Patterns of Plant Diversity in the Hantam-Tanqua-Roggeveld Subregion of the Succulent Karoo, South Africa

Home , Pteronia, Pteronia glauca, Pteronia pallens

Patterns of plant diversity in the Hantam-Tanqua-Roggeveld Subregion of the Succulent Karoo, South Africa

Helga van der Merwe

Submitted in partial fulfilment of the requirements for the degree Philosophiae Doctor in the Faculty of Natural and Agricultural Science Department of Plant Science University of Pretoria Pretoria

Supervisor: Prof. M.W. van Rooyen

July 2009

I hitched my wagon to a daisy Direction vague and destination hazy But, Could any other star have guided me more exactly to where I most dearly loved to be?

- Cythna Letty

Table of contents

Abstract ………………………………………………………………...... vii

Chapter 1: General introduction …………………………………………… 1 References ……………………………………………………………………… 5

Chapter 2: Study area ……………………………………………….…….. 9 2.1 Location and topography ………………………………………………… 9 2.2 Geology and soils …………………………………………………………. 10 2.3 Climate …………………………………………………………………….... 14 2.4 Vegetation ………………………………………………………………….. 17 2.4.1 Succulent Karoo Biome ………………………………………. 17 2.4.2 Fynbos Biome …………………………………………………... 18 2.4.3 Phytogeographical affinities ………………………………….. 18 2.4.4 Vegetation classification ………………………………..……… 21 References ……………………………………………………………………... 21

Chapter 3: Materials and methods ………………………………..………. 25 3.1 Introduction ……………………………………………………………..…. 25 3.2 Vegetation mapping of the Hantam-Tanqua-Roggeveld subregion ……………………………………………………………………..… 25 3.3 Plant diversity studies …………………………….………………….…. 26 3.3.1 Species-area relationships …………………………………….. 27 3.3.2 Diversity parameters …………………………………………… 28 3.3.3 Life form spectra ………………………………………………… 29 3.4 Life form and species diversity on abandoned croplands in the Roggeveld ………………………………………………………..…….. 29 3.5 Vegetation trends following fire in the Roggeveld ……………..….. 30 References ………………………………………………………………….….. 32

Chapter 4: Vegetation of the Hantam-Tanqua-Roggeveld subregion, South Africa. Part 1. Fynbos Biome related vegetation (VAN DER MERWE, H., VAN ROOYEN, M.W. AND VAN ROOYEN, N. 2008a. Vegetation of the Hantam-Tanqua-Roggeveld subregion, South Africa. Part 1. Fynbos Biome related vegetation. Koedoe 50, 61-71.) …………………….. 35 ABSTRACT ……………………………………………………………….……. 36 STUDY AREA ……………………………………………………………….…. 36 METHODS AND MATERIALS ………………………………………………… 37

iii

RESULTS ………………………………………………………………………. 41 DISCUSSION ………………………………………………………………….. 44 ACKNOWLEDGEMENTS ………………………………………………….... 45 REFERENCES …………………………………………………………..……. 45

Chapter 5: Vegetation of the Hantam-Tanqua-Roggeveld subregion, South Africa. Part 2. Succulent Karoo Biome related vegetation (VAN DER MERWE, H., VAN ROOYEN, M.W. AND VAN ROOYEN, N. 2008b. Vegetation of the Hantam-Tanqua-Roggeveld subregion, South Africa. Part 2. Succulent Karoo Biome related vegetation. Koedoe 50, 160-183.) ………………………………………………………... 47 ABSTRACT ………………………………………………………………..….. 48 STUDY AREA ………………………………………………………………….. 48 METHODS AND MATERIALS ……………………………………………... 49 RESULTS …………………………………………………...………………… 49 DISCUSSION ……………………………………………...………………….. 64 ACKNOWLEDGEMENTS ……………………………………………………. 70 REFERENCES …………………………………………………………….……. 70

Chapter 6: Plant diversity in the Hantam-Tanqua-Roggeveld, Succulent Karoo, South Africa: Species-area relationships ……….... 72 Abstract ………………………………………………………………..………… 72 6.1 Introduction …………………..…………………………………………….. 73 6.2 Study area ………………………………………………………………….. 74 6.3 Materials and methods ……………………………………….………… 76 6.4 Results and discussion ……………………………………………….. 78 6.5 Conclusions ………………………………………………..………………. 90 6.6 Acknowledgements ………………………………………..……………… 90 6.7 References …………………………………..……………………………… 90

Chapter 7: Plant diversity in the Hantam-Tanqua-Roggeveld, Succulent Karoo, South Africa: Diversity parameters ………………… 95 Abstract ……………………………………………………..…………………… 95 7.1 Introduction ………………………………………………………………… 96 7.2 Study area ……………………………………………...…………………… 97 7.3 Materials and methods …………………………………………………… 98 7.4 Results and discussion …………………………….…………………. 100 7.5 Conclusions …………………………………………………...………… 105 7.6 Acknowledgements …………………………………………………….. 107

7.7 References ……………………………………………………….………. 107

Chapter 8: Plant diversity in the Hantam-Tanqua-Roggeveld, Succulent Karoo, South Africa: Life form spectra ………………...… 111 Abstract …………………………………………………………..……………. 111 8.1 Introduction ……………………………………………………………….. 113 8.2 Study area ……………………………………………..………………… 114 8.3 Materials and methods ………………………...………………………. 116 8.4 Results and discussion ………………………….………………….….. 117 8.5 Conclusions …………………………………...………………………… 126 8.6 Acknowledgements ……………………….…………………………….. 127 8.7 References ……………………………..………………………………… 127

Chapter 9: Life form and species diversity on abandoned croplands in the Roggeveld, South Africa ……………………………….. 132 Abstract ……………………………………………………………………….. 132 9.1 Introduction …………………………..………………………………….. 133 9.2 Study area ……………………………...………………………………… 135 9.3 Materials and methods ………………………………………………….. 136 9.4 Results and discussion ………………..………………………………. 137 9.5 Conclusions ……………………………………………………………… 148 9.6 Acknowledgements …………..………………………………………… 150 9.7 References ……………………………………………………………….. 150

Chapter 10: Vegetation trends following fire in the Roggeveld, South Africa ………………………………..………………………….. 155 Abstract ………………………………..……………………………………… 155 10.1 Introduction ………………………..…………………………………… 156 10.2 Study area ……………………………………………………………….. 157 10.3 Materials and methods ………..……………………………………… 159 10.4 Results and discussion …………..………………………………….. 160 10.5 Conclusions …………………….……………………………………… 167 10.6 Acknowledgements …………………………………………………… 168 10.7 References …………………..………………………………………….. 168

Chapter 11: General discussion and synthesis ……………………….. 172 11.1 Introduction …………………………………………………………… 172 11.2 Vegetation mapping of the Hantam-Tanqua-Roggeveld subregion ……………………………………..…………………..…………… 172

11.3 Plant diversity studies ……………………………………………….. 177 11.3.1 Species-area relationships ……………….………………. 177 11.3.2 Diversity parameters ……………………….……………… 177 11.3.3 Life form spectra ……………………………………………… 179 11.4 Life form and species diversity on abandoned croplands in the Roggeveld ………………………………………………..………..……… 180 11.5 Vegetation trends following fire in the Roggeveld ……………… 182 11.6 Succulent Karoo and Fynbos affinities: a synthesis of results ……………………………………………………..... 183 11.6.1 Phytosociology …………………………………………….… 183 11.6.2 Environmental parameters ………………………………….. 183 11.6.3 Diversity parameters …………………………….………….. 184 11.6.4 Life forms ………………………………………..…………… 184 References …………………………………………………………..……….. 191

Chapter 12: Summary …………………………….………………………… 195

Chapter 13: Acknowledgements …………...…………………………….. 197

Chapter 14: References …………………...……………………………….. 198

Appendix 1 …………………………………………………………………...………………….. 212 Appendix 2 ………………………………………………………………………………………. 213 Appendix 3 …………………………………………………………………………….………… 248

Chapter 13

Acknowledgements

I would like to thank the Critical Ecosystem Partnership Fund (CEPF) through the Succulent Karoo Ecosystem Plan/Program (SKEP) initiative for funding the project. The Critical Ecosystem Partnership Fund is a joint initiative of Conservation International, the Global Environmental Facility, the Government of Japan, the MacArthur Foundation and the World Bank. A fundamental goal is to ensure civil society is engaged in biodiversity conservation. The various people who assisted with the fieldwork are gratefully acknowledged. This research was also supported by the National Research Foundation under grant number 61277.

CapeNature, the Department of Tourism, Environment and Conservation (Northern Cape) and SANParks are thanked for the necessary permits and permission to conduct the research. Mr. L. Powrie and Dr. M.C. Rutherford of the South African Biodiversity Institute are gratefully acknowledged for their assistance in obtaining weather data for Elandsvlei in the Tanqua Karoo. Dr M. van der Linde and Dr L. Debusso of the Statistics Department at the University of Pretoria are thanked for their assistance with the statistical analysis.

Also, a very special thanks to my husband (Jac van der Merwe), daughter (Vera van der Merwe); and my parents and other family for their support and patience; my friend (Huibrey Theron) for all the hours of assistance in the field and behind the scenes and my supervisor (Gretel van Rooyen) without whom I would not have tackled or completed this thesis.

197

Patterns of plant diversity in the Hantam-Tanqua-Roggeveld Subregion of the Succulent Karoo, South Africa

Helga van der Merwe

Supervisor: Prof. M.W. van Rooyen

Department of Plant Science PhD

Abstract

The Hantam-Tanqua-Roggeveld subregion is located within the Succulent Karoo and Fynbos Biomes, in the predominately winter rainfall area of the Northern and Western Cape Provinces. A phytosociological analysis identified and mapped eight plant associations and 25 subassociations. Forty Whittaker plots were surveyed to quantify the botanical wealth in the area. Each plant association produced its own species-area curves, with the curves of the Mountain Renosterveld and Winter Rainfall Karoo more similar to one another than to the Tanqua Karoo.

Species richness was highest for Mountain Renosterveld, intermediate for Winter Rainfall Karoo and lowest for Tanqua Karoo vegetation. The Mountain Renosterveld and Winter Rainfall Karoo values for evenness, Shannon and Simpson indices were not significantly different, but these values were significantly higher than for the Tanqua Karoo. An ordination of diversity data confirmed a clear Tanqua Karoo cluster, but the Mountain Renosterveld could only be partially separated from the Winter Rainfall Karoo.

Chamaephyte, cryptophyte and therophyte species dominated the study area. Comparisons of life form spectra among associations showed clear differences at a species and vegetation cover level. The percentage contribution of succulent species was low in Mountain Renosterveld, intermediate in Winter Rainfall Karoo and highest in the Tanqua Karoo. Results confirmed the Tanqua Karoo and Winter Rainfall Karoo inclusion into the Succulent Karoo Biome and the strong karroid affinities of the Mountain Renosterveld.

vii

Abandoned croplands of various ages surveyed in the Roggeveld revealed that species richness increased with age yet no similar increase in evenness, Shannon or Simpson indices was found. An abandoned cropland of approximately 33-years should be as species rich as the natural vegetation but was floristically still very different. Recovery rates of the different life forms varied across the different ages of the abandoned croplands.

A ten-year post-fire study in the Mountain Renosterveld indicated that species richness and Shannon index values usually reached a maximum within three years and then declined. A Principal Co-ordinate Analysis of species compositional data separated the first two years from the following eight years. Succession seemed to follow the ‘initial floristic composition’ model of Egler (1954).

viii

Chapter 1

General introduction

The Succulent Karoo stretches along the western side of South Africa and Namibia and is recognised as one of the global hotspots of diversity (Myers et al. 2000, Critical Ecosystem Partnership Fund 2003), being one of only two hotspots that are entirely arid (Conservation International – website 2006). Despite a general lack of structural diversity, plant species diversity at both local and regional scales is undoubtedly the highest recorded for any arid region in the world (Cowling et al. 1989).

To the south the Succulent Karoo lies adjacent to South Africa’s Cape Floristic Region (CFR), a region that has one of the highest species densities and levels of endemism at both local and regional scales for any temperate or tropical continental region (Cowling et al. 1989, 1992). The Cape Floristic Region has the distinction of being the world’s smallest floristic kingdom (Good 1947) and is also recognised as a global hotspot of diversity (Cowling & Hilton-Taylor 1994). On a biome scale the major part of the CFR is classified as the Fynbos Biome (Low & Rebelo 1996, Mucina et al. 2005, Rutherford et al. 2006).

In 2002 the Succulent Karoo Ecosystem Plan (SKEP) initiative was launched (with the sponsorship of the Critical Ecosystem Partnership Fund (CEPF)) to identify and generate consensus for a 20-year conservation and sustainable land-use strategy for the Succulent Karoo hotspot of biodiversity (Conservation International – website 2006). For management purposes, the SKEP planning domain was subdivided into four subregions, one of which was formed by the Hantam-Tanqua-Roggeveld. As little information was available on the biodiversity of the subregion that could be used for future planning, conservation and development (Cilliers et al. 2002, Critical Ecosystem Partnership Fund 2003) collecting botanical data for the subregion was of paramount importance.

The inclusion of the Roggeveld area into the SKEP planning domain (Succulent Karoo) can be questioned because many authors classify the vegetation as part of the Fynbos Biome and not the Succulent Karoo Biome (Low & Rebelo 1996, Mucina et al. 2005, Rutherford et al. 2006). However, there is also a group of researchers that contend that the Roggeveld is a part of the Succulent Karoo Biome (Hilton-Taylor 1994, Jürgens 1997, Van Wyk & Smith 2001) and yet others that classify it as part of the Nama Karoo Biome (Acocks 1988, Rutherford & Westfall 1994). At the time the SKEP initiative was launched it was included in the SKEP domain primarily because it had not been included in the Cape Action Plan for the Environment (C.A.P.E) which was a similar programme to SKEP but focused on the CFR. However, there is a strong relationship between the Succulent Karoo Biome and the Fynbos Biome. The close relationship between these two biomes was advocated by Jürgens (1997)

who proposed the recognition of the Floristic Kingdom of the Greater Cape Flora including at least two partial areas, i.e. the Cape Floristic Region and the Succulent Karoo Region. Born et al. (2007) recently investigated support for an area named the Greater Cape Floristic Region which includes the whole winter rainfall area (arid and mesic climates).

Amongst the earliest references to the botanical wealth of the Hantam-Tanqua-Roggeveld subregion dates from the early 1900s when Diels (1909) mentions the high levels of endemism on the Hantam Mountain. Both Marloth (1908) and Weimarck (1941) expand on the unique character of the vegetation in the subregion. In his phytogeographical analysis Weimarck (1941) treated the Hantam-Roggeveld as a subcentre of his North-Western Centre stating that the subcentre constituted the last outlier of the Cape element in the inner parts of western South Africa. Hilton-Taylor (1994) identified three centres of endemism within his Western Cape Domain, namely: the Western Mountain Karoo, Roggeveld and Tanqua Karoo which all occur within the study area. The botanical importance of the Hantam-Roggeveld was later emphasised by Van Wyk and Smith (2001) who identified it as one of 13 principal centres of plant endemism in southern Africa.

Identification, classification and description of vegetation units across the landscape comprise the first critical steps to improve understanding, protection and management of natural resources. As the first step in the current botanical study of the Hantam-Tanqua-Roggeveld subregion, a broad scale vegetation survey of the entire subregion of approximately three million hectares was undertaken. This vegetation survey, identifying eight plant associations (Van der Merwe et al. 2008a, 2008b), was used as the basis for further detailed botanical investigations.

Since biodiversity hotspots are biologically the richest, yet among the most threatened places on earth, because large numbers of endemic species are undergoing exceptional loss of habitat (Broennimann et al. 2006), a study to investigate various diversity parameters (i.e. species richness, evenness, Shannon and Simpson index of diversity) was initiated. Diversity has two components: species richness, or the number of plant species in a given area, and species evenness, or how well abundance or biomass is distributed among species within a community (Whittaker 1977, Shmida 1984, Magurran 1988, Stohlgren et al. 1995, Wilsey & Potvin 2000). Numerous indices exist which use either species richness or evenness as well as a combination of these two components (Shmida 1984, Magurran 1988, Clinebell et al. 1995, Stirling & Wilsey 2001). In spite of various criticisms, these indices have sparked renewed interest in handling problems associated with the conservation of natural heritage or the changes in global ecology (Mouillot & Leprêtre 1999).

Species richness is currently the most widely used diversity measure (Magurran 1988, Mouillot & Leprêtre 1999, Wilsey & Potvin 2000, Gotelli & Colwell 2001, Stirling & Wilsey

2001) because it is relatively easy to measure, is comparable across communities, and is well understood by researchers, managers, and the public (Hellman & Fowler 1999). However, species richness per se does not imply any standardisation of sampling area (Whittaker 1977, Whittaker et al. 2001). By adding a spatial scale, species-area curves and species accumulation curves can provide more information on the nature of the differences between vegetation types in different geographical areas than mere measures of species richness (Lomolino 2000, Cam et al. 2002, Ugland et al. 2003, Colwell et al. 2004, Drakare, et al., 2006). In recent years species-area curves have been applied successfully to examine the effects of habitat loss on species diversity (Pimm et al. 1995), the effect of invasions on species diversity (Vitousek et al. 1996), the identification of hotspots (Veech 2000), and to set baseline targets for conservation (Desmet & Cowling 2004).

Life forms (and life history patterns) give preliminary information concerning the habitat and the adaptive suite of plant traits and indicate that certain life forms are restricted to growing in particular habitats (Barbour et al. 1999). The most common, parsimonious and accepted plant life form classification is Raunkiaer’s (1934) (Van Rooyen et al. 1990, Pavón et al. 2000) who suggested that the location of a plant’s renewal buds, as differentiated in various life forms, best expresses its adaptation to the unfavourable season for plant life (Danin & Orshan 1990). Life form studies have shown that the distribution and abundance of different plant life forms have well defined limits along an altitudinal gradient studied and suggest that environmental heterogeneity in semi-arid environments was important for the establishment of the different life forms, thus affecting community physiognomy and structure (Pavón et al. 2000). Additionally, assessing effects of land-use change and climate change by life form or plant functional types (PFTs) should facilitate the identification of future trends in ecosystem structure (Smith et al. 1997, Epstein et al. 2002, Broennimann et al. 2006).

Within the study area, the first European farmers settled along the northern slopes of the Roggeveld Mountains in the 1740s (Van der Merwe 1938). These first farmers cultivated crops on a small scale to be self-sustainable. In later years the farmers ploughed large tracts of land to cultivate crops since the Roggeveld has a higher rainfall than the surrounding Succulent Karoo areas. An increase in production costs have forced farmers to cultivate fewer crops leading to the abandonment of various croplands throughout the Roggeveld. Plant community succession is one of the most important aspects of vegetation ecology (Zhang 2005) since successional plant communities provide a model system for testing a variety of ecological hypotheses regarding the controls on biodiversity that could be applied to the management and restoration of plant communities (Huberty et al. 1998). Additionally, with the current predictions of climate change the study of plant succession and vegetation recovery take on an even stronger urgency (Bazzaz 2000). Climate change will add another layer of complexity to the restoration of old fields and could exacerbate the ecological thresholds to plant community assembly (Cramer et al. 2007).

Although fire is not a key environmental parameter in the Succulent Karoo the renosterveld is fire-prone. Mediterranean-type shrublands belong to the world’s major fire-prone biomes with fire a crucial process whereby gaps are created controlling vegetation dynamics and structure (Capitanio & Carcaillet 2008). Various studies have found that in such fire-prone ecosystems species composition and structure rapidly recover after fire (Hanes 1971, Lloret & Vilá 2003). Classifying species into fire life forms or fire response type has been found to provide a useful framework for describing post-fire chaparral succession because these response groups affect ecosystem processes in some predictable ways and may also reflect the underlying environmental changes after fire (Guo 2001).

Due to the lack of botanical information in the Hantam-Tanqua-Roggeveld subregion the main objectives of the current study were to: a) gather botanical information on a regional scale by identifying, classifying and describing plant associations and subassociations present in the Hantam-Tanqua- Roggeveld subregion; b) gain information on plant diversity in each of the vegetation units by studying species richness and species-area relationships; c) analyse patterns of plant diversity using various diversity parameters such as species richness, evenness, Shannon’s index of diversity and Simpson’s index of diversity; d) compare life forms at a species and vegetation cover level on a broad vegetation scale in an attempt to provide clarity on the Succulent Karoo vs. Fynbos Biome status of the subregion; e) follow the recovery of vegetation on abandoned croplands in the Mountain Renosterveld vegetation of the Roggeveld by evaluating the rate of recovery in terms of species composition and various parameters of species and life form diversity; and f) report on post-fire vegetation trends over a 10-year period in the Mountain Renosterveld of the Roggeveld using changes in species numbers, vegetation cover and life form categories.

The main body of this thesis is presented in the form of papers. Chapters 4 and 5 are two published articles that have been included in their published format. Chapters 6 to 10 are to be submitted for publication in various scientific journals and thus the formatting differs between them. Since these chapters have been prepared as independent articles some degree of repetition is inevitable. Except for the papers, a general introduction, study area, materials and methods, general discussion and synthesis, summary, acknowledgements and combined reference list are included. A CD containing various vegetation maps are included for larger scale printing purposes.

References

ACOCKS, J.P.H. 1988. Veld types of South Africa. 3rd edn. Memoirs of the Botanical Survey of South Africa 57, 1-146. BARBOUR, M.G., BURK, J.H., PITTS, W.D., GILLIAM, F.S. AND SCHWARTZ, M.W. (Eds) 1999. Terrestrial plant ecology (3rd edition). Benjamin/Cummings Publishing Company Inc., California. BAZZAZ, F.A. 2000. Plants in changing environments. Linking physiological, population and community ecology. Cambridge University Press, Cambridge. BORN, J., LINDER, H.P. AND DESMET, P. 2007. The Greater Cape Floristic Region. Journal of Biogeography 34, 147-162. BROENNIMANN, O., THUILLER, W., HUGHES, G., MIDGLEY, G.F., ALKEMADE, J.M.R. AND GUISAN, A. 2006. Do geographic distribution, niche property and life form explain plants’ vulnerability to global change? Global Change Biology 12, 1079-1093. CAM, E., NICHOLS, J.D., HINES, J.E., SAUER, J.R., ALPIZAR-JARA, R. AND FLATHER, C.H. 2002. Disentangling sampling and ecological explanations underlying species- area relationships. Ecology 83, 1118-1130. CAPITANIO, R. AND CARCAILLET, C. 2008. Post-fire Mediterranean vegetation dynamics and diversity: A discussion of succession models. Forest Ecology and Management 255, 431-439. CILLIERS, C., THERON, H., RÖSCH, H. AND LE ROUX, A. 2002. Succulent Karoo Ecosystem Plan, Sub-regional report, Hantam/Tanqua/Roggeveld. Succulent Karoo Ecosystem Plan report. CLINEBELL, H.R.R., PHILLIPS, O.L., GENTRY, A.H., STARK, N. AND ZUURING, H. 1995. Prediction of neotropic tree and liana richness from soil and climatic data. Biodiversity and Conservation 4, 56-90. COLWELL, R.K., MAO, C.X. AND CHANG, J. 2004. Interpolating, extrapolating, and comparing incidence-based species accumulation curves. Ecology 85, 2717-2727. CONSERVATION INTERNATIONAL WEBSITE: http://www.biodiversityhotspots.org (accessed 2006). COWLING, R.M., GIBBS RUSSEL, G.E., HOFFMAN, M.T. AND HILTON-TAYLOR, C. 1989. Patterns of plant species diversity in southern Africa. In: B.J. Huntley (Ed.). Biotic diversity in southern Africa. Concepts and Conservation, pp. 19-50. Oxford University Press, Cape Town. COWLING, R.M. AND HILTON-TAYLOR, C. 1994. Patterns of plant diversity and endemism in southern Africa: an overview. In: B.J. Huntley (Ed.). Botanical diversity in southern Africa. Strelitzia 1, pp. 31-52. National Botanical Institute, Pretoria. COWLING, R.M., HOLMES, P.M. AND REBELO, A.G. 1992. Plant diversity and endemism. In: R.M. Cowling (Ed.). The Ecology of Fynbos: Nutrients, fire and diversity, pp. 62- 112. Oxford University Press, Cape Town.

CRAMER, V.A., HOBBS, R.J. AND STANDISH, R.J. 2007. What’s new about old fields? Land abandonment and ecosystem assembly. Trends in Ecology and Evolution 23, 104- 112. CRITICAL ECOSYSTEM PARTNERSHIP FUND, 2003. Ecosystem Profile: The Succulent Karoo hotspot, Namibia and South Africa. Critical Ecosystem Partnership Fund report. DANIN, A. AND ORSHAN, G. 1990. The distribution of Raunkiaer life forms in Israel in relation to the environment. Journal of Vegetation Science 1, 41-48. DESMET, P. AND COWLING, R.M. 2004. Using the species-area relationship to set baseline targets for conservation. Ecology and Society 9, 11 [online] URL: http://www.ecologyandsociety.org/vol9/iss2/art11 DIELS, L. 1909. Formationen und Florenelemente im nordwestlichen Kapland. Botanische Jahrbücher 44, 91-124. DRAKARE, S., LENNON, J.J. AND HILLEBRAND, H. 2006. The imprint of geographical, evolutionary and ecological context on species-area relationships. Ecology Letters 9, 215-227. EPSTEIN, H.E., GILL, R.A., PARUELO, J.M., LAUENROTH, W.K., JIA, G.J. AND BURKE, I.C. 2002. The relative abundance of three plant functional types in temperate grasslands and shrublands of North and South America: effects of projected climate change. Journal of Biogeography 29, 875-888. GOOD, R. 1947. The geography of flowering plants. Longmans, Green & Co., New York. GOTELLI, N.J. AND COLWELL, R.K. 2001. Quantifying biodiversity: procedures and pitfalls in the measurement and comparison of species richness. Ecology Letter 4, 379-391. GUO, Q. 2001. Early post-fire succession in California chaparral: changes in diversity, density, cover and biomass. Ecological research 16, 471-485. HANES, T.L. 1971. Succession after fire in the chaparral of southern California. Ecological Monographs 41, 27-52. HELLMANN, J.J. AND FOWLER, G.W. 1999. Bias, precision, and accuracy of four measures of species richness. Ecological Applications 9, 824-834. HILTON-TAYLOR, C. 1994. Western Cape Domain (Succulent Karoo). In: S.D. Davis, V.H. Heywood and A.C. Hamilton (Eds). Centres of plant diversity. A guide and strategy for their conservation, pp. 201-203. IUCN Publications Unit, Cambridge. HUBERTY, L.E., GROSS, K.L. AND MILLER, C.J. 1998. Effects of nitrogen addition on successional dynamics and species diversity in Michigan old-fields. Journal of Ecology 86, 794-803. JÜRGENS, N. 1997. Floristic biodiversity and history of African arid regions. Biodiversity and Conservation 6, 495-514. LLORET, F. AND VILÁ, M. 2003. Diversity patterns of plant functional types in relation to fire regime and previous land use in Mediterranean woodlands. Journal of Vegetation Science 14, 387-398.

LOMOLINO, M.V. 2000. Ecology’s most general, yet protean pattern: the species-area relationship. Journal of Biogeography 27, 17-26. LOW, A.B. AND REBELO, A.G. 1996. Vegetation of South Africa, Lesotho and Swaziland. Department of Environmental Affairs and Tourism, Pretoria. MAGURRAN, A.E. 1988. Ecological Diversity and its measurement. Cambridge University Press, Cambridge. MARLOTH, R. 1908. Das Kapland, insonderheit das Reich der Kapflora, das Waldgebiet und die Karoo, pflanzengeografisch dargestellt. Wissenschaftliche Ergebnisse der Deutscher Tiefsee-Expedition ‘Waldivia’, 1898 – 1899. 2, T. 3, Fischer, Jena. MOUILLOT, D. AND LEPRÊTRE, A. 1999. A comparison of species diversity estimators. Research on Population Ecology 41, 203-215. MUCINA, L., RUTHERFORD, M.C. AND POWRIE, L.W. (Eds) 2005. Vegetation map of South Africa, Lesotho and Swaziland, 1 : 1 000 000 scale sheet maps. South African National Biodiversity Institute, Pretoria. MYERS, N., MITTERMEIR, R.A., MITTERMEIR, C.G., DE FONSECA, G.A.B., AND KENT, J. 2000. Biodiversity hotspots for conservation priorities. Nature 403, 853-858. PAVÓN, N.P., HERNÁNDEZ-TREJO, H. AND RICO-GRAY, V. 2000. Distribution of plant life forms along an altitudinal gradient in the semi-arid valley of Zapotitlán, Mexico. Journal of Vegetation Science 11, 39-42. PIMM, S.L., RUSSEL, G.J., GITTLEMAN, J.L. AND BROOKS, T.M. 1995. The future of biodiversity. Science 269, 347-350. RAUNKIAER, C. 1934. The life forms of plants and statistical plant geography. Oxford University Press, Oxford. RUTHERFORD, M.C. AND WESTFALL., R.H. 1994. Biomes of Southern Africa. An objective characterisation. Memoirs of the Botanical Survey of South Africa 63, 1-94. RUTHERFORD, M.C., MUCINA, L. AND POWRIE, L.W. 2006. Biomes and Bioregions of Southern Africa. In: L. Mucina and M.C. Rutherford (Eds). The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19, pp. 30-51. South African National Biodiversity Institute, Pretoria. SHMIDA, A. 1984. Whittaker’s plant diversity sampling method. Israel Journal of Botany 33, 41-46. SMITH, T.M., SHUGART, H.H. AND WOODWARD, F.I. (Eds) 1997. Plant functional types: their relevance to ecosystem properties and global change. Cambridge University Press, Cambridge. STIRLING, G. AND WILSEY, B. 2001. Empirical relationships between species richness, evenness, and proportional diversity. The American Naturalist 158, 286-299. STOHLGREN, T.J., FALKNER, M.B. AND SCHELL, L.D. 1995. A modified-Whittaker nested vegetation sampling method. Vegetatio 117, 113-121. UGLAND, K.I., GRAY, J.S. AND ELLINGSEN, K.E. 2003. The species-accumulation curve and estimation of species richness. Journal of Animal Ecology 72, 888-897.

VAN DER MERWE, H., VAN ROOYEN, M.W. AND VAN ROOYEN, N. 2008a. Vegetation of the Hantam-Tanqua-Roggeveld subregion, South Africa. Part 1. Fynbos Biome related vegetation. Koedoe 50, 61-71. VAN DER MERWE, H., VAN ROOYEN, M.W. AND VAN ROOYEN, N. 2008b. Vegetation of the Hantam-Tanqua-Roggeveld subregion, South Africa. Part 2. Succulent Karoo Biome related vegetation. Koedoe 50, 160-183. VAN DER MERWE, P.J. 1938. Die trekboer in die geskiedenis van die Kaapkolonie (1657- 1842). Nasionale Pers Beperk, Kaapstad. VAN ROOYEN, M.W., THERON, G.K. AND GROBBELAAR, N. 1990. Life forms and dispersal spectra of the flora of Namaqualand, South Africa. Journal of Arid Environments 19, 133-145. VAN WYK, A.E. AND SMITH, G.F. (Eds) 2001. Regions of Floristic Endemism in Southern Africa: A review with emphasis on succulents, pp. 1-199. Umdaus Press, Pretoria. VEECH, J.A. 2000. Choice of species-area function affects identification of hotspots. Conservation Biology 14, 140-147. VITOUSEK, P.M., D’ANTONIO, C.M., LOOPE, L.L. AND WESTBROOKS, R. 1996. Biological invasions as global environmental change. American Scientist 84, 468-478. WEIMARCK, H. 1941. Phytogeographical groups, centres and intervals within the Cape flora. Lunds Universitets Årsskrif Avd. 2. 37, 1-143. WHITTAKER, R.H. 1977. Evolution of species diversity on land communities. Evolutionary Biology 10, 1-67. WHITTAKER, R.J., WILLIS, K.J. AND FIELD, R. 2001. Scale and species richness: towards a general, hierarchical theory of species diversity. Journal of Biogeography 28, 453- 470. WILSEY, B.J. AND POTVIN, C. 2000. Biodiversity and ecosystem functioning: importance of species evenness in an old field. Ecology 81, 887-892. ZHANG, J-T. 2005. Succession analysis of plant communities in abandoned croplands in the eastern Loess plateau of China. Journal of Arid Environments 63, 458-474.

Chapter 2

Study area

2.1 Location and topography

The Hantam-Tanqua-Roggeveld subregion, an area of approximately 3 million hectares, is situated within the predominately winter rainfall region of the Northern and Western Cape Provinces of the Republic of South Africa (Figure 2.1). The southwestern border of the study area is found just east of the Cederberg Mountains from where it stretches northwards along the eastern fringe of the Bokkeveld Mountains to just north of Loeriesfontein. Escarpment forming mountains such as the Roggeveld, Komsberg, Klein Roggeveld and Nuweveld Mountains to just southwest of Fraserburg delineate the eastern boundary. The southern limit includes the Tanqua and Ceres Karoo to where the Swartrug and Bontberg Mountains meet north of Ceres.

Figure 2.1 Subregions in the SKEP planning domain (Critical Ecosystem Partnership Fund

2003).

Topographically the physical geography of the Hantam, Tanqua and Roggeveld areas varies greatly. From the level plains of the Tanqua Karoo (Figure 2.2a – c) at about 290 m above sea level, the landscape rises steeply up the escarpment formed by the Roggeveld,

Komsberg and Nuweveld Mountains (to approximately 1800 m above sea level) onto the inland plateau of South Africa (Figure 2.3a – c). The Hantam area is characterised by a gently undulating to steeply rolling topography (Figure 2.4a – c).

The Hantam is known for its Hantam Mountain which is situated to the north of the town of Calvinia, the Tanqua Karoo is known for its aridity and barren landscapes while the star- gazing town of Sutherland is situated in the Roggeveld. The name Roggeveld was derived from the ‘wilde rog’ or wild rye (Secale strictum subsp. africanum) which used to abound in the area, but is now on the brink of extinction.

2.2 Geology and soils

Geologically, the study area is dominated by the Ecca Group (Rubidge & Hancox 1999, Council for Geoscience 1973, 1983, 1989, 1991, 1997, 2001, 2008, Johnson et al. 2006). In the east, mudstones of the Abrahamskraal Formation in the Beaufort Group are found, while in the west the Dwyka Group (tillite, sandstone, mudstone and shale) crops out. The Ecca Group includes sediments of the Tierberg (shale), Prince Albert (mudrock), Kookfontein (shale, siltstone and sandstone) and Skoorsteenberg Formations (mudstone, siltstone and sandstone). Igneous intrusions of dolerite occur throughout the subregion, being easily recognisable as very hard dark grey to nearly black rocks (Van Wyk & Smith 2001). An array of land types are found in the study area (Agricultural Research Council 1986a, 1986b, 1995, 1999a, 1999b, 2002, 2003), with some of the prominent land types being Ag, Da, Db, Dc, Fb, Fc, Ia and Ib (Du Plessis 1987).

Soils of the Tanqua Karoo are shallow lithosols that often include a desert pavement and deep unconsolidated deposits in the alluvial parts. To the west of Calvinia, the Hantam is characterised by shallow lithosols and duplex soils, but where dolerite occurs the soils are red structured and red vertic clays. The Hantam Mountain and mountains of the great escarpment are shallow stony lithosol soils, and duplex soils are found in the occasional lowlands (Francis et al. 2007).

Figure 2.2 A general impression of the low-lying Tanqua Karoo: a) An unusual scene, the Tanqua Karoo covered in carpets of annual flowers; b) the usually barren landscape of the Tanqua Karoo; and c) Tanqua Karoo low-lying plains covered with annuals in an extraordinarily good rainfall year.

Figure 2.3 Scenes of the high-lying Roggeveld Mountains: a) the Roggeveld escarpment and plains of the Tanqua Karoo; b) Roggeveld vegetation receives a higher rainfall than the surrounding areas; and c) the typical vegetation of the Sutherland area dominated by Rosenia oppositifolia.

Figure 2.4 A photographic impression of the Hantam: a) fallow lands covered in the annuals of the Hantam; b) annuals interspersed with perennial shrubs in the Hantam following good rains; and c) red dolerite derived Hantam soils rich in geophytes and annuals.

2.3 Climate

Rainfall ranges from 50 to 300 mm a year with a mean of 228 mm measured at Calvinia and 266 mm measured at Sutherland (Weather Bureau 1988). A maximum of 472 mm in 1976 for Calvinia and 467 mm for Sutherland in 1976 have been recorded (Weather Bureau 1988). The majority of the rainfall is received in winter, however, a few summer thunderstorms do contribute to the total annual rainfall. The mean annual precipitation for the Tanqua Karoo ranges from <100 mm to 200 mm, while the Hantam receives between 100 mm and 400 mm per year and the Roggeveld between 200 mm and 400 mm annually (Schulze 1997). The coefficient of variation of the annual precipitation for the Tanqua Karoo and Hantam is generally 35% to 40% and a few isolated areas with a coefficient of variation of >40% (Schulze 1997). The Roggeveld’s coefficient of variation for the annual precipitation is also between 35% to 40% however, the higher-lying areas have a coefficient of variation of between 25% and 30% (Schulze 1997). Snowfalls usually occur on the high-lying areas and over a 20-year period a mean of one snow day per year was recorded for Calvinia and over a 24-year period, a mean of six snow days per year were recorded for Sutherland (Weather Bureau 1988).

January and February have a mean daily maximum of 30.8°C, with an extreme maximum of 41.2°C recorded for Calvinia in February 1990. The summer month of January is the warmest month in Sutherland with a mean daily maximum of 27.1°C and an extreme maximum of 35.5°C recorded in January 1980 (Weather Bureau 1988). For Calvinia, the coldest months are June and July, with a mean daily minimum of 4.4°C and mean daily maximum of 17.1°C for June and for July a mean daily minimum of 3.5°C and mean daily maximum of 17.2°C (Weather Bureau 1988). June and July are also the coldest months in Sutherland with a mean daily minimum of -1.2°C and mean daily maximum of 12.7°C for June, while July’s mean daily minimum is -2.4°C and mean daily maximum 12.8°C (Weather Bureau 1988). The lowest temperature recorded for Calvinia was -6.5°C in June 1978 and for Sutherland -13.6°C in July 1970 and August 1978 (Weather Bureau 1988).

Walter diagrams (Figure 2.5) were constructed using data obtained from the Weather Bureau (1998) and electronic data extracted by the South African National Biodiversity Institute from the following two reports: 1) Lynch, S.D. 2004. Development of a raster database of annual, monthly and daily rainfall for southern Africa. WRC Report No. 1156/1/04. School of Bioresources Engineering and Environmental Hydrology, University of KwaZulu-Natal, Pietermaritzburg, South Africa. 78 pp. CD included in report; and 2) Schulze, R.E & Maharaj, M. 2004. Development of a database of gridded daily temperatures for southern Africa. WRC Report No. 1156/2/04. School of Bioresources Engineering and Environmental Hydrology, University of KwaZulu-Natal, Pietermaritzburg, South Africa. 82 pp. CD included in report.

Calviniaa (974 m)b °C 30 d e 40 [29 - 29]c 16.4° 228 mm 41.2i 30.8h

j 15.8 m n 20 n

10 l k

3.5f

-6.5g 0 0 July Aug Sept Oct Nov Dec Jan Feb Mar April May June

Figure 2.5 Walter diagrams for a) Calvinia (Hantam); b) Elandsvlei (Tanqua Karoo); and c) Sutherland (Roggeveld). Superscripts denote the following: a = station, b = height above sea level, c = durations of observations in years (of two figures the first indicates temperature, the second precipitation), d = mean annual temperature in °C, e = mean annual precipitation in mm, f = mean daily minimum of the coldest month, g = lowest temperature recorded, h = mean daily maximum for the warmest month, I = highest temperature recorded, j = mean daily temperature variations, k = curve of mean monthly temperature, l = curve of mean monthly precipitation, m = relative period of drought, n = relative humid season and q = months with mean daily minimum below 0°C. Some values are missing, where no data is available for the stations (After Cox and Moore 1994).

