Myiarchus magnirostris Gould and Mimus melanotis Gould

(Heleno et al., 2013a). If not eaten, they dry and remain on the shrub for weeks.

Abiotic dispersal happens occasionally like for example flash floods in South Africa, caused by cyclone Demoina in 1983, transported seeds and deposited them on the flood plain of the Ndumu Game Reserve (Bromilow, 1995). 27

Germination rate of fresh seed is generally low, it is reduced by low light condition but studies have shown that germination is more likely to occur if the seed has travelled through the gut of a bird or mammal. High light intensity and soil temperature will stimulate germination of seeds which means that human disturbances such as clearing of forest areas, construction and inappropriate burning; seeds are capable of surviving the hottest fires (Moody et al., 1984). Seeds need warm temperatures and sufficient moisture to germinate. Lantana seed survival of 21.3 percent has been recorded after 36 months under natural rainfall conditions and 27.2 percent after 24 months for seeds placed under irrigated conditions. Various studies have attributes seed viability ranging from 2-5 years. Lantana vegetation can spread via a process known as layering, where horizontal stems take root when they are in contact with moist soil. It can also re-shoot vigorously from the base of vertical stems and more slowly from the rooted horizontal stems, although it does not sucker from damaged or broken roots (Neena and Joshi, 2013).

2.6 The Effects of Lantana camara on Soil Chemistry

Lantana camara is an insidious weed and thought to have been introduced in East

Africa in the 1930s (Verdcourt, 1992), and has since naturalized itself in many of the region’s semi-arid areas including nature reserves where its negative effects cannot be under-estimated. Invasion of Lantana camara on native plant communities is capable of causing changes in soil properties and species composition. Different levels of

Lantana cover affect soil properties and herbaceous species composition (Sharma and

Raghubanshi, 2011). Results from the same study indicate that, as the Lantana cover increases some of the species get locally extinct while some are indeed favoured by its invasion. Sharma and Raghubanshi (2011) also show that the invasive L. camara affected soil nutrient levels such that concentration of organic carbon and total 28

nitrogen were significantly higher in places that were heavily invaded. Ragubanshi

(1992) shows a strong positive relation between total nitrogen content and organic carbon content of soil in a deciduous forest ecosystem. According to Rawat et al.

(1994) superiority in nitrogen extraction from the soil along with an efficient re- translocation of nitrogen from the senescing leaves enables Lantana camara to perform better as an invasive species. Several studies have shown that soil nutrient levels play an important role in determining community invisibility (Ehrenfeld, 2003;

Hawkes et al., 2005; Reinhart and Callaway, 2006). This self-perpetuating changed micro-habitat could probably provide L. camara with increased resources leading to its successful proliferation.

Some studies have shown that Lantana camara has allelopathic properties that inhibit seed germination and seedling growth of several native herbaceous plants and woody plants (An et al., 1998; Lwando, 2009). Allelopathy is a form of chemical competition or interference competition, whereby a plant secretes chemicals that make the surrounding soil un-inhabitable, or at least inhibitory, to competing species (Elisante and Ndakidemi, 2014). Allelo-chemicals potentially alter the basic nature of substrates in which plants grow and in turn increase the competitive ability of the invasive species for certain nutrients uptake. This could be the possible force behind the declines of nitrogen and phosporous in L. camara invaded sites (Simba et al., 2013).

Soil bacteria and fungi found in areas invaded by L. camara have been found to change nutrient cycling leading to competitive outcomes in their favour and against natives (Hawkes et al., 2005). According to Simba et al. (2013), macro-nutrient concentrations of magnesium, calcium and potasium were higher in L. camara invaded sites than non-invaded ones while phosporous and nitrogen generally 29

indicated low concentrations in the invaded sites relative to non-invaded ones. Also high pH levels were recorded in soils invaded by L. camara. High soil pH is known to accelerate litter decomposition and thus plays a crucial role in regulating nutrient availability. Generally fungi have a greater carbon to nitrogen ratio than bacteria hence fungi are expected to have lower nutrient requirements and bacteria are expected to have higher requirements. If access to carbon is equal but nitrogenis limiting then a shift toward fungal dominance is expected but if nitrogen is not limited then a shift toward bacterial dominance is expected.

CHAPTER THREE

STUDY AREA AND METHODOLOGY

3.1 Study Site – Ol Donyo Sabuk National Park

Source: Wambua, 2012

Figure 3.1: Ol Donyo Sabuk National Park; Vegetation Distribution (KWS, 1999)

The study was carried out in Ol-Donyo Sabuk National Park (Figure 3.1). The Park is located about 85km North-East of Nairobi in Machakos County. It lies between latitudes 1005’ and 10010’S, longitudes 37010’ and 37020’E and altitude 2145 metres above sea level at the peak of the hill. The Park was gazetted by the Kenya Wildlife

Service (KWS) in December 1967 and covers an area of 2068 hectares (KWS, 1999).

Ol Donyo Sabuk, means large mountain in Maasai language. The Kikuyu community 31

call it Kilimambogo ‘the Mountain of the Buffalo’ while traditionalists call the mountain by Kea-Njahe, meaning the 'Mountain of the Big Rain', one of Ngai's (God) lesser homes. The vegetation of the Park is mountane forest vary with climatic conditions of the area. According to the National Atlas of Kenya, rainfall is between

30-40 inches or 760-1015 mm per annum (KWS, 1999). The regime is bimodal with rains occurring mostly from March to May and October to December with peaks in

April and November. The maximum annual mean temperature is between 220C-260C and minimum annual mean is 100C-140C. January-March is hot and dry, April-June is hot and wet, July-October is very warm and dry, November - December is warm and wet (KWS, 1999).

3.1.1 Topography

Topography of the area is characterized by plains at the altitude 1400m-1700m above sea level with a solitary Ol-Donyo Sabuk Hill (Kilimambogo) situated in the middle of

Ol-Donyo Sabuk National Park rising to 2147m above sea level from an otherwise flat area and forms a substantial part of the study area. Several water streams originate from the Park forming a critical source of water for the local communities and the inherent wildlife (Ngoru et al., 2007).

3.1.2 Soils

The hill is surrounded by the expansive lava of the Athi plains with fertile black cotton soils which are well drained. The soils are varied due to geological formation; shallow and stony with rocky outcrops which have been subjected to geological and recent accelerated erosion losing their original characteristics. They are deeply weathered, except where they are eroded on steep slopes or where un-weathered rock outcrops occur (KWS, 1999). 32

3.1.3. Wildlife Communities The Park ecosystem is made of a hill which is mainly covered with dense montane forest except for a small area at the top of the hill. Forests and other vegetated habitats especially Ol-Donyo Sabuk National Park are important wildlife sanctuaries (KWS,

1999).

3.1.3.1 Animals

The Park has twenty species of mammals which are documented as resident in the study area. Bovids are the most dominant species and they include antelopes,

Buffaloes (Bubalus bubalis Linnaeus, 1758), Bushpig (Potamochoerus larvatus

Cuvier, 1822) and giant forest hog (Hylochoerus meinertzhageni Thomas, 1904). The carnivores are the second and comprise of African civets (Civettictis civetta Schreber,

1776), Dwarf mongoose (Helogale parvula Sundevall, 1847) and the Family

Hyaenidae comprising of spotted hyena (Crocuta crocuta Erxleben, 1777). Reptiles include the common python (Python sabae Gmelin, 1788), green mamba

(Dendroaspis angusticeps Smith, 1849), Black necked spitting cobra (Naja nigricollis

Reinhardt, 1843) and Nile monitor lizards (Veranus niloticus Linnaeus, 1758) (KWS,

1999). Ol-Donyo Sabuk National Park also has 45 bird species including, Rufous

Chatterer (Turdoides rubiginosa Rüppell, 1845), Greater Blue-eared Starling

(Lamprotornis chalybaeus Hemprich and Ehrenberg, 1828), Olive Thrush (Turdus olivaceus Rüppell, 1845 ), Collared Sunbird (Hedydipna collaris Vieillot, 1819),

White-browed Sparrow Weaver (Plocepasser mahali Smith, 1836), (Chebures, 1983).

3.1.3.2 Vegetation

The vegetation at Ol-Donyo Sabuk National Park can broadly be characterized into four types; closed bushland, closed trees, open grassland/bushland, and the herbaceous 33

crop (Ngoru et al., 2007). Parts of the middle section of the hill except the summit is

covered by the forest consisting of tall trees which include Croton macrostachyus and

Albizia schimperiana. Others include Cassipourea malosana L., Cussonia holstii L,

Albizia gumifera L., Tabernaemontana stapfiana L., Teclea simplicifolia and

Clausena ansata L. The undergrowth includes Landolfia buchananii L., Periplocca

linearifolia L. and Aneilema pedunculata L. The forest vegetation serves as water

catchment maintaining condition and flow of the several streams that emanate from

the hill (KWS, 1999).

Table 3.1: Study Sites and Some of the Native Plant Species

Name of Native species Point Coordinates Elevation Site (m.a.s.l) Baringo Terminalia catappa L. 9877038N, 1513 Croton megalocarpus Hutch 0306294E Euclea divinorum Hiern Pterolobium stellatum (Forssk) Brenan Park HQ Combretum molle R.Br 9877749N, 1694 Rhus ovata S.Watson 0304365E Olea africana Vahl Carrisa edulis Forssk Kitambasie Combretum molle R.Br 9874331N, 1700 Olea africana Vahl 0308316E Acacia hockii De Wild Euclea divinorum Hiern Nzambani Rhus vulgaris Meikle 9876633N, 1776 Combretum molle R.Br 0304660E Erythrina abyssinica Lam. ex DC Pterolobium stellatum (Forssk) Brenan Isooni Rhus natalensis Bernh ex Krauss 9873448N, 1831 Osyris lanceolata Hochst and Steud 0305275E Pterolobium stellatum (Forssk) Brenan Rhus natalensis Krauss

3.1.3 Land Use

Currently, the Park is an ‘island’ conservation area since it is surrounded by human settlements. The major land-use in the area bordering the Park is agriculture thus creating an island conservation area. These adjacent settlements are mainly involved in intensive smallholder rain-fed agriculture, irrigation farming growing horticultural crops such as peas and tomatoes and keeping of livestock. Nearby also, are large scale plantation farms which are involved in large-scale commercial horticulture of pineapples (Ngoru et al., 2007). Communities utilise the Park resources for basic needs like firewood, grass, meat from wildlife, medicinal plants, honey and pollinator services; in some cases illegal charcoal burning and livestock grazing. Encroachment and edge effects resulting from increased population and associated disturbances are some of the activities are some of the factors that have contributed to the introduction of invasive plant species into the National Park. The most common invasive plant species found at Ol-Donyo National Park are: Lantana camara, Datura stramonium L,

Solanum incanum L, Dovyalis caffra Warb, Ricinus communis L, Tagetes minuta L,

Caesalpinia decapelata (Roth) Alston (Faden, 1974).

3.2. Materials and Methods

A field reconnaissance survey was done in the Park to establish the areas invaded by

Lantana camara. The Park covers an area of 2068 hectares (KWS, 1999) and has been divided into 10 control blocks; Baringo, Itetani, Park Head quarter, Kitambaase,

Kumbu, Kivani, Isooni, Maindandu, Zambani and the Summit each measuring 500 meters by 500 meters (Figure 3.2) for the purpose of eradicating Lantana camara manually by uprooting and burning by the Kenya Wildlife Service.

The study was undertaken in the months of November to December 2014 for the wet season sampling while the dry season sampling was carried out between January and

March 2015. By purposeful sampling, 5 blocks were selected out of the 10 blocks for study. This was done to achieve a clear representation of the sample site of the whole

Park. They represented varying altitudes, the blocks selected are; Baringo, Park Head quarter, Isooni, Zambani, Kitambasie. The blocks were selected based on security- wild animals in the Park, accessibility-some parts had dense thickets making them inaccessible and areas invaded by L. camara and non-invaded. In each block, 2 quadrants (10mx10m) at interval of >100m were established randomly using tape measure and wooden pegs. One quadrant was invaded by L. camara and the other with native plant species (non-invaded). Each quadrant was sub-divided into 2 strata due to steep terrain (upper and lower) (Figure 3.3). In each stratum of the quadrant 20 sampling points were randomly established (Figure 3.3). Soil samples were then collected using a soil auger to a depth of 30 centimetres. 36

Figure 3.2: Map of Ol-Donyo Sabuk National Park Control Blocks for L. camara

(KWS, 1999)

The soil samples were collected directly under the canopies of Lantana camara and that of native species (Plate 3.1); this was done so as to get the direct effects of the plant. The soil samples from the 20 sampling points from each stratum were composited to one sample in a bucket to reduce variability within each stratum.

