Range condition assessment to document the extent of ...

Range condition assessment to document the extent of degradation on

selected semi-arid rangelands of the Eastern Cape, South Africa.

By

NDANDANI AKHONA

Dissertation submitted in fulfillment of the requirements for the degree of

Master of Science in Agriculture - Pasture Science

University of Fort Hare

Department of Livestock and Pasture Science

School of Agriculture and Agribusiness

Faculty of Science and Agriculture

University of Fort Hare

P/ Bag X1314

Alice

South Africa

Supervisor: Dr K Mopipi

Co-Supervisor: Prof S. T Beyene

i

DECLARATION

I, Akhona Ndandani declare that “Range condition assessment to document the extent of

degradation on the selected semi-arid rangelands of the Eastern Cape, South Africa” has not

been submitted to any University and that it is my original work conducted under the

supervision of Dr K. Mopipi and Prof. S.T Beyene. All assistance towards the production of

this work and all references contained herein have been duly accredited.

_________________________ _______________________

Miss Akhona Ndandani Date:

Approved as to style and content by:

Dr K. Mopipi (Supervisor)

Prof S.T Beyene (Co-supervisor)

ii

DEDICATION

I would like to dedicate this dissertation to my mother (Nosimo), father (Mawethu), siblings

(Lusanda, Sophakama and Lunga). Most importantly, my son Nikho (Nhinhi kamama) and

nephew Iyazi (Yaya). It is with great pleasure that I have these people in my life (Humbled).

Thank you to you all.

iii

ACKNOWLEDGEMENTS

First and foremost, I would like to thank the almighty God for the strength and courage to

finish my dissertation. His grace has been with me since the beginning. A sincere gratitude

also goes to my supervisors Dr K Mopipi and Prof S T Beyene. Dr Mopipi has managed to

support me fully and was patient enough during the hardships of this study. I extend my

gratitude to Mr Wellington Monwabisi Sibanga and Mr Mweli Nyanga for technical support.

Many thanks to the village leaders who approved conducting research in their villages. I give

much appreciation to the manager of the Reserve Dr Gavin Shaw for giving us permission to

conduct the study in the Reserve. I acknowledge the financial support that I got from received

from DST/NRF South African Research Chairs Initiative (SARCHI) co-hosted by the

University of Fort Hare as well as the Govan Mbeki Research and Developments Centre

(GMRDC) of the Universitry of Fort Hare (Project T358). I thank God for the sweet souls

who have been with me through and out data collection “My colleagues” namely: Yonela

Maziko, Ntomboxolo Mamayo, Thando Ntutha, Odwa Armstrong Ngcofe, Sive Tokozwayo,

Thabo Magandana, Siphamandla Huza and Sinethemba Matshawule.

Lastly, I would like to thank my family for allowing me such an opportunity and in the end

understanding the importance of learning and getting education. Not forgetting my friends for

their support and encouraging words when one felt like giving up.

iv

ABSTRACT

The assessment of communal rangelands and Nature Reserve capability is crucial in order to

prevent resource degradation and facilitate adaptive management practice. This study was

conducted to document the extent of land degradation in three selected semi-arid rangelands

of the Eastern Cape, South Africa. These comprised the Great Fish River Nature Reserve,

Glenmore and Ndwayana communal rangelands. Each rangeland was demarcated into two

homogenous vegetation units (HVU’s) toplands, bottomlands and a benchmark site.

Botanical composition (woody and herbaceous), aboveground biomass production, soil seed

bank composition and density and soil micro nutrients (Cu, Mn, Zn) and macro nutrients (N,

P, K, OC, Mg, Ca, Na) were determined.

Twenty two (22) perennial grass species and some forbs were recorded in all the HVU’s. In

general the grass species composition consisted of 59% pioneer (Increaser II) species, 36.4%

mesophytes (Decreaser) species and the remaining were 4.54% sub-climax/climax (Increaser

I) species. The grazing value of the grass species was: High 41%, Moderate 14% and Low

45%. Six dominant grass species and were recorded, comprising mainly of Increaser species

in all the HVU’s, (except for Digitaria eriantha). Biomass production in the benchmark

(2700 kg/ha) was significantly higher (p<0.05) in summer than all the other HVU’s, but in

winter (1715 kg/ha) it was not significantly different (p>0.05) from the bottomlands of the

Great Fish RNR. There was an increasing trend in mean basal cover from the benchmark to

Ndwayana toplands (0.0-15.75cm). The results showed that the benchmark had higher dense

cover (0.0 to 1.5cm) than all of the other HVU’s. There were 27 woody species, where 56%

were acceptable to browsers while 44% were not acceptable. Of these woody plants 41% had

thorns or spines whilst 59% had no thorns or spines. Ptaeroxylon obliquum (14%) was the

most dominant species and the least dominant being Pappea capensis (0.05%)

respectively.Glenmore had significantly higher (p<0.05) bush density (1181.25 and

1337.5Trees/ha) and TE (1069 TE/ha) than all the other HVU’s.

Soil samples from each sample plot were collected with an auger from a 20 cm layer with the

use of a 0.25m2 quadrat distributed within the four 100 m transects in each sample plot. The

samples were analyzed for N, P, K, OC, Na, Ca, Mg, Zn, Cu and Mn and pH using

photospectrometer. There were significant differences (p<0.05 in the concentration levels of

all the macro nutrients N, OC, P, K, Ca, and Na (except Mg) in different HVU’s. There were

v

significant differences (p<0.05) in the soil content of Cu, Mn, Zn and pH between the

different HVU’s.

In the soil seed bank experiment, a total of 21 species found (9 grasses, 9 forbs and 3 sedges).

Most of the grass species identified in the seed bank were mainly found in the Benchmark

site while the rest of the homogenous vegetation units were dominated by either forbs or

sedges. Seed bank grass composition comprised 67% perennial and 33% annual species. Of

these grasses, 29% were unpalatable, 48% of low, 14% high and 9% moderately palatability.

Pseudognaphalium undulataum (14.59%) was the most abundant species, followed by

Medicago laciniata (8.44%), Hypertelis bowkeriana (8.41%) and Sutera campulata (8.36%)

with Tragus species (0.23%) followed by Panicum stapfianum (0.5%) being the least

abundant species. There were significant differences (p<0.05) in the seed bank density

between the Great Fish RNR and the communal areas of Glenmore and Ndwayana (both

toplands and bottomlands). Similarities between the seed bank and the above ground

vegetation were tested using Sorensen’s Similarity Index. The coefficients were as follows;

Glenmore toplands (40%), Glenmore bottomlands (37.5%), Ndwayana toplands (25%),

Ndwayana bottomlands (28.57%), Great Fish RNR toplands and bottomlands were (0%) with

the benchmark comprising of (80%). Rangaland degradation is found in all the study sites

and it was more in the communal areas than in the Great Fish RNR excluding the benchmark.

Key words: Land degradation, rangeland condition, botanical composition, biomass

production, vegetation cover. Soil seed bank, seasonal, micro and macro nutrients

vi

Table of Contents

DECLARATION ............................................................................................................................... i

DEDICATION ................................................................................................................................. i

ACKNOWLEDGEMENTS ............................................................................................................... iii

ABSTRACT ................................................................................................................................... iv

LIST OF TABLES ........................................................................................................................... ix

LIST OF FIGURES ........................................................................................................................ xiii

LIST OF ABBREVIATIONS ............................................................................................................ xiv

CHAPTER 1. INTRODUCTION ......................................................................................................... 1

1.1 Background ............................................................................................................................. 1

1.2 Problem statement ................................................................................................................... 4

1.3 Justification ............................................................................................................................. 4

1.4 Objectives of the study ............................................................................................................ 5

1.5 Research questions .................................................................................................................. 5

References .............................................................................................................................................. 6

CHAPTER 2. LITERATURE REVIEW ...................................................................................... 10

2.1 Introduction ................................................................................................................................. 10

.......................................................................................................................................................... 13

2.2 Causes of land degradation ......................................................................................................... 14

2.3 Rangeland condition as an indicator of degradation. .................................................................. 15

2.5 Biomass production .................................................................................................................... 18

2.6 Basal cover .................................................................................................................................. 19

2.7 Soil quality and soil properties.................................................................................................... 20

2.9 Soil seed bank composition in rangelands .................................................................................. 22

2.8 Rationale for the study ................................................................................................................ 23

References ......................................................................................................................................... 25

CHAPTER 3. BOTANICAL COMPOSITION IN NDWAYANA, GLENMORE AND THE GREAT FISH RNR. . 37

ABSTRACT ..................................................................................................................................... 38

3.1 INTRODUCTION ...................................................................................................................... 39

vii

3.2 Description of the study sites ...................................................................................................... 41

3.3. Experimental layout ................................................................................................................... 44

3.4 Data collection ............................................................................................................................ 44

3.4.1 Determination of botanical composition and basal cover. ....................................................... 44

3.4.2 Determination of biomass production ...................................................................................... 46

3.4.4 Determination of the woody species composition ................................................................... 46

3.4.4 Statistical analysis .................................................................................................................... 46

3.5. RESULTS .................................................................................................................................. 47

3.5.1. Overall herbaceous species composition in the selected semi-arid rangelands. ..................... 47

3.5.2. Species abundances across Homogenous Vegetation Units ................................................... 52

3.5.3. Biomass production in summer and winter seasons. .............................................................. 54

3.5.4 Basal cover in different homogenous vegetation units. ........................................................... 56

3.5.5. Woody species abundances across Homogenous Vegetation Units. ...................................... 58

3.5.6. The dominant woody species at Glenmore, Ndwayana and the Great Fish RNR. .................. 59

3.5.7. Tree equivalents and bush density across homogenous vegetation units................................ 60

3.6. Discussion ....................................................................................................................................... 61

3.6.1 Species seasonal abundances across HVU’s ............................................................................ 61

3.6.2 Seasonal biomass production across the HVU’s. ..................................................................... 63

3.6.3 Basal cover in the toplands and bottomlands of Glenmore, Ndwayana and the Great Fish

RNR. ................................................................................................................................................. 64

3.6.5 Woody species composition in Glenmore, Ndwayana and the Great Fish RNR. .................... 66

4.7 Conclusion .................................................................................................................................. 68

References ......................................................................................................................................... 68

CHAPTER 4. SOIL CHEMICAL PROPERTIES IN GLENMORE, NDWAYANA AND THE GREAT FISH RIVER

NATURE RESERVE. ................................................................................................................................ 78

Abstract ............................................................................................................................................ 78

4.1.1 Introduction .............................................................................................................................. 79

4.1.2 Soil sampling in Glenmore, Ndwayana and the Great Fish RNR. ........................................... 80

4.1.3 Statistical analysis .................................................................................................................... 80

viii

4.2 Results ......................................................................................................................................... 80

4.2.1 Soil macro nutrient contents .................................................................................................... 80

2.2 Soil micro nutrient contents ........................................................................................................ 89

4.3 Discussion ................................................................................................................................... 91

4.3.1 Soil macronutrients across homogenous vegetation units ....................................................... 91

4.3.2 Soil micro nutrients and soil pH (KCL) across homogenous vegetation units. ....................... 94

4.4 Conclusion .................................................................................................................................. 96

References ......................................................................................... Error! Bookmark not defined.

CHAPTER 5. THE SEED BANK COMPOSITION AND DENSITY IN GLENMORE, NDWAYANA AND THE

GREAT FISH RIVER NATURE RESERVE. ............................................................................................... 102

Abstract .......................................................................................................................................... 102

5.1.1 Introduction ............................................................................................................................ 104

5.1.2 Data collection ......................................................................... Error! Bookmark not defined.

5.1.2.1 Determination of soil seed bank composition and plant density ......................................... 105

5.1.3 Statistical analysis .................................................................................................................. 106

5.2 RESULTS ................................................................................................................................. 107

5.2.1 Seed bank composition .......................................................................................................... 107

5.2.2 The abundances of dominant species in the soil seed bank ................................................... 109

5.2.3 Soil Seed bank density (plants/m2) ........................................................................................ 115

5.2.4 Comparison between soil seed bank composition and standing herbage composition. ......... 116

5.3 Discussion ................................................................................................................................. 118

5.3.1 Soil seed bank composition.................................................................................................... 118

5.3.2 The effect of homogenous vegetation units on the seed bank density ................................... 119

5.3.3 Comparison between the above ground vegetation and the seed bank composition ............. 119

5.4 Conclusions ............................................................................................................................... 120

References ....................................................................................................................................... 121

CHAPTER 6.GENERAL DISCUSSION AND CONCLUSIONS. .................................................................. 124

6.1 General discussion .................................................................................................................... 125

6.2 General conclusions .................................................................................................................. 127

ix

6.3 Recommendations ..................................................................................................................... 129

References ....................................................................................................................................... 130

APPENDICES ........................................................................................................................................ 132

Appendix A: Herbaceous and woody composition ......................................................................... 132

Appendix B: Soil properties and pH ............................................................................................... 137

Appendix C: Seed bank composition and density .......................................................................... 140

xiii

LIST OF TABLES

Table 3.1: Herbceous species composition in Glenmore, Ndwayana and the Great Fish RNR

.................................................................................................................................................. 51

Table 3.2: Mean abundance of the dominant species found in the homogenous vegetation

units .......................................................................................................................................... 53

Table 3.3: Mean (S.E) of biomass production in different homogenous vegetation units. ..... 55

Table 3.4: % abundance, acceptability and the availability of thorns/spines of the woody

species ...................................................................................................................................... 58

Table 3.5: Woody species abundances across HVUs .............................................................. 59

Table 3.5.6 Woody density and tree equivalents ..................................................................... 60

Table 4.1 : Soil macro nutrient status in Glenmore, Ndwayana and the Great Fish RNR. ..... 88

Table 4.2: Soil micro nutrient status of Glenmore, Ndwayana and the Great Fish RNR. ....... 90

Table 5.1: Overall mean abundances of the soil seed bank composition in the selected semi-

arid rangelands. ...................................................................................................................... 108

Table 5.2: Mean (S.E) abundances of the dominant species in the soil seed bank . .............. 114

xiv

LIST OF FIGURES

Figure 2.1: Land degradation index of South Africa ............................................................... 14

Figure 3.1: Mean basal cover of all the homogenous vegetation units. ................................... 56

Figure 3.2: Mean basal cover of season in all the homogenous vegetation units. ................... 57

Figure 5.1: Effect of seedbank density on the homogenous vegetation units of Glenmore,

Ndwayana and the Great Fish RNR ....................................................................................... 115

Figure 5.2: Comparison between above ground vegetation and the seed bank composition.116

xiv

LIST OF ABBREVIATIONS

HVUs Homogenous Vegetation Units

SAS Statistical Analysis System

GLM General Linear Model

S.E Standard Error

LBM Lowest Browsable material

TE Tree Equivalents

DTPA Diethylenetriamenepentaacetic

GREAT FISH RNR Great Fish River Nature Reserve

GRBOT Great Fish River Nature Reserve Bottomlands

GRTOP Great Fish River Nature Reserve Toplands

GLENBOT Glenmore bottomlands

GLENTOP Glenmore toplands

NDWABOT Ndwayana bottomlands

NDWATOP Ndwayana toplands

1

CHAPTER 1. INTRODUCTIOn

1.1 Background

Land degradation poses a serious threat to the natural resources and economic development

of South Africa (Hoffman et al., 1999; Hoffman and Todd, 2000). Approximately 91% of

South Africa is potentially susceptible to degradation (Hoffman and Ashwell, 2001).

Moreover, a large proportion of the population are dependent on the services derived from

dryland ecosystems for their livelihood. The sustainable land use of communal rangelands

depends on the understanding of the extent of land degradation, and how restoration of these

areas can be applied (Solomon et al., 2006). Most of the farmers working in communal areas

have underestimated the existing degradation problems (Meadows and Hoffman, 2003). The

biophysical and climatic environment appears crucial for any model of land degradation

(Hoffman and Todd, 2000). Rangeland degradation is not a spatially uniform process, there

are substantial side effects and some landscapes are more prone to land degradation than

others because they have erodible soils and palatable species, which results in more attraction

for grazing activities or both (Pickup, 1998). Land degradation has affected two billion

hectares of the world agricultural land, rangelands, forests and woodland (Al Dousari et al.,

2000). High extents of land degradation in an area are attributed to the disappearance of about

5-10 million hectares of agricultural land annually (Al Dousari et al., 2000). Dryland areas

are environmentally fragile and therefore more prone to degradation (Gao and Liu, 2010).

Hoffman and Todd (2000) characterized land degradation in South Africa into soil and

rangeland degradation. When soil and rangeland degradation were combined the extent of

degradation was mostly found on the steeply sloping environments along the eastern

escarpment, incorporating the communal areas. This was observed in the former Ciskei,

Transkei in the Eastern Cape and Kwa-Zulu homelands which were seen as the most

degraded areas all over South Africa (Hoffman and Todd, 2000). However, the extent of land

degradation differs with the management history of the farming areas (Lesoli, 2011). There

are severely degraded districts and these are commonly categorized by the communal land

tenure system and formed part of the former “homelands” of the apartheid state of South

Africa (Meadows and Hoffman, 2003). Additionally, the degree of land degradation also

varies with land ownership and practise. Consequently, if soil and rangeland degradation are

the main assessment criteria, largely communally farmed area of South Africa are perceived

2

to be significantly more degraded than commercial areas (Hoffman et al., 1999; Hoffman and

Todd, 2000).

With the identification of a structural, socio-political foundation to the land degradation

problem; the role of physical environmental factors on degradation should not be

underestimated (Lesoli, 2011). Hoffman et al. (1999) highlighted that the distribution of

communal and commercial agricultural land in South Africa is itself reinforced by physical

environmental circumstances. Commercial farms are likely to be found in areas characterized

by greater aridity and gentler slopes than the communal system. On the other hand, rural

South Africa dominated by communal land is subject to higher levels of land degradation

susceptibility because it is characterized by higher rainfall and steeper slopes (Meadows and

Hoffman, 2003). Land degradation has also been reported from other parts of the world and

the extent varies with biophysical socio- economic factors (Lesoli, 2011).

Knowledge obtained from secondary education and to some extent, tertiary education is used

by most South African farmers in managing commercial livestock and game ranches (Oztas

et al., 2003). South African agricultural research institutes have a long history of rangeland

management research and extension in commercial ranching areas. On the other hand,

communal livestock management has largely been based on traditional management systems

without the livestock owners having any formal training in animal husbandry or rangeland

management. Lack of education by communal farmers has long been considered a major

cause of the perceived mismanagement of communal rangelands (Behnke and Scoones 1992).

However, lack of education is not likely to be the main cause of rangeland degradation on

communal ranches due to several reasons. Communal farmers face a number of problems,

one being the fact that ranches are often managed by more than one manager (Smet and

Ward, 2004). Management by different managers on the same rangeland has been considered

in the “Tragedy of commons” (Hardin, 1968). The “Tragedy of commons states that”, “it is

more profitable for an individual to over stock the ‘commons’ because he derives the entire

benefit from each additional animal, but the cost is shared by all.” This has been said to be

the main cause of rangeland degradation on the communal ranches (Ellis and Swift 1988,

Ward et al., 2000). This phenomenon is proof that, the increase in the livestock population

will lead to a decrease in the rangelands ecological capacity and promote rangeland

degradation (Hardin, 1968).

Grazing is generally considered to be the most economical way of utilizing rangeland

vegetation and it is the most dominant use of rangeland resources in the communal areas of

3

the Eastern Cape (Lesoli, 2011). This is mainly because climatic, topographic and geological

factors limit crop production (de Wet and van Averbeke, 1995). Overgrazing or uncontrolled

grazing always reduces plant cover that protects the soil and generally results in soil erosion

and compaction (Oztas et al., 2003). The factors that affect runoff and erosion are a

consequence of complex interactions of vegetation and soil characteristics (Thurow et al.,

1986). The occurrence of soil erosion varies widely at different rates over the landscape

(Foster, 1988). The differences in soil formation result in significant differences in soil

properties (Brubaker et al., 1993), plant production (Jones et al., 1989) and vegetation

(Bragg, 1978). Changes in soil properties and vegetation can also be altered over time under

different land uses, management systems and soil erosion (Oztas et al., 2003). Biodiversity is

reduced and the biomass production is the lowest on communal areas compared to

commercial farming areas (Fabricius, 1997). Land degradation is one of the main limiting

ecological factors in the communal areas of the Eastern Cape (Trollope and Coetzee, 1975).

Vegetation in terms of species composition, soil cover and standing biomass production is

indicative of the potential primary productivity and soil protection of the rangelands (Oztas et

al., 2003). Biological complexity and diversity, essential components for sustainable

production of rangeland ecosystems require maintenance of a wide range of vegetation and

various habitats within a production system (Snyman, 1998). Sustainability in communal

rangeland resource utilization, management and conservation requires the responsibility of all

the stakeholders (Lesoli, 2011). Attainment of sufficient information about a particular

rangeland vegetation variation and distribution between vegetation types and local landscapes

would make a difference in the sustainability processes of these rangelands.

Plants establish themselves by the expansion and subsequent fragmentation of vegetative

parts such as tillers, rhizomes or runners, or by the successful establishment of a soil seed

bank or bulbils (Freedman et al., 1982). Soil seed banks are important in rangeland

ecosystems where grasses will count as a large part of the vegetation and their role is

threefold (Solomon et al. 2006). Firstly, it is a potential pool of propagules for regeneration

of grasses after disturbance (Snyman, 1998; Laura and Brenda, 2000). Secondly, the seed

banks may reduce the probability of population extinction of plants (Venable and Brown,

1988). Lastly, it is likely to be the major source in establishing aboveground plant

communities following environmental changes such as rainfall (Wilson et al., 1993; Hayatt,

1999). High grazing pressure by livestock introduce a disturbance to rangelands, which can

4

negatively affect the size and composition of grasses in the seed bank, both in space and time

(Solomon, 2003; Snyman, 2004b).

1.2 Problem statement

Land degradation is a socio-economic problem that leads to the reduction of livestock

numbers and to the loss of land for agricultural purposes. Land degradation therefore poses a

serious threat to the livelihoods of people relying on these rangelands. The extent of land

degradation results in declining functional capacity, increased poverty, and food insecurity

(Cohen et al. 2006). Major changes caused by land degradation in rangeland surface

morphology and soil characteristics have a drastic effect on the primary productivity of the

rangeland ecosystem, and in turn on livestock production (Payton et al., 1992). Land

degradation results in the loss of vegetation cover, promoting soil nudity, soil erosion, poor

water infiltration, water runoff, reduced forage productivity for the animals due to the

increase in the less palatable species such as the woody species and soil erosion. There are a

number of factors responsible for degradation; among others are climate, grazing (Arnalds

and Barkarson 2003), soil quality, and landform and its influence on rangeland ecosystem

hydrology (Garcia-Aguirre et al., 2007) needs to be addressed. This study focused on

documenting land degradation on communal rangelands and the impacts it has over the

natural vegetation and food security for livestock production.

1.3 Justification

Rangelands are very important to the communal and commercial farmers and their

sustainability through evaluating the extent of degradation leads to a better understanding of

the causative factors. Moreover, it, provides recommendations on how to reduce land

degradation in the selected semi-arid rangelands of the Eastern Cape. A better understanding

of the causes of land degradation promoted an increase in the functional capacity, reduce

poverty and ensure food security to both communal and commercial farmers.

Recommendations would contribute towards improved vegetation cover and forage

productivity and reduce soil erosion, soil nudity, water runoff, and poor water infiltration.

The importance of these rangelands is based on the fact that they are the major grazing

resource for livestock and for crop production. Assessing these rangelands will helps address

the issue of land degradation for sustainable land use. Sustainable conservation and

5

utilization of the remaining dryland vegetation resources and rehabilitation of those that have

already been degraded provided economic, social and ecological benefits. There is

insufficient scientific documentation of the extent of degradation in many communal

rangelands of South Africa, especially in the Eastern Cape Province. The study will provide

recommendations on how to rehabilitate and sustain these rangelands.

1.4 Objectives of the study

The main objectives are:

To document the extent of land degradation and seasonal variation on the selected

semi-arid rangelands of the Eastern Cape.

The specific objectives are:

To conduct a full veld condition assessment and determine botanical composition,

basal cover and biomass production in Ndwayana, Glenmore and the Great Fish

River Nature Reserve..

To determine the soil properties in the selected rangelands of the Eastern Cape.

To determine the soil seed bank composition and density of the areas under study.

1.5 Research questions

What is rangeland condition in terms of species composition, vegetation cover,

biomass production, soil seed bank composition and density in communal rangelands

of Glenmore, Ndwayana and Great Fish RNR?

What seasonal variations occur relative to the abundances of palatable/acceptable

grass species? Do the palatable species disappear with season as land degradation

increases?

How does the soil seed bank composition compare with that of the standing

vegetation in the study areas?

What impact does land degradation have on soil macro and micro nutrients?

6

References

Al Dousari A M, Misak R and Shahid S 2000. Soil compaction and sealing in Al-Salmi area,

Western Kuwait. Land Degradation and Development 11: 401- 418.

Arnalds O and Barkarson B H 2003. Soil erosion and land use policy in Iceland in relation to

sheep grazing and government subsidies. Environmental Science and Policy, 6:105-

113.

Behnke R H and Scoones I 1992.Rethinking Range Ecology Implications for Rangeland

Management in Africa. International Institute for Environment and Development

(IIED), London.

Bragg T B 1978. Effect of burning cattle grazing and topography on vegetation of the choppy

sand range site in the Nebraska sandhills prairie. Proceedings of the First International

Rangeland Congress, Society for Range Management, Denver, Colorado, USA, pp.

248–253.

Brubaker S C, Jones A J, Frank K, Lewis D T 1993. Regression models for estimating soil

properties by landscape position. Soil Science Society of America Journal 58, 1763–

1767.

De Wet C and van Averbeke W 1995. Regional overview of land reform related issues in the

Eastern Cape Province. ARDRI/ BRC/ISER. University of Fort Hare, Alice and

Rhodes University, Grahamstown.

Ellis JE and Swift DM 1988. Stability of African pastoral ecosystems: alternate paradigms

and implications for development. Journal of Range Management 41: 450–459.

Fabricius C A 1997. The impact of land use on biodiversity in Xeric Succulent Thicket,

South Africa. Bulletin of the Southern African Institute of Ecologists and

Environmental Scientists 16 (2): 15- 17.

7

Foster G R 1988. Modeling soil erosion and sediment yield. In: R. Lal (Ed.), Soil Erosion

Research Methods. Soil & Water Conservation Service, Soil and Water Conservation

Society, Ankeny, Iowa, USA, pp. 97–117.

Freedman B, Hill N, Henry G, Svoboda J 1982.Seed banks and seedling occurrence in a

high Arctic oasis at Alexandra Fjord, Ellesmere Island, Canada. Canadian Journal of

Botany 60, 2112– 2118.

Gao J and Liu Y 2010. Determination of land degradation causes in Tongyu County,

Northeast China via land cover change detection. International Journal of Applied

Earth Observation and Geoinformation. 12: 9-16.