Elandsvleia (274 m)b °C 30 d e 40 [14 - 108]c 20.7° 72 mm 45.9i 36.4h

j 14.9 m 20 k

5.8f l -1.3g 0 0 July Aug Sept Oct Nov Dec Jan Feb Mar April May June

Sutherlanda (1459 m)b °C 30 d e 40 [27 - 26]c 11.8° 266 mm 35.5i 27.1h

j m 16.7 n 20 n

-2.4f

-13.6g k 0 0 July Aug Sept Oct Nov Dec Jan Feb Mar April May June

Figure 2.5 (continued)

An interesting climatic characteristic of the Roggeveld is the low number of heat units (degree days) that are encountered during a year. The concept of the heat unit (or degree day) revolves around the development of a plant or organism being dependent upon the total heat to which it was subjected during its lifetime, or else during a certain developmental stage (Schulze 1997). Heat units are expressed as degree days, where these are an accumulation of mean temperatures above a certain lower threshold value (below which active development is considered not to take place) and below an upper limit (above which growth is considered to remain static or even decline) over a period of time (Schulze 1997). The degree days for October to March above 10°C vary in the Roggeveld from 1400 to 1800 however, for April to September these degree days vary from <200 to 400, some of the lowest values for the entire South Africa (Schulze 1997).

Some plants, which have a dormant season during winter, may have to accumulate chill units with temperatures below a threshold in order to stimulate growth, develop leaves, flowers or set fruit (Steyn et al. 1996, Schulze 1997). The required amount of chilling for completion of the rest period varies between species, cultivars and different locations. Chill units have been derived from models using threshold temperatures. Positive chill units for the Roggeveld for May range from 200 to >250, June 300 to >350, August 300 to >350 and September 150 to >250. The accumulated positive chill units from May to September range from 1250 to >1750, some of the highest values for the entire South Africa (Schulze 1997).

2.4 Vegetation

2.4.1 Succulent Karoo Biome

In the Succulent Karoo Biome, succulents and non-succulent chamaephytes, geophytes and therophytes are unusually common relative to trees and grasses (Milton et al. 1997). The Succulent Karoo has a remarkable dominance and unique diversity of short to medium-lived leaf-succulent shrubs as well as a rich geophyte flora (Jürgens et al. 1999, Esler et al. 1999a). The high diversity of dwarf leaf-succulent shrubs is the biome’s most distinctive character (Mucina et al. 2006) with most of the species contained in two families i.e. Mesembryanthemaceae and Crassulaceae (Milton et al. 1997). While geophytes are as successful in the Succulent Karoo Biome as they are in the Fynbos Biome in terms of abundance and diversity (Snijman & Perry 1987, Esler et al. 1999b, Mucina et al. 2006). A feature of the Succulent Karoo Biome are the spring floral displays of winter-growing annuals (Milton et al. 1997), that are relatively predictable and often extravagant (Cowling et al. 1999, Van Rooyen 1999).

Many of the biologically unique features of the Succulent Karoo Biome have been attributed to its climatic conditions (Cowling et al. 1999, Mucina et al. 2006). Firstly, the highly effective

and relatively predictable seasonal rainfall (Desmet & Cowling 1999) and secondly, the relatively moderate winter and early spring temperatures (Mucina et al. 2006). Globally there are few other places what can claim to be as biologically distinct as the Succulent Karoo Biome with the biome experiencing numerous adaptive radiations and associated endemism for a wide range of faunal and floral groups (Mucina et al. 2006).

2.4.2 Fynbos Biome

Fynbos is the predominant vegetation type in the Fynbos Biome (Cowling et al. 1997) however, the Fynbos Biome strictly comprises three quite different, naturally fragmented vegetation types (fynbos, renosterveld and strandveld) that occur in the winter and summer rainfall areas, are dominated by small-leaved, evergreen shrubs and whose generation is intimately related to fire (Rebelo et al. 2006).

Renosterveld is an evergreen, fire-prone vegetation dominated by small-leaved, asteraceous shrubs (especially Dicerothamnus rhinocerotis, renosterbos) and has an understory of Poaceae and geophytes (Moll et al. 1984, Cowling et al. 1997). Moll et al. (1984) differentiated renosterveld into four, more or less biogeographically defined types, however, these exclude the escarpment types, which show strong karroid affiliations (Rebelo et al. 2006).

Four factors separate the Fynbos Biome from the other biomes of southern Africa. These are: 1) nutrient-poor soils supporting fynbos; 2) hot, dry summers alternating with cool, wet winters; 3) recurrent fires at five to 50 year intervals in fynbos and two to 10 year intervals in renosterveld (in comparison to no regularly occurring annual fires or the absence of fire); and 4) a complex of plant-animal interactions (Rebelo et al. 2006).

2.4.3 Phytogeographical affinities

The phytogeographical affinities of the Hantam-Tanqua-Roggeveld area have been under revision for more than a century and still remain uncertain. Some of the first detailed accounts of the area were provided by Marloth (1908) and Diels (1909).

Marloth (1908) placed the Tanqua Karoo within his ‘Karroo’ phytogeographic guild describing it as a semi-desert steppe with a dominance of dwarf shrubs and succulents, and dry riverbeds lined with trees of Acacia spp. and Searsia spp. Additionally, he divided this ‘Karroo’ into four zones one of which he called the Bokkeveld and Tanqua Karoo, which was synonymous to the historically delineated western Karoo and what we would today consider to be the Tanqua Karoo. The great escarpment was used by Marloth (1908) to separate the ‘Karoo’ guild from the ‘Karroides Hochland’ (Karoo highland) thereby placing the Roggeveld

and Roggeveld Karoo areas into the ‘Karroides Hochland’ phytogeographic guild. Marloth (1908) considered the vegetation of the three subdivisions of the Roggeveld (Klein, Middel and Lower Roggeveld) similar, yet different from the vegetation of the Nuweveld Mountains. The Hantam is not specifically mentioned by Marloth (1908), however, his map of the phytogeographic guilds indicates the Hantam included into the ‘Karroides Hochland’ guild.

Diels (1909) visited the Hantam in 1900 which was a good rainfall year and provided additional information on the area. He concurred with Thunberg that as soon as the Bokkeveld ended the landscape became more arid and that the further one traveled it eventually became completely Karoo. As Marloth (1908), Diels (1909) stated that the Karoo vegetation was sparse and that it was characterised by succulents and Asteraceae (Compositae).

This early traveller, Diels (1909), gave a very specific account of his investigations up the Hantam Mountain broadly stating the following: 1) the perennial vegetation above 1000 m above sea level was not found at lower altitudes; 2) on south-facing slopes, at approximately 1400 m above sea level, many new species were encountered including the Cliffortia arborea mentioned in Marloth’s (1908) descriptions of the Roggeveld vegetation; and 3) the highest parts of the mountain were similar to the Cape flora, however, Diels (1909) was not able to sight any Proteaceae or Ericaceae but found Restionaceae and the Cliffortia arborea. Diels (1909) suggested that the Hantam Mountain was perhaps one of the last outliers of the southwestern flora (Cape flora), however, stated that this would have to be researched. Additionally, the mountains to the north, towards Loeriesfontein, and into Namaqualand would have to be studied since these areas were totally unknown at the time. The flora of the Hantam Mountain showed links with the flora he found on Roepmyniet and the flora described by Marloth (1908) for the Roggeveld Mountains.

Weimarck (1941) proposed a classification of the Cape species into five phytogeographical groups and treated the Hantam-Roggeveld as a subcentre of his North-Western Centre and stated that the subcentre constituted the last outlier of the Cape element in the inner parts of western South Africa. He did this somewhat hesitantly, as he stated that perhaps this region did not deserve to be classified as a subcentre on its own, as the species belonging to the Cape element are comparatively few. But the types represented in the Hantam-Roggeveld are so peculiar and often systematically so singular, that such a view could be admitted (Weimark 1941, Van Wyk & Smith 2001).

In his book, Veld types of South Africa, Acocks (1988) included a map of the vegetation in A.D. 1400? stating that there was no direct historical evidence to work with yet, he compiled a map he considered to indicate the vegetation of South Africa at that time. Acocks (1988) placed the Hantam and slopes of the Roggeveld escarpment within a Karoo veld type

(including Karroid Bushveld), the Tanqua Karoo within the Succulent Karoo and the Roggeveld, including the Hantam Mountain and Nuweveld Mountains, within a Scrubby Mixed Grassveld veld type (Nama Karoo Biome). A second map of the vegetation in A.D. 1950, no longer places the Hantam and Roggeveld vegetation as a type of shrubby grassland but within the Karoo veld types and the Tanqua Karoo within the Succulent Karoo and Desert veld types.

The Fynbos Biome Project (1977-1980) resulted in a report on the description of the major vegetation categories in and adjacent to the Fynbos Biome (Moll et al. 1984). Unfortunately, this document did not include all the areas mapped by Acocks, specifically, the Roggeveld and Kamiesberg mountains covered by Mountain Renosterveld (Moll & Bossi 1984).

Rutherford and Westfall (1994) placed the Hantam and Tanqua Karoo within the Succulent Karoo Biome and the Roggeveld within the Nama Karoo Biome stating that one of the four anomalies found while classifying the biomes was an area of Mountain Renosterveld which was physically outside the Fynbos Biome on the Roggeveld escarpment. These limited areas contained vegetation different to the surrounding biomes and were only marginally similar in vegetation structure to that of the Fynbos Biome but showed some floristic affinity to the Fynbos Biome (Olivier et al. 1983 in Rutherford & Westfall 1994). They contend that the life form combination nevertheless precluded this area from being considered as part of the Fynbos Biome (Rutherford and Westfall 1994).

Low and Rebelo (1996) placed the Hantam and Tanqua Karoo within the Succulent Karoo Biome with its low presence of winter rainfall and extreme summer aridity while, the Roggeveld was included within their renosterveld vegetation group of the Fynbos Biome. Furthermore, Low and Rebelo (1996) confirm that the Cape Floral Kingdom traditionally does not include the fynbos and renosterveld outliers to the north and east.

Jürgens (1997) mapped the entire Hantam-Tanqua-Roggeveld subregion as belonging to the Succulent Karoo Biome and listed various studies indicating that some relationships exist between the Karoo and Cape flora arguing for the recognition of the Floristic Kingdom of the Greater Cape Flora with two subdivisions, the Cape Floristic Region and the Succulent Karoo Region.

Mucina et al. (2005), Rebelo et al. (2006), Rutherford et al. (2006) and Mucina et al. (2006) place the Hantam and Tanqua Karoo within the Succulent Karoo Biome and the Roggeveld within the Fynbos Biome. However, they clearly state that they did not apply the explicit and globally derived definition of a biome and only considered botanical elements (Rutherford et al. 2006).

A recent study by Born et al. (2007) evaluates the floristic support for expanding the delimitation of the CFR to include the whole winter-rainfall area into a Greater Cape Floristic Region. Their main conclusions were that the CFR constitutes a valid floristic region, however, the total endemism is higher for the whole winter-rainfall area and this supports the recognition of the larger (Greater Cape Floristic Region) unit. If floristic regions were to be delimited only on endemism, then the Greater Cape Floristic Region is to be preferred (Born et al. 2007).

2.4.4 Vegetation classification

Three of Acocks’s (1953) vegetation types are found within the study area namely: Mountain Renosterveld, Succulent Karoo and Western Mountain Karoo (Acocks 1988). Acocks’s Mountain Renosterveld is equivalent to the Escarpment Mountain Renosterveld of Low and Rebelo (1996), and his Succulent Karoo and Western Mountain Karoo form part of the Lowland Succulent Karoo and Upland Succulent Karoo of Low and Rebelo (1996) respectively. Mucina et al. (2005) and Mucina and Rutherford (2006) recognised twelve vegetation types in the study area which include: the Nieuwoudtville Shale Renosterveld (FRs 2), Roggeveld Shale Renosterveld (FRs 3), Central Mountain Shale Renosterveld (FRs 5), Nieuwoudtville Roggeveld Dolerite Renosterveld (FRd 1), Hantam Plateau Dolerite Renosterveld (FRd 2), Roggeveld Karoo (SKt 3), Hantam Karoo (SKt 2), Tanqua Escarpment Shrubland (SKv 4), Tanqua Karoo (Skv 5), Tanqua Wash Riviere (AZi 7), Namaqualand Riviere (AZi 1) and Bushmanland Vloere (AZi 5).

References

ACOCKS, J.P.H. 1953. Veld types of South Africa. Memoirs of the Botanical Survey of South Africa 28, 1-192. ACOCKS, J.P.H. 1988. Veld types of South Africa. 3rd edn. Memoirs of the Botanical Survey of South Africa 57, 1-146. AGRICULTURAL RESEARCH COUNCIL 1986a. Land type map 3220 Sutherland. Institute for Soil, Climate and Water, Pretoria. AGRICULTURAL RESEARCH COUNCIL 1986b. Land type map 3018 Loeriesfontein. Institute for Soil, Climate and Water, Pretoria. AGRICULTURAL RESEARCH COUNCIL 1995. Land type map 3118 Calvinia. Institute for Soil, Climate and Water, Pretoria. AGRICULTURAL RESEARCH COUNCIL 1999a. Land type map 3120 Williston. Institute for Soil, Climate and Water, Pretoria. AGRICULTURAL RESEARCH COUNCIL 1999b. Land type map 3218 Clanwilliam. Institute for Soil, Climate and Water, Pretoria.

AGRICULTURAL RESEARCH COUNCIL 2002. Land type map 3319 Worcester. Institute for Soil, Climate and Water, Pretoria. AGRICULTURAL RESEARCH COUNCIL 2003. Land type map 3320 Ladismith. Institute for Soil, Climate and Water, Pretoria. BORN, J., LINDER, H.P. AND DESMET, P. 2007. The Greater Cape Floristic Region. Journal of Biogeography 34, 147-162. COUNCIL FOR GEOSCIENCE 1973. Geological map 3218 Clanwillliam. Council for Geoscience, Pretoria. COUNCIL FOR GEOSCIENCE 1983. Geological map 3220 Sutherland. Council for Geoscience, Pretoria. COUNCIL FOR GEOSCIENCE 1989. Geological map 3120 Williston. Council for Geoscience, Pretoria. COUNCIL FOR GEOSCIENCE 1991. Geological map 3220 Ladismith. Council for Geoscience, Pretoria. COUNCIL FOR GEOSCIENCE 1997. Geological map 3319 Worcester. Council for Geoscience, Pretoria. COUNCIL FOR GEOSCIENCE 2001. Geological map 3118 Calvinia. Council for Geoscience, Pretoria. COUNCIL FOR GEOSCIENCE 2008. Geological data 1: 250 000. CD data provided by the Council for Geoscience, Silverton, Pretoria. COWLING, R.M., ESLER, K.J. AND RUNDEL, P.W. 1999. Namaqualand, South Africa – an overview of a unique winter-rainfall desert ecosystem. Plant Ecology 142, 3-21. COWLING, R.M., RICHARDSON, D.M. AND MUSTARD, P.J. 1997. Fynbos. In: R.M. Cowling, D.M. Richardson and S.M. Pierce (Eds). Vegetation of southern Africa, pp. 99-129. Cambridge University Press, Cambridge. COX, C.B. AND MOORE, P.D. 1994. Biogeography: an ecological and evolutionary approach. 5th Edition. Blackwell Scientific Publishing, Oxford. CRITICAL ECOSYSTEM PARTNERSHIP FUND, 2003. Ecosystem Profile: The Succulent Karoo hotspot, Namibia and South Africa. Critical Ecosystem Partnership Fund report. DESMET, P.G. AND COWLING, R.M. 1999. The climate of the Karoo. A functional approach. In: W.R. Dean and S.J. Milton (Eds). The Karoo: Ecological patterns and processes, pp. 3-16. Cambridge University Press, Cambridge. DIELS, L. 1909. Formationen und Florenelemente im nordwestlichen Kapland. Botanische Jahrbücher 44, 91-124. DU PLESSIS, H.M. 1987. Land Types of the maps 2816 Alexander Bay, 2818 Warmbad, 2916 Springbok, 2918 Pofadder, 3017 Garies, 3018 Loeriesfontein. Memoirs on the Agricultural Natural Resources of South Africa 9, 1-538. ESLER, K.J., RUNDEL, P.W. AND COWLING, R.M. 1999a. The Succulent Karoo in a global context: plant structural and functional comparison with North America winter-rainfall

deserts. In: W.R. Dean and S.J. Milton (Eds). The Karoo: Ecological patterns and processes, pp. 123-144. Cambridge University Press, Cambridge. ESLER, K.J., RUNDEL, P.W. AND VOSTER, P. 1999b. Biogeography of prostrate-leaved geophytes in semi-arid South Africa: hypotheses on functionality. Plant Ecology 142, 105-120. FRANCIS, M.L., FEY, M.V., PRINSLOO, H.P., ELLIS, F., MILLS, A.J. AND MEDINSKI, T.V. 2007. Soils of Namaqualand: Compensations for aridity. Journal of Arid Environments 70, 588-603. JOHNSON, M.R., VAN VUUREN, C.J., VISSER, J.N.J., COLE, D.I., WICKENS, H. DE.V., CHRISTIE, A.D.M., ROBERTS, D.L. AND BRANDL, G. 2006. Sedimentary rocks of the Karoo Supergroup. In: M.R. Johnson, C.R. Anhaeusser and R.J. Thomas (Eds). The geology of South Africa, pp. 461-500. The Geological Society of South Africa, Johannesburg/Council for Geoscience, Pretoria. JÜRGENS, N. 1997. Floristic biodiversity and history of African arid regions. Biodiversity and Conservation 6, 495-514. JÜRGENS, N., GOTZMANN, I.H. AND COWLING, R.M. 1999. Remarkable medium-term dynamics of leaf succulent Mesembryanthemaceae shrubs in the winter-rainfall desert of northwestern Namaqualand, South Africa. Plant Ecology 142, 87-96. LOW, A.B. AND REBELO, A.G. 1996. Vegetation of South Africa, Lesotho and Swaziland. Department of Environmental Affairs and Tourism, Pretoria. MARLOTH, R. 1908. Das Kapland, insonderheit das Reich der Kapflora, das Waldgebiet und die Karoo, pflanzengeografisch dargestellt. Wissenschaftliche Ergebnisse der Deutscher Tiefsee-Expedition ‘Waldivia’, 1898 – 1899. 2, T. 3, Fischer, Jena MILTON, S.J., YEATON, R.I., DEAN, W.R.J. AND VLOK, J.H.J. 1997. Succulent Karoo. In: R.M. Cowling, D.M. Richardson and S.M. Pierce (Eds). Vegetation of southern Africa, pp. 99-129. Cambridge University Press, Cambridge. MOLL, E.J. AND BOSSI, L. 1984. Assessment of the extent of the natural vegetation of the Fynbos Biome of South Africa. South African Journal of Science 80, 355-358. MOLL, E.J., CAMPBELL, B.M., COWLING, R.M., BOSSI, L., JARMAN, M.L. AND BOUCHER, C. 1984. A description of major vegetation categories in and adjacent to the Fynbos Biome. South African National Scientific Programmes Report 83, 1-29. MUCINA, L., RUTHERFORD, M.C. AND POWRIE, L.W. (Eds) 2005. Vegetation map of South Africa, Lesotho and Swaziland, 1 : 1 000 000 scale sheet maps. South African National Biodiversity Institute, Pretoria. MUCINA, L., JÜRGENS, N., LE ROUX, A., RUTHERFORD, M.C., SCHMIEDEL, U., ESLER, K.J.,POWRIE, L.W., DESMET, P.G. AND MILTON, S.J. 2006. Succulent Karoo Biome. In: L. Mucina and M.C. Rutherford (Eds). The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19, pp. 220-299. South African National Biodiversity Institute, Pretoria.

MUCINA, L. AND RUTHERFORD, M.C. (Eds) 2006. The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19. South African National Biodiversity Institute, Pretoria. REBELO, A.G., BOUCHER, C., HELME, N., MUCINA, L. AND RUTHERFORD, M.C. 2006. Fynbos Biome. In: L. Mucina and M.C. Rutherford (Eds). The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19, pp. 52-219. South African National Biodiversity Institute, Pretoria. RUBIDGE, B.S. AND HANCOX, P.J. 1999. The Karoo – a palaeontological wonderland. In: M.J. Viljoen and W.U. Reimold (Eds). An introduction to South Africa’s geological and mining heritage, pp. 83-91. Published by the Geological Society of South Africa and Mintek. RUTHERFORD, M.C., MUCINA, L. AND POWRIE, L.W. 2006. Biomes and bioregions of southern Africa. In: L. Mucina and M.C. Rutherford (Eds). The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19, pp. 30-51. South African National Biodiversity Institute, Pretoria. RUTHERFORD, M.C. AND WESTFALL., R.H. 1994. Biomes of Southern Africa. An objective characterisation. Memoirs of the Botanical Survey of South Africa 63, 1-94. SCHULZE, R.E. 1997. South African Atlas of Agrohydrology and – Climatology. Water Research Commission, Pretoria, Report TT82/96. SNIJMAN, D. AND PERRY, P. 1987. A floristic analysis of the Nieuwoudtville Wildflower Reserve, north-western Cape. South African Journal of Botany 53, 445-454. STEYN, H.M., VAN ROOYEN, N. AND VAN ROOYEN, M.W. 1996. The prediction of phenological stages in four Namaqualand ephemeral species using thermal unit indices. Israel Journal of Plant Sciences 44, 147-160. VAN ROOYEN, M.W. 1999. Functional aspects of short-lived plants. In: W.R. Dean and S.J. Milton (Eds). The Karoo: Ecological patterns and processes, pp. 107-122. Cambridge University Press, Cambridge. VAN WYK, A.E. AND SMITH, G.F. (Eds) 2001. Regions of Floristic Endemism in Southern Africa: A review with emphasis on succulents, pp. 1-199. Umdaus Press, Pretoria. WEATHER BUREAU 1998. Climate of South Africa. Climate statistics up to 1990. WB 42. Government Printer, Pretoria. WEIMARCK, H. 1941. Phytogeographical groups, centres and intervals within the Cape flora. Lunds Universitets Årsskrif Avd. 2. 37, 1-143.

Chapter 3

Materials and Methods

3.1 Introduction

This chapter provides a brief outline of the methods applied in chapter 4 to 10. For more detail on the methods the reader is referred to the relevant chapter.

3.2 Vegetation mapping of the Hantam-Tanqua-Roggeveld subregion

Colour, texture and topography were used as a basis to visually stratify satellite images (Bands: 4,5,3 (R,G,B)) of the Hantam-Tanqua-Roggeveld subregion. A total of 390 sample plots were surveyed from August to October 2004 in the stratified homogeneous units close to any national or provincial road or farm track. Most of the surveys were conducted in a 10 m x 10 m plot but larger plots (20 m x 20 m) were used in more denuded areas (Rubin 1998). Geographic Positioning System (GPS) co-ordinates were taken at each site and each species present in a plot was noted and a cover-abundance value assigned according to the Braun- Blanquet cover-abundance scale (Werger 1974). Environmental characteristics such as altitude, topography, aspect, slope, position on the slope, soil type and colour, an estimation of rock cover, rock size and erosion were noted at each survey plot. Biotic effects for example, trampling, small mammal activity and invasion by alien plants, were also recorded.

Analysis of floristic data was undertaken using the TURBOVEG and MEGATAB computer packages (Hennekens & Schaminée 2001). The TURBOVEG software was used to capture the vegetation data and a Two Way Indicator Species Analysis (TWINSPAN) (Hill 1979) was run in MEGATAB. The resulting TWINSPAN on the entire data set of 390 relevés indicated the presence of two distinct floristic groups and enabled the data set to be split into two. A TWINSPAN was then run separately on each data set. The resulting tables were further refined to obtain clear species assemblages using Braun-Blanquet procedures.

The major vegetation units distinguished were termed associations following the definition by Nelder et al. (2005). Species presence and abundance, vegetation structure and spatial distribution of individuals in the dominant layer were used as a basis for the description of the vegetation units. The subassociations were described in terms of a list of species within the subdominant structural layer, together with its canopy cover.

Botanical survey data as well as supporting data such as satellite images, 1:250 000 topocadastral maps, land type maps (Agricultural Research Council 1986a, 1986b, 1995, 1999a, 1999b, 2002, 2003), geological maps (Council for Geoscience 1973, 1983, 1989,

1991, 1997, 2001) and electronic information supplied by the Council for Geoscience (2008) were used to map the vegetation associations and subassociations. Mosaics of specific vegetation types were mapped where two or more subassociations were present in a mapping unit but where it was not possible to map them separately as a result of a high spatial diversity.

Unidentified species were collected and herbarium specimens sent to the Compton Herbarium, Kirstenbosch, for identification. Specimen collection code (HR) and numbers were kept throughout the process since all species have not yet been positively identified, especially those within the Aizoaceae (Mesembryanthemaceae). Voucher specimens are housed at the H.G.W.J. Schweickerdt Herbarium (PRU), University of Pretoria. Nomenclature follows that of Germishuizen and Meyer (2003) and the South African National Biodiversity Institute’s ‘Plants of South Africa’ electronic database (http://posa.sanbi.org).

3.3 Plant diversity studies

Within each of the eight vegetation associations described in the Hantam-Tanqua-Roggeveld subregion (Van der Merwe et al. 2008a, 2008b), Whittaker plots were surveyed. A total of 40 plots were surveyed using Whittaker’s plant diversity plot technique (Shmida 1984). The ease with which the plot can be set up and sampled relative to other techniques such as the Modified Whittaker Nested Vegetation Sampling Technique of Stohlgren et al. (1995) and facilitation of comparisons with other diversity studies were the main reasons for using the Whittaker plot technique. Also, Wilson and Shmida (1984) concluded that the Whittaker method came close to fulfilling four criteria of ‘good’ performance of beta diversity measures.

The only modification to the methodology described by Shmida (1984) related to the field form and notations used on the field form. Each size quadrat was noted in a separate column on the field form with the vegetation of the two 5 m² quadrats noted in two separate columns and the 10 m x 10 m square separated into two 5 m x 10 m rectangles with the species noted apart from one another in two columns on the field form. The columns thus read: ten 1 m², two 5 m², two 50 m² and one 1000 m². Each species in a quadrat was noted within a column and a percentage cover value was assigned for each species in the 1000 m² quadrat. Thus, each column contained a list of all the species present in that quadrat enabling additional calculations for quadrats of a different size than actually measured. At each survey plot, various environmental data were collected, for example, altitude, aspect, slope, geology, various soil characteristics and biotic effects such a small mammal activity and trampling.

m² plots for the 10 m² plot, total of the ten 1 m² and the two 5 m² plots for a 20 m² plot, mean of the two 50 m² plots for a 50 m² plot, the total of the two 50 m² plots for a 100 m² plot and the total for the 1000 m² plot.

Throughout the chapters in this thesis comparisons are made with respect to the vegetation associations as described in Van der Merwe et al. (2008a, 2008b). For convenience, the three Mountain Renosterveld associations are grouped together and called Mountain Renosterveld vegetation, the Escarpment Karoo, Hantam Karoo and Roggeveld Karoo are collectively referred to as the Winter Rainfall Karoo vegetation and the Tanqua and Loeriesfontein Karoo together with the Central Tanqua Grassy Plains are collectively termed the Tanqua Karoo vegetation.

The STATISTICA computer package (StaSoft, Inc. Version 7 and Version 8, 2300 East 14th Street, Tulsa, OK 74104) were used to determine species-area equations, r-values and p- values (significance). ANOVAs were conducted to compare slopes of species-area curves between the three vegetation groups for each of the three functions as well as to determine the significant difference in diversity parameters between the vegetation associations or between different plot sizes. All ANOVAs were preceded by a test for normality. Life form data was compared at species and cover levels across the broad vegetation groups and plant associations using an analysis of variance was performed using the GLM (General Linear Model) Procedure in SAS (SAS® Version 8.2 running on an IBM z9 mainframe computer under z/VM 5.3.0 at the University of Pretoria). Since assumption that the variances among treatment levels were constant was violated, the data were transformed. A power transformation test indicated that the appropriate transformation would be of the form: log10 (life form + 1). The transformed life form values were then used in the statistical analysis. Statistical analyses of the data to investigate a degree of succulence were conducted using the STATISTICA computer package (StaSoft, Inc. Version 8, 2300 East 14th Street, Tulsa, OK 74104) (ANOVA’s – Kruskal-Wallis test) since the data were not normally distributed.

Differences in the slope and intercept values of the species-area curves were analysed by an Analysis of Covariance (Quinn & Keough 2002) with GraphPad Prism 4.03 for Windows (GraphPad software, San Diego, California, USA, www.graphpad.com). The SYN-TAX computer program (Podani 2001) was used to ordinate the total number of species for all seven plot sizes for the 40 survey plots using Principal Co-ordinate Analysis (PCoA).

3.3.1 Species-area relationships

Type II species-area curves (Scheiner 2003, 2004) for each of the 40 Whittaker plots were constructed using the seven different plot sizes. Three different functions were used to

construct these species-area curves namely: 1) the untransformed linear function between species richness (S) and area (A): S = zA + c where c and z are constants for the slope and y-intercept respectively; and 2) the power function, typically expressed as the log transformation: log S = log c + z log A, and 3) the exponential function, expressed as a semilog function: S = z log A + c (Veech 2000). A fourth function, the logistic function, was not used in the study since the whole community was not sampled. If sampling covers the whole of a community, the logistic is expected to be the best model to describe the species-area relationship (He & Legendre 1996).

The species-area curves produced using all three functions were calculated for each sample plot and mean values derived for each subassociation using the function which produced the best fit to the data. Additionally, species-area curves along a transect of ten survey plots running from west to east through the study area was compiled to illustrate the changes in species-area relationships along such an environmental gradient. The transect begins in the Tanqua Karoo and stretches eastwards across the Roggeveld escarpment onto the Roggeveld plateau and crosses five of the eight plant associations.

3.3.2 Diversity parameters

The PC-ORD computer program (PC-ORD Version 4 for Windows, MjM Software design) was used to calculate species richness (S), Shannon’s index of diversity (H’), Simpson’s index of diversity (D) and a measure of evenness (E) for each 1000 m² (0.1 ha) plot sampled. PC- ORD calculates these four diversity measures as follows:

S = richness = number of non-zero elements in a row. H’ = Shannon diversity

H’ = - ∑ pi log pi i

Where pi = importance probability in column i. E = Evenness (equitability) = H’ / ln (richness) D = Simpson’s index of diversity for an infinite population. This is the complement of Simpson’s original index and represents the likelihood that two randomly chosen individuals will be different species.

S 2 D = 1 - ∑ pi i i

Mean values for each of these parameters were calculated for each association as well as for the three vegetation groups, i.e. Mountain Renosterveld, Winter Rainfall Karoo and Tanqua Karoo.

3.3.3 Life form spectra

The species noted in the 1000 m² Whittaker plots were classified into broad life form categories following Raunkiaer’s (1934) classification as modified by Mueller-Dombois and Ellenberg (1974) (Appendix 1). Relative life form contribution at both a species and vegetation cover level, was calculated for each plot using Raunkiaer’s classic forms. Comparisons of each life form were made across the eight plant associations as well as for the three broad vegetation groups (Mountain Renosterveld, Winter Rainfall Karoo and Tanqua Karoo) found in the region. An analysis of variance was performed using the GLM (General Linear Model) Procedure in SAS (SAS® Version 8.2 running on an IBM z9 mainframe computer under z/VM 5.3.0 at the University of Pretoria). The assumption that the variances among treatment levels were constant was violated and thus the data were transformed. A power transformation test indicated that the appropriate transformation would be of the form: log10 (life form + 1). The transformed life form values were then used in the statistical analysis. The complete SAS outputs are included in Appendix 2 for the association level output and Appendix 3 for the vegetation group level output.

A measure of the succulence in the vegetation was determined by calculating the percentage contribution of succulent species to the total species as well as the percentage contribution by succulents in terms of vegetation cover. Statistical analyses of these data were conducted using the STATISTICA computer package (StaSoft, Inc. Version 8, 2300 East 14th Street, Tulsa, OK 74104) (ANOVA’s – Kruskal-Wallis test) since the data were not normally distributed.

3.4 Life form and species diversity on abandoned croplands in the Roggeveld

Whittaker’s plant diversity plot technique (Shmida 1984) was used to sample eight abandoned croplands of various ages (3-, 4-, 8-, 10- 15- and 20-years old) and an undisturbed plot of natural vegetation close to the 20-year old abandoned cropland. All surveys were conducted in one season in the same vegetation type and on the same geological substrate, on one farm in the Roggeveld. The same modification from Shmida’s (1984) methodology was used for the field form and notations on the field form as described in section 3.3 of this chapter were applied.

Raunkiaer’s life form categories (1934) as modified by Mueller-Dombois and Ellenberg (1974) were used to classify the species encountered in the surveys into broad life from categories

(Appendix 1). The relative contribution of each life form, in terms of species as well as vegetation cover, to the 1000 m² (0.1 ha) sample plot were calculated.

Species number totals for seven plot sizes (1 m², 5 m², 10 m², 20 m², 50 m², 100 m² and 1000 m²) were determined and used to construct Type II species-area curves (Scheiner 2003, 2004) for each of the nine plots sampled using the exponential function, since this function produced the best results across the study area (Chapter 6). The exponential function is expressed as a semilog function: S = z log A + c (Veech 2000).

The PC-ORD computer program (PC-ORD Version 4 for Windows, MjM Software design) was used to calculate species richness (S), Shannon’s index of diversity (H’), Simpsons index of diversity (D) and a measure of evenness (E) for each 1000 m² (0.1 ha) plot sampled as set out in section 3.3.2. The Shannon index was also used to calculate a life form diversity index using frequency of life forms instead of species.

Life form distributions were compared using the Chi-square test in the STATISTICA computer package (StaSoft, Inc. Version 7, 2300 East 14th Street, Tulsa, OK 74104). The SYN-TAX computer program (Podani 2001) was used to ordinate the floristic data for all nine plots surveyed using Principal Co-ordinate Analysis (PCoA). Principal Co-ordinates Analysis is a more general form of Principal Component Analysis (PCA) that can give a marked improvement over PCA by allowing the use of a wide array of distance measures (McCune & Grace 2002).

An Analysis of Covariance (Quinn & Keough 2002) in GraphPad Prism 4.03 for Windows (GraphPad software, San Diego, California, USA, www.graphpad.com) was used to analyse differences between slope values and intercepts of the exponential function curves between the abandoned croplands of different ages.

3.5 Vegetation trends following fire in the Roggeveld

On 26 January 1999 more than 10 000 ha burnt in a lightning induced fire in the Roggeveld. The local farmers requested that the, then, Northern Cape Nature Conservation Service conduct surveys to track vegetation changes following the fire. Since the department was under serious financial strain at the time, a small project was initiated in October 1999. Five monitoring sites were selected and surveyed yearly in order to monitor trends in species composition and vegetation cover.

A point or plotless method was used to acquire ten years of data on vegetation changes that followed the fire. Due to the steep slopes and rock-strewn areas, the descending point

method (Roux 1963, Novellie & Strydom 1987) was deemed the most appropriate method to track post-fire vegetation trends at the five sites.

The post-fire transects were permanently marked with iron poles (‘droppers’) indicating the position of the beginning and end points of a 50 m rope which was marked at 1 m intervals. Four lines, 1 m apart and parallel to one another were surveyed in order to limit the chance of surveying transitional areas. A total of 200 points were surveyed per locality. Monitoring was conducted yearly in the last week of September or the first week of October from 1999 to 2008.

The number of strikes per species were expressed as a percentage of the 200 points surveyed and these totals added to determine the percentage vegetation cover since the number of strikes on a species was calculated as a percentage of the total number of point observations made and were not expressed as a percentage of only the strikes (Du Toit 1998a). The sum of the individual plant species percentages obtained rarely totals one hundred because the number of strikes observed are fewer than the total number of point observations made (Du Toit 1998b).

Additionally, the species were classified among the classic life forms as defined by Raunkiaer (1934) and modified by Mueller-Dombois and Ellenberg (1974) (Appendix 1). The post-fire vegetation was investigated in terms of (a) total vegetation cover, (b) total species richness, (c) Shannon-Wiener index of diversity, (d) vegetation cover per life form, (e) species richness per life form, and (f) changes in species composition over the 10 year period.

Shannon’s index of diversity (H’) was calculated for each sampled plot using the PC-ORD computer program (PC-ORD Version 4 for Windows, MjM Software design) which calculates this diversity measure as follows: H’ = Shannon diversity

H’ = - ∑ pi log pi i

Where pi = importance probability in column i.

The species compositional data for each of the five post-fire monitoring plots were ordinated using Principal Co-ordinate Analysis (PCoA) in the SYN-TAX computer program (Podani 2001) in an attempt to visualise vegetation recovery over time.

References

AGRICULTURAL RESEARCH COUNCIL 1986a. Land type map 3220 Sutherland. Institute for Soil, Climate and Water, Pretoria. AGRICULTURAL RESEARCH COUNCIL 1986b. Land type map 3018 Loeriesfontein. Institute for Soil, Climate and Water, Pretoria. AGRICULTURAL RESEARCH COUNCIL 1995. Land type map 3118 Calvinia. Institute for Soil, Climate and Water, Pretoria. AGRICULTURAL RESEARCH COUNCIL 1999a. Land type map 3120 Williston. Institute for Soil, Climate and Water, Pretoria. AGRICULTURAL RESEARCH COUNCIL 1999b. Land type map 3218 Clanwilliam. Institute for Soil, Climate and Water, Pretoria. AGRICULTURAL RESEARCH COUNCIL 2002. Land type map 3319 Worcester. Institute for Soil, Climate and Water, Pretoria. AGRICULTURAL RESEARCH COUNCIL 2003. Land type map 3320 Ladismith. Institute for Soil, Climate and Water, Pretoria. COUNCIL FOR GEOSCIENCE 1973. Geological map 3218 Clanwillliam. Council for Geoscience, Pretoria. COUNCIL FOR GEOSCIENCE 1983. Geological map 3220 Sutherland. Council for Geoscience, Pretoria. COUNCIL FOR GEOSCIENCE 1989. Geological map 3120 Williston. Council for Geoscience, Pretoria. COUNCIL FOR GEOSCIENCE 1991. Geological map 3220 Ladismith. Council for Geoscience, Pretoria. COUNCIL FOR GEOSCIENCE 1997. Geological map 3319 Worcester. Council for Geoscience, Pretoria. COUNCIL FOR GEOSCIENCE 2001. Geological map 3118 Calvinia. Council for Geoscience, Pretoria. COUNCIL FOR GEOSCIENCE 2008. Geological data 1: 250 000. CD data provided by the Council for Geoscience, Silverton, Pretoria. DU TOIT, P.C.V. 1998a. Research note: Grazing-index method procedures of vegetation surveys. African Journal of Range and Forage Science 14, 107-110. DU TOIT, P.C.V. 1998b. Description of a method for assessing veld condition in the Karoo. African Journal of Range and Forage Science 14, 90-93. GERMISHUIZEN, G. AND MEYER, N.L. (Eds) 2003. Plants of southern Africa: an annotated checklist. Strelitzia 14. National Botanical Institute, Pretoria. HE, F. AND LEGENDRE, P. 1996. On species-area relations. The American Naturalist 4, 719-737.