Therefore for each quadrant two composite soil samples were obtained and 37

subsequently, 4 composite samples per block. For the 5 blocks under study, a total of

20 composite soil samples were collected during the wet season and a similar number during the dry season. Desired amount of the collected soil samples was scooped and packed in airtight polythene bags. They were then clearly labelled (Plate 3.2) and sent to the laboratory for analysis. Random complete block design was used.

Upper stratum

Lower stratum

Figure 3.3: Divided quadrant for random composite sampling

Plate 3.1: Soil sampling using an auger

Plate 3.2: Collected soil samples (yellow labels for dry season and blue for wet season)

3.3 Laboratory Analysis of the Soil Samples

The soil samples were analysed at the Kenya Agricultural and Livestock Research

Organization (KALRO) Kabete laboratories in three replicates and averages obtained.

The following nutrients and parameters were determined: soil pH, potassium (K), calcium (Ca) magnesium (Mg), total nitrogen (N), phosphorous (P), total organic carbon (c), manganese (Mn), copper (Cu), iron (Fe), zinc (Zn), sodium (Na) and soil texture.

Soil pH value was determined using a pH meter – Geo-technical Engineering Bureau method (State of New York Department, 2007). The soil samples were separated on the ¼ inch (6.3 mm) sieve. 30 g of soil was weighed and placed into the glass beaker,

30g of distilled water was added to the soil sample and stirred to obtain soil slurry and then covered with watch glass. The samples were left to stand for a minimum of one hour. This is to allow the pH of the soil slurry to stabilize. The temperature of the soil samples was measured and the temperature controller of the pH meter was adjusted to 39

that of the soil sample temperature. The pH meter was standardized by means of the standard solutions. The soil samples were stirred well with a glass rod then the electrodes were placed into the soil slurry solution and the beaker turned gently to make good contact between the solution and the electrodes. The pH value was read and recorded.

Phosphorous, potassium, sodium, calcium, magnesium and manganese nutrient elements (P, K, Na, Ca, Mg and Mn) were analysed using the Mehlich Double Acid

Method (Mehlich, 1953). The soil samples were oven dried to remove moisture. For each nutrient analysed, a soil sample was extracted in the ratio of 1:5 (weight/volume) and introduced in a mixture of 0.1M dilute hydrochloric acid and 0.025M dilute sulphuric acid. Na, Ca, and K elements were determined using a flame photometric method (Toth and Prince, 1949) while P, Mg and Mn were determined using a

Calorimeter.

Total organic carbon was determined using calorimetric method (Charles and

Simmons, 1986); all the organic carbon in the soil sample was oxidised by acidified potassium dichromate at 1500oC for 30 minutes to ensure complete oxidation then barium chloride was added to the cool digests. The digests were then mixed thoroughly and allowed to cool overnight. The total carbon concentration was then read on the spectrophotometer at 600 nanometres.

Total nitrogen was determined using the Kjeldahl method (Bremner, 1996); the soil samples were digested with concentrated sulphuric acid containing potassium sulphate, selenium and copper sulphate hydrated to approximately 350oC. Total 40

nitrogen was determined by distillation followed by titration with dilute standardized

0.007144M dilute sulphuric acid.

Copper and iron were determined by atomic absorption spectrometry method (Charles and Simmons, 1986). 1.0g of oven-dry soil was weighed, 300 mesh size (0.05 mm sieve opening) into a 100 ml teflon beaker. 10 ml of concentrated nitric acid was added and boiled gently for about 30 minutes on a hot plate at 100-150˚ C, content allowed to cool. 5 ml perchloric acid and 15 ml hydrofluoric acid was added in a stainless steel fume hood and boiled gently for 60 minutes on a hot plate at 150-225˚

C until the volume is reduced to 2-3 ml. The contented was cooled and washed down with 5-10 ml of deionized water boiled for about 20 minutes. The contents were transferred into 100 ml polypropylene bottles for storage and determination of elements.

Soil texture is the relative proportion of sand, silt and clay hence sand, silt, and clay are the only particles used to determine soil texture. Soil texture affects the amount of nutrients and water that a soil can hold and be utilized by the plants. Physical properties of soil such as structure, and movement of air and water through the soil are also affected by soil texture. Soil textural classes are used to classify soil texture and are based on the relative proportions of the various soil separates it into sand, silt, and clay. There are 12 different soil textural classes which are determined using the texture triangle (Figure 3.4). Soil texture triangle is a tool that is used by soil scientists to understand the meaning of soil texture names. The triangle shows how each of these

12 textures is classified based on the percent of silt, sand and silt, and clay in each.

These percentages are based on the definition of sand and silt only. Alongside the 41

sides of the triangle percentage units (0-100%) of sand, silt, and clay are listed and the relative proportion of sand, silt, and clay always adds up to 100%.

Figure 3.4: Soil Textural Triangle http://www.ext.colostate.edu/mg/gardennotes/214.html.

Soil texture was determined using the Bouyoucos hydrometer method (Allen et al.,

1974) that was used to calculate the size of the soil particle. 100 ml of 5% dispersing solution of 50g of Sodium hexa-metaphosphate dissolved in deionised water was used to obtain dispersion.

3.4 Data Analysis

The difference in soil nutrient composition during the wet and dry season in areas invaded by L. camara and areas with native plants (non-invaded) was analysed using

Two-Way ANOVA. Three-Way ANOVA was also used to show analysis of interaction between wet and dry seasons; study sites and Lantana invaded and non- 42

invaded areas tested using Statistical Analytical Software (S.A.S) portable version 9.4 software. It was also used to determine how the nutrients varied during the dry and wet seasons in various sites outside and under the Lantana camara canopy. A post-

ANOVA test; Tukey's Honest Significant Difference (H.S.D) at 95% means was used to separate the means .The data was presented in descriptive tables.

CHAPTER FOUR

RESULTS

4.1 Soil Characteristics During the Wet Season

During the wet season soil pH varied significantly among sites (P-value<0.0001)(

Figure 4.1) . The pH value of Baringo study site was the highest with a mean (7.23) whereas Isooni had the lowest pH value with a mean value (5.94). There was significant difference in soil pH between the areas invaded by Lantana and non- invaded areas. The invaded areas had higher pH value mean (6.48) compared to the non-invaded with a pH value mean (6.30). Interaction between site and invasion had a significant effect on soil pH (P-value<0.0001).

Total nitrogen content varied significantly (P-value <0.0001) among the study sites

(Figure 4.1). The highest total nitrogen mean recorded at Isooni study site (0.20%) followed by Baringo site mean (0.19%) and the lowest at Kitambasie mean (0.13%).

There was significant difference (P-value <0.0001) in total nitrogen concentration between areas invaded by Lantana and the non-invaded. High mean (0.18%) of total nitrogen was recorded in areas not invaded by Lantana compared to the invaded areas that had a mean (0.16%). Interaction between site and invasion had a significant effect on total nitrogen (P-value<0.0001).

Total organic carbon content varied significantly (P-value <0.0001) among the study sites (Figure 4.1). Baringo study site had the highest mean (1.89%) followed by

Nzambani (1.83%) and Kitambasie had the lowest mean (1.10%). There was significant difference (P-value= 0.0020) in total organic carbon between areas invaded by L. camara and the non-invaded areas in the study sites. The non-invaded areas had 44

higher mean (1.73%) in total organic carbon compared to the invaded (1.49%).

Interaction between site and invasion had a significant effect on total organic carbon

(P-value<0.0001).

Phosphorous concentration varied significantly (P-value <0.0001) among the study sites (Figure 4.1). Isooni study site had the highest mean (27.51me%) in phosphorous followed by Park Headquarter study site which had a mean (27.49me%) while

Kitambasie study site had the lowest mean (12.72me%) in phosphorous. There was significant difference (P-value <0.0001) in phosphorous concentration between the

Lantana invaded areas and non-invaded. Areas invaded by Lantana had significantly higher phosphorous (21.11me%) compared to non-invaded areas 18.81me%.

Interaction between site and invasion had a significant effect on Phosphorous (P- value<0.0001).

Concentration of potassium varied significantly with the study sites (P-value <0.0001) and was significantly higher in areas invaded by L. camara mean (2.49me%) than the non-invaded that had mean (1.89me%). There was high mean (7.60me%) in

Potassium at Isooni study site and Nzambani study site had the lowest mean

(0.49me%). Further, interaction between the site and invasion resulted in high concentration of potassium (P-value<0.0001).

Calcium concentration did not vary significantly with the study sites nor was it different between the areas invaded by L. camara and those not invaded.

Magnesium content varied significantly (P-value<0.0001) among the study sites

(Figure 4.1).Baringo study site had a high mean (2.59me %) in magnesium followed by Isooni site with mean (2.56me%) and Kitambasie had the least mean (0.30me%).

There was also significant difference (P-value 0.0045) in magnesium concentration between the invaded and non-invaded areas. The areas non-invaded by Lantana in the study sites had high mean (3.05me%) in magnesium concentration compared to the invaded areas (3.00me%). Interaction between site and invasion had a significant effect on magnesium (P-value = 0.001).

Figure 4.1: Comparison of pH and soil nutrients during wet season 48

Iron content varied significantly (P-value<0.0001) among the study sites (Figure 4.1).

Kitambasie site had a high mean (21.75ppm) in iron and Baringo site had the least mean (13.41ppm) in iron. There was significant difference (P-value 0.0007) in iron concentration between the Lantana invaded and non-invaded areas. However the non- invaded areas had a high mean (16.92ppm) in iron concentration compared to the invaded areas that had a lower mean (16.48ppm). Interaction between site and invasion had a significant effect on iron (P-value<0.0001).

Concentration of zinc varied significantly with the study sites (P-value <0.0001) and was significantly higher (P-value <0.0001) in areas non-invaded by L. camara mean

(10.79ppm) than the invaded that had mean (8.75ppm). Baringo site had a high mean

(20.45ppm) in zinc concentration and Kitambasie site had the least mean (2.42ppm).

Interaction between site and invasion had a significant effect on zinc (P- value<0.0001).

Sodium content varied significantly (P-value<0.0001) among the study sites (Figure

4.1). Kitambasie study site had a high mean (21.75me%) in sodium concentration followed by Nzambani site (16.54me%) and Baringo study site had the least mean

(13.41me%). Sodium concentration was significantly higher (P-value<0.0001) in the

L. camara non-invaded areas mean (16.92me%) than the invaded areas mean

(16.48me%). Further, interaction between the site and invasion resulted in high concentration of sodium (P-value<0.0001).

Copper content varied significantly (P-value<0.0001) among the study sites (Figure

4.1). Nzambani study site had a high mean (11.56ppm) in copper concentration 49

followed by Baringo site mean (9.88ppm) and Kitambasie study site had the least mean (7.14ppm) (Figure 4.1). There was significant difference (P-value <0.0001) in copper concentration between the Lantana invaded and non-invaded areas. The non- invaded areas in the study sites had high mean (10.66ppm) in copper concentration compared to the invaded areas that had low mean (8.18ppm). Interaction between site and invasion had a significant effect on copper (P-value<0.0001).

4.2 Soil Characteristics During the Dry Season

During dry season soil pH varied significantly among sites (P-value <0.0001) (Figure

4.2). Park Headquarter had high pH value mean (6.88) followed by Nzambani mean

(6.86) and Isooni the least pH mean (6.21). There was significant difference (P-value

<0.0001) in pH between the invaded and non-invaded areas. However the areas invaded by L. camara had a high mean of 6.88 in pH compared to the non-invaded areas that had a mean of 6.30. Interaction between site and invasion had a significant effect on pH (P-value<0.0001).

Total nitrogen content had no significant difference (P-value= 0.0862) among the study sites (Figure 4.2). Kitambasie site had high mean (0.49%) in total nitrogen concentration then Park Headquarter study site had the least mean (0.17%). Total nitrogen concentration was significantly higher (P-value = 0.0222) in the L. camara invaded areas mean (0.36%) than the non-invaded areas mean (0.18%). Interaction between site and invasion had a significant effect on total nitrogen (P-value 0.0462).