Garcia-Aguirre M C, Ortiz M A, Zamorano J J and Reyes Y 2007. Vegetation and land

relationships at Ajusco volcano Mexico, using a geographic information system

(GIS).Forest Ecology and Management 239:1-12.

Hardin, G 1968. The tragedy of the commons.Science162: 1243–1248.

Hayatt, L.A., 1999. Differences between seed bank composition and field recruitment in a

temperate zone deciduous forest. American Middle land Naturalist 142, 31– 38.

Hoffman, T and Todd, S 2000. National review of land degradation in South Africa: The

influence of biophysical and socio-economic factors. Journal of Southern African

studies 26:743-758.

Hoffman M T and Ashwell A 2001.Nature Divided: Land degradation in South Africa: the

influence of biophysical and socio-economic factors. Journal of Southern African

studies, 26, 743-758.

Hoffman T , Todd S, Ntoshona Z and Turner S 1999.Land degradation in South Africa, (pp

245) Cape Town National Botanical Institute.

8

Jones A J, Mielke L N, Bartles C A and Miller C A 1989.Relationship of landscape position

and properties to crop production. Journal of Soil and Water Conservation 44, 328–

332.

Laura A H and Brenda B C 2000. Seed bank formation during early secondary succession in

a temperate deciduous forest. Journal of Ecology 88, 516– 527.

Lesoli M 2011. Characterization of communal rangeland degradation and evaluation of

vegetation restoration techniques in the Eastern Cape, South Africa. PhD. Thesis,

University of Fort Hare, Alice, South Africa.

Meadows M E and Hoffman T M 2003. Land degradation and climate change in South

Africa. The Geographical Journal 169 (2): 168-177.

Oztas T, Koc A, and Comakli B 2003.Changes in vegetation and soil properties along a slope

on overgrazed and eroded rangelands. Journal of Arid Environments 55: 93- 100.

Payton R W, Christiansson C, Shishira E K, Yanda P and Eriksson M G 1992. Landform,and

erosion in the north eastern Iringi Hills, Kondoa, Tanzania. Geografiska. Series A,

Physical Geography. 74 (2): 65-79.

Pickup G, Bastin G N and Chewings V H 1998.Identifying trends in land degradation in non-

equilibrium rangelands. Journal of Applied Ecology 35: 365- 377.

Smet M and Ward D 2005. A comparison of the effects of different rangeland management

systems on plant species composition, diversity and vegetation structure in a Semi-

Arid Savanna. African Journal of Range and Forage Science 22 (1): 59- 71.

Snyman H A 1998. Dynamics and sustainability of the rangeland ecosystem in an arid and

semi-arid climate of southern Africa. Journal of Arid Environments 39, 655– 666.

Snyman H A 2004b. Short-term influence of fire on seedling establishment in a semi-arid

grassland of South Africa. South African Journal of Botany 70, 215– 226.

9

Solomon T B, Snyman H A and Smit G N 2006. Soil seed bank characteristics in relation to

land use systems and distance from water in a semi-arid rangeland of southern

Ethiopia. South African Journal of Botany 72: 263- 271.

Solomon T B 2003. Rangeland evaluation and perceptions of the pastoralists in the Borana

Zone of Southern Ethiopia. PhD thesis, University of the Free State, Bloemfontein,

South Africa, p. 329.

Thurow T L, Blackburn W H, Taylor CA 1986. Hydrological characteristics of vegetation

types as affected by livestock grazing systems. Edwards, Plateau Texas. Journal of

Range Management 39 (6), 505–509.

Tongway D J, Sparrow A D, Friedel M H 2003. Degradation and recovery process in arid

grazing lands of central Australia: Part 1. Soil and land resources. Journal of Arid

Environments 55: 301- 326.

Trollope W S W and Coetzee P G F 1975. Vegetation and veld management. In: Laker MC

(ed) The Agricultural Potential of the Ciskei: A preliminary report. pp. 71-124.

Faculty of Agriculture, University of Fort Hare, Alice.

Venable D L, Brown .S 1988. The selective interactions of dispersal, dormancy, and seed size

as adaptations for reducing risk in variable environments. American Naturalist 131,

360– 384.

Ward D, Ngairorue B T, Karamata J, Kapofi I, Samuels R and Ofran Y 2000. Effects of

communal pastoralism on vegetation and soil in a semi-arid and in an arid region of

Namibia. Proceedings IAVS Symposium 1998: 344–347.

Wilson S D, Moore D K J, Keddy P.A 1993. Relationships of marsh seed banks of vegetation

patterns along environmental gradients. Freshwater Biology 29, 361–370.

10

CHAPTER 2. LITERATURE REVIEW

2.1 Introduction

There is no single, readily identifiable definition for land degradation but all of them describe

how one or more of the land resources (soil, vegetation, water, rocks, air) has changed from

better to worse (Stocking and Murnaghan, 2001). The Food and Agriculture Organisation of

the United Nations cited by Stocking and Murnaghan (2001), states that land degradation is a

temporal or permanent decline in the productive capacity of land. World meteorological

Organisation (WMO) (2005) defined land degradation as the reduction or loss in arid, semi-

arid and dry sub-humid areas, of the biological or economic productivity and complexity of

rain fed cropland, irrigated cropland or range, pasture, forest and woodlands as a result of

land uses or from processes or combination of processes. The processes arising from human

activities and habitation patterns such as the following:

a) Soil erosion caused by wind or water

b) Deterioration of the physical, chemical and biological or economical properties of the

soil and lastly

c) Long term loss of natural vegetation

Hoffman and Todd (2000) stated that vegetation degradation through long term reduction in

biomass is also a form of land degradation. However, it is acknowledged that vegetation

degradation is not easily recognisable (Hoffman and Todd, 2000). The changes in vegetation

are revealed gradually, sometimes not in terms of biomass decrease in an area but through the

loss of species diversity, increase in invasive species and reduction of the woody species

(Kakembo et al., 2007 and Wessels et al., 2004).

South Africa has a long history of research into land degradation. Numerous publications

(Kokot, 1948; Acocks, 1953) official investigations and government and public intervention

schemes (Du Toit, 1991) have demonstrated the concern shown by South Africans towards

the issue (Hoffman and Todd, 2000). It is estimated that approximately 66% of rangelands in

South Africa are moderately or seriously degraded (Snyman, 1998). Similar vegetation

changes have been reported for arid and semi-arid rangelands throughout the world (Milton

and Hoffman, 1994). The result of such changes is a reduction in the grazing capacity of

11

natural rangelands for domestic livestock (Milton and Hoffman, 1994), which is comparable

to a loss in agricultural production.

Land-use change is estimated to remain the dominant driver of biodiversity loss in southern

Africa over the next century (Biggs et al., 2008). The main cause of biodiversity loss in the

arid and semi-arid rangelands of South Africa is land degradation (Scholes and Biggs, 2005)

and the levels of degradation may have been seriously underestimated (Rouget et al., 2006).

Local extinction of susceptible plant species may more readily be expected in semi-arid than

moist regions owing to the combination of droughts and sustained grazing (O’Connor, 1991).

In an inclusive review of land degradation in the arid rangelands of South Africa (Dean et al.,

1995a) it is clear that earlier work emphasized links of land degradation to less productive

rangeland and gave very little attention to implications of degradation for biodiversity.

Agricultural extension officers and resource conservation technicians have had an important

influence on other major assessments of land degradation in South Africa (Hoffman and

Ashwell, 2001). Even in some recent works on biodiversity assessment, land degradation has

been defined as land uses that lead to a persistent loss in ecosystem productivity (Scholes and

Biggs, 2005).

The term ‘land degradation’ is viewed differently by different stakeholders (Reynolds and

Stafford, 2002) and remains disreputably difficult to quantify. Even where site-based studies

have addressed the relationship between degradation and plant species assemblages a number

of these studies are compromised by use of sampling methods which are designed to include

common species but leave out many less common species (Reynolds and Stafford, 2002).

Many studies include relatively mild levels of degradation in which local extinction is either

not detected or only in the limited areas close to stock water points (Hoffman and Todd,

2001). An assessment of the impact of extreme levels of degradation on comprehensive plant

diversity, where extirpation of at least some species may be expected, is lacking across the

biomes of South Africa. The aim of this study is to quantify the floristic differences that we

could find between rangeland vegetation in ‘good condition’ and ‘severely degraded’ (Esler

et al., 2006) in the Albany Thicket Biome in communal rangelands and a Nature Reserve.

Attempts to establish which species undergo local extinction are important. This study will

also help to qualify the knowledge base of Increaser and Decreaser species response to

grazing since their classification can vary according to region or habitat (Sullivan and Rohde,

12

2002). It has been shown in Australia that a significant proportion of species respond

inconsistently to grazing and are context dependent (Vesk and Westoby, 2001).

13

14

Figure 2.1: The relative of soil degradation (top), vegetation degradation (middle) a

combined index of soil degradation (bottom) in South Africa as perceived by Agricultural

extension officers and resource conservation technicians (Hoffman et al., 2001)

2.2 Causes of land degradation

2.2.1 Natural processes and human activities

Land degradation is a natural process that may also be induced by human activities (Barrow,

2001). Biggs et al. (2008) argued that the relationship of humans with nature has in certain

instances increased the rate of land degradation and therefore undermined nature’s ability to

recover. A number of studies have been cited by Stocking and Murnaghan (2001) which

identify poor land management, inappropriate technology, overpopulation, poverty and

decisions of social and political structures as human factors associated with land degradation.

However, some studies have argued that land degradation can occur independently from

human activities. Todd and Gobena (2003) stated that natural land degradation processes are

slow and are often unnoticed.

2.2.2 Rain and soil erosion

Rainfall is regarded as the most crucial climatic factor in determining areas at risk of land

degradation by the World Meteorological Organisation (WMO), 2005. This is mainly due to

the vital role that rainfall plays towards development and distribution of plant life. The areas

having little or no vegetation, erosion is forced into the soil by the raindrops, surface and sub-

surface run-off and by river flooding (WMO, 2005). Climate change is also a factor that

clearly can increase the rate of land degradation through the alteration of spatial and temporal

patterns in temperature, rainfall and wind. Soil erosion by water is also recognised as a factor

that results to the land degradation problem worldwide (Stocking and Murnaghan, 2001.)

2.2.3 INVASIVE SPECIES

The increase in land degradation is also due to the increase of the invasive species with the

argument that in areas with a deep water-table invasive species gain better competitive

15

advantage in obtaining water to grow (due to their long tap roots) than the indigenous species

(that have a short root system) (Kakembo et al., 2007). Kakembo et al. (2007) also indicated

that factors influencing invasive environment are still a great challenge. The issue of land

degradation is very complex and an understanding of the problem requires a multi-faceted

approach. According to Stocking and Murnaghan (2001), the identification and analysis of

social factors that contribute to land degradation deserve particular attention because they

often set the stage for correcting actions and policies.

2.2.4 Overgrazing

Overgrazing of rangelands has often been mentioned as one of the major causes of land

degradation (Versbug and van Keulen, 1999). The grazing impacts on watershed properties

vary naturally over time due to the normal variability of climate, vegetation, intensity and

duration of livestock use (Blackmburn, 1983). Some of the concerns with livestock grazing in

the arid and semi-arid rangelands are the result of uneven grazing distribution (Bailey, 2004).

Cattle graze areas with gentle terrain and near water more heavily than rugged terrain or areas

far from water. Livestock directly affects plant species composition by grazing and the

trampling effect although the impacts vary with animal density and distribution (Belsky and

Bluementhal, 1997). The awareness of the importance of grazing and the grazing animals

should be increased in the dynamics of ecological systems. There is an increasing interest in

the role played by large herbivores in shaping and maintaining vegetation formation

(Schuman et al., 2002; Maki et al., 2007). The interrelationships between herbivores and

vegetation are more complex than many models recognised (Vernamkhasti et al., 1995).

They are mainly influenced by the behaviour and ecology of the herbivores and by the

ecological response of the different plant species to trampling and defoliation. It is therefore

generally perceived that land degradation in communal areas is caused by overgrazing

(Vernamkhasti et al., 1995).

2.3 Rangeland condition as an indicator of degradation.

Assessment of rangeland condition is very crucial to devise management practices (Rezaei et

al., 2006) and to estimate the extent of land degradation in the semi-arid rangelands. Mention

of the three tier system that involves consideration of the species composition, woody

16

components and soil properties has been indicated as a good assessment system (Friedel,

1991; Solomon et al., 2007). Different methods have been used to assess rangeland condition

namely the benchmark and ecological index, key species method, degradation gradient

method (Friedel, 1991). The benchmark method requires a comparison between the

benchmark site and the sample site (Friedel, 1991). The ecological index method suggests that

the weightings be given to each ecological group of grasses such as Decreasers and Increasers

(Van Oudtshoorn, 2006). The Decreaser species are found to be desirable than Increaser

species in the rangeland, and they decrease with the increase in the Increaser species as a

result of poor range management (either underutilization or over utilization) (Tainton, 1999).

They increase with proper range management. Increaser I species are less desirable and

increase with underutilization. The Increaser II species increase with over utilization

(Tainton, 1999). The index is not calculated based on the bench mark but at the end it is

compared to bench mark (Hurt and Bosch, 1991).

An assumption is that, different grazing regimes differ in species composition and grasses are

categorised into ecological groupings as whether Decreasers or Increasers/Invaders (Tainton,

1999). The interpretation for the benchmark method is reliant on ecological groupings as a

result it provides bias estimates. Noting that, species respond differently to grazing pressure

(Tainton, 1999). Moreover, climatic variation and fire regimes found in the benchmark may

differ to those of the sample sites. In the key species method acknowledges that the

distribution of other species is not grazing dependent. Rangeland condition is indicated by the

relative abundances of the key species in the sample site and the index helps to estimate the

grazing history of the rangeland (Smet and Ward, 2005). On the other hand, degradation

gradient and weighted key species methods provide measurements of the trend through site

positioning along gradient of degradation. These methods are suitable where vegetation in a

sample site is homogenous to minimise ecotypical differences in species. These two methods

are not reliant on ecological groupings but weightings are species based (Hurt and Bosch,

1991).

2.4 Species composition as an indicator of degradation

Botanical composition is one of the means of studying ecological changes in the development

of rangelands (Malan and Van Nierkerk, 2005). Any change in the grazing practice will cause

a change in plant species composition (Hayes and Holl, 2003). According to Oztas et al.

17

(2003) and Maki et al. (2007), any change in grazing pressure will result in a change in

vegetation structure, composition and productivity. The increase in the grazing pressure

results in the disappearance of the Decreaser species and they are replaced by the Increaser

or Invader species (Sisay and Baars, 2002). Decreaser species tend to decrease with over and

underutilization while Increaser II species are favored by overutilization (Kioko et al., 2012).

The replacement is results to the reduction of tuft size (Kioko et al., 2012) and a remarkable

decline in forage quality and quantity of the grasses (Retzer, 2006). However, Laughlin and

Abella (2007) indicated that the composition change is determined by rainfall than by grazing

pressure. In addition to that, the transition models emphasized that species composition

change from one state to the other due to unpredictable climatic variations (Hoffman and

Milton, 1994) and rainfall variation, competition between grass species are major

determinants of the differences on plant species composition in the rangelands. Moreover,

according to Fynn and O’Connor, 2000, rangelands with a high rainfall are predominated by

perennial plants and annual dominate in rangelands with a low rainfall. Noting that,

investigations to document land degradation should not only be based on anthropogenic

influences but should also consider environmental disturbances (Hoffman and Milton, 1994).

Species composition can be used as an indicator of rangeland condition because species vary

significantly in their acceptability and response to grazing (Abule et al., 2007). Herbivores

affect rangeland ecosystems directly through defoliation of vegetation and trampling (Lesoli,

2011). Physically the animals damage the plants by cutting, bruising and debarking. Certain

plants may be dislodged or uprooted during grazing. The trampling effect causes a change in

species composition, certain species are resistant while others are more vulnerable (Solomon

et al., 2007). There are positive effects of herbivores on vegetation such as plant distribution,

promotion of seed dispersal and soil nutrient cycling through excretion (Schuman et al.

2002). Certain plant species have different successional stages during grassland retrogression

and they can be used as indicators of the rangeland condition (Malan and Van Nierkerk,

2005). High and intense grazing leads to excessive removal of the most palatable species,

which are usually the perennial grasses (Todd and Hoffman, 1996; Anderson and Hoffman,

2006). This results in the establishment of the less palatable species which are the annuals

and forbs (Nsinamwa et al., 2005). The constant fading of the highly desirable species

(Malan and Van Nierkerk, 2005) can result in rangeland degradation.

18

2.5 Biomass production

Forage yield or biomass production generally refers to the above ground herbaceous material

(Lesoli, 2011). It is expressed as dry matter weight per area (Abule et al., 2007). Biomass

production is used to determine the amount of available forage for grazing animals, to

measure the effects of management on vegetation and to assess the rangeland condition

(Abule et al., 2007). Forage yield in rangelands may be described in terms of soil quality and

biomass production of the dominant species (Peden, 2005). The rangeland that produces

biomass less than 1500 kg/ha/year is well recognized as in poor condition for livestock

purposes. However, the rangeland with ≥800 kg/ha/yeaer biomass has high protection

potential against erosion (Teague et al., 2009). The quality of forage is mainly influenced by

factors such as type and amount of nutrients, fibre content, unpalatable chemical substances

and percentage moisture (it varies with species) (Peden, 2005). The palatable species occur

naturally in the rangelands that are well managed and decreases with poor management such

as over grazing (Morris and Kotze, 2006). Biomass production of natural grassland systems

varies according to available moisture (Noellmeyer et al., 2006). Perennial grasses produce

more foliage than annual grasses and thus provide more forage yield than annuals (Peden,

2005). Perennial grasses have extensive root systems and protect the soil from erosion more

effectively than the annual species. The annual species replace perennial species as the

grazing intensity increases (Maki et al., 2007).

Climatic conditions and grazing have marked influences on biomass production (Fynn and

O`Connor, 2000; Savadogo et al., 2006; Angassa and Oba, 2010). It was reported that

biomass production during the dry season is less when compared to the wet season (Angassa

and Oba, 2010). This provides an indication that seasonal variation is a major driving force to

the variation in biomass production. The rainfall inefficiency is the main driver to this

phenomenon per growing season (Angassa and Oba, 2010). Positive correlations between

rainy days and biomass production have been reported in a study in Burkina Faso (Savadogo

et al., 2006). Adjustment of the stocking rate should be according to the response of forage

due to seasonal variations.

19

2.6 Basal cover

In consideration of the high stocking rates, herbivores alter the plant distribution resulting in

substitution of perennials by annuals (Tessema et al., 2011; Savadogo et al., 2006) and forbs,

resulting to a reduction in forage production (Savadogo et al., 2006). Moreover, this

promotes the reduction of species diversity, increase exposure to bare ground and leads to

increased runoff and soil erosion, which in turn results to reduced water availability, nutrient

retention and plant establishment (Mganga et al., 2011). The most important single factor

affecting water run-off is the amount and type of vegetative cover (Malan and Van Nierkerk,

2005). Soil cover provided by vegetation maybe in basal or aerial terms (Lesoli, 2011). The

base of a rooted plant provides basal cover and it depends on the thickness of the tuft and

plant density (Lesoli, 2011). The higher the basal cover, the lower the run-off rate and the

lower the basal cover the higher the run-off rate. The run-off rate is one of the factors

responsible for soil transportation. Herbaceous plants provide more soil protection against

rain drops and run-off than the non-herbaceous ones (Tainton, 1999). According to Lesoli

(2011), this is mainly because herbaceous species provide a complex network of roots

immediately below the ground surface, which hold the soil particles together unlike deep

rooted trees. Stands of the perennial species were more stable than stands of the annual

species and provided stable soil cover (Lesoli, 2011). The influence of basal cover and bare

ground on grass yield was reported to be higher on forage biomass production meaning that a

higher proportion of basal cover leads to a high forage yield (Fahnestock and Detling, 2000).

Baars et al. (1997) indicated that under proper rangeland management practices, basal cover

of excellent vegetation is expected to be greater than 12%.

The diminution in vegetation cover by overgrazing put soil at risk for runoff to takeover,

thereby aggravating the extent of soil erosion (Oztas et al., 2003). Lutge et al., (1998) In a

report in Kokstad Research Station, 90% reduction in basal cover was linked to patch grazing

by livestock as the animals tend to focus on one portion thereby ruining the grass sward.

According to Snyman, 2009, there is a correlation between changes in plant species

composition and basal cover. Highlighting that, a healthy basal cover should be characterized

by perennial grasses because annual grasses die after completion of lifespan leaving the bare

soils in favor of soil erosion (Malan and Van Niekerk, 2005). Basal cover increases with a

decrease in rangeland condition due to the fact that the low creeping grasses tend to take over

20

when the tall, erect grasses decline (Sisay and Baars, 2002). Bare ground is a good indicator

of over utilization and the degree of degradation of the vegetation (Abule et al., 2007). Lack

of managed grazing or uncontrolled grazing may result in poor basal cover, change in species

composition and low biomass production, which in turn leads to rangeland degradation (Smet

and Ward, 2005).

2.7 Soil quality and soil properties

Soil quality can be defined as the capacity of a soil to function, within ecosystem and land

use boundaries, to sustain biological productivity, maintain environmental quality and

promote plant and animal health (Corwin et al., 2003). Soil quality has been typically equated

with soil organic matter or its indicator elements, carbon and nitrogen. Goldschmidt (1987),

reported a nitrogen problem in pasture soils of the Natal Sour Veld. Du Toit, (1990) noted

that because of the widespread aridity and low humus content, South African soils generally

have an extremely delicate nature and lack resilience compared to soils in temperate areas.

Penzhorn (1991) spoke of the thin, vulnerable and unstable soil mantle in South Africa. Only

recently have quantitative begun to emerge: Du toit et al. (1994) found 5-90 years of

cultivation in the Free State resulted in a loss of 10-73% of C and N relative to natural

rangelands. The parent material and inherent diversity is considered as the major causes of

high abundance of macro (N, P, K, Mg and Ca) and micro (Fe, Cu and Zn) nutrients in

hardveld than in sandveld. High amounts of potassium in the cultivated and grazing areas can

be attributed to higher clay mineral content in the soil. The nutrient availability is positively

correlated with soil clay content (Mills and Fey, 2005). The mafic rocks that are originally

from basalts are the sources of Mg and Zn and they easily weather to form soils that are rich

in clay minerals (Grant et al., 2000).

Salinity and alkalinity (especially alkalinity) have major impacts on plant production (Lesoli,

2008). Extreme values of soil pH (which after solubility of most of the elements necessary for

plant growth) are an insidious problem in some regions. According to Rezaei and Gilkes

(2005), Soil pH affects the solubility of nutrients and uptake by plants. Soil pH often affects

plant community composition because plants differ in nutrient requirement and soil acidity or

basicity tolerance and soil pH is influenced by elevation (Lesoli, 2008). The soil parent

material of higher pH occurs at lower elevation (Laughlin and Abella, 2007). Salinity is a

dynamic soil property and it varies temporally and spatially with depth and across the

21

landscape (Lesoli, 2011). Corwin et al. (2003) stated that salinity varies primarily due to the

process of leaching with topographic effects to this variation. Surface topography plays a

vital role in influencing spatial electrical conductivity variation. The difference in the cation

exchange capacity (CEC) of the soils is influenced by organic carbon and clay content. The

CEC values indicate the capacity of soil to retain nutrient cation against leaching (Ludwig et

al., 2001). There is a positive relationship between soil organic carbon and the capacity of the

soil to supply essential plant nutrients including nitrogen, phosphorus and potassium (Rezaei

and Gilkes, 2005).

Soil nitrogen content follows soil carbon content in grassland soil (Conant and Paustial,

1998). Moreover, the relationship between organic carbon and landscape attributes as well as

the positive relationship between organic carbon and the nutrient elements, indicates the

usefulness of the organic carbon as a reliable and sensitive indicator of rangeland health

(Rezaei and Gilkes, 2005). The soil found under managed rangelands has high levels of

organic carbon and almost all organic constituents (Lu et al., 2007). On the other hand, Li et

al. (2007) showed that soil organic carbon played an important role in improving soil

physical, chemical and biological properties for sustained plant growth. Rangeland

sustainability is related to soil carbon and nutrient balance and the capability to maintain

adequate soil conditions for water availability and root development (Noellemeyer et al.,

2006). The soil carbon balance is maintained by plant litter inputs, which enter the soil as

particulate organic carbon (Lesoli, 2011). Soil under shade such as tree canopy, accumulates

more soil organic carbon due to the influence of the tree canopy on the soil temperature

regime. According to Simion et al. (2003), the different carbon dynamics are the result of a

high proportion of woody debris under shade and different removal rates of aboveground

biomass by grazing in the open communities. Changes in the soil carbon may occur in

response to a wide range of management and environmental factors hence, rotational grazing

management will provide enough time between occupation period and in turn stimulating

growth of the herbaceous species and improve nutrient cycling in the rangelands (Schumen et

al., 2002). The disturbance of the rangelands has negative impacts on soil structural

properties and water holding capacity, which are related to losses of the soil organic content

(Li et al., 2007) and the deterioration of the soil properties results to a decrease in soil

infiltration and water retention accelerating soil erosion.

22

2.8 Soil seed bank composition in rangelands

Plants establish by the expansion and subsequent fragmentation of vegetative parts such as

tillers, rhizomes, runners or by the successful establishment of a soil seed bank (Freedman et

al., 1982). Seeds may be introduced to the soil seed bank during different times may be

during the current or previous year. They may also be removed through germination,

predation, and senescence and by pathogens (Solomon et al., 2006). Soil seed bank plays a

significant role in restoration of degraded rangelands through seedling recruitment. Plenty of

studies during rangeland condition assessments usually focus on the aboveground vegetation

overlooking the importance of soil seed bank in resistance and resilience of the rangelands

(Drbber et al., 2011). Soil seed bank recruitment is restricted to periods with favourable

conditions of the soil parameters that may control seed germination and these parameters

include soil water, pH, temperature and light (Solomon et al., 2006). In addition to that,

drought and heavy grazing adversely affect the size and composition of grasses in the seed

bank, both spatially and temporally (Solomon et al., 2006). The evaluation of soil seed banks

can give an idea of the recovery potential of a particular degraded rangeland (Tongway et al.,

2003). Moreover, Tongway et al., 2003 suggested that the soil seed bank is not the reason for

lack of vegetation in degraded lands. Even though soil seed banks in the degraded areas are

generally low, the major factor that can determine vegetation germination is soil moisture.

Therefore, soil seed bank evaluation may be used as a valuable tool to assess rangeland

condition and potential (Snyman, 2004; Solomon et al., 2006; Dreber et al., 2011). Soil seed

banks may be composed of viable seeds which may either be persistent (Shaukat and

Siddiqui, 2004) or those that are transient. However, the efficacy of recruitment from seed

bank is largely dependent on moisture and nutrient status of the soil (Snyman, 2004).