HENNEKENS, S.M. AND SCHAMINÉE, J.H.J. 2001. TURBOVEG, a comprehensive data base management system for vegetation data. Journal of Vegetation Science 12, 589-591. HILL, M.O. 1979. TWINSPAN – A FORTRAN program for arranging multivariate data in an ordered two-way table by classification of the individuals and attributes. Ecology & Systematics, Cornell University, Ithaca, New York. McCUNE, B. AND GRACE, J.B. 2002. Analysis of Ecological Communities. MjM Stoftware Design, Gleneden Beach, Oregon. MUELLER-DOMBOIS, D. AND ELLENBERG, H. (Eds) 1974. Aims and methods of vegetation ecology. Wiley, New York. NELDER V.J., WILSON, B.A., THOMPSON, E.J. AND DILLEWAARD, H.A. 2005. Methodology for Survey and Mapping of Regional Ecosystems and Vegetation Communities in Queensland. Version 3.1. Updated September 2005. Queensland Herbarium, Environmental Protection Agency, Brisbane. NOVELLIE, P. AND STRYDOM, G. 1987. Monitoring the response of vegetation to use by large herbivores: an assessment of some techniques. South African Journal of Wildlife Research 17, 109-117. PODANI, J. 2001. SYN-TAX 2000 Computer programs for data analysis in ecology and systematics. Scientia publishing, Budapest. QUINN, G.P. AND KEOUGH, M.J. 2002. Experimental design and data analysis for biologists. Cambridge University Press, Cambridge. RAUNKIAER, C. 1934. The life forms of plants and statistical plant geography. Oxford University Press, Oxford. ROUX, P.W. 1963. The descending point method of vegetation survey. A point-sampling method for the measurement of semi-open grasslands and Karoo vegetation in South Africa. South African Journal of Agricultural Science 6, 273-288. RUBIN, F. 1998. The physical environment and major plant communities of the Tankwa Karoo National Park. Koedoe 41, 61-94. SCHEINER, S.M. 2003. Six types of species-area curves. Global Ecology and Biogeography 12, 441-447. SCHEINER, S.M. 2004. A mélange of curves – further dialogue about species-area relationships. Global Ecology and Biogeography 13, 479-484. SHMIDA, A. 1984. Whittaker’s plant diversity sampling method. Israel Journal of Botany 33, 41-46. STOHLGREN, T.J., FALKNER, M.B. AND SCHELL, L.D. 1995. A modified-Whittaker nested vegetation sampling method. Vegetatio 117, 113-121. VAN DER MERWE, H., VAN ROOYEN, M.W. AND VAN ROOYEN, N. 2008a. Vegetation of the Hantam-Tanqua-Roggeveld subregion, South Africa. Part 1. Fynbos Biome related vegetation. Koedoe 50, 61-71.

VAN DER MERWE, H., VAN ROOYEN, M.W. AND VAN ROOYEN, N. 2008b. Vegetation of the Hantam-Tanqua-Roggeveld subregion, South Africa. Part 2. Succulent Karoo Biome related vegetation. Koedoe 50, 160-183. VEECH, J.A. 2000. Choice of species-area function affects identification of hotspots. Conservation Biology 14, 140-147. WERGER, M.J.A. 1974. On concepts and techniques applied in the Zürich-Montpellier method of vegetation survey. Bothalia 11, 309–323. WILSON, M.V. AND SHMIDA, A. 1984. Measuring Beta diversity with presence-absence data. Journal of Ecology 72, 1055-1064.

Chapter 4

Vegetation of the Hantam-Tanqua-Roggeveld subregion, South Africa. Part 1. Fynbos Biome related vegetation

(VAN DER MERWE, H., VAN ROOYEN, M.W. AND VAN ROOYEN, N. 2008a. Vegetation of the Hantam-Tanqua-Roggeveld subregion, South Africa. Part 1. Fynbos Biome related vegetation. Koedoe 50, 61-71.)

35 African Protected Area Conservation and Science 61 , KOEDOE Original Research egion r . in press) will report on sub

et al eld v STUDY AREA egetation

ogge . -R ica r Information on the spatial, temporal and ecological properties understanding,improvedthe to vegetationunitstheleadcan of protection and management of natural resources. Progression of the SKEP available on the biodiversity initiative of the Hantam-Tanqua-Roggeveld soon future showed subregion,to keywhich planning,was conservation and the paucity TheCriticaldevelopment. Ecosystem Partnership of Fund(CEPF), data which is a joint initiative Global of Environmental Conservation Facility, International, the Government the of MacArthur Japan, Foundation the and Bank,the World therefore granted funding for botanical studies in the subregion. The first stepsubregionentirethe of of surveyvegetationbroad-scalesystematic the botanical study of approximately three was million hectares, towhich could be undertakeused as the a basis for further detailed botanical survey revealed investigations.two distinct vegetationThe groups, thei.e. Fynbos Biome related Mountain Renosterveld vegetation the groupSucculent Karoo and Biome related vegetation group. The aim the of present article is reportto theon Mountain Renosterveld vegetation, depicting its component vegetation units on a map. A second article (Van der Merwe the latter vegetation group. The Hantam-Tanqua-Roggeveld subregion (Fig. 1), as defined asin 1), subregion (Fig. Hantam-Tanqua-Roggeveld The rainfallregion winterpredominantly the in lies study, current the of the Northern and Western Cape Provinces of South Africa, and covers an area of approximately In the threewest it stretches million from east hectares. of the Cederberg Mountains elated f r

anqua

A T - South AfricaSouth ABSTRACT iome University of Pretoria . 2000, outh B Department Science of Plant yen Noel van Roo Vol. 50 No. 1 pp. 61 - 71 50 61 No. 1 pp. Vol. e-mail: [email protected] et al S Merwe Helga van der antam Correspondence Helga van to: der Merwe an Rooyen Margaretha W. V epartment Plant of Science, University Pretoria, of Pretoria, 0002 H ynbos dress: D the

1: F 1: t of

r Postal Ad a P The Succulent Karoo Hotspot stretches along the western Namibia. side of A the lack Republic of of South botanical Africa informationand on Karoo the Hotspot Hantam-Tanqua-Roggeveld was identified area during of thethe SKEP Succulent(Succulent Karoofrom CEPF (Critical Ecosystem Ecosystem PartnershipPlan) process. Fund) funded a study to Aproduce a vegetation map of grantthe area to serve as baseline for ecosystem management. Vegetation surveys were conducted over an area of more than three million hectares from August to October 2004.major floristicTwo unitswere identified, namely the FynbosBiome related(Mountain Renosterveld) and Succulent Karoo Biome predominantly Mountain related Renosterveld vegetation unit units. is presented in this An Three paper. associations analysis which contains subdivided subassociations,variants.nine were whichinto four of identified, wereone of the floristic dataThe of vegetation units are described the in terms of their species composition and their relationshipsdepictingprovidedgeographical thewithis environment. distributionvegetation map physical the the A of different vegetation types. The main threat to the vegetation of the regioncommunity identified was a lackinfrastructure. of by the farming Keywords: Mountain Renosterveld, vegetation map phytosociology, Succulent Karoo, vegetation classification, egetation V CEPF CEPF 2003) and one of only two hotspots that are entirely arid (Conservation International – website). In 2002 the launched Succulent to Karoo identify EcosystemSucculent the and for strategysustainableland-useconservation and Plan generate consensus (SKEP) for was Karoo a Hotspot. SKEP aims 20-yearto meet the quantitative targets for the conservation of vegetation andendemic species particularat globally sites, as well as critical ecological threatened and and evolutionary processes that must be conserved to theensure persistence of these – website). species (Conservation International For management purposes, the SKEP initiative subdivided the Succulent Karoo into four subregions, of which Tanqua-Roggeveld the constituted Hantam- one. In common with the of the Succulent rest Karoo, the vegetation of the Hantam-Tanqua- Roggeveld subregion includes a geophytes wide and range annuals. of autumn After succulents, and spring good displays rains, of region wild attract the large flowers numbers of in spectaculartourists. parts Unlike many parts of of Namaqualand, the such brilliant shows of annuals and geophytes are not only a feature of undisturbed fallow natural fields, vegetation butin the alsoHantam and Roggeveld occur & Smith in Wyk 2001). (Van the The identification,description and classification of vegetation units across the landscape comprise the building critical first a steps framework in for ecosystem management planning. The Succulent Karoo, which stretches along the of western South side Africa and Namibia, is one recognised of the by the global IUCN hotspots as of biodiversity (Myers http://www.koedoe.co.za Original Research Van der Merwe, Van der Merwe & Van Rooyen

in the southwestern corner, northwards along the Bokkeveld with the mean of six snow days recorded over a 24-year period Mountains to just north of Loeriesfontein. The eastern border by the Weather Bureau (1998). At Sutherland the mean maximum includes the Roggeveld and Nuweveld Mountain Ranges to just for the warmest month, January, is 27.1°C, while, the extreme southwest of Fraserburg, while the southern limit includes the maximum recorded was 35.5°C in January 1980 (Weather Tanqua and Ceres Karoo to where the Swartrug Mountains and Bureau 1998). The mean minimum for the coldest month, July, the Bontberg Mountains meet north of Ceres. is –2.4°C, while the extreme minimum, -13.6°C, was recorded in July 1970 and August 1978 (Weather Bureau 1998). The Mountain Renosterveld discussed in the current article is found on the Roggeveld, Nuweveld, Komsberg, Klein Rocks of the Ecca Group cover most of this area with Dwyka Roggeveld, Koedoesberg and Hantam Mountains. In general, (consisting of tillite, sandstone, mudstone and shale) cropping this is the higher-lying part of the larger subregion that is out in the west and the Beaufort Group in the east (Council for actually situated in the Fynbos Biome (Rutherford & Westfall Geoscience 1973, 1983, 1989, 1991, 1997, 2001). The Ecca Group 1986). This area includes Acocks’s (1988) Mountain Renosterveld includes sediments of the Koedoesberg Formation (consisting of (Veld Type 43), which is equivalent to the Escarpment Mountain sandstone and shale) and the Tierberg Formation (consisting of Renosterveld (Unit 60) of Low and Rebelo (1998). According to shale) (Council for Geoscience 1973, 1983, 1989, 1991, 1997, 2001). Mucina et al. (2005), six vegetation types are represented in the Mudstones of the Beaufort Group are found on the eastern side area, namely the Nieuwoudtville Shale Renosterveld (FRs 2); of the study area (Council for Geoscience 1973, 1983, 1989, 1991, the Roggeveld Shale Renosterveld (FRs 3); the Central Mountain 1997, 2001). Igneous intrusions of dolerite occur throughout Shale Renosterveld (FRs 5); the Nieuwoudtville Roggeveld the region, being easily recognisable as very hard, dark grey Dolerite Renosterveld (FRd 1); the Hantam Plateau Dolerite to nearly black rocks (Van Wyk & Smith 2001). The soils of the Renosterveld (FRd 2); and the Roggeveld Karoo (SKt 3). Roggeveld consist primarily of clays and silts derived from the Karoo sequence shales (Low & Rebelo 1998) and are found The earliest references to the botanical wealth of the Hantam- on the slopes and foothills of the Great Escarpment along the Roggeveld date from the early 1900s. Diels (1909 in Van Wyk various mountain ranges. & Smith 2001) mentioned the high levels of endemism in the Hantam-Roggeveld and provided a useful floristic analysis of the region. He concurred with Marloth (1908) that the region METHODS AND MATERIALS is floristically more closely related to the Succulent Karoo and the Great Karoo than to the Cape Floristic Region, although Satellite images (Bands: 4, 5 and 3 (R,G,B)) of the study area Cape floristic elements are clearly present, especially on the were visually stratified into relatively homogeneous units on Hantamsberg (Van Wyk & Smith 2001). The Roggeveld was the basis of colour, texture and topography. This stratification also one of the three centres of endemism that Hilton-Taylor was used to select the sites at which sample plots were surveyed (1994) identified within the Western Cape Domain, the other from August until October 2004. At each site GPS (Global two centres being the Western Mountain Karoo and Tanqua Positioning System) coordinates were taken and each species Karoo, which also fall within the Hantam-Tanqua-Roggeveld present in the plot was noted and assigned a cover-abundance subregion. Van Wyk and Smith (2001) combined the Hantam- value according to the Braun Blanquet cover-abundance scale Roggeveld into one of their 13 principal centres of plant (Werger 1974). Various environmental characteristics, such as endemism in southern Africa and stressed the unique botanical altitude, topography, aspect, slope, an estimation of rock cover, importance of this area. the size of the rocks, soil type and colour, and the degree of erosion were noted at each sampling point. Biotic effects, such The rainfall ranges from 132–467 mm per year (Weather Bureau as trampling, small mammal activity, or invasion by alien 1998), which, although falling mainly in winter, includes a few plants, were also recorded. summer thunderstorms. In 2004 the rainfall season was poor and the usual winter snowfalls on the high-lying areas were A total of 390 sample plots covering the entire Hantam-Tanqua- limited to the light snow that fell on one occasion, compared Roggeveld subregion were surveyed in 2004. An analysis of the floristic data was undertaken using the TURBOVEG and MEGATAB computer package (Hennekens & Schaminée 2001). Vegetation data were captured with the TURBOVEG software and the data were classified with the aid of MEGATAB.

As a first step to the classification of the floristic data, a Two-Way Indicator Species Analysis (TWINSPAN) (Hill 1979) was run in MEGATAB. The result of the TWINSPAN on the entire data set confirmed the presence of two distinct floristic groups, which enabled the data set to be split into two. A TWINSPAN was then run separately on each data set, with the resulting tables being further refined to obtain clear species assemblages. The first phytosociological table, which characterises the vegetation of the predominately Mountain Renosterveld as defined by Acocks (1988), is discussed in the current article.

The major vegetation units distinguished in the Mountain Renosterveld were termed associations following the use as defined by Nelder et al. (2005). Associations are produced on the basis of the presence and abundance of species, vegetation structure and the spatial distribution of individuals in the dominant layer. Subassociations are generally distinguished on the basis of elements in the subdominant layers. The subassociations are described in terms of a list of species featuring each structural layer, together with its canopy cover. FIGURE 1 Subregions in the SKEP planning domain (CEPF 2003). African Protected Area Conservation and Science

62 KOEDOE Vol. 50 No. 1 pp. 61 - 71 http://www.koedoe.co.za Fynbos biome related vegetation Original Research

FIGURE 2* Vegetation map the of Mountain Renosterveld vegetation the of Hantam-Tanqua-Roggeveld subregion. *Enlarged figure is available online. African Protected Area Conservation and Science and Conservation Area Protected African

http://www.koedoe.co.za Vol. 50 No. 1 pp. 61 - 71 KOEDOE 63 Original Research Van der Merwe, Van der Merwe & Van Rooyen

.3 3 . 2 4 2 .1. 2 .1.3 2 .1. 1 2 2.1. BLE 1 a T 1.3 1.4 1.5 1.2 Phytosociological table the of predominantly Mountain Renosterveld vegetation the of Hantam-Tanqua-Roggeveld subregion 331232312122231332333333223233332221 1 1111 12331 333 2222323 1 2 113 11111333323311133113333 1.1 1.aa....1.a...... ++..+...+...... +...... +..++...... ++...... ++..+++.+...... 11..1...... +...... r..+...... +...... + ...... +.+...... +.1...... +...... +...... a...... a...... ++...... +...... +1+++...... +..+...... ++.++..+.....+...... ++...... r....+...... +..+...... r...... ++...... +...... +...... ++...... +...... +...... +...... +...... +...... +....++.++.+++...... +...... +.+...... +...... +...... +.....+...... +++.+++.+++.+.+..++.+++..+.++..+1a+.+...+++++....+...... +.+..+...... +...+...+++...... +++.+++.+++..+++.++..+..+....+r....+...... +...... ++...... +++...... +.+.....+++.1+.....+.+++.+++...... ++...... +...... b...... ++.+....++...... +..+.1...... +..+...... +...... +...... 1...... 1...... 1...... +...... 3 3 3 311123111112464904838774500703322287392234788979439539233344383366785883832528888446204589666374459946096767 3 3 334512012670591274483153056 3 3 490334922515662358529616328540345124935592167089227513794408523127498644206108091 ...... +...... r.r.+r++...... +...+...... +..++...... +...r...... +...a++...... +...... +...... 1.1..++.1+...... +...... ++...... ++....++r...... 1...... +..a.+.....+.+...... +....+...... +.....1...++...... ++...... ++...+...... +...... +...... +...... +...... +++....+..+...... +...... +...... ++...... +...... +...... +...... +.+...... ++...... +...... +...... ++...... +...... r+....+...... +...... +...... +...... +++...... ++...... +...... ++1++b...++a1..+..++111.+++++a.3.a3..a+33..1...... +...+...... +...... +...... E C B G I sp. sp. GROUP A GROUP D GROUP GROUP F GROUP GROUP GROUP GROUP H GROUP GROUP J s s s s s s s s s s a genistifolia ra gnaphalodes ia incana nia glomerata ephalus microphyllus ELEVÉ NUMBER PECIE PECIE PECIE PECIE PECIE Erioc Pentzia (HRp317) Euryops imbricatus Diospyros austro-africana S Salvia disermas Chrysanthemoides incana Aptosimum indivisum Stipagrostis namaquensis Braunsia sp. S Euryops multifidus Phyllobolus tenuiflorus SPECIE Antimima cf. granitica (HR248) Pelargonium Senecio erosus Lotononis sp. Asparagus asparagoides S Pentz Rosenia oppositifolia Ptero Mesembryanthemum guerichianum Androcymbium volutare SPECIE Lycium spp. Karroochloa schismoides Eriocephalus decussatus Pentzia cf. sphaerocephala Ruschia cradockensis S R Galenia africana Medicago sp. SPECIE Oeder Ursinia pilifera Ruschia intricata Ursinia calenduliflora SPECIE Poa bulbosa Galenia sarcophylla Cromidon varicalyx Eriocephalus paniculatus Lotononis hirsuta S Plantago cafra SPECIE Leyse Selago sp. African Protected Area Conservation and Science

64 KOEDOE Vol. 50 No. 1 pp. 61 - 71 http://www.koedoe.co.za Fynbos biome related vegetation Original Research

3 .1.4 2 .2 2.3 2 .1.3 2 .1.2 2 2.1.1 .5 1.3 1.4 1 1.2 African Protected Area Conservation and Science and Conservation Area Protected African 3331232312122231332333333223233332221 1 1111 12331 333 2222323 1 2 113 11111333323311133113333 1.1 ...... b+++...... +...... +...... +...... +.r...... +...... +r...... +...... +...+...... +...... r....1..+...... 1...... +.++.+b...... +...... +.+...... +.+...... +...... +....+...... 1...... r..a...... +...... ++.1r..++++++..+...... +...... r...+...... +...... r..++...... +r...... +...... +...... +..r...++.++.+...+.....+...... r....+...... +...... +...... r.+...++.r...... +...... +...... +...+...... +...++..+.+...... +...... +++..r...... ++1...... +...+...... r...... +....+.+...... +...... +...... +....+...... +.....+.+...... +..+...... +...... +....+...... +.....+...... +...... +...... +...... ++...... ++...... +.+...... +.+.+.+...... +...... +...... +..r...... +...... +...... r....r...+...... +...... +.....+.+...... +..+...... 13....++...... ++....+++++++...+.+.+++...... +...... ++...... +.+...... +..+...... +.++...+...... +++..+.++...... +..+.....+.+...... ++...... +.++.+.+...+.+.++.+....++++..++++.+..+...... +...... ++...... +...+.+.....+...... +...+++...... +.....+.++..+.+..+.+..+...... +...... +.+...... +...... +.+...... ++...+.+.+.+...... +...... ++...... +..+...... +...... +...... +..+...... +...... +.....r..++...... +...... +..+...... +...... +...... +...... +...... +...... a1...... +...... b...... 1....+...... +...... +..+...+...... +.1...... +...... +...... +.+.+...... +...... ++...... +....+...... +...... +...... +...... +...... +.+...... +.. 3 3 3 311123111112464904838774500703322287392234788979439539233344383366785883832528888446204589666374459946096767 3 3 334512012670591274483153056490334922515662358529616328540345124935592167089227513794408523127498644206108091 3 ...... +...... +..+....+...... +...... +.....+...... +...+...... +....+.+...... +...... +....+...... +...... r...... +...... +...... +...+...... +...... +...... +....+...... ++...... +....+...... +...++.+.+++...... +...r...... +...... ++....+...+.+..+..+.+.+...... +...... +...... +...... ++.+....++.....+.+..+...... +...... P O N M L sp. sp. sp. GROUP GROUP GROUP GROUP GROUP GROUP K spp. s s s s s s sp. E I la nudicaulis C E ELEVÉ NUMBER PECIE PECIE PECIE P Table 1 (cont...) Table Passerina truncata Cliffortia arborea Agathosma Othonna auriculifolia Hesperantha sp. Pteronia glauca Helichrysum asperum Androcymbium sp. Stachys rugosa Leipoldtia (HRp347) sp. S Felicia filifolia Polygala scabra Ehrharta melicoides SPECIE Hermannia cuneifolia Dorotheanthus Helichrysum hamulosum Oedera sedifolia Tripteris aghillana Ursinia nana Cyperaceae Tetragonia microptera Tylecodon wallichii Dorotheanthus maughanii Scleranthus annuus Pelargonium grossularioides Festuca scabra S Erodium cicutarium Bromus pectinatus Senecio cakilefolius Felicia australis Heliophila spp. Arctotheca calendula Helichrysum lucilioides Eriocephalus pauperrimus Leysera tenella Gazania rigida Adenogramma glomerata S R Tribolium hispidum Albuca Ehrharta calycina SPECIE Cotu Pharnaceum aurantium Berkheya sp. S Polycarena aurea Nemesia http://www.koedoe.co.za Vol. 50 No. 1 pp. 61 - 71 KOEDOE 65 Original Research Van der Merwe, Van der Merwe & Van Rooyen

Using the distribution of the sample plots, supported by 1:250 000 topocadastral maps, land type maps (Agricultural Research Council 1986a, 1986b, 1995, 1999a, 1999b, 2002, 2003), geology maps (Council for Geoscience 1973, 1983, 1989, 1991, 1997, 2001) and satellite images, the stratified units were assigned to a vegetation unit in the floristic classification. Where two or more subassociations were present in a mapping unit, but it was not possible to map them separately as a result of high spatial diversity, they were mapped as mosaics of specific vegetation types.

Species that were unidentifiable during the field surveys were

collected and the herbarium specimens sent to the Compton Herbarium, Kirstenbosch, for identification. The collection code (HR) and numbers of the specimens were kept throughout the process as not all the species, especially within the Mesembryanthemaceae, have yet been positively identified. All voucher specimens are lodged at the Schweickerdt Herbarium (PRU), University of Pretoria, Pretoria. Nomenclature follows that of Germishuizen and Meyer (2003).

RESULTS

The floristic data analysis resulted in two phytosociological tables. The first table (Table 1) contains the predominantly Mountain Renosterveld veld type, as defined by Acocks (1988), or the Escarpment Mountain Renosterveld vegetation type, as defined byL ow and Rebelo (1998), and is described in this article. Three associations were identified, which were subdivided into nine subassociations, one of which contains four variants, as set out in the following scheme: 1. Rosenia oppositifolia Mountain Renosterveld 1.1 Eriocephalus microphyllus – Rosenia oppositifolia Mountain Renosterveld 1.2 Antimima cf. granitica (HR248) – Rosenia oppositifolia Mountain Renosterveld 1.3 Pentzia incana – Rosenia oppositifolia Mountain Renosterveld

1.4 Euryops multifidus – Rosenia oppositifolia Mountain Renosterveld 1.5 Pteronia glomerata – Rosenia oppositifolia Mountain

Renosterveld 2. Dicerothamnus rhinocerotis Mountain Renosterveld 2.1 Erodium cicutarium – Dicerothamnus rhinocerotis Mountain Renosterveld 2.1.1 Galenia africana – Dicerothamnus rhinocerotis Mountain Renosterveld 2.1.2 Oedera genistifolia – Dicerothamnus rhinocerotis 1.3 1.4 1.5 2.1.1 2.1.2 2.1.3 2.1.4 2.2. 2.3 3 Mountain Renosterveld 2.1.3 Senecio cakilefolius – Dicerothamnus rhinocerotis Mountain Renosterveld 2.1.4 Euryops lateriflorus – Dicerothamnus rhinocerotis

1.2 Mountain Renosterveld 2.2 Dimorphotheca cuneata – Dicerothamnus rhinocerotis Mountain Renosterveld

33331232312122231332333333223233332221 1 1111 12331 333 2222323 1 2 113 11111333323311133113333 2.3 Merxmuellera stricta – Dicerothamnus rhinocerotis Mountain Renosterveld 1.1 3. Passerina truncata Mountain Renosterveld +3...... r+...... a...... r..++.r...1.a+1.4a+a.a+1++..ara1a+.44.+.b1++a.1a.1bb1a131a..b..4.b+3...... +...... 3....+....11++3.a....+..3.+..a1+++++.+a..r13+.+..ab.4+.a+.+....+...... +.+++...+...... +...... ++.+++++1+..+.++.+.++++.++++..++.++.+..+.+.++.+....+....+++.+..++++...... +...... +++++++++++.++.++++++++++....++.+++++++.++.+...+.+...+++..+...+...... +...... +...... +.+++.+.+....+...... ++..+.a1+++.+...11+++r...++...+...... ++...... +...... +...... +.+.++...+...... +.....+.+...... ++.+...+...... +...... +..+...... +.+.+.+...... ++.+.+...... a...+.+...... +.+...... +...... +.+.+++++++...+.1++..++.+++a++...a.++....+.++++...+..+.+...... ++.+a...+....++.++++.r+..1.+++.r.1.1.+r...r.+....++..+..+++...... +...... ++..+.+++...... r+..+...++..+.+.+...r.r...... r.....+..+r.+..+r...... +.1+.+.++++++.+..++.....++.+....++.+.++.+....++..+r..+++++....+...++...... +....+....+.++.+..+.+...a1..+.+...... 1.+a...... a+11...... ++.1.....++1a+1....1.++11..1.1....1++.+1..++11.....r..+1.b...1a1..+.....a..1...+ab...+...+...... a.+...... +.....+.+1....+.1....+...++...... ++...... +.r+...... ++.+...... +...... +.. 3 3 3 311123111112464904838774500703322287392234788979439539233344383366785883832528888446204589666374459946096767 3 3 3 34512012670591274483153056490334922515662358529616328540345124935592167089227513794408523127498644206108091 With the exception of unit 1.2, all vegetation units could be mapped (Fig. 2). Three additional mosaics were also mapped: • the Nieuwoudtville mosaic, consisting of vegetation units

2.1.1, 2.1.4, 2.2 and 5.1 (Van der Merwe et al. in press); • the Soekop mosaic, consisting of units 2.1.3 and 2.2; and the • Welgemoed mosaic, consisting of units 1.2, 1.3 and 1.5.

R A large number of different land types are found in the study

ies in a taxonomic group area and therefore only the predominant types are listed for an one species in a taxonomic group. These species, even although they are grouped together, are included in the table since they occur in different species groups. However, they are not used in the descriptions in the text. each vegetation unit (Agricultural Research Council 1986a, group Q group s inadequate for identification yet different from species that could be identified are indicated with a plot number(p…) s s 1986b, 1995, 1999a, 1999b, 2002, 2003). Table 2 summarises the

erothamnus rhinocerotis most important features of the various land type symbols that elevé number pecie Specie Dic Merxmuellera stricta Geophytic spp. Oxalis spp. Dimorphotheca cuneata Heliophila crithmifolia Zaluzianskya benthamiana Gnidia scabra S Asparagus capensis Moraea spp. Euryops lateriflorus Eriocephalus ericoides Helichrysum obtusum Chrysocoma ciliata R Table 1 (cont...) Table * Non-diagnostic species are excluded * HR collection code and numbers are included for future reference, if necessary * Specimen * sp. = one spec * spp. = more th have been used in the text. African Protected Area Conservation and Science 66 KOEDOE Vol. 50 No. 1 pp. 61 - 71 http://www.koedoe.co.za Fynbos biome related vegetation Original Research

Description of plant communities (Table 1, Fig. 2) Table 2 Land type symbols and their meaning within the text (Du Plessis 1987) The 2004 winter season was extremely dry, resulting in annuals and geophytes being poorly represented in the survey. The Land type Meaning of symbol following description will therefore focus on the perennial D Prismacutanic and/or pedocutanic diagnostic horizons dominate. plant species with permanent above-ground organs. After subtracting exposed rock, stones or boulders, more than half of the remaining land must consist of duplex soils.

1. Rosenia oppositifolia Mountain Renosterveld Da Refers to land where duplex soils with red B horizons comprise more than half of the area covered by duplex soils. This plant association is located at the southern end of the Db Refers to land where duplex soils with non-red B horizons Roggeveld and Nuweveld Mountains as well as in the vicinity comprise more than half of the area covered by duplex soils. of the farms Onderplaas and Droëkloof further north and F Glenrosa and/or Mispah forms (though other soils may occur). The occurs predominantly on Land Types Fc and Da. Mudstones group accommodates pedologically young landscapes that are not of the Beaufort Group as well as dolerites are found underlying predominantly rock, alluvial or aeolian and in which the dominant soil-forming processes have been rock weathering, the formation this association. The association is generally found on level of orthic topsoil horizons and clay illuviation. terrain, gentle or moderate sloping ridges with a low rock Fa Refers to land in which lime is rare or absent from the entire cover from 0 to 10% or a high rock cover of 70 to 90%. Brown or landscape. light brown sandy soils are prevalent in this high-lying plant Fb Indicates land where lime occurs regularly (though possibly in association. Although a high shrub cover is present, the grass small quantities) in one or more valley bottom soils. and annual component only sometimes feature, usually with Fc Refers to land where lime is generally present throughout the less than 5% cover. entire landscape.

I Miscellaneous land classes. The vegetation of this association is characterised by species Ia Refers to land types with a soil pattern difficult to accommodate group F with a high cover of Rosenia oppositifolia and includes elsewhere, at least 60% of which comprises pedologically youthful, species such as Pteronia glomerata and Karroochloa schismoides. deep (more than 1 m to underlying rock) unconsolidated deposits. Common species include Chrysocoma ciliata, Euryops lateriflorus Ib Indicates land types with exposed rock (country rock, stones or and Eriocephalus ericoides (species group R). This association has boulders) covering 60–80% of the area. been subdivided into five subassociations.

1.1 Eriocephalus microphyllus – Rosenia oppositifolia 1.3 Pentzia incana – Rosenia oppositifolia Mountain Mountain Renosterveld Renosterveld This unit is found in the region of the Nuweveld Mountains Subassociation 1.3 is located around Sutherland and east of and covers an area of 106 454 ha (13.2% of the total area covered Sutherland on mudstones of the Beaufort Group. It also occurs by Mountain Renosterveld vegetation). Geologically, this further south in combination with subassociations 1.2 and 1.5 in subassociation is found on mudstones of the Beaufort Group the vicinity of the farm Welgemoed, at the foot of the Komsberg and predominantly on Land Type Fc, indicating that there is Mountains. This subassociation, excluding the mosaic unit, lime present in the entire landscape. This high-lying vegetation covers an area of 44 499 ha (5.5% of the total area covered by occurs at an altitude of > 1400 m above sea level on ridges with Mountain Renosterveld vegetation). Land types include Fc and level terraces to gentle slopes. The rock cover varies from zero Da and altitude ranges from approximately 1300–1500 m above to > 85%, and is usually comprised of stones (> 50–200 mm) and sea level. The high-lying ridges on level terrain to moderate boulders (> 200 mm). slopes are usually covered with brown or light brown sandy soils. Shrub cover is generally high (mean cover 66%) and is characterised by species such as Rosenia oppositifolia and Pteronia The high shrub cover is attributed to species such as Pentzia glomerata (species group F) as well as the diagnostic species incana (species group D), Rosenia oppositifolia and Pteronia Eriocephalus microphyllus, Pentzia sp. (HRp317) and Euryops glomerata (species group F) as well as Chrysocoma ciliata and imbricatus (species group A). Other shrubs also present include Euryops lateriflorus (species group R). This subassociation shows Chrysocoma ciliata, Euryops lateriflorus and Eriocephalus ericoides local variations resulting from a low constancy of such species of species group R. Grasses are either absent or their cover as Stipagrostis namaquensis, Braunsia sp. and Chrysanthemoides is limited to less than 5%, while annuals are seldom present. incana (species group C). When the perennial shrub cover is This phenomenon could, however, be a result of the drought high, species in group C do not occur, however, when the shrub conditions experienced during the time in which the surveys cover is lower, species in group C can dominate. Generally, the were conducted. cover of the grass and non-grassy herbaceous layer is limited, except in the case of the grass species Stipagrostis namaquensis African Protected Area Conservation and Science and Conservation Area Protected African 1.2 Antimima cf. granitica (HR248) – Rosenia oppositifolia that occurred in a single relevé sampled in a drainage line. Mountain Renosterveld 1.4 Euryops multifidus – Rosenia oppositifolia Mountain Dolerite-derived B horizon soils on Land Types Da and Fc Renosterveld characterise this subassociation that is scattered throughout the Roggeveld Mountains and has not been mapped as a Located just north of the Komsberg, subassociation 1.4 is found separate unit. The altitude generally ranges from 1200 to 1350 m predominantly on mudstones of the Beaufort Group and covers above sea level. The high-lying ridges with level terrain to an area of 106 189 ha (13.1% of the total area covered by Mountain moderate slopes on brown to red brown sandy soils are usually Renosterveld vegetation). Land types present include Fc and covered with stones (> 50–200 mm in size). Shrub cover in this Da, with, occasionally, deep deposits of the Ia land type. This unit is high (with a mean value of 72%), while the herbaceous subassociation is found at an altitude of between 1400–1500 m component is generally < 5%. Almost no grasses contribute to above sea level. The level to gently sloped ridges and light the herbaceous cover. brown soils in this subassociation support a high shrub canopy cover of between 60 and 90%. The shrub layer is characterised by species such as Rosenia oppositifolia, Pteronia glomerata (species group F), Eriocephalus The diagnostic species Euryops multifidus and Phyllobolus pauperrimus (species group L) as well as Asparagus capensis and tenuiflorus (species group E), together with Rosenia oppositifolia Eriocephalus ericoides (species group R). Diagnostic perennial (species group F), Chrysocoma ciliata and Eriocephalus ericoides species include Antimima cf. granitica (HR248) and Pelargonium (species group R) characterise this subassociation. The cover of sp. (species group B). the herbaceous component (including grasses) is usually limited http://www.koedoe.co.za Vol. 50 No. 1 pp. 61 - 71 KOEDOE 67 Original Research Van der Merwe, Van der Merwe & Van Rooyen

to < 5%, which could be the result of the drought conditions in as well as Euryops lateriflorus (species group R). Common annual the year of survey. species occurring in the unit include Erodium cicutarium, Bromus pectinatus, Senecio cakilefolius and Felicia australis (species group 1.5 Pteronia glomerata – Rosenia oppositifolia Mountain L). This subassociation has been subdivided into four variants. Renosterveld Geologically, this subassociation occurs predominantly 2.1.1 Galenia africana – Dicerothamnus rhinocerotis Mountain on mudstones of the Beaufort Group and is similar to Renosterveld subassociations 1.3 and 1.4. It is found on Land Types Fc, Da and This variant is floristically very diverse and occurs on the Db on the southwestern extreme of the Roggeveld Mountains. mudstones of the Beaufort Group and the shales of the Ecca It also occurs in a mosaic with subassociations 1.2 and 1.3 in Group. It is located in the region of the farms M’Vera and the vicinity of the farm Welgemoed at the foot of the Komsberg Vondelingsfontein at the northern extreme of the Roggeveld Mountains. This subassociation covers an area of 69 233 ha Mountains and Kareebos and Rooiwal west of the Klein (8.6% of the total area of the Mountain Renosterveld vegetation), Roggeveld Mountains. It also forms a mosaic in combination excluding the mosaic unit. This high-lying (1200–1600m above with variant 2.1.4, subassociation 2.2, and subassociation 5.1 sea level) subassociation is found on level terrain and gentle (Van der Merwe et al. in press) in the Nieuwoudtville area. slopes on a range of different rock sizes, varying from gravel Excluding mosaics, this variant covers an area of 25 369 ha (< 10 mm) to boulders (> 200 mm). The soil colour also varies (3.1% of the total area of the Mountain Renosterveld vegetation). substantially from light brown to brown to red brown. Various land types are present, predominantly of the Da and Fb types. The altitude is notably lower than for the other Shrub cover is generally high (> 60%) and the grassy component vegetation units, and varies from 600 to 1300 m above sea level. has a higher presence and cover compared with the previous This variant occurs on undulating terrain. The light brown to subassociations. Likewise, annual species are present in all brown coloured sandy soils are usually not covered by much relevés, with their cover generally being higher than in the rock, however, boulders (> 200 mm) do occur locally. previous subassociations. This was probably the case due to the local rain showers received in the area during the year of A high shrub cover results primarily from the presence of survey. Dicerothamnus rhinocerotis (species group Q) as well as the diagnostic species Galenia africana (species group G). Various No diagnostic species group separates this subassociation. The annual species, such as Cotula nudicaulis and Polycarena aurea most prominent species present include Rosenia oppositifolia, (species group K) and Erodium cicutarium, Senecio cakilefolius, Pteronia glomerata (species group F), Erodium cicutarium (species Felicia australis and Leysera tenella (species group L), are present. group L) and Euryops lateriflorus (species group R). The grass The annual grass Bromus pectinatus (species group L) contributes component is represented by Karroochloa schismoides (species to the low cover of the grass component in the variant. The group F) and Bromus pectinatus (species group L) with, in one absence of species group J in this variant differentiates it from relevé, a high cover of Merxmuellera stricta (species group Q). variant 2.1.2. The presence of Galenia africana and various Erodium cicutarium and Felicia australis (species group L) as well annuals indicates the increased disturbance that has taken as Heliophila crithmifolia (species group Q) represent some of the place in this variant in the past. annual species. 2.1.2. Oedera genistifolia – Dicerothamnus rhinocerotis 2. Dicerothamnus rhinocerotis Mountain Renosterveld Mountain Renosterveld This plant association is located in the Roggeveld, Klein Variant 2.1.2 occurs on the mudstones of the Beaufort Group in Roggeveld, Koedoesberg and Komsberg Mountains and has the Klein Roggeveld Mountains, and covers an area of 46 797 ha been further subdivided into three subassociations. Generally (5.8% of the total area of the Mountain Renosterveld vegetation). it can be found on the mudstones of the Beaufort Group or the It is found at an altitude between 1000 and 1300 m above sea shales of the Ecca Group on Land Types Da, Fb, Fc, Ib and Fa. level on level terrain to gentle slopes. The light brown sandy The level terrain and gentle slopes of subassociations 2.1 and soils of this variant are covered with gravel (< 10 mm), small 2.2 as well as the gentle to moderate slopes of subassociation 2.3 stones (> 10–50 mm) and boulders (> 200 mm), which are typical are usually comprised of light brown or brown sandy soils. The of Land Type Ib. shrub cover is high (50–95%) and a grass and annual component are generally present throughout the association. The very The high shrub cover (more than 70%) is primarily a result of the high cover of Dicerothamnus rhinocerotis, Merxmuellera stricta presence of Dicerothamnus rhinocerotis (species group Q) as well and Dimorphotheca cuneata (species group Q) distinguishes this as Oedera genistifolia (species group H) and Euryops lateriflorus association from the Rosenia oppositifolia Mountain Renosterveld (species group R). Merxmuellera stricta (species group Q), a (association 1). perennial grass, dominates the grass component of this variant. Annual species are consistently present, however, their cover 2.1 Erodium cicutarium – Dicerothamnus rhinocerotis is low due to the drought conditions in the year in which the Mountain Renosterveld surveys were conducted. Subassociation 2.1 generally occurs on mudstones of the Beaufort Group in the Roggeveld, Klein Roggeveld and Komsberg The presence of species groups G and J distinguishes variant Mountains on Land Types Da, Fb, Fc and Ib, and excluding 2.1.1 from variant 2.1.2, whereas the absence of species group I mosaic units, covers an area of 213 410 ha (26.4% of the total area distinguishes variant 2.1.2 from variant 2.1.3. of Mountain Renosterveld vegetation). The altitude varies from 600 to about 1600 m above sea level and the landscape is gently 2.1.3 Senecio cakilefolius – Dicerothamnus rhinocerotis undulating. Soils are light brown sandy soils with a varying Mountain Renosterveld rock cover consisting predominantly of boulders (> 200 mm). This variant, excluding mosaics, covers an area of 13 654 ha (1.7% The shrub cover is generally high (50–95%), with grass and of the total area covered by Mountain Renosterveld vegetation) other herbaceous species consistently occurring across all the and is found on mudstones of the Beaufort Group and shales of surveyed sites. the Ecca Group in the region of the farms Botuin, Blomfontein and De Hoop in the Roggeveld Mountains, predominantly on Prominent perennial species in this subassociation include Land Types Da and Fc. In combination with subassociation Dicerothamnus rhinocerotis, Merxmuellera stricta (species group Q) 2.2 in the region of the farm Soekop, it is found in a mosaic African Protected Area Conservation and Science 68 KOEDOE Vol. 50 No. 1 pp. 61 - 71 http://www.koedoe.co.za Fynbos biome related vegetation Original Research

vegetation unit. It occurs at altitudes higher than 1200 m above rhinocerotis and other prominent species include Merxmuellera sea level on level terrain to gently sloping landscapes with stricta, Dimorphotheca cuneata (species group Q) and Chrysocoma light brown to brown coloured soils. Rocks are mostly absent, ciliata, Euryops lateriflorus and Eriocephalus ericoides (species although boulders do occasionally occur. group R).