Total organic carbon content varied significantly (P-value<0.0001) among the study sites (Figure 4.2). Baringo study site had high mean (2.11%) in total organic carbon concentration and Kitambasie site least mean (1.48%). Total organic carbon 50

concentration was significantly higher (P-value<0.0001) in the L. camara invaded areas mean (1.78%) than the invaded areas mean (1.72%). Further, interaction between the study site and invasion significantly (P-value<0.0001) affected concentration of total organic carbon.

Phosphorous content varied significantly (P-value<0.0001) among the study sites

(Figure 4.2). Phosphorous concentration mean (30.01me%) was high in Isooni site followed by Baringo site mean( 27.04me%) and Nzambani site had the least mean

(7.02me%) (Figure 4.2). There was significant difference (P-value<0.0001) in phosphorous concentration between the areas invaded by Lantana and the non- invaded areas. The L. camara invaded areas had high mean (20.76me%) compared to the non-invaded areas that had low mean (18.81me%) in phosphorous concentration.

Interaction between site and invasion had a significant effect on phosphorous (P- value<0.0001).

Concentration of potassium varied significantly (P-value <0.0001) among the study sites (Figure 4.2).Mean (2.63me%) was high in Kitambasie site while Park

Headquarter site had the least mean (0.63me%). It was significantly higher (P-value

<0.0001) in areas non-invaded by L. camara mean (1.89me%) than the invaded that had mean (1.08me%). Interaction between site and invasion had a significant effect on phosphorous (P-value<0.0001).

Concentration of calcium varied significantly (P-value <0.0001) among the study sites

(Figure 4.2). Kitambasie site had high mean ( 8.95me%) and Baringo site had the least mean (5.01me%). There was significant difference (P-value <0.0001) in calcium concentration between the areas invaded by Lantana and the non-invaded areas. The 51

non-invaded areas had high mean (6.87me%) in calcium compared to the Lantana invaded mean (5.65me%). Interaction between site and invasion had a significant effect on calcium (P-value<0.0001). 52

Figure 4.2: Comparison of pH and soil nutrients during dry season 53

Magnesium concentration varied significantly (P-value <0.0001) among the study

(Figure 4.2). Magnesium mean (3.71me%) was high in Kitambasie site while Park

Headquarter site had the least mean (2.45me%) . There was significant difference (P- value <0.0001) in magnesium concentration between the areas invaded by Lantana and the non-invaded areas. The Lantana non-invaded areas had high mean of

3.03me% in magnesium than the invaded that had a mean of 2.43me%. Interaction between site and invasion had a significant effect on magnesium (P-value<0.0001).

Content of manganese varied significantly (P-value <0.0001) among the sites of study

(Figure 4.2). Isooni site had high mean (1.31me%) in manganese concentration and

Kitambasie site had the least mean (0.60me%). There was significant difference (P- value <0.0001) in manganese between the areas invaded by Lantana and the non- invaded areas. The Lantana invaded areas had high mean (1.03me%) in manganese than the non-invaded mean (0.84me%). Interaction between study sites and invasion affected manganese significantly (P-value <0.0001).

Copper concentration varied significantly (P-value <0.0001) among study (Figure

4.2). Isooni site had high mean (11.09ppm) in copper concentration and Kitambasie site had the least mean (8.56ppm). There was significant difference (P-value <0.0001) in copper concentration between the areas invaded by Lantana and the non-invaded areas. The Lantana non-invaded areas had high mean (11.05ppm) in copper concentration than the non-invaded that had a lower mean (9.17ppm). Interaction between study sites and invasion affected copper significantly (P-value <0.0001).

Iron concentration varied significantly (P-value <0.0001) among the study sites

(Figure 4.2). Kitambasie site had high mean (26.62ppm) in iron concentration followed by Nzambani site mean (26.54ppm) and Baringo site the least mean

(15.79ppm). There was significant difference (P-value <0.0001) in iron concentration between the Lantana invaded areas and the non-invaded areas. The Lantana invaded areas had high mean (24.97ppm) in iron compared to the non-invaded that had lower mean (17.72ppm). Interaction between study sites and invasion affected iron significantly (P-value <0.0001).

Zinc content varied significantly (P-value <0.0001) among the study sites (Figure 4.2).

Baringo site had high mean (9.51ppm) in zinc while Baringo site had the least mean

(2.49ppm). There was significant difference (P-value <0.0001) in zinc concentration between the Lantana invaded areas and the non-invaded areas. The Lantana non- invaded areas had high mean (8.20ppm) in zinc compared to the invaded areas that had lower mean (4.62ppm). Interaction between the study sites and invasion significantly (P-value<0.0001) affected zinc concentration.

Concentration of sodium varied significantly (P-value <0.0001) among the study sites

(Figure 4.2). The mean of sodium was high at Kitambasie study site (1.32me%) while

Baringo study site had the least mean (0.37me%). There was significant difference (P- value <0.0001) in sodium concentration between the Lantana invaded areas and the non-invaded areas. The areas not invaded Lantana had high mean (0.77me%) compared to the invaded areas that had lower mean (0.46me%). Interaction between the study sites and invasion significantly (P-value<0.0001) affected sodium concentration. 55

4.3 Effect of Seasons and Invasion on Soil Nutrients and pH

During the wet and dry season pH value varied significantly (P-value <0.0001)

(Figure 4.3). The pH value was higher during the dry season compared to wet season

(6.59 and 6.39 respectively). There was significant difference (P-value <0.0001) in pH among the study sites. Baringo study site had the highest pH value (6.95) and Isooni with the least pH value (6.07). There was significant difference (P-value <0.0001) between the Lantana invaded and non-invaded areas. The Lantana invaded areas had high pH value (6.68) compared to the non-invaded that had pH value (6.30). There was significant interaction (P-value <0.0001) among the study sites and the wet and dry season in influencing pH in the soil. There was significant interaction (P-value

<0.0001) between wet and dry season and Lantana invaded areas in influencing pH in the soil. There was significant interaction (P-value <0.0001) among the study sites and

Lantana invaded areas in influencing pH in the soil. Interaction between the study sites, wet and dry season and invasion significantly (P-value<0.0001) affected pH in the soil.

Total nitrogen concentration varied significantly (P-value=0.0111) during both wet and dry season (Figure 4.3). Total nitrogen concentration was higher during the dry season mean (0.27%) than the wet season mean (0.17%). There was no significant difference (P-value = 0.1476) in total nitrogen among the study sites. Kitambasie study site had high mean (0.31%) and Park Headquarter had the lowest mean (0.17%).

There was significant difference (P-value 0.0271) between the Lantana invaded and non-invaded areas. The Lantana invaded areas had higher mean (0.26 %) in total nitrogen concentration compared to the non-invaded (0.18%). There was significant interaction (P-value = 0.0228) between all the study sites and the wet and dry season 56

in influencing total nitrogen in the soil. Concentration of total nitrogen varied significantly (P-value = 0.0111) between wet and dry season and Lantana invaded areas. There was significant interaction (P-value = 0.0401) among the study sites and

Lantana invaded areas in influencing total nitrogen concentration in the soil. There was significant interaction (P-value = 0.0228) among the study sites, wet and dry season and Lantana invaded areas in influencing total nitrogen concentration in the soil.

Total organic carbon concentration varied significantly (P-value <0.0001) during both wet and dry season (Figure 4.3). Total organic carbon concentration was high during the dry season mean (1.85%) compared to the wet season that had lower mean

(1.61%). There was significant difference (P-value <0.0001) in total organic carbon among the study sites, Baringo study site had high mean (2.00%) and Park

Headquarter had the lowest mean of 1.29%. There was no significant difference (P- value = 0.8377) between the Lantana invaded and non-invaded areas (Figure 4.3). The

Lantana invaded areas had higher mean (1.73 %) in total organic carbon concentration compared to the non-invaded that had lower mean (1.72%). The concentration of total organic carbon varied significantly (P-value = 0.0163) among the study sites and the wet and dry season (Figure 4.3). There was significant interaction (P-value <0.0001) between wet and dry season and Lantana invaded areas in influencing total organic carbon concentration in the soil. There was significant interaction (P-value < 0.0001) among the study sites and Lantana invaded areas in influencing organic carbon concentration in the soil. There was significant interaction (P-value = 0.0163) among the study sites, wet and dry season and Lantana invaded areas in influencing total organic carbon concentration in the soil. 57

Phosphorous concentration varied significantly (P-value = 0.0004) during both wet and dry season (Figure 4.3). The phosphorous concentration was higher during the wet season compared to the dry season (mean of 19.78me% and 20.00me% respectively).

There was significant difference (P-value <0.0001) in phosphorous concentration among the study sites. Isooni study site had higher mean (28.76me%) in phosphorous and Nzambani study site had the lowest mean (1.29me%). Phosphorous concentration varied significantly (P-value <0.0001) in the Lantana invaded and non-invaded areas.

The Lantana invaded and non-invaded areas had the same mean (20.93me%) in phosphorous concentration. There was significant interaction (P-value <0.0001) among the study sites and the wet and dry season in influencing phosphorous concentration in the soil). There was significant interaction (P-value = 0.0004) between wet and dry season and Lantana invaded areas in influencing phosphorous concentration the soil. Phosphorous concentration varied significantly (P-value <

0.0001) among the study sites and Lantana invaded areas in influencing phosphorous concentration in the soil. Interaction between the study sites, wet and dry season and invasion significantly (P-value<0.0001) affected phosphorous in the soil.

During the wet and dry season potassium concentration varied significantly (P-value

<0.0001) (Figure 4.3). The concentration of potassium was higher during the wet season compared to the dry season (mean of 2.18me% and 1.49me% respectively).

There was significant difference (P-value <0.0001) in potassium concentration among the study sites. Kitambasie study site had higher mean (4.07me%) in potassium and

Nzambani had the lowest mean (1.79%). There was significant difference (P-value =

0.0037) between the Lantana invaded and non-invaded areas (Figure 4.3). The areas not invaded by L. camara had higher mean (1.89me%) in potassium concentration 58

compared to the invaded areas that had lower mean (1.79me%). There was significant interaction (P-value <0.0001) among the study sites and the wet and dry season in influencing potassium concentration in the soil. There was significant interaction (P- value <0.0001) between wet and dry season and Lantana invaded areas in influencing potassium concentration in the soil. There was significant interaction (P-value

<0.0001) among the study sites and Lantana invaded areas in influencing potassium concentration in the soil. The concentration of potassium varied significantly (P-value

<0.0001) among the study sites, wet and dry season and Lantana invaded areas in influencing potassium concentration in the soil.

The concentration of calcium did not vary significantly (P-value = 0.2054) during both wet and dry season (Figure 4.3), among the study sites neither was it different between the L. camara invaded and non- invaded areas. Magnesium concentration varied significantly (P-value <0.0001) during both wet and dry season (Figure 4.3). The concentration of magnesium was higher during the wet season mean (3.02me%) compared to the dry season mean (2.73me%). There was significant difference (P- value <0.0001) in magnesium concentration among the study sites. Kitambasie study site had higher mean (4.34me%) in magnesium followed by Baringo (2.56me%) and

Park Headquarter had the least mean (2.45me%). There was significant difference (P- value <0.0001) between the Lantana invaded and non-invaded areas. The L. camara non-invaded areas had higher mean of (3.04me%) inmagnesium concentration compared to the non-invaded areas that was lower mean (2.71me%). There was significant interaction (P-value <0.0001) between all the study sites and the wet and dry season in influencing magnesium concentration in the soil. 59

Invaded, non-invaded, seasons and sites

Figure 4.3: Interactions between Lantana invaded, non-invaded, wet and dry seasons and sites 60

There was significant interaction (P-value <0.0001) between wet and dry season and

Lantana invaded areas in influencing magnesium concentration in the soil. There was significant interaction (P-value <0.0001) among the study sites and Lantana invaded areas in influencing magnesium concentration in the soil. Interaction between the study sites, wet and dry season and invasion significantly (P-value<0.0001) affected magnesium in the soil.