Rangelands that have a large, persistent seed bank, often have species composition but that

does not resemble the aboveground vegetation (Thompson and Grime, 1997; Amaha

Kassahun et al., 2009), but these seeds can state the successional trends that occured

following large-scale disturbances (Bekker et al., 1997; Edwards and Crawley, 1999).

Worldwide, rangelands are subject to active management and these practices are based on a

variety of criteria and constraints (Snyman, 2009). High grazing pressure has been considered

the most important cause of rangeland degradation in South Africa. The ecologically

sensitive semi-arid rangelands are increasingly susceptible to severe grazing pressures, which

23

results to rapid deterioration (van der Westhuizen, 1999). In some communal grazing areas,

plants are not allowed to seed due to continuous heavy grazing (Solomon et al., 2006;

Rutherford and Powrie, 2011). Understanding the function and dynamics of seed banks has

become a great challenge to ecologists working in plant communities. The understanding of

the function and dynamics of seed banks is necessary to determine the role of the seed bank

in ecosystem function and to improve the integrated management of ecosystems (Luzuriaga

et al., 2007; Snyman, 2009; Dreber, 2011).

It is of high importance to know the degree to which species in a system depend on specific

forms of disturbance or whether various types of disturbance have equivalent effects on the

soil seed bank (Bekker et al., 1997; Page et al., 2006; Ma et al., 2010). Some authors have

argued that prescribed burning should be the preferred form of rangeland management

(Everson, 1999; Trollope. 1999), some postulated that a variety of forms of disturbances can

have equivalent effects (Collins et al., 1998; Jutila and Grace. 2002). The potentially adverse

effects of disturbances, particularly when intense and/or frequent, must also be given careful

consideration (Jutila and Grace, 2002; Laterra et al., 2006). In a study by Mndela (2013) in

the communal areas of Eastern Cape, South Africa, reliance on soil seed bank for restoration

of degraded rangelands was not recommended as the composition was dominated by forbs

and sedges. Similarly to that, Solomon et al. (2006) found that reliance on the seed bank in

his study between communally grazed areas was not recommended which was in contrast to

the results of the controlled grazing areas where reliance on seed bank was of importance in

restoring the rangelands. Unfortunately, there are only few studies about the regenerative

potential of seed banks (Luzuriaga et al., 2007), the longevity of the seeds for each species

under specific climatic conditions, and the quantification of seed rain in arid and semi-arid

areas (Snyman, 2010).

2.9 Rationale for the study

This study was aimed at evaluating the vegetation dynamics through comparison of different

homogenous vegetation units, seasonal variation and examination of soil seed banks.

Evaluation of these parameters gave an indication of the extent of degradation in the selected

rangelands. Recommendations on management at the selected semi-arid rangelands will be

formulated using the scientific understanding of rangeland utilization so as to improve the

24

condition of communal rangelands at Glenmore, Ndwayana and the Great Fish River Nature

Reserve.

25

References

Abel N O J 1993. Reducing cattle numbers on Southern African communal range: is it worth it? In:

Behnke Jr, R H Scoones, I., Kerven, C. (Eds.), Range Ecology at Disequilibrium: New

Models of Natural Variability and Pastoral Adaptation in African Savannas. Overseas

Development Institute, London, pp. 173–195.

Abule E, Snyman H A, and Smit G N 2007. Rangeland evaluation in the middle Awash valley of

Ethiopia: I. Herbaceous vegetation cover. Journal of Arid Environments, 60 (343-358).

Acocks J P 1953‘The veld types of South Africa’, Memoirs of the Botanical Survey of South Africa,

8, pp. 1–128.

Amaha Kassahun H A, Snyman H A and Smit G N 2009. Soil seed bank evaluation along a

degradation gradient in arid rangelands of the Somali region, eastern Ethiopia. Agricultural,

Ecosystems and Environment 129:428–436.

Anderson P M L and Hoffman M T 2006. The impacts of sustained heavy grazing on plant diversity

and composition in lowland and upland habitats across the Kamiesberg mountain range in the

Succulent Karoo, South Africa. Journal of Arid Environments.2006.05.017.

Angassa A and Oba G 2010. Effects of grazing pressure, age of enclosures and seasonality on bush

cover dynamics and vegetation composition in southern Ethiopia. Journal of Arid

Environments. 74: 111 – 120.

Baars R M T, Chileshe E C, and Kalokoni 1997. Technical notes: range condition in high cattle

density areas in the Western Province of Zambia. Tropical Grasslands 31: 569 -573.

Behnke R H, Scoones I 1992. Rethinking Range Ecology. Implications for Rangeland Management

in Africa. International Institute for Environment and Development (IIED), London.

Bekker R, Verweij GL, Smith R E N, Reine R, Bakker J P and Schneider S 1997. Soil seedbanks in

European grasslands: does land use affects regeneration perspective. Journal of Applied

Ecology 34: 1293- 1310.

26

Belsky A J and Bluementhal D M 1997. Effects of livestock grazing on Stand Dynamics and Soils in

Upland Forest of the Interior West.

Biggs R , Simons H, Bakkenes M , Scholes R J, Eickhout B, van Vuuren D, Alkemade R 2008.

Scenarios of biodiversity loss in southern Africa in the 21st century. Global Environmental

Change 18, 296–309.

Blackburn W H 1983. Livestock Grazing Impacts on Watersheds. Rangelands 5 (3): 123-125

Collins S L, Knapp A K, Briggs J M, Blair J M and Steinauer E M 1998. Modulation of diversity by

grazing and mowing in native tallgrass prairie. Science 280:745–747.

Conant R T and Paustial K 1998.Carbon sequestration in pastures. The Ruminant Livestock

Efficiency Program annual Conference 32- 40.

Corwin D L, Kaffka S R, Hopmans J W, Mori Y, van Groenigen J W and van Kessel C 2003.

Assessment and field-scale mapping of soil quality properties of a saline-sodic soil.Geodema

114: 231-2 59.

Dean W R J, Hoffman M T, Meadows M E, Milton S J 1995a. Desertification in the semi-arid

Karoo, South Africa: review and reassessment. Journal of Arid Environments 30, 247–264.

Dougill A J, Thomas D S G, Heathwaite A L 1999. Environmental change in the Kalahari:

integrated land degradation studies for nonequilibrium dryland environments. Annals of

American Geographers 89, 420–442.

Dreber N and Esler K J 2011. Spatio-temporal variation in soil seed banks under contrasting grazing

regimes following low and high seasonal rainfall in arid Namibia. Journal of Arid


Du Toit A L 1990.Geological and Soil characteristics in relation to deterioration and conservation of

Natural resources, etc in the Union, South Africa. Journal of Science.35,470-476.

27

Du Toit A L 1991.Geological and Soil characteristics in relation to deterioration and conservation of

Natural resources, etc in the Union, South Africa. Journal of Science.35,470-476.

Du Toit M E, Du Preez C C, Hnesley M and Bennie A T P 1994. Effek van beweking of die organise

materuaalinhoud van geselekteerde aroeland grande in Suid Afrika. South African Journal of

Plant Science.11,71-79.

Edwards, G.R., and Crawley, M.J., 1999. Herbivores, seed bnks and seedling recruitment in Mesic

grassland. Journal of Ecology. 87: 423 – 435.

Ellis J E and Swift M 1988. Stability of African pastoral ecosystems: alternate paradigms and

implications for development, Journal of Range Management 41:-450 459.

Esler K J, Milton S J, Dean W R J 2006 Karoo Veld, Ecology and Management. Briza, Pretoria.

Everson T M and Hatch G P 1999. Managing veld (rangeland) in communal areas of southern Africa

In: In: Tainton N M (ed) Veld Management in South Africa. University of Natal Press,

Pietermaritzburg, pp. 381- 388.

Fahnestock J T and Detling J K 2000. Morphological and Physiological response of perennial grass

to long-term grazing in the Pryor Mountains, Montana. Animal Midland Naturalist143: 312 -

320.

Freedman B, Hill N, Henry G and Svoboda J 1982.Seed banks and seedling occurrence in a high

Arctic oasis at Alexandra Fjord, Ellesmere Island, Canada. Canadian Journal of Botany 60:

2112– 2118.

Friedel M H 1991. Range condition assessment and the concept of thresholds: a viewpoint. Journal

Range Management 44: 422- 426.

Fynn R W S and O'connor T G 2000. Effect of stocking rate and rainfall on rangeland dynamics and

cattle performance in a semi-arid savanna, South Africa. Journal of Applied Ecology. 37: 491

– 507.

28

Goldshmedt W B 1987 Preliminary report the Nitrogen problem in Pasture soils of the Natal Sour

Veld. South African Journal of Science. 35,244-245.

Govers G, Vandaele K, Desmet P, Poesen,J and Bunte K 1994.The role of tillage in soil

redistribution on hillslopes. European Journal of Soil Science.45 469-478.

Grant C C, Peel M J S and Van Ryssen J B J 2000. Nitrogen and phosphorus concentration in faeces:

an indicator of range quality as a practical adjunct to existing range evaluation methods.

African Journal of Range and Forage Science. 17: 81 – 92.

Hayes G E and Holl D K 2003.Cattle grazing impacts on annual forbs and vegetation composition of

mesic grassveld in California. Conservation Biology 17(6): 1694- 1702.

Hoffman I, Gerling D, Kyiogwom U B and Mané-Bielfedt A 2001.Farmers’ management strategies

to maintain soil fertility in a remote area in northwest Nigeria. Agriculture, Ecosystems and

Environment 86: 263 – 275.

Hoffman M.T and Ashwell A 2001 Nature Divided: Land degradation in South Africa: the influence

of biophysical and socio-economic factors. Journal of Southern African studies,26,743-758

Hoffman T and Todd S 2000. National review of land degradation in South Africa: The influence of

biophysical and socio-economic factors. Journal of Southern African studies 26:743-758.

Hoffman T, Ashwell A 2001. Nature Divided. Land Degradation in South Africa. University of Cape

Town Press, Cape Town.

Huntington T G 2006 Evidence for intensification of the global water cycle: Review and synthesis.

Journal of Hydrology. 319 (83-95).

Hurt C R and Bosch O J H 1991. A comparison of some range condition assessment techniques used

in Southern African grasslands. Journal of the Grassland Society of Southern Africa. 8: 131 –

137.

Jutila H M and Grace J B 2002. Effects of disturbance on germination and seedling establishment in

a coastal prairie grassland: a test of the competitive release hypothesis. Journal of Ecology

99:291–302.

29

Kakembo V, Rowtree K and Palmer V 2007 Topographic controls on the invasion of Pteronia incana

(blue bush) onto hillslopes in Ngqushwa (Formely Peddie) district Eastern Cape, South

Africa.

Kassahun A, Snyman H A and Smit G N 2008. Impact of rangeland degradation on the pastoral

production systems, livelihoods and perceptions of the Somali pastoralists in Eastern

Ethiopia. Journal of Arid Environments. 72: 1265 – 1281.

Kioko J and Okello M M 2010. Land use cover and environmental changes in semi-arid rangeland,

Southern Kenya. Journal of Geography and Regional Planning. 3: 322 – 326.

Kioko J, Kiringe J W and Seno S O 2012. Impacts of livestock grazing on a savannah grasslands of

Kenya. Journal of Arid land. 41: 29 – 35.

Kokot D F 1948.An Investigation into the Evidence Bearing on Recent Climatic Changes over

Southern Africa (Pretoria, Irrigation Department Memoir).

Lal R and Pimentel D 2007 Biofuels from Crop Residues. Soil and Tillage Research. 93 (2), 237.

Lal R 2004. Soil Carbon Sequestration Impacts on Global Climate Change and Food Security.

Science.34, 1623–1627.

Lal R 2001.Soil degradation by erosion. Land degradation and development. 12 519-539

Laughlin D C and Abella S R 2007. Abiotic and biotic factors explain independent gradients of plant

community composition in ponderosa pine forests. Ecological modelling, doi: 10.1016/j.

ecolmodel.2007.02.018.

Lesoli M S 2011. Characterization of communal rangeland degradation and evaluation of vegetation

restoration techniques in the Eastern Cape, South Africa.PhD. Thesis, University of Fort

Hare, Alice, South Africa.

Lesoli M S 2008. Vegetation and soil status and human perceptions on the condition of communal

rangelands of the Eastern Cape, South Africa. MSc. Dissertation, University of Fort Hare,

Alice, South Africa.

30

Leterra. P, Ortega, E. Z,Ochoa, M. D. Vignolio. C O. R and Fern`andez. O. N. 2006. Interactive

influences of fire intensity and vertical distribution of seed banks in post-fire recolonization

of a tall-tussock grassland in Argentina. Austral Ecology 31:608–622.

Li X Y, Liu L Y, Gao S Y, Shi P J, Zou X Y and Zhang C L 2005. Microcatchment water harvesting

for growing Tamarixramosissima in the semiarid loess region of China. Forest Ecology and

Management 214: 111 – 117.

Lobb D A, Kachanoski R G, Miller M H , 1999. Tillage translocation and tillage erosion in the

complex upland landscapes of southwestern Ontario, Canada. Soil Tillage Res.

Lu G, Sakagami K, Tanaka H and Hamada R 1998.Role of soil Organic Matter in stabilization of

water-stable aggregates in soils under different types of land use. Soil science and plant

nutrition 44: 147- 155. 272

Ludwig B, Khanna P K, Anurugsa and Fölster H 2001.Assessment of cation and anion exchange and

pH buffering in an Amazonian Ultisol.Geoderma 102: 27- 40.

Lutge B U , Hardy M B, and Hatch G P 1998. Soil and sward characteristics of patches and non-

patches in the Highland Sourveld of South Africa. Tropical Grasslands. 32: 64 – 71.

Luzuriaga A L, Escudero A, Olano J M and Loidi J 2007. Regenerative of seedbanks following an

intense soil disturbance. Acta Oecologica 27:57–66.

Ma M J, Zhou X H and Du G Z 2010. Role of soil seed bank along a disturbance gradient in an

alpine meadow on the Tibet plateau. Flora 205:128–134.

Maki A, Kenji T, Kiyokazu K and Teruo H 2007.Morphological and physico-chemical

characteristics of soils in a steppe region of the Kherlen River basin, Mongolia. Journal of

Hydrology 333: 100- 108.

31

Malan P W and Van Niekerk 2005.The extent of grass species composition in Braklaagte, Zeerust

District, North-West Province, South Africa. African Journal of Range and forage Science

22(3): 177- 184.

Mndela M 2013. Evaluation of range condition, soil properties and seed banks, and farmer’s

perceptions in Peddie communal rangeland of the Eastern Cape, South Africa. MSc Thesis

University of Fort Hare, Alice.

Mganga Z, Musimba, K R N, Nyariki D M, Nyangito M M , Ekaya W N, Muiru W M and

Mwang‟omb A W 2011. Different land use types in the semi-arid rangelands of Kenya

influence soil properties. Journal of Soil Science and Environmental Management. 2: 370 –

374.

Mills A J and Fey M V 2005. Interactive response of herbivores, soils and vegetation to annual

burning in a South African Savanna. Austral Ecology.30: 435 – 444.

Milton S J , Dean W R J , du Plessis M A, Siegfried W R 1994. A conceptual model of arid

rangeland degradation. Bio Science 44, 70–76.

Milton S J, Hoffman, M.T., 1994. The application of state-and-transition models to rangeland

research and management in arid succulent and semi-arid grassy Karoo, South Africa.

African Journal of Range and Forage Science 11, 18–26.

Milton S J, Hoffman, M.T 1994. The application of state-and-transition models to rangeland research

and management in arid succulent and semi-arid grassy Karoo, South Africa. African Journal

of Range and Forage Science 11, 18–26.

Morgan 2009 Soil Erosion and Conservation.3rd

Edition. National Soil Resource Institute,Cranfield

University.

Morris C and Kotze D 2006. Introduction to Veld care (1). Agricultural research council, University

of Kwazulu Natal.

32

Noellemeyer E, Quiroga A R and Estelrich D 2006.Soil quality in three range soils of the semi-arid

Pampa of Argentina. Journal of Arid Environments 65: 142- 155.

Nsinamwa M, Moleele N M and Sebego R J 2005. Vegetation patterns and nutrients in relation to

grazing pressure and soils in the sandveld and hardveld communal grazing areas of Botswana

22 (1): 17- 28.

O’Connor T G 1991. Local extinction in perennial grasslands: a life-history approach. The American

Naturalist 137, 753–773.

Oztas T, Koc A, and Comakli B 2003.Changes in vegetation and soil properties along a slope on

overgrazed and eroded rangelands. Journal of Arid Environments 55: 93- 100.

Page M J, Baxter G S and Lisle A T 2006. Evaluating the adequacy of sampling germinable soil seed

banks in semi-arid systems. Journal of Arid Environments 64:323–341.

Palmer A R, Ainslie, A and Hoffman M T 1998 Sustainability of commercial and communal

rangeland systems in southern Africa, Townsville, Australia.

Peden M I 2005. Tackling the most avoided issue: Communal rangeland management in Kwazulu-

Natal, South Africa. African Journal of Range and Forage Science 22 (3) 167-175.

Pezhorn K E 1991.Soil Conservation =: Where do we stand? South African Journal of Science. 68

(177-181).

Retzer V 2006. Impacts of grazing and rainfall variability on the dynamics of a Sahelian rangeland

revisited (Hein): New insights from old data. Journal of Arid Environments. 67: 157 – 164.

Reynolds J F, Stafford Smith D M 2002. Do humans cause deserts? In: Reynolds J F, Stafford Smith

D M (Eds.), Global Desertification: Do Humans Cause Deserts? Dahlem University Press,

Berlin, pp. 1–21.

33

Reynolds J F 2001 In Encyclopedia of biodiversity.Vol.2 (Ed.Levin.S),(PP.61-78).San Diego:

Academic Press.

Rezaei A S and Gilkes R J 2005.The effects of landscape attributes and plant community on soil

chemical properties in rangelands.Geodema 125: 167- 176.

Rouget M, Cowling R M, Vlok J, Thompson M, Balmford A 2006. Getting the biodiversity

intactness index right: the importance of habitat degradation data. Global Change Biology 12,

2032–2036.

Mucina L, Rutherford M C 2004. Terrestrial Component. South African National Spatial

Biodiversity Assessment 2004: Technical Report, vol. 1.

Rutherford M C and Powrie L W 2011. Can heavy grazing on communal land elevate plant species

richness levels in the Grassland Biome of South Africa? Plant Ecology 212:1407–1418.

Savadogo P, Sawadogo L and Tiveau D 2006. Effects of grazing intensity and prescribed fire on soil

physical and hydrological properties and pasture yield in the savanna woodlands of Burkina

Faso. Agriculture, Ecosystems and Environment. 11: 80 – 92.

Scholes R J, Biggs R 2005. A biodiversity intactness index. Nature 434, 45–49.

Schuman G E, Janzen H H and Herrick J E 2002.Soil carbon dynamics and potential carbon

sequestration by rangelands. Environmental pollution 116; 391- 396.284 Science.34, 1623–

1627.

Scoones I 1993. Why are there so many animals? Cattle population dynamics in communal areas of

Zimbabwe. In: Behnke, Jr., R.H., Scoones, I., Kerven, C. (Eds.), Range Ecology at

Disequilibrium: New Models of Natural Variability and Pastoral Adaptation in African

Savannas. Overseas Development Institute, London, pp. 62–76.

Shaukat S S and Siddiqui A I 2004. Spatial pattern analysis of seeds of an arable soil seed bank and

its relationship with above-ground vegetation in an arid region. Journal of Arid


34

Simion G, Gignoux J and Le Roux X 2003. Tree layer spatial structure can affect Savanna

production and water budget: Results of a 3-D Model. Ecology 84 (7): 1879- 1894.

Sisay A and Baars R M T 2002.Grass composition and rangeland condition of the major grazing

areas in the mid Rift Valley, Ethiopia. African of range and forage Science 19: 161- 166.

Smet M and Ward D 2005. A comparison of the effects of different rangeland management systems

on plant species composition, diversity and vegetation structure in a Semi-Arid Savanna.

African Journal of Range and Forage Science 22 (1): 59- 71.

Snyman H A 2010. Longevity of grass seeds in a semi-arid grassland. Grassroots10 (2):8–15.

Snyman H A 1998. Short-term effects of soil water, defoliation and rangeland condition on

productivity of a semi-arid rangeland in South Africa. Journal of Arid Environments 43, 47–

62.

Snyman H A 2004. Soil seed bank evaluation and seedling establishment along a degradation

gradient in a semi-arid rangeland. African Journal of Range and Forage Science. 21: 37 – 47.

Snyman H A 2009. Root studies on grass species in a semi-arid South Africa along a degradation

gradient. Agriculture, Ecosystems and Environment. 130: 100 – 108.

Solomon T B, Snyman H A and Smit G N 2006. Soil seed bank characteristics in relation to land use

systems and distance from water in a semi-arid rangeland of southern Ethiopia. South African

Journal of Botany 72: 263- 271.

Solomon T B, Snyman H A and Smit G N 2007. Rangeland dynamics in outhern Ethiopia: (3).

Assessment of rangeland condition in relation to land-use and distance from water in semi-

arid Borana rangelands. Journal of Environmental Management. 85: 453 – 460.

Stocking M and Murnaghan N 2001. Land degradation – Guidelines for field assessment. Overseas

Development Group, University of East Anglia, Norwich, UK.288

35

Sullivan S and Rohde, R., 2002 On non-equilibrium in arid and semi-arid grazing systems. Journal

of Biogeography 29, 1595–1618.

Tainton N M 1999 The ecology of the main grazing lands of South Africa. In: Tainton N M (ed)

Veld Management in South Africa. University of Natal Press, Pietermaritzburg, pp. 23- 53.

Tainton N M, Aucamp A J, Danckwerts J E 1999 Principles of managing veld. In: Tainton, N.M.

(Ed.), Veld Management in South Africa. University of Natal Press, Pietermaritzburg, pp.

169–193.

Tapson D 1993 Biological sustainability in pastoral systems: the KwaZulu case. In: Behnke R H

Scoones I, Kerven, C. (Eds.), Range Ecology at Disequilibrium: New Models of Natural

Variability and Pastoral Adaptation in African Savannas. Overseas Development Institute,

London, pp. 118–135.

Teague W R 1989 Monitoring vegetation in thorn/ grass communities. In: Danckwertz, J. E and

W.R. Teague (Eds). Veld management in the Eastern Cape. Government Printer, Pretoria. Pp.

99 – 100.

Teague W R, Kreuter U P, Grant W E, Diaz-solis H and Kothmann M M 2009. Economic

implications of maintaining rangeland ecosystem health in a semi-arid savanna. Ecological

Economics. 68: 1417 – 1429.

Tessema Z K, de Boer F W, Baars M R T and Prins H H T 2011. Changes in soil nutrients,

vegetation structure and herbaceous biomass in response to grazing in semi-arid savanna of

Ethiopia. Journal of Arid Environment. 75: 662 – 670.

Thompson K and Grime J P 1997. Seasonal variation in seed banks of herbaceous species in ten

contrasting habitats. Journal of Ecology 67:893–921.

Todd S W and Hoffman M T 1999. A fence-line contrast reveals effects of heavy grazing on plant

diversity and community composition in Namaqualand, South Africa. Plant Ecology 142:

169- 178.

36

Tongway D J, Sparrow A D, Friedel M H 2003. Degradation and recovery process in arid grazing

lands of central Australia: Part 1. Soil and land resources. Journal of Arid Environments 55:

301- 326.

Trollope W S W 1999. Fire Behaviour. In: N M Tainton (ed) Veld Management in South use

efficiency in short-grass steppe: grazing history, defoliation, and water resource.

Van der westhuizen H C, Van rensburg W L J and Snyman H A 1999. The quantification of

rangeland condition in a semi-arid grassland of South Africa. African Journal of Range and

Forage Science 16:49–61.

Van Muysen W, Govers G and Van Oost 2002. Identification of important factors in the process of

tillage erosion: The case of mouldboard tillage.65 77-93

Van Oudtshoorn F P 2006.A guide to grasses of South Africa. Second edition, Fifth impression.

Briza Publications. Cape Town, South Africa.

Varnamkhasti A S, Milchunas D G, Lauenroth W K, and Goetz H 1995. Production and rain

use efficiency in short-grass steppe: grazing history, defoliation, and water resource.

Journal of Vegetation Science 6: 787- 796.

Verburg P H and van Keulen H 1999. Exploring changes in the spatial distribution of livestock in

China. Agricultural systems 62: 51- 67.

Ward D, Ngairorue B T, Kathena J, Samuels R, Ofran Y 1998. Land degradation is not a necessary

outcome of communal pastoralism in arid Namibia. Journal of Arid Environments 40, 357–

371.

37

Wessels K J, Prince S D, Frost, P E and Van Zyl D 2004.Assessing the effects of human-induced

land degradation in the former homelands of northern South Africa with1km AVHRR NDVI

time-series. Remote sensing of Environment, 91.47-67.

Wiegleb G and Felinks B 2001. Primary succession in post-mining landscapes of lower Lusatia

chance or necessity. Ecological Engineering 17 (2–3): 199–217.

WMO 2005. Climate and Land degradation world meteorological Organization, WMO, No. 989

38

CHAPTER 3. BOTANICAL COMPOSITION, BIOMASS

PRODUCTION AND BASAL COVER IN NDWAYANA,

GLENMORE AND THE GREAT FISH RNR.

ABSTRACT

A study was conducted to document the extent of land degradation through range condition

assessment in three selected semi-arid rangelands of the Eastern Cape, South Africa. These

comprised the Great Fish River Nature Reserve (also had a site considered a benchmark), as

well as Glenmore and Ndwayana communal rangelands. Twenty-two (22) perennial

herbaceous species and forbs were recorded in all the HVUs. In general the herbaceous

species composition consisted of 59% pioneer (Increaser II) species, 36.4% mesophytes

(Decreaser) species and the remaining were 4.54 % sub-climax/climax (Increaser I) species.

The grazing value amongst the species were as follows High 41%, Moderate 14% and Low

45%. There was a significant variation (p<0.05) in biomass production between the HVUs

during the different seasons. Biomass production in the benchmark (2700 kg/ha) was

significantly higher (p<0.05) than in Glenmore & Ndwayana in summer, while no significant

differences (p>0.05) occurred between the bottomlands (1992 kg/ha) in summer and winter

(1715 kg/ha) at the Great Fish RNR. There were 27 woody species recorded from all study

sites. Thorns or spines were present on 41% of the species while they were absent on 59% of

the species. Ptaeroxylon obliquum (14%) was the most dominant woody species, and the

least dominant was Pappea capensis (0.05%) respectively.Glenmore bottomlands and

toplands had significantly higher (p<0.05) bush density (1181.25 and 1337.5Trees/ha) than

all the other HVU’s. The trend between the three sites was that: the benchmark site had the

best, followed by the toplands and bottomlands of the Great Fish RNR in terms of botanical

composition, ecological stability and biomass production. These were subsequently followed

by toplands and bottomlands at Glenmore and Ndwayana respectively. The communal

rangelands are more degraded than the Great Fish RNR in terms of biomass production and

species composition. Glenmore top-lands and bottom-lands had higher tree equivalents and

density when compared to the other rangelands but bush encroachment was not a problem in

these areas. The results for both bush density and tree equivalents was less than the threshold

given to determine whether an area is bush encroached or not (<1500 TE/ha). Therefore,

39

reseeding of the rangelands, herding of the livestock, and application of the correct stocking

rates, demarcation and fencing of camps are strict control measures that are recommended to

halt degradation in Glenmore, Ndwayana and the Great Fish RNR.