Shrub cover varies considerably, with the main contributors 2.3 Merxmuellera stricta – Dicerothamnus rhinocerotis being Dicerothamnus rhinocerotis and Dimorphotheca cuneata Mountain Renosterveld (species group Q) as well as Chrysocoma ciliata, Asparagus This subassociation is located in the region of the farms Piet capensis, Euryops lateriflorus and Eriocephalus ericoides (species se Nuplaas, Droëberg, Nuwepos, Soekop and Vaalhoek in the group R). The grass component varies considerably depending Roggeveld Mountains and includes the higher-lying vegetation on the presence or absence of the perennial grass Merxmuellera of the Koedoesberg and Basterberg Mountains. It covers an stricta (species group Q). The most prominent annual species area of 230 838 ha (28.5% of the total Mountain Renosterveld include Cromidon varicalyx and Plantago cafra (species group I), vegetation). Geologically, it occurs on the mudstones of the Cotula nudicaulis and Polycarena aurea (species group K) as well Beaufort Group, the shales of the Ecca Group and even, as Erodium cicutarium and Senecio cakilefolius (species group L). occasionally, on dolerites occurring within the mudstones and The cover of this component is highly variable, depending on shales. Land types include Fc, Da and occasionally Ib at an the amount of rainfall received locally during the season. altitude of 900 to 1600 m above sea level. The high-lying gentle to moderately steep slopes are usually covered with stones Variant 2.1.3 has a close affinity with variant 2.1.2 due to their (> 50–200 mm) or boulders (> 200 mm). The soils are generally sharing species group J, however, they differ as a result of the brown sandy soils. Shrub and grass cover vary considerably, presence of species group I that is confined to variant 2.1.3. whereas the annual component is either absent or covers less than 1% of the area. 2.1.4 Euryops lateriflorus – Dicerothamnus rhinocerotis Mountain Renosterveld Three variations are distinguished in this subassociation. Variant 2.1.4 occurs on Land Types Da and Fc in the Komsberg The first variation is differentiated by the presence of species Mountains and southwest of the Basterberg Mountains and group N, which is shared with subassociation 2.2. The second covers an area of 127 590 ha (15.8% of the total Mountain variation is characterised by the perennial shrub Pteronia Renosterveld vegetation), excluding the mosaic vegetation glauca (species group O), whereas the third variation does unit. The mosaic is found in combination with variant 2.1.1, not include species group N or O. In all of these variations subassociation 2.2, and subassociation 5.1 in the Nieuwoudtville Dicerothamnus rhinocerotis, Merxmuellera stricta (species group Q) area (Van der Merwe et al. in press). This variant is generally and Chrysocoma ciliata (species group R) dominate with a very found at high altitudes on level terrain to gentle slopes. The light high cover (60–95%). Other species present include Asparagus brown soils are derived from mudstones of the Beaufort Group. capensis, Euryops lateriflorus and Eriocephalus ericoides (species Rocks are generally absent, although boulders (> 200 mm) may group R). occur locally. 3. Passerina truncata Mountain Renosterveld The high shrub cover is due primarily to Dicerothamnus The third plant association, which is found exclusively on rhinocerotis and Dimorphotheca cuneata (species group Q) as well dolerites on Land Type Ia, occurs at high altitudes (approximately Chrysocoma ciliata, Asparagus capensis Euryops lateriflorus as and 1500 m above sea level and higher) on the Hantam Mountain (species group R). The grass cover is generally low, except where as well as at various locations scattered throughout high-lying Merxmuellera stricta (species group Q) dominates. The cover of areas in the Roggeveld Mountains. It covers an approximate the annual component is generally low. area of 17 982 ha (2.2% of the total area of the Mountain Renosterveld vegetation). The high-lying terraces and plateaux Two forms of variant 2.1.4 occur as a result of the presence or consist of red brown sandy clay soils, with the rock cover absence of species group K, which mainly consists of annual varying from 20– 80%. The shrub cover is very high, except species. Such species might have occurred throughout the where a high cover of exposed rocks occurs. Compared to the region in a normal rainfall year. high shrub cover, the cover of the grass and annual species is generally very low. 2.2 Dimorphotheca cuneata – Dicerothamnus rhinocerotis Mountain Renosterveld This association is differentiated by species group P, which This high-lying subassociation can be found in the Keiskie includes diagnostic species such as Passerina truncata and Mountains, at the northern extreme of the Roggeveld Mountains, Othonna auriculifolia. Other common species present include and excluding mosaics, covers an area of 20 196 ha (2.5% of Dicerothamnus rhinocerotis, Merxmuellera stricta and Dimorphotheca Science and Conservation Area Protected African the total area covered by Mountain Renosterveld vegetation). cuneata (species group Q) and Eriocephalus ericoides (species It also occurs in combination with variant 2.1.3 in the region group R). of the farm Soekop and additionally, it forms a mosaic in the Nieuwoudtville area in combination with variants 2.1.1 and DISCUSSION 2.1.4 and subassociation 5.1 (Van der Merwe et al. in press). The land types present include Da, Fa and Fc and the altitude varies According to Rutherford and Westfall (1986), Low and Rebelo from 700–1400 m above sea level. The undulating terrain is (1998) and Mucina et al. (2005) the vegetation of the subregion, usually covered by a high percentage of boulders (>200 mm). as discussed in the present article, is situated predominantly The light brown to brown coloured sandy soils are derived in the Fynbos Biome. However, Diels (1909 in Van Wyk & from Ecca shales. Smith 2001) concurred with Marloth (1908) that the region is floristically more closely related to the Succulent Karoo than The shrub cover varies greatly (20–80%), whereas the cover to the Cape Floristic Region. This area was also included in the of both the grass and annual species remains low. Diagnostic SKEP initiative and not in the CAPE (Cape Action Plan for the species include Hermannia cuneifolia, Helichrysum hamulosum Environment) initiative. and Oedera sedifolia (species group M). Felicia filifolia, Polygala scabra and Ehrharta melicoides (species group N) are common The clear split between Table 1 and the table presented on the to both subassociation 2.2 and 2.3, although subassociation 2.3 Succulent Karoo related vegetation (Van der Merwe et al. in lacks species group M. The dominant species is Dicerothamnus press) reveals that most of the species in species group F (Table 1) http://www.koedoe.co.za Vol. 50 No. 1 pp. 61 - 71 KOEDOE 69 Original Research Van der Merwe, Van der Merwe & Van Rooyen

are found in the general species group in the upper portion Roggeveld subregion in terms of their species composition, of the Succulent Karoo table, whereas most of the species in environmental parameters and relationships to one another species group Q (Table 1) are not found in the Succulent Karoo as well as to map their geographical distribution. Such an table. Such a finding indicates association 1’s affinity with the inventory of vegetation types should aid future planning, Succulent Karoo Biome vegetation of the Escarpment Karoo, resource management and biodiversity conservation, which Roggeveld Karoo and Hantam Karoo, as described in Van der should encourage sustainable land use practices, reducing the Merwe et al. (in press). The true Renosterveld of associations negative impact on the environment. 2 and 3, as defined by species group Q (Table 1), is, however, lacking from the Succulent Karoo table and belongs to the Fynbos Biome related vegetation. This Mountain Renosterveld ACKNOWLEDGEMENTS is probably distinct from other Renosterveld vegetation types The authors would like to thank the Critical Ecosystem in any case and could be studied in the future. Partnership Fund (CEPF) for funding the project by way of the SKEP (Succulent Karoo Ecosystem Plan/Program) initiative. Grazing and cropping are the main land-uses in the Mountain The Critical Ecosystem Partnership Fund is a joint initiative of Renosterveld. Sustainable land management tries to minimise Conservation International, the Global Environmental Facility, the risk of veld degradation or species extinction by managing the Government of Japan, the MacArthur Foundation and the populations of plants and animals within an area to ensure that World Bank. A fundamental goal is to ensure that civil society is they can continue to reproduce and function normally, even after engaged in biodiversity conservation. The assistance of Hennie stressful conditions such as drought (Esler et al. 2006). Although van den Berg of Iris International in compiling the vegetation damage can happen fast, recovery in the Karoo is very slow, map is gratefully acknowledged. CapeNature, the Department as it depends mainly upon unpredictable rainfall events (Esler of Tourism, Environment and Conservation (Northern Cape) et al. 2006). Sustainable farm management planning is critical and SANParks are thanked for the necessary permits and for ensuring a productive, profitable future in the region. permission to conduct this research. Inadequate farming practices, resulting from a severe lack of infrastructure, especially fencing, pose a serious threat to the REFERENCES vegetation. Farms in the region yield a low income as a result of the harsh environmental conditions and the unpalatable Acocks, J.P.H. 1988. Veld types of South Africa. 3rd ed. Memoirs grazing caused by the dominance of Dicerothamnus rhinocerotis. of the Botanical Survey of South Africa, 57: 1–146. Because of the low monetary value of the land and the high Agricultural Research Council 1986a. Land type map 3220 cost of infrastructure it is not financially viable for a farmer to Sutherland. Pretoria: Institute for Soil, Climate and Water. invest too much in infrastructure, as the ability to recover such Agricultural Research Council 1986b. Land type map 3018 costs is limited. Although the farmers are generally willing Loeriesfontein. Pretoria: Institute for Soil, Climate and to implement improved veld management and infrastructure Water. development, their financial means hinder their doing so. Agricultural Research Council 1995. Land type map 3118 Calvinia. Pretoria: Institute for Soil, Climate and Water. According to Low and Rebelo (1996) the degree of transformation Agricultural Research Council 1999a. Land type map 3120 in the Escarpment Mountain Renosterveld, which closely Williston. Pretoria: Institute for Soil, Climate and Water. Agricultural Research Council 1999b. Land type map 3218 corresponds to the Mountain Renosterveld as described in Clanwilliam. Pretoria: Institute for Soil, Climate and the current article, is unknown. However, many large tracts of Water. land cultivated in the past are still cultivated due to the higher Agricultural Research Council 2002. Land type map 3319 rainfall in the region compared with that experienced in the Worcester. Pretoria: Institute for Soil, Climate and Water. surrounding areas of the Hantam and Tanqua Karoo. Agricultural Research Council 2003. Land type map 3320 Ladismith. Pretoria: Institute for Soil, Climate and Water. Invasive species in the vegetation type described are CEPF, 2003. Ecosystem profile: The Succulent Karoo Hotspot, predominantly annuals that were brought into the region with Namibia and South Africa. Critical Ecosystem Partnership fodder from other parts of the world, and of which many have Fund report. been naturalised over the centuries. The isolated individuals of Conservation International: http://www.biodiversityhotspots. Prosopis species present are usually limited to highly disturbed org Accessed July 2006). areas alongside water points and feeding areas. The unpalatable ( Council for Geoscience 1973. Geological map 3218 Clanwillliam. renosterbos, Dicerothamnus rhinocerotis, which dominates Pretoria: Council for Geoscience. large sections of the Mountain Renosterveld is considered an Council for Geoscience 1983. Geological map 3220 Sutherland. encroacher by most farmers with its dominance being blamed Pretoria: Council for Geoscience. on centuries of incorrect management practices in the region. Council for Geoscience 1989. Geological map 3120 Williston. Also, overgrazing is thought to have substantially reduced the Pretoria: Council for Geoscience. grassy component in the vegetation. Council for Geoscience 1991. Geological map 3220 Ladismith. Pretoria: Council for Geoscience. The protected area network for the Mountain Renosterveld Council for Geoscience 1997. Geological map 3319 Worcester. is severely under-represented. Two local municipal nature Pretoria: Council for Geoscience. reserves, namely the Nieuwoudtville Wildflower Reserve Council for Geoscience 2001. Geological map 3118 Calvinia. (115 ha) and the Akkerendam Nature Reserve (230 ha), fall Pretoria: Council for Geoscience. within the region. A natural heritage site at Banksgate, near Du Plessis, H.M. 1987. Land Types of the maps 2816 Alexander Sutherland, protects the rare sterboom, Cliffortia arborea. The Bay, 2818 Warmbad, 2916 Springbok, 2918 Pofadder, 3017 Tanqua National Park has substantially expanded during the Garies, 3018 Loeriesfontein. Memoirs on the Agricultural last 3–5 years, with the latest land acquisitions including a Natural Resources of South Africa, 9: 1–538. section of Mountain Renosterveld vegetation. Esler, K.J., Milton, S.J. & Dean, W.R.J. 2006. Karoo veld ecology and management. Pretoria: Briza Publications. In conclusion, the aims of the project described in this article Germishuizen, G. & Meyer, N.L. (eds). 2003. Plants of southern were to classify and describe the various vegetation units present Africa: An annotated checklist. Strelitzia, 14. Pretoria: in the Mountain Renosterveld part of the Hantam-Tanqua- National Botanical Institute. African Protected Area Conservation and Science 70 KOEDOE Vol. 50 No. 1 pp. 61 - 71 http://www.koedoe.co.za Fynbos biome related vegetation Original Research

Hennekens, S.M. & Schaminée, J.H.J. 2001. TURBOVEG, a Myers, N., Mittermeir, R.A., Mittermeir, C.G., De Fonseca, comprehensive database management system for vegetation G.A.B. & Kent, J. 2000. Biodiversity hotspots for conservation data. Journal of Vegetation Science, 12: 589–591. priorities. Nature, 403: 853–858. Hill, M.O. 1979. TWINSPAN – A FORTRAN program for Nelder V.J., Wilson, B.A., Thompson, E.J. & Dillewaard, arranging multivariate data in an ordered two-way table H.A. 2005. Methodology for survey and mapping of regional by classification of the individuals and attributes. Ithaca: ecosystems and vegetation communities in Queensland. Version Ecology & Systematics, Cornell University. 3.1. Updated September 2005. Brisbane: Queensland Hilton-Taylor, C. 1994. Western Cape Domain (Succulent Karoo). Herbarium, Environmental Protection Agency. In Davis, S.D., Heywood, V.H. & Hamilton, A.C. (eds), Centres Rutherford, M.C. & Westfall, R.H. 1986. Biomes of Southern of plant diversity: A guide and strategy for their conservation, 1. Africa. An objective characterisation. Memoirs of the Botanical Cambridge: IUCN Publications Unit. pp. 201– 203. Survey of South Africa, 54: 1–98. Low, A.B. & Rebelo, A.G. 1998. Vegetation of South Africa, Lesotho Van der Merwe, H., Van Rooyen, M.W. & Van Rooyen, N. and Swaziland. Pretoria: Department of Environmental In press. Vegetation of the Hantam-Tanqua-Roggeveld Affairs and Tourism. subregion, South Africa. Part 2. Succulent Karoo Biome related vegetation. Koedoe. Marloth, R. 1908. Das Kapland, insonderheit das Reich der Van Wyk, A.E. & Smith, G.F . 2001. Regions of floristic endemism in Kapflora, das Waldgebiet und die Karoo, pflanzengeografisch southern Africa: A review with emphasis on succulents. Pretoria: dargestellt. Wiss. Ergebn. Deutsch. Tiefsee-Exped. Umdaus Press. ‘Waldivia’, 1898 – 1899. 2, T. 3, Jena: Fischer. Weather Bureau 1998. Climate of South Africa: Climate statistics up Mucina, L., Rutherford, M.C. & Powrie, L.W. (eds). 2005. to 1990. WB 42. Pretoria: Government Printer. Vegetation map of South Africa, Lesotho and Swaziland, Werger, M.J.A. 1974. On concepts and techniques applied in the 1 : 1 000 000 scale sheet maps. Pretoria: South African Zürich-Montpellier method of vegetation survey. Bothalia, Biodiversity Institute. 11(3): 309–323.

African Protected Area Conservation and Science and Conservation Area Protected African

http://www.koedoe.co.za Vol. 50 No. 1 pp. 61 - 71 KOEDOE 71

Chapter 5

Vegetation of the Hantam-Tanqua-Roggeveld subregion, South Africa. Part 2. Succulent Karoo Biome related vegetation

(VAN DER MERWE, H., VAN ROOYEN, M.W. AND VAN ROOYEN, N. 2008b. Vegetation of the Hantam-Tanqua-Roggeveld subregion, South Africa. Part 2. Succulent Karoo Biome related vegetation. Koedoe 50, 160-183.)

47 African Protected Area Conservation and Science 160 , ) was 2 KOEDOE yk yk and al Research W 000 km

Origin e g i o n 30 oggeveld as one of f the present paper r R the Western Mountain lies in the predominantlythe in lies 000 ha ( viz. s u b

he aim o nt of Agriculture, and throughAgriculture,and of nt T 000 . 2008) 2008) . ld ld subregion (see figure of

e g e t a t i o n cluster, predominantly Mountain omain, et al et v e l d D

epartme v oggeve . (2008). D STUDY AREA R he first T et al

anqua- hey identified the Hantam- T o g g e estern Cape T e l a t e d W oggeveld was also emphasised by Van

oria, Pretoria, 0002, South Africa roo, roo, where the Swartrug and Bontberg Mountains orthernCape r R antam region, the Tanqua and Ceres Karoo regions rthwards into the Tanqua Basin. The eastern border a -R

study a broad-scale vegetation survey of the entire rainfall region of the Northern and Western Cape econd vegetation cluster, i.e. the vegetation related to N culent Karoo Biome, and to create a map depicting the aken. Analysis of the data revealed two distinct uweveld uweveld Mountains to just southwest of Fraserburg, ces of South Africa. In the south it stretches from the es the Hantam, Roggeveld, Komsberg, Klein Roggeveld s to the region as identifiedby the farming community ent vegetation units. A brief discussion of the various N i c a y of Prety of enosterveld, is related to the Fynbos Biome, and is described he Hantam- oeriesfontein form the western boundary. r Karoo and Tanqua Karoo, area both covered of in this which article. occur The botanical within importance the of the undert provin Ceres Ka meet, no includ and while the Cederberg and Bokkeveld Mountains to just north of L The smaller study area discussed of in the this H paper covers parts within the Hantam- Smith (2001). theirprincipal 13 centres plant of endemism in southern Africa and also classifiedit partas of Succulentthe Karoo Biome. In this subregion of approximately 3 vegetation clusters. R in Van der Merwe is to present of a the s phytosociological analysis the Suc and description differ threat the and personal observations, is presented. T Merwe der Van insubregion winter f i o m e a n q u a

B A of Science Plant . 1 pp. 160 - 183 160 pp. . 1 -T South Africa e o o he he ABSTRACT BI), BI), T T r ama University of Pretoria N N a o u t h Department en Noel van Rooy ue ue to th e-mail: [email protected] S D Vol. 50 NoVol. K r Merwe Helga van de a n t a m Correspondence Helga van to: der Merwe . van Rooyen Margaretha W H and the conservation os related vegetation, or vegetation,related os 998). However, 998). theHowever, park l 1986) meet. l Botanicall Institute (now epartment Plant of Science, Universit t h e

eserve R estfal oggeveld subregion as an area ubin (1 ationa he Fynb he R u c c u l e n t W R o f T N

l Biodiversity Institute, SA wer wer l Biodiversity Institute, 2006). : Hantam, phytosociology, Roggeveld, Succulent Karoo, Tanqua Karoo, vegetation map, ) who argue for the recognition of a his article will focus on the Succulent anqua- T ord & egion. o n T ationa ationa ild Flo R 2: S N l Park by N Postal address: D W . (2007 t utherf Western Mountain Karoo The Hantam-Tanqua-Roggeveld subregion lies within the along Succulent the Karoo western side Hotspot of that the stretches Republicdocument the of botanical Southdiversity in Africathe Hantam-Tanqua-Roggeveld subregion, and was part of Namibia.a projectidentified This as a project,priority during carried the out SKEP (Succulent to KarooHotspot. Botanical Ecosystem surveys Programme) were conducted initiativein an area incovering over thisthree million hectares.images Satellite of the area and topocadastral, land into relatively homogeneous type units. An analysis and of the floristic geology dataof 390 sampleplots identified maps two were used major to floristic stratify units,i.e. theFynbos the Biome relatedvegetation area and theSucculent KarooBiome related vegetation. A description of the vegetation related to the Succulent Karoo Biome is presented in article.thisSeven associations, subassociations 16 and several mosaic vegetation units, consisting more of than one vegetation unit, were identifiedVarious and mapped. threats to vegetationthe in the region were identifiedduring the survey and are briefly discussed. Keywords r (2008). (2008). R et al ationa a N et al. P tam-Tanqua-Roggeveld tam-Tanqua-Roggeveld subregion occurs where three ether the vegetation, especially that of the Roggeveld, aylor aylor (1994) considered the subregion as part of the we we luded in Rubin’s original inluded Rubin’s study. e g e t a t i y Born as a study by Snijman and Perry (1987) on the floristicsof the on (1987) PerrySnijmanand by study a as lthough a phytosociological study was conducted in T ieuwoudtville ankwa V T N ithin this area the Succulent Karoo Ecosystem Plan (SKEP) nly a few studies have been conducted in the area, most of reater Cape FloristicCape reater he Succulent Karoo (CEPF 2003) stretches along the western area inc Hilton- SucculentKarooidentified He Biome. centrestwo of endemism Tanqua Tanqua Karoo has been less studied than the Nieuwoudtvillearea, a the subsequentlysubstantially extended boundariesits beyondthe farmingproject the of former the South African resulting in a host of research projects in area the(South AfricanNieuwoudtville O whichwere concentrated in the Nieuwoudtville area. Amongst thesew the fire-pronevegetation within subregion,the discussedis Van by der Mer Karoo related vegetation, which is not fire-prone. as to wh should be classifiedtransitional nature as of the Fynbos subregion isstudy b supported or in a Succulent recent G Karoo. The Han biomes, namely the Fynbos, Succulent KarooKaroo Biomes and ( the transitionalnature the of area there has been some controversy for which there is little information on the floral diversity, and consequently the Critical Ecosystem Partnership Fund (CEPF) funded botanical studies in the subregion. side Southof side Africa and Namibia and is one of only two global International(Conservation 2006). aridentirely are thathotspots W identified the Hantam- T http://www.koedoe.co.za

Original Research Van der Merwe, Van Rooyen & Van Rooyen

as well as the area east of the Roggeveld Mountains. This area Schaminée 2001). TURBOVEG software was used to capture includes two of Acocks’s veld types (Acocks 1953, 1988), namely: the vegetation data and a TW INSPAN (Two Way Species Succulent Karoo (Veld Type 31) and Western Mountain Karoo Indicator Analysis (Hill 1979)) was run in MEGATAB as a first (Veld Type 28). These two veld types of Acocks are equivalent to step to the classification of the vegetation data.T he result of the the Lowland Succulent Karoo (Unit 57) and Upland Succulent TW INSPAN (Two Way Species Indicator Analysis (Hill 1979)) Karoo (Unit 56) of Low and Rebelo (1998), respectively. Mucina on the entire data set of 390 relevés confirmed the presence of et al. (2005) recognised the following large vegetation types two distinct floristic groups, which enabled the data set to be in the study area: Hantam Karoo (SKt 2), Roggeveld Karoo split into two, and a TW INSPAN was then run on each data (SKt 3), Tanqua Escarpment Shrubland (SKv 4), Tanqua Karoo set separately. The resulting differential tables were further (Skv 5), Tanqua Wash Riviere (AZi 7) and Namaqualand Riviere refined using Braun-Blanquet procedures. The first data set (AZi 1). Smaller patches of Nieuwoudtville Roggeveld Dolerite (107 relevés) characterised the vegetation of the predominantly Renosterveld (FRd 1), Hantam Plateau Dolerite Renosterveld Mountain Renosterveld with Fynbos affinities, and is described (FRd 2), Nieuwoudtville Shale Renosterveld (FRs 2) and by Van der Merwe et al. (2008). The second data set (283 relevés) Bushmanland Vloere (AZi 5) are also mapped in the area. characterised the vegetation with predominantly Succulent Karoo affinities and is discussed further in this article. The physical geography of the region differs greatly. From the level plains of the Tanqua Karoo the landscape in the east The basic unit used here in the classification of the vegetation rises steeply up the escarpment to the plateau formed by the is the association. An association, as defined by Nelder et al. Roggeveld, Komsberg, Koedoesberg and Nuweveld Mountains. (2005), describes vegetation units on the basis of the presence The Hantam is characterised by a gently undulating to a steeply and abundance of species, vegetation structure and spatial rolling topography. distribution of individuals in the dominant layer. Species within a subdominant structural layer and their canopy cover Geologically, the Beaufort and Ecca Groups dominate the study are used to describe the subassociations. area (Rubidge & Hancox 1999, Council for Geoscience 1973, 1983, 1989, 1991, 1997, 2001, 2008). The Ecca Group covers most of the Associations and subassociations were mapped using the study area, with the Dwyka Group (tillite, sandstone, mudstone botanical survey data as well as 1:250 000 topocadastral maps, and shale) cropping out in the west. The Ecca Group includes land type maps, geology maps (and electronic information sediments of the Tierberg (shale), Prince Albert (mudrock), supplied by the Council of Geoscience, 2008), and satellite Kookfontein (shale, siltstone and stone) and Skoorsteenberg images. Several mosaic units, which include more than one Formations (mudstone, siltstone and sandstone). The mudstones association or subassociation due to the transitional nature of the Abrahamskraal Formation in the Beaufort Group are of the area, were also identified and mapped. (An electronic found on the eastern side of the study area, while igneous ArcView compatible version (raster or vector format) of the intrusions of dolerite occur throughout the region. An array vegetation map is available from the authors on request. The of land types is represented in the study area (Agricultural pdf version provided here should best be printed to A3.) Research Council 1986a, 1986b, 1995, 1999a, 1999b, 2002, 2003). The prominent land types include Ag, Da, Dc, Fb, Fc, Ia and Ib Herbarium specimens were sent to Compton Herbarium, (Du Plessis 1987). Kirstenbosch, for identification if identification of species in the field was in doubt.T he collection code (HR) and numbers of the Rainfall ranges from 50 to 300 mm a year, with a mean of specimens were kept throughout the process. All the species, 228 mm, as measured at Calvinia (SA Weather Bureau 1998). especially within the Aizoaceae (Mesembryanthemaceae), have Although the rain falls mainly in winter it does include a few not yet been positively identified. The H.G.W.J. Schweickerdt Herbarium (PRU), University of Pretoria, houses the summer thunderstorms. The highest annual rainfall, measured voucher specimens collected. Nomenclature follows that of over a 29-year period, at Calvinia, was 472 mm in 1976 Germishuizen and Meyer (2003). (SA Weather Bureau 1998). January and February have a mean maximum temperature of 30.8°C while an extreme maximum of 41.2°C was recorded in February 1990. The coldest months RESULTS are June and July with a mean maximum of 17.1°C and 17.2°C respectively. The lowest temperature recorded was –6.5°C in The phytosociological analysis of the floristic data is given in June 1978 (SA Weather Bureau 1998). Tables 1a, 1b and 1c. In order to prevent confusion and to aid in the compilation of a vegetation map of the entire subregion the plant associations have been numbered sequentially following METHODS AND MATERIALS the three associations (1–3) described in Part 1 of this series (Van der Merwe et al. 2008). Seven associations were identified On the basis of the colour, texture and topography, satellite in the area under discussion and these were subdivided into 16 images (Bands: 4,5,3 (R,G,B)) of the study area were visually subassociations, as set out in the following scheme: stratified into relatively homogeneous units. Floristic surveys 4. Pteronia glauca – Euphorbia decussata Escarpment Karoo were carried out during the spring of 2004 at 390 sites selected 4.1 Montina caryophyllacea – Pteronia glauca Roggeveld in the stratified homogeneous units close to any national or Escarpment Karoo provincial road or farm track. Plot sizes 10 x 10 m were used 4.2 Galenia africana – Pteronia glauca Escarpment Karoo in most of the surveys but larger plots (20 x 20 m) were used in 4.3 Euphorbia decussata – Ruschia cradockensis Escarpment more denuded areas (Rubin 1998). Global Positioning System Karoo (GPS) coordinates were taken at each site. A cover-abundance 5. Eriocephalus purpureus Hantam Karoo value, according to the Braun-Blanquet scale (Werger 1974), was 5.1 Erodium cicutarium – Eriocephalus purpureus Hantam noted for each species present in a plot. Aspect, slope, position Karoo in the landscape, soil type and colour, an estimation of rock 5.2 Ruschia cradockensis – Eriocephalus purpureus Hantam cover, rock size and erosion were noted at each sampling point. Karoo Rock size was categorised into gravel (< 10 mm), small stones 5.3 Leipoldtia schultzei – Eriocephalus purpureus Hantam (10–50 mm), stones (> 50–200 mm) and boulders (> 200 mm). Karoo Trampling, small mammal activity and invasion by alien plants 6. Pteronia glomerata Roggeveld Karoo were some of the biotic effects also recorded. 6.1 Phyllobolus tenuiflorus – Pteronia glomerata Roggeveld Karoo Using the TURBOVEG and MEGATAB computer package an 6.2 Eriocephalus pauperrimus – Pteronia glomerata analysis of the floristic data was conducted (Hennekens & Roggeveld Karoo African Protected Area Conservation and Science 161 KOEDOE Vol. 50 No. 1 pp. 160 - 183 http://www.koedoe.co.za

Succulent Karoo Biome related vegetation – Part 2 Original Research

6.3 Pentzia incana – Pteronia glomerata Roggeveld Karoo its tributaries, the Biedouw Valley and the western extreme of 6.4 Ruschia intricata – Pteronia glomerata Roggeveld the brackish system of the Hantam River in the vicinity of Van Karoo Rhynsdorp. The first unit is characterised by a strong presence 6.5 Eriocephalus ericoides – Pteronia glomerata Roggeveld of species group A, and a weak presence of species group B, Karoo while the second unit comprises species group B. At present it is 7. Aridaria noctiflora Tanqua and Loeriesfontein Karoo not clear whether the unmapped units form part of association 4 7.1 Ruschia robusta – Aridaria noctiflora Tanqua Karoo or whether they have closer affinities to associations outside the 7.2 Drosanthemum (HR217) sp. – Aridaria noctiflora study area. Tanqua Karoo 7.3 Malephora crassa – Aridaria noctiflora Tanqua Karoo 4. Pteronia glauca – Euphorbia decussata Escarpment Karoo 7.4 Atriplex lindleyi – Aridaria noctiflora Loeriesfontein This plant association is located along the Roggeveld Karoo Escarpment, the slopes of the Hantam Mountains, on the slopes 7.5 Ruschia intricata – Aridaria noctiflora Tanqua Karoo alongside the Hantam River in the Agter-Hantam and to the 8. Stipagrostis obtusa Central Tanqua Grassy Plains west of the Stinkfontein, Lang and Boegoeberg Mountains, and 9. Mesembryanthemaceae (HRp359) sp. Ceres Karoo has links to associations along the western part of the Doorn Vygieveld River and its tributaries (Fig. 1). Geologically, the Ecca Group 10. Pteronia (HRp118) sp. Tanqua Brackish Flats dominates this association while tillite and dolerite intrusions are present at times. A variety of land types are found in this Two vegetation units falling outside the present study area have plant association, including Fc, Fb, Ia, Ib, Da and Ag. The been included in Table 1a. With the exception of these units and association generally occurs in lower-lying areas on level associations 9 and 10 all vegetation units could be mapped terrain, gentle or moderate to steep slopes. The percentage rock (Fig. 1). Ten mosaics found in areas where the landscape is cover varies tremendously (0–99%) and includes gravel, small spatially diverse were also mapped. The vegetation units stones, stones and boulders. Shrub cover is usually high, while making up the mosaics and the most important environmental grass and annual cover is limited. features distinguishing the units are set out below: • the Nieuwoudtville mosaic consists of vegetation Prominent species included in this plant association are Pteronia units 2.1.1, 2.1.4, 2.2 of the Mountain Renosterveld glauca (species group C, Table 1a), Euphorbia decussata (species vegetation (Van der Merwe et al. 2008.) and 5.1 group D), Pentzia incana and Ruschia cradockensis (species predominantly found on the dolerite outcrops; group n), as well as Tripteris sinuata, Drosanthemum (HR217) • the Grootfontein mosaic consists of vegetation units 4.2, sp. and Galenia africana (species group V). Plant associations 4, 4.3 on undulating ridges and 5.1 and 5.2 on dolerite 5 and 6 are closely related because they share the common outcrops and dolerite derived soils; species in species group N, yet associations 4 and 5 (Table 1a) • the Houhoek mosaic consists of vegetation unit 4.1 at a differ from association 6 (Table 1b) since both associations 4 and higher altitude and higher rock cover and unit 4.3 at a 5 lack species group U. lower altitude and lower rock cover; • the Adamsfontein/Noodsaaklikheid mosaic with 4.1. Montina caryophyllacea – Pteronia glauca Roggeveld vegetation unit 5.2 predominantly on dolerites and Escarpment Karoo 6.5 on shales and dolerite intrusions that have been disturbed; This subassociation characterises the vegetation of the • the Calvinia mosaic consists of vegetation units 6.5 on Roggeveld Escarpment, i.e. the west-facing slopes of the Ecca shales and dolerite intrusions and 7.4 on brackish Roggeveld Mountains (Fig. 1), and also forms part of the soils (historically overutilised); Houhoek, Kalkgat as well as Windheuwel/Rooiheuwel mosaics. • the Kalkgat mosaic consists of vegetation units 4.1 on Subassociation 4.1, excluding the various mosaic vegetation rocky ridges, 7.2 on shales and sedimentary deposits units, covers an area of 152 057 ha (9.2% of the total area in and 7.3 on brackish soils; Fig. 1). The subassociation occurs at intermediate altitudes • the Tanqua Karoo Inselberg mosaic with vegetation of 700 to 1 100 m above sea level, on gentle to moderate, and unit 5.2 on dolerites and 7.5 on Ecca shales; sometimes steep, slopes. A high rock cover of generally more • the Tanqua pan mosaics consist of vegetation units 7.2 than 90%, comprising small stones, stones and boulders, is on shales and alluvial deposits and 10 on brackish found in this subassociation. The land type is predominantly soils; Fc. The light brown to brown coloured loamy soils are derived • the Windheuwel/Rooiheuwel mosaic consists of from Ecca shales. vegetation units 4.1, 4.2 on the rocky ridges and 7.3 on the brackish plains; The vegetation is characterised by a high shrub cover (50–90%) • the Naresie mosaic consists of elements of the and grasses are usually absent. A very low cover of annuals can Science and Conservation Area Protected African Namaqualand Brokenveld as defined by Acocks (1953, be found in certain areas. Subassociation 4.1 comprises species group B and there is a strong presence of the diagnostic species 1988) and vegetation unit 6.4. group C (Table 1a). The vegetation is dominated by Pteronia glauca (species group C) that is constantly present and often The year in which the surveys were conducted (2004) was a poor has a very high canopy cover. Other prominent species present rainfall year, and would have substantially reduced the annual include Tylecodon wallichii and Montinia caryophyllacea (species and geophyte components of the vegetation. The resulting group B). phytosociological tables thus poorly represent these two components. In normal or good rainfall years the contribution 4.2. Galenia africana – Pteronia glauca Escarpment Karoo of the annuals and geophytes to the vegetation cover would be Plant subassociation 4.2 is located on the slopes of the Hantam considerably higher. Mountain, the undulating slopes of the escarpment in the Platberg and surrounding area southwest of Calvinia, and the Description of plant associations (Tables 1a, 1b, 1c; Fig. 1) slopes where the Roggeveld and Klein Roggeveld Mountains Two vegetation units, included in Table 1a, have not been meet (Fig. 1). Subassociation 4.2, excluding the mosaic vegetation mapped as they fall outside the present study area and their units, covers an area of 45 047 ha (2.7% of the total area in linkages with the vegetation in the Van Rhynsdorp region still Fig. 1). It is also found in the mosaic between the Roggeveld requires further study. These two units are related to what and Koedoesberg Mountains in the vicinity of the farms Acocks (1953, 1988) described as Succulent Karoo in the Van W indheuwel and Rooiheuwel. Elements of this subassociation Rhynsdorp region, and are found along the Doorn River and are found on the ridges between the dolerite derived soils in the http://www.koedoe.co.za Vol. 50 No. 1 pp. 160 - 183 KOEDOE 162

Original Research Van der Merwe, Van Rooyen & Van Rooyen

Figure 1 Vegetation map of the Succulent Karoo Biome related vegetation of the Hantam-Tanqua-Roggeveld subregion African Protected Area Conservation and Science 163 KOEDOE Vol. 50 No. 1 pp. 160 - 183 http://www.koedoe.co.za

Succulent Karoo Biome related vegetation – Part 2 Original Research

Grootfontein mosaic in the Hantam along the R27 national road, to the common species in species group N. This association is east of Nieuwoudtville. Land types are numerous, and include subdivided into three subassociations. Fb, Ia, Ib and Da, at altitudes ranging from 700 to 1 200 m above sea level. Ecca shales and dolerite intrusions predominate in 5.1. Erodium cicutarium – Eriocephalus purpureus Hantam this vegetation unit on gently sloping terrain. A low rock cover Karoo (< 10%) or a high rock cover (> 80%), consisting of gravel, small The two variations of this subassociation (Table 1a) are mainly stones, stones and boulders covers the light brown or brown due to the presence or absence of annuals and/or geophytes in loamy soils. species groups E and F, which could be as a result of the drought year in which the surveys were conducted. These life forms are Shrub cover in this subassociation is high (> 70%), while the rainfall dependent and thus by conducting the same surveys grass and annual components are not well represented. Pteronia in a good rainfall year it might not be possible to distinguish glauca (species group C) is diagnostic of this subassociation (Table between the two variations. 1a). Other species present include Pentzia incana, Eriocephalus ericoides (species group N) and Tripteris sinuata, Drosanthemum This subassociation occurs in the vicinity of the farm (HR217) sp. and Galenia africana (species group V). Matjiesfontein, close to Calvinia, southwards to the farm Klipbak and northwards to Loeriesfontein (Fig. 1). Subassociation 5.1, 4.3. Euphorbia decussata – Ruschia cradockensis Escarpment excluding the mosaic vegetation units, covers an area of Karoo 92 682 ha (5.6% of the total area in Fig. 1). It also occurs as part of the Grootfontein mosaic, as well as westwards to the This small subassociation occurs at lower altitudes of 400 to Nieuwoudtville mosaic where it occurs in combination with 1 200 m above sea level, to the west of the Hantam Mountain and, variants 2.1.1, 2.1.4 and subassociation 2.2 (Van der Merwe et al. excluding the mosaic vegetation units, covers an area of 5 196 ha 2008). The subassociation is found on the Ecca Group where (0.3% of the total area in Fig. 1). Elements of this subassociation dolerite intrusions occur in the landscape, on level to gently can also be found closer to Nieuwoudtville in the Grootfontein sloping areas, at an altitude from 700 to 1 000 m above sea level mosaic as well as in the Agter-Hantam in the Houhoek mosaic on Land Types Fb, Ea, Dc or Da. Soils are clays or loams of red where it covers the lower slopes. The Ecca Group dominates brown or dark brown colour, with little or no rocks. in this subassociation, while tillite and dolerite intrusions are present in the mosaics. Land types include Fb, Fc, Da, Ag Shrub cover is highly variable (< 5–75%) and the contribution and Ib, resulting in the mosaic vegetation. The level, gently or of the grass component is very small. Annual cover varies from moderately sloping ridges are usually covered with either very < 1% to 95%, indicating its high variability. Subassociation 5.1 little rock (0–5%) or much rock (60–99%). Small stones, stones differs from 5.2 by the presence of species groups E, F and G, and boulders are present in the light brown to brown loamy which occur only in subassociation 5.1 (Table 1a). Common soils. species include Erodium cicutarium (species group G), Ehrharta calycina (species group H), Eriocephalus purpureus and Tetragonia Shrub cover is high (45–95%), whereas the grass and annual microptera (species group I), Pentzia incana, Ruschia cradockensis components are generally weakly developed. The dominant and Asparagus capensis (species group N). species are Euphorbia decussata (species group D, Table 1a) and Ruschia cradockensis (species group N), and both these 5.2 Ruschia cradockensis – Eriocephalus purpureus Hantam species generally have a high cover. Other species present Karoo include Tetragonia fruticosa (species group N), Tripteris sinuata, Drosanthemum (HR217) sp. and Galenia africana (species Subassociation 5.2 occurs as a clearly defined vegetation unit group V). as well as in various mosaic vegetation units. As a clearly defined unit it occurs in the Agter-Hantam near Klipwerf 5. Eriocephalus purpureus Hantam Karoo (Fig. 1) and, excluding the mosaic vegetation units, covers an area of 22 353 ha (1.3% of the total area in Fig. 1). Here the Association 5 occurs in the vicinity of the farm Matjiesfontein, subassociation is found on the Ecca Group while dolerite close to Calvinia, northwards to Loeriesfontein and intrusions are scattered throughout the landscape. As part southwards to the farm Klipbak. It also occurs as part of the of a mosaic, it occurs north and south of the Hantam River in Grootfontein mosaic, westwards along the R27 national road the Adamsfontein/Noodsaaklikheid mosaic. It also occurs on to the Nieuwoudtville mosaic, as described by Van der Merwe dolerites on the inselbergs south of Calvinia into the Tanqua et al. (2008). Additionally, it is found in the Agter-Hantam Karoo (Tanqua Karoo Inselbergs mosaic). These inselbergs (Adamsfontein/Noodsaaklikheid mosaic) and on small include the Nuwewater se berg, Elandsberg, Eselberg, Leeuberg, inselbergs into the Tanqua Karoo, such as the Nuwewater Potkleiberg and Sterretjieberg. Additionally, subassociation 5.2 se berg, Elandsberg, Eselberg, Leeuberg, Potkleiberg and occurs in the Grootfontein mosaic vegetation unit on the African Protected Area Conservation and Science and Conservation Area Protected African Sterretjieberg (Tanqua Karoo Inselbergs mosaic) (Fig. 1). The dolerite derived soils. This subassociation generally occurs on altitude is lower than 1 000 m and an array of land types occur, Land Types Fb and Fc at altitudes from 700 to 1 000 m above sea including Dc, Fc, Da and Ag. The level to gently sloping ridges level. The rock cover varies considerably from zero to 90% and usually have a low rock cover. The brown, red brown or light includes small stones, stones and boulders. The underlying red brown loams and clays are derived from Ecca shales and Dwyka brown or brown soils are clays and loams, primarily derived tillite, as well as dolerite intrusions that scatter the landscape. from the dolerites and the Ecca shales.