Manganese concentration varied significantly (P-value <0.0001) during both wet and dry season (Figure 4.3). The concentration of manganese was higher during the dry season with a mean (0.96me%) compared to the wet season mean (0.86me%). There was significant difference (P-value <0.0001) in manganese concentration among the study sites. Isooni study site had higher mean (1.33me%) in manganese and

Kitambasie study site had the least mean (0.45me%). There was significant difference

(P-value <0.0001) between the Lantana invaded and non-invaded areas. The L. camara invaded areas had higher mean (0.99me%) in manganese concentration compared to the non-invaded areas mean (0.83me%). There was significant interaction (P-value <0.0001) among the study sites and the wet and dry season in influencing manganese concentration in the soil. There was significant interaction (P- value <0.0001) between wet and dry season and Lantana invaded areas in influencing manganese concentration in the soil. There was significant interaction (P-value =

0.0508) among the study sites and Lantana invaded areas in influencing manganese concentration in the soil. There was significant interaction (P-value <0.0001) among the study sites, wet and dry season and Lantana invaded areas in influencing manganese concentration in the soil.

Concentration of copper varied significantly (P-value <0.0001) during both wet and dry season (Figure 4.3). The concentration of copper was higher during the dry season mean (10.13ppm) compared to the wet season mean (9.42ppm). There was significant difference (P-value <0.0001) in copper concentration among the study sites.

Nzambani study site had higher mean (10.69ppm) in copper and Kitambasie study site had the least mean (7.85ppm). There was significant difference (P-value <0.0001) between the Lantana invaded and non-invaded areas. The concentration of copper was higher in the L. camara non-invaded areas than the invaded (mean of 10.88ppm and

6.68ppm respectively). There was significant interaction (P-value <0.0001) among the study sites and the wet and dry season in influencing copper concentration in the soil.

There was significant interaction (P-value <0.0001) between wet and dry season and

Lantana invaded areas in influencing copper concentration in the soil. There was significant interaction (P-value <0.0001) among the study sites and Lantana invaded areas in influencing copper concentration in the soil. There was significant interaction

(P-value <0.0001) among the study sites, wet and dry season and Lantana invaded areas in influencing copper concentration in the soil.

During the wet and dry season the concentration of Iron varied significantly (P-value

<0.0001) (Figure 4.3). The concentration of Iron was higher during the dry season mean (21.19ppm) compared to the wet season mean (16.70ppm). There was significant difference (P-value <0.0001) in iron concentration among the study sites

(Table 4.3). Kitambasie study site had higher mean (24.18ppm) in iron followed by

Nzambani mean (21.54ppm) and Baringo site had the least mean (14.60ppm). There was significant difference (P-value <0.0001) between the Lantana invaded and non- invaded areas. Iron concentration was higher in the Lantana invaded areas compared 62

to the non-invaded (mean of 20.57ppm and 17.32ppm respectively). There was significant interaction (P-value <0.0001) among the study sites and the wet and dry season in influencing iron concentration in the soil (Figure 4.3). There was significant interaction (P-value <0.0001) between wet and dry season and Lantana invaded areas in influencing iron concentration in the soil. There was significant interaction (P-value

<0.0001) among the study sites and Lantana invaded areas in influencing iron concentration in the soil. There was significant interaction (P-value <0.0001) between all the study sites, wet and dry season and Lantana invaded areas in influencing iron concentration in the soil.

Zinc concentration varied significantly (P-value <0.0001) during both wet and dry season (Figure 4.3). The concentration of zinc was higher during the wet season mean

(9.77ppm) compared to the dry season mean (6.41ppm). There was significant difference (P-value <0.0001) in zinc concentration among the study sites. Baringo study site had higher mean (14.98ppm) in zinc followed by Kitambasie mean

(2.46ppm) and Baringo site had the least mean (14.60ppm). There was significant difference (P-value <0.0001) between the Lantana invaded and non-invaded areas.

The L. camara non-invaded areas had higher mean (9.50ppm) in zinc concentration compared to the invaded areas mean (6.69ppm). There was significant interaction (P- value <0.0001) between all the study sites and the wet and dry season in influencing zinc concentration in the soil. There was significant interaction (P-value <0.0001) between wet and dry season and Lantana invaded areas in influencing zinc concentration in the soil. There was significant interaction (P-value <0.0001) between all the study sites and Lantana invaded areas in influencing zinc concentration in the soil. There was significant interaction (P-value <0.0001) among the study sites, wet 63

and dry season and Lantana invaded areas in influencing zinc concentration in the soil.

Sodium concentration varied significantly (P-value <0.0001) during both wet and dry season (Figure 4.3). The concentration of sodium was higher during the wet season mean (0.67ppm) compared to the dry season mean (0.61ppm). There was significant difference (P-value <0.0001) in sodium concentration among the study sites.

Kitambasie study site had highest mean in sodium and Nzambani site had the least mean. (1.53ppm and 0.34ppm respectively). There was no significant difference (P- value = 0.8931) between the Lantana invaded and non-invaded areas. Both L. camara invaded and non-invaded areas had the same concentration of sodium with a mean of

0.64ppm. There was significant interaction (P-value <0.0001) among the study sites and the wet and dry season in influencing sodium concentration in the soil). There was significant interaction (P-value <0.0001) between wet and dry season and Lantana invaded areas in influencing sodium concentration in the soil. There was significant interaction (P-value <0.0001) among the study sites and Lantana invaded areas in influencing sodium concentration in the soil. There was significant interaction (P- value <0.0001) between all the study sites, wet and dry season and Lantana invaded areas in influencing sodium concentration in the soil.

In summary comparison of nutrients between the Lantana invaded and non-invaded areas showed that Lantana invaded areas had higher levels of pH, total nitrogen,

Phosphorous, iron, calcium, copper, zinc, magnesium, potassium and total organic carbon but the concentration of sodium was the same in both Lantana invaded and non-invaded areas (Figure 4.3). The nutrients concentrations varied between seasons.

The wet season had higher concentrations of phosphorous, potassium, calcium, 64

magnesium, zinc and sodium (Figure 4.3).The nutrients concentrations also varied among the five study sites with the seasons. During the wet season Baringo study site had the highest levels of pH, total carbon and zinc; Kitambasie had calcium, magnesium. Iron and sodium; Isooni had total nitrogen, phosphorous, manganese and potassium; Nzambani recorded the highest concentration in copper (Figure 4.1).

During the dry season Kitambasie had the highest concentrations in total nitrogen, potassium, sodium, iron, calcium and magnesium; Park Headquarter recorded the highest pH level Isooni had phosphorous, manganeses and copper while Baringo had the highest concentration in zinc and total organic carbon (Figure 4.2). Kitambasie recorded the highest concentration of nutrients during both wet and dry seasons while

Park Headquarter and Nzambani sites each had only one parameter as the highest

(Figure 4.1 and Figure 4.2).

4.4 Consistency of pH and soil nutrients in the Lantana invaded areas.

It was found out that there was consistency in nutrients in the Lantana invaded areas in some of the study sites. The pH of soils in the Lantana invaded patches was high at

Isooni, Park Headquarter and Nzambani while Baringo and Kitambasie study sites had low pH values in comparison to the non-invaded areas (Figure 4.4). Three out of five study sites had high pH values in the Lantana invaded areas compared to the non- invaded areas this showed inconsistency.

The concentration of total nitrogen in the Lantana invaded areas was higher in three out of five sites in comparison to non-invaded areas. Isooni, Kitambasie and

Nzambani sites had higher concentration of total nitrogen in the Lantana invaded areas while Park Headquarter and Baringo had low total nitrogen concentration 0.19% 65

in invaded areas in comparison to non-invaded areas that had 0.21%. This showed inconsistency across all study all the sites (Figure 4.4). Concentration of organic carbon showed consistency across all the study sites. The concentration of organic carbon in the Lantana invaded areas was higher in four sites Kitambasie, Isooni,

Nzambani and Park Head quarter except for Baringo that had the concentration organic carbon high in non-invaded areas 0.95% in comparison to Lantana invaded that had 2.05% (Figure 4.4).

There was inconsistency in the concentration of phosphorous in Lantana invaded areas. Three sites had high concentrations of phosphorous in Lantana invaded areas compared to non-invaded. These were; Kitambasie (invaded 25.09me %, non-invaded had 9.97me %), Nzambani (invaded 14.54me %, non-invaded 10.03me %) and Park

Headquarter (invaded 20.12me%, non-invaded 19.99me %). Three sites had the

Lantana invaded areas having higher concentration of phosphorous than the non- invaded areas (Figure 4.4). 66

Figure 4.4: Consistency of nutrients levels in invaded and non-invaded areas 67

There was inconsistency in the concentration potassium in the Lantana invaded areas in comparison to the non-invaded across all the study sites. Three out of the five sites

Baringo, Isooni and Kitambasie had low concentration of potassium in the Lantana invaded areas compared to non-invaded. Study sites of Nzambani (invaded 1.55me %, non-invaded 1.39me %) and Park Headquarter (invaded 1.24me %, non-invaded

1.00me %) had high concentration of potassium the invaded areas compared to the non-invaded Figure 4.4).

The concentration of calcium in the Lantana invaded areas was recorded high at four sites out of five compared to the non-invaded that had only one site with low concentration calcium in invaded areas. Nzambani (invaded 6.30me% and non- invaded 2.47me %), Kitambasie (invaded 52.94me% and non-invaded 14.38me %),

Baringo (invaded 1.19me%, non-invaded 0.002me %) and Park Headquarter (invaded

6.89me% and non-invaded 4.91me %) (Figure 4.4).This showed consistency in the concentration of calcium in the Lantana invaded areas across all the study.

The concentration of magnesium in the Lantana invaded areas was high in one site

Park Headquarter (invaded 2.50me % and non-invaded 2.40me %) while the other four sites, the non-invaded areas had higher concentration of magnesium than the

Lantana invaded areas (Figure 4.4). This showed consistency in the concentration of magnesium in the Lantana invaded areas across all the study sites that had low concentrations in magnesium in four sites. Manganese content in the Lantana invaded areas was higher in two sites Kitambasie (invaded 0.65me% and non-invaded

0.25me%) and Nzambani (invaded 1.06me% and non-invaded 0.04me%) the remaining three sites, the non-invaded areas had higher concentration of manganese 68

than the Lantana invaded areas (Figure 4.4).This showed inconsistency in the concentration of manganese in the Lantana invaded areas across all the study sites.

Copper content in the Lantana invaded areas was high in only two sites Isooni

(invaded 11.09ppm and non-invaded 10.25ppm) and Park Headquarter (invaded

12.80ppm and invaded 6.03ppm). The other three sites Baringo, Kitambasie and

Nzambani had the non-invaded areas having a high concentration of copper than the invaded (Figure 4.4).There was inconsistency in the concentration of copper in the

Lantana invaded areas across all the study sites.

Iron concentration in the Lantana invaded areas was higher in three sites Kitambasie

(invaded 30.10ppm and non-invaded 18.27ppm), Nzambani (invaded 27.01ppm and non-invaded 16.07ppm and Isooni (invaded 20.26ppm and non-invaded 16.21ppm) the remaining two sites the non-invaded areas had higher concentration of iron than the Lantana invaded areas (Figure 4.4). There was inconsistency in the concentration of iron in the Lantana invaded areas across all the study sites.

The concentration of zinc in the Lantana invaded areas was higher in three sites

Kitambasie (invaded 3.02ppm and non-invaded 1.90ppm), Nzambani (invaded

7.20ppm and non-invaded 7.01ppm) and Park Headquarter (invaded 6.68ppm and non-invaded 4.27ppm) compared to the other two sites where the non-invaded areas had higher concentration of zinc than the Lantana invaded areas (Figure 4.4).There was inconsistency in the concentration of zinc in the Lantana invaded areas across all the study sites.

Sodium content in the Lantana invaded areas was higher in one sites Headquarter

(invaded 0.53 me% and non-invaded 0.36 me%) compared to the other three sites where the non-invaded areas had higher concentration of sodium than the Lantana invaded areas (Figure 4.4). Concentration of sodium in Lantana invaded and non- invaded were the same Lantana invaded 0.37 me% and non-invaded 0.37 me% at

Baringo. There was consistency in the concentration of sodium in the Lantana invaded areas across all the study sites ranging.

4.5 Soil Texture Characteristics.

Results of soil texture analysis using hydrometer method obtained in the five study site (Figure 4.5). The main component of soil in all the study sites was clay. Isooni and Park Headquarter had the same type of soil; clay both in the areas invaded by

Lantana and non-invaded areas (Figure 4.5). Nzambani had clay soil in the invaded areas and clay-loam in the non-invaded (Figure 4.5). Baringo had sand-clay-loam soil both in the invaded and non-invaded areas. Kitambasie had sand-clay in the invaded areas and clay-loam soil in areas not invaded by Lantana (Figure 4.5).