Key words: Land degradation, rangeland condition, biomass production, vegetation cover.

40

3.1 INTRODUCTION

Rangeland condition is the state of health of the rangeland in terms of ecological status,

resistance to soil erosion and the potential for producing forage for sustained optimum

livestock production (Trollope, 1990). Subjective and quantitative techniques have been used

in rangeland condition assessments and the choice of the method to be used depends on the

factors and local conditions (Jordaan, 1997). Generally, in conducting assessments in any

rangeland ecosystem composed of different vegetation components, rangeland monitoring

must incorporate three tiers of assessment namely, the herbaceous layer, the soil and the

woody component (Dankwarts, 1982; Friedel, 1987, 1991) as cited by (Abule et al., 2007).

An indication was made by Pratt and Gywanne (1977), that the use of species composition

alone as an index of rangeland condition rating is unsatisfactory, and hence suggested the

inclusions of other parameters like basal cover, plant vigour, percentage bare ground,

biomass, estimated soil erosion and soil compaction as deemed necessary. Moreover, the use

of species composition and biomass production provide proper estimates of stocking rate for

sustainable grazing management (Kunst et al., 2006). The extent of soil loss through surface

runoff in a rangeland is largely reliant on the ecological stability (basal cover) of that

rangeland (Rowntree et al., 2004).

Botanical composition is one of the means of studying ecological changes in the development

of a rangeland (Malan and Van Niekerk, 2005). As a result, this reflects many factors that

include past management (Whalley and Hardy, 2000). Any change in the grazing practices

will result in the change in species composition (Hayes and Holl, 2003). The amount of

grazing pressure in the rangelands cause changes not only in species composition but to the

vegetation structure and productivity (Oztas et al., 2003; Maki et al., 2007). Moreover, Sisay

and Baars, (2002) stated that a long term increase or relaxation of grazing pressure changes

plant community also concluding that under heavy grazing pressure Decreaser species

disappear and are replaced by Increaser or Invader species. Contrary to that, Laughin and

Abella (2007) indicated that the change in composition is determined more by rainfall than by

grazing pressure. Species composition is an indicator of rangeland condition mainly because

species vary significantly in their acceptability and response to defoliation (Abule et al.,

2007).

41

High intensity grazing leads to excessive removal of the most desirable species, which are

usually perennial grasses (Todd and Hoffman 1999; Anderson and Hoffman 2006). This

opens the way for less palatable and faster establishing annual grasses and forbs to take over

(Nsinamwa et al., 2005). Constant diminishing of the highly desirable species (Malan and

Van Niekerk, 2005) can result in rangeland deterioration. On the other hand, heavy grazing

depletes foliage of the palatable species, which results in reduced plant vigor (Morris and

Kotze, 2006). Single species grazing systems can have dramatic negative effects on

vegetation composition due to selective grazing (Smet and Ward, 2005). The composition of

the dry matter of the rangeland is very variable depending on the physiological stage of the

grass, species dominating and soil nutritional status (McDonald et al., 1987). Rangeland

forage quality has spatial and temporal variation (Arzani et al., 2006; Laughlin and Abella,

2007). Rangelands that are properly managed normally have more of acceptable species and

higher biomass production (Sisay and Baars, 2002).

Communal rangelands and their associated residential areas make up 13% of the land surface

of South Africa and support a quarter of the country's population and half the country's

livestock (Ward et al., 1998). There has been concern about the state of communally grazed

rangelands in Africa and other parts of the world (Vetter, 2003). Examination on degradation

and productivity of communal rangelands has been based on comparisons between communal

and commercial farming areas (Todd et al., 1998, Ward et al., 1998, Todd and Hoffman,

1999). The communal rangelands are commonly perceived as overstocked, overgrazed,

degraded and unproductive (Lamprey 1983, Sinclair and Fryxell, 1985). Currently, this way

of screening communal grazing systems has come under considerable criticism regarding its

economic and ecological assumptions, and the idea that communal rangelands are necessarily

degraded is now widely challenged (Ellis and Swift 1988, Behnke and Scoones 1993, Behnke

and Abel 1996, Sullivan and Rohde 2002). This study aimed at determining the impacts that

land degradation has on different rangeland parameters namely; species composition,

biomass production and basal cover.

3.2 Description of study sites

The study was conducted in three sites namely the Great Fish RNR, and communal

rangelands of Glenmore (S:33˚07.574’ and E:026˚52.731’) and Ndwayana (S:33˚09.691’ and

42

E:026˚53.035’). The Great Fish River Nature Reserve complex is situated approximately 40

km north of Grahamstown, located between 33º04’ and 33º09’S and 26º49’E. The three areas

under study falls under the Albany Thicket Biome with the specific vegetation types

classified as “Great Fish Noorsveld”, “Bhisho Thornveld” and “Great Fish Thicket” with the

latter being the dominant vegetation type throughout the Great Fish River Reserve complex

(Hoare et al., 2006). The Albany Thicket is a structurally unusual vegetation of the steeply

sloping, semi-arid, river valleys and was first described as Valley Bushveld (Acocks 1953).

Albany Thicket is typically found in semi-arid areas of the Eastern Cape, with between 200

mm and 950 mm mean annual rainfall (Vlok and Euston-Brown, 2002). The areas experience

mean annual precipitation (MAP) that varies from 350-550mm, co-efficient of variation in

MAP is 28-32%, elevation varies from sea-level to 500m, rainfall is bi-modal with peaks in

October-November and then March-April, mean monthly maximum temperatures is 29-32°C

and mean monthly minimum is 4-6°C.

The understorey typically hosts a relatively high diversity of dwarf succulent shrubs and

forbs (mainly Crassulaceae, Aizoaceae), many of which are locally endemic and rare

(Cowling 1983, Johnson et al., 1999, Vlok & Euston-Brown 2002, Vlok et al., 2003), but few

perennial grasses. The wide range of growth forms and taxa typical of Albany Thicket is a

reflection of the transitional nature of thicket vegetation; being an interface between

indigenous forests, Fynbos, Nama-Karoo and Grassland Biomes (Cowling, 1983, Everard,

1987, Palmer, 1990, Kerley et al., 1995, Vlok and Euston-Brown, 2002).The vegetation

comprises of the short thicket type where Portulacaria afra is replaced by Euphorbia bothae,

with increasing aridity (Palmer, 1982; Palmer et al., 1988, Evans et al., 1997). This area

occurs on very shallow clay soils (<1m) derived from the Ecca formations. The vegetation of

the area is mainly charactized by Aloe ferox, Euphorbia bothae and Portulacaria afra

succulents; Rhizogum obovatum shrub and lastly trees Euclea undulate, Boscia oleoides and

Pappea capensis. Satellite maps of the Great Fish RNR, Glenmore and Ndwayana are

presented on figure 3.1, 3.2 and 3.3 respectively.

43

Figure 3.1: Satellite map of the Great Fish River Nature Reserve. (Google Earth, 2015)

Figure 3.2 Sattelite map of Ndwayana. (Google Earth, 2015)

44

Figure 3.3: Glenmore. (Google Earth, 2015)

3.3. Experimental layout

Through visual observations seven Homogenous Vegetation Units (HVU’s) were identified,

comprising of three sites in the Reserve and four outside the Great Fish RNR (two in

Glenmore and two in Ndwayana respectively). These HVU’s were used as experimental sites,

and were demarcated with respect to slope (i.e toplands and bottomlands). One site in the

Great Fish RNR was considered to be the benchmark where conditions were regarded as

optimum for animal production in terms of species composition, ecological stability (basal

cover) and biomass production and this site was compared with the other experimental sites

(Friedel,1999). The extent of land degradation was assessed in terms of botanical

composition (herbaceous and woody species), basal cover, biomass production, soil seed

bank composition and density, and soil characteristics.

3.4 Data collection

3.4.1 Determination of botanical composition and basal cover.

Rangeland condition assessment was conducted on each site using the method described by

Trollope (1990). This method measures the botanical composition of the grass sward and

45

compares it with a reference benchmark site. A benchmark site should possess optimum

vegetation condition for livestock production relative to the specific veld type. Botanical

composition of herbaceous species was determined using the step-point method which was

used according to the condition of the rangelands (Trollope, 1986). The step point method

requires that a determined number of steps be taken on a line (transect) inside a given plot.

The points in each plot are sampled to prevent bias between plant species. A 50m by 100 m

transect was demarcated and four parallel transects (100 m× 20m) were laid in each HVU. To

determine the species composition, 200 points were sampled to make sure that most of the

species in the area were included making that 50 points in each transect. The point in each

transect was marked with a pointer after every two steps and the type of species found there

was identified and recorded.

The grasses were identified to species level, while the other herbaceous plants that belonged

to other families were categorized as forbs, sedges and the karroid species. The grasses were

further classified according to their ecological status such as Decreasers, Increaser I and

Increaser II species and their life forms whether a species is an annual or a perennial. The

Decreaser species are found to be more desirable to grazers than Increaser species in the

rangeland, and they decrease with poor range management (either underutilization or over

utilization). Decreasers increase with proper range management. The Increaser I species are

the less desirable and increase with underutilization or selective grazing. The Increaser II

species increase with over utilization. The plant life form was included as another

classification criterion such as annual and perennial (Solomon et al., 2010). Another species

distinguishing criterion that was used was the grazing value using the following groupings: H

= high grazing value, M = moderate grazing value and L = low grazing value.

The mean point-to-tuft distance provides an estimate for basal cover and an indication of the

vulnerability to soil erosion (Bennet et al. 2012). The basal cover was determined using the

step-point method (Mentis, 1981), where every grass was identified and recorded. If the

pointer points on a bare area the distance between bare and the nearest plant species was

recorded and if the pointer hits the tuft of the plant then zero distance was recorded and

considered as a strike. If the rod struck on a bare area exceeding 40 cm, the area was recorded

as bare (Solomon et al., 2010). The mean point-tuft distance was determined in 200 points

collected per sample site.

46

3.4.2 Determination of biomass production

Seven sample sites were selected and 14 100×50m plots were demarcated according to their

slope (toplands and bottomlands). Biomass production was measured in each sample site and

placed approximately five times along a 100m long transect. These were replicated four

times, making a total of 20 samples in each site. In each plot, two 0.25m2 quadrats were laid

randomly in 10m intervals and any herbaceous material within these quadrants was clipped at

a stubble height of 30mm with hand shears and placed in well labeled sample paper bags. The

moribund material from the previous season was separated from fresh herbaceous plant

material. The harvested samples were oven-dried to a constant mass at 60°C for 48 hours

expressed in kg/ha on a dry matter basis.

3.4.4 Determination of the woody species composition

Woody species composition was determined by identifying and recording all woody plants

within a 200 m2 transect. This was done with the use of a 2 m long calibrated aluminium rod,

which was used to measure the bush height, canopy diameter and the height of the lowest

browsable material (LBM). Four 100 m× 2 m belt transects were laid in the centre of the 50

m×100 m transect. All the woody plants found within the 2 m x 100m transect were

identified and recorded. After species identification, tree height and that of the LBM were

measured according to procedure described by Teague, 1989 by a well calibrated aluminum

rod. Tree density and physiognomic structure were estimated by counting all trees within

200m2

belt transects and the density was expressed as woody plants/hectare. The tree

phytomass was estimated from tree equivalents (TE/ha), a tree which is 1.5 m high (Teague

et al, 1981).

3.4.4 Statistical analysis

Analysis of Variance (ANOVA) was conducted and the Fischer least test using general linear

model (GLM) procedure of SAS (2007) to compare means at (P ≤ 0.05) on herbaceous

species composition, biomass production, basal cover, woody composition, density and tree

equivalents between different HVU’s. The data were also log transformed for mean point to

tuft distance. The interactions between homogenous vegetation units and season on basal

47

cover and biomass production were also tested using general linear model (GLM) procedure

of SAS (2007). For quantitative field data, a completely randomized design (CRD) was

employed. Each of the seven homogenous vegetation units was replicated 4 times.

Outline of the model employed: Yιj (k) = µ + αι (K) + ειj (K)

Where Yιj= Response variables (species composition, biomass production, basal cover,

season).

µ= overall mean

αι(k)=effect of the ιth

HVU,s

ειj(K) = effect of a Random error.

3.5. RESULTS

3.5.1. Overall herbaceous species composition in the selected semi-arid

rangelands.

There were 22 herbaceous species found in Glenmore, Ndwayana and the Great Fish RNR.

The different levels of grazing value amongst the species were as follows High 41%,

Moderate 14% and Low 45%. When considering the life forms, all the grass species were

perennials (95%) with the exception of the forbs as their life form is unknown. In general, the

herbaceous species composition consisted of 59% Increaser II species, 36.4% Decreaser

species and the remaining were 4.54% Increaser 1 species. (Table 4.1)

51

Table 3.1: Herbaceous species composition in Glenmore, Ndwayana and the Great Fish RNR.

ES-Ecological status, GV-Grazing value , LF-Life forms. A-Absent (0%), R-Rare (1-4%), LC- Less common (5-10%), C-

Common(10-15%),D-Dominant(>15%),P-Present(<1%).

Species Ecological Status Grazing value Life Form Top Bottom Benchmark

Cynodon dactylon Increaser II High Perennial R R LC

Sporobolus nitens Increaser II Low Perennial LC LC A

Aristida conjesta Increaser II Low Perennial C LC A

Eragrostis obtusa Increaser II Moderate Perennial A R A

Digitaria eriantha Decreaser High Perennial LC A C

Panicum stapfianum Decreaser High Perennial R A R

Themeda triandra Decreaser High Perennial R A R

Sporobolus fimbriatus Increaser II High Perennial R A D

Sporobolus africanus Increaser II Low Perennial A A D

Heteropogon contortus Decreaser High Perennial R A R

Eustachys paspeloides Decreaser High Perennial R A R

Cymbopogon popischilii Increaser I &III Low Perennial LC A P

Microchloa caffra Increaser II Low Perennial R A P

Setarria sp. Decreaser High Perennial R A A

Eragrostis curvula Increaser II Moderate Perennial A A A

Eragrostis plana Increaser II Low Perennial A A C

Panicum maximum Decreaser High Perennial A A A

Brachiaria serrata Decreaser Moderate Perennial A A A

Eragrostis chloromelas Increaser II Low Perennial A A LC

Forb Increaser II Low Unknown LC LC R

Karroo Increaser II Low Perennial D D P

Sedge Increaser II Low Perennial A A A

52

3.5.2. Species abundances across Homogenous Vegetation Units

Out of the 22 herbaceous species identified, six were the most abundant all HVU’s and these

species were selected to represent the dominant species. These species were Aristida

congesta, Karroid species, Digitaria eriantha, Eragrostis plana, Sporobolus africanus and

Sporobolus fimbriatus. Some of these dominant species were found in the Benchmark and did

not or rarely occurred in the other HVU’s namely Sporobolus fimbriatus, Sporobolus

africanus, Digitaria eriantha and Eragrostis plana (Table 3.2). On the contrary, A.congesta

and the Karroid species were found in all the other HVU’s either on bottomlands and

toplands of the Great Fish RNR, Glenmore and Ndwayana but did not or rarely occurred in

the benchmark. The abundance of A. congesta in Ndwayana toplands and bottomlands was

significantly higher (p<0.05) than in all the other HVU’s (Table 3.2). The Great Fish RNR

bottomlands had the most significantly higher (p<0.05) the abundance of the Karroid species

in all the other HVU’s. The abundance of S. fimbriatus, S. africanus and E. plana was

significantly higher (p<0.05) in the benchmark than in all the other HVU’s, while D. eriantha

was significantly higher (p<0.05) in the Great Fish RNR toplands and the benchmark, but its

occurrence between the two HVU’s was significantly different from each other (Table 3.2).

53

Table 3.2: Mean abundance of the dominant species found in the homogenous vegetation

units

Different superscripts across a column denote significant differences at p<0.05

Site HVUs Aristida

congesta

Digitaria

eriantha

Eragrostis

plana

Karroid

species

Sporobolus

africanus

Sporobolus

fimbriatus

Glenmore Bottomlands 2.25b

0.5c 0.0

b 29.75

c 0.0

b 0.0

c

Toplands 1.75b 0.0

c 0.0

b 40.37

b 0.0

b 0.0

c

Ndwayana Bottomlands 23.75a 0.0

c 0.0

b 17.00

d 0.0

b 0.0

c

Toplands 21.0a

0.0c 0.0

b 7.0

e 0.0

b 0.0

c

Great Fish

RNR

Bottomlands 0.75c 0.0

c 0.0

b 80.25

a 0.0

b 0.0

c

Toplands 2.75b 24

a 0.0

b 19.0

d 0.5

b 1.5

b

Benchmark 0.0d 13.75

b 12.00

a 0.25

f 32.12

a 15.75

a

S.E 1.91 1.98 1.50 4.23 3.09 1.62

54

3.5.3. Biomass production in summer and winter.

Biomass production (kg ha-1

) in summer was greater than in winter in all the HVU’s, except

for Glenmore toplands (Table 3.3) where it was higher in winter than in summer. The

Biomass (kg ha-1

) in both seasons showed no significant differences (p>0.05) between all the

other HVU’s, with the exception of the benchmark and Great Fish RNR toplands, where there

were significant differences (p<0.05) between seasons (Table 3.3). The biomass production in

the benchmark site (2700 kg/ha) was higher in summer than in winter (1715 kg/ha) and was

significantly higher (p<0.05) than all the other HVU’s when compared in both seasons. The

biomass production for the benchmark site in summer was significantly higher (p<0.05) from

the benchmark site in winter (Table 3.3). Both the HVU’s in Great Fish RNR bottomlands

and toplands in summer (1992 and 1209.50 kg/ha) were significantly different (p<0.05)

(Table 3.3).

55

Table 3.3: Mean (S.E) of biomass production in different homogenous vegetation units.

Site HVU’s Summer (kg/ha) Winter (kg/ha)

Great Fish RNR Benchmark 2700.50aA

1715.25aB

Great Fish RNR Bottomlands 1992.50bA

1105.75bA

Great Fish RNR Toplands 1209.50cA

629.00cB

Glenmore Bottomlands 270.25eA

257.25dA

Glenmore Toplands 228.50eA

291.75dA

Ndwayana Bottomlands 318.75eA

293.00dA

Ndwayana Toplands 429.50dA

269.25dA

Standard error 135.39 135.39

Different superscripts across a row indicates significant differences at p<0.05 with small

letters across columns and Capital letters across rows.

56

3.5.4 Basal cover in different homogenous vegetation units.

There was an increasing trend in mean basal cover from the benchmark to Ndwayana

toplands in mean basal cover (0.0-15.75cm) (Figure 3.1). The basal cover for the benchmark

site was significantly lower (p<0.05) than all the other HVU’s, and differed between summer

and winter (Figure 3.1). These results clearly show that the benchmark had more dense cover

(0.0 and 1.5cm) than all of the other HVU’s by having the lowest means for both seasons

(Figure 3.1). In the bottomlands and toplands of Glenmore, basal cover (cm) was

significantly lower (p<0.05) than bottomlands and toplands of Ndwayana. In Glenmore

bottomlands and toplands the basal cover (cm) was significantly different (p<0.05) during

both seasons. Basal cover (cm) in Ndwayana bottomlands were significantly higher (p<0.05)

than Ndwayana top in both seasons and between the HVUs (Figure 3.1).

Figure 3.1: Mean basal cover of all the homogenous vegetation units.

Different superscripts indicate significant differences at p<0.05 between HVU’s.

g

f e

d

c

a b

g

e f

d c

a b

-5

0

5

10

15

20

BA

SA

L C

OV

ER

(C

M)

HOMOGENOUS VEGETATION UNITS

SUMMER

WINTER

57

Seasonality had an effect on the basal cover between the HVU’s (Figure 3.2). There was

significant interaction (p<0.05) between seasonality and HVU on mean basal cover. The

basal cover (cm) was significantly higher in summer than in winter in all of the Homogenous

vegetation units (Figure 3.2). The benchmark sites were significantly different (p<0.05) from

each other when compared in both seasons (Figure 3.2). Similarly to this, the bottomlands of

the Great Fish RNR, Glenmore and Ndwayana were significantly different (p<0.05) between

the two seasons (Figure 3.2).

Figure 3.2: Mean basal cover of season in all the homogenous vegetation units.

Different superscripts indicate significant differences at p<0.05 between seasons.

b

b a

b a

b a

a

a a

a a

a a

-5

0

5

10

15

20

BA

SA

L C

OV

ER

(C

M)

HOMOGENOUS VEGETATION UNITS

SUMMER

WINTER

58

3.5.5. Woody species abundances across Homogenous Vegetation Units.

There were 27 woody species identified at Ndwayana, Glenmore and the Great Fish RNR.

The availability of thorns/spines was prevalent on 41% of the species whilst 59% was the

absence of thorns/spines on the species. Ptaeroxylon obliquum (14%) was the most dominant

and the least dominant was Pappea capensis (0.05%) respectively.

Table 3.4: % abundance, acceptability and the availability of thorns/spines of the woody

species

Species Acceptability Thorns/spines

Abundance

(%)

Coddia ruddis + - 11.84

Grewia robasta + - 13.03

Maytenus capitate - - 1.45

Jatrova capensis - - 10.28

Acacia karroo + + 5.55

Scutia affra + + 0.6

Scutia maytina + + 0.32

Leucas capensis - - 2.65

Rhus Refrecta - + 1.04

Azima tetracantha - + 1.2

Maytenus policantha - - 0.74

Ehretia rigida + - 1.60

Lippia javanica + - 10.99

Ptaeroxylon obliquum - - 14.55

Phyllanthus verrocosus + - 9.25

Brachylaena ilicifolia + - 1.72

Rhizogum obovatum - + 6.03

Caressa haematocarpa + + 0.62

Dovyalis caffra - + 0.1

Pappea capensis + - 0.05

Euphobia triangularis - + 0.54

Opuntia ficus indica - + 2.03

Plambago auriculata - - 0.74

Portulacaria affra + - 1.28

Grewia occidentalis + - 0.93

Diospyros Lycioides + - 1.09

Acacia caffra + + 0.12

59

3.5.6. The dominant woody species at Glenmore, Ndwayana and the Great

Fish RNR.

Six dominant woody species were found in all the homogenous vegetation units namely,

Coddia rudis,Grewia robusta,Jatrova capensis, Ptaeroxylon obliquum, lippia javania and

Phyllanthus verrucosus (Table 3.5). Post-hoc analysis showed that the abundanc of L.

javanica (43%) was significantly higher (p<0.05) in the benchmark site from all the other

HVU’s except for Great Fish RNR topland (3.12%) and bottomland (0.00%). P. obliquum

(38%)was higher (p<0.05) in Ndwayana bottomlands than all the other HVU’s, while

P.verrocosus (27%) was higher (p<0.05) in Ndwayana toplands than all the other HVU’s

(Table 3.5). The abundance of J.capensis (19%) was higher in Great Fish RNR bottomlands

and was significantly higher (p<0.05) from the benchmark and Great fish RNR toplands

while not significantly diffent (p>0.05) from the remaing HVU’s (Table 3.5.6). There was no

significant difference (p>0.05) in the abundance of C.ruddis in the HVU’s (Table 3.5).

Table 3.5: Woody species abundances across HVUs.

Sites HVUs Coddia

ruddis

Grewia

robasta

Jatrova

capensis

Lippia

javanica

Phyllanthus

verrocosus

Ptaeroxylon

obliquum

Glenmore Bottomlands 10.78c 13.31

e 15.47

b 1.29

d 14.38

b 20.53

b

Toplands 7.24d 22.39

b 12.12

c 0.00

e 14.12

b 19.47

b

Ndwayana Bottomlands 10.55c 12.80 10.57

c 14.19

b 2.96

c 37.58

a

Toplands 7.36d 10.57

c 10.57

c 6.44

c 26.72

a 21.97

b

Great Fish

RNR

Bottomlands 6.12d 28.99

a 18.68

a 0.00

e 5.66

c 2.69

d

Toplands 18.82b 1.87

d 0.00

d 3.12

c -0.00

d -0.00

d

Benchmark 20.28a -0.00

e 0.00

d 44.92

a 0.00

d 0.00

d

S.E 5.40 3.51 2.81 4.57 3.89 3.24

60

Different superscripts across a row denote significant differences at p<0.05

3.5.7. Tree equivalents and bush density across homogenous vegetation

units.

Glenmore toplands had significantly (p<0.05) higher tree equivalents (1069 TE/ha) as

compared to the other homogenous vegetation units (Table 3.5.7). Bottomlands and toplands

at Ndwayana had significantly lower (p<0.05) tree equivalent (331.24 and 310.02TE/ha)

respectively (Table 3.6). The bottomlands and toplands of Glenmore had the most highly

significant (p<0.05) bush density (1181.25 and 1337.5Trees/ha) than all the other

homogenous vegetation units and Great Fish RNR toplands had the significantly lower

(p<0.05) bush density (612.5Trees/ha) (Table 3.6).

Table 3.6: Woody density and tree equivalents

Sites HVU’s Tree

equivalents

(TE/ha)

Bush

density

(trees/ha)

Glenmore Bottomlands 371.08c 1181.25

a

Toplands 1069.29a 1337.50

a

Ndwayana Bottomlands 331.25c 725.00

c

Toplands 310.03c 768.75

c

Great Fish

RNR

Bottomlands 402.54b 843.75

b

Toplands 489.04b 612.50

d

Benchmark 424.42b 1000.0

b

S.E 88.68 156.65

Different superscripts denote significant differences across columns at p<0.05

61

3.6. Discussion

3.6.1 Species seasonal abundances across HVU’s

Plant species may vary significantly in their acceptability and response to grazing herbivores

due to the differences in palatability (Abule et al., 2007). Although the grazing practices are

the same in communal rangelands, such practices lead to the variation on species composition

at the different vegetation types; this is ascribed to the climatic variation between the

vegetation types (Abule et al., 2007). Perennial grasses (95%) were more dominant in all the

HVU’s under study namely Glenmore, Ndwayana, the Great Fish RNR and the benchmark

(Table 3.1) and these results are in contrast to the findings by Solomon et al. (2007), where

he compared different grazing systems and reported that the frequency of annuals was higher

in communal rangelands when compared to other grazing systems. There was also a high

prevalence of species with low grazing value (45%) when comparing all the Homogenous

vegetation units (Table 3.1). Solomon et al. (2007) further indicated that, this could be as a

result of perennial grasses being replaced by annual grasses in communal rangelands than in

the other grazing systems due to the higher grazing pressure.