The shrub cover varies considerably (0–80%) and grasses are Shrub cover ranges from 40 to 90%, while the grass and annual generally absent. The cover of the annual component ranges components are usually absent or < 5%. Ehrharta calycina (species from very little to 95%. It is assumed that the annual component group H), Eriocephalus purpureus (species group I), Pentzia incana contributes substantially to the total aboveground biomass in and Ruschia cradockensis (species group N), as well as Aridaria good rainfall years since the area is renowned for its annual noctiflora (species group V) are prominent in this subassociation and geophyte spring displays. Eriocephalus purpureus (species (Table 1a). group I) and Leipoldtia schultzei (species group J) characterise the vegetation of this association (Table 1a). Other prominent Subassociation 5.2 is closely related to subassociation 5.1 as species include Pentzia incana, Ruschia cradockensis, Asparagus they both include species groups H, I, J, N and V, but it differs capensis, Tetragonia fruticosa and Eriocephalus microphyllus from from the latter by the absence of species groups E, F and G. species group N, as well as Aridaria noctiflora and Galenia Subassociation 5.2 is also closely related to subassociation 5.3, africana from species group V. Plant association 5 is closely from which it only differs by the absence of species group H in related to both associations 4 (Table 1a) and 6 (Table 1b) due the latter. http://www.koedoe.co.za Vol. 50 No. 1 pp. 160 - 183 KOEDOE 164

Original Research Van der Merwe, Van Rooyen & Van Rooyen

. . . . 9 ...... 8 ...... 5 ...... 3 ...... 2 ...... 6 ...... 1 ...... 2 ...... 5 ...... r . . . 1 ...... 3 ...... 8 ...... 7 ...... 1 ...... 5 ...... 8 ...... 5 ...... 5 ...... 5.3 ...... 2 ...... 8 ...... 2 ...... 7 ...... 2 ...... 6 ...... 1 ...... 6 ...... 5 ...... 1 ...... 4 ...... 9 + ...... + ...... 4 ...... 8 ...... 4 ...... 5 ...... 1 . + ...... 9 ...... 1 ...... 6 ...... 9 ...... 6 ...... 8 ...... 6 ...... 5 ...... 1 ...... 1 ...... 5.2 ...... 1 ...... 7 ...... + . . . 9 ...... + . . . . . 2 ...... 5 ...... 1 ...... 1 ...... 3 ...... 3 ...... 3 . r ...... + 2 ...... 2 ...... 3 ...... + ...... 1 ...... 3 ...... 1 ...... 2 ...... + ...... 1 ...... 0 ...... 3 . . 5.1 ...... r . 2 ...... + 0 ...... 3 ...... r . 0 ...... 2 ...... 1 . . . . . + . . + . . . . . r ...... + ...... + . 8 . . . . . + . . + ...... + . . . . 6 . . + . . + ...... + . . . . . + ...... + . 4 . . + ...... + . . . . r . + . . . + . . + + . . + . 3 ...... + . . + ...... 1 ...... 1 . . 8 ...... + ...... 9 ...... r + . . 0 ...... 9 ...... + . . 0 . . . . 3 ...... + . . 5 ...... 1 . . 0 . . . . 2 ...... 1 ...... + . . 3 . . . . 1 . . . + ...... 4 ...... + . . 0 ...... + ...... 0 ...... 1 . . 3 . . . 1 ...... 1 . . . 8 . + ...... 1 . . 2 . . . 3 ...... 4.3 ...... 6 ...... + . . 1 . . . 3 ...... + . . . 4 . + ...... + . . 7 . . . 2 ...... 0 ...... + 1 . . 8 . . . 2 ...... 6 ...... + 1 . . 1 . . . 1 ...... 5 ...... 1 . . 7 . . . . . + . . . . a ...... 4 ...... 1 . . 6 ...... 1 ...... + . . 5 ...... ble 1 ...... 1 ...... 1 . . 4 ...... a T ...... 2 ...... + . . . 2 . . . 2 ...... 3 ...... 1 . . . 4 . . . 2 ...... 8 ...... 1 . . . 6 ...... 5 ...... + . . . 4 ...... 0 ...... + . . . 7 . . 2 ...... 8 ...... + . . . 9 ...... 5 ...... b . . . 2 . . . . . 4.2 ...... 5 ...... 4 . . . 0 ...... 2 ...... + . . . 1 2 ...... 0 . . + . . . . . 1 . . . 5 2 ...... 8 ...... + . . . 9 1 ...... 2 ...... 1 . . . 9 1 ...... 1 ...... 1 . . . 9 ...... 7 ...... + . + . . . 2 3 ...... + ...... + . . . 4 ...... + . + . . . 9 3 ...... + ...... 4 ...... 1 . . . 7 3 ...... + . . . . 8 . . + . . + + . a . . . 1 1 . + . 4.1 . . . + ...... 3 ...... 1 . . . 3 1 ...... 8 ...... + . 1 . . . 8 . b . . . . + ...... r . . . . 8 ...... + . 1 . . . 7 r ...... 7 . . r ...... 9 3 ...... 5 . . + ...... 1 2 ...... + . . . . 1 ...... 5 2 ...... 5 . . r . . . . . r . . . 6 ...... 4 . + . . . . + . . . . . 1 ...... + ...... 3 ...... + . . + . . 9 . . 1 . . . . + ...... + . . . . 3 ...... 6 Phytosociological table the of Succulent Karoo Biome related vegetation the of Hantam-Tanqua-Roggeveld subregion – associations 4 and 5 ...... 8 + . + . . . + 3 . . . . . 0 ...... + ...... 9 ...... 1 . . . . + 7 Not mapped ...... + ...... 9 ...... 1 . . . . . 1 ...... + ...... 4 ...... 2 . . . . . 5 . 1 ...... r . . . . 6 . . . . . + + 1 . . . . . 5 . 2 1 ...... r . . . . 6 ...... 1 ...... 7 b ...... r . . . . 2 ...... 1 ...... 4 . . . . + ...... + + . . . . . 8 + ...... 3 . . r . . . 1 . . + ...... 8 . . . . + + . 3 . . 1 . . + 7 ...... + . 7 + . . . . + . 3 . . . . . + 3 a . . . + . . . . + ...... 4 . . . . . + ...... 3 . . + . 1 ...... 1 + . . . . . 8 . . . . . 3 ...... Not mapped . 5 + . . . + + . . . + . . + + . . . . 3 . . r . . + ...... sp. sp. sp. (HRp378) sp. spp. (HRp35) sp. sp. s group A s group B s group C s group D s group E s group F sp. (HRp35) sp. elevé number Didelta spinosa Montinia caryophyllacea Hermannia cuneifolia Pteronia (HRp382) sp. Euphorbia Crassula alpestris Lachenalia Sarcocaulon Babiana Heliophila collina Ruschia Lachenalia violacea Vygie (HRp384) sp. Stapelia Rhus burchellii Antimima cf. granitica (HR248) Thesium cf. hystrix Vygie (HRp378) sp. Othonna filicaulis Asparagus Euphorbia hamata Phyllobolus Trachyandra falcata Vygie (HRp382) sp. R Specie Zygophyllum foetidum Specie Tylecodon wallichii Specie Lapeirousia montana Pteronia glauca Specie Euphorbia decussata Specie Arctotis acaulis Specie Dorotheanthus African Protected Area Conservation and Science 165 KOEDOE Vol. 50 No. 1 pp. 160 - 183 http://www.koedoe.co.za

Succulent Karoo Biome related vegetation – Part 2 Original Research

. . + . 9 ...... 8 . + . . . . . + ...... 5 ...... 3 ...... 1 ...... 2 . . . . a . . 6 ...... 1 ...... + 2 . . . . 1 . . 5 ...... 1 ...... 1 . 3 + ...... 8 ...... 7 . . . a + . . 1 + ...... + ...... + ...... 5 ...... 8 ...... + ...... + ...... + . 5 . . . . . + . 5 ...... + + ...... 5.3 . . . . . 2 ...... 8 . . + . . . . + ...... + . . + . . . . . 2 ...... 7 . . + . . . . . + ...... + ...... 2 . . . . + + . 6 ...... + + ...... 1 ...... 6 ...... + + ...... + . . . . . + . 5 . . . 1 1 . . 1 ...... 4 . . . . + . . 9 ...... + + ...... 4 . . . . + . . 8 . . + ...... + . 4 + ...... 5 ...... + ...... + ...... 1 ...... 9 ...... 1 . . . + + + . . . . . + . 6 ...... 9 ...... + + ...... 6 ...... 8 ...... + + . . + . . . . . + . . . . . r . . 6 ...... 5 ...... r ...... + + . 1 ...... 1 ...... + + . . . . 5.2 ...... a 3 . . 1 ...... r + ...... + . . . . 7 ...... + ...... + + . . . 9 ...... + ...... + a r . . 2 . . + . . . + . . . . . + ...... 5 . . + ...... + . 1 . . . . . + . . . + . 1 + + . . + . . + + 3 . . . . . + . 1 . . . 3 + . . . + + . . + . . . . 1 . . . . . + 3 . . . . + + . + . . . 2 r ...... + . . . . 1 . . + . . . + 2 ...... 3 . . . . + . . + . + . + + ...... 1 . . . . . + + + . . . 3 . + . . . . . + + . . . + ...... 1 . + . . . . + . . + . 2 ...... + . . . + ...... 1 + ...... + . 0 ...... b 5.1 . . . + . . . . 2 + ...... a . . . 0 ...... + + . . . . + . . + . . . . 3 . + . . + ...... 0 . . . . + . . + . . . . + . . + . . . . 2 . . + ...... + . 1 . . . . + + . + + + . . + . + ...... + . . . + . . . . . 8 ...... + + . . + ...... + . . . . 6 ...... + + . . . . . + ...... + . . . . + . . . . 4 . . . . + . + + + . . . . . + + ...... + + + . . . 3 . . + . + . . + + . . . . . + . . . . . 1 ...... + . . 8 ...... + ...... r . . . . . 9 ...... 0 ...... 9 ...... 0 . . . . 3 ...... 5 . + ...... 0 . . . . 2 ...... 1 ...... 3 . . . . 1 ...... 4 ...... 0 ...... 0 ...... 3 . + . . 1 . . . + ...... 8 ...... 2 . . . . 3 ...... 4.3 ...... 6 ...... 1 . . . . 3 ...... 4 ...... 7 . . . . 2 ...... 0 ...... 8 . . . . 2 ...... ) t ...... 6 ...... + . . 1 . . . . 1 ...... + . . . 5 . . + . . . . . + . . 7 ...... 4 ...... 1 . . 6 ...... + ...... + . . . 1 . . + ...... 5 ...... + . . a (con . . . . . 1 . . . 1 ...... + . + . . 4 ...... 2 ...... 2 . . . . 2 ...... ble 1 ...... 3 ...... 1 + . . 4 . . . . 2 . . . . . + . . . . + ...... 8 ...... 6 ...... ta . . + ...... 5 . . . . + ...... 4 ...... 0 ...... 7 . 2 ...... 8 ...... 9 ...... 5 . . . . + ...... 2 . . . . 4.2 . + ...... + . . . . 5 ...... 0 ...... 2 ...... + . 1 2 ...... + . . . 0 ...... 5 2 ...... 8 ...... 9 1 ...... 2 ...... 9 1 . . . . + ...... 1 ...... + + . . 9 ...... 7 ...... 2 3 ...... 4 ...... 9 3 ...... 4 ...... 7 3 ...... + . . . 8 ...... 1 1 . . . . 4.1 ...... 3 ...... 3 1 ...... 8 ...... 8 ...... 8 . . . . + ...... 7 ...... + ...... 7 . . . . + ...... 9 3 ...... 5 . . + ...... 1 2 ...... 1 ...... a . . 5 2 . . . . African Protected Area Conservation and Science and Conservation Area Protected African ...... + . . . 5 ...... 6 ...... + ...... 4 . . . . + ...... 1 ...... 3 ...... 1 . . 9 ...... r . . . 3 ...... + . . 6 ...... + + ...... + . . . 8 . . . . + . . 3 . . . . 0 . . . + . + ...... 9 ...... 1 . . . . 7 Not mapped ...... 9 ...... 1 . . . . 1 ...... 4 ...... 2 . . . . 5 ...... + . . 6 ...... 1 . . . . 5 . 2 ...... + . . 6 ...... 1 . . . . . 7 ...... 2 ...... 1 . . . . . 4 ...... + ...... 8 ...... 3 . . . . . 1 ...... 8 ...... 3 . . . . . 7 ...... 1 . . . 7 ...... 3 . . . . . 3 ...... + ...... + + . . . 4 ...... + . . . 3 ...... 8 ...... 3 . . . . . Not mapped . 5 . . . . . + + ...... + . + . . . 3 ...... 1 . .

sp. sp. sp. s group G s group H s group I s group J s group K s group L elevé number Zygophyllum pygmaeum Bulbine succulenta Lotononis Helichrysum obtusum Dimorphotheca sinuata Cyphia digitata Tetragonia microptera Gorteria diffusa Bromus pectinatus Rosenia glandulosa Senecio cakilefolius Lotononis hirsuta Erodium cicutarium Merxmuellera stricta Microloma sagittatum Senecio cardaminifolius Euphorbia mauritanica Oncosiphon grandiflorum Othonna auriculifolia Osteospermum Osteospermum acanthospermum Moraea Berkheya fruticosa Cotula barbata Eriocephalus namaquensis Drosanthemum cf. latipetalum R Specie Bulbinella Specie Ehrharta calycina Specie Eriocephalus purpureus Specie Leipoldtia schultzei Specie Phyllobolus tenuiflorus Specie Pentaschistis patula http://www.koedoe.co.za Vol. 50 No. 1 pp. 160 - 183 KOEDOE 166

Original Research Van der Merwe, Van Rooyen & Van Rooyen

. . . a . 9 + ...... 8 . 1 . . . + ...... + . . . . . + ...... 5 ...... 3 ...... 1 . . . . . + . . r . . . . 2 ...... 6 ...... a . . . . . 1 ...... 2 ...... + 5 ...... + . . . . . 1 . . . . + ...... 3 + . . . . + . 8 ...... + ...... + . 7 . . . + . + . 1 . + ...... 5 + ...... 8 . 1 . . . . . + . . . . + ...... 5 + . . + . . . 5 ...... 5.3 ...... 2 . . . + . . . 8 . + . . + + . + ...... 2 + ...... 7 . . + ...... r . . . . . + . . . . 2 ...... 6 . . + . . . . 1 ...... 1 . . . + . . . 6 ...... + . . . . 5 + . . + . . . 1 . . . . . + ...... + ...... 4 ...... 9 ...... + ...... + ...... 4 . . + . . . . 8 ...... + . + ...... + + . . . 4 ...... 5 ...... + ...... 1 ...... + 9 ...... 1 ...... r . . + . 6 + ...... 9 . 1 . . . + ...... + . 6 + . . + . . . 8 . + . . . + . . . . . + + ...... 6 ...... 5 . . . . + r + 1 . . . r ...... + ...... 1 + . . + . . . 1 ...... 5.2 . . . . . + + ...... + . . . 1 . + . . . r . + . . . . . + ...... + . + . . . + . . . . . 7 . + ...... + + . . + ...... + ...... + . . . 9 ...... + . . . . . + ...... + . . . . + . . . 2 . . . . . + . 3 ...... 5 . . . . . + ...... 1 . . 1 ...... 1 . + ...... 3 ...... + . . . 3 . + . . . + . + . . . . + ...... 3 + ...... 2 . . . . . + ...... 2 . . . . . + ...... 3 ...... 1 . . . + . + . . + . . . 3 . . . . . + . + ...... 1 . . . . . + ...... 2 ...... 1 . . . + ...... 0 ...... 5.1 ...... 2 . . . . . + . . + . . . 0 . . b . . + ...... r . . r . 1 . . 3 ...... 0 . + . . . + ...... + . . 2 . . . . . + . . + . . . 1 . . . . . + . + ...... + . + . . + . . . 8 . + . . . + ...... + . + + . + . . . 6 ...... + ...... + ...... + . . . 4 . 1 . . . . . 1 ...... + . + . . . . + . . + . . . 3 . . . . . + ...... 1 r ...... + . 8 . + . . . . . + ...... + . . 9 ...... 0 ...... + ...... 9 ...... 0 . . . . . 3 + . 4 . . . . + ...... + 5 + ...... 0 . . . . . 2 . + 1 ...... 1 . . . . + ...... 3 . . . . . 1 . . 1 . . . . + + ...... 4 + ...... + . 0 . . . . . + ...... + . . . . . + + . 0 . . . . . + ...... 3 . . . . . 1 . + ...... 8 ...... 2 . . . . . 3 ...... 4.3 ...... 6 ...... 1 . . . . . 3 . . a ...... 1 . . . . 4 + ...... 7 . . . . . 2 ...... + ...... 0 ...... + . . . 8 . . . . . 2 . . a ...... ) t ...... + . . . 6 ...... + . . . 1 . + . . . 1 + . a . . . . + ...... 5 ...... + . 7 ...... + ...... 4 + ...... 6 ...... + ...... 1 ...... + . . . 5 ...... + . . . a (con . . . . + . . . . . 1 ...... + . . . 4 . . + ...... + . . . . 2 ...... + . . . 2 . . 1 . . 2 ...... ble 1 . + . . . + + . . . 3 ...... + . . . 4 . . . . . 2 + ...... 8 + . . . . + . . + . . . 6 . + ...... ta . 1 ...... 5 ...... + . . . 4 . + . . . + . . + ...... + . . . . 0 ...... 7 . . 1 . 2 ...... 8 + . . . . + ...... 9 ...... 5 . . . . . + . . + . . . 2 . . + . . 4.2 ...... 5 ...... 0 . . . . . + ...... 2 . . . + ...... 1 2 . . b ...... + . 0 ...... + . . . 5 2 . . + ...... + 8 + ...... 1 . . . 9 1 . . . . . r . . + . . . . + ...... + 2 . . . + ...... 9 1 . + ...... 1 + ...... + . . . 9 ...... r ...... 7 ...... + . . . 2 3 . . . . . + . . . . . + ...... 4 ...... 9 3 . . 1 ...... 4 ...... 7 3 ...... 8 ...... 1 1 . . + . . 4.1 + ...... 3 ...... + . . . 3 1 . r . . . + . . . . . + ...... 8 ...... 8 ...... + . 8 ...... + . . . 7 ...... + ...... 7 ...... 9 3 . . . . . + ...... + . . . . 5 ...... + . . . 1 2 . . + ...... 1 ...... + . 5 2 ...... + ...... 1 5 . . . . . + . . . . + . 6 ...... + . + ...... 4 + ...... 1 ...... + ...... + . . . 3 + . . . . + . . . . + . 9 ...... 3 ...... 6 ...... + ...... 8 ...... 3 . . . 0 ...... 1 . . . . 9 ...... 1 . . . 7 Not mapped ...... a . . . + 9 . . . . . + . . . 1 . . . 1 ...... b ...... 4 ...... 2 . . . 5 ...... + . . . r + . . 6 ...... + . 1 . . . 5 . . . . 2 ...... r ...... 6 ...... 1 ...... 7 . r . . + . . r ...... + 2 ...... + 1 ...... 4 ...... + . . . . + + . . . 8 + ...... 3 ...... 1 ...... 8 ...... 3 ...... 7 ...... 7 ...... 3 ...... + 3 ...... + ...... 4 ...... + ...... 3 ...... 8 . . . . . 3 ...... Not mapped . . . + 5 ...... + ...... 3 ...... sp. (HR219) sphaerocephala s group M s group N s group O s group P s group Q s group R spp. (HRp171) sp. (HRp171) sp. elevé number Eriocephalus microphyllus Pteronia glabrata Rosenia oppositifolia Eriocephalus ericoides Euryops annuus Asparagus capensis Gazania rigida Pteronia villosa Ruschia cradockensis Melolobium candicans Malephora crassa Thesium lineatum Leysera tenella Hirpicium alienatum Euryops multifidus Pteronia glomerata Felicia Galenia fruticosa Tribolium hispidum Chrysocoma ciliata Karroochloa schismoides Tetragonia fruticosa Pentzia cf. Moraea Felicia Augea capensis R Arctotheca calendula Specie Eriocephalus pauperrimus Specie Pentzia incana Specie Ruschia robusta Specie Cephalophyllum Specie Gazania lichtensteinii Specie Drosanthemum sp. African Protected Area Conservation and Science 167 KOEDOE Vol. 50 No. 1 pp. 160 - 183 http://www.koedoe.co.za

Succulent Karoo Biome related vegetation – Part 2 Original Research

. . . . . 9 . . + . + . . . 8 ...... + . . + ...... 5 . . . . 1 . . . 3 ...... 2 ...... 6 . + . . . . + + + . . . . 1 ...... 1 . . . . . 2 . . r . . . . . 5 ...... + . . . . . + 1 ...... r . 3 . . + . . + . . 8 . . . . + . + . . + . . + . + ...... 7 . . + . . 1 . . 1 . + ...... 5 . . . . . + . . 8 . . . . + . + . . . . . + . + ...... 5 . . + . + . . . 5 . . . . + . + . . . . . + . . . . + . 5.3 ...... 2 . . + . . + . . 8 . + . . + . + ...... + . . 1 . . . + . 2 . . . . . + . . 7 . + . . + . . . . + . . + ...... + . . . . . 2 ...... 6 . . . . + . + . . + . . . . + . + ...... 1 . . + . . + . . 6 . + . . + . + . . . + . + . + . + ...... 5 . . + + . . . . 1 ...... + . . . . . + ...... 1 . . . . . 4 ...... 9 ...... + . . . . . + ...... 4 . . + + . . . . 8 . . . . + . + . . + ...... + ...... + . 4 . . + . . . . . 5 . + . . . . + . . . . . + ...... + . . 1 . . . . . 1 . . 9 . . . . + . + . . . . . 1 1 . . + ...... 6 . . . . . + . . 9 . . . + . . + . . . . . + . + . + ...... 6 . . + . . . . . 8 . + . . + . + ...... 6 . . + . . + . . 5 . . . + . . . . . + ...... + . + . . . . + . . . 1 . . + + . + . . 1 . + . . . . + . . . + . + . . . . 5.2 ...... 1 . . r . . . + ...... + ...... + . . + . . 7 ...... + ...... + . . 9 ...... + ...... + . . . . . 2 ...... + . + . . . . . + ...... + 5 ...... 1 ...... 1 ...... + . . . . 3 . . . . . + . . . . . 3 ...... + ...... 3 . . . . . + . . . . . 2 ...... 2 . . . . . + . . . . . 3 ...... 1 . . b ...... + + . . . . 1 . . . . . + . . . . . 3 ...... + . . . . . + ...... + 1 . . . . . + + . . . . 2 . + . . . . + ...... + 1 . . . . . 1 . . . . . 0 ...... 5.1 . . . . r . . . . 2 ...... + . . 0 ...... + ...... + 3 . . . . . + . . . . . 0 ...... + ...... r . + . . . + 2 . . . . . + . . . . . 1 ...... + . . + ...... + . . . . . 8 . + . . . . + ...... + . . 6 ...... + . . . + ...... + . . . . . 4 ...... + ...... + ...... + . . + . . 3 ...... + ...... 1 ...... 8 ...... + . . . . . + ...... a 9 . . . + . . . . . + + . . 0 ...... + . + . . . + ...... 9 . + ...... + . . 0 . . . . + 3 . . . . + . . + . . . + + . . . 5 . . . . . + . + . . . . . 0 . . . . + 2 . + ...... 1 ...... 3 . . . . . 1 . . . . . + ...... 4 ...... 0 . . . . + . + ...... + + + . . 0 . + . . + . 1 + . . . . . 3 . . . . . 1 . + . . + . . . . . + . . . . 8 ...... 2 . . . . . 3 ...... 4.3 . . . + . . r . 6 . . . . . + ...... 1 . . . . + 3 ...... + ...... 4 ...... 7 . . . . + 2 ...... + . . + . + . . 0 ...... + ...... 8 . . . . . 2 ...... ) t ...... + 6 ...... 1 . . + . . 1 ...... 5 . . . + . . . + . + . . . 7 . . . . + . + . . . . . + . . + . . . . 4 . . . r . . . + . . + . . 6 ...... + . . + ...... 1 ...... + . . + . . 5 . + . . + . . . . + + a (con . + ...... 1 ...... + . . . . . 4 . . . . + . + ...... + . . 2 ...... 1 . . . 2 . . . . + 2 ...... ble 1 . + . . + . . . . 3 ...... + . . . . . 4 . . . . . 2 . + ...... + . . . . 8 ...... 6 ...... ta ...... 1 5 ...... + . . . . . 4 . . . . + ...... + + ...... 0 ...... + . . + . . 7 . . . 2 . + . + . + . . . 1 . . + . + . + 8 . + ...... 1 . . 9 ...... + . . + . . + . . + . r . . 5 ...... + . . + . . 2 ...... + 4.2 ...... 5 ...... + . . . . . 0 . . . . . + . + ...... + . . . . 2 . + ...... 1 . 2 ...... + . . + . + . . 0 ...... 5 . 2 . . . + . + . . . . . + . . + . . . . 8 ...... + . . 9 . 1 . . . . . + ...... + . . . . 2 ...... 9 . 1 ...... 1 ...... 9 . . . . + ...... + . . . 3 7 ...... + . . . 2 . 3 ...... 4 ...... 9 . 3 ...... + . . + 4 ...... 7 . 3 . . . . . + ...... 8 ...... + . . . . . 1 . 1 . . . + 4.1 ...... + 3 ...... + . . . 3 . 1 . . . . . + ...... + . . . . 8 ...... 8 ...... + . . . . 8 ...... + . . . . . 7 . . . . + ...... 7 ...... + + . + . . 9 . 3 . . . . . + ...... 5 ...... + . 1 . . . 1 . 2 . . . . . + . + . + ...... 1 . . . . . + . + . . + . . 5 . 2 . . . . African Protected Area Conservation and Science and Conservation Area Protected African . + . . . . . + . . + + . . . 5 ...... + . . . . . 6 . . . . + . + . . . . + . . + ...... 4 . . . + . . . + + . . . . 1 . . . . . + . + . . . . . + . . + . . . + 3 ...... + . . . . . 9 . + ...... 3 ...... 6 ...... + . . + . . . . 8 ...... + 3 . . . . 0 ...... + ...... r . . 9 ...... + . 1 1 . . . 7 . . Not mapped . . . . . + . . . . . r . . r + r . . 9 ...... + . 1 . . . . 1 ...... + ...... 4 ...... 2 . . . . 5 . . . . . + . + . . . . . + ...... 6 ...... + . 1 . . . . 5 . . . . 2 ...... + . . . . 6 ...... 1 ...... 7 . . + ...... + 2 ...... 1 ...... 4 + . . . . + . . . . . + . . . . 8 ...... + . 3 ...... 1 ...... + . . . . 8 ...... 3 ...... 7 ...... + 1 . . . 7 ...... 3 . . . . . + . . 3 . . + . . + . . . . . + . . . . 4 ...... + ...... 3 ...... + . . . . 8 ...... 3 ...... Not mapped . . . . 5 ...... + . . . . 3 ...... vaginatus (HR217) sp. cf. (HRp65) sp. apiculata (p91) sp. (p91) sp. cf. sp. s group S s group T s group U s group V s group W s group X sp. spp. rostis brevifolia rostis ciliata spp. Lampranthus elevé number Salsola Osteospermum pinnatum Exomis microphylla Tripteris sinuata Zygophyllum Brownanthus Oxalis Phyllobolus Felicia australis Galenia sarcophylla Zygophyllum retrofractum Pentzia spinescens Drosanthemum Pteronia Galenia africana Psilocaulon Eriocephalus decussatus Lycium Stipag Pteronia pallens cf. otzenianum Mesembryanthemum guerichianum Salsola aphylla Salsola tuberculata Foveolina dichotoma Stipag R Specie Braunsia Geophytic spp. Specie Atriplex lindleyi Specie Ruschia intricata Specie Aridaria noctiflora Specie Cladoraphis spinosa Specie Stipagrostis obtusa http://www.koedoe.co.za Vol. 50 No. 1 pp. 160 - 183 KOEDOE 168

Original Research Van der Merwe, Van Rooyen & Van Rooyen

. 9 . . 8 . . . . . 5.3 Leipoldtia schultzei – Eriocephalus purpureus Hantam . 5 . . 3 . . . . . Karoo . 2 . . 6 . . . 1 . . . 2 . . 5 . . . 1 . . This subassociation is located around the farms Soetwater . 3 . . 8 ...... 7 . . 1 . . . . . and Leeuriet southwest to Mensieskraal and stretches . 5 . . 8 . . . . . southeastwards to approximately the farm Soutpan (Fig. 1). . 5 . . 5 . . . . . 5.3 . 2 . . 8 . . . . . It covers an area of 58 214 ha (3.5% of the total area in Fig. 1). . 2 . . 7 . . . . . Subassociation 5.3 occurs on Dwyka tillites and Ecca shales . 2 . . 6 ...... 1 . . 6 . . . . . at an altitude ranging from 300 to 1 000 m, and the dominant . 5 . . 1 . . . . . land types are Da and Fc. The ridges of this subassociation are . 4 . . 9 ...... 4 . . 8 . . . . . generally level, occasionally gentle, with a low rock cover. The . 4 . . 5 . . . . . loamy soils are red brown or light brown in colour. . 1 . . 9 . . . 1 . . . 6 . . 9 ...... 6 . . 8 . . . . . Shrub cover in this subassociation varies from 45 to 80%. The . 6 . . 5 ...... 1 . . 1 . . . . grass component is absent while the annual component varies 5.2 . . . . 1 . . . . greatly, from < 1% to 60%. Subassociation 5.3 shares species . . . . 7 ...... 9 . . . groups I and J with subassociations 5.1 and 5.2 but lacks its own . . . . . 2 . . . defining group (Table 1a). The dominant species are Leipoldtia . . . 5 . . 1 . . . 1 . . . 3 . . 3 . . . schultzei and Gorteria diffusa (species group J), while other species . . . 3 . . 2 . . . describing this subassociation include Pentzia incana, Ruschia . . . 2 . . 3 ...... 1 . . 3 . . . cradockensis and Eriocephalus microphyllus (species group n), . . . 1 . . 2 . . . as well as Aridaria noctiflora, Tripteris sinuata, Drosanthemum . . . 1 . . 0 . . . 5.1 . . . 2 . . 0 . . . (HR217) sp. and occasionally a high cover of Pteronia pallens . . . 3 . . 0 . . . (species group V). . . . 2 . . 1 ...... 8 ...... 6 . . . 6. Pteronia glomerata Roggeveld Karoo . . . . . 4 ...... 3 . . . This association is predominantly found along the eastern . . 1 . . . 8 . . . . . 9 . . . 0 boundary of the study area (Fig. 1) and has been termed . . . . . 9 . . . 0 3 . . . Roggeveld Karoo. The Ecca Group, characterised by shales, . . 5 . . . 0 2 . . . and the Beaufort Group, characterised by mudstones, meet in . . 1 . . . 3 1 . . . . . 4 . . . 0 . . . this transitional area. Additionally, this area is the transition . . 0 . . . 3 1 . . . between the winter rainfall region to the west and the summer . . 8 . . . 2 3 . . . 4.3 . . 6 . . . 1 3 . . . rainfall region to the east. In general, this association occurs . . 4 . . . 7 2 . . . at an altitude greater than 1 000 m above sea level and . . 0 . . . 8 2 ...... ) t . . 6 . . . 1 1 . . . predominantly on Land Types Fc or Da. The level to gently . . 5 . . . 7 . . . sloping ridges, plains and valleys commonly have light brown . . 4 . . . 6 . . . . . or brown loamy soils. 1 . . . 5 . . . a (con . . 1 . . . 4 . . . . . 2 . . . 2 2 . . . A high shrub cover (usually > 60%) is found, whereas the grass ble 1 . . 3 . . . 4 2 ...... 8 . . . 6 . . ta and annual components are absent or < 5%. Association 6 . . . 5 . . . 4 . . (Table 1b) is related to associations 4 and 5 (Table 1a) through . . . 0 . . . 7 2 ...... 8 . . . 9 common species such as Pentzia incana, Ruschia cradockensis . . . . . 5 . . . 2

4.2 and Pteronia glomerata (species group N), but also related to . . . . . 5 . . . 0 . . . . . 2 . . . 1 2 association 7 (Table 1c) through common species such as Ruschia . . . . . 0 . . . 5 2 intricata and Eriocephalus decussatus (species group U)...... 8 . . . 9 1 . . . . . 2 . . . 9 1 . . . . . 1 . . . 9 6.1. Phyllobolus tenuiflorus – Pteronia glomerata Roggeveld . . . . . 7 . . . 2 3 Karoo . . . . . 4 . . . 9 3 . . . . . 4 . . . 7 3 Located in the vicinity of the farms Rooiwal and Weltevrede on . . . . . 8 . . . 1 1 4.1 . . . . . 3 . . . 3 1 dolerite soils, southeast of Calvinia (Fig. 1), subassociation 6.1 is . . . . . 8 . . . 8 found predominantly on Land Types Fc and Ag, and covers an . . . . . 8 . . . 7 . . . . . 7 . . . 9 3 area of 9 172 ha (0.6% of the total area in Fig. 1). It is generally . . . . . 5 . . . 1 2 found on level terrain to gently sloping ridges and plains at an . . . . . 1 . . . 5 2 . . . . . 5 . altitude varying from 1 000 to 1 300 m above sea level. Soils are . . 6 . . . . . 4 . . . 1 loamy and usually brown, with a rock cover that ranges from . . . . . 3 . . . 9 zero to 80%, comprising small stones and stones...... 3 . . . 6 . . . . . 8 . . . 3 0 . . . . . 9 . . . 1 7 Shrub cover in this subassociation varies from 5 to 90%. The Not mapped . . . . . 9 . . . 1 1 . . . . . 4 . . . 2 5 grass component is generally absent except in isolated cases . . . . . 6 . . . 1 5 where it can be as high as 25 to 40%. The annual component 2 . . . . . 6 . . . 1 7 . . . . . 2 . . . 1 varies considerably from < 1% to 60%. Subassociation 6.1 4 . . . . . 8 . . . 3 comprises species groups K, N, U and V but lacks species groups 1 . . . . . 8 . . . 3

7 L and M, which are present in subassociations 6.2, 6.3 and 6.4 . . . . . 7 . . . 3 3 . . . . . 4 . . . (Table 1b). Annual species include Phyllobolus tenuiflorus (species 3 . . . . . 8 . . 3 .

Not mapped group K), Felicia australis and Foveolina dichotoma (species group 5 . . . . . 3 . . . U), and Galenia sarcophylla (species group V), while perennial species include Pentzia incana, Ruschia cradockensis and Pteronia glomerata (species group N) (Table 1b).

(HRp386) sp. 6.2. Eriocephalus pauperrimus – Pteronia glomerata Roggeveld (HRp118) sp. (HRp118) (HRp118) sp. (HRp118) s group Y s group Z Karoo (HRp118) sp. (HRp118) (HRp118) sp. (HRp118) This subassociation is located within the Basterberg Mountains elevé number Lycium Mesembryanthemaceae 2 (HRp359) sp. Blepharis pruinosa Salsola Zygophyllum Pteronia R Specie Mesembryanthemaceae 1 (HRp359) sp. Specie Pteronia * Non-diagnostic species are excluded * HR collection code and numbers are included for future reference, if necessary* Specimens inadequate for identification yet different from species that could* sp. = one be speciesidentified in a taxonomic are indicated group with a plot number(p…) * spp. = more than one species in a taxonomic group. These species, even although they are grouped together, are included in the table since they occur in different species groups. However, they are not used in the descriptions in the text. (Fig. 1) and covers an area of 50 790 ha (3.1% of the total area African Protected Area Conservation and Science 169 KOEDOE Vol. 50 No. 1 pp. 160 - 183 http://www.koedoe.co.za

Succulent Karoo Biome related vegetation – Part 2 Original Research

in Fig. 1). It is predominantly found on Land Types Da and Fc interrupting the shales. The subassociation occurs at altitudes at an altitude of 1 100 to 1 400 m. These level to gently sloping ranging from 900 to 1 200 m above sea level on level to gently mudstone ridges of the Beaufort Group are covered with brown sloping ridges, with generally a higher rock cover than in or light brown coloured loamy soils with a rock cover varying subassociations 6.1, 6.2 and 6.3. from zero to 95% stone. The shrub layer covers 40 to 80% of the light brown and brown The shrub cover is high while the grasses and annuals make loamy soils, while the grass and annual component is either little contribution to the overall plant cover. Conspicuous absent or covers < 5%. Subassociation 6.5 differs from the species in this subassociation include Phyllobolus tenuiflorus previous subassociations because it lacks species groups K, L (species group K) and Eriocephalus pauperrimus (species and M (Table 1b). Dominant species in this subassociation are group M) (Table 1b). Other common species are Pentzia incana, Pentzia incana, Pteronia glomerata, Eriocephalus ericoides (species Ruschia cradockensis, Pteronia glomerata, Karroochloa schismoides group N), Ruschia intricata (species group U), Aridaria noctiflora, (species group N), Eriocephalus decussatus (species group U) and Tripteris sinuata and Galenia africana (species group V). These Tripteris sinuata (species group V). species are all common to the entire association 6.