Invaded Non-invaded

Figure 4.5: Soil types percentage values for L. camara in invaded and non-invaded areas 71

CHAPTER FIVE

DISCUSSION, CONCLUSIONS AND RECOMMENDATIONS

5.1 Discussion

From the results obtained in this study invasive plant species, Lantana camara changes physical and chemical attributes of soils including pH, soil nutrients and soil organic matter. The soil pH and most of nutrients varied significantly (P-value <0.05) among the study sites except for

Calcium and Nitrogen (P-value >0.05). Soil pH value was higher in the soils invaded by L. camara during this study. This is consistent with the results reported by Osunkonya et al. (2010), following L. camara invasion in South Eastern Queensland, Australia, Fan et al. (2010) due to

Lantana invasion in China and Sharma and Raghubanshi (2009) following Lantana invasion in

India. However, both increases and decreases in soil pH have also been reported following plant invasions (Kourtev et al., 2002). Previous studies have also shown that plant invasion significantly increase soil pH (Chen et al., 2012). High soil pH is known to accelerate litter decomposition and thus plays a crucial role in regulating nutrient availability (Simba et al.,

2013). These results may be attributed to the invasive plants having high nitrate uptake rates, which increase soil pH because decrease in soil nitrates are known to increase soil pH (Ehrenfeld et al., 2001). Weidenhamer and Callaway (2010) stated that soil pH has a major role in controlling nutrient bio-availability.

There was significant difference (P-value <0.05) in total nitrogen during both seasons but the concentration was higher in the Lantana invaded areas during dry season than in non-invaded areas. This is in agreement with (Rout and Callaway, 2009) who have reported that many plant invasions, despite dramatic decreases in local diversity, appear to increase local soil nitrogen 72 pools and total ecosystem nitrogen stocks. High nitrogen availability favours invasive plants, and low nitrogen availability favours native plants (Laungani and Knops, 2009). Also Dassonville et al. (2011) reported that invasive plants could create a micro-environment that facilitates its further invasion by increasing nitrogen availability in soil ecosystem through changed community structure and metabolic activities of soil micro-organisms which contribute to

Nitrogen cycles like nitrogen fixation and nitrifying. Also studies of L. camara invasion in India correlates with increases in available soil nitrogen, ammo-nitrification and nitrogen mineralization, which in turn correlates with high nitrogen, low lignin: nitrogen ratios and low carbon: nitrogen ratios in litter of L. camara (Sharma and Rhaghubanshi, 2009).

The L. camara invaded areas in study sites had leaf litter under the canopies slightly more than

2.5cm this is probably because, as reported by Simba et al. (2013) L. camara being a highly branched species with a lot of leaf biomass released its leaf tissues with time and accumulated a lot of litter within its understory approximately more than 2.5cm where there is 100% invasion and where there is 30-50% invasion the litter is about less than 2.5cm. There was high concentration total nitrogen in Lantana invaded areas in the study sites. This can be attributed to

Lantana leaf litter. The concentration of litter nitrogen of invasive plants is generally often higher than that of native plants (Liao et al., 2008). The findings of this study are also in agreement with Mandal and Guantum (2014) who reported that increase in nitrogen and phosphorus levels with increase in Lantana intensity could be due to decrease in nutrient impounding followed by the displacement of native species or reduction in their recruitment and growth rates. In this study there were high concentrations of nitrogen and phosphorous in

Lantana invaded areas. It has also been reported by Sharma and Raghubanshi (2009) that L. 73 camara architecture promotes accumulation of litter under the shrub, resulting to build up of organic carbon and nitrogen and carbon and nitrogen levels were elevated in the Lantana invaded areas in comparison to the non-invaded in the study sites during this study.

Concentration of total organic carbon, zinc, copper, phosphorous, potassium, sodium, magnesium, manganese and iron varied significantly (P-value <0.05) among the study sites during wet and dry season. The nutrient that was highest in concentration during wet season was calcium followed by phosphorous, sodium, manganese and potassium in the Lantana invaded.

During the dry season the concentration of nutrients varied significantly (P-value <0.05) among the study sites. As stated earlier Lantana has a lot of leaf biomass that falls off the plant and form a lot of leaf litter. According to Wang et al. (2011) who studied ecological effects of invasive plants on soil nitrogen cycles reported that leaf litter of invasive plants undergoes high rate of decomposition accompanied by the release of nutrients and formation of soil organic matter which are important processes in ecosystem nutrient cycles, carbon changes and humus formation. This could be the reason as to why there was increase in the concentrations of phosphorous, calcium, sodium, manganese and potassium in the Lantana invaded areas during wet season. Also during the wet season there is a substantial amount of water in the soil compared to the dry season. Presence of organic matter in the soil during wet season will influence water retention and combination of organic matter from Lantana and water availability will speed up the rate of decomposition and this will also contribute to the increase of nutrients in the soil.

These results are also in agreement with Simba et al. (2013) findings on L. camara at Nairobi

National Park who reported that Lantana leaf litter on decomposition release cations into the soil within its root rhizosphere and subsequently increased the concentrations of magnesium, calcium and potassium. The accumulated organic matter also mulches the soil surface under the invasive plant species hence inhibiting leaching of nutrients from the soil surface (Simba et al., 2013).

According to the findings of this study there was high concentrations of K were found in

Lantana invaded areas, this concur with Basumatary and Bordolo (1992) who found out that a layer of organic matter increases the retention of potassium in the soil. In the current study high soil pH was found in the Lantana invaded areas and according to Simba et al. (2013) high soil pH accelerates litter decomposition. The results are also in agreement with Osunkonya et al.

(2010) study on L. camara who reported that soil within Lantana patches had greater air dried water content, higher organic and total organic carbon, higher exchangeable calcium, and higher pH than soils from adjacent vegetation lacking the weed. Sharma and Raghubanshi (2011) also reported higher concentrations of magnesium, calcium and potassium following Lantana camara invasion in Southern India.

There was significant difference (P-value <0.05) in phosphorus across all the study sites and during dry and wet season. Areas invaded by Lantana had high concentration of P compared to the non-invaded. These findings are in agreement with Martin et al. (2009) who stated that higher soil phosphorus often is correlated with invasion. This could be attributed to high soil pH which was found in the current study and is in agreement with Hinsinger (2001) who reported that phosphorous is generally most available to plants when the soil pH is between 6.0 and 6.5 and it is more soluble at higher pH levels. When the soil pH is less than 6.0 the potential for 75 phosphorous deficiency increases for most plants hence phosphate ions readily precipitate with metal cations, forming a range of Phosphorous minerals. Also according to Mandal and Guantum

(2014) large amount of litter dropped beneath Lantana is probably responsible for the elevated phosphorous levels.

There was significant difference (P-value <0.05) in zinc, iron, copper magnesium and manganese across all the study sites and the L. camara invaded and non-invaded areas. Sodium showed no significant difference in the invaded and non-invaded areas. Manganese and iron were in high concentration in Lantana invaded areas while magnesium, copper and zinc were in high concentration in the non-invaded areas. The concentration of sodium was the same in both invaded and non-invaded areas. This results are contrary to Osunkonya et al. (2010) who reported that there was lower soil copper, iron and manganese recorded within L. camara infestations than from non-infested soils; salt concentrations in the form of total chlorine, total and exchangeable sodium were also lower within L. camara patches. These results are not in agreement with Osunkonya et al. (2010) who reported that presence of Lantana camara actually decreases the concentrations of iron copper and manganese. In the current study the concentration of manganese and iron were found to be in high in Lantana invaded areas but for copper the results are in agreement with Osunkonya et al. (2010). According to Liptzin and

Silver (2009) high levels of iron and copper in the are known to decrease microbial activity and availability of free nutrients for plant root uptake generating oxidative stress hence negatively affecting plant growth. This could be a characteristic adapted by the invasive plant Lantana to increase the concentration of iron as obtained in the results so that it can colonise the habitats of native plants eliminating them. According to Li et al. (2013) soil macro-nutrients form a large 76 proportion of soil quality indicators therefore their measurements are essential in biodiversity conservation that’s why there soil concentrations were measured. There is no documented data on the soil pH and nutrient composition of Ol-Donyo Sabuk National Park before the Park was invaded by L. camara to be used for comparison with the current study.

There was consistency in the concentrations of sodium, magnesium, calcium and organic carbon in the study sites while pH and the other nutrients showed inconsistency in concentrations. Due to Lantana invasion this consistency can be expected. According to Osunkonya et al. (2010) study on Lantana invasion under increased pH, the soil exchange complex is no longer controlled by hydrogen ions, a process that might make mineral ions easily accessible for plant function and growth. Across all the five study sites the results showed an increase in the concentration of soil nutrients and pH in the Lantana invaded areas. This consistency whether in the invaded or non-invaded could mean that indeed Lantana changes the soil nutrients and this is what is accelerating its invasion in OL-Donyo Sabuk National Park.

This study did not find major difference in soil texture of invaded and non-invaded areas. The soils consisted more of clay than silt and sand particles. Lantana invasion was restricted to specific areas with specific soil texture; either clay or clay -loam soil texture during both the wet and dry season. Clay tightly binds soil water than sand and silt and has more sites for cations hence more rich in nutrients than sand (Simba et al., 2013). When L. camara invades a habitat it changes the soil texture to suit its survival and affects the native plant species (Ehrenfeld, 2003).

It does so by changing the mico-habitat. Some native plants like East African Sandlewood –

Osyris Lanceolata Hochst and Steud, Rhus natalensis Krauss and Pterolobium stellatum 77

(Forssk.) Brenan; their population in the Lantana invaded sites was between counts of 3-5 in an area of 10m by 10m. In some studies it was found that soil texture is a useful indicator of soil permeability, soil water retention capacity, and soil capacity to retain cations and influences plant available moisture and plant available nutrients (White, 1997). Clay content has also been considered as an index of nutrient availability (Scholes and Walke, 1993) because when Lantana invades an area it changes the nutrients in the soil and the texture. Lantana changes the soil texture to clay or clay loam for more moisture retention for its use because it is a short rooted plant which maximises use of moisture on top layers of the soil from which soil samples were collected.

5.2 Conclusion

This study provides evidence that Lantana invasion is changing the soil properties and negatively affecting ecosystem processes in Ol-Donyo Sabuk National Park and possibly the nearby areas. The results show a change in the soil minerals of most of the Lantana invaded areas of the Park increasing its invasion and this could give room to other invasive plants into the

Park and this could affect the wildlife and in-directly affect tourism which is a source of foreign exchange in Kenya. The invasive success of Lantana depends on soil properties like soil pH, total organic carbon, total nitrogen, phosphorus, magnesium, calcium, phosphorous, texture of soil and water availability but anthropogenic factors also contribute to Lantana invasion. Soils are the main source of nutrients for plant species hence these findings are important for the Park management in the effective conservation of the habitat and its bio-diversity.

5.3 Recommendations

 Soil samples from invaded and non-invaded areas to be collected and propagules in their soil

seed bank determined through germination trials to aid in restoration programmes and

determine the soil seed bank affected by L. camara invasion.

 More research to investigate if L. camara leaf litter causes changes in the composition of the

soil in the Park.

 Further studies should be carried to determine whether other factors like different tree species

are affecting the soil properties.

REFERENCES

Achhireddy, N. and Singh, M. (1984). Allelopathic Effects of Lantana (Lantana camara) on

milkweed vine (Morrenia odorata). Weed Science , 32: 757-761.

Allen, S., Grimshaw, Η., Parkinson, J., and Quarnby, C. (1974). Chemical Analysis of Ecological

Materials. London: Blackwell: Oxford.

An, M., Pratley, J. E. and Haig, T. (1998). Allelopathy: From Concept to Reality.Report on 9th

Australian Agronomy Confrence. Wagga Wagga: Charles Sturt University.

Atkinson, I. and Atkinson, T. (2000). Land Vertebrates as Invasive Species on the Islands of the

South Pacific Regional Environment Programme. Invasive Species in the Pacific: A

Technical Review and Draft Regional Strategy , 19-84.

Bais, H., Vepachedu, R., Gilroy, S., Callaway, R. and Vivanco, J. (2003). Allelopathy and Exotic

Plant Invasion From Molecules and Genes to Species Interactions . Science , 301:1377–

1380.