The benchmark site was in an open/grassy area and it was more diverse in species

composition and had less woody species were available. The results of this study showed that

the benchmark site was mostly dominated by Sporobolus fimbriatus, Sporobolus africanus,

Eragrostis plana and Digitaria eriantha (Table 3.2). A benchmark site is where conditions

are regarded optimum for animal production in terms of species composition, ecological

stability (basal cover) and biomass production and this site was compared with the other

experimental sites (Friedel, 1999). In this particular HVU, species such as Themeda triandra

are present but are not dominant. The species found in the benchmark were Increaser II

species (except for D. eriantha) which was almost the least dominant in the benchmark.

Increaser II species are said to dominate in a veld that is poorly managed and increase with

an increase in stocking rate (overutilization) (Lesoli, 2008). Increaser II species dominate in

areas that are overgrazed. Decreaser species are dominant in a veld that is in a good

condition but decrease when the veld is overutilized or underutilized (Van Oudtshoorn,

2006). These types of grasses such as (Themeda triandra) are palatable and preferred by

grazing animals (Van Oudtshoorn, 2006). These results found in the benchmark can be

62

ascribed to the fact that in rangelands, depending on the history of management there are

plant species that are more palatable (Decreaser) to grazing and those that are less acceptable

(Increaser). Therefore, when the animals are grazing in one area they tend to select species

that are more palatable leaving the less acceptable species not grazed (Lesoli, 2008).

Vegetation type also plays a role in determining species abundances, and in this type (valley

bushveld) perennial grasses such as T.triandra are not common, even when the rangeland is

properly managed. Depending on the severity of defoliation and reserved carbohydrates some

species that are grazed repetitively are not given sufficient time to recover and they lose plant

vigor. In supporting of these results, the benchmark site was in an open area that means that

the grazers preferred it for grazing than the other areas which had more trees which resulted

to selective grazing of the more acceptable species. The veld type also played a significant

role in these results as the Valley Bushveld is under the Albany Thicket Biome. The

dominant species in the Albany Thicket Biome are a high standing biomass of woody and

succulent shrubs (Aucamp et al., 1982). While conducting the study, visual observations were

made and there was a prevalence of some Decreaser species such as D. eriantha, T. triandra,

Panicum species but they were not dominant. Plant species that are not grazed increase in

numbers and vigour because they get more time to grow and produce more seeds and

subsequently become more dominant in an area (Lesoli, 2008). The higher grazing intensity

in rangelands result in changes on the vegetation structure, composition and productivity

(Oztas et al., 2003 and Maki et al., 2007).

Degradation was mainly expected in the communal areas than in the reserve. Surprisingly,

the bottomlands of the Reserve were dominated by Aristida congesta and the Karroid species

(Table 3.2) which were the same as the results found in both communal areas Glenmore and

Ndwayana. The toplands were dominated by D.eriantha followed by the Karroid species then

A.congesta. A.Congesta is an Increaser II species with a low grazing value and is said to be

more prevalent in disturbed soils. In most areas it is a good indicator of veld degradation

when common (Van Oudtshoorn, 2006). D. eriantha is a Decreaser species which is more

prevalent in the toplands of the Great Fish Reserve is a palatable grass that is regarded as one

of the best natural and cultivated pastures in Southern Africa (Van Oudtshoorn, 2006). Its

dominance in a rangeland indicates a veld in good condition and is suitable for grazing

animals. Under heavy grazing pressure Decreaser species disappear and are replaced by

Increaser or invader species (Sisay and Baars, 2002).

63

The dominance of A. congesta and the Karroid species in these rangelands whether in

communal or the Great Fish RNR can be ascribed to localized grazing patterns, which have

resulted in the variation in degradation intensity along the landscape (Lesoli, 2008). As a

result of this, different grazing pressure within the different landscapes of the rangelands

could result in changing the species composition (Hays and Holl, 2003). According to the

results of species composition, all the communal rangelands as well as the Great Fish RNR

bottomlands were degraded as they were dominated by Increaser II species (A.congesta and

Karroid species respectively). Moreover, literature states that degradation in the communal

grazing areas could be a result of overgrazing (Versbug and van Keulen, 1999; Lesoli, 2008).

The impact of high grazing pressures have been explained by researchers and conclusions

made were that African communal rangelands frequently support high numbers of livestock,

and often exceed advised carrying capacity levels (Abel, 1993; Scoones, 1993; Tapson, 1993;

Ward et al., 1998). Contrastingly, commercial or game ranchers apply much lower stocking

rates in order to produce high-quality products for markets. Therefore, degradation in the

Reserve could be ascribed to the phenomenon of selective grazing and climatic variations.

Moreover, degradation in the Great Fish RNR could mainly be because of natural factors,

such as excessive proliferation of game animals, soil innate properties and climatic variables

(Smet and Ward, 2006). The abundance of the less palatable grasses in these rangelands can

be used as an indicator of degradation, a method which is in agreement with Malan and Van

Nierkerk (2005) who indicated that certain species characterize different succession stages

during grassland retrogression and they could be severe as characteristic attributes of

rangeland degradation. Vegetation indicators for rangeland degradation serve as the early

warning system for degradation and can subsequently justify early intervention (Ward and

Smet, 2006).

3.6.2 Seasonal biomass production across the HVU’s.

Biomass production (kg/ha) was higher in summer than in winter (Table 3.3) which can be a

result of the high rainfall and temperatures in summer (Sherry et al., 2008).This is to be

expected, considering the differences between the dormant and growing seasons. Similarly,

Angassa and Oba (2010), reported that biomass production during the dry season is less when

compared to the wet season. The high summer rainfall has been ascribed to an increase in leaf

64

area and leaf production (Angassa and Oba, 2010). The acceptable state for biomass

production as stated by Teague et al. (2009) is 800kg/ha while the recommended threshold

for livestock production in the rangelands is (1500kg/ha). Homogenous vegetation units had

an effect on biomass production whish was similar to the results by Lesoli (2008) where he

reported biomass production was different between communities in a study around Alice,

Eastern Cape. The benchmark in summer (2700 kg/ha) had the greatest biomass production

than the benchmark in winter (1715 kg/ha) and both HVU’s were significantly different from

each other (Table 3.3). Following the benchmark was the HVU’s that were also selected in

the Reserve (Great Fish RNR bottomlands in summer and winter). Both these HVU’s were

not significantly different from each other as shown by (Table 3.3). Season had an effect in

biomass production (kg/ha) in summer and winter in Great Fish toplands and they were

significantly different. Great Fish RNR toplands (summer and winter) had followed Great

Fish RNR bottomlands in biomass production (table 3.3). There was no seasonal variation in

biomass production in these HVU’s. Ndwayana toplands and bottomlands in summer had the

greater biomass production than Ndwayana toplands and bottomlands in winter. These results

(Table 3.3) show that season did not have an effect on the biomass production of these

HVU’s. The bottomlands of the Great Fish RNR and Glenmore had high biomass production

when compared to the toplands in both seasons (Table 3.3). In contrast to the other results,

Ndwayana bottomlands had low biomass production when compared to Ndwayana toplands

in both seasons (Table 3.3).The results in (Table 3.4.3) showed that the sites found in the

Great Fish RNR had higher biomass production when compared to the HVU’s that were

found in the communal areas in both seasons. Climatic conditions and grazing have marked

influences on biomass production (Fynn and O`connor, 2000; Savadogo et al., 2006; Angassa

and Oba, 2010).

3.6.3 Basal cover in the toplands and bottomlands of Glenmore, Ndwayana

and the Great Fish RNR.

The results (Figure 3.1) indicate that the benchmark had the highest basal cover in both

seasons (summer and winter) than all the other homogenous vegetation units, followed by the

Great fish bottomlands and toplands. When comparing the HVU’s in the Great Fish RNR,

seasonality had an effect because in all three including the benchmark, the basal cover

65

obtained was greater in summer than in winter (Figure 3.1) which is in contrast to the results

of Sisay and Baars (2002), where they found no significant differences in basal cover

between benchmarks and seasonally grazed areas on the one hand, and roadsides and

lakeshores on the other hand in the Rift Valley, Ethiopia. The results concur with the results

where bare ground was more common under communal grazing than the other land use

systems (Solomon et al., 2007). In the communal rangelands Glenmore had the highest basal

cover than Ndwayana and these areas comprised of four HVUs namely Glenmore toplands

and bottomlands and Ndwayana toplands and bottomlands. The results in (Figure 3.1)

provide scientific evidence of the extent of degradation which was observed in these

rangelands when considering ecological stability as one of the indicators of land degradation.

The results revealed that basal cover differed between homogenous vegetation units in the

different sites (Figure 3.1).

The results of this study concur with those of Lesoli (2008) where it was found that basal

cover was significantly different between different communities and between different

slopes. This could be ascribed to topography: Ndwayana had greater landscape heterogeneity

leading to spatial preferential grazing and is more characterized by steeper slopes than the

other communities which lead to accelerated runoffs. The implication made by these results is

that the grazing pattern and landform had an effect on basal cover possibly due to increased

plant interspaces’ and run off rate (Lesoli, 2008). Parsons et al. (1997) reported that the

bottomlands have higher tuft density, basal area, and abundance of poorly palatable species

which implies that selective grazing leads to reduction on the basal cover.

The results of the study revealed that season had an effect on basal cover (cm) in the HVU’s

found in Glenmore communal rangeland. In winter Glenmore bottomlands had greater

ecological stability when compared to Glenmore bottomlands in summer (Figure 3.1). The

results show that (Figure 3.1), for all the HVU’s found in this communal area, Glenmore

bottomlands had the lowest mean in both seasons than Glenmore toplands. Therefore, the

bottomlands of Glenmore had more basal cover than the toplands. The toplands and

bottomlands of Ndwayana were also affected by season where, the basal cover was less in

winter than in summer. These results show that there is more degradation in the lower parts

of Ndwayana than the upper parts which is in contrast to what the results show in Glenmore

and the Great Fish RNR. Where, the toplands had more basal cover than the bottomlands

(Figure 3.1).

66

The ecological status of these rangelands refers to the grouping of these grasses based on

their reaction to different levels of grazing (Lesoli, 2011). Grass species react to grazing in

one of two ways, it can either increase or decrease which supports the concept of ecological

stability indicated by (Van Oudtshoorn, 2009). The mechanism through which rangeland

vegetation species change as a result of grazing pressure could be related to the repetitive

removal of leaves from acceptable species, which weakens the plant reserves useful for

recovery after defoliation. In agreement to the foregoing assertion, grasses lose their vigor

because of the repeated removal of leaves and constant draining of their nutrient reserves as

indicated by (Malan and Van Nierkerk, 2005). As a result of this, a plant will be unable to

replenish the stored resources resulting in the failure to produce new leaves eventually

reduces the plants photosynthetic power (Morris and Kotze, 2006). As the desirable species

become weaker and die off, the number of roots in the upper layer of the soil decreases

resulting to a reduced competitive ability of grasses later forming bare areas (Sisay and

Baars, 2002). The dominance of certain species and their density in communal rangelands

bears implications to basal cover, which in turn indicates rangeland degradation (Lesoli,

2011). Therefore, poor/low basal cover, low plant density and poor botanical composition

could be considered the characteristic features of land degradation in communal areas and

part of the Great Fish RNR.

3.6.5 Woody species composition in Glenmore, Ndwayana and the Great

Fish RNR.

The woody species composition of Glenmore, Ndwayana and the Great Fish RNR were

represented by a high abundance of browsable species namely Lippia.javanica,

Grewia.robasta and Coddia.ruddis and some thicket species (Table 3.5). The six dominant

species found in all the homogenous vegetation units namely, C.ruddis,G.robasta,Javanese.

capensis, Ptaeroxylon.obliquum, L.javania and Phyllanthus.verrucosus were signs of change

in vegetation structure in these areas. The benhcmark and Great Fish RNR toplands were

mainly dominated by C.ruddis and L.javanica which are highly acceptable to browsers. The

Great Fish RNR bottomlands was dominated by G.robusta and J.capensis. The two

communal areas were dominated by G.robusta, P.obliquum and P.verrocosus. Ndwayana

communal area had more of P.obliquum which in not acceptable for browsing by livestock.

G.robusta is an acceptable species to browsing game animals and its valued for different uses

67

even as fodder for livestock (Van Wyk et al., 2000). Ptaroxylon obliquum is a legume that is

not browsed by animals and its mainly used for medicinal and traditional purposes. The wood

contains highly irritant resin and the resin is used to kill ticks on cattle and warts in humans

(Van Wyk et al.,2000). Powdered wood is used as snuff to treat headaches, while infusions

are taken for rheumatism and heart disease. To some extent, clearing of this species would be

appropiate as it is not browsed by livestock. Its abundance is only reducing the rangelands

capacity for browsing but is also of high economic and health importance to the community.

The results of this study showed that bush encroachment in these rangelands is not the main

factor leading to land degradation. This notion is supported by the fact that, Abate et al.

(2010) and Gemedo Dalle et al. (2006) considered threshold of 2400 trees ha-1

as a barrier

between bush encroached and non-encroached rangeland. However, woody density alone

cannot be considered as a single factor affecting competitive behavior against herbaceous

species (Abule et al., 2007). Therefore, tree equivalents beyond the threshold (2500 TE ha-1

)

are a true mirror of highly encroached condition in a given rangeland (Richter et al., 2001).

True, but vegetation type is also considered, where some naturally have dense tall trees. The

sites under study have tree density range of 612.50-1181 trees/ha and a range of 310. -1069.

TE/ha tree equivalents. The results of the study are in contrast to the given thresholds for both

tree density and tree equivalents. A curve which relates to tree density and biomass

production was developed which indicates that biomass production increases proportionally

with bush density up to a certain point and later declines (Aucamp et al., 2001). The results of

the study are in contrast to the results conducted by other scientific researchers specifically

that of ( Mndela, 2013 and Libala,2014) where they found that bush encroachment was a

problem in the communal rangelands of the Eastern Cape and the species that was dominant

in their study was Acacia karroo and however A.karroo is not a common species study areas.

The continuation of bush encroachment has been associated with a number of factors

(Glasscock et al., 2005).

Overgrazing and the suppression of fire could be the most probable causes of degradation.

Scholes and Archer (1997) indicated that there is a strong, negative correlation between tree

density or cover, grass cover and biomass. Glenmore toplands had high tree equivalents and

tree density than all the other sites. The increase in woody plant encroachment jeopardizes

grassland productivity and species biodiversity and threatens the sustainability of pastoral

subsistence and commercial livestock grazing (Richter et al., 2001). Bush encroachment has

been found to have an adverse influence on grass biomass production and decreases potential

68

grazing capacity of rangeland (Richter et al., 2001) which could be a result of the low

biomass production in Glenmore (toplands) when compared to the Great Fish RNR.

4.7 Conclusions

The study showed that there was a high abundance of perennial grasses in the species

composition yet these species were mainly the Increaser II species. Glenmore and Ndwayana

communal areas, as indicated in the literature, were more degraded than the sites in the Great

Fish RNR when looking at species abundance. The benchmark site was also dominated by

Increaser II species, which is not the case in more productive vegetation types, but can be

attributed to the vegetation type and selective grazing. Ndwayana, Glenmore and the Great

Fish RNR areas were dominated by the unpalatable Karroid species and Aristida.congesta.

Degradation in these areas was also a result of the reduction in biomass production and basal

cover as we moved from the Great Fish RNR to the communal areas. The increase in

degradation due to the reduction of biomass production causes measurable impacts on

livestock production as it leads to the decrease in herbage production.

There was no remarkable prevalence of bush encroachment from the communal areas

entering the Reserve. Degradation as a result of bush encroachment was not the case in these

areas and therefore more evaluation based on other factors such as soil erosion need to be

considered as the main cause of degradation. In addition to that, the farmer’s perceptions

need to be included. Ndwayana and Glenmore provide clear signs of high degradation when

considering species composition, biomass production and basal cover. Furthermore, the Great

Fish RNR was also degraded in terms of species composition but had high biomass

production and basal cover as compared to the communal rangelands. The benchmark site

had proved not to be very ideal for livestock production as the study revealed that the

dominant species observed were indeed perennial grasses but were also Increaser II species.

The prevailing condition of the benchmark site was attributed to the issue of selective grazing

and veld type. Land use management history is very significant to incorporate better

understanding and rehabilitation of these rangelands. Therefore, control measures to halt

degradation need to be taken into consideration in these semi-arid rangelands noting that bush

encroachment is not a problem.

69

References

Abate T, Ebro A and Nigatu L 2010. Traditional rangeland resource utilisation practices and

pastoralists perceptions on land degradation in South-east Ethiopia. Tropical Grasslands. 44:

202 – 212.

Abel N O J 1993. Reducing cattle numbers on Southern African Communal Range: Is itworth it? In:

Behnke, R.H., Scoones, I. & Kerven, C. (Eds), Range Ecology at Disequilibrium, pp. 173–

195. London: Overseas Development Institute.

Abule E, Snyman H A, and Smit G N 2007. Rangeland evaluation in the middle Awash valley of

Ethiopia: I. Herbaceous vegetation cover. Journal of Arid Environments, doi: 10.1016

2006.12.008.

Acocks JPH 1953. Veld Types of South Africa. Botanical Survey of South Africa Memoir No.28.

Government Printer, Pretoria. 1-192.

Anderson P M L and Hoffman M T 2006. The impacts of sustained heavy grazing on plant diversity

and composition in lowland and upland habitats across the Kamiesberg mountain range in the

Succulent Karoo, South Africa. Journal of Arid Environments.

Doi:10.1016/j.jaridenv.2006.05.017.

Angassa A and Oba G 2010. Effects of grazing pressure, age of enclosures and seasonality on bush

cover dynamics and vegetation composition in southern Ethiopia. Journal of Arid


Arzani H, Basiri M, Khatibi F, and Ghorbani G 2006.Nutritive value of some Zagros Mountain

rangeland species. Small ruminant research 65: 128- 135.

Aucamp A J 1976. The role of the browser in the Bushveld of the Eastern Cape. Proceedings of the

Grassland Society of Southern Africa. 11: 135 – 138.

Aucamp A J, Danckwerts J E, Teague W R and Venter J J 1982. The role of Acacia karroo in the

False Thornveld of the Eastern Cape. Proceedings of Grassland Society of Southern Africa.

18: 151 – 154.

70

Aucamp A J, Danckwerts J E , Teague W R and Venter J J 1983. The role of Acacia karoo in the

False Thornveld of the Eastern Cape. Proceedings of Grassland Society of Southern Africa.

18: 151 – 154.

Behnke R H and Abel N O J 1996. Revisited: the overstocking controversy in semi-arid Africa.

World Animal Review 87, 3–27.

Behnke R H, Scoones I and Kerven C.(Eds.) 1993, Range Ecology at Disequilibrium: New Models

of Natural Variability and Pastoral Adaptation in African Savannas. Overseas Development

Institute, London (248pp).

Bennett J E, Lent P C and Harris P J C 2007. Dry season foraging preferences of cattle and sheep in

a communal area of South Africa. African Journal of Range and forage Science 24 (3):109-

121.

Bennett J, Ainslie A and Davis J 2012.Fenced in: Common property struggles in the management of

communal rangelands in central Eastern Cape Province, South Africa. Land Use Policy 27:

340 – 350.

Cowling RM 1983. Phytochorology and vegetation history in the south-eastern Cape South Africa.

Journal of Biogeography: 393-419.

Ellis J E and Swift D M 1988. Stability of African pastoral ecosystems: alternate paradigms and

implications for development. Journal of Range Management 41, 450–459.

Evans NV, Avis AM and Palmer AR 1997 Changes to the vegetation of the mid-Fish River valley,

Eastern Cape, South Africa, in response to land-use, as revealed by a direct gradient analysis.

African Journal of Range and Forage Science 14, 68-74.

Everard DA 1987. A classification of the subtropical transitional thicket in the Eastern Cape, based

on syntaxonomic and structural attributes. South African Journal of Botany 53: 329-340.



71



Friedel M H 1997. Discontinuous change in arid woodland and grassland vegetation along Gradients

of cattle grazing in central Australia. Journal of Arid Environments 37: 145-164.

Fynn R W S and O'connor T G 2000. Effect of stocking rate and rainfall on rangeland dynamics and

cattle performance in a semi-arid savanna, South Africa. Journal of Applied Ecology. 37: 491

– 507.

Gemedo-Dalle 2004. Vegetation ecology, rangeland condition and forage resources evaluation in the

Borana lowlands, southern Oromia, Ethiopia. Ph.D Thesis, Georg- August Universitat

Gottingen, Germany, Cuvillier Verlag Gottingen.

Gemedo-Dalle, Maass B L and Isselstein J 2006. Rangeland condition and trend in the semiarid

Boran lowlands, southern Oromia, Ethiopia. African Journal of Range and Forage Science

23 (1): 49- 58.

Glasscock S N, Grant W E, Drawe D L 2005. Simulation of vegetation dynamics and management

strategies on South Texas, semi-arid rangeland. Journal of Environmental Management. 75:

379 – 397.

Hayes G E and Holl D K 2003.Cattle grazing impacts on annual forbs and vegetation composition of

mesic grassveld in California. Conservation Biology 17(6): 1694-1702

Hoare DB 1997. Syntaxonomy and synecology of the grasslands of the southern parts of the Eastern

Cape. M.Sc thesis, Department of Botany, University of Pretoria.

Hoare DB and Bredenkamp G. 1999 Grassland communities of the Amatola/Winterberg mountain

region of the Eastern Cape, South Africa. South African journal of Botany, Vol 65, Issue 1,

75-82

Johnson CF, Cowling RM and Phillipson PB 1999. The flora of the Addo Elephant National .Park,

South Africa: are threatened species vulnerable to elephant damage? Biodiversity and

Conservation 8, 1447-1456.

72

Jordaan F P 1997. Implementation of a conceptualized system for assessing rangeland condition and

monitoring in a number of key grazing areas of the western grassland biome. PhD Thesis,

Potchefstroom University for Christian Higher Education,

Kerley G I H, Knight M H & de Kock M 1995. Desertification of Subtropical thicket in the Eastern

Cape, South Africa: are there alternatives? Environmental Monitoring and Assessment 37:

211-230

Kunst C, Monti E, Perez H and Godoy J 2006. Assessment of the rangelands of Southwestern

Santiago del Estero, Argenina, for grazing management and research. Journal of

Environmental Management. 80: 248 – 265.

Lamprey H F 1983. Pastoralism yesterday and today: the overgrazing controversy. In: Bourlie` re, F.

(Ed.), Tropical Savannas. Ecosystems of the World, vol. 13. Elsevier, Amsterdam, 730pp.

Laughlin D C and Abella S R 2007. Abiotic and biotic factors explain independent gradients of plant

community composition in ponderosa pine forests. Ecological modelling,

ecolmodel.2007.02.018.

Lesoli M 2011. Characterisation of communal rangeland degradation and evaluation of vegetation

restoration techniques in the Eastern Cape, South Africa. PhD. Thesis, University of Fort

Hare, Alice, South Africa.

Lesoli M 2008. Vegetation and soil status and human perceptions on the condition of communal

rangelands of the Eastern Cape, South Africa. MSc. Dissertation, University of Fort Hare,


Maki A, Kenji T, Kiyokazu K and Teruo H 2007.Morphological and physico-chemical charateristics

of soils in a steppe region of the Kherlen River basin, Mongolia. Journal of Hydrology 333:

100- 108.

Malan P W and Van Niekerk 2005.The extent of grass species composition in Braklaagte, Zeerust

District, North-West Province, South Africa. African Journal of Range and forage Science

22(3): 177- 184.

73

McDonald P, Edwards R A and Greenhalgh J F 1987.Animal Nutrition.4th ed. Longman group,

Essex.

Mentis, M.T., 1981. Evaluation of the wheel point and step-point methods of veld condition

assessment. Proceedings of the Grassland Society South Africa 16, 89–94.

Mndela M. 2013. Evaluation of range condition, soil properties and seed banks, and farmer’s



Morris C and Kotze D 2006. Introduction to Veld care (1). Agricultural research council, University

of Kwazulu Natal.



22 (1): 17- 28.

Oba G, Stenseth N C and Lusigi W J 2000.Grazing management in arid zones of Sub- Saharan

Africa. BioScience 50 (1): 35- 51.


overgrazed and eroded rangelands. Journal of Arid Environments 55: 93-100.

Palmer A R, Tanser F and Hintsa M D 1997. Using satellite imagery to map and inventorise

vegetation status for Eastern Cape Province. Unpublished report, Range and Forage Institute,

Grahamstown.

Palmer AR (1982) A study of the vegetation of the Andries Vosloo Kudu Reserve. MSc Thesis

Rhodes University, Grahamstown.

Palmer AR 1990 A qualitative model of vegetation history in the Eastern Cape midlands, South

Africa. Journal of Biogeography 17, 35-46.

Palmer AR, Crook BJS, Lubke RA (1988) Aspects of the vegetation and soil relationships in the

Andries Vosloo Kudu Reserve, Cape Province. South African Journal Botany. 54, 309-314.

74

Person B T, Herzog M P, Ruess R W, Sedinger J S, Anthony R M and Babcock C A 2003. Feedback

dynamics of grazing lawns: coupling vegetation change with animal growth. Oecologia 135:

583-592.

Pratt D J and Gwynne M D 1977. Range Management and Ecology in East Africa. Hodder and

Stoughton, London, UK. 310pp.

Richter C G F, Snyman H A, and Smit G N 2001.The influence of tree on the grass layer of three

semi-arid savanna types of Southern Africa. African Journal of range and forage science 18:

103- 109.

SAS Institute Inc. 2007. SAS software. Frame entry usage and reference. Version 8. Cary, NC.

Savadogo P, Sawadogo L, and Tiveau D 2006. Effects of grazing intensity and prescribed fire on

soil physical and hydrological properties and pasture yield in the savanna woodlands of

Burkina Faso. Agriculture, Ecosystems and Environment. 11: 80 – 9.

Scholes R J and Archer S R 1997.Tree-grass interactions in Savannas.Annual review of Ecology and

systematics 28: 517- 544.

Scoones I 1992. Coping with drought: responses of headers and livestock in contrasting

savanna environments in southern Zimbabwe. Human Ecology 20 (1): 293- 314.

Sherry R A, Weng E, Arnone J A, ohnson D W, Schimel D S, Verburg P S, Wallace L L and Luo Y

2008. Lagged effects of experimental warming and doubled precipitation on annual and

seasonal aboveground biomass production in a tall grass prairie. Global Change Biology.14:

2923 – 2936.

Sinclair A R E and Fryxell J M 1985. The Sahel of Africa: ecology of a disaster. Canadian Journal of

Zoology 63, 987–994.

Sisay A and Baars R M T 2002.Grass composition and rangeland condition of the major grazing

areas in the mid Rift Valley, Ethiopia. African of range and forage Science 19: 161- 166.

Smet M and Ward D 2005. A comparison of the effects of different rangeland management systems

on plant species composition, diversity and vegetation structure in a Semi-Arid Savanna.

African Journal of Range and Forage Science 22 (1): 59- 71.