6.3. Pentzia incana – Pteronia glomerata Roggeveld Karoo 7. Aridaria noctiflora Tanqua Karoo Subassociation 6.3 is located in the vicinity of Elandsoog se berg, Association 7 is located primarily in the Tanqua Karoo Basin around the farm De Puts, southwest of the Droëberg Mountains and the adjacent foothills of the Koedoesberg Mountains. and northeast of Loeriesfontein (Fig. 1), and covers an area of Large areas east and northeast of Loeriesfontein, as well 31 594 ha (1.9% of the total area in Fig. 1). This subassociation as on the western extreme of the Hantam River, a brackish is found on mudstones of the Beaufort Group and shales of the system, are also included in this association, as well as the Ecca Group, and predominantly on Land Type Da. The altitude vegetation mosaic located around the town of Calvinia (Fig. 1). ranges from 1 000 to 1 200 m and the subassociation is found Geologically, shale of the Ecca Group and tillite of the Dwyka on relatively level areas of ridges, plains and valleys. Generally, Group dominate association 7, while alluvial deposits are the light brown loamy soils have little rock cover but when rock present at times. Limited intrusions of dolerite also occur. This occurs it is in the form of stones or boulders. association is found predominantly on Land Types Ia and Fc at an altitude ranging from 200 to 1 100 m. The level to gently The shrubby component in this subassociation is well developed sloping ridges and plains occur on brown and light brown soils, but grasses and annuals are usually absent or contribute very with generally a low rock cover. little to the overall cover. Eriocephalus pauperrimus (species group M) has a high cover in this subassociation, which is closely Shrub cover is intermediate (ca. 40%), while the grass and related to subassociations 6.2 and 6.4 through the presence annual components are absent or poorly developed. The of species group M (Table 1b). Other species present include annual component is expected to be much higher in normal Pentzia incana and Eriocephalus ericoides (species group N) and and good rainfall years. Association 7 (Table 1c) differs from Ruschia intricata (species group U), as well as various Lycium associations 4 to 6 (Table 1a, 1b) as it lacks species group N but species (species group V). shares species groups U and V with association 6 (Table 1b). Associations 4 to 7 (Table 1a, 1b, 1c) are related through the 6.4. Ruschia intricata – Pteronia glomerata Roggeveld Karoo shared presence of species groups V, and have no species groups in common with associations 8, 9 and 10 (Table 1c). Common Subassociation 6.4 is located north to south along the eastern species in association 7 are Ruschia intricata (species group boundary of the study area, predominantly on mudstones U), Aridaria noctiflora, Salsola tuberculata, Tripteris sinuata and of the Beaufort Group (Fig. 1) and, excluding the mosaic Drosanthemum (HR217) sp. (species group V). This association vegetation units, covers an area of 265 394 ha (16% of the total is subdivided into five subassociations (Table 1c). area in Fig. 1). It forms the transition with the Arid Karoo and False Succulent Karoo, as described by Acocks (1953, 1988). 7.1. Ruschia robusta – Aridaria noctiflora Tanqua Karoo Elements of this subassociation are also found in the transition with the Namaqualand Broken Veld (Acocks 1953, 1988), west of Subassociation 7.1 is located at the eastern edge of the Tanqua the Hantam Mountain, in the Naresie mosaic vegetation unit, Basin at the foot of the Roggeveld Escarpment and in the predominantly on Dwyka tillite. This subassociation is found vicinity of the farm Twee Damme, and covers an area of 79 395 on Land Types Fc and Da at an altitude ranging from 800 to ha (4.8% of the total area in Fig. 1). This low-lying (300–800 m 1 400 m above sea level. The level terrain and gentle slopes are above sea level) subassociation is found on the Ecca Group on usually covered in light brown or brown coloured soils with gently sloping ridges and plains with brown coloured loamy either a low rock cover (0–15%) or a high rock cover (80–95%). soils of Land Types Ia and Fc. Rock cover varies considerably, Science and Conservation Area Protected African from no rocks to 50 to 99% consisting of small stones (> 10–50 Shrub cover usually ranges from 50 to 90%, while grass and mm). annual cover is usually absent or < 5%. Common species include Rosenia glandulosa (species group L), Eriocephalus pauperrimus Shrub cover is generally low, but can be as high as 75%. (species group M), Pentzia incana, Ruschia cradockensis, Pteronia Grass cover is low or absent, while annual cover ranges from glomerata, Eriocephalus ericoides (species group N), Ruschia very low (the norm) to isolated dense patches (50–80%). This intricata, Eriocephalus decussatus (species group U), Aridaria subassociation is characterised by perennials such as Ruschia noctiflora and Tripteris sinuata (species group V) (Table 1b). robusta, Augea capensis and the annual Euryops annuus (species group O) (Table 1c). Other perennial species are Aridaria 6.5. Eriocephalus ericoides – Pteronia glomerata Roggeveld noctiflora, Drosanthemum (HR217) sp. and Pteronia pallens (species Karoo group V), and annual species include Gazania lichtensteinii (species group Q) and Felicia australis (species group U). This subassociation, located on Land Type Fc, is found northeast and southwest of the Hantam Mountain (Fig. 1) and, excluding 7.2. Drosanthemum (HR217) sp. – Aridaria noctiflora Tanqua the mosaic vegetation units, covers an area of 69 356 ha (4.2% of Karoo the total area in Fig. 1). It is also located in a mosaic vegetation unit around the town of Calvinia (Calvinia mosaic) and in the Subassociation 7.2 is located on the alluviums in the vicinity of Adamsfontein/Noodsaaklikheid mosaic. Geologically, the the farms Weltevreden and Platfontein, southwest of Calvinia Ecca Group dominates these areas, with dolerite intrusions (Fig. 1) and, excluding the mosaic vegetation units, covers an http://www.koedoe.co.za Vol. 50 No. 1 pp. 160 - 183 KOEDOE 170

Original Research Van der Merwe, Van Rooyen & Van Rooyen

...... 5 ...... 2 ...... 2 ...... 3 ...... 2 ...... 1 ...... 6 ...... 2 ...... 0 ...... 0 ...... 7 ...... 4 ...... 6 ...... 4 . . . . + ...... 3 ...... 7 ...... 4 ...... 2 ...... 5 ...... 0 ...... 2 ...... 1 ...... 9 ...... 1 ...... 3 ...... 7 ...... 6 6.5 ...... 0 ...... 3 ...... 0 ...... 6 ...... 2 ...... 4 ...... 0 ...... 6 ...... 5 ...... + ...... 8 ...... 3 ...... 7 ...... 2 ...... 2 ...... 1 ...... 8 ...... 1 ...... 5 ...... 6 ...... 1 . + ...... 5 ...... 3 ...... 1 ...... 8 ...... 1 ...... 6 ...... 5 ...... 2 ...... 5 ...... 3 ...... 2 ...... 0 ...... 3 ...... 1 ...... 4 ...... 3 ...... 3 ...... 0 ...... 7 ...... 3 ...... 2 ...... 2 ...... 3 ...... 2 1 ...... 9 ...... 3 ...... 1 ...... 8 ...... 3 ...... 0 ...... 8 ...... 3 ...... 0 ...... 4 ...... 3 ...... 0 ...... 2 ...... 3 ...... 9 ...... 8 ...... 2 ...... 9 ...... 7 ...... 2 ...... 9 ...... 6 ...... 2 ...... 6 ...... 5 ...... 2 ...... 6.4 . 6 ...... 4 ...... 2 ...... 6 ...... 2 ...... 2 ...... 6 ...... 1 ...... 2 ...... 5 ...... 6 ...... 2 ...... 3 ...... 1 ...... 2 ...... 2 ...... 0 . . . . + ...... 2 ...... 0 ...... 4 ...... 2 ...... 8 ...... 8 ...... 1 ...... 6 ...... 0 ...... 1 ...... 4 ...... 2 ...... 1 ...... 4 ...... 1 ...... 1 ...... 4 ...... 0 ...... 1 ...... 3 ...... 7 ...... 1 ...... 3 ...... 6 ...... 1 ...... 7 ...... 9 ...... ble 1b a ...... 5 ...... 8 ...... 2 T ...... 7 ...... 6 ...... 0 ...... 2 ...... 2 ...... 3 ...... 5 ...... 1 ...... 7 ...... 8 ...... 2 ...... 7 ...... 7 ...... 2 ...... 3 ...... 0 ...... 2 ...... 2 ...... 9 ...... 2 . 6.3 ...... 3 ...... 4 ...... 1 ...... 8 ...... 0 ...... 7 ...... + 7 ...... 7 ...... 4 ...... 7 ...... 3 ...... 7 ...... 8 . . . . r ...... 5 ...... 9 . . . . . 2 ...... 3 ...... 1 . . . . . 3 ...... 6 ...... 3 . . . . . 2 ...... 2 ...... 8 . . . . . 3 ...... 0 ...... 1 . . . . . 3 ...... 9 ...... 4 . . . . . 2 ...... 6.2 ...... 7 ...... 6 . . . . . 2 ...... 9 ...... 3 . . . . . 2 ...... 9 ...... 0 . . . . . 2 ...... 6 ...... 8 . . . . . 2 ...... 6 ...... 7 . . . . . 2 ...... 6 ...... 6 . . . . . 2 ...... Phytosociological table the of Succulent Karoo Biome related vegetation the of Hantam-Tanqua-Roggeveld subregion – association 6 ...... 1 ...... 7 ...... 5 . . . . . + . . . 7 1 ...... 9 . . . . + . . . . 9 1 ...... 5 ...... 7 2 ...... 2 ...... 8 2 ...... 1 ...... 1 2 ...... 6.1 ...... 1 ...... 6 2 ...... 0 ...... 0 2 ...... 0 ...... 9 1 ...... 0 ...... 2 1 ...... 7 ...... 0 2 ...... (HR248) sp. sp. sp. (HRp378) sp. spp. (HRp35) sp. sp. s group A s group B s group C s group D s group E s group F sp. (HRp35) sp. elevé number Asparagus Phyllobolus Vygie (HRp378) sp. Euphorbia hamata Trachyandra falcata Vygie (HRp384) sp. Othonna filicaulis Euphorbia Didelta spinosa Crassula alpestris Sarcocaulon Montinia caryophyllacea Vygie (HRp382) sp. Babiana Heliophila collina Hermannia cuneifolia Lapeirousia montana Stapelia Rhus burchellii Ruschia Antimima cf. granitica Thesium cf. hystrix Senecio cakilefolius Pteronia (HRp382) sp. Lachenalia violacea Lachenalia R Specie Zygophyllum foetidum Specie Tylecodon wallichii Specie Pteronia glauca Specie Euphorbia decussata Specie Arctotis acaulis Specie Dorotheanthus African Protected Area Conservation and Science 171 KOEDOE Vol. 50 No. 1 pp. 160 - 183 http://www.koedoe.co.za

Succulent Karoo Biome related vegetation – Part 2 Original Research

...... 5 ...... 2 ...... 2 ...... 3 ...... 2 ...... 1 ...... 6 ...... 2 ...... 0 ...... 0 ...... + ...... 7 ...... 4 ...... 6 ...... 4 ...... 3 ...... 7 ...... 4 ...... 2 ...... 5 ...... 0 ...... 2 ...... 1 ...... 9 ...... 1 ...... 3 ...... 7 ...... 6 . 6.5 + ...... 0 ...... 3 ...... 0 ...... 6 . . . + ...... 2 ...... 4 ...... 0 ...... + ...... + ...... 6 ...... 5 . . + ...... + ...... + . . . 8 ...... 3 ...... 7 ...... 2 ...... 2 ...... 1 ...... 8 ...... 1 ...... 5 ...... 6 ...... 1 ...... 5 ...... 3 ...... 1 ...... + . . . . 8 ...... 1 ...... 6 ...... 5 ...... 2 ...... 5 ...... 3 ...... 2 ...... 0 ...... 3 ...... 1 ...... 4 ...... 3 ...... 3 ...... 0 ...... 7 ...... 3 ...... 2 ...... 2 ...... 3 ...... 2 . . . 1 ...... 9 ...... 3 ...... 1 ...... 8 ...... 3 ...... 0 ...... 8 ...... 3 ...... 0 ...... 4 ...... 3 ...... 0 ...... 2 ...... 3 ...... 9 ...... 8 ...... 2 ...... 9 ...... 7 ...... 2 ...... 9 ...... 6 ...... 2 ...... 6 ...... 5 ...... 2 ...... 6.4 . . 6 ...... 4 ...... 2 ...... 6 ...... 2 ...... 2 . . . . + ...... 6 ...... 1 ...... 2 ...... 5 ...... 6 ...... 2 ...... 3 ...... 1 ...... 2 ...... + ...... 2 ...... 0 ...... 2 ...... 0 ...... 4 ...... 2 ...... 8 ...... 8 ...... 1 ...... 6 ...... 0 ...... 1 ...... 4 ...... 2 ...... 1 ...... 4 ...... ) . 1 ...... 1 . . t . . . . + . . + . . . 4 ...... 0 ...... 1 ...... 3 ...... 7 ...... 1 ...... 3 ...... 6 ...... 1 ...... 7 ...... 9 ...... 5 ...... 8 ...... 2 . . . . + ...... 7 . . . . . + . 6 ...... ble (con 1b ...... 0 . . . . . + . 2 ...... 2 . . a T ...... 3 ...... 5 ...... + . 1 ...... + . . . 7 . . . . . + . 8 ...... + . . . 2 ...... + . . + . . . 7 ...... 7 ...... + . . . 2 ...... + . . . . . 3 ...... 0 ...... 2 ...... + . . . . . 2 ...... 9 ...... 2 . . 6.3 ...... 3 ...... 4 ...... + . . . 1 ...... 8 ...... 0 ...... 7 . . . . . + . 7 ...... + ...... 7 . . . . . + . 4 ...... + ...... + 7 ...... 3 ...... + . + ...... 1 . + . . . 7 ...... 8 ...... + ...... 5 ...... 9 . . . 2 ...... + ...... 3 ...... 1 . . . 3 . + ...... + ...... 6 ...... 3 . . . 2 ...... 2 ...... 8 . . . 3 . + ...... + . . 0 ...... 1 . . . 3 ...... + ...... 9 ...... 4 . . . 2 ...... 6.2 ...... 7 ...... 6 . . . 2 . + ...... + . . 9 . . . . + . . 3 . . . 2 . + ...... + . . . + . . 9 + ...... 0 . . . 2 . + . . . . . African Protected Area Conservation and Science and Conservation Area Protected African ...... + + . . + . . 6 . . . . + . . 8 . . . 2 . + ...... + . . . 6 ...... 7 . . . 2 . + . . . . a ...... + . . 6 ...... 6 . . . 2 . + ...... + ...... 1 ...... 7 . . . . + ...... + . . + . . . 5 ...... 7 1 ...... + + . . . . . 9 ...... 9 1 . . . . + ...... 5 ...... 7 2 ...... 1 ...... + ...... 2 + ...... 8 2 . . . . + ...... 1 . . . . + . . 1 2 ...... 6.1 ...... 1 ...... 6 2 . . . . + ...... + ...... 0 ...... 0 2 . . . . + . . . . + ...... + ...... 0 . . . . + . . 9 1 . . . . + . . . . a ...... + . . . 0 ...... 2 1 . . . . + ...... + ...... 7 ...... 0 2 . . . . + . latipetalum sp. sp. sp. s group G s group H s group I s group J s group K s group L elevé number Berkheya fruticosa Cotula barbata Eriocephalus namaquensis Drosanthemum cf. Lotononis hirsuta Merxmuellera stricta Dimorphotheca sinuata Cyphia digitata Tetragonia microptera Gorteria diffusa Bromus pectinatus Rosenia glandulosa Bulbine succulenta Helichrysum obtusum Lotononis Moraea Osteospermum acanthospermum Erodium cicutarium Microloma sagittatum Senecio cardaminifolius Euphorbia mauritanica Oncosiphon grandiflorum Othonna auriculifolia Zygophyllum pygmaeum R Specie Bulbinella Specie Ehrharta calycina Specie Eriocephalus purpureus Specie Leipoldtia schultzei Specie Phyllobolus tenuiflorus Specie Pentaschistis patula http://www.koedoe.co.za Vol. 50 No. 1 pp. 160 - 183 KOEDOE 172

Original Research Van der Merwe, Van Rooyen & Van Rooyen

...... + . 5 ...... 2 ...... 2 . + ...... 3 . . + ...... + 2 ...... 1 . . . . . 1 ...... 6 . . + + ...... 2 ...... 0 . + . . . . . 1 0 . . + ...... 7 . . . . . + . . 4 ...... + ...... r . 6 ...... 4 . . + ...... + 3 ...... 7 . + ...... 4 . . + ...... 2 ...... 5 ...... 0 . . + ...... 2 ...... + . . . . + 1 . + . . . + . . 9 . . + ...... + . . 1 ...... + . . . . + 3 ...... 7 . . + + ...... + . . . . 1 . . + . . 6 6.5 . + ...... 0 . . + ...... + . . 3 . a ...... 0 . 1 ...... 6 ...... + . 2 ...... 4 ...... 0 ...... + ...... 6 ...... 5 . . + ...... + ...... 8 ...... + . 3 ...... 7 . . + ...... 2 . . + ...... + 2 ...... 1 + . + ...... 8 ...... + 1 . + . . . . . + . . . . 5 . + + ...... 6 . . + ...... 1 ...... + . . . . 5 ...... 3 . . . + ...... 1 ...... + . . . . 8 ...... + . 1 . . 1 ...... r + + ...... 6 ...... 5 + ...... + 2 ...... 5 ...... + 3 . . + + ...... 2 . . + ...... + . 0 . . + ...... 3 1 . + + ...... 1 . . + ...... + . 4 ...... 3 + . . + ...... 3 . . . . . + ...... 0 . . + ...... 7 . + ...... + 3 ...... + ...... 2 . + + ...... 2 . + + + ...... 3 ...... + ...... 2 . 1 r ...... 9 . . a ...... + 3 ...... + . + ...... 1 1 + . . . . . + 8 . . + ...... + 3 ...... + . . . . . 0 + ...... 8 . + ...... + 3 ...... + . . . . . 0 r . . . . + . . 4 . . 1 + ...... + 3 ...... 1 . . . . . 0 . + ...... 2 . . 1 + ...... 3 . . + ...... 9 . + . . . . + + 8 . + ...... 2 ...... + . . . . . + + 9 + ...... 7 . . . + ...... 2 . . . + . . . . 1 . . . . . 9 + ...... 6 . . + ...... 2 . . . + ...... 6 . + ...... 5 1 + + + ...... 2 . . . + . . . . + . . . 6.4 . . . 6 + + . . . . + . 4 . . + ...... 2 . . r 1 . . . . + . . + . . 6 ...... 2 + . + + ...... + 2 . . + ...... 6 . + ...... 1 1 . + ...... 2 . + + ...... 5 + . . . . + . . 6 . . 1 + ...... + 2 ...... 3 ...... 1 . . + ...... 2 . . + ...... + . 2 ...... 0 1 . + ...... 2 + . . 1 ...... 0 . + ...... 4 . . + ...... 2 + . + . . . . + ...... 8 . + . . . . + . 8 . . + ...... 1 ...... 6 . . . . + . + . 0 + . + ...... 1 ...... 4 . . . . . + . . 2 + 1 ...... 1 . . + ...... 4 . . . . . + ...) . . 1 + . + ...... 1 . . t . + ...... + . . 4 ...... 1 0 . . + + ...... + 1 . . + . . + ...... 3 . + . . . . + . 7 . . + ...... 1 . . . . . + ...... 3 + + ...... 6 . . + ...... 1 . + + . . . r ...... 7 ...... 9 . . 1 + ...... 5 ...... 8 + ...... + 2 . . + ...... + . . 7 . + ...... 6 . . . + ...... ble (con 1b ...... 0 . + . . . . + 1 2 ...... 2 . . a T ...... 3 . . . . . + r . 5 . . . + ...... + 1 ...... + . . 7 . . . . . + . . 8 + . 1 + ...... 2 ...... + . + 7 . a . . . + . . 7 + . + + ...... 2 ...... 3 . + ...... 0 . + ...... 2 ...... 2 . + . . + . . . 9 a . + ...... 2 . . 6.3 + ...... 3 . . . . . + . . 4 . . 1 + ...... + 1 . . + ...... + + . 8 . + . . . + . . 0 . . + 1 ...... + ...... + . 7 ...... 7 + ...... + ...... 7 . + ...... 4 + ...... 7 . + ...... 3 + ...... + ...... + . . 7 ...... 8 1 . + ...... + ...... 5 + ...... 9 1 + . . 2 ...... 1 ...... 3 + + ...... 1 . + . . 3 ...... + . . . . + ...... 6 . + ...... 3 1 . + + 2 ...... 1 ...... 2 . + . . . . . + 8 . + . . 3 ...... 1 ...... 0 + . . . . + . . 1 + . + . 3 ...... + . . + . a ...... 9 . + ...... 4 . . + . 2 ...... + . . 6.2 + + ...... + 7 . + ...... 6 . . + + 2 ...... + ...... 9 . . . . . + . a 3 . . . 1 2 ...... + . . + ...... 9 . . . . . + . . 0 a . . + 2 ...... + . . . . + ...... 6 . + ...... 8 a + . 1 2 ...... + . . + ...... + . . 6 . r ...... 7 . . . 1 2 ...... + ...... 6 . . . . . + . . 6 1 . . + 2 ...... 1 ...... + . 1 ...... + . 7 ...... b ...... 5 . . . . . + . 1 7 1 . . . + ...... + . 9 . 1 ...... 9 1 . . 1 + ...... + ...... 5 . + ...... 7 2 . . + + ...... + . . 2 . + ...... 8 2 ...... + ...... 1 ...... 1 2 . . . + ...... 6.1 + ...... 1 ...... + . 6 2 ...... + ...... 0 ...... 0 2 . . + ...... + ...... 0 ...... 9 1 ...... + . . 0 . . . a . . . . 2 1 ...... + . + . . . . . + ...... 7 . + . . . . + . 0 2 ...... sp. (HR219) sp. sphaerocephala s group M s group N s group O s group P s group Q s group R spp. sp. (HRp171) sp. (HRp171) elevé number Pentzia cf. Moraea Felicia Augea capensis Karroochloa schismoides Hirpicium alienatum Thesium lineatum Gazania rigida Rosenia oppositifolia Ruschia cradockensis Eriocephalus ericoides Melolobium candicans Pteronia villosa Pteronia glabrata Malephora crassa Euryops annuus Arctotheca calendula Tetragonia fruticosa Chrysocoma ciliata Euryops multifidus Pteronia glomerata Felicia Tribolium hispidum Galenia fruticosa Leysera tenella Asparagus capensis Eriocephalus microphyllus R Specie Eriocephalus pauperrimus Specie Pentzia incana Specie Ruschia robusta Specie Cephalophyllum Specie Gazania lichtensteinii Specie Drosanthemum African Protected Area Conservation and Science 173 KOEDOE Vol. 50 No. 1 pp. 160 - 183 http://www.koedoe.co.za

Succulent Karoo Biome related vegetation – Part 2 Original Research

. . . + . . . . 5 . . . 1 . . . + . . . . 2 ...... + . . . 2 ...... + . 3 . . . . . + ...... 2 ...... + + . 1 . 1 + . . . + + . 6 . . . . . + . . . . . + 2 ...... 0 . . . . r . . + . 0 . . . + ...... + . . . . + . . . + . . 7 . . . . r . . . + 4 ...... + . . . + + . 6 . . . + . . . . . 4 . . . 1 . + ...... 3 ...... + . + . . 7 . . . + + . . . . 4 1 . . + ...... 2 ...... 5 . . . + + . . . . 0 1 ...... + . . . + 2 ...... + + . . 1 . . + . + . . . . 9 . . . 1 . + ...... 1 ...... + . . 3 . . . . + . . + . 7 + ...... 6 6.5 . . . + . . . . . 0 ...... 3 ...... 0 ...... 6 . . . . . + . . . . . + 2 . . . . . + . . . . . 4 ...... + . 0 . 1 ...... + . . . . + . 6 . . . . + . + + . 5 . . . . . + . + ...... + . + . + ...... 8 ...... + . . . + 3 . . . . + . . . . . + 7 . . . + . . . + . 2 . . . a ...... + 2 ...... + . . 1 ...... 8 . . . 1 ...... 1 ...... + . . 5 ...... + + . 6 . . . 1 ...... + . 1 ...... + . . 5 ...... + . 3 . . . a ...... 1 ...... 8 + ...... 1 . . . . . + . r ...... + . + . + + + 6 . . . . + . + . . 5 ...... + + . 2 ...... 1 + + . . 5 . . . . 1 . + . + 3 . + . . . . . 1 . + . 2 . . . . + ...... 0 . . . + . . . . . 3 . . . + . + . . . . . 1 . . . . + ...... + 4 ...... 3 ...... 3 ...... 0 . . . . + . . . . 7 + ...... + . . . 3 ...... 2 . . . a . . . . . 2 . . . + . + . . . . . 3 ...... 2 1 . . . + + . . . . 9 ...... 3 ...... 1 . . . . + . . . . 8 ...... 3 ...... 0 . . . . + . . . . 8 . . . 1 ...... 3 ...... 0 . . . . + . . . . 4 ...... 3 ...... + + . 0 . . . + . . . . . 2 . . . + . + . + . . . 3 ...... 9 ...... 8 . . . . . + . . . . . 2 ...... 9 . . . 1 + . . . . 7 + . . + . + . . . . . 2 ...... 9 . . . + + . . . . 6 ...... 2 ...... + . . . . 6 ...... 5 ...... 2 ...... + . 6.4 . 6 . . . . + . . . . 4 ...... 2 ...... + . . . 6 . . . + r . . . . 2 . . . a . . . r . + . 2 ...... r . . 6 . . . + + . + . . 1 ...... 2 ...... + . + 5 . . . + 1 . + . . 6 + ...... + . 2 ...... + . + + . . 3 . . . + + . . . . 1 . . . . . + . . + . . 2 ...... + . . . 2 . . . . + . . . . 0 . . . . . + . . . . . 2 ...... 1 . . . 0 . . . + + . + . . 4 . . . 1 . + . . . . . 2 + ...... + . . . . 8 + . . 1 + . . + . 8 . . . 1 . + . . . . . 1 . + ...... 6 . . . . + . . . . 0 . . . 1 ...... 1 1 ...... + . . . . 4 ...... 2 ...... + . . . 1 + ...... 4 . . . + + . . ...) . . 1 . . . 1 . . . + . . . 1 . . t ...... 4 ...... + . . 0 . . . . . + . . . . . 1 + ...... + . . . . 3 . . . . + . . + . 7 . . . 1 ...... 1 ...... + . . . . 3 . . r . + . . . . 6 . . . 1 . r . . . . . 1 + ...... 7 . . . . + . . + . 9 . . . + . r ...... + . . . + 1 . + 5 . . . . + . . . . 8 ...... + . 2 ...... + . . . + . 7 . + . 1 + . . + . 6 . . . . . + ...... ble (con 1b ...... + . . 0 . + . . + . . . . 2 + ...... 2 . a T . + . . . . . + . + . 3 . . . + + + . . . 5 . . . . . r . . + . . 1 ...... 7 ...... 8 . . . + ...... 2 ...... 7 . . . . + . . . . 7 + ...... + . . . 2 ...... + + + . . 3 . . . . + . . . . 0 + ...... + . . . 2 . . . . + . . + . + + . 2 . . . . + . + . . 9 ...... r . . . 2 . 6.3 . . . + . . + . . . . 3 . . . + . . + + + 4 . . . + . . . r . . . 1 ...... r . . + . 8 . . . 1 + . . + . 0 . . . + ...... + . . . . . + . 7 . . . 1 . . . . . 7 . . . + ...... + . . . . . + . 7 ...... 4 . + . . . + ...... + . 7 . . . . r . + . . 3 ...... + . + . . . + . 7 . . . . + . . + . 8 . . . + . . . + ...... + . . . . 5 . . . . + . + . . 9 ...... 2 . . + ...... + . + . 3 . . . . + . . . . 1 ...... 3 ...... + . . r . . . . 6 . . . . + . . . . 3 ...... 2 . . . + ...... 2 ...... + . 8 ...... 3 . + . . . . + . . . . . + . . . . 0 . . . + . . . . . 1 ...... 3 . . . + ...... + . . . . 9 . . . + + . + . . 4 . . . . . + 2 ...... 6.2 . . . + . . + . . + . 7 . . . 1 + . . . . 6 ...... 2 ...... 9 . . . . + . . . . 3 ...... 2 ...... + . . . . . 9 . . . . + . . . . 0 ...... 2 ...... African Protected Area Conservation and Science and Conservation Area Protected African . . . + ...... 6 . . . . + . . . . 8 ...... 2 ...... + . . . . . 6 ...... 7 ...... 2 ...... + . . . 6 ...... 6 1 . . . . . 2 . . . + . . . + . . . + + . . + . 1 ...... + . 7 . + ...... r ...... + . 5 ...... 7 1 ...... r . . . + . + . . . . . 9 ...... 9 1 ...... + ...... 5 ...... + . + 7 2 . . . + ...... + . + . . + + + 2 . . + . . . a . + 8 2 . . . . . + . + . + . . . . . + . . + + . + . 1 . . . . + . . . . 1 2 ...... + . 6.1 ...... + + . . . 1 ...... + 6 2 . + ...... + ...... + . . . . + 0 ...... 0 2 ...... + ...... + . . + + . . . 0 . . . . r . . . . 9 1 ...... + . 0 . . . . r . . . . 2 1 ...... 7 ...... 0 2 . . + . . . . vaginatus (HR217) sp. cf. (HRp65) sp. apiculata (p91) sp. (p91) sp. cf. sp. s group S s group T s group U s group V s group W s group X sp. spp. rostis brevifolia spp. Lampranthus otzenianum elevé number Salsola aphylla Foveolina dichotoma Salsola tuberculata Pteronia pallens Galenia africana Brownanthus Phyllobolus Exomis microphylla Felicia australis Salsola Galenia sarcophylla Tripteris sinuata Zygophyllum retrofractum Zygophyllum Geophytic spp. Mesembryanthemum guerichianum Pteronia Pentzia spinescens Psilocaulon Eriocephalus decussatus Lycium Stipag Drosanthemum Oxalis Osteospermum pinnatum R cf. Specie Braunsia Specie Atriplex lindleyi Specie Ruschia intricata Specie Aridaria noctiflora Specie Cladoraphis spinosa Specie Stipagrostis obtusa http://www.koedoe.co.za Vol. 50 No. 1 pp. 160 - 183 KOEDOE 174

Original Research Van der Merwe, Van Rooyen & Van Rooyen

area of 103 665 ha (6.2% of the total area in Fig. 1). It is also found near Kalkgat in the north of the Tanqua Basin (Kalkgat mosaic) and in the pan system in the Tanqua Basin (Tanqua pan mosaics). . 5 . . . . 2 . . . 2 It occurs predominantly on Land Types Ia and Fc. Geologically, . 3 . . . . 2 . . . 1 . 6 . . . . 2 . . . 0 alluvial deposits dominate, while shales of the Ecca Group are . 0 ...... 7 present at times. Limited intrusions of dolerite also occur. This . 4 ...... 6 . 4 . . . . 3 . . low-lying (400–800 m above sea level) subassociation is found . 7 . 4 . . . . 2 . . . 5 on level to gently sloping ridges and plains with no rock cover . 0 . . . . 2 . . . 1

. or a 50 to 98% rock cover comprising small stones. The sands 9 . . . . 1 . . . 3 . 7 ...... 6

6.5 are brown, light brown or red brown in colour. . 0 . . . . 3 . . . 0 . 6 . . . . 2 . . . 4 . 0 ...... 6 The shrub cover is intermediate to high (30–80%), the grassy . 5 ...... component is absent, and annual cover is very low. The . 8 . . . 3 . . . 7 . . 2 . . . 2 . . . 1 . subassociation is characterised by a weak species group P with . 8 . . . 1 . . . 5 . species such as Cephalophyllum sp., Pteronia villosa and Pteronia . 6 . . . 1 . . . 5 . . 3 . . . 1 . . . 8 . glabrata (Table 1c). Other common species include Malephora . 1 ...... 6 . crassa (species group R), Ruschia intricata (species group U) as . 5 . . . 2 . . . 5 . . 3 . . . 2 . . . 0 . well as Aridaria noctiflora, Salsola tuberculata, Tripteris sinuata . 3 . . . 1 . . . 4 . and Pteronia pallens (species group V). Subassociations 7.1 and . 3 . . . 3 . . . 0 . . 7 . . . 3 . . . 2 . 7.2 share Gazania lichtensteinii of species group Q. . 2 . . . 3 . . . 2 . . 1 . 9 . . . 3 . . . . 1 . 8 . . . 3 . . . 7.3. Malephora crassa – Aridaria noctiflora Tanqua Karoo . 0 . 8 . . . 3 . . . . 0 . 4 . . . 3 . . . This subassociation is found predominantly at the southern . 0 . 2 . . . 3 . . . extreme of the Tanqua Basin, i.e. Ceres Karoo, as well as on the . 9 . 8 . . . 2 . . . . 9 . 7 . . . 2 . . . western extreme of the Hantam River where the river system . 9 . 6 . . . 2 . . . is brackish (Fig. 1). Subassociation 7.3, excluding the mosaic . 6 . 5 . . . 2 . . . 6.4 . 6 . 4 . . . 2 . . . vegetation units, covers an area of 176 425 ha (10.6% of the total . 6 . 2 . . . 2 . . . area in Fig. 1). It also occurs in the Kalkgat mosaic southwest of . 6 . 1 . . . 2 . . . . 5 . 6 . . . 2 . . . Calvinia and in the Windheuwel/Rooiheuwel mosaic between . 3 . 1 . . . 2 . . . the Roggeveld and Koedoesberg Mountains. Shales of the Ecca . 2 . 0 . . . 2 . . . . 0 . 4 . . . 2 . . . Group and tillite of the Dwyka Group are found in these areas. . 8 . 8 . . . 1 . . . Land Types Da, Ia and Fc dominate the area that ranges in . 6 . 0 . . . 1 ...... 4 . 2 . . . 1 . altitude from 200 to 1 000 m above sea level. These generally . . . 4 . ...) 1 . . . 1 .

t level ridges and plains have a low rock cover on light brown . . . 4 . 0 . . . 1 . . . . 3 . 7 . . . 1 . loamy soils. . . . 3 . 6 . . . 1 . . . . 7 . 9 . . . . Shrub cover averages 30 to 40%, while the cover of grasses and . . . 5 . 8 . . . 2 . . . . 7 . 6 . . . .

ble (con 1b annuals is absent or very low. There is no diagnostic species . . . 0 . 2 . . . 2 . a

T group defining subassociation 7.3 (Table 1c). Prominent species . . . 3 . 5 . . . 1 . . . . 7 . 8 . . . 2 . include Malephora crassa (species group R), Atriplex lindleyi . . . 7 . 7 . . . 2 . (species group T), Ruschia intricata (species group U), Aridaria . . . 3 . 0 . . . 2 . . . . 2 . 9 . . . 2 . noctiflora, Salsola tuberculata, Drosanthemum (HR217) sp. and 6.3 . . . 3 . 4 . . . 1 . Pteronia pallens (species group V). The absence of species groups O, . . . 8 . 0 ...... 7 . 7 . . . . P and Q distinguishes this subassociation from subassociations . . . 7 . 4 . . . . 7.1 and 7.2, but the presence of species group R shows affinity . . . 7 . 3 ...... 7 . 8 . . . . with them. A relationship with subassociation 7.4 is indicated . . . 5 . 9 . . . 2 . by the shared presence of species group T (Table 1c). . . . 3 . 1 . . . 3 . . . . 6 . 3 . . . 2 . . . . 2 . 8 . . . 3 . 7.4. Atriplex lindleyi – Aridaria noctiflora Loeriesfontein Karoo . . . 0 . 1 . . . 3 . . . . 9 . 4 . . . 2 . Subassociation 7.4 is located around the town of Loeriesfontein 6.2 . . . 7 . 6 . . . 2 . . . . . 9 . 3 . . + 2 and east thereof (Fig. 1) and, excluding the mosaic vegetation . . . . 9 . 0 . . . 2 unit, covers an area of 184 612 ha (11.1% of the total area in Fig. 1). . . . . 6 . 8 . . + 2 . . . . 6 . 7 . . . 2 In the Calvinia mosaic subassociation 7.4 (Table 1c) is found in . . . . 6 . 6 . . . 2 combination with subassociation 6.5 (Table 1b). These areas are . . . . 1 . 7 . . . closely associated with what Acocks (1953, 1988) termed False . . . . 5 . 7 1 ...... 9 . 9 1 . . . Succulent Karoo. The vegetation has a desert character and is . . . . 5 . 7 2 . . . sparsely populated with mesembs (vygies) and relics of the Arid . . . . 2 . 8 2 ...... 1 . 1 2 . . . Karoo. The degradation of the vegetation has been ascribed to 6.1 . . . . 1 . 6 2 . . . the excessive grazing pressure and, consequently, the species . . . . 0 . 0 2 ...... 0 . 9 1 . . . that are of value for grazing are precisely the ones that no . . . . 0 . 2 1 . . . longer appear in it (Acocks 1953, 1988). This subassociation . . . . 7 . 0 2 . . . occurs on the shales of the Ecca group and a network of Karoo dolerites, predominantly on Land Types Ia, Fc and Da, at an altitude ranging from 400 to 1 000 m. These usually level ridges generally have a low rock cover on the brown to light brown coloured loamy soils.