Basumatary, A. and Bordolo, I. P. (1992). Forms of Potassium in Some Soils of Assam in

Relation to Soil Properties. Journal Indian Society of Soil Science , 40(3): 443–446.

Batianoff, G. and Butler, D. (2003). Impact Assessment and Analysis of Sixty-Six Priority

Invasive weeds in South-East Queensland. Plant Protection Quartery , 18: 11-17.

Batten, K. M., Scow, K. M., Davies, K. F. and Harrison, S. P. (2006). Two Invasive Plants Alter

Soil Microbial Community Composition in Serpentine Grasslands. Biological Invasions ,

8: 217–230.

Bever, J., Dickie, I., Facelli, E., Facelli, J., Klironomos, J. and Moora, M. (2010). Rooting

Theories of Plant Community Ecology in Microbial Interactions. Trends in Ecology and

Evolution , 25: 468-478. 80

Bhagwat, S., Breman, E., Thekaekara, T. and Thornton, T. (2012). PLoS ONE. A Battle Lost?

Report of Two Centuries of Invasion and Management of Lantana camara L. in

Australia, India and South Africa , 7(3): e32407.

Bharath, S., Siddhartha, K., Ankila, J. and Gladwin, J. (2012). Ecology and Impacts of the

Invasive Species, Lantana camara, in a Social-Ecological System in South India. Human

Ecology , 40: 931–942.

Bremner, J. M. (1996). Total Nitrogen. (D. Sparks, A. Page, P. Helmke, R. Loeppert, P.

Soltanpour, M. Tabatabai, Eds, Methods of Soil Analysis- Chemical Methods , Part 3:

1085-1122.

Brocque, A., Case, H. and Dearnaley, J. (2013). Response of Soil Fungal Richness and

Composition to Lantana camara L. Infestation in the Toowoomba Region, South-East

Queensland, Australia. Australasian Mycologist , 31: 17-23.

Bromilow, C. (1995). Problem Plants of South Africa. Arcadia: Briza Publications.

Brooks, M., D’Antonio, C., Richardson, D., Grace, J. and Pyke, D. (2004). Effects of Invasive

Alien Plants on Fire. Bioscience , 54: 677-688.

Callaway, R. and Reinhart, K. (2006). Soil Biota and Invasive Plants. New Phytology , 170: 445-

457.

Callaway, R. and Ridenou, R. (2004). Novel Weapons; Invasive Success and the Evolution of

Increased Competitive Ability. Frontiers in Ecology and the Enviroment , 2: 436-443.

Carlsson, N. B. (2004). Invading Herbivory: the Golden Apple Snail Alters Ecosystem

Functioning in Asian wetlands. Ecology , 85: 1575–1580.

Chapin, F. Z. (2000). Consequences of Changing Biodiversity. Nature , 405: 234–242. 81

Charles, H. and Dukes, J. (2007). Impacts of Invasive Species on Ecosystem Services. (W.

Nentwig, Ed.) Ecological Studies , 193: 217-237.

Charles, M. and Simmons, M. (1986). Methods for the Determination of Total Carbon in Soils

and Sediments. A Review. Analyst , 111: 385-390.

Chebures, J. B. (1983). Ecological Studies Investigation plan Ol-Donyo Sabuk National Park.

Nairobi: Kenya Wildlife Service.

Chen, T., Liu, W., Zhang, C. and Wang, J. (2012). Effects of Solidago canadensis Invadation on

Dynamics of Native Plant Communities and their Mechanisms. China Journal of Ecology

, 36: 253–261.

Choyal, R. and Sharma, S. (2011). Evaluation of Allelopathic Effects of Lantana camara L. on

Regeneration of Pogonatum Alloides in Culture Media. Asian Journal of Plant Science

and Research , 1(3): 41–48.

Cilliers, C. (1983). The Weed, Lantana camara L and the Insect Natural Enemies Imported for

its Biological Control in South Africa. Journal of the Entomological Society of Southern

Africa , 46:131-138.

Colautti, R. I. and MacIsaac, H. J. (2004). A Neutral Terminology to Define Invasive Species.

Diversity and Distribution, 10: 135-141.

Corlett, R. (1998). Frugivory and Seed Dispersal by Vertebrates in the Oriental (Indomalayan)

Region. Biological Reviews of the Cambridge Philosophical Society , 413-448.

Dassonville, N., Guillaumaud, N., Piola, F., Meerts, P. and Poly, F. (2011). Niche Construction

by the Invasive Asian Knotweeds (Species Complex Fallopia): Impact on Activity,

Abundance and Community Structure of Denitrifiers and Nitifiers. Biological Invasions ,

13: 1115-1133. 82

Davis, M., Grime, J. and Thompson, K. (2000). Fluctuating Resources in Plant Communities: A

General Theory of Invasibility. Journal Ecology , 88: 528-538.

Day, M., Wiley, C., J., P. and Zalucki, M. (2003). Lantana: Current Management, Status and

Future Prospects. . Australian Centre for International Agricultural Research , 5: 1-20.

Drake, S., Weltzin, J. and Parr, P. (2003). Assessment of Non-native Invasive Plant Species on

the United States Department of Energy Oak Ridge National Environmental Research

Park. Castanea , 68(1): 15-30.

Dukes, J. and Mooney, H. (2004). Disruption of Ecosystem Processes in Western North America

by Invasive Species. Nature , 77: 411–437.

Ehrenfeld, J. (2003). Effects of Exotic Plant Invasions on Soil Nutrient Cycling Processes.

Ecosystems , 6: 503-523.

Ehrenfeld, J. G., Kourtev, P. and Huang, W. (2001). Changes in Soil Functions Following

Invasions of Exotic Understory Plants in Deciduous Forests. Ecological Applications ,

11: 1287–1300.

Elisante, F. and Ndakidemi, P. (2014). Allelopathic Effect of Datura stramonium on the Survival

of Grass and Legume Species in the Conservation Areas. American Journal of Research

Communication , 2(1), 2 (1): 27-43.

Englberger, K. (2009). A Guide for Identification and Public Awareness. Invasive weeds of

Pohnpe: Conservation Society of Pohnpei , 29.

Everham., E. B. and Brokaw (1996). Forest Damage and Recovery from Catastrophic Wind.

Botany Revision , 62: 114-185. 83

Ewell, D.J., O., Bergelson, J., Daehler, C., D’Antonio, C. and Gomez, L . (1999). Deliberate

Introductions of Species: Research Needs-Benefits can be Reaped, but Risks are High.

Bioscience , 4: 619-630.

Faden, R. B. (1974). Commelinaceae. Upland Kenya Wild Flowers. (A. D. Agnew, Ed.) New

York: Oxford University Press.

Fan, L., Chen, Y., Yuan, J., Yang, Z. (2010). The Effect of Lantana Invasion on the Soil

Chemical, Microbial Properties and Plant Biomass accumulationin Southern China.

Geoderma , 154: 370-378.

Farrell, G., Kibata, G. and Sutherland, J. (1995). A Review of Crop Protection Research in

Kenya. Crop Protection Project , 1: 165.

GISP. (2003, May 21). The Invasive Alien Species Problem. Retrieved October 10, 2015, from

The Global Invasive Species Programe: http//www.org/about/AIS.asp

Gordon, R. D. (1998). Effects of Invasive, Non-indigenous Plants on Ecosystem Processses.

Ecological Application , 975-989.

Graaff, J. L. (1986). Lantana camara, the Plant and Some Methods for its Control. South African

Forestry Journal , 136: 26-30.

Grayston, S., Wang, S., Campbell, C. and Edwards, A. (1996). Selective Influence of Plant

Species on Microbial Diversity in the Rhizosphere. Soil Biology and Bio-chemistry , 30:

369-378.

Harley, K. (1973). Biological Control of Lantana in Australia. Wapshere[edited], Proceedings,

III International Symposium on Biological Control of Weeds (pp. 23-29). United

Kingdom: Commonwealth Institute of Biological Control, Farnham Royal. 84

Hawkes, C., Wren, I., Herman, D. and Firestone, M. (2005). Plant Invasion Alters Nitrogen

Cycling by Modifying the Soil Nitrifying Community. Ecology Letters , 8: 976–985.

Heleno, R., Ramos, J. and Memmott, J. ( 2013a ). Integration of Exotic Seeds into an Azorean

Seed Dispersal Network. Biological Invasion , 15: 1143–1154.

Hinsinger, P. (2001). Bioavailability of Soil Inorganic Phosphorous in the Rhizosphere as

Affected by Root-induced Chemical Changes. Soil Science and Plant Nutrition , 237:

173-195. .

Huenneke, l., Hamburg, S., Koide, R., Mooney, H. and Vitousek, P. (1990). Effects of Soil

Resourses on Plant Invasion and Community Structure in Carlifornia USA Serpentine

Grassland. Ecology , 71: 478-491.

Islam, K. A. (2001). Deforestation Effects on Vegetration and Soil Quality in Tropical Semi-

evergreen Degraded and Protected Forests of Bangladesh. Land Degradation and

Development , 12: 45-56.

IUCN. (2004, May 21). IUCN-the World Conservation Union Species Survival Commission.

Retrieved November 12, 2015, from Invasive Species Specialist Group:

http://www.issg.org/database/spp

Jones, R., Sharitz, R., Dixon, P., Segal, D. and Schneider, R. (1994). Woody Flant Regeneration

in Four Flood Plain Forests. Ecology , 64: 345-367.

Kedera, C. K. (2005). Invasive Alien Species in Kenya: Status and Management. Retrieved

October 6, 2015, from Food and Agricultural Organization of the United Nations:

Http://www.fao.org/docrep/008/y5968e10.htm

Kolar, C. L. (2001). "Progress in Invasion Biology: Predicting Invaders". Trends in Ecology and

Evolution , 16 (4): 199–204. 85

Kourtev, P., Ehrenfeld, J. and Haggblom, M. (2002). Exotic Plant Species Alter the Microbial

Community Structure and Function in the Soil. Ecology , 83: 3152-3166.

Kulmatiski, A., Beard, K., Stevens, J. and Cobbold, S. (2008). Plant–Soil Feedbacks: A Meta-

analytical Review. Ecology Letters , 11: 980–992.

Kumar, S. and Rohatgi, N. (1999). The Role of Invasive Weeds in Changing Floristic Diversity.

Annals of Forestry , 7(1): 147-150.

Kumar, S., K.M, S., Karthick, S., Mythili, S. and Sathiavela, A. (2011). Lantana Invasion: An

Overview Weed Biology. Asian Journal of Plant Science and Research , 1 (1): 48–56.

KWS. (1999). Fourteen falls/Oldonyo Sabuk Integrated Management Plan. Nairobi: Kenya

Wildlife Service.

Laungani, R. and Knops, J. (2009). Species-Driven Changes in Nitrogen Cycling Can Provide a

Mechanism for Plant Invasions. Proceedings of the Natural Academy of Sciences of the

United States of America, (pp. 106: 12400-12405).

Levine, J., Vila, M., D’Antonio, C., Dukes, J., Grigulis, K. and Lavorel, S. (2003). Mechanism

underlying the Impacts of Exotic Plant Invasions. Ecology , 270: 775–781.

Li, P., Zhang, T. and Wang, Y. D. (2013). Development of Biological Soil Quality Indicator

System for Sub-Tropical China. Soil and Tillage Research , 126: 112-118.

Liao, C., Peng, R., Luo, Y., Zhou, X., Wu, X. and Fang, C. (2008). Altered Ecosystem Carbon

and Nitrogen Cycles by Plant Invasion. New Phytologist , 177: 706-714.

Liptzin, D. and Silver, W. (2009). Effects of Carbon Additions on Iron Reduction and

Phosphorus Availability in Humid Tropical Forest Soils. Soil Biology and Bio-Chemistry,

41: 1696–1702. 86

Lockwood, J., Hoopes, M. and Marchetti, M. (2007). Invasion Ecology. United Kingdom:

Blackwell Publishing Ltd: Oxford.

Lwando, C. (2009). Effect of Lantana camara on Plant Diversity in Zambia. WWF- Education

for Nature Program , 1-4.

Lyons, E. E. (1999). Preliminary Survey of Invasive Species in Eastern Africa. Invasive species

in Eastern Africa (pp. 65-70). Nairobi: International Centre of Insect Physiology and

Ecology (ICIPE).

Mack, M. and D’Antonio, C. (1998). Impacts of Biological Invasions on Disturbance Regimes.