75

Smet M and Ward D 2006. Soil quality around water points under different management systems in

a semi-arid savanna, South Africa. Journal of Arid Environment 64, 251–269.


systems and distance from water in a semi-arid rangeland of southern Ethiopia. South African

Journal of Botany 72: 263- 271.

Solomon T B, Snyman H A and Smit G N 2007.Cattle-rangeland management practices and

perceptions of pastoralists towards rangeland degradation in the Borana zone of southern

Ethiopia. Journal of Environmental Management 82: 481- 494.

Solomon, T.B., and Mlambo, V., 2010. Encroachment of Acacia brevispica and Acacia

drepanolobiumin semi-arid rangelands of Ethiopia and their influence on sub-canopy grasses.

Research Journal of Botany. 5: 1 – 13.

Solomon, T.B., and Mlambo, V., 2010a. Encroachment of Acacia brevispica and Acacia

drepanolobium in Semi-Arid rangelands of Ethiopia and their influence on sub-canopy

grasses. Research Journal of Botany. 5: 1 – 13.

Solomon, T.B., Dlamini, B.J., and Dlamini.A.M. 2010b. Changes in soil characteristics and grass

layer condition in relation to land management systems in the semi-arid savannas of

Swaziland. Journal of Arid Environments. 74: 675 – 684.

Sullivan, S., Rohde, R., 2002. On non-equilibrium in arid and semi-arid grazing systems. Journal of

Biogeography 29, 1595–1618.

Tapson D 1993. Biological sustainability in pastoral systems: the KwaZulu case. In: Behnke, Jr.,

R.H.,Scoones, I., Kerven, C. (Eds.), Range Ecology at Disequilibrium: New Models of

Natural Variability and Pastoral Adaptation in African Savannas. Overseas Development

Institute, London, pp. 118–135.

76

Teague W R, Kreuter U P, Grant W E, Diaz-solis, H and Kothmann M M 2009. Economic

implications of maintaining rangeland ecosystem health in a semi-arid savanna. Ecological

Economics. 68: 1417 – 1429.

Teague W R, Trollope W S W and Aucamp A J 1981. Veld management in the semi-arid bush-grass

communities of Eastern Cape. Proceedings of Grassland society of Southern Africa. 16: 23 –

28.

Todd S W and Hoffman M T 1999. A fence-line contrast reveals effects of heavy grazing on plant

diversity and community composition in Namaqualand, South Africa. Plant Ecology 142:

169- 178.

Trollope W S W 1986. Land use surveys: Assessment of veld condition in Ciskei. In: Republic of

Ciskei National Soil Conservation Strategy (1). Department of Agriculture and Forestry,

Ciskei.

Trollope W S W, Trollope L A and Bosch O J H 1990.Veld and Pasture management terminology in

southern Africa. Journal of Grassland Society of Southern Africa 7: 52 –61.

Van Oudtshoorn. F P. 2006.A guide to grasses of South Africa. Second edition, Fifth impression.


Van Wyk B, Van Wyk P and Van Wyk B E 2004. Guide to Trees of South Africa (2nd Edition).

Briza Publications, Pretoria, South Africa.

Verburg P H and van Keulen H 1999. Exploring changes in the spatial distribution of

Vetter S Goqwana W M Bond W J and Trollope W S W. 2006. Effects of land tenure, geology and

topography on vegetation and soils of two grassland types in South Africa. African Journal of

Range and Forage Science. 23: 13 – 27.

Vlok J H J and Euston-Brown D I W 2002. The patterns within, and the ecological processes that

sustain, the Subtropical Thicket Vegetation in the planning domain for the Subtropical

Thicket Ecosystem Planning (Step) Project. Terrestrial Ecology Research Unit, University of

Port Elizabeth, Port Elizabeth. Report No. 40

77

Walley R D B and Hardy M B 2000. Measuring botanical composition of grasslands. In: Mannetje,

L., and R.M. Jones (Eds). Field and laboratory methods for grassland and animal production

research. CABI publishing, Pp 67 – 102.

Ward D ,Ngairorue B T, Kathena J, Samuels R and Ofran, Y., 1998. Land degradation is not a

necessary outcome of communal pastoralism in arid Namibia. Journal of Arid Environments

40, 357–371.

78

CHAPTER 4. SOIL CHEMICAL PROPERTIES AT

GLENMORE, NDWAYANA AND THE GREAT FISH RIVER

NATURE RESERVE.

ABSTRACT

A study was conducted to document the extent of land degradation in Glenmore, Ndwayana

and the Great Fish River Nature Reserve. The objective was to evaluate the effect of seasonal

variation and vegetation type on soil macro and micro nutrients and pH in rangelands of

Glenmore, Ndwayana and the Great Fish RNR. The soil chemical characteristics were

determined following the methods of soil chemical analysis and the concentrations were read

under the photo-spectrometer. The macro and micro nutrients determined were N, P, K, Mg,

Na, Ca, Cu, Mn, Zn pH and Organic carbon. The results of the study showed that season had

no effect on the concentration levels of these nutrients in all the study sites. There were

significant differences (p<0.05) in the occurrence of these nutrients in Glenmore and

Ndwayana rangelands. There were significant differences (p<0.05) of N, P, K, Mg, OC, Na

and Ca in the toplands and bottomlands found in the Great Fish Nature Reserve.

Homogenous vegetation units had an effect on the micro nutrients and there were significant

differences (p<0.05) in the occurrence of Cu, Zn, Mn and pH in the toplands and bottomlands

of the Great Fish RNR. There were no clear trends of these nutrients but fluctuations when

compared between the different sites according to whether its toplands or bottomlands. The

concentration levels of Mn and Zn were higher in the communal areas than in the Great Fish

RNR.

Keywords: Homogenous vegetation units, season, micro and macro nutrients

79

4.1.1 Introduction

Assessment of communal and commercial rangeland capability is crucial in order to prevent

resource degradation and facilitate adaptive management practices (Rezaei et al., 2006), and

soil properties form an important part of range condition assessment. Soil forms the basis for

all vegetation growth and plays a key role in the hydrological, carbon and nutrient cycles of

ecosystems (Li et al., 2007). It is an important bio-physical rangeland resource (Rezaei et al.,

2006). Soil is characterized by three quality indicators which are physical, chemical and

biological component and these require attention to evaluate the functional capacity of the

soil resource in rangelands. The modification of these components in a short period of time is

a result of the quality indicators (Mojiri et al. 2012.). There is little information available on

soil properties in communal rangelands (Snyman, 1998), and the use of the quality indicators

is mostly of importance to add value and to obtain high precision of evaluation and trend

analysis (USDA-NRCS,2001). The study of the variations in soil properties resulting from

topographic aspect and vegetation changes later has implications on the proper management

and environmentally sensitive areas (Yimer et al.,2006).

Overgrazing, transformation of rangelands, forests and deforestation are considered as the

driving forces of the diminution in soil quality indicators (Nael et al., 2004). In rangeland

science, soil chemical properties that are of interest are soil acidity and salinity as they

indirectly or directly impact the vegetation growth through elimination of vital nutrients in

the soil and they mask the access of these nutrients to vegetation (Herrick, undated). Soil

properties on rangelands may vary in space and time due to natural and anthropogenic

disturbances (Kariuki et al., 2010). Dry matter and mineral availability can vary according to

soil (Ramirez et al., 2004), season (Scholes and Walker, 1993; Mcdonald et al., 1996) species

composition (Tefera et al., 2009) and topography (Gizashaw et al., 2002). Gizashaw et al.

(2002), reported that during the wet season, the mean concentration of most minerals in

forages tend to be higher for low lying areas than uplands. They further reported that the high

mineral concentrations of pasture in bottom land is due to the relatively higher soil mineral

and organic matter levels at the lower slopes and the predominance of species adapted to high

fertility environments. This study focused on determining the macro and micro nutrients and

soil pH and their contribution to soil degradation at Glenmore, Ndwayana & the Great Fish

RNR.

80

4.1.2 Soil sampling in Glenmore, Ndwayana and the Great Fish RNR.

Soil samples were collected using a soil auger from the different HVU’S with the use of four

0.25 m2 quadrats at a depth of 200mm. The soil samples were placed in well labeled brown

paper bags. Four soil samples from each HVU were oven dried to constant mass at 600C for

48 hours, and then pulverized to pass through a 2mm sieve, they were then blended, weighed

and analyzed for macro-and-micro nutrients and soil pH. Samples were then sent to

Elsenburg Department of Agriculture analytical laboratory where they were analyzed for

Organic carbon, P, N, K, Mg, Ca and Na, Zn, Cu, and Mn contents. Organic carbon was

analyzed using the Waltley-Black method (Nelson and Somers, 1982), while nitrogen was

analyzed in a solution prepared with concentrated sulphuric acid, selenium powder and

hydrogen peroxide as described by (Okalebo et al., 2002). Other nutrients including Mg, Na

and Ca were analyzed using the ammonium acetate method. The soil pH was determined with

an electrode pH-meter in a soil: water slurry. The analysis of Zn, Cu and Mn was achieved by

the DTPA (Diethylenetriamenepentaacetic) extraction method (Linday and Norvell, 1978)

and their concentrations were observed under photospectrometer.


A Fischer least test and Analysis of variance were used with the use of SAS (2007) to find

significant differences between the mean concentrations of each nutrient and pH across

homogenous vegetation units and season. The significant differences between means were

recognized at a confidence level of 95% (p<0.05).

4.2 Results

4.2.1 Soil macro nutrient contents

Seasonal variation did not have significant effects on the concentrations of all macro and

macronutrients in the three rangelands (p>0.05). The concentration levels on N (%) were

affected by homogenous vegetation units in all the other sites except in the benchmark in

81

both seasons (summer and winter). There were significant differences (p<0.05) in the levels

of OC (%) in the bottomlands of Ndwayana, where bottomland in winter was lower than

(OC=0.85%) all the other sites (Table 4.1). The sites in the Great Fish RNR were all

significantly different (p<0.05) from each other in the concentration levels of K, with the

bottomland in winter (471.75mg/kg) greater than the other three homogenous vegetation units

(Table 4.1). The concentration levels of P (mg/kg) were significantly different in the

bottomlands of the Great Fish RNR with the following concentrations, bottomlands

(236.00mg/kg) in less than the bottomlands (358.75mg/kg) in winter. The concentrations of

(Ca) were higher in bottomland (winter) of the Great Fish RNR was significantly higher from

all the other sites in the concentration levels of Ca (18.05 Cmol (+)/kg). The toplands

(summer) of Ndwayana was significantly higher (p<0.05) from all the other sites in the

concentration levels of Na (213.00mg/kg).

88

Table 4.1: Soil macro nutrient status in Glenmore, Ndwayana and the Great Fish River Nature Reserve.

Site HVU Season N (%) OC (%) K(mg/kg) P(mg/kg) Mg

(cmol(+)/kg)

Ca(cmol (+)/kg) Na (mg/kg)

Glenmore Toplands Summer 0.180aA

1.42aA

176.00eA

58.25cA

1.87aA

5.64bA

55.50cA

Glenmore Toplands Winter 0.142cA

1.48aA

140.25eA

43.75cA

1.78aA

6.05bA

45.50cA

Glenmore Bottomlands Summer 0.160bA

1.56aA

168.00eA

85.75cA

2.54aA

12.96bA

74.25cA

Glenmore Bottomlands Winter 0.140cA

1.30aA

190.00eA

54.00cA

3.10aA

11.23bA

89.25cA

Ndwayana Bottomlands Summer 0.137cA

1.12aA

92.00eA

67.50cA

3.48aA

6.73bA

213.00aA

Ndwayana Toplands Winter 0.115dA

1.13aA

88.00eA

75.25cA

2.99aA

6.64bA

160.50bA

Ndwayana Bottomlands Summer 0.145cA

1.25aA

97.00eA

83.50cA

3.39aA

8.68bA

189.75bA

Ndwayana Bottomlands Winter 0.107dA

0.85bA

84.50eA

50.50cA

2.64aA

7.30bA

172.50bA

GFRNR Toplands Summer 0.140cA

1.64aA

290.50cA

19.00cA

1.48aA

2.88bA

28.00cA

GFRNR Toplands Winter 0.162bA

1.78aA

248.25dA

20.00cA

1.65aA

3.21bA

61.00cA

GFRNR Bottomlands Summer 0.120dA

1.27aA

333.5bB

236.00bA

3.85aA

13.46bA

34.75cA

GFRNR Bottomlands Winter 0.140cA

1.35aA

471.75aA

358.75aA

4.17aA

18.05aA

48.75cA

89

Benchmark Summer 0.140cA

1.47aA

186.50eA

12.5cA

1.03aA

2.19bA

36.25cA

Benchmark Winter 0.142cA

1.14aA

120.00eA

7.75cA

1.18aA

2.00bA

64.75cA

±S.E 0.01 0.17 40.18 39.01 0.63 2.65 22.75

Different superscripts across the columns denote significant differences (p<0.05) of soil macro nutrients in each HVU and season. Capital letters

denote season and small letters HVUs.

89

2.2 Soil micro nutrient contents

Season had no effect (p>0.05) on the soil micro nutrient levels. There were significant

differences (p<0.05) on the concentration levels of Cu (mg/kg) based on the different HVU’s

at Glenmore and the Great Fish RNR (Table 4.2). The Great Fish River Nature Reserve

toplands had these concentration levels of Cu (0.85 and 0.84 mg/kg) and bottomlands Cu

(1.28mg/kg and 1.29 mg/kg) (Table 4.2). On the other hand, the concentration levels of Cu

differed Glenmore toplands (1.39mg/kg and 1.34mg/kg) and Glenmore bottomlands

(2.53mg/kg and 2.19mg/kg).

The concentration levels of Mn (mg/kg) differed between the HVU’s of Glenmore

(bottomlands) and Ndwayana (bottomlands) (Table 4.2). Glenmore bottomlands (Mn=

626.40 and 487.45mg/kg) with bottomlands in summer being higher than bottomlands in

winter. Ndwayana (Mn=341.05 and 416.05mg/kg) with bottomlands in winter greater than

bottomlands in summer (Table 4.2).

The concentrations of Zn differed between some homogenous vegetation units in both

summer and winter (Table 4.2). There was a significant difference (p<0.05) in the

concentration levels of Zn in Glenmore bottomlands with winter greater than summer

(Zn=3.55 and 4.41mg/kg). There were significant differences (p,0.05) in the concentration

levels of Zn between toplands and bottomlands of the Great Fish River RNR (Toplands Zn=

1.48 and 1.17mg/kg) and (bottomlands Zn=3.10 and 3.35mg/kg).

There were significant differences (p<0.05) in soil pH of the Great Fish RNR (Table 4.2)

whereby the pH was higher in winter than in summer.

90

Table 4.2: Soil micro nutrient status of Glenmore, Ndwayana and the Great Fish River Nature

Reserve.

Season HVU Season Cu (mg/kg) Mn (mg/kg) Zn (mg/kg) pH (KCL)

Glenmore Toplands Summer 1.39bA

352.57cA

2.56bA

5.62Ba

Glenmore Toplands Winter 1.34bA

406.60cA

3.27bA

5.72bA

Glenmore Bottomlands Summer 2.53aA

626.40aA

3.55bA

6.27bA

Glenmore Bottomlands Winter 2.19bA

487.45bA

4.41aA

6.47bA

Ndwayana Bottomlands Summer 1.98bA

335.17cA

1.49cA

5.72bA

Ndwayana Bottomlands Winter 1.81bA

304.62cA

1.67cA

5.47bA

Ndwayana Bottomlands Summer 1.87bA

341.05cA

1.41cA

6.00bA

Ndwayana Bottomlands Winter 1.73bA

416.05bA

1.20cA

5.72bA

GFRNR Toplands Summer 0.85cA

280.97cA

1.48cA

5.17bA

GFRNR Toplands Winter 0.84cA

260.00cA

1.17cA

5.07bA

GFRNR Bottomlands Summer 1.28bA

253.67cA

3.10bA

6.97aA

GFRNR Bottomlands Winter 1.29bA

343.87cA

3.35bA

6.35bA

Benchmark Summer 1.39bA

157.20dA

0.96cA

4.90bA

Benchmark Winter 1.46bA

110.97dA

0.69cA

5.10bA

±S.E 0.29 80.32 0.63 0.30

Different superscripts across the columns denote significant differences (p<0.05) of soil

micro nutrients in each HVU and season.

91

4.3 Discussion

4.3.1 Soil macronutrients across homogenous vegetation units

The study showed no seasonal variation in the occurrence of N (Table 4.1). These results

were in contrast to the findings of Gwelo, (2012), in a study conducted around Alice, Eastern

Cape, where seasonal variation occurred in the levels of N with high concentration in winter

than in summer. Homogenous vegetation units had an effect on the concentration of N (%) in

all the sites except in the benchmark (Table 4.1). There was no definite trend in the

occurrence of N accept for fluctuations in all the sites. The inconsistencies in trends make it

difficult to distinguish whether soil N is higher in the communal areas or in the Great Fish

RNR. Studies around South Africa have been conducted investigating the effects of land

degradation on soil C and N (Phesheya et al., 2014; Fatunbi and Dube, 2008). These studies

were conducted in different Biomes but under the semi-arid region. In a study conducted by

Phesheya et al. (2014), there was a decrease in soil organic carbon stocks by 89% with the

decrease in aerial basal cover.

92

The nutrients differed between the HVU’s and the concentration levels of organic carbon in

the bottomlands (winter) of Ndwayana was less significant from all the others sites in the

concentration levels of OC (0.85%) (Table 4.1) The results in Ndwayana bottomlands

indicated that it was more degraded than the other areas of the study. This is similar to the

results by Fatunbi and Dube (2008), who reported that degraded soils had low organic carbon

(poorly, moderately and highly degraded respectively). The concentration of OC ranges from

0.85-1.78% and the normal organic values for a rangeland with a low rainfall are 1.9-2.8

(Baker and Gourley, 2011). The differences in carbon concentrations could be a result of a

high proportion of woody debris under shade and different removal rates of aboveground

biomass by grazing in the open communities (Simion et al., 2003). A change in carbon occurs

due to a wide range of management and environmental factors (Schuman et al., 2002). In a

study by Snyman and Du Preez (2005) in one of the grasslands of . a semi-arid climate in

Bloemfontein, South Africa, the results revealed that degradation of the rangeland from good

to poor condition, with species composition and basal cover used to characterize grassland

condition decreased soil organic carbon and N stocks by 22% and 13%, respectively in fine

sandy loamy grassland soils Giving a clear indication that, a number of factors could be

responsible for the decrease of organic carbon in the soil. There is a positive relationship

between soil organic carbon and the capacity of the soil to supply essential nutrients

including nitrogen, phosphorus and potassium (Rezaei and Gilkes, 2005). Li et al. (2007)

indicated that organic carbon plays an important role in improving soil physical, chemical

and biological properties for sustained plant growth. The relationship between OC and

landscape attributes as well as the positive relationship between OC and nutrient elements

indicated the usefulness of organic carbon as a reliable and sensitive indicator for rangeland

health (Rezaei and Gilkes, 2005).

There was no variation in the soil P contents between HVU’s in the communal areas of

Glenmore and Ndwayana, but P levels differed significantly between the HVU’s found in the

Reserve. Moreover, P levels in the HVU’s found in the Reserve differed seasonally between

the bottomlands (Table 4.1). Similar to these results Oztas et al. (2003) found that the

concentration levels of P in his study were high of foot-slopes than back-slopes. On the

contrary, Lesoli (2008) found no significant differences in the occurrence of P between

communities and between top, slope and valley sites. According to Lesoli (2008) further

supported his results to the fact that identical grazing practices at the different vegetation

types did not affect the concentration of P. Nevertheless, there were differences in the levels

93

of P between the toplands and bottomlands found in the Great Fish RNR (Table 4.1).

Furthermore, there were differences between the bottom sites with winter greater than

summer. The results showed that the concentration levels of P was high except for the

benchmark which had the least concentration in both seasons (12.5 and 7.75) respectively. In

support of these results, (Sigua et al., 2011) reported that the spatial variation in the Great

Fish RNR may be as a result of the input and output of the livestock-pasture system that

includes excretion of urine by animals, phosphorus loss into the soils and eventually harvest

by the animals. Sigua et al. (2011) also reported that livestock grazing plays an important role

in soil P dynamics as it affects P cycling because of the return of P through mineral excretion.

The benchmark site had lower P levels which are in contrast to the results of Congdon and

Herbohn (1993), who reported that the available soil P level is low at the disturbed areas.

Calcium is essential to reduce soil acidity and is also a major nutrient for normal plant growth

(Ashraf et al., 2006).In this study, Seasonal variation had no significance in the levels of Ca

in all the HVU’s, which was in agreement to the findings by Ashraf et al.(2006) in Pakistan.

Similarly, Khan et al. (2004) found little or no effect of seasonal variation on Ca levels in the

soil in a study conducted in a semi-arid region in Pakistan. These results are in contrast to

those of Tiffany et al. (2000) in North Florida (USA) where seasonal variations in the levels

of Ca were observed, with higher levels in summer than in winter. HVU’s had no effect on

the concentration levels of Ca in the communal areas of Glenmore and Ndwayana (Table

4.2.1). There were marked differences in the concentration levels of Ca between the HVU’s

found in the Great Fish RNR. As a result, the bottomlands were highly significant (18.05

cmol(+)/kg) from the toplands when compared in both seasons (Table 4.1).

94

The concentration levels of K were significantly different on the toplands and bottomlands of

the Great Fish RNR in both seasons (Table 4.1).These results were in contrast to those

reported by Fardous et al. (2010) which stated that there was no spatial variation in the

concentration levels conducted in a semi-arid rangeland of Pakistan. Seasonal variability of

soil Mg was not found in the study sites of Glenmore, Ndwayana and The Great Fish RNR,

which was also contrary to the findings of Khan et al. (2008) in the semi-arid rangelands of

Pakistan. There were high concentration levels of Na in the sites of Ndwayana when

compared to the other sites. The concentration levels of Na (213.00 mg/kg) were significantly

higher in the toplands of Ndwayana when compared to all the other sites (Table 4.1). These

results are in accordance with those of Fatunbi and Dube, 2008 who found that there was

high Na in the degraded sites in their study. These results are proof of solidification on the

breakdown and erosion of the soil aggregates.

4.3.2 Soil micro nutrients and soil pH (KCL) across homogenous vegetation

units.

In this study, there was no seasonal variation (p>0.05) in the occurrence of Cu (mg/kg) in the

different sites (Table 4.2.). These findings are in agreement with those of Khan et al. (2006)

in Pakistan who reported no marked seasonal variation in the concentration of soil Cu in all

study sites and concluded that they were all above the critical level of 0.3 mg/kg suggested

for normal plant growth (Rhue and Kidder, 1983) in both summer and winter. The different

sites under study had high concentration levels of Cu and during data collection, the toplands

and bottomlands of Ndwayana had less organic matter and more bare/eroded areas. These

results are in contrast to those of Khan et al. (2006) who related soil Cu availability to soil

organic matter and concluded that the levels of Cu in his study above the critical value

indicate the presence of high organic matter. Moreover, Kabata-Pendias and Pendias (1992)

reported that Cu-binding capacity of any soil and Cu solubility are highly dependent on the

amount of organic matter. There were different fluctuations in the Cu levels when looking at

the results (Table 4.2).

Seasonal variation had no effect in the concentration of Mn in all the study sites as illustrated

in Table 4.2 These results are in conformity with study by Fardous et al. (2011) where

season did not have an effect on Mn levels in the semi-arid regions of Pakistan. All the

values in the sites were above the critical value for plant growth 5mg/kg (Rhue and Kidder,

95

1983) and the same results have been reported by Ferdous et al. (2011). The values (table

4.2) showed that there were high concentration levels of Mn in the communal area when

compared to the sites in the Great Fish RNR giving a clear indication that degradation favors

Mn concentration. There was no seasonal variation in the concentration levels of Zn (Table

4.2) which is in contrast with the study by Gwelo, (2012) where seasonal variation had an

effect on the concentration levels of Zn. The critical growth level for Zn is 1mg/kg (Rhue and

Kidder, 1983). Similar results were found by Ahmad et al. (2012) who found that season had

an effect on the concentration levels of Zn in the semi-arid rangelands in Pakistan. All the

levels resulting from this study except for the benchmark site were above the critical level for

plant growth 1mg/kg (Rhue and Kidder, 1983). Homogenous vegetation units had an effect

on the levels of Zn in the toplands and bottomlands of Glenmore and the Great Fish RNR

under study. The bottomlands of the Great Fish RNR had higher levels when compared to the

toplands in both seasons. The bottomlands of Glenmore differed in the concentration levels of

Zn with winter greater than summer. There were no clear trends in the levels of Zn across all

the sites.

The results of this study revealed that there was no seasonal variation (p>0.05) in the

concentration of pH (KCL) in all the homogenous vegetation units (table 4.2.). The pH

(KCL) ranges from 4.90-6.97 showing that all the areas are slightly acidic and these results

are in contrast to the results of Fatunbi and Dube, 2008. Angassa et al. (2012) postulated that

the acidic state of the rangelands could be a consequence of high leaching of bases in favour

of the acidic compounds. Previous studies have also shown that in acidic soils, base cations

such as Ca, K and Mg are weakly bound to the soil (Berthrong et al., 2009), causing weak

interactions with soil organic matter in the soil. There were no significant differences in the

concentration of pH (KCL) in the communal areas Glenmore and Ndwayana which was the

same discovery for Lesoli (2008) where it was found that in his study the pH did not vary

between communities and between the sites. Later explaining the results by postulating that

identical grazing practices at different vegetation types did not have an effect on soil pH

therefore they could be attributed to the similarities in terms of herbivore grazing intensity,

trampling, defecation and urination. Moreover, Killham, 1994 and Zhao et al.(2007) reported

that herbivore grazing, trampling, defecation and urination could affect soil pH. Homogenous

vegetation unit had an effect in the Great Fish RNR where the bottomlands (summer) were

significantly different from the bottomlands (winter). Adding to that, the bottomlands in the

Great Fish RNR were less acidic when compared to the toplands. The pH levels were not

96

affected by HVU’s in the benchmark site during the different seasons. This implies that

identical grazing practices at different vegetation types did not have effect on soil pH.

4.4 Conclusion

The study indicated no clear trends or relationship between soil micro and macro nutrients

across the different homogenous vegetation units. It was evident that seasonal variation had

no effect in the concentration levels of micro and macro nutrients. Land degradation plays a

significant role in changing the soil nutrients status of these rangelands. There was a marked

reduction in most of the soil macro nutrients namely, Nitrogen, carbon and Phosphorus.

There were high concentrations of Potassium in the sites found the Great Fish RNR when

compared to the other sites. The study revealed a positive relationship between the change in

species composition and the concentration levels of this element. Moreover, there was a

positive correlation between the trace elements and land degradation because all the values

were beyond the critical growth levels of rangelands for plant growth. The relationship

between soil acidity and land degradation was also found to be positive (pH ranged from

4.90-6.67). There were no clear trends with regards to the variations in pH across HVUs

97

References

Ahmad K, Khan Z I, Ashraf M and Ejaz A 2012.Appraisal of temporal variation in soil and forage

iron and zinc in a pasture under semi–arid environmental conditions.Pakistan Journal of

Botany 44 (1): 157–160.