(HRp386) sp. An intermediate shrub cover (40–60%), and grass and annual components that are absent or very low, describe this (HRp118) sp. (HRp118) (HRp118) sp. (HRp118) s group Y s group Z (HRp118) sp. (HRp118) (HRp118) sp. (HRp118) subassociation. Species group S with species such as Braunsia cf. apiculata and Brownanthus cf. vaginatus characterise this elevé number Pteronia Blepharis pruinosa Mesembryanthemaceae 2 (HRp359) sp. Lycium Zygophyllum Salsola R Specie Mesembryanthemaceae 1 (HRp359) sp. Specie Pteronia * Non-diagnostic species are excluded * HR collection code and numbers are included for future reference, if necessary* Specimens inadequate for identification yet different from species that could* sp. = one be speciesidentified in a taxonomic are indicated group with a plot number(p…) * spp. = more than one species in a taxonomic group. These species, even although they are grouped together, are included in the table since they occur in different species groups. However, they are not used in the descriptions in the text. subassociation (Table 1c). Atriplex lindleyi (species group T), African Protected Area Conservation and Science 175 KOEDOE Vol. 50 No. 1 pp. 160 - 183 http://www.koedoe.co.za

Succulent Karoo Biome related vegetation – Part 2 Original Research

Ruschia intricata (species group U), as well as Aridaria noctiflora, In some areas the pan systems contain no vegetation. This low- Salsola tuberculata and Tripteris sinuata (species group V) are also lying association occurs on level valley floors or on the edges abundant. of the pans. Shrub cover can be as high as 50% while grasses and annuals are not usually present. Species characterising this 7.5 Ruschia intricata – Aridaria noctiflora Tanqua Karoo association are represented in species group Z (Table 1c). This subassociation is located at the foothills of the Koedoesberg Mountains (Fig. 1) and, excluding the mosaic vegetation The three major river systems in the region, namely the Hantam, unit, covers an area of 73 290 ha (4.4% of the total area in Tankwa and Doorn Rivers, were not sampled, but were mapped Fig. 1). It is also found in the Tanqua Karoo Inselberg mosaic (Fig. 1). These drainage systems occur on Land Type Ia. Many south of Calvinia. These inselbergs include the Nuwewater tributaries originating to the northwest of the Hantam Mountain se berg, Elandsberg, Eselberg, Leeuberg, Potkleiberg and converge to form the Hantam River, which only flows after Sterretjieberg. The subassociation occurs predominantly on heavy rains. Patches of Salsola tuberculata and Salsola cf. aphylla the Skoorsteenberg Formation of the Ecca Group in Land Types dominate this drainage system. Other species include Pentzia Fc and Ia. In the mosaics it co-occurs with subassociation 5.2 incana, Ruschia cradockensis, Atriplex lindleyi, Aridaria noctiflora (Table 1a), which occupies the dolerites of these inselbergs. The and Galenia africana. There are numerous flood-irrigated lands level to occasionally gently sloping ridges on light brown or as a result of the more favourable water conditions along the brown coloured loams have either a low rock cover of < 5% or a drainage system. Other forms of transformation evident are high rock cover of 70 to 90%. the presence of the invasive alien Prosopis species and the naturalised species Atriplex lindleyi. An intermediate (40–45%) shrub cover is present in this subassociation and the grass and annual components are The Succulent Karoo vegetation (Acocks 1953, 1988) of the usually absent. There is no diagnostic species group defining Tankwa and Doorn Rivers comprises species such as Acacia this subassociation and species groups U and V are the only karoo and Malephora crassa, and is usually dominated by Salsola links with the other subassociations in association 7 (Table 1c). tuberculata. Transformation of the natural vegetation into flood- Ruschia intricata (species group U) and Aridaria noctiflora(species irrigated lands next to these drainage systems has occurred in group V) dominate the subassociation. the past. Few of these lands are still utilised for cropping and now lie barren with limited vegetation cover to combat erosion. 8. Stipagrostis obtusa Central Tanqua Grassy Plains The invader Prosopis species is also a serious problem, and has taken over vast areas of these drainage systems. Association 8 is located in the central Tanqua Basin (Fig. 1) and covers an area of 239 781 ha (14.5% of the total area in Fig. 1). It occurs from 200 to 1 000 m above sea level, predominantly on DISCUSSION Land Type Fc. Tillite of the Dwyka Group and mudrock and shales of the Ecca Group comprise the geology of this area. The The clear difference in species composition of the Fynbos sandy plains and ridges generally have no rock cover or have related vegetation units and the Succulent Karoo vegetation a high rock cover in localised patches, with a 60 to 99% cover units was supported by the TW INSPAN on the entire data set of gravel, small stones and/or stones. The sandy soils vary in of 390 relevés. This difference subsequently led to the final two colour from light brown to brown to red brown. phytosociological tables and associated maps as discussed in this article and in Van der Merwe et al. (2008). However, since the Shrub cover is less than 20%, while grass cover ranges from 10 Hantam-Tanqua-Roggeveld is an area where three biomes meet, to 90%. Annuals are generally absent or have < 5% cover. Two namely the Fynbos, Succulent Karoo and Nama Karoo Biomes variations can be distinguished (Table 1c). The first variation (Rutherford & Westfall 1986), it is expected that some vegetation occurs at a low altitude (200–600 m) on deep sandy plains units are transitional between two biomes. For example, Rosenia without any rock cover. The second variation occurs at an oppositifolia is a prominent Mountain Renosterveld (Fynbos) altitude of 300 to 1 000 m above sea level on level ridges and related species (Van der Merwe et al. 2008), but it also occurs in plains, and in washes with no rock cover or on ridges with the Roggeveld Karoo subassociations 6.2, 6.3 and 6.4 (species a rock cover of 60 to 99%. The first variation of association 8 group M, Table 1b). Furthermore, Nama Karoo elements such as is dominated by Cladoraphis spinosa (species group W ) but Euryops multifidus are also found in these subassociations. Stipagrostis obtusa, Stipagrostis brevifolia and Stipagrostis ciliata (species group X) may be present, while the second variation While the relationship between the Tanqua Karoo grasslands lacks Cladoraphis spinosa (species group W ) and is characterised and grasslands of the Bushmanland, Nama Karoo Biome, by Stipagrostis species of species group X (Table 1c). was not investigated, common species with a high cover and constancy are found in both areas. Further studies of these Science and Conservation Area Protected African 9. Mesembryantheceae (HRp359) sp. Ceres Karoo Vygieveld affinities are necessary before the relationship between these two areas can be quantified. A single relevé was surveyed defining this plant association that has not been mapped separately but occurs in isolated patches Comparisons between previous maps published by Acocks within subassociation 7.3 in the southern part of the Tanqua (1953, 1988), Low and Rebelo (1998) and Mucina et al. (2005) Basin on shales of the Ecca Group and tillites of the Dwyka show similarities and some interesting dissimilarities. Since Group. This vygieveld (succulent veld) is found on level ridges the scale of these maps and the current map, and the reasons with a 99% cover of small stones. These small black stones and for compiling the maps, all differ; only generalisations can be brown chips are responsible for creating a unique microhabitat made when comparing the various maps. The area covered in which these succulents grow. The succulent and shrub cover in the present paper basically includes two of Acocks’s veld is less than 15%, with no grass or annual species present. Species types (Acocks 1953, 1988). Associations 4 (Escarpment Karoo), group Y characterises this association (Table 1c). 5 (Hantam Karoo) and 6 (Roggeveld Karoo) combined form part of the Western Mountain Karoo (Veld Type 28), which is 10. Pteronia (HRp118) sp. Tanqua Karoo Brackish Flats incorporated into Low and Rebelo’s Upland Succulent Karoo Association 10 occurs in combination with association 7.2 in the (Unit 56) (Low & Rebelo 1998), while associations 7 (Tanqua and Tanqua pan mosaic and is found in the central region of the Loeriesfontein Karoo) and 8 (Central Tanqua Grassy Plains) are Tanqua Karoo on alluvial deposits. The alluvial soils are brackish both included into Acocks’s Succulent Karoo (Veld Type 31) and depending on the salinity of the soils the vegetation differs. (Acocks 1953, 1988), which relates to Low and Rebelo’s Lowland http://www.koedoe.co.za Vol. 50 No. 1 pp. 160 - 183 KOEDOE 176

Original Research Van der Merwe, Van Rooyen & Van Rooyen

10 ...... 8 ...... 1 ...... 1 ...... 9 ...... 8 ...... 5 + ...... 3 ...... 7 ...... 8 ...... 3 ...... 5 ...... 8 ...... 3 ...... 1 ...... 7 ...... 1 ...... 6 ...... 8 ...... 3 ...... 0 ...... 4 ...... 2 ...... 9 ...... 3 ...... 2 ...... 1 ...... 2 ...... 1 ...... 6 ...... 1 ...... 1 ...... 5 ...... 1 ...... 1 ...... 4 ...... 1 ...... 1 ...... 8 . . 2 ...... 1 ...... 1 . . . + ...... 7 ...... 5 ...... 3 ...... 5 ...... 9 ...... 2 ...... 8 ...... 3 ...... 2 ...... 3 ...... 7 ...... 3 ...... 1 ...... 1 ...... 1 ...... 2 ...... 2 ...... 1 ...... 4 ...... 5 ...... 3 ...... 4 ...... 2 ...... 1 ...... 2 ...... 7 ...... 3 . . . . . + ...... 7 ...... 7 ...... 3 ...... 6 ...... 7 ...... 3 ...... 5 ...... 5 ...... 3 ...... 1 ...... 6 ...... 2 ...... 0 ...... 7 ...... 1 ...... 6 ...... 5 ...... 3 ...... 0 7.5 ...... 9 ...... 2 ...... 9 ...... 5 ...... 3 . + ...... 5 ...... 4 ...... 2 ...... 4 ...... 4 ...... 2 ...... 1 ...... 3 ...... 2 ...... 5 ...... 2 ...... 2 ...... 6 ...... 0 ...... 2 ...... 1 ...... 8 ...... 1 ...... 5 ...... 4 ...... 2 ...... 2 . . . . . 2 ...... 2 ...... 7 . . . . . 2 ...... 2 ...... 4 . . . . 8 ...... 3 ...... 8 ...... 3 ...... 2 . . + ...... 6 ...... 3 ...... 2 ...... 2 ...... 2 ...... 2 ...... 3 ...... 0 ...... 2 ...... 9 . . . . 7.4 . 8 ...... 1 ...... 7 ...... 8 ...... 1 ...... 4 ...... 5 ...... 1 . . 1 ...... 9 . . . . . 6 ...... 6 ...... 6 ...... 2 ...... 8 ...... 3 ...... 9 ...... 3 ...... 2 ...... 2 ...... 1 ...... 2 ...... 9 ...... 2 ...... 4 ...... 1 ...... 1 ...... 9 ...... 7 ...... r ...... 1 ...... 9 ...... 9 ...... 5 r ...... 1 . . + ...... 8 ...... 2 ...... 3 . + ...... 6 ...... 5 ble 1c ...... 3 ...... 3 ...... 5 a T ...... 2 ...... 3 ...... 5 ...... 1 ...... 6 ...... 8 . . r . . . . 3 ...... 2 ...... 5 ...... 3 ...... 9 ...... 5 7.3 ...... 2 ...... 9 ...... 4 ...... 2 ...... 8 ...... 4 ...... 1 ...... 8 ...... 7 ...... 1 ...... 4 ...... 6 ...... 1 ...... 3 ...... 6 ...... 1 ...... 0 ...... 1 ...... 4 ...... 4 . . . . 1 ...... 0 ...... 8 ...... 1 . + . + ...... 7 ...... 6 ...... + . . . . . 7 ...... 4 ...... 2 ...... 8 ...... 1 ...... 1 . . . a ...... + . . . . . 7 ...... 7 ...... 1 ...... 4 ...... 7 ...... 1 ...... 8 ...... 6 ...... 1 ...... 6 ...... 6 ...... 1 ...... 2 ...... 7 ...... 1 ...... 0 ...... 7 ...... 7.2 . 1 ...... 5 ...... 7 ...... 1 ...... 3 ...... 7 ...... 1 ...... 9 ...... 6 ...... 1 ...... 4 ...... 2 ...... 1 ...... 3 ...... 2 ...... 1 ...... 6 ...... 0 ...... 1 ...... 4 ...... 0 ...... 6 ...... 9 ...... 3 ...... 9 ...... 5 ...... 9 ...... Phytosociological table the of Succulent Karoo Biome related vegetation the of Hantam-Tanqua-Roggeveld subregion – associations 8, 9 and 7, 10 ...... 3 ...... 6 ...... 2 ...... 3 ...... 4 ...... 1 ...... 0 ...... 2 ...... 1 ...... 7 ...... 1 ...... 1 ...... 8 ...... 0 ...... 7.1 1 ...... 5 ...... 0 ...... 1 ...... 7 ...... 0 ...... 7 ...... 9 ...... 4 ...... 9 ...... 1 ...... 2 ...... 9 ...... (HR248) sp. sp. (HRp378) sp. spp. (HRp35) sp. s group A s group B s group C s group D s group E sp. sp. (HRp35) sp. elevé number R Specie Zygophyllum foetidum Specie Tylecodon wallichii Specie Pteronia glauca Specie Euphorbia decussata Specie Arctotis acaulis Montinia caryophyllacea Hermannia cuneifolia Didelta spinosa Lachenalia Pteronia (HRp382) sp. Phyllobolus Trachyandra falcata Asparagus Euphorbia hamata Lapeirousia montana Vygie (HRp382) sp. Rhus burchellii Stapelia Antimima cf. granitica Thesium cf. hystrix Othonna filicaulis Vygie (HRp378) sp. Euphorbia Crassula alpestris Sarcocaulon Babiana Lachenalia violacea Ruschia Vygie (HRp384) sp. African Protected Area Conservation and Science 177 KOEDOE Vol. 50 No. 1 pp. 160 - 183 http://www.koedoe.co.za 37

Succulent Karoo Biome related vegetation – Part 2 Original Research

10 ...... 8 ...... 1 ...... 1 ...... 9 ...... 8 ...... 5 ...... 3 ...... 7 ...... 8 ...... 3 ...... 5 ...... 8 ...... 3 ...... 1 . . . . + . . . 7 . + ...... 1 ...... 6 ...... 8 ...... 3 ...... 0 ...... 4 ...... 2 ...... 9 ...... 3 ...... 2 ...... 1 ...... 2 ...... 1 ...... 6 ...... 1 ...... 1 ...... 5 ...... 1 ...... 1 ...... 4 ...... 1 ...... 1 ...... 8 . 2 ...... 1 ...... 1 ...... 7 . . . . + . . . 5 ...... 3 ...... 5 ...... 9 ...... 2 ...... 8 ...... 3 ...... 2 ...... 3 ...... 7 ...... 3 ...... 1 ...... 1 ...... 1 ...... 2 ...... 2 ...... 1 ...... 4 . . . . + . . . 5 ...... 3 ...... 4 ...... 2 . . . . + . . . 1 ...... 2 ...... 7 . . + . . . . . 3 ...... + ...... 7 ...... 7 ...... 3 ...... 6 ...... 7 ...... 3 ...... 5 . . . + . . . . 5 ...... 3 ...... 1 ...... 6 ...... 2 ...... 0 ...... 7 ...... 1 ...... 6 ...... 5 ...... 3 ...... 0 7.5 ...... 9 ...... 2 ...... 9 ...... 5 ...... 3 ...... 5 ...... 4 ...... 2 ...... 4 ...... 4 . + ...... 2 ...... 1 ...... 3 ...... 2 ...... 5 ...... 2 ...... 2 ...... 6 ...... 0 . . . + . . . . 2 ...... + . . 1 ...... 8 ...... 1 ...... 5 ...... 4 ...... 2 ...... 2 . . . 2 ...... 2 ...... 7 ...... 2 ...... 2 ...... 4 ...... 8 ...... 3 ...... 8 ...... 3 ...... 2 ...... 6 ...... 3 ...... 2 ...... 2 ...... 2 ...... 2 ...... 3 ...... 0 ...... 2 ...... 9 ...... 7.4 . 8 . + ...... 1 ...... 7 ...... 8 ...... 1 . . . . . + ...... 4 ...... 5 ...... 1 ...... 9 . . + . . . . . 6 ...... 6 ...... 6 ...... 2 ...... 8 ...... 3 ...... + . . . . . 9 ...... 3 ...... 2 . . . . . + ...... 2 ...... 1 ...... 2 ...... 9 ...... 2 . . . . . r ...... 4 ...... 1 ...) ...... 1 ...... 9 ...... 7 t . + ...... 1 . + ...... 9 . + ...... + . . . 9 ...... 5 ...... 1 ...... 8 . . + . . . . . 2 ...... 3 ...... 6 . . . 1 . . . . 5 ...... 3 ...... 3 ...... 5 . . . + . . . . 2 ...... 3 ...... 5 ble (con 1c ...... 1 ...... 6 ...... 8 a ...... 3 ...... 2 ...... 5 T ...... 3 ...... 9 ...... 5 7.3 ...... 2 ...... 9 ...... 4 . + ...... 2 ...... 8 ...... 4 . . . + . . . . 1 ...... 8 ...... 7 ...... 1 . . . . . + ...... 4 ...... 6 ...... 1 ...... 3 ...... 6 ...... 1 ...... 0 ...... 1 ...... + . . . . . + . . . . . 4 ...... 4 ...... 1 . . . . . + ...... 0 ...... 8 ...... 1 ...... 7 ...... 6 ...... + . . . . . + . . . . . 7 ...... 4 . . . . . + . . 2 ...... 8 ...... 1 ...... 1 ...... 7 ...... 7 ...... 1 ...... 4 ...... 7 ...... 1 ...... 8 ...... 6 ...... 1 ...... 6 ...... 6 ...... 1 ...... 2 ...... 7 . . . + . . . . 1 ...... 0 ...... 7 ...... 7.2 1 ...... 5 ...... 7 ...... 1 ...... 3 ...... 7 ...... 1 ...... 9 ...... 6 ...... 1 ...... 4 ...... 2 ...... African Protected Area Conservation and Science and Conservation Area Protected African 1 ...... 3 . . . . + . . . 2 ...... 1 ...... 6 . . . . + . . . 0 ...... 1 ...... 4 ...... 0 . . . + ...... + . . . . . 6 ...... 9 . . . + ...... 3 ...... 9 . . . + ...... 5 ...... 1 9 . . . + ...... 3 ...... 6 ...... 2 ...... 3 . . . . + . . . 4 . . . + . . . . . 1 ...... 0 ...... 2 . . . + . . . . . 1 ...... 7 ...... 1 ...... 1 ...... 8 ...... 0 . . . 1 . . . . 7.1 . 1 ...... 5 ...... 0 ...... 1 ...... 7 ...... 0 . . . + ...... + . . . 7 ...... 9 ...... 4 ...... 9 . . . + ...... 2 ...... 9 ...... talum latipe cf. sp. sp. sp. s group F s group G s group H s group I s group J s group K s group L sp. a auriculifolia themum rysum obtusum elevé number R Specie Dorotheanthus Specie Bulbinella Specie Ehrharta calycina Specie Eriocephalus purpureus Specie Leipoldtia schultzei Specie Phyllobolus tenuiflorus Specie Pentaschistis patula Bulbine succulenta Lotononis Moraea Helich Drosan Berkheya fruticosa Cotula barbata Eriocephalus namaquensis Merxmuellera stricta Osteospermum acanthospermum Othonn Microloma sagittatum Senecio cardaminifolius Euphorbia mauritanica Oncosiphon grandiflorum Senecio cakilefolius Erodium cicutarium Lotononis hirsuta Dimorphotheca sinuata Heliophila collina Cyphia digitata Tetragonia microptera Gorteria diffusa Bromus pectinatus Rosenia glandulosa Zygophyllum pygmaeum http://www.koedoe.co.za Vol. 50 No. 1 pp. 160 - 183 KOEDOE 178

Original Research Van der Merwe, Van Rooyen & Van Rooyen

10 ...... 8 ...... 1 ...... 1 ...... 9 ...... 8 ...... 5 ...... 3 ...... 7 ...... 8 ...... 3 ...... 5 ...... 8 ...... 3 ...... 1 . . . . + . . 7 ...... 1 ...... 6 ...... 8 ...... 3 ...... 0 ...... 4 ...... 2 ...... 9 ...... 3 ...... 2 ...... 1 ...... 2 . . . . + . . . 1 ...... 6 ...... 1 ...... 1 ...... 5 ...... 1 ...... 1 ...... 4 . . . + . . . 1 ...... 1 ...... 1 8 . . 2 ...... 1 . 1 ...... 1 ...... 7 ...... 5 ...... 3 ...... 5 ...... 9 ...... 2 . + ...... 8 ...... 3 ...... 2 ...... 3 ...... 7 ...... 3 ...... 1 ...... 1 ...... 1 ...... 2 . . . . . + . 2 ...... 1 ...... + . . . . 4 ...... 5 ...... 3 ...... 4 ...... 2 ...... 1 ...... 2 . . . . + ...... 7 ...... + 3 ...... 7 . . . . . 1 . 7 ...... 3 ...... + . . . . 6 ...... 7 ...... 3 ...... + . . . . 5 ...... 5 ...... 3 ...... 1 ...... 6 ...... 2 ...... 0 ...... 7 ...... 1 ...... 6 ...... 5 ...... 3 ...... 0 7.5 ...... 9 ...... + 2 ...... + 9 ...... 5 ...... + 3 ...... 5 ...... 4 ...... 2 ...... 4 ...... 4 ...... 2 ...... 1 ...... 3 ...... 2 ...... 5 ...... 2 ...... 2 ...... 1 . . . . 6 ...... 0 ...... 2 ...... + 1 . + . . . . + 8 ...... 1 ...... 5 . . . + . . . 4 ...... 2 . . + ...... + ...... 2 . . . . 2 ...... 2 . a ...... 7 . + . . . . 2 + ...... 2 . . + ...... 4 ...... 8 ...... 3 ...... + . . 8 ...... 3 ...... + . 2 ...... + ...... 6 . . r . . . . 3 ...... 2 ...... 2 ...... 2 ...... 2 ...... 3 ...... 0 ...... 2 ...... 9 . . . . . 7.4 . 8 ...... 1 ...... + . 7 ...... 8 ...... 1 . . + ...... 4 ...... 5 ...... 1 . + ...... 9 ...... 6 ...... + + ...... 6 ...... 6 ...... 2 ...... 8 ...... 3 . . . . + ...... 9 ...... 3 . . . . . + . . 2 + ...... 2 ...... 1 ...... 2 ...... + ...... 9 . . . . a . . + ...... 2 . . . . . + ...... 4 . . + . . . . 1 ...) . . . + . . . . 1 ...... 9 . . . . + . . 7 t ...... + . 1 ...... 9 ...... 9 ...... 5 . . . . + . . . 1 ...... 8 ...... + 2 ...... 3 . . . . . + ...... 6 ...... 5 ...... 3 . . . . . + ...... 3 . . . . + . . 5 + ...... 2 . . . . . + ...... 3 ...... 5 ble (con 1c . . . . 1 . . . 1 ...... 6 ...... 8 a . . . . + . . . 3 ...... 2 ...... 5 T . . . . + . . . 3 . . . . . + ...... 9 ...... 5 7.3 ...... 2 . . . . . + ...... 9 ...... 4 . a ...... 2 . . . . . + ...... + . 8 . . . . + . . 4 ...... 1 ...... 8 ...... 7 . . . . 1 . . . 1 . . . . . + . . . + ...... 4 ...... 6 ...... 1 ...... 3 . . . . 1 . . 6 . . . . + . . . 1 ...... 0 ...... 1 . + ...... 4 ...... 4 ...... 1 . . . . . + ...... 0 ...... 8 ...... 1 . . . . + ...... + . . 7 . . . . + . . 6 r ...... + . . . . 1 ...... 7 ...... 4 . . . . + . . . 2 + . . . + ...... 8 ...... 1 ...... 1 . . . . + + ...... 7 ...... 7 ...... 1 . . . . + ...... 4 + ...... 7 ...... 1 1 . . . . + ...... 8 ...... 6 + ...... 1 a ...... 6 ...... 6 1 ...... 1 ...... 2 + ...... 7 ...... 1 ...... 0 ...... 7 ...... 7.2 . . 1 . . . . + . . . . 1 ...... 5 + ...... 7 ...... 1 ...... 3 + ...... 7 . . . . 1 . . 1 . . . . + + ...... 9 ...... 6 + ...... 1 . . . . + ...... + . . . . . 4 . . . . + . . 2 . . a . . . . . 1 . . . . + . + . . . . . + . . . . . 3 ...... 2 ...... 1 . . . . + ...... 6 ...... + 0 . . . . 1 . . . 1 . . . . 1 + ...... 4 ...... 0 ...... + ...... 6 ...... 9 ...... r . . . . + + ...... 3 ...... 9 ...... + ...... + + . 5 ...... 9 . . . . . + ...... + . + ...... 3 ...... 6 ...... 2 ...... + . . 3 ...... 4 . . . r . . . . . 1 . . 1 + + ...... 0 ...... 2 ...... + 1 . . . + + + ...... 7 . . . + . . . 1 + . . . + . . . . 1 . . . 1 . + ...... + . . 8 . . . . + . . 0 ...... 7.1 . 1 . . . + + ...... 5 . . . + . . + 0 . . . + . . . . . 1 . . . 1 . + ...... 7 ...... 0 ...... r + . + ...... + . . 7 ...... 9 . . . . a ...... 1 + + ...... 4 . . . + . . . 9 . . . + ...... a . + ...... 2 . . . . + . . 9 + ...... sp. (HR219) sp. sphaerocephala s group M s group N s group O s group P s group Q s group R spp. sp. (HRp171) sp. (HRp171) elevé number R Specie Eriocephalus pauperrimus Specie Pentzia incana Specie Ruschia robusta Specie Cephalophyllum Specie Gazania lichtensteinii Specie Drosanthemum Euryops annuus Eriocephalus ericoides Rosenia oppositifolia Pteronia glabrata Asparagus capensis Thesium lineatum Pentzia cf. Moraea Augea capensis Felicia Arctotheca calendula Eriocephalus microphyllus Euryops multifidus Pteronia glomerata Tribolium hispidum Felicia Galenia fruticosa Karroochloa schismoides Tetragonia fruticosa Pteronia villosa Gazania rigida Ruschia cradockensis Melolobium candicans Malephora crassa Leysera tenella Hirpicium alienatum Chrysocoma ciliata African Protected Area Conservation and Science 179 KOEDOE Vol. 50 No. 1 pp. 160 - 183 http://www.koedoe.co.za

Succulent Karoo Biome related vegetation – Part 2 Original Research

10 . . . . . 8 ...... 1 ...... + 1 ...... 9 . . . . . 8 ...... 5 ...... 3 ...... 7 ...... 8 ...... 3 ...... 5 ...... 8 ...... 3 ...... + ...... 1 . + . . . . . 7 ...... 1 ...... 1 ...... 6 ...... 8 ...... 3 ...... 1 ...... 0 ...... 4 . . . 1 . . . . . 2 ...... + ...... 9 ...... 3 . . . + . . . . . 2 ...... a ...... 1 ...... 2 . . . r . . . . . 1 ...... + ...... 6 ...... 1 . . . + . . . . . 1 ...... 1 . . . . . r . . . 5 ...... 1 ...... 1 ...... + ...... 4 ...... 1 . . . + . . . . . 1 ...... 1 ...... 8 . 2 ...... 1 ...... 1 ...... + ...... 7 . 3 . + . . . 5 ...... 3 . . . . . + ...... 5 ...... 9 ...... 2 + ...... 8 . + . . . . . 3 . . . 1 . . . . . 2 ...... 3 . . + + . . . 7 ...... 3 . + . + . . 1 ...... 1 . 1 . . . . . 1 . . . + . . . . + 1 . . . . . 1 + ...... 2 ...... 2 ...... 1 ...... 1 ...... 4 . . + . . . . 5 ...... 3 . . . . + 1 + . . . . . + ...... 4 ...... + . . . . 1 a ...... r . . . 2 ...... 1 + + . + . . . . . 2 . . . . + 1 ...... + 7 ...... 3 ...... + . . . . . b ...... + 7 . + . r . . . 7 ...... 3 . . . + + ...... 6 . 1 . . . . . 7 ...... 3 . . . 1 + ...... 5 ...... + 5 + ...... 3 . . . 1 + ...... r . . . 1 . + . + . . . 6 . . . + . . . . . 2 . . . a + ...... 0 ...... a 7 ...... 1 . . . + + ...... + . . . 6 . + . . . . r 5 ...... 3 . . . a ...... 0 . 7.5 . . r . . . 9 ...... 2 . . . a + . . + ...... 9 . . . r . . + 5 ...... 3 . . . + ...... 5 ...... 4 ...... 2 . . . . + ...... + . . . 4 ...... 4 ...... 2 . . . . + ...... + 1 . . . 1 . + . . . . . 3 ...... + 2 . . . . + ...... 1 5 . . . + . . . 2 ...... 2 . . + ...... + . . 6 . . . + . + . 0 ...... + . 2 . . + . + ...... 1 . . . + . + . 8 . . . . + + . . . 1 . . + . . . . + ...... 5 . . . + + . . 4 . . . + + + . + . 2 + + . . + . . + . . . . + . . . 2 . + . . . 2 ...... + . . 2 + 1 . . + ...... + . . . 7 ...... 2 . . . + + + . . . 2 + . . . + . a + . . . . + . . . 4 . . . + . . 8 . . . + + . . . . 3 . . + . + . . . . + . . . . + . 8 ...... 3 ...... 2 + . . a . . . + ...... 6 . . . + . . . 3 . . . . + + . . . 2 . . + . + ...... 1 . . . 2 . + . . + . 2 . . . . . + . . . 2 . + + a . . . r ...... + . 3 ...... 0 . + ...... 2 + . + + + ...... 1 . . . 9 . . . . + . 7.4 . 8 . . . . . + 1 . . 1 . . + ...... + . . . . 7 . . . . . + 8 ...... + . . 1 . . 1 . . . . + + ...... 4 . + . . . . . 5 . . + . + . + . . 1 . . + . + . . + . . . + + . . . 9 . . + . . . . 6 . . . + . . . . . + 1 ...... + 6 ...... 6 . . a . . 1 ...... 1 . . + . . + 2 . . . . . + . 8 . . . . + . . . . 3 . . + ...... + . . . 9 . . . . . + . 3 ...... 2 . . + + . . . . . 1 ...... 2 ...... 1 ...... 2 . + + + ...... + + . . . 9 . + . . . + ...... + . . 2 . + 1 . + . . . a . . + + . . . 4 . + + . . + . 1 ...) ...... 1 . 1 . . + ...... + + . . . 9 . 1 + . . . . . 7 t + + . . . 1 . + ...... + . . 1 1 . 1 . + . . . . + . 9 . . . . . + . . . . + . + . . . . . + . 1 a + . 9 + . . . . + . . 5 + . . + . + . . 1 . . + . + . . . + . + + . 1 . . 8 ...... 1 . 2 . . . . . + . . 3 . . . . 1 ...... 6 . . . . . + + . 5 . . . . + . . . 3 . . . + + ...... + . . . 3 ...... a . 5 . + . + . 1 . . 2 . . + ...... 3 . . . . . + . . 5 ble (con 1c + . . . . + . . 1 ...... + 1 . . + . . . . 6 . . . . + + . . 8 a ...... 3 . . . 1 + . . + ...... 2 ...... + . 5 T ...... 3 . . . 1 + ...... + . . . 9 ...... 5 7.3 . . . + . . . . 2 . . . . + ...... + . . . 9 ...... 1 . 4 . . . + . . . + 2 . . + . + ...... 8 . + . . . . + + 4 a + . + . 1 . . 1 . . . . + . . + . . + . + . . . 8 + . . . . . + . 7 . . . . . + . . 1 . . . . + ...... 4 . + ...... 6 ...... 1 . . . . + ...... + . . . 3 ...... 6 ...... 1 ...... 1 . . . 0 ...... 1 ...... + . + . . + . . + . + . . + 4 + ...... 4 . . . . + . . 1 . . . + . . . + ...... 0 . + . . . . 4 + 8 ...... 1 . . . 1 . . . + ...... 7 ...... 6 . . . . . + . . . . + . 1 . . + . . . . + . . . 7 + ...... 4 . . + . . . r 2 . . . b ...... 8 . . . . + . . . 1 ...... + 1 . . . 1 + . . + ...... 7 ...... + . 7 ...... 1 . . . 1 ...... 4 r . . . . . + . 7 . . . . . + . . 1 . . . + r ...... + . . . . 8 ...... 1 . 6 ...... 1 . . . 1 ...... r . . . 6 ...... 6 ...... 1 . . . . 1 . + . . . . . + . . + 2 ...... 7 ...... + 1 . . . . + ...... + + . . 0 ...... + 7 . . . . . a 7.2 . . 1 . . . 1 + ...... + . . . 5 . . . . . r . . 7 . . . . . + . . 1 . . . . + ...... + . . . . 3 ...... 7 . . . + . + . 1 . . . . + . . . . . + + + . . + 9 ...... r . 6 . . . . . + . . 1 ...... + ...... 4 . . . . . r . . 2 . . . . r . . African Protected Area Conservation and Science and Conservation Area Protected African 1 . . . . 1 . . + ...... 3 . . . . . + . + 2 . + . + + . . . 1 . . + . + ...... + . . . 6 ...... 0 ...... 1 . . . . + ...... + + . . . 4 . . + . . . . . 0 . + + ...... 1 . . . + . . . + . . . + 6 . . . . + . . . 9 . . . . . + . + . . . + a ...... + . . . 3 + ...... + 9 . . + . . + ...... + . . . . 5 1 + . . . . . + 9 . . + + ...... 1 . . . + ...... + + 3 . . . . + . . . 6 . . . . + 1 . . . 2 . . . 1 ...... + + . . . 3 . a . . . . . + 4 . . . + + . . + . 1 . . . + ...... + . . . . 0 + . . . . . 1 . 2 . . + ...... 1 . . . + ...... + 7 ...... 1 ...... 1 ...... + . . + . . . . . 8 ...... 0 . . . . + . . r 7.1 . 1 . . 1 + ...... + . . + 5 . . + . . . . . 0 . + + + . . . . . 1 . . . . . + ...... + 7 . . . r . . . . 0 . . + . + . . + ...... + . . . . . 7 ...... r . 9 ...... + . . + . . + + + . . . 4 + . . . . . 1 . 9 . . + ...... + . . + + . . . . 2 . + ...... 9 ...... vaginatus (HR217) sp. cf. (HRp65) sp. apiculata (p91) sp. (p91) sp. cf. sp. s group S s group T s group U s group V s group W s group X sp. spp. rostis brevifolia spp. Lampranthus otzenianum elevé number R Specie Braunsia Specie Atriplex lindleyi Specie Ruschia intricata Specie Aridaria noctiflora Specie Cladoraphis spinosa Specie Stipagrostis obtusa Zygophyllum Tripteris sinuata Exomis microphylla Salsola Oxalis Foveolina dichotoma Salsola tuberculata Salsola aphylla cf. Geophytic spp. Osteospermum pinnatum Stipag Eriocephalus decussatus Lycium Psilocaulon Pteronia Pteronia pallens Mesembryanthemum guerichianum Brownanthus Phyllobolus Felicia australis Galenia sarcophylla Zygophyllum retrofractum Drosanthemum Pentzia spinescens Galenia africana http://www.koedoe.co.za Vol. 50 No. 1 pp. 160 - 183 KOEDOE 180

Original Research Van der Merwe, Van Rooyen & Van Rooyen

Succulent Karoo (Unit 57) (Low & Rebelo 1998). Mucina et al. 10 + 8 + 1 . . + 1 1 . . . 9 . 8 . 5 . 1 . 3 . . 1 . (2005) mapped approximately ten vegetation units within the . 7 . 8 + . . 3 . . . . . 5 . 8 + . . 3 . . . . study area. There is good agreement between association 4 . 1 . 7 . . . 1 . . . . (Escarpment Karoo) and the Tanqua Escarpment Shrubland . 6 . 8 . . . 3 . . . . . 0 . 4 . . . 2 . + . + (SKv 4) of Mucina et al. (2005). In general, the delineation of the . 9 . 3 . . . 2 . . . a . 1 . 2 . . . 1 . . . . Hantam Karoo (association 5), Roggeveld Karoo (association 6) . 6 . 1 . . . 1 . . . . and Tanqua Karoo (association 7) in this study is more restricted 5 . 1 . . . 1 . . . . . 4 . 1 . . . 1 . . . . . than that of the Hantam Karoo (SKt 2), Roggeveld Karoo (SKt 3) 8 2 . 1 . . . 1 . . . . . 7 . 5 . . . 3 . . . . . and Tanqua Karoo (Skv 5) of Mucina et al. (2005). Comparisons 5 . 9 . . . 2 . . . b . of the Acocks (1953, 1988), Low and Rebelo (1998) and Mucina 8 . 3 . . . 2 . + . . . 3 . 7 . . . 3 . . . a . et al. (2005) maps with the map presented in this article reveal 1 . 1 . . . 1 . . . . . 2 . 2 . . . 1 . . . r

. that the maps of Acocks (1953, 1988) and Low and Rebelo (1998) 4 . 5 . . . 3 . . . + . differ appreciably in the Van Rhynsdorp/Doorn River region . 4 ...... 2 . 1 . . . 2 . . . 1 . 7 while the Mucina et al. (2005) map is more closely related to . 3 ...... 7 . 7 . . . 3 . . . . . 6 the map presented in this paper. The differences in this region . 7 . . . 3 . . . . . 5 require additional study to determine the boundaries of each . 5 . . . 3 . . . . . 1 . 6 . . . 2 . . . . . 0 different vegetation type since the region is situated in the . 7 . . . 1 . . . . . 6 . 5 . . . 3 . . . . . 0 transition between the Tanqua Karoo and Van Rhynsdorp 7.5 . 9 . . . 2 . . . . . 9 Succulent Karoo and the Fynbos Biome. . 5 . . . 3 . . . . . 5 . 4 . . . 2 . . . . . 4 . 4 . . . 2 . . . . . 1 . 3 . . . 2 . . . . . 5 In the Sutherland area, the region is classified by Acocks (1953, . 2 . . . 2 . . . . . 6 1988) as Western Mountain Karoo and by Low and Rebelo . 0 . . . 2 . . . . . 1 . 8 . . . 1 . . . . . 5 (1998) as Upland Succulent Karoo, yet the present study found . 4 . . . 2 . . . . . 2 . 2 . . . 2 . . . . . 7 this vegetation to be more closely related to the Mountain . 2 . . . 2 . . . . . 4 Renosterveld vegetation of the Fynbos Biome, and is described . 8 . . . 3 . . . . . 8 3 . . . 2 . . . . . 6 . by Van der Merwe et al. (2008) as association 1. Also, north and 3 . . . 2 . . . . . 2 . 2 . . . 2 . . . . . 3 . east of Sutherland, parts of the Mucina et al. (2005) Roggeveld 0 . . . 2 . . . . . 9 . Karoo (SKt3) were incorporated into the Mountain Renosterveld 7.4 8 . . . 1 . . . . . 7 . 8 . . . 1 . . . . . 4 . (association 1 in Van der Merwe et al. 2008). 5 . . . 1 . . . . . 9 . 6 ...... + . 6 . 6 ...... 2 . Two major threats to the vegetation in the study area were 8 . . . 3 . . . . . 9 . 3 . . . 2 . . . . . 2 . identified by the farming community and the Northern Cape 1 . . . 2 . . . . . 9 . Department of Agriculture, and through personal observation . . . 2 . . . . . 4 . 1 ...) . . . 1 . . . . . 9 . 7 t during field surveys. The first threat is that of invasive alien ...... 1 . 9 ...... 9 . 5 species, especially Prosopis species. Prosopis glandulosa was . . . 1 . . . . . 8 . 2 . . . 3 . . . . . 6

. introduced to the Karoo to aid in fodder production for small 5 . . . 3 . . . . . 3 . 5 stock in the drier months of the year when little natural . . . 2 . . . . . 3 . 5 ble (con 1c . . . 1 . . . . . 6 . 8 vegetation is available (Zimmermann 1991). The pods of these a . . . 3 . . . . . 2 . 5 T trees are high in protein and many farmers rely on them to . . . 3 . . . . . 9 . 5 7.3 . . . 2 . . . . . 9 . 4 carry their stock through the dry months. However, more than . . . 2 . . . . . 8 . 4 . . . 1 . . . . . 8 . 7 one species of Prosopis was introduced and these have now . . 1 . . . . . 4 . 6 . hybridised. These hybrids, with their hybrid strength, have . . 1 . . . . . 3 . 6 . . . 1 . . . . . 0 . 1 . invaded large tracts of land and are now a serious threat in ...... 4 . 4 . . . 1 . . . . . 0 . 8 . many areas. 1 . . . . . 7 . 6 ...... 7 . 4 . . . 2 . . . . . 8 . 1 . . . Other alien invasive species include Nerium oleander and 1 . . . . . 7 . 7 . . . 1 . . . . . 4 . 7 . . . Nicotiana glauca, which are usually restricted to drainage lines. 1 . . . . . 8 . 6 . . . These species are a serious problem in some areas, although 1 . . . . . 6 . 6 . . . 1 . . . . . 2 . 7 . . . they are not as widespread as the Prosopis species that threaten 1 . . . . . 0 . 7 . . . 7.2 1 . . . . . 5 . 7 . . . not only drainage lines but also the natural vegetation between 1 . . . . . 3 . 7 . . . the drainage lines. 1 . . . . . 9 . 6 . . . 1 . . . . . 4 . 2 . . . 1 . . . . . 3 . 2 . . . 1 . . . . . 6 . 0 . . . The second serious threat is that of less than ideal farming 1 . . . + . 4 . 0 . . . practices. Due to a lack of infrastructure, especially fencing, . . . . . 6 . 9 ...... 3 . 9 . . . optimal farm management is not implemented. The main . . . . . 5 . 9 ...... 3 . 6 . . . reason for this is that farms in the region have a low income . 2 . . . . 3 . 4 . . . as a result of the unfavourable and harsh environmental . 1 . . . . 0 . 2 . . . . 1 . . . . 7 . 1 . . . conditions. Additionally, the monetary value of the land is low . 1 . . . . 8 . 0 . . . 7.1 . 1 . . . . 5 . 0 . . . and the cost of infrastructure so high that it is not financially . 1 . . . . 7 . 0 . . . viable for a farmer to invest too much in infrastructure as it . . . . 7 . 9 ...... 4 . 9 . . . . will not be possible to recover these costs. There is definitely . . . . 2 . 9 . . . . willingness amongst farmers for improved farm management and infrastructure development; however, their financial means usually do not allow it. Although damage can happen fast, recovery in the Karoo is very slow, because it depends upon unpredictable rainfall events (Esler et al. 2006).