Trends Ecolological Evolution , 13: 195–198.

Mack, R., Simberloff, D., Lonsdale, W., Evans, H., Clout, M. and Bazzaz, F. (2000). Biotic

Invasions: Causes, Epidemiology, Global Consequences, and Control. Ecological

Applications, 10: 689-710.

Mandal and Gautam. (2014). The Role of Soil Nutrient Availability on the Invasion

Establishment and Carbon Sequestration Potential of an Invasive Shrub from Doon

Valley, India. Journal of Advanced Botany and Zoology , 14: 8.

Marler, M., Zabinski, C. and Callaway, R. (1999). Mycorrhizae Indirectly Enhance Competitive

Effects of an Invasive Forb on a Native Bunchgrass. Ecology , 80: 1180–1186.

Martin, M. R., Tipping, P. W. and Sickman, J. O. (2009). Invasion by an Exotic Tree Alters

Above and Below Ground Ecosystem Components. Biological invasions , 11: 1883–

1894.

Mehlich, A. (1953). Determination of P, Ca, Mg, K, Na and NH4. North Carolina Soil Test

Division , 2: 31-36. 87

Mersie, W. and Singh, M. (1987). Allelopathic Effect of Lantana on Some Agronomic Crops

and Weeds. Plant and Soil , 98: 25-30.

Mishra, A. and Singh, R. (2010). Comparative Study of Effect of Lantana camara Extract of

Different Parts on Seed Germination of Parthenium hysterophorus. International Journal

of Plant Science , 5 (1): 74–75.

Moody, K., Monroe, C., Lubigan, R. and Paller, E. (1984). Major Weeds of the Phillipines. Los

Banos: University of Phillipines Laguna.

Morton. (1994). Lantana, or Red Sage (Lantana camara L. Vergenaceae), Notorious Weed and

Popular Garden Flower; Some Cases of Poisoning in Florida. Economic Botany , 48(3):

259-270.

Munyua, S. J., Karitu, M., Kimoro, T. P., Kiptoon, J. E. and Buoro, I. B. (1990). A Note on

Clinical-Pathological Findings and Serum Enzyme Activity in Sheep, Goats and Friesian

Calves with Acute Lantana camara Poisoning. Bulletin of Animal Health and Production

in Africa , 38(3): 275-279.

Nannipieri, P., Asche, J., Ceccherini, M., Landi, L., G, P. and Renella, G. (2003). Microbial

Diversity and Soil Functions. European Journal of Soil Science , 54: 655-670.

Neena, P., and Joshi, P. K. (2013). A Review of Lantana camara Studies in India. International

Journal of Scientific and Research Publications , 3: 4.

Ngoru, B. A., M., R., Charles, L., Joseph, M., Reuben, M. and Samuel, M. (2007). Ol-Donyo

Sabuk National Park Biological Resource Inventory. Nairobi: Kenya Wildlife Service.

Osunkoya, O. and Perret, C. (2010). Lantana camara L. (Verbenaceae) Invasion Effects on Soil

Physicochemical Properties. Biological Fertility Soils , 47: 349-355.

Parsons, W. and Cuthbertson, E. (2001). Noxious weeds of Australia. Collingwood: CSIRO 88

Ragubanshi, A. (1992). Effect of Topography on Selected Soil Properties and Nitrogen

Mineralization in a Dry Tropical Forest. Soil Bio-Chemistry , 145-150.

Rawat, Y., Bhatt, Y., Pande, P. and Singh, S. (1994). Production and Nutrient Cycling in

Arundinaria Falcata and Lantana camara. The Two Converted Ecosystems in Central

Himalaya-Tropical Ecology , 35(1): 53-67.

Reichard, S. and Hamilton, C. (1997). "Predicting Invasions of Woody Plants Introduced into

North America". Conservation Biology, 11 (1): 193–203.

Reinhart, K. and Callaway, R. (2006). Soil Biota and Invasive Plants. New Phytology , 170: 445-

457.

Richardson, D. P. (2000). Naturalization and Invasion of Alien Plants: Concepts and Definitions.

Diversity and Distributions , 6:93–107.

Rose, S., Perry, D., Pilz, D. and Schoeneberger, M. (1983). Allelopathic Effects of Litter on the

Growth and Colonization of Mycorrhizal Fungi. Journal of Chemical Ecology , 9: 1153–

1162.

Rosswall. (1976). The Internal Nitrogen Cycle between Microorganisms, Vegetation and

Soil.Ecology Bulletin, 1: 157-167.

Rout, M. E. and Callaway, R. (2012). Interactions Between Exotic Invasive Plants and Soil

Microbes in the Rhizosphere Suggest that "Everything is not Everywhere". Annals of

Botany , 1-10.

Rout, M. E., Callaway, R. M., Holben, W. E., Chrzanowski, T. H., Westlie, T. K. and DeLuca, T.

H. (2013). Bacterial Endophytes Enhance Competition by Invasive Plants. American

Journal of Botany , 100.

Rout, M. and Callaway, R. (2009). An Invasive Plant Paradox. Science , 324: 724–725. 89

Scholes, R. and Walke, B. H. (1993). Synthesis of the Nysvley Study. An African Savanna , 1:

33-77.

Sharma, G., Raghubanshi, A. and Singh, J. (2005). Lantana invasion: An Overview. Weed

Biology Management , 5: 157-167.

Sharma, G., and Raghubanshi, A. (2009). Lantana Invasion Alters Soil Nitrogen Pool and

Processes in the Tropical Dry Deciduous Forest of India. Applied Soil Ecology , 42: 134-

140.

Sharma, G. and Raghubanshi, A. (2011, July 8). Lantana Camara L. Invasion and Impact on

Herb Layer Diversity and Soil Properties in a Dry Deciduous Forest of India. Applied

Ecology and Environmental Research , 3: 253-264.

Sharma., G. R. (2007). Effect of Lantana camara L. Cover on Plant Species Depletion in the

Vindhyan Tropical Dry Deciduous Forest of India. Applied Ecology and Environmental

Research , 5(1): 109-121.

Shaukat, S. and Siddiqui, I. (2001). Lantana camara in the Soil Changes the Fungal Community

Structure and Reduces Impact of Meloidogyne javanica on mungbean. Phytopathology

Meditteranea , 40: 245–252.

Shea, K. and Chesson, P. (2002). Community Ecology Theory as a Framework for Biological

Invasions. Trends Ecology and Evolution , 17(4): 170-176.

Simba, Y. R., Kamweya, A. M., Mwangi, P. N. and Ochora, J. M. (2013). Impact of the Invasive

shrub, Lantana camara L. on Soil Properties in Nairobi National Park, Kenya.

International Journal of Biodiversity and Conservation , 5 (12), 803-809.

Sinha, B. and Sharma, A. (1995). Chromosome composition of Lantana camara L. karyotype,

Basic Number and DNA Diversity. Nucleus Calcutta , 38(1-2): 16-22. 90

State of New York Department, T. D. (2007, April 4). Geotechnical Engineering Bureau.

Retrieved August 13th, 2015, from GTM-24:

http//www.dot.ny.gov/divisions/engineering/technical-services/technical-services-

respository/GTM-24b.pdf

Stinson, K., Campbell, S., Powell, J., Wolfe, B., Callaway, R. and Thelen, G. (2006). Invasive

Plant Suppresses the Growth of Native Tree Seedlings by Disrupting Below Ground

Mutualisms. PloS Biology , 4: 727-731.

Stohlgren, T., D. Binkley, G. C., Bull, K., Y. Otsuki, G. N., Bashkin, M. and Son, a. Y. (1999).

Exotic Plant Species Invade Hot Spots of Native Plant Diversity. Ecological Monographs

, 25-46.

Thakur, M., Ahmad, M. and Thakur, R. (1992). Lantana Weed and its Possible Management

Through Natural Insect Pests in India. Indian Forester , 118: 466–488.

Thebaud, C., Finzi, A., Affre, L., Debussche, M. and Escarre, J. (1996). "Assessing Why Two

Introduced Conyza Differ in Their Ability to Invade Mediterranean Old Fields". Ecology

, 77(3): 791–804.

Tilman, D. (1982). Resource Competition and Community Structure. Princeton University

press.Princeton

Toth, S. and Prince, A. L. (1949). Estimation of Carbon Exchange Capacity and Exchangeable

Ca, K and Na Contents of Soils by Flame Photometric Techniques. Soil Science , 67: 439

– 445.

Turton., S. D. (1992). Light Environments and Floristic Composition Across an Open Forest-

Rain Forest Boundary in Northeastern Queensland. Australian Journal of Ecology , 17:

415-423. 91

Van der Heidjen, M., Bardgett, R. and Van Straalen, N. (2008). The Unseen Majority. Soil

Microbes as Drivers of Plant Diversity and Productivity in Terrestrial Ecosystems , 11:

296–310.

Verdcourt, B. (1992). Flora of Tropical East Africa. Verbenaceae , 37-47.

Vitousek, P. (1990). Biological Invasions and Ecosystem Processes: Towards an Integration of

Population Biology and Ecosystem Studies. Oikos , 57: 7-13.

Vivrette, N. and Muller, C. (1997). Mechanism of Invasion and Dominance of Coastal

Grassland. Ecogical Monograph , 47: 301–318.

Vogelsang, K. and Bever, J. (2009). Mycorrhizal Densities Decline in Association with Non-

native Plants and Contribute to Plant Invasion. Ecology , 90: 399–407.

Walters, M. R. (1996). Are Shade Tolerance Survival and Growth Linked? Low Light and

Nitrogen Effects on Hardwood Seedlings. Ecology , 77: 841-853.

Walton, C. (2006). Lantana camara (shrub). Retrieved December 27th, 2015, from Global

Invasive Species Database: Global Invasive Species Databas

http://www.issg.org/database/species/ecology.asp?si=56andfr=1andsts=sss

Wambua, J. (2012). The Distribution, Abundance and Ecological Impacts of Invasive Plant

Species at Oldonyo Sabuk National Park. Nairobi: University of Nairobi.

Wang, C., Han, G., Jia, Y., Feng, X., Guo, P. and Tian, X. (2011). Response of Litter

Decomposition and Related Soil Enzyme Activities to Different Forms of Nitrogen

Fertilization in a Sub-tropical Forest. Ecological Research , 26: 505-513.

Weidenhamer, J. and Callaway, R. (2010). Direct and Indirect Effects of Invasive Plants on Soil

Chemistry and Ecosystem Function. Chemical Ecology , 36: 59–69. 92

Wellington, Z. R. and Amel, A. (2004). Poisonous Weed Prominent in Rangelands and

Grassland Areas. Lantana and Hagonoy , 1: 3-4.

Westover, K., Kennedy, A. and Kelley, S. (1997). Patterns of Rhizosphere Microbial Community

Structure Associated with Co-occurring Plant Species. Journal of Ecology , 85: 863–873.

White, R. E. (1997). Principles and Practices of Soil Sciences. London: Blackwell: Oxford.

Zavaleta, E. (2000). The Economic Value of Controlling an Invasive Shrub. Ambio. , 29: 462–

467.