Angassa A, Sheleme B, Oba G, Treydte A C, Lisntadter A and Sauerborn J 2012b. Savanna land use

and its effect on soil characteristics in Southern Ethiopi. Journal of Arid Environments. 81:

67 – 76.

Ashraf M Y, Khan A, Ashraf M and Zafar S 2006.Studies on the Transfer of mineral nutrients from

feed, water, soil and plants to buffaloes under arid environments.Journal of Arid

Environments 65: 632–643.

Congdon A R and Herbohn J L 1993. Ecosystem dynamics of disturbed and undisturbed sites in

north Queensland wet tropical rain forest. Floristic composition, climate, and soil

chemistry.Journal of Tropical Ecology 9: 349- 363.

Fardous A, Ahmad K, Gondal S, Khan Z I, Ejaz A and Valeem E E 2011.Assessment of iron, cobalt

and manganese in soil and forage: A case study at a rural livestock farm in Sargodha,

Pakistan. Pakistan Journal of Botany 43 (3): 1463–146.

Fardous A, Gondal S, Shar A Z, Ahmmad K, Khan Z I, Ibrahim M, Ejaz A, Ahmad W, Ullah S and

Valeem E E 2010.Sodium Potassium and Magnesium Dynamics insoil–plant–animal

continuum.Pakistan Journal of Botany 42 (4): 2411–2421.

Gizachew L, Hirpha A, Jalata F and Smit G N 2002. Mineral element status of soils, native pastures

and cattle blood serum in the Mid–altitude of Western Ethiopia. African Journal of Forage

Science 19: 147–155.

Kabata-Pendias A and Pendias H 1992 Trace Elements in Soils and Plants; CRC Press: Boca Raton,

Florida.

98

Kariuki K S, Schroder J L, Zhang H, Hanks T, Mcgrath M J and Payton M E 2010.Temporal

variability of soil property dynamics in a grazed pasture. Communications in Soil Science

and Plant Analysis 41 (22): 2744–2754.

Khan I K, Ashraf M, Ahmad K, McDowell L R and Valeem E E 2008. Transfer of magnesium from

soil and forage to goats grazing in a semi–arid region of Pakistan: Influence of seasons and

sampling periods. Pakistan Journal of Botany 40 (5): 2127–2133.

Khan Z I, Hussain A, Ashraf M Koyro H W, Huchzermeyer B, Yousaf M and Akhtar S 2004.

Seasonal variation in Sodium status of soil, forage, feed, water and small Ruminants grazing

in the Semi–arid Region of Pakistan.Journal of Animal and Veterinary Advances 3: 228–240.

Khn Z I, Ashraf M, Hussain A and McDowell L R 2006.Seasonal Variation of Trace Elements in a

Semiarid Veld Pasture. Communications in Soil Science and Plant Analysis. 37: 1471 – 1483.

Khan Z I, Hussain A, Ashraf M and McDowell L R 2008. Mineral Status of soils and forages in

Southwestern Punjab–Pakistan: Macro–minerals. Asian–Australian Journal of Animal

Science 19(8): 1139–1147.

Killham K 1994. Soil Ecology. Cambridge University Press, Cambridge.

Lesoli M S 2008. Vegetation and soil status and human perceptions on the condition of communal

rangelands of the Eastern Cape, South Africa.MSc. Dissertation, University of Fort Hare,


Li X.-G, Li F M, Zed R, Zhan Z Y, and Singh B 2007.Soil physical properties and their relations to

organic carbon pools as affected by land use in an alpine pastureland.Geoderma,

2007.01.006.

Lindsay W. L and Norvell 1978.Development of a DPTA soil test for zinc, iron, Manganese and

Copper Journal of Soil Science Society of America. 42: 421 – 428.

McDonald P, Edwards R A, Greenhalgh J F D and Morgan C A 1996.Animal Nutrition, 5th ed.

Longman, U.K. pp 109.

99

Gwelo F A 2012. Farmer’s perception of livestock feeding and rangeland management; dynamics of

soil, forage and cattle blood serum mineral levels in two communal areas of the Eastern

Cape. University of Fort Hare.

Mojiri A, Aziz H A, and Ramaji A 2012. Potential decline in soil quality attributes as a result of land

use change in a hill slope in Lordegan, Western Iran. African Journal of Agricultural

Research. 7: 577 – 582.

Nael M, Khademi H, and Hajabbasi M A 2004.Response of soil quality indicators and their spatial

variability to land degradation in central Iran.Applied Soil Ecology. 27: 221 – 232.

Nelson D D, and Somers L E 1982.Total carbon, orgarnic carbon and organic matter. In. Page, A.l.,

(ED), Methods of soil analysis, American Society of Agronomy, Soil Science Society of

America, Madison, Wis.

Okalebo J R, Gathua K W and Woomer P L 2002. Laboratory Methods of Soil and Plant Analysis: A

Working Manual. TSBF Program UNESCO – ROSTA Soil Science Society of East Africa

technical Publication No. 1. Marvel EPZ Ltd; Nairobi, Kenya p. 127.


overgrazed and eroded rangelands.Journal of Arid Environments 55: 93- 100.

Ramirez R G, Haenlin G F W, Carcia–Castillo C G and Nunez–Ganzales M A 2004.Protein, lignin

and mineral contents and in situ dry matter digestibility of native Mexican grasses consumed

by range goats. Small Ruminant Research 52: 261–269.

Rasaei A, Ghobadi M E, Jalali–Honarmand S, Ghobadi M and Saeidi M 2006. Waterlogging and its

effects on nitrogen of soil and plant.Annals of Biological Research 3 (1):119–124.

Rezaei A S and Gilkes R J 2005.The effects of landscape attributes and plant community on soil

chemical properties in rangelands.Geodema 125: 167- 176.

Rhue R D and Kidder G 1983 Analytical Procedures Used by the IFAS Extension Soil Laboratory

and the Interpretation of Results; Soil Science Department, University of Florida:

Gainesville, Florida.

100

SAS 2007. SAS Institute Inc. Cary, NC USA.

Scholes R J and Walker B H 1993. An African savanna: Synthesis of the Nylsvley study. Cambridge

University Press, UK. pp 306.

Schuman G E, Janzen H H and Herrick J E 2002. Soil carbon dynamics and potential carbon

sequestration by rangelands. Environmental pollution 116; 391- 396.

Sigua G C, Coleman S W, Albano J and Williams M 2011. Spatial distribution of soil phosphorus

and herbage mass in beef cattle pastures: effects of slope aspect and slope position. Nutrient

Cycling in Agro–ecosystems 89: 59–70.

Simion G, Gignoux J and Le Roux X 2003. Tree layer spatial structure can affect Savanna

production and water budget: Results of a 3-D Model. Ecology 84 (7):1879- 1894.

Snyman H A 1998. Dynamics and sustainable utilization of the rangeland ecosystem in arid and

semi-arid climates of southern Africa. Journal of Arid Environments 39: 645- 666.

Tefera S, Mlambo V, Dlamini B J, Koralagama K D N and Mould F L 2009.Chemical composition

and in vitro ruminal fermentation of selected grasses in the semi–arid savannas of

Swaziland.African Journal of Range and Forage Science 26: 1, 9–17.

Tiffany M.E, McDowell L R, O’Connor G A, Nguyen H, Martin FG, Wilkinson N S and Katzowitz

N A 2001 Effects of residual and reapplied biosolids on forage and soil concentrations over a

grazing season in north Florida,II: Microminerals. Communications in Soil Science and Plant

Analysis, 32:2211–2226.

USDA-NRCS, 2001. Rangeland Soil Quality-Organic Matter.Soil Quality Information Sheet.

Accessed (15-09-2013).

Yimer F, Ledin S, and Abdelkadir A 2006.Soil property variability in relation to topography aspect

and vegetation community in the southern-eastern highlands of Ethiopia. Forest Ecology and

Management 232: 90- 99.

101

Zhao Y, Peth S, Krummelbein J, Horn R, Wang Z, Steffens M, Hoffman C and Peng X 2007. Spatial

variability of soil properties affected by grazing intensity in Inner Mongolia

grassland.Ecological Modelling 205: 241- 254.

102

CHAPTER 5. SOIL SEED BANK COMPOSITION AND

DENSITY IN GLENMORE, NDWAYANA AND THE GREAT

FISH RIVER NATURE RESERVE.

ABSTRACT

A study was conducted in Glenmore, Ndwayana and the Great Fish River Nature Reserve to

document the extent of degradation on the semi-arid rangelands of the Eastern Cape. One of

the objectives of the study was to determine the soil seed bank composition and density in

Glenmore, Ndwayana, Great Fish RNR and the benchmark. An experiment was conducted in

a glass house to determine soil seed bank composition, density and the potential of the soil

seed bank for possible rangeland restoration at Glenmore, Ndwayana and the Great Fish

RNR. A total of 112 soil samples were collected in seven homogenous vegetation units at

Glenmore, Ndwayana and the Great Fish RNR rangelands in 2014. These samples were

collected at a depth of 80 mm with the use of Auger. Sampling points were indicated by

randomly throwing ten 0.25 m-2

quadrats along 100m transects in each homogenous

vegetation unit. A total of 21 herbaceous species were found in the soil seed bank comprising

9 grasses, 9 forbs and 3 sedges. Most of the grass species found in the seed bank were in the

Benchmark site, while the rest of the homogenous vegetation units were dominated by either

forbs or sedges. The study area comprised 67% perennial and 33% annual grass species. In

terms of palatability, there were 29%= unpalatable, 48% low, 14% high and 9% moderately

palatable species .Pseudognaphalium undulataum (14.59%) was the most abundant species

followed by Medicago laciniata (8.44%),Hypertelis sbowkeriana (8.41%) and Sutera

campulata (8.36%) with Tragus sp (0.23%) followed by Panicum stapfianum (0.5%) being

the least abundant species. There was no clear trend in botanical composition of soil seed

banks, but fluctuations between the sites. There were significant differences (p<0.05) in the

soil seed bank density of the Great Fish RNR when compared to the communal areas which

are Glenmore and Ndwayana (toplands and bottomlands) sites. The benchmark and Great

Fish RNR bottomlands were not significantly different (p>0.05) from three homogenous

vegetation units found in the communal rangelands at Glenmore (toplands and bottomlands)

and Ndwayana (bottomlands) but they were significantly different (p<0.05) from Ndwayana

toplands.

103

Similarities between the seed bank and the above ground vegetation were tested using

Sorensen’s similarity index. The similarity indices were as follows; Glenmore toplands

(40%), Glenmore bottomlands (37.5%), Ndwayana toplands (25%), Ndwayana bottomlands

(28.57%), Great Fish RNR toplands and bottomlands were (0%) with the Benchmark

comprising of (80%). The poor relationship between the seed bank composition and above

ground vegetation indicate that reliance on the soil seed bank for restoration of these

rangelands would not be practically viable as it cannot change the state of the rangelands

from Increaser species to Decreaser species these rangelands.

Key words: Soil seed bank composition, plant density, homogenous vegetation units, and

Sorensen’s similarity index

104

5.1.1 Introduction

Soil seed banks play a significant role in the composition of different plant communities and

also in their conservation (Shauhat and Siddiqui, 2004). The structure of the seed bank

depends on the production and composition of the present and previous communities

(Harrington et al., 1984; Fenner, 1985), and on the longevity of the seeds of each species

under local conditions (Bekker et al., 1997; Thompson and Grime, 1997). Studies made in the

vegetation grazing systems are mostly restricted to the aboveground vegetation, and often

ignore the role that soil seed banks could play in the restoration of degraded vegetation

communities after disturbance (Solomon et al., 2006; Amaha-Kassahun et al., 2009; Dreber

et al., 2011). Trampling and removal of vegetation by animals or fire have an important

impact on the number of seeds produced by a plant and released as seed rain (Page and

Beeton, 2000; Snyman. 2005). Lower seed densities are associated with smooth bare soil

surfaces whereas areas with perennial vegetation, depressions or surfaces covered with plant

litter have higher seed densities. The occurrence of seeds in disturbed habitats is determined

by the relationship between the original plant assemblages, the amount of propagule

production, and the capacity to build up seed reserves in the soil (Kinucan and Smeins, 1992;

Chang et al., 2001).

In rangeland management, it is precarious to establish how far an ecosystem can deviate from

a reference state before being at risk to cross a threshold into an alternative stable state from

which it is unable to revert without active intervention (Briske et al., 2008; Dreber and Esler,

2011). If the seed bank changes, the resulting community structure will be different and

therefore seed banks have the potential to represent a threshold. Rangelands have a large,

persistent seed bank, often with a species composition that does not resemble the

aboveground vegetation (Thompson and Grime, 1997; Amaha-Kassahun et al., 2009), and it

is well documented that these seeds can dictate the successional trends that occur following

large-scale disturbances (Bekker et al., 1997; Edwards and Crawley, 1999). Rangeland

degradation as a result of heavy grazing can decrease species richness both in the seed bank

and seedling establishment in the field (Snyman, 2004), while fire can over the short-term

motivates seedling density from the seed bank (Snyman, 2004). Understanding of the

function and dynamics of seed banks has become a great challenge to ecologists working in

plant communities, as this understanding is obligatory to determine the role of the seed bank

in ecosystem functioning and to improve the integrated management of ecosystems (Snyman

105

, 2009; Dreber, 2011). It is significant to know the degree to which species in a system

depend on specific forms of disturbance or whether innumerable types of disturbance have

equivalent effects on the soil seed bank (Page et al., 2006). The numbers of studies dealing

with vegetation in the arid and semi-arid pastoral Africa are restricted to the aboveground

vegetation community, and ignore seed bank stored in the soil apart of the plant diversity the

recordings of which require more time and effort (Solomon et al., 2006).This study aimed at

determining the impacts that degradation has on soil seed bank composition and density of

the rangelands of Glenmore, Ndwayana and the Great Fish RNR. Comparisons between the

seed bank composition and above ground vegetation were done.

5.1.2. Determination of soil seed bank composition and plant density

A total of 112 soil samples were collected in seven homogenous vegetation units of in

Glenmore, Ndwayana and the Great Fish RNR rangelands in September 2014. These samples

were collected at a depth of 80mm with the use of Auger. Sampling points were indicated by

randomly throwing ten 0.25 m-2

quadrats along 100m transects in each homogenous

vegetation unit. The seedling method was employed to determine botanical composition and

density of the seed banks. 3000 g of sterile composite growth medium was placed into each

plant pots, and the soil samples were evenly placed in 112 plastic pots at a depth of 10 cm.

The pots were divided into 16 pots per HVU. For control measures twelve pots were used on

which no soil sample was added to evaluate whether or not the composite is contaminated

(Solomon, 2011). The visible litter, roots, stolons, rhizomes and tubers were carefully

removed in soil samples (Loydi et al., 2012).The soil samples collected were mixed and 300

g of soil was scooped and evenly spread above the composite in 60 pots to make a thin layer

of 3 cm. Then, all pots were labeled and placed in the agro-forestry nursery of the university

of Fort Hare and were automatically irrigated three times a day (at 9:15 o’clock, 12:15

o’clock and 3:15 o’clock respectively) in an overhead computerized sprinkler irrigation

system. The first germination took place on 7th

day after experimental inception. All emerged

seedlings were counted (Scott et al., 2010) and marked with great care using sharp tooth

picks (Jones and Esler, 2004) to avoid double recording of the same seedling. In the case

where germination occurred in excess of the area of a pot, transplanting was done to avoid

overcrowding which could result in retarded horizontal root growth. The specimens were

compressed and submitted to Selmar Schonla nd herbarium in Grahamstown to confirm the

plant identification. The plants were then categorized according to life forms as annual or

106

perennial graminoids, forbs, and sedges neglecting the woody plant seedlings. The plant

density was calculated as the number of plants relative to the area of a pot.


The Analysis of Variance (ANOVA) and a Fischer least test were used to get mean

abundances of herbaceous species and plant density of the soil seed bank through the use of

SAS (2007). The significant differences of means were tested at 95% confidence level

(p<0.05). Similarity between seed bank and aboveground vegetation was calculated using the

Sorensen’s similarity index (Graig Smith, 1983). The Sorensen’s index (β = 2𝑐 ÷ 𝐴 + 𝐵 ×

100%) was used to evaluate similarities and difference between above ground and below

herbaceous composition. Where:

β = the similarity index,

2c the shared species between site

and A and B the number of species from each sample site.

107

5.2 RESULTS

5.2.1 Seed bank composition

The results for the seed bank composition show a total of 21 species regardless of the plant

class. The soil seed bank comprised nine grasses, 9 forbs and three sedges excluding the

woody species that were recorded. The species comprised 67% perennial species and 33%

annual species. In terms of palatability, there were 29% unpalatable, 48% low, 14% high and

9% moderately palatable species. All the nine grasses recorded were perennials, while sedges

comprised of 67% annuals and 33% perennials while forbs had 56% annuals and 44%

perennials. All the sedges were unpalatable with grass species comprising of 45%low

palatability, 33%high palatability and 22% moderate palatability. The forbs had 33% low

palatability and 67% unpalatable species. Pseudognaphalium undulataum(14.59%) was the

most abundant species followed by Medicago laciniata (8.44%), Hypertelis bowkeriana

(8.41%) and Sutera campulata (8.36%) with Tragus species (0.23%) followed by Panicum

stapfianum (0.5%) being the least abundant species.

108

Table 5.1: Overall mean abundances of the soil seed bank composition in the selected semi-

arid rangelands.

Species Life form

Plant

class Palatability %Abundance

Eragrostis obtuse Perennial Grass Moderate 0.94

Digitaria eriantha Perennial Grass High 3.53

Sporobolus africanus Perennial Grass Moderate 4.25

Sporobolus fimbriatus Perennial Grass High 3.45

Aristida congesta Perennial Grass Low 4.49

Sporobolu snitens Perennial Grass Low 2.55

Eragrostis chloromelas Perennial Grass Low 0.82

Panicum stapfianum Perennial Grass High 0.5

Tragus species Perennial Grass Low 0.23

Bulbostylis humilis Annual Sedge Unpalatable 7.92

Oenothera sp. Annual Forb Unpalatable 0.93

Hypertelis bowkeriana Perennial Sedge Unpalatable 8.41

Medicago laciniata Annual Forb Low 8.44

Sonchus oleraceus Annual Forb Low 5.39

Senecio ilicifolia Perennial Forb Unpalatable 7.59

Oxalis pes-caprea Perennial Forb Low 4.91

Pseudognaphalium undulatum Annual Forb Unpalatable 14.59

Anagallis arvensis Annual Forb Unpalatable 3.56

Senecio inaequidens; Perennial Forb Unpalatable 4.53

Sutera campanula Perennial Forb Low 8.36

Matricaria sp. Annual Sedge Unpalatable 4.46

109

5.2.2 The abundances of dominant species in the soil seed bank

The results in Table 5.2 show that out of the 21 herbaceous species found in the soil seed

bank, nine of these were consistently dominant in all the HVU’s. These comprised of

Bulbostylis humilis, Senecio ilicifolia, Medicago lacinata, Hypertelis bowkeriana, Sutera

campanula, Pseudognaphalium undulatum, Sporobolus fimbriatus, Sporobolus africanus and

Digitaria eriantha. There was no clear trend in the abundances of these species across all

sites, but fluctuations in abundances occurred between the sites. There were significant

differences (p<0.05) in the occurrence of S. fimbriatus, S. africanus and D. eriantha found in

the benchmark when compared to the rest of the other sites. The occurrence of H. bowkeriana

was highly significant (p<0.05) in Ndwayana toplands when compared to the benchmark.

The observation of B.humilis was significantly lower (p<0.05) in the benchmark when

compared to the other sites. Moreover, the occurrence of B.humilis in the Great Fish RNR

(bottomlands) was also significantly different when compared to Ndwayana (toplands).

P.undulatum showed a significant difference (p<0.05) between Ndwayana (toplands and

bottomlands) sites and Glenmore (bottomlands) when compared to the benchmark. This

species increased in abundance from the Great Fish RNR to both communal areas. S.ilicifolia

proved to be significantly higher (p<0.05) in Glenmore (toplands) when compared to the rest

of the sites but contrary to this, there was no significant difference (p<0.05) from the

benchmark. M.lacinata occurring in Glenmore (toplands) and the Great Fish RNR was

significantly lower (p<0.05) from Ndwayana (bottomlands). The abundance of S. campulata

in the benchmark was significantly lower (p<0.05) from the abundance in the Great Fish

RNR (bottomlands).

114

Table 5.2: Mean (S.E) abundances of the dominant species in the soil seed bank.

Different superscripts in the rows denote significant differences (P<0.05).

Sites HVUs Bulbostylis

humilis

Digitaria

eriantha

Hypertelis

bowkeriana

Medicago

lacinilata

Pseudognaphalium

undulataum

Senecio

ilicifolia

Sporobolus

africanus

Sporobolus

fimbriatus

Sutera

campanula

Glenmore Bottomlands 5.77ab

0.0b 10.81

ab 7.19

abc 19.51

a 8.25

a 0.0

cb 0.0

cb 9.39

ab

Toplands 12.75ab

0.0b 6.75

ab 14.23

a 18.65

ab 1.61

b 1.70

b 0.0

cb 8.31

ab

Ndwayana Bottomlands 15.17a 0.0

b 8.15

ab 3.82

b 20.05

a 11.38

a 0.0

cb 1.16

b 4.20

ab

Toplands 3.45b 0.0

b 16.28

a 11.75

ab 19.75

a 10.98

a 0.0

cb 0.0

cb 9.25

ab

Great Fish

RNR

Bottomlands 16.70a 0.0

b 8.89

ab 9.28

a 11.86

ab 6.62

a 0.0

cb 0.0

cb 14.88

a

Toplands 7.68ab

0.0b 7.25

ab 10.38

ab 15.21

ab 12.65

a 0.0

cb 0.0

cb 6.53

ab

Benchmark 0.0cb

24.74a 0.0

b 0.0

cb 0.96

b -0.0

cb 28.01

a 24.15a 2.63

b

S.E 3.81 1.23 4.07 3.03 6.21 2.82 1.33 0.91 3.82

115

5.2.3 Soil Seed bank density (plants/m2)

There were significant differences (p<0.05) in the seed bank density of the Great Fish RNR

but there was no significant differences (p>0.05) in seed bank density of the Glenmore and

Ndwayana communal between HVU’s (Figure 5.1). Two sites in the Great Fish RNR namely

the benchmark and Great Fish RNR (bottomlands) were not significantly different (p>0.05)

from three homogenous vegetation units found in the communal rangelands at Glenmore

(toplands and bottomlands) and Ndwayana (bottomlands). On the contrary, they were

significantly different (p<0.05) from Ndwayana (toplands). There was no clear trend on the

seed bank density on the homogenous vegetation units but when looking at (Figure 5.1) it can

be said that the Great Fish RNR had more seed bank density than the communal areas except

for Ndwayana (toplands).

Figure 5.1: Effect of seed bank density on the homogenous vegetation units of Glenmore,

Ndwayana and the Great Fish RNR.

0

200

400

600

800

1000

1200

1400

See

d b

ank

de

nsi

ty (

pla

nt/

m2

)

Homogenous vegetation units

116

5.2.4 Comparison between soil seed bank composition and standing

herbage composition.

Comparisons between the seed bank and the above ground vegetation are presented in Figure

5.2. The coefficients were as follows; Glenmore toplands (40%), Glenmore bottomlands

37.5%, Ndwayana toplands 25%, Ndwayana bottomlands 28.57%, The Great Fish RNR

toplands and bottomlands were 0% with the benchmark comprising of 80%.These results

proved that, in the communal areas of Glenmore and Ndwayana there were slight similarities

between the seed bank and the above ground vegetation. On the other hand, the sites found in

the Great Fish RNR showed no similarities between the seed bank and the standing herbage

production. The toplands and bottomlands of the Great Fish RNR comprised of different

species when compared to those of the above ground vegetation. Surprisingly, the

Benchmark site which was also found in the Great Fish RNR gave a clear indication that the

seed bank and above ground vegetation were similar by 80%. (Figure 5.2).

Figure 5.2: Comparison between above ground vegetation and the seed bank composition in

Glenmore, Ndwayana and the Great Fish RNR.

0

10

20

30

40

50

60

70

80

90

Sore

nse

n's

ind

ex (

%)

Homogenous Vegetation units

117

118

5.3 Discussion

5.3.1 Soil seed bank composition.

The seed bank composition of the current study was approximately nine grass species, nine

forbs and three sedges (Table 5.1). Noting that, all the nine grass species were found in the

benchmark site inside the Reserve but were not present in the other sites. Glenmore (toplands

and bottomlands), Ndwayana (toplands and bottomlands) and the Great Fish RNR (toplands

and bottomlands) showed similarities in vegetation as they were dominated by forbs and

sedges (table 5.1). Similarly, Mndela (2013) reported that in a study conducted in the

communal areas of the Eastern Cape, South Africa the soil seed bank was dominated by forbs

and sedges. Concluding that reliance on the soil seed bank for restoration of degraded

rangelands is not of significance because of it can depend on a number of factors such the

type of soil, veld type and soil nutrient status. Furthermore Solomon et al. (2006) observed

that, the seed bank composition in the communal ranches was poor indicating that reliance of

the seed bank for the restoration of degraded rangelands would not be effective. Grazing by

large herbivores has been reported as being able to alter the composition and density of the

seed bank (Major and Pyott, 1966; Snyman, 2004) which could be a result to the change in

the seed bank composition of this study. Kinucan and Smeins, (1992) and Tessema et al.

(2012) reported that the substitution of the perennial grasses by the annual forbs in the soil

seed bank is related to high grazing pressures. This was a result of the fact that, when the

grasses are heavily grazed, the forbs were ignored promoting their seed production because

they are undisturbed (Koc et al., 2013).The change in seed bank composition and density

alters the abundance relationships within the plant community and subsequent seed output of

each component (Kinucan and Smeins, 1992). Therefore, herbivory can modify plant

successional processes (Kinucan and Smeins, 1992). Snyman, (2004), it was reported that

rangelands in poor condition were characterized by a significantly higher seed bank and more

seedling establishment than the rangelands in good condition. This is in contrast to the results

of this study as the benchmark had more seedling establishment than the rest of these

homogenous vegetation units (Table 5.2).

119

5.3.2 The effect of homogenous vegetation units on the seed bank density

The soil seed bank density showed no significant differences between the sites found in the

Reserve including the benchmark (Figure 5.3). Similarly, there was no significant difference

in the seed bank density of the communal areas when compared with each other (Figure 5.3).