APPENDICES

APPENDIX I: SOIL TYPES FOR THE FIVE STUDY SITES

Soil Texture % Sand Clay Silt Grade

Site Kitambasie Invaded 50 36 14 SC Non-invaded 42 38 20 CL Site Nzambani Invaded 36 44 20 C Non-invaded 42 38 20 CL Site Park Headquaters Invaded 38 46 16 C Non-invaded 26 56 18 C Site Baringo Invaded 56 28 16 SCL Non-invaded 50 34 16 SCL Site Isooni Invaded 26 56 18 C Non-invaded 26 54 20 C

Key: SCL----Sand clay loam SC----Sand clay CL----Clay loam C----Clay

Two-Way ANOVA of Comparison of pH and nutrients of the 5 study sites during the wet season

Treatment Soil pH N % Organic P me% K me% Ca Mg Mn Cu ppm Fe ppm Zn ppm Na C % me% me% me% me% Site Baringo 7.23±0.06 0.19±0.0 1.89±0.07 14.51±4.2 1.34±0.0 5.41±0.0 2.59±0.0 1.04±0.0 9.88±0.64 13.41±0.8 20.45±1.9 0.37±0.0 a+ 1a a 3c 3c 4a 1b 1b b 1d 5a 0d Isooni 5.94±0.00 0.20±0.0 1.62±0.28 27.51±1.1 7.60±0.1 2.56±0.0 2.56±0.0 1.35±0.1 10.25±0.4 16.21±0.9 13.85±1.5 0.56±0.0 d 1a a 2a 3b 2a 2b 3b 8b 0b 0b 1b Kitambasie 6.38±0.13 0.13±0.0 1.10±0.06 12.72±1.2 5.52±0.3 58.37±3 4.98±0.0 0.30±0.0 7.14±0.16 21.75±1.5 2.42±0.24 1.74±0.2 b 1d b 4d 6a 9a 3a 3d d 7b e 7a Nzambani 6.12±0.04 0.18±0.0 1.83±0.05 17.55±3.3 0.49±0.0 3.09±0.2 2.53±0.0 0.55±0.0 11.56±0.4 16.54±0.2 7.85±0.38 0.23±0.0 cd 0b a 6b 4d 7a 5b 7c 7a 1b c 3e Park HQ 6.30±0.19 0.16±0.0 1.59±0.07 27.49±3.3 1.60±0.2 6.29±0.6 2.45±0.0 1.05±0.0 8.27±2.03 15.57±2.3 4.27±0.55 0.45±0.0 cb 1c a 4a 7c 2a 2c 5b c 0c d 4c Treatment Invaded 6.48±0.14 0.16±0.0 1.49±0.12 21.11±2.7 2.49±0.5 25.4±16. 3.00±0.2 0.94±0.1 8.18±0.74 16.48±1.4 8.75±1.20 0.82±0.2 a 1a a 1a 3a 5a 6a 1a a 1a a 0a Non- 6.30±0.13 0.18±0.0 1.73±0.10 18.81±2.1 1.89±0.4 6.87±1.0 3.05±0.2 0.77±0.1 10.66±0.5 16.92±0.6 10.79±2.3 0.51±0.0 invaded b 1b b 0b 2b 8a 6b 0b 0b 1b 1b 9b P values Site <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 0.1503 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 Treatment 0.0051 <0.0001 0.0020 <0.0001 <0.0001 0.2412 0.0045 <0.0001 <0.0001 0.0007 <0.0001 <0.0001 Site* <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 0.3120 0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 Treatment +Mean ±SD value followed by the same letter(s) within the same column do not differ significantly from one another (Two-way

ANOVA, Tukey HSD-test, α = 0.05), ppm stands for Parts per million while me stands for miliequivalent.

Two-Way ANOVA of Comparison of pH and nutrients of the 5 study sites during the dry season

Treatment Soil pH Total N% Organic C P me% K me% Ca me% Mg me% Mn me% Cu ppm Fe ppm Zn ppm Na me% % Site Baringo 6.68± 0.19a+ 0.21±0.00a 2.11±0.05a 27.04±1.6b 1.11±0.08d 5.01±0.15d 2.54±0.33b 0.92±0.04c 10.63±0.30b 15.79±0.26d 9.51±2.94a 0.37±0.00c

Isooni 6.21± 0.13b 0.28±0.08a 1.94±0.10b 30.01±0.0a 1.61±0.18b 6.18±0.51b 2.47±0.02c 1.31±0.15b 11.09±0.11a 20.26±0.91b 7.01±1.56b 0.47±0.05b

Kitambasie 6.33±0.12b 0.49±0.21a 1.48±0.11d 22.34±5.3c 2.63±0.97a 8.95±2.43a 3.71±0.57a 0.60±0.16d 8.56±0.48d 26.62±3.74a 2.49±0.0.26e 1.32±0.45a

Nzambani 6.86±0.34a 0.20±0.01a 1.97±0.11b 7.02±1.35e 1.46±0.48c 5.69±1.44c 2.48±0.07c 0.91±0.23c 9.83±1.21c 26.54±4.68a 6.35±0.29d 0.45±0.13b

Park HQ 6.88±0.45a 0.17±0.02a 1.75±0.14c 12.51±3.6d 0.64±0.16e 5.51±0.27c 2.45±0.02c 1.07±0.04b 10.56±1.00b 16.76±1.82c 6.68±0.53c 0.45±0.04b

Treatment Invaded 6.88±0.19a 0.36±0.09a 1.98±0.06a 20.76±3.58a 1.08±0.21b 5.65±0.49b 2.43±0.02b 1.03±0.06a 9.17±0.35b 24.97±2.61a 4.62±0.51b 0.46±0.04b

Non- 6.30±0.13b 0.18±0.01b 1.72±0.09b 18.81± 2.10b 1.89±0.41a 6.87±1.08a 3.03±0.26a 0.84±0.14b 11.09±0.51a 17.72±0.53b 8.20±1.29a 0.77±0.21a invaded P-values Site <0.0001 0.0862 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 Treatment <0.0001 0.0222 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001

Site* <0.0001 0.0462 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 Treatment

+Mean ±SD value followed by the same letter(s) within the same column do not differ significantly from one another (Two-way ANOVA, Tukey HSD-test, α =

0.05), ppm

Three-Way ANOVA of interaction between season, sites and invaded and non-invaded areas

Treatment Soil pH Total N Organic C P K Ca Mg Mn Cu Fe Zn Na % % me% me% me% me% me% ppm ppm Ppm me%

Season

Dry 6.59±0.13a 0.27±0.05a 1.85±0.06a 19.78±2.05a 1.49±0.24a 6.26±0.60aa 2.73±0.14a 0.96± 0.07a 10.13±0.35a 21.19±1.46a 6.41±0.76a 0.61±0.11a Wet 6.39±0.09b 0.17±0.01b 1.61±0.08b 20.00± 1.70b 2.19±0.33b 16.15±8.34a 3.02±0.18b 0.86± 0.08b 9.42 ±0.51b 16.70±0.77b 9.77±1.32b 0.67± 0.11b Site

Baringo 6.95±0.12a 0.20±0.00ab 2.00±0.05aa 20.77±2.84b 1.23±0.05cc 5.21±0.10bb 2.56±0.02b 0.98±0.03c 10.25±0.36b 14.60±0.54e 14.98±2.31a 0.37±0.00d

Isooni 6.07±0.07d 0.24±0.04ab 1.78±0.15bb 28.76± 0.65a 1.80±0.10b 6.89±0.33bb 2.51±0.02cc 1.33±0.10a 10.67±0.27a 18.23±0.86c 10.43±1.40b 0.52±0.03b Kitambasi 6.36±0.08c 0.31±0.12a 1.29±0.08c 17.53±3.07d 4.07±0.66a 33.66±20.32a 4.34±0.33a 0.45±0.09e 7.85±0.32d 24.18±2.07a 2.46±0.17e 1.53±0.26a Nzambani 6.48±0.20b 0.19±0.01b 1.90±0.06a 12.28±2.35e 0.97±0.27d 4.37±0.80b 2.50±0.04c 0.73±0.13d 10.69±0.67aa 21.54±2.70b 7.10±0.32c 0.34±0.07e Park HQ 6.60±0.25 bb 0.17±0.01b 1.67±0.08b 20.00±3.19c 1.12±0.21c 5.90±0.34bb 2.45±0.06d 1.06±0.03b 9.41±1.13c 16.17±1.43d 5.48±0.51d 0.45±0.02c

Treatment Invaded 6.68±0.12a 0.26±0.05a 1.73±0.08aa 20.93±2.21a 1.79±0.31a 15.54±8.35aa 2.71±0.14a 0.99±0.06a 8.68 ±0.41a 20.57±1.65a 6.69±0.75a 0.64±0.11aa Non- 6.30±0.09b 0.18±0.01b 1.72±0.07a 20.93±2.21b 1.89±0.29b 6.87±0.75a 3.04±0.18b 0.83±0.08b 10.88±0.37b 17.32±0.42b 9.50±1.34b 0.64±0.12a invaded P-values

Season <0.0001 0.0111 <0.0001 0.0004 <0.0001 0.2054 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 Site <0.0001 0.1476 <0.0001 <0.0001 <0.0001 0.0927 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 Treatment <0.0001 0.0271 0.8377 <0.0001 0.0037 0.2659 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 0.8931

Site*seaso <0.0001 0.0228 0.0163 <0.0001 <0.0001 0.1774 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 n <0.0001 Season*Tr <0.0001 0.0111 <0.0001 0.0004 <0.0001 0.2054 <0.0001 0.0508 <0.0001 <0.0001 <0.0001 <0.0001 Site*Tr <0.0001 0.0401 <0.0001 <0.0001 <0.0001 0.4411 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 Site * Sea <0.0001 0.0228 0.0163 <0.0001 <0.0001 0.1774 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 <0.0001 son* Tr +Mean ±SD value followed by the same letter(s) within the same column do not differ significantly from one another (Three-way ANOVA, Tukey HSD-test, α = 0.05), ppm stands for Parts per million, me stands for miliequivalent while Tr stands for treatment.

Variation of pH and nutrients across sites in the Lantana invaded and non-invaded areas

SITE Treatment Soil pH Total N Organi P K me Ca me Mg me Mn Cu ppm Fe ppm Zn Na % c C% me% % % % me ppm me% % Baringo Invaded 6.81±0. 0.19±0. 1.95±0. 17.54± 1.17±0. 5.11±0.1 2.52±0. 0.95±0. 9.20±0. 13.99±1 9.51±2. 0.37±0. 24 01 11 5.61 11 9 02 05 34 .06 94 00 Non- 7.09±0. 0.21±0. 2.05±0. 24.01± 1.28±0. 5.31±0.0 2.61±0. 1.02±0. 11.30±0 15.21±0 20.45± 0.37±0. invaded 20 00 00 0.00 00 0 01 00 .00 .00 1.95 00 Isooni Invaded 6.21±0. 0.26±0. 1.39±0. 27.51± 1.60±0. 6.48±0.6 2.47±0. 1.31±0. 11.09±0 20.26±0 7.01±1. 0.47±0. 12 09 21 1.12 18 4 02 15 .11 .92 56 05 Non- 5.94±0. 0.22±0. 2.16±0. 30.01± 2.01±0. 7.30±0.0 2.56±0. 1.35±0. 10.25±0 16.21±0 13.85± 0.56±0. invaded 00 00 00 0.00 00 0 02 13 .48 .90 1.50 01 Kitambasi Invaded 6.21±0. 0.47±0. 1.34±0. 25.09± 3.37±1. 52.94±4 3.70±0. 0.65±0. 8.21±0. 30.10±2 3.02±0. 1.32±0. e 07 22 17 4.30 30 0.85 57 14 64 .18 05 45 Non- 6.50±0. 0.14±0. 1.24±0. 9.97±0. 4.79±0. 14.38±0. 4.99±0. 0.25±0. 7.50±0. 18.27±0 1.90±0. 1.74±0. invaded 14 00 01 02 00 13 00 01 02 .25 00 27 Nzambani Invaded 6.85±0. 0.20±0. 2.07±0. 14.54± 1.55±0. 6.30±1.1 2.38±0. 1.06±0. 8.85±0. 27.01±4 7.20±0. 0.51±0. 34 01 06 4.71 43 6 03 16 79 .47 67 10 Non- 6.11±0. 0.17±0. 1.73±0. 10.03± 0.39±0. 2.47±0.0 2.63±0. 0.40±0. 12.54±0 16.07±0 7.01±0. 0.16±0. invaded 04 00 01 0.02 00b 2 02 00 .02 .06 04 00 Park HQ Invaded 7.31±0. 0.19±0. 1.90±0. 19.99± 1.24±0. 6.89±0.3 2.50±0. 0.96±0. 6.03±1. 11.51±0 6.68±0. 0.53±0. 26 01 07 6.70 43 5 00 01 03 .53 53 01 Non- 5.87±0. 0.14±0. 1.44±0. 20.02± 0.10±0. 4.91±0.0 2.40±0. 1.15±0. 12.80±0 20.82±0 4.27±0. 0.36±0. invaded 00 00 00 0.00 01 0 00 02 .03 .23 55 00 P-value Site <0.0001 0.1476 0.0001 <0.000 <0.000 0.0927 <0.0001 <0.000 <0.000 <0.0001 <0.0001 <0.000 <0.000 1 1 1 1 1 1 Treatment <0.0001 0.0271 0.8377 <0.000 0.0037 0.2659 <0.0001 <0.000 <0.000 <0.0001 <0.0001 <0.000 0.8931 1 1 1 1

+Mean ±SD value ,ppm stands for Parts per million and me stands for mili-equivalent. 98