Bakoglu et al. (2009) reported a close relationship between the soil seed bank composition

and density. Therefore, the higher the abundance of forbs in a given community, the higher

will be their contribution to plant seed bank density. This was in contrast to the results of this

study because there were no clear trends in the seed bank density (Figure 5.3) or there is no

clear indication of the report made by Bakoglu et al. (2009) when looking at the seed bank

composition (Table 5.3). There are few (if any) studies that give soil seed bank standards that

can be used as reference to conclude that plant density alone is realistic enough to give clear

indications for reclamation purposes. Physiographic factors of the study area and grazing

largely affect the soil seed bank density (Bakoglu et al., 2009).

The low seed bank density measured in the communal areas and the other two sites in the

Great Fish RNR can be can be attributed to continuous grazing, as the above ground grass

vegetation has decreased because of heavy utilization (Bekker et al., 1997; Snyman, 1998).

Also the destruction of grass roots by trampling livestock (Thompson et al., 1997; Quinfeng

et al. 1999). Furthermore, other research in Southern Africa and Europe has reported the

influences of increased grazing on the seed bank population (O’Connor and Pickett, 1992;

Bertiller, 1996; Bekker et al., 1997). Conclusions by Kinloch and Friedel (2005a) were that,

the impact of grazing on the seed bank and standing herbage depends on the extent of over

utilization and the coincidence with drought in a given time.

5.3.3 Comparison between the above ground vegetation and the seed bank

composition.

Species composition and abundances varied between the seed bank and the above ground

vegetation varied between the sites (figure 5.1). The resulting coefficients were Glenmore

toplands (40%), Glenmore bottomlands (37.5%), Ndwayana toplands (25%), Ndwayana

bottomlands (28.57%), Great Fish RNR toplands and bottomlands were (0%) with the

120

Benchmark comprising of (80%) respectively. These results show that there were little or no

resemblance between the above ground and seed bank vegetation contrary to the benchmark.

Similar results were reported by Solomon et al. (2006) and Lemenih and Teketay, (2006).

The seed bank was mainly prevalent of the annual forbs and sedges which again were

contrary to the benchmark site. This variation was not only based on the type of species

found in these sites, but their abundances also differed greatly. From the results of this study,

it was apparent that the seed bank was persistent (Solomon et al., 2006). Variations in the

values of Sorensen’s similarity index were likely to be affected by the diversity experienced

over the time of sampling (Solomon et al., 2006). Kinloch and Friedel (2005a) reported that,

sampling over a the sample sites longer time frame will increase the chance of detecting an

extended number of species in both the seed bank and above ground vegetation and this may

occur when germination is stimulated by rainfall at different times of the year.

5.4 Conclusions

In conclusion, the seed bank of composition of Glenmore, Ndwayana and the Great Fish

RNR was represented by a poor species composition and low density. The plants species

dominating in the seed bank were forbs and sedges except in the seed bank of the benchmark

where the seed bank was dominated by the grass species. Despite the fact that the

bottomlands and toplands of the Great Fish RNR had a seed bank composition composed of

the forbs and sedges as in the communal rangelands, they had high seed bank density

followed by the benchmark site. In terms of species composition, biomass production and

basal cover Glenmore and Ndwayana were more degraded as compared to the Great Fish

RNR. In conclusion, seed bank composition showed that more seeds persist in the rangelands

that are not severely degraded as compared to the less degraded sites. Seed bank density

varied between the homogenous vegetation units with most of the communal areas having

less seed bank density than the sites found in the Great Fish RNR. The fact that there was a

large number of annual forbs was proof enough that, seed bank composition and density

could not be used as one of the reclamation techniques in degraded rangelands. High

utilization by grazers in these rangelands in the previous years was one of the attributes that

could have promoted the change in the seed bank. The event of heavy grazing was speculated

to be one of the reasons why there was little or no resemblance in the seed bank and standing

herbage production in Glenmore, Ndwayana and Great Fish RNR. Contrary to the results of

121

the other sites, the benchmark showed great results where the seed bank and the above

ground vegetation were very similar.

122

References

Amaha Kassahun H, Snyman A and Smit G N 2009. Soil seed bank evaluation along a degradation

gradient in arid rangelands of the Somali region, eastern Ethiopia. Agricultural, Ecosystems

and Environment 129:428–436.

Bakker J.P and Van Diggelen R 2006. Restoration of dry grasslands and heartlands: Restoration

Perspectives (1stedition). (EDS): Van Andel, J., and Aronson, J. Blackwell Publishing

Company, Victoria, Ausrtalia.

Bakoglu A, Bagci E, Erkovan, H I, Koc A and Kocak A 2009.Seed stocks of grazed and ungrazed

rangelands on Palandoken Mountains of Eastern Anatolia. Journal of Food, Agriculture &

Environment. 7: 674 – 678.

Bertiller M B 1996. Grazing effects on sustainable semi-arid rangelands in Patagonia: the state and

dynamics of the soil seed bank. Environmental Management 20, 123– 132.

Briske D D and Wilson A M 1977. Temperature effects on adventitious root development in blue

grama seedlings. Journal of Range Management 31,174– 178.

Chang E R, Jefferies R L, and Carleton T J 2001. Relationship between vegetation and soil seed

banks in an arctic coastal marsh. Journal of Ecology 89:367–384.

Dreber N and Esler K J 2011. Spatio-temporal variation in soil seed banks under contrasting grazing

regimes following low and high seasonal rainfall in arid Namibia. Journal of Arid

Environments 75:174–184.

Dreber N, Oldeland J and Van Rooyen G M W 2011. Species, functional groups and community

structure in seed banks of the arid Nama Karoo: grazing impacts and implications for

rangeland restoration. Agriculture, Ecosystems and Environment 141:399–409.

Edwards GR and Crawley MJ 1999. Herbivores, seed banks and seedling recruitment in mesic

grassland. Journal of Ecology 87.

123

Fenner M 1985. Seed ecology. New York, NY, USA: Chapman and Hall. 144 p.

Harrington G N, MillsD M D, Pressland A J, and Hodgkinson K C 1984. Semi-arid woodlands, p.

189-206. In: G.N. Harrington, A.D. Wilson, and M.D. Young (eds), Management of

Australia's Rangelands. CSIRO Australia, Melbourne.

Jones F E and Esler K J 2004. Relationship between soil-stored seed banks and degradation in

eastern Nama Karoo rangelands (South Africa).Biodiversity and Conservation 13: 2027 –

2053.

Kinucan R.J and Smeins F E 1992. Soil seed Bank of a Semiarid Texas Grassland under Three Long-

Term (36- Years) Grazing Regimes. American Midland Naturalist. 128: 11 – 21.

Koc A, Gullap M K andErkovan H I 2013. The soil seed pattern in Highland rangelands of Eastern

Anatolian region of Turkey under different grazing systems. Turkish Journal of Field Crops.

18: 109 – 117.

Loydi A, Zalba S M and Distel R A 2012. Viable seed banks under grazing and exclosure conditions

in montane mesic Grasslands of Argentina. Acta Oecologia. 45: 8 – 15.

Majori J, Pyott W T 1966. Buried viable seeds in two California bunchgrass sites and their bearing

on the definition of a flora. Vegetation 13, 253– 282.




O’Connor T G, Pickett G A 1992. The influence of rainfall and grazing on seed production and seed

banks of some African savanna grasslands. Journal of Applied Ecology 29, 247–260.

Occonor T G, Martindale G, Moris C D, Short A, Witkowski E T F and Scott-shaw R 2011.

Influence of grazing management on plant diversity of Highland Sourveld Grassland,

Kwazulu Natal, South Africa. Rangeland Ecology Management. 64: 196 – 207.

Page M J and Beeton R J S 2000. Is the removal of domestic stock sufficient to restore semi-arid

conservation areas? Pacific Conservation Biology 6:245–253.

124

SAS 2007. SAS Institute Inc. Cary, NC USA.

Scott K, Setterfield S, Douglas M and Anderson A 2010. Soil seed banks confer resilience to

Savanna grass-layer plants during seasonal disturbance. Acta Oecologia. 36: 202 – 210.

Shaukat S S and Siddiqui A I 2004. Spatial pattern analysis of seeds of an arable soil seed bank and

its relationship with above-ground vegetation in an arid region. Journal of Arid


Snyman H A 1998. Dynamics and sustainable utilization of the rangeland ecosystem in arid and

semi-arid climates of southern Africa. Journal of Arid Environments 39: 645- 666.

Snyman H A 2004a. Soil seed bank evaluation and seedling establishment along a degradation

gradient in a semi-arid rangeland. African Journal of Range and Forage Science 21, 37–47.

Snyman H A 2009. Disturbance impacts on seed longevity of grass seeds, Semi-Arid South African

Rangeland. Rangeland Ecology & Management. 66: 143 – 156.


systems and distance from water points in a semi-arid rangeland of Southern Ethiopia. South

African Journal of Botany. 72: 263 – 271.

Tessema Z K, de Boer W F, Baars R M T and Prins H H T 2012. Influence of grazing on soil seed

banks determines the restoration potential of above ground vegetation in a semi-arid savanna

of Ethiopia. Biotropica. 44: 211 – 219.

Thompson K and Grime J P 1997. Seasonal variation in seed banks of herbaceous species in ten

contrasting habitats. Journal of Ecology 67:893–921.

125

CHAPTER 6.GENERAL DISCUSSION AND CONCLUSIONS.

6.1 General discussion

The main objective of the study was to do a full rangeland condition assessment to document

the extent of degradation in the selected semi-arid rangelands of the Eastern Cape. The study

sites of choice were two communal areas and a Game Reserve namely Glenmore, Ndwayana

and the Great Fish RNR. The assessment was based on the botanical composition, biomass

production, basal cover, seed bank composition and density and lastly, the macro-and-micro

nutrients in the soil. Findings of the current study on the botanical composition of these

rangelands are as follows:

The rangelands were dominated by Increaser II species in all the homogenous vegetation

units (Table 3.1) and this implies that the rangelands are in poor condition. A long-term

increase of grazing pressure changes a plant community. Under heavy grazing pressures,

palatable plants (Decreasers) disappear and are replaced by less palatable plants (Increasers

or Invaders) (Sisay and Baars, 2002). Under low grazing pressure, the reverse might happen

(Dyksterhuis, 1949). In this study high grazing pressures increased the unpalatable species

over the palatable species which is in agreement with (Sisay and Baars, 2002). All the grass

species were perennials and most of them had a high grazing value (Table 3.1) which was in

contrast to report by Sisay and Baars (2002). They reported that, intense grazing leads to

excessive removal of the most palatable species, which are usually perennial grasses”.

Msinamwa et al. (2004) further explained that intense grazing reduces ground cover, but

ultimately opens the way for less palatable and faster establishing annual grasses and forbs to

take over. The sites found in the communal areas and the Great Fish RNR was dominated by

A.congesta and the Karroid species. In addition to that, Great Fish RNR (toplands) was also

dominated by D.eriantha (Table 3.2). Increaser II species are known to be less desirable and

they increase with overutilizationn of the rangelands (Van Oudtshoorn, 2006). These are the

first signs of degradation as a result of heavy grazing pressures inside and outside the Great

Fish RNR. There was a remarkable shift from the palatable species to the less palatable

species as can be referenced by the Great Fish RNR (toplands) site where there was a

dominance D.eriantha (Table 3.2). The benchmark site was expected to be in a much better

condition than the other sites but surprisingly it was also dominated by Increaser II species

126

namely S.fimriatus, S.africanus, E.plana then one Decreaser species D.eriantha which was

even less dominant. In this study, it was of pivotal importance to note the method used for

site selection during the research of this study. Visual observations were made and the

benchmark site was selected based on the absence of bare patches meaning high ecological

stability, high biomass production and the availability of the Decreaser species such as

Themeda triandra, Heteropogon contortus, Digitaria eriantha e.t.c. Therefore, it was during

the actual study when the results proved to be the dominance of Increaser II species. The

presence of these Decreaser species (but are not dominant in the site) can be underpinned to

the issue of selective grazing and veld type.

Selective grazing involves the foraging of the most palatable species over the least palatable

species by grazing animals. Furthermore, the benchmark site was inside the Great Fish RNR.

The veld type for rangelands under study was first known as the Xeric Thicket but now is

known the Valley Bushveld under the biome Albany Thicket (Mucina and Rutherford, 2006).

In this veld type one can expect to see bush clumps or shrubs (Thickets) (Palmer, 2004),

highlighting that, there was a large number of the woody species over the herbaceous species

in Glenmore and Ndwayana.

Moreover, season had an effect on the biomass production and ecological stability (basal

cover) of these rangelands. There was high biomass production and high ecological stability

in summer than in winter (Table 3.3). There was no clear trend between the sites as to

whether the toplands or bottomlands had more biomass and basal cover or less. The

correlation between biomass and range condition corresponds to the presumption that forage

production is low if the range condition is low (Snyman and Fouché 1993) as was the case in

Glenmore and Ndwayana where bare patches were prevalent. The small amount of vegetation

for poor range condition may largely be attributed to high surface runoff due to soil

compaction and absence of litter, which results in poor water use (Sisay and Baars, 2002).

There is a positive linear relationship for range condition rating and total biomass minus the

unpalatable species, and a negative linear relationship for range condition and total biomass

Tiedeman et al. (1991). There was an increasing trend of biomass production and basal cover

as we moved from the communal areas to the Great Fish RNR (Table 3.3). These results

agree with that of Lamphrey, 1983 which stated that, the communal areas were characterized

by continuous grazing and over stocking which lead to heavy grazing pressures. The results

showed that bush encroachment was not one of the problems in these rangelands. Other

127

factors such as soil erosion whether by natural or human induced activities need to be

considered as the major cause of degradation in these rangelands?

Season had no effect while homogenous vegetation units had an effect on the soil nutrient

status of these rangelands (Table 4.1). There was no clear trend as to whether toplands or

bottomlands had more or less of the macro and micro nutrients. There was no clear trend as to

whether the Great Fish RNR or the communal areas had a better nutrient status. The soil

properties did not give a clear indication of the extent of degradation which is in contrast to

other studies by Fatunbi and Dube, (2008) who found that soil C and N were less in degraded

sites when compared to the non-degraded sites. Moreover, Phesheya et al. (2014) reported

decreases in OC and N as cover decreased. There was a great increase of the micro nutrients

in most of the rangelands meaning that poor condition favored the increase of these nutrients.

Heavy grazing in communal areas exerts negative impacts through repeated consumption of

plant seeds thereby reducing the seed number of grazed plants (Snyman, 2004). The seed

bank composition revealed that reliance on the soil seed bank in these rangelands would not

favor restoration of these rangelands as it was mainly forbs and sedges (Table 5.1). These

results agree that of Mndela (2013) in a study in Peddie communal area, Eastern Cape, South

Africa. The seed bank in the toplands and bottomlands showed no resemblance of the above

ground vegetation which was not the case for the benchmark. The seed bank of the

benchmark was fairly good and reliance on it for restoration purposes would promote good

ecological stability and prevent soil erosion.

6.2 General conclusions

The study tried to document the extent of degradation in the selected semi-arid rangelands of

the Eastern Cape (Glenmore, Ndwayana and the Great Fish RNR respectively). Determining

the impact that land degradation has on different parameters such as species composition

biomass production, basal cover, soil seed bank composition and density and soil nutrients

status of these rangelands in consideration with micro and macro nutrients and pH. The

important question kept to mind was to what extent does degradation impact each of these

parameters and how in the rangelands of Glenmore, Ndwayana and the Great Fish RNR? In

terms of species composition, all the rangelands of Glenmore, Ndwayana and the Reserve

(bottomlands) were dominated by A.congesta and the Karroid species and are not even

128

acceptable to grazing animals. Exceptional to these results the Reserve toplands were

dominated by D.eriantha. These results were enough proof that indeed these rangelands were

degraded and the communal areas had more degradation than the Great Fish RNR in

consideration to whether it is toplands or bottomlands. Moreover, these results indicate signs

of high grazing pressure in these rangelands resulting from previous or current land use

systems. When comparing the other sites with the benchmark site, the benchmark was also

dominated by three Increaser II species, one of which was highly palatable to livestock (S.

fimbriatus) and one Decreaser species (D. eriantha).

The results on species composition on the benchmark were underpinned to the issue of

selective grazing and veld type. The veld type plays a significant role and species such as T.

triandra are not dominant. Season had an effect on biomass production and basal cover and

summer had more than winter in both parameters. There were no clear trends on the extent of

degradation relative to biomass production and basal cover between the homogenous

vegetation units. Using biomass production and basal cover as indicators of degradation, the

communal rangelands proved to be more degraded than the sites in Reserve inclusive of the

benchmark site. There was more degradation in these communal rangelands and Ndwayana

showed to be more degraded than Glenmore. The benchmark site had higher biomass

production and basal cover. There were no signs of bush encroachment as an indicator of

degradation in these rangelands when looking at the tree equivalents and bush density per

homogenous vegetation unit. This indicated that degradation could be a result of other factors

such as soil erosion.

There were fluctuations in concentration levels of the macro and micro nutrients and pH

when looking at the soil nutrients status of these rangelands and season did not have an

effect. The different fluctuation made it difficult to state whether which rangelands were

degraded than which and under which homogenous unit. When looking at the concentration

levels of the nutrients there were marked deficiencies of N, OC and P in these rangelands.

There were high concentration levels of K in the sites found in the Reserve (bottomlands and

toplands respectively). Moreover, the micro nutrients were beyond the critical levels for plant

growth in these rangelands except Zn in the benchmark site. The soils in these rangelands

were slightly acidic as the pH ranged from 4.90-6.97. There were no clear trends as which

site had high pH than which when looking at the different sites but fluctuations. The soil

properties did not give clear signs of the extent of degradation between homogenous units

whether found in the Great Fish RNR or communal rangelands. This can further be explained

129

to the issue of soil type under the type of veld type in the whole area. The results of this study

support conventional wisdom, which states that effects of high stocking rates are generally an

undesirable change in species composition, reduced productivity and increased erosion

(Pluhar et al., 1987). Therefore, the utilization of rangeland according to management

principles must be established, soil erosion must be controlled, and degraded rangelands must

be taken into rehabilitation program.

6.3 Recommendations

The study has revealed severe degradation which needs urgent attention. The development of

rangeland access and utilization policies, capacity building of farmers on livestock-rangeland

management, strengthening farmers’ responsibility on livestock grazing movement and

institutionalization of communal system could assume some positive results. The

responsibility of farmers could be strengthened through introducing kraaling and herding and

later influence the practicality of rotational grazing in communal areas. The employment of

people for herding will promote livestock production when compared to fencing. Considering

the size of the communal areas of South Africa, it will be unsustainably expensive to fence

and maintain the fencing. There were clear signs of vandalism of the previous fencing which

indicates lack of responsibility amongst residents- ‘the tragedy of commons’.

Considering unemployment rates and low livestock production in the communal areas, the

employment of people for herding would be beneficial addressing household income,

promote livestock and rangeland management. Therefore, livestock herders would be trained

on basic rangeland and livestock management practices. Linkages between livestock-

rangeland management will help identify early problems that might occur in livestock

production and rangeland condition. This will influence different grazing patterns between

the rangelands and promote resting of the other areas that are mostly susceptible. Restoration

of the rested sites will occur quickly and reduce rangeland degradation. Further

recommendations in promoting restoration of the degraded land would be to introduce

techniques that improve soil-water collection and retention such as development of micro-

catchment, brush packs and the use of water spreading systems (diversion/conversation

furrows). This is mainly because degraded land may have less vegetation cover which

predisposes the land to accelerated runoff resulting to soil loss in the system.

130

References

Dyksterhuis E J 1949. Condition and management of rangeland based on quantitative ecology.

Journal of Range Management 2: 104–112.

Fatunbi A O and Dube S 2008. Land degradation evaluation in a game reserve in Eastern Cape,

South Africa: Soil properties and vegetation cover.






22 (1): 17- 28.

Mucina L and Rutherford M C 2006. The vegetation of South Africa, Lesotho and Swaziland.

Strelitzia 19. Pretoria, South Africa: South African National Biodiversity Institute. 808 p.

Lamprey H F 1983. Pastoralism yesterday and today: the overgrazing controversy. In: Bourlie` re, F.

(Ed.), Tropical Savannas. Ecosystems of the World, vol. 13. Elsevier, Amsterdam, 730pp.

Palmer T 2004. Comprehensive Environment Audit: Vgetation of Makana.ARC-Range and Forage

Institute, Grahamstown.

Pluhar J J Knight R W and Heitschmidt R K 1987. Infiltration rates and sediment production as

influenced by grazing systems in Texas Rolling Plains. Journal of Range Management40:

240–243.

Sisay A and Baars R M T 2002. Grass composition and rangeland condition of the major slope on

overgrazed and eroded rangelands. Journal of Arid Environments. 55: 93 – 100.

Snyman H A and Fouché HJ 1993. Estimating seasonal herbage production of a semi-arid grassland

based on veld condition, rainfall, and evapotranspiration. African Journal of Range & Forag

eScience10: 21–24.

131

Snyman H A 2004. Soil seed bank evaluation and seedling establishment along a degradation

gradient in a semi-arid rangeland. African Journal of Range and Forage Science 21, 37–47.

Tiedeman J A, Beck R and Vanhorn Ecret R 1991. Dependence of standing crop on range condition

rating in New Mexico. Journal of Range Management 44: 602–605.

Van Oudtshoorn. F P. 2006.A guide to grasses of South Africa. Second edition, Fifth impression.


132

APPENDICES

Appendix A: Herbaceous and woody composition

Dependent Variable: Aristida congesta

Sum of

Source DF Squares Mean Square F Value Pr > F

Model 6 5050.428571 841.738095 28.93 <.0001

Error 49 1425.500000 29.091837

Corrected Total 55 6475.928571

Dependent Variable: Karroo

Sum of


Model 6 34388.17857 5731.36310 40.08 <.0001

Error 49 7006.37500 142.98724


Dependent Variable: Digitaria eriantha

Sum of


Model 6 4450.428571 741.738095 23.52 <.0001

133

Error 49 1545.500000 31.540816


Dependent Variable: Sporobolus fimbriatus

Sum of


Model 6 1662.428571 277.071429 13.24 <.0001

Error 49 1025.500000 20.928571


Dependent Variable: Sporobolus africanus

Sum of


Model 6 7041.67857 1173.61310 15.40 <.0001

Error 49 3734.87500 76.22194


Dependent Variable: Biomass production

Sum of


134

Model 13 32874321.30 2528793.95 34.49 <.0001

Error 42 3079506.25 73321.58


Dependent Variable: Basal cover

Sum of


Model 13 1492.928571 114.840659 130.36 <.0001

Error 42 37.000000 0.880952


Dependent Variable: Eragrostis plana

Sum of


Model 6 987.428571 164.571429 9.08 <.0001

Error 49 888.000000 18.122449


135

Dependent Variable: Lippia javanica

Sum of

Source DF Squares Mean Square F Value Pr> F

Model 6 12579.99465 2096.66578 13.20 <.0001

Error 49 7785.54553 158.88868


Dependent Variable: Ptaeroxylon obliquum

Sum of


Model 6 9676.57232 1612.76205 19.25 <.0001

Error 49 4105.80075 83.79185


Dependent Variable: Phyllanthus verrocosus

Sum of


Model 6 4630.77714 771.79619 6.36 <.0001

Error 49 5945.05305 121.32761


Dependent Variable: Jatrova capensis

136

Sum of


Model 6 2476.071347 412.678558 6.54 <.0001

Error 49 3089.901127 63.059207


Dependent Variable: Grewia robasta

Sum of


Model 6 5104.775549 850.795925 8.60 <.0001

Error 49 4844.954545 98.876623


Dependent Variable: Coddia ruddis

Sum of


Model 6 1570.50590 261.75098 1.12 0.3642

Error 49 11442.65147 233.52350


Dependent Variable: Tree equivalents

137

Sum of


Model 6 3352900.382 558816.730 8.88 <.0001

Error 49 3083248.665 62923.442


Dependent Variable: Tree density

Sum of


Model 6 3280892.86 546815.48 2.79 0.0207

Error 49 9619062.50 196307.40


Appendix B: Soil properties and pH

Dependent Variable: pH

Source DF Type I SS Mean Square F Value Pr> F

HVU 13 27.44928571 2.11148352 5.83 <.0001

138

HVU*Season 0 0.00000000 .

Dependent Variable: Ca


HVU 13 1186.589186 91.276091 3.25 0.0019

HVU*Season 0 0.000000 .

Dependent Variable: Mg


HVU 13 95.62168036 7.35551387 4.68 <.0001

HVU*Season 0 0.00000000 . . .

Dependent Variable: Na


HVU 13 212907.7321 16377.5179 7.91 <.0001

HVU*Season 0 0.0000 . . .

Dependent Variable: K


HVU 13 645095.8750 49622.7596 7.68 <.0001

HVU*Season 0 0.0000 . . .

Dependent Variable: P


HVU 13 489982.0000 37690.9231 6.19 <.0001

139

HVU*Season 0 0.0000 . . .

Dependent Variable: Cu


HVU 13 12.19977321 0.93844409 2.80 0.0058

HVU*Season 0 0.00000000 . . .

Dependent Variable: Zn


HVU 13 70.94403571 5.45723352 3.47 0.0011

HVU*Season 0 0.00000000 .

. .

Dependent Variable: Mn


HVU 13 872593.8344 67122.6026 2.60 0.0096

HVU*Season 0 0.0000

Dependent Variable: C


HVU 13 3.03169286 0.23320714 2.04 0.0404

HVU*Season 0 0.00000000 .

140

Appendix C: Seedbank composition and density

Dependent Variable: Sporobolus fimbriatus

Sum of


Model 6 1971.757736 328.626289 82.35 <.0001

Error 21 83.803150 3.990626


Dependent Variable: Sporobolus africanus

Sum of


Model 6 2660.590393 443.431732 62.23 <.0001

Error 21 149.638275 7.125632


Dependent Variable: Digitaria eriantha

Sum of


Model 6 2098.517486 349.752914 57.26 <.0001

141

Error 21 128.280400 6.108590


Dependent Variable: Bulbostylis humillis

Sum of


Model 6 939.507271 156.584545 2.70 0.0420

Error 21 1217.356100 57.969338


Dependent Variable: Senecio ilicifolia

Sum of


Model 6 583.171486 97.195248 3.06 0.0260

Error 21 666.980325 31.760968


Dependent Variable: Hypertelis bowkeriana

Sum of

142


Model 6 573.081371 95.513562 1.44 0.2462

Error 21 1391.937800 66.282752


Dependent Variable: Medicago laciniata

Sum of


Model 6 569.078486 94.846414 2.57 0.0503

Error 21 775.606125 36.933625


Dependent Variable: Sutera campanula

Sum of


Model 6 385.076950 64.179492 1.10 0.3962

Error 21 1227.848550 58.468979


Dependent Variable: Pseudognaphalium undulatum

143

Sum of


Model 6 1154.622100 192.437017 1.21 0.3385

Error 21 3330.787625 158.608935


Dependent Variable: Density

Sum of


Model 6 1135600.027 189266.671 2.18 0.0864

Error 21 1823911.147 86852.912


Date post:	16-Oct-2021
Category:	Documents
Upload:	others
View:	1 times
Download:	0 times

Range condition assessment to document the extent of ...

Documents