(HRp386) sp. Apart from the transformation as a result of alien invasive (HRp118) sp. (HRp118) (HRp118) sp. (HRp118) s group Y s group Z (HRp118) sp. (HRp118) vegetation, transformation of natural vegetation into flood- (HRp118) sp. (HRp118) irrigated lands next to the major drainage systems has taken

elevé number place in the past. Few of these lands, especially in the Tanqua R Specie Mesembryanthemaceae 1 (HRp359) sp. Specie Pteronia * Non-diagnostic species are excluded * HR collection code and numbers are included for future reference, if necessary* Specimens inadequate for identification yet different from species that could* sp. = one be speciesidentified in a taxonomic are indicated group with a plot number(p…) * spp. = more than one species in a taxonomic group. These species, even although they are grouped together, are included in the table since they occur in different species groups. However, they are not used in the descriptions in the text. Stipagrostis ciliata Pteronia Salsola Lycium Zygophyllum Mesembryanthemaceae 2 (HRp359) sp. Blepharis pruinosa African Protected Area Conservation and Science 181 KOEDOE Vol. 50 No. 1 pp. 160 - 183 http://www.koedoe.co.za

Succulent Karoo Biome related vegetation – Part 2 Original Research

Karoo, are still utilised for cropping; they now lie barren, Agricultural Research Council. 2003. Land type map 3320 with little vegetation cover to combat erosion, or the lands are Ladismith. Pretoria, Institute for Soil, Climate and Water. infested by invasive species, particularly Prosopis species. Born, J., Linder, H.P. and Desmet, P. 2007. The Greater Cape Floristic Region. Journal of Biogeography, 34: 147–162. Formal conservation within the Hantam-Tanqua-Roggeveld CEPF, 2003. Ecosystem Profile: The Succulent Karoo hotspot, subregion is limited. There are two local municipal reserves, Namibia and South Africa. Critical Ecosystem Partnership the Nieuwoudtville W ildflower Reserve (115 ha) and the Fund report. Akkerendam Nature Reserve (230 ha) that qualify as formally Conservation International. 2006. http://www. protected areas. However, most of the Akkerendam Nature biodiversityhotspots.org [Accessed 20 February 2006]. Reserve is located on the plateau of the Hantam Mountain Council for Geoscience. 1973. Geological map 3218 Clanwillliam. within the Mountain Renosterveld, Fynbos related, vegetation Pretoria, Council for Geoscience. of association 3 (Van der Merwe et al. 2008). The Nieuwoudtville Council for Geoscience. 1983. Geological map 3220 Sutherland. W ildflower Reserve is situated in the Nieuwoudtville mosaic Pretoria, Council for Geoscience. consisting of vegetation units 2.1.1, 2.1.4, 2.2. (Van der Merwe Council for Geoscience. 1989. Geological map 3120 W illiston. et al. 2008) and 5.1 (Fig.1), which are a combination of Mountain Pretoria, Council for Geoscience. Renosterveld (Fynbos related vegetation) and Succulent Karoo Council for Geoscience. 1991. Geological map 3320 Ladismith. vegetation. The conservation status of the Tanqua Karoo has Pretoria, Council for Geoscience. increased considerably with the expansion of the Tankwa Council for Geoscience. 1997. Geological map 3319 Worcester. Karoo National Park over the last few years. This National Pretoria, Council for Geoscience. Park currently protects 92 495 ha of land; however large tracts Council for Geoscience. 2001. Geological map 3118 Calvinia. thereof are highly degraded. A few private nature reserves and Pretoria, Council for Geoscience. conservancies are also found in the Tanqua Karoo, but these Council for Geoscience. 2008. Electronic data supplied by the are generally located at the southern end of the Tanqua Basin, Council for Geoscience, Silverton, Pretoria. closer to Ceres. Du Plessis, H.M. 1987. Land Types of the maps 2816 Alexander Bay, 2818 Warmbad, 2916 Springbok, 2918 Pofadder, 3017 In conclusion, this project aimed to classify and describe the Garies, 3018 Loeriesfontein. Memoirs on the Agricultural Natural Resources of South Africa, vegetation units in the Hantam-Tanqua-Roggeveld subregion 9: 1–538. Esler, K.J., Milton, S.J. & Dean, W.R.J. 2006. Karoo veld ecology and using species composition, environmental parameters and management. Pretoria, Briza Publications. vegetation unit relationships to one another to map their Germishuizen, G. & Meyer, N.L. (eds.). 2003. Plants of southern geographical distribution. This map could serve as a basis to Africa: An annotated checklist. Strelitzia 14. Pretoria: aid future planning and biodiversity conservation through National Botanical Institute. sustainable land use practices to reduce the impact on the Hennekens, S.M. & Schaminee, J.H.J. 2001. TURBOVEG, a land. comprehensive data base management system for vegetation data. Journal of Vegetation Science, 12: 589–591. ACKNOWLEDGEMENTS Hill, M.O. 1979. TWINSPAN – A FORTRAN program for arranging multivariate data in an ordered two-way table by classification The authors would like to thank the Critical Ecosystem of the individuals and attributes. Ithaca, NY, Ecology & Partnership Fund (CEPF) through the SKEP (Succulent Karoo Systematics: Cornell University. Ecosystem Plan/Program) initiative for funding the project. Hilton-Taylor, C. 1994. Western Cape Domain (Succulent The Critical Ecosystem Partnership Fund is a joint initiative of Karoo). In Davis, S.D., Heywood, V.H. & Hamilton, A.C. Conservation International, the Global Environmental Facility, (eds.), Centres of plant diversity, A guide and strategy for their the Government of Japan, the MacArthur Foundation and conservation, volume 1, Cambridge: IUCN Publications Unit, the World Bank. Its fundamental goal is to ensure that civil pp. 201–203. society is engaged in biodiversity conservation. CapeNature, Low, A.B. & Rebelo, A.G. 1998. Vegetation of South Africa, Lesotho Department of Tourism, Environment and Conservation and Swaziland. Pretoria: Department of Environmental (Northern Cape) as well as SANParks are thanked for the Affairs and Tourism. necessary permits and permission to conduct this research. Mucina, L., Rutherford, M.C. & Powrie, L.W. (eds). 2005. The Council for Geoscience is thanked for providing geological Vegetation map of South Africa, Lesotho and Swaziland, data of the study area. The assistance of Hennie van den Berg of 1 : 1 000 000 scale sheet maps. Pretoria, South African Iris International for compiling the vegetation map is gratefully National Biodiversity Institute. acknowledged. Nelder V.J., W ilson, B.A., Thompson, E.J. & Dillewaard, H.A. 2005. Methodology for Survey and Mapping of Regional Ecosystems and Vegetation Communities in Queensland. African Protected Area Conservation and Science and Conservation Area Protected African REFERENCES Version 3.1. Updated September 2005. Brisbane, Queensland Herbarium: Environmental Protection Agency. Acocks, J.P.H. 1953. Veld types of South Africa. Memoirs of the Rubidge, B.S. & Hancox P.J. 1999. The Karoo – a palaeontological Botanical Survey of South Africa, 28: 1–192. wonderland. In: Viljoen, M.J. & Reimold W.U. An introduction Acocks, J.P.H. 1988. Veld types of South Africa. 3rd ed. Memoirs to South Africa’s geological and mining heritage. Pretoria: of the Botanical Survey of South Africa, 57: 1–146. Published by the Geological Society of South Africa and Agricultural Research Council. 1986a. Land type map 3018 Mintek. Loeriesfontein. Pretoria, Institute for Soil, Climate and Rubin, F. 1998. The physical environment and major plant Water. communities of the Tankwa Karoo National Park. Koedoe, Agricultural Research Council. 1986b. Land type map 3220 41: 61–94. Sutherland. Pretoria, Institute for Soil, Climate and Water. Rutherford, M.C. & Westfall, R.H. 1986. Biomes of Southern Agricultural Research Council. 1995. Land type map 3118 Africa. An objective characterisation. Memoirs of the Botanical Calvinia. Pretoria, Institute for Soil, Climate and Water. Survey of South Africa, 54: 1–98. Agricultural Research Council. 1999a. Land type map 3120 South African National Biodiversity Institute. 2006. h t t p :// W illiston. Pretoria, Institute for Soil, Climate and Water. www.sanbi.org/consfarm/ [Accessed 2 October 2006]. Agricultural Research Council. 1999b. Land type map 3218 SA Weather Bureau 1998. Climate of South Africa. Climate Clanwilliam. Pretoria, Institute for Soil, Climate and statistics up to 1990. W B 42. Pretoria, Government Printer. Water. Snijman, D. & Perry, P. 1987. A floristic analysis of the Agricultural Research Council. 2002. Land type map 3319 Nieuwoudtville W ild Flower Reserve, north-western Cape. Worcester. Pretoria, Institute for Soil, Climate and Water. South African Journal of Botany, 53: 445–454. http://www.koedoe.co.za Vol. 50 No. 1 pp. 160 - 183 KOEDOE 182

Original Research Van der Merwe, Van Rooyen & Van Rooyen

Van der Merwe, H., Van Rooyen, M.W. & Van Rooyen, N. 2008. Werger, M.J.A. 1974. On concepts and techniques applied in the Vegetation of the Hantam-Tanqua-Roggeveld subregion, Zürich-Montpellier method of vegetation survey. Bothalia, South Africa. Part 1. Fynbos Biome related vegetation. 11: 309–323. Koedoe, 50: 61–71. Zimmermann, H.G. 1991. Biological control of mesquite, Prosopis Van W yk, A.E. & Smith, G.F. 2001. Regions of Floristic Endemism spp. (Fabaceae), in South Africa. Agriculture, Ecosystems and in Southern Africa: A review with emphasis on succulents. Environment, 37: 175–186. Pretoria, Umdaus Press. African Protected Area Conservation and Science 183 KOEDOE Vol. 50 No. 1 pp. 160 - 183 http://www.koedoe.co.za

Chapter 6

Plant diversity in the Hantam-Tanqua-Roggeveld, Succulent Karoo, South Africa: Species-area relationships

Abstract

The Hantam-Tanqua-Roggeveld subregion is part of the Succulent Karoo hotspot of biodiversity which stretches along the southwestern side of South Africa and Namibia. Forty Whittaker plots were surveyed in the spring of 2005, in eight vegetation associations, to gather diversity data for the Hantam, Tanqua Karoo and Roggeveld areas. Seven plot sizes were used to construct species-area curves using three different models namely: the untransformed linear function, the power function and the exponential function. In general, the power and exponential functions produced a more significant fit to the data than the untransformed linear function.

To illustrate the variation in species-area curves and species richness across the landscape, a transect through the study area is discussed. The transect stretches eastwards from the Tanqua Karoo across the escarpment into the Roggeveld and crosses five different vegetation associations. Differences between associations were found in species richness in the 1000 m2 plots. Each association also produced species-area curves with their own characteristics. Slope values for the samples within an association did not differ significantly, although the intercept value often did. Comparisons between associations along the transect revealed significant differences in the slope value between the associations, except for the Dicerothamnus rhinocerotis Mountain Renosterveld which did not differ significantly from the associations bordering it on either side.

Keywords: biodiversity hotspot, Hantam, Roggeveld, species-area curves, species richness, Tanqua Karoo, Whittaker plots

6.1 Introduction

One of the few laws that has emerged in ecology is that there is an increase in the number of species as the sampled area increases (Tjørve 2003). This pattern was already recognised by De Candolle in 1855 (in Scheiner 2003) and formalised as the species-area curve in the early twentieth century (Scheiner 2003). Recent years have seen a renewed interest in species- area curves. In particular it holds promise as a tool in conservation biology (Palmer & White 1994, Tjørve 2003). Species richness, or the number of species, is currently the most widely used diversity measure (Stirling & Wilsey 2001) because it is relatively easy to measure, is comparable across communities, and is well understood by researchers, managers, and the public (Hellman & Fowler 1999). However, species richness per se does not imply any standardisation of sampling area (Whittaker 1977, Whittaker et al. 2001). By adding a spatial scale, species-area curves can provide more information on the nature of the differences between vegetation types in different geographical areas than mere measures of species richness. Amongst other applications, species-area curves have been applied successfully to examine the effects of habitat loss on species diversity (Pimm et al. 1995), the effect of invasions on species diversity (Vitousek et al. 1996) and the identification of hotspots (Veech 2000).

In spite of the seeming simplicity of the relationship between species number and area there are numerous ways to construct species-area curves (Palmer & White 1994, Tjørve 2003) and an analysis of 100 species-area curves by Connor and McCoy (1979) indicated that there was no single best-fit model. This statement is supported by He and Legendre (1996) whose results suggest that for any particular data set a ‘best’ model can be constructed but that there is no model that is universally best, and that all depends on sampling scale.

The Succulent Karoo Biome is a semi-desert region that occupies 111 000 km² on the arid fringes of South Africa’s Cape Floristic Region (Mucina et al. 2006). Despite a general lack of structural diversity, plant species diversity at both the local and regional scales in the Succulent Karoo is undoubtedly the highest recorded for any arid region in the world (Cowling et al. 1989). Approximately 30 percent of the world’s succulent species are located in this small area (Van Jaarsveld 1987, Smith et al. 1993). On account of its exceptional biodiversity, this region was the first arid land to qualify as a global biodiversity hotspot (Cowling & Hilton- Taylor 1994).

In 2002 the Succulent Karoo Ecosystem Plan (SKEP) was launched to determine the state of biodiversity in this hotspot. For management purposes, the SKEP planning domain subdivided the Succulent Karoo into four subregions, of which the Hantam-Tanqua- Roggeveld constituted one. However, little information was available on the biodiversity of the subregion that could be used for future planning, conservation and development (Cilliers et al.

2002, Critical Ecosystem Partnership Fund 2003). As the first step to gather botanical information the vegetation associations and subassociations in the subregion were identified, classified and described (Van der Merwe et al. 2008a, 2008b). The next step was to gain information on the ecological properties of each of the vegetation units to improve our understanding, conservation and management of this unique arid area. This paper reports on the analysis of patterns of plant diversity by means of species richness and species-area curves in the Hantam-Tanqua-Roggeveld subregion.

6.2 Study area

The Succulent Karoo stretches along the southwestern side of South Africa and Namibia (Figure 6.1). The Hantam, Tanqua and Roggeveld areas lie within the predominantly winter rainfall region of the Succulent Karoo in the Northern and Western Cape Provinces of South Africa. In the south the Hantam-Tanqua-Roggeveld subregion stretches from the Ceres Karoo, where the Swartrug and Bontberg mountains meet, northwards into the Tanqua basin. The eastern border includes the Hantam, Roggeveld, Komsberg and Klein Roggeveld mountains to just southwest of Fraserburg, while the Cederberg and Bokkeveld Mountains, falling outside of the study area, form the western boundary which stretches to just north of Loeriesfontein. Rainfall ranges from 25 mm per year in parts of the Tanqua Karoo to 467 mm per year, the annual maximum recorded for Sutherland, on the Roggeveld plateau (Weather Bureau 1998). Although the rain falls mainly in winter it does include a few summer thunderstorms.

The physical geography of the region differs greatly. From the level plains of the Tanqua Karoo the landscape rises steeply to the escarpment formed by the Roggeveld, Komsberg and Nuweveld Mountains, from where it stretches eastwards along the Roggeveld plateau. The Hantam is characterised by a gently undulating to a steeply rolling topography. The altitude across the study area varies from approximately 290 m above sea level in the Tanqua Karoo to about 1800 m above sea level on the Roggeveld plateau.

The soils of the Tanqua Karoo are shallow lithosols that often include a desert pavement and deep unconsolidated deposits in the alluvial parts. Hantam Karoo soils to the west of Calvinia are composed of shallow lithosols and duplex soils, but where dolerite occurs the soils are red structured and red vertic clays. The mountains of the great escarpment, as well as the Hantam Mountain near Calvinia, are comprised of shallow stony lithosols and duplex soils in the occasional lowlands (Francis et al. 2007).

The study area has a high level of endemism. Hilton-Taylor (1994) identified three centres of endemism namely the Western Mountain Karoo, Roggeveld and Tanqua Karoo within the

study area whereas Van Wyk and Smith (2001) recognised the Hantam-Roggeveld as one of the 13 principal regions and centres of plant endemism in southern Africa.

Figure 6.1 Location of the eight plant associations identified in the Hantam-Tanqua- Roggeveld subregion of the Succulent Karoo, South Africa (after Van der Merwe et al. 2008a, 2008b).

The Hantam-Tanqua-Roggeveld includes three of Acocks’s (1953) vegetation units namely: Mountain Renosterveld, Succulent Karoo and Western Mountain Karoo (Acocks 1988). Acocks termed his vegetation units veld types which he defined as units having the ‘same

farming potentialities’. His vegetation map and subsequent hierarchy was an intuitive, expert based system (Rutherford et al. 2003) and in the Hantam, Tanqua and Roggeveld regions, Acocks’s three veld types can clearly be distinguished. Acocks’s Mountain Renosterveld is equivalent to the Escarpment Mountain Renosterveld of Low and Rebelo (1996), and his Succulent Karoo and Western Mountain Karoo form part of the Lowland Succulent Karoo and Upland Succulent Karoo (Low and Rebelo 1996) respectively. Mucina et al. (2005) and Mucina et al. (2006) recognised twelve vegetation types in the study area. The classification of the vegetation used in this paper follows Van der Merwe et al. (2008a, 2008b) who recently classified and mapped the subregion into eight large vegetation associations and 25 subassociations (Figure 6.1). These associations are as follows: 1. Rosenia oppositifolia Mountain Renosterveld, 2. Dicerothamnus rhinocerotis Mountain Renosterveld, 3. Passerina truncata Mountain Renosterveld, 4. Pteronia glauca - Euphorbia decussata Escarpment Karoo, 5. Eriocephalus purpureus Hantam Karoo, 6. Pteronia glomerata Roggeveld Karoo, 7. Aridaria noctiflora Tanqua and Loeriesfontein Karoo, and 8. Stipagrostis obtusa Central Tanqua Grassy Plains.

In this paper the associations have also been grouped at a higher hierarchical level into three vegetation groups. The three Mountain Renosterveld vegetation associations are grouped together and called Mountain Renosterveld, the Escarpment Karoo, Hantam Karoo and Roggeveld Karoo are collectively termed Winter Rainfall Karoo and the Tanqua and Loeriesfontein Karoo together with the Central Tanqua Grassy Plains are termed Tanqua Karoo.

6.3 Materials and methods

Whittaker’s plant diversity plot technique (Shmida 1984) was used to conduct field surveys. The main reasons for using the Whittaker plot technique were to facilitate comparisons with other diversity studies as well as the ease with which the plot can be set up and sampled relative to other techniques such as the Modified Whittaker Nested Vegetation Sampling Technique of Stohlgren et al. (1995). Stohlgren et al. (1995) found that their long-thin plot design was cumbersome for field crews, and the increase in perimeter to area ratio of the subplots made it more difficult for field crews to identify which plants should be included or excluded. Also, Wilson and Shmida (1984) concluded that the Whittaker method came close to fulfilling four criteria of ‘good’ performance of beta diversity measures. The only modification of the methodology described by Shmida (1984) related to the field form and notations used on the field form. A separate column for each size quadrat was provided for on the field form. Additionally, the vegetation of the two 5 m² quadrats were noted in two

separate columns and the 10 m x 10 m square was separated into two 5 m x 10 m rectangles and the species were noted apart from one another in two columns on the field form. The columns thus read as follows: ten 1 m², two 5 m², two 50 m² and one 1000 m² columns. The presence of each species encountered in a quadrat was noted within each column and a percentage cover value given for each species in the 1000 m² quadrat.

By ensuring that each column contained a list of all species present in that quadrat, more freedom was gained to calculate the number of species present in a quadrat of a different size than actually measured. For example, if the number of species for a 15 m² quadrat is required, the total number of different species across, for example, the ten 1 m² and a one 5 m² quadrat could be tallied.

Forty sample plots were surveyed across the Hantam, Tanqua and Roggeveld regions in the different vegetation associations prevalent in the area (Van der Merwe et al. 2008a, 2008b). Environmental data such as altitude, aspect, slope, position on the slope, geology, topography, percentage stone and stone size, soil type and colour, drainage, erosion, trampling and soil compaction were noted at each site.

The total species number for seven plot sizes (1 m², 5 m², 10 m², 20 m², 50 m², 100 m² and 1000 m²) were determined by using the mean of the ten 1 m² plots for the 1 m² plot, the mean of the two 5 m² plots for the 5 m² plot, mean of the total of ten 1 m² plots and the total of two 5 m² plots for the 10 m² plot, total of the ten 1 m² and the two 5 m² plots for a 20 m² plot, mean of the two 50 m² plots for a 50 m² plot, the total of the two 50 m² plots for a 100 m² plot and the total for the 1000 m² plot. These seven plot sizes were used to construct Type II species- area curves (Scheiner 2003, 2004) for each of the 40 plots sampled using three different functions namely: 1) the untransformed linear function between species richness (S) and area (A): S = zA + c where c and z are constants for the slope and y-intercept respectively; 2) the power function, typically expressed as the log transformation: log S = log c + z log A, and 3) the exponential function, expressed as a semilog function: S = z log A + c (Veech 2000). A fourth function, the logistic function, was not used in the study since the whole community was not sampled. If sampling covers the whole of a community, the logistic is expected to be the best model to describe the species-area relationship (He & Legendre 1996).

For each site the equation, r-value and p-value (significance) were determined using the STATISTICA computer package (StaSoft, Inc. Version 7 and Version 8, 2300 East 14th Street, Tulsa, OK 74104). Upon testing for normality and finding that the distribution was normal, an Analysis of Variance (ANOVA, Bonferroni’s post-hoc test) was conducted to compare the

slopes of the species-area curves between the associations and three vegetation groups for the exponential function. The statistical significance of the differences between the slope values and intercepts of the exponential curves were analysed by an Analysis of Covariance (Quinn & Keough 2002) with GraphPad Prism 4.03 for Windows (GraphPad software, San Diego, California, USA, www.graphpad.com).

To illustrate the variation in species-area curves produced using the exponential function and species richness in relation to environmental data a transect of ten survey plots running from west to east through the study area was compiled. This transect stretches eastwards from the Tanqua Karoo, across the Roggeveld escarpment onto the Roggeveld plateau.

6.4 Results and discussion

Whittaker (1977) contended that patterns of plant diversity can be elucidated only by systematic surveys and by sampling at multiple spatial scales. Multiple scale sampling of vegetation allows for: 1) evaluating the influence of spatial scale on local species richness patterns (Podani et al. 1993); 2) improved comparisons of community richness compared with single scale measurements (Whittaker 1977); and 3) the development of species-area curves to estimate larger-scale species richness patterns (Shmida 1984, Stohlgren et al. 1995). Whittaker’s plant diversity plot method has proved to be an efficient method of sampling used around the world, especially in semi-arid environments (Shmida 1984, Stohlgren et al. 1995).

Since most biological communities demonstrate natural patchiness and clumping of species, samples drawn from natural communities are not random samples from a multinomial population (Heltshe & Forrester 1983). Consequently, samples taken are not random samples of individuals but random samples of space (Heltshe & Forrester 1983), leading to an array of possible species-area curves.

In general, the exponential and power functions provided a better fit to the data than the untransformed linear function (see r-values and p-values in Table 6.1) except in plots W17 and W18 (Table 6.1), located in the Tanqua Karoo, where the untransformed linear function produced the best fit. The exponential function performed marginally better than the power function with 23 out of the 40 plots producing the best fit (r-value) with the exponential function. These findings support Tjørve (2003) who stated that the power and exponential models not only dominate the literature but are proven models that perform well, but contrast with the findings of Connor and McCoy (1979) who studied 100 cases and concluded that the power function is usually superior in linearising species-area relationships. Veech (2000) indicates the importance of not choosing just one function and suggests investigating the power, linear and exponential functions, especially given that none of these three functions consistently fits species-area data better than the others.

Table 6.1 Vegetation group, plant association number, survey plot number, untransformed linear function, exponential function, power function, r-value and p- value and species richness (at 1000 m²) for 40 modified Whittaker diversity plots surveyed in the Hantam-Tanqua-Roggeveld subregion in 2005

Vegetation group Plant Survey Untransformed linear function Exponential (Semi-log) function Power (Log-log) function Species asso- plot no. Linear equation r- p-value Linear equation r- p-value Linear equation r- p-value richness ciation value value value at no. 1000 m2 Mountain Renosterveld 1 W23 y=49.1038+0.0521x 0.7987 0.0312* y=24.0807+24.4289x 0.9862 0.0000*** y=1.4529+0.1976x 0.9567 0.0007*** 99 Mountain Renosterveld 1 W24 y=29.0398+0.0455x 0.8813 0.0087** y=10.0469+19.2775x 0.9821 0.0000*** y=1.1934+0.2327x 0.9875 0.0000*** 72 Mountain Renosterveld 1 W25 y=21.6580+0.0446x 0.9130 0.0041** y=4.6157+17.7535x 0.9568 0.0007*** y=1.0403+0.2599x 0.9815 0.0000*** 65 Mountain Renosterveld 1 W26 y=39.4736+0.0445x 0.8016 0.0302* y=18.2448+20.7637x 0.9846 0.0000*** y=1.3506+0.2042x 0.9629 0.0005*** 82 Mountain Renosterveld 1 W27 y=37.7687+0.0340x 0.7464 0.0539ns y=20.0244+16.9636x 0.9806 0.0001*** y=1.3542+0.1827x 0.9513 0.0010*** 70 Mountain Renosterveld 2 W3 y=18.7349+0.0573x 0.9535 0.0009*** y=-0.8280+21.1255x 0.9256 0.0028** y=0.9086+0.3221x 0.9745 0.0002*** 75 Mountain Renosterveld 2 W4 y=24.3334+0.0496x 0.8578 0.0135* y=2.7603+21.6392x 0.9847 0.0000*** y=0.9742+0.3247x 0.9669 0.0004*** 72 Mountain Renosterveld 2 W11 y=19.3549+0.0445x 0.8589 0.0133* y=0.0385+19.3867x 0.9844 0.0000*** y=0.8293+0.3556x 0.9752 0.0002*** 62 Mountain Renosterveld 2 W12 y=32.0579+0.0347x 0.7739 0.0412* y=14.4695+16.9413x 0.9937 0.0000*** y=1.2444+0.2121x 0.9605 0.0006*** 65 Mountain Renosterveld 2 W20 y=31.6149+0.0492x 0.8376 0.0187* y=9.5255+21.9633x 0.9837 0.0000*** y=1.1713+0.2650x 0.9568 0.0007*** 79 Mountain Renosterveld 2 W21 Y=31.1870+0.0557x 0.8083 0.0278* y=4.6272+25.9831x 0.9922 0.0000*** y=1.0822+0.3176x 0.9692 0.0003*** 84 Mountain Renosterveld 2 W28 y=45.5203+0.0533x 0.7720 0.0420** y=18.4944+26.0171x 0.9930 0.0000*** y=1.3691+0.2307x 0.9568 0.0007*** 96 Mountain Renosterveld 2 W29 y=36.0150+0.0509x 0.8585 0.0134* y=13.8923+22.1935x 0.9851 0.0000*** y=1.2967+0.2221x 0.9831 0.0000*** 85 Mountain Renosterveld 2 W30 y=41.5343+0.0536x 0.8358 0.0192* y=17.3209+24.0287x 0.9867 0.0000*** y=1.3596+0.2174x 0.9719 0.0003*** 93 Mountain Renosterveld 2 W40 y=24.2869+0.0433x 0.8504 0.0153* y=5.3234+18.9851x 0.9810 0.0001*** y=1.0237+0.2898x 0.9608 0.0006*** 66 Mountain Renosterveld 3 W7 y=15.1610+0.0653x 0.9551 0.0008*** y=-7.1957+24.1232x 0.9286 0.0025** y=0.7054+0.4058x 0.9975 0.0000*** 79 Mountain Renosterveld 3 W8 y=19.6701+0.0807x 0.9639 0.0005*** y=-7.0353+29.1419x 0.9165 0.0037** y=0.8691+0.3751x 0.9895 0.0000*** 99

Figure 6.1 (continued)

Winter Rainfall Karoo 4 W5 y=36.8267+0.0556x 0.8177 0.0246* y=10.6242+25.7069x 0.9959 0.0000*** y=1.2064+0.2814x 0.9454 0.0013** 90 Winter Rainfall Karoo 4 W6 y=27.0250+0.0751x 0.9210 0.0032** y=-1.2689+29.6096x 0.9555 0.0008*** y=1.0432+0.3289x 0.9869 0.0000*** 100 Winter Rainfall Karoo 4 W38 y=22.5390+0.0495x 0.8600 0.0130* y=1.4013+21.3106x 0.9745 0.0002*** y=0.9461+0.3244x 0.9727 0.0002*** 70 Winter Rainfall Karoo 4 W39 y=21.8865+0.0482x 0.8367 0.0189* y=0.3194+21.4625x 0.9804 0.0001*** y=0.8525+0.3680x 0.9562 0.0008*** 68 Winter Rainfall Karoo 5 W9 y=23.7785+0.0407x 0.8637 0.0122* y=6.3768+17.5341x 0.9800 0.0001*** y=1.0580+0.2639x 0.9750 0.0002*** 63 Winter Rainfall Karoo 5 W10 y=27.5470+0.0484x 0.8502 0.0154* y=6.1057+21.3899x 0.9898 0.0000*** y=1.0776+0.2902x 0.9657 0.0004*** 74 Winter Rainfall Karoo 5 W31 y=25.3184+0.0328x 0.8813 0.0087** y=11.7322+13.8133x 0.9779 0.0001*** y=1.1873+0.1928x 0.9882 0.0000*** 57 Winter Rainfall Karoo 5 W32 y=30.1233+0.0182x 0.6154 0.1413ns y=18.3652+10.7066x 0.9552 0.0008*** y=1.2718+0.1614x 0.9030 0.0053** 47 Winter Rainfall Karoo 5 W33 y=25.0023+0.0375x 0.8241 0.0226* y=7.7520+17.0365x 0.9855 0.0000*** y=1.0659+0.2653x 0.9526 0.0009*** 61 Winter Rainfall Karoo 5 W34 y=24.6008+0.0386x 0.8445 0.0168* y=7.9234+16.7581x 0.9649 0.0004*** y=1.0759+0.2569x 0.9450 0.0013** 62 Winter Rainfall Karoo 5 W35 y=13.9400+0.0218x 0.9113 0.0043** y=5.5902+8.6884x 0.9575 0.0007*** y=0.9287+0.2001x 0.9875 0.0000*** 35 Winter Rainfall Karoo 5 W36 y=28.9600+0.0443x 0.7715 0.0422* y=6.6151+21.5481x 0.9872 0.0000*** y=1.0447+0.3123x 0.9364 0.0019** 71 Winter Rainfall Karoo 5 W37 y=27.7537+0.0540x 0.8268 0.0218* y=3.7763+23.9062x 0.9640 0.0005*** y=1.0826+0.2910x 0.9678 0.0004*** 79 Winter Rainfall Karoo 6 W1 y=13.5255+0.0472x 0.9522 0.0009*** y=-2.3938+17.2654x 0.9164 0.0037** y=0.7286+0.3506x 0.9706 0.0003*** 60 Winter Rainfall Karoo 6 W2 y=23.5017+0.0438x 0.9016 0.0055** y=6.2029+17.8348x 0.9677 0.0004*** y=1.0776+0.2543x 0.9823 0.0000*** 66 Winter Rainfall Karoo 6 W22 y=16.3957+0.0156x 0.6563 0.1093ns y=6.9467+8.7301x 0.9656 0.0004*** y=0.8811+0.2469x 0.9029 0.0054** 31 Tanqua Karoo 7 W13 y=4.7679+0.0256x 0.9779 0.0001*** y=-3.1667+8.8532x 0.8914 0.0070** y=0.2360+0.4044x 0.9888 0.0000*** 30 Tanqua Karoo 7 W14 y=11.3370+0.0155x 0.7842 0.0368* y=3.6387+7.4471x 0.9943 0.0000*** y=0.7216+0.2617x 0.9589 0.0006*** 26 Tanqua Karoo 7 W17 y=1.8790+0.0112x 0.9818 0.0000*** y=-1.3900+3.7252x 0.8621 0.0126* y=-0.2211+0.4402x 0.9480 0.0012** 13 Tanqua Karoo 7 W19 y=9.0794+0.0174x 0.9067 0.0049** y=2.2380+7.0661x 0.9690 0.0003** y=0.6615+0.2578x 0.9879 0.0000*** 26 Tanqua Karoo 8 W15 y=4.4008+0.0100x 0.8660 0.0117* y=0.1601+4.2870x 0.9746 0.0002*** y=0.2469+0.3211x 0.9859 0.0000*** 14 Tanqua Karoo 8 W16 y=3.8269+0.0084x 0.9364 0.0019** y=0.8376+3.1794x 0.9380 0.0018** y=0.3121+0.2488x 0.9764 0.0002*** 12 Tanqua Karoo 8 W18 y=0.7566+0.0082x 0.9955 0.0000*** y=-1.0362+2.2944x 0.7350 0.0598ns y=-0.2429+0.2737x 0.7350 0.0598ns 9 ns Not significant, * p < 0.05 Significant, ** p < 0.01 Highly significant, *** p < 0.001 Very highly significant

As a consequence of the assumed linearity of the models the rate at which species accumulate with increments in area (the slope of the line) is a constant value and this can be used to make comparisons between different areas. The slope of the exponential function obtained for the 40 plots surveyed, varied widely from 2.2944 for plot W18 in the Tanqua Karoo to 29.6096 on the Roggeveld escarpment for plot W6 (Table 6.1). Slopes for the Tanqua Karoo vegetation ranged from 2.2944 (plot W18) to 8.8532 (plot W13), whereas for the Winter Rainfall Karoo the range was from 8.6884 (plot W35) to 29.6096 (plot W6) and in the Mountain Renosterveld vegetation, slopes varying from 16.9413 (plot W12) to 29.1419 (plot W8) were found. The range of slopes was the largest for the Winter Rainfall Karoo, which is the most diverse vegetation group, and smallest for the Tanqua Karoo vegetation. Both the untransformed linear function and power function demonstrated the same trends as the exponential function with the Tanqua Karoo having the lowest slope values, the winter Rainfall Karoo showing a large range of slope values and the Mountain Renosterveld having a smaller range, but values generally being quite high. A large range of slopes within a vegetation unit was also reported by Palmer and White (1994) who state that there is no such thing as the species-area curve for a given location. Rather, there is a suite of species-area curves, each with its own characteristics (Palmer & White 1994).

Since the exponential function produced marginally better results than the other functions these results were further analysed. The ANOVA on the exponential curves revealed that the slopes obtained for the Tanqua Karoo were significantly shallower than those obtained for the Winter Rainfall Karoo (p < 0.0001) and Mountain Renosterveld (p < 0.0001) (Table 6.2). No significant difference was found between the slopes of the Mountain Renosterveld and Winter Rainfall Karoo (p = 0.099). Within the Mountain Renosterveld group, no significant difference was found between associations 1 and 2, these associations were also related to association 3 of the Mountain Renosterveld group which in turn was not significantly different from the association 4 (Escarpment Karoo) of the Winter Rainfall Karoo (Table 6.2). The other two associations within the Winter Rainfall Karoo were not significantly different yet, the transitional Roggeveld Karoo (association 6) was also similar to association 7 of the Tanqua Karoo (Table 6.2). Associations 7 and 8 of the Tanqua Karoo did not differ significantly (Table 6.2).

The large difference between species-area curves within the study area can be ascribed to factors such as differences in patch or habitat size and structural differences in community organisation resulting from different species abundance distributions (see Scheiner et al. 2000, He & Legendre 2002, Scheiner & Jones 2002, Keeley 2003, Scheiner 2003). These may be tied to patterns of life form distributions, which result from different environmental conditions (Keeley 2003). The Tanqua Karoo is a more arid environment with a mean annual precipitation varying from <100 mm to 200 mm (Schulze 1997) while rainfall in the Winter Rainfall Karoo and Mountain Renosterveld ranges from 100 mm to 400 mm in normal years

and to more than 400 mm in above-normal years (Schulze 1997). In general, the habitat is more homogeneous at the Whittaker plot size in the Tanqua Karoo than in the other vegetation groups, resulting in shallower slopes and lower intercepts in this vegetation group.

Table 6.2 Mean slope values for the eight associations and three broad vegetation groups in the Hantam-Tanqua-Roggeveld subregion

Association Vegetation group Mean slope value 1. Rosenia oppositifolia Mountain Renosterveld 19.837cd Mountain Renosterveld 2. Dicerothamnus rhinocerotis Mountain Renosterveld 21.826cd Mountain Renosterveld 3. Passerina truncata Mountain Mountain Renosterveld 26.633d Renosterveld Mountain Renosterveld 21.807a 4. Pteronia glauca – Euphorbia Winter Rainfall Karoo 24.522d decussata Escarpment Karoo 5. Eriocephalus purpureus Winter Rainfall Karoo 16.820c Hantam Karoo 6. Pteronia glomerata Winter Rainfall Karoo 14.610bc Roggeveld Karoo Winter Rainfall Karoo 18.331a 7. Aridaria noctiflora Tanqua Tanqua Karoo 6.773ab and Loeriesfontein Karoo 8. Stipagrostis obtusa Central Tanqua Karoo 3.254a Tanqua Grassy Plains Tanqua Karoo 5.265b p-value P < 0.05 P < 0.0001

A west to east transect of ten survey plots through the Tanqua Karoo, up the Roggeveld Escarpment, across the Roggeveld Mountains and Roggeveld plateau towards the strongly summer rainfall Nama Karoo Biome (Rutherford & Westfall 1994) is presented to illustrate the most prominent changes in the vegetation and how these are reflected in the species-area curves. A profile of this transect illustrates the altitudinal variation, species richness at the 1000 m2 scale and the sequence of vegetation associations (Figure 6.2). Differences in the slopes and intercepts of the species-area curves among the various sites are discussed only for the exponential function due to its slightly better fit to the data.

1600

Tanqua Karoo Winter Rainfall Karoo Mountain Renosterveld Winter Rainfall Karoo 1400 72 75

66 60 1200

1000

100 800

600 26 Altitude(m above sealevel)

12 9 400 Stipagrostis obtusa Aridaria noctiflora Pteronia glauca - Dicerothamnus rhinocerotis Pteronia glomerata Central Tanqua Tanqua Karoo Euphorbia decussata Mountain Renosterveld Roggeveld Karoo Grassy Plains Roggeveld Escarpment 200 Karoo

0 W16 W18 W19 W14 W6 W5 W3 W4 W2 W1 Survey plot number Figure 6.2 Profile from west to east along a transect through three vegetation groups and five associations in the Hantam-Tanqua-Roggeveld subregion. Values for species richness are indicated for each survey plot.

Species richness provides an instantly comprehensible expression of diversity (Magurran 1988), and clearly illustrates the changes along the altitudinal gradient. The lowest species counts (1000 m2 plots) were found in the low-lying Tanqua Karoo, the intermediate counts on the high-lying inland plateau and the highest counts on the steep escarpment (Figure 6.2). Altitude is an important contributor to explaining species diversity (Körner 2000) and altitudinal gradients have been recognised as an important buffer in the event of climate change (Bond & Richardson 1990). Because mountain habitats could provide refugia for species during climate change (Midgley et al. 2000) it emphasizes the importance of conserving mountain communities (Burke et al. 2003).

120

100 W6 W5

60 W1 Species richness

0 0 200 400 600 800 1000 1200 Area (m²)

120

100

80 W4 W3

60 Species richness

0 0 200 400 600 800 1000 1200 Area (m²)

120

100

60 Species richness

W14 W19 20

W16 W18

0 0 200 400 600 800 1000 1200 Area (m²)

Figure 6.3 Species-area curves for 10 plots along a transect from west to east through the study area comparing the three vegetation groups as well as the three plotting functions: a) Winter Rainfall Karoo, untransformed linear function; b) Mountain Renosterveld, untransformed linear function; c) Tanqua Karoo, untransformed linear function; d) Winter Rainfall Karoo, exponential function; e) Mountain Renosterveld, exponential function; f) Tanqua Karoo, exponential function; g) Winter Rainfall Karoo, power function; h) Mountain Renosterveld, power function; and i) Tanqua Karoo, power function.

120

100 W5

W6 80

60 W2

40 Species richness

0 00.511.522.533.5

-20 Area (log m²)

120

100

60 W3

40 Species richness

0 00.511.522.533.5

-20 Area (log m²)

120

100

40 Species richness W14 W19 20

W16

W18 0 00.511.522.533.5

-20 Area (log m²)

Figure 6.3 (continued)

2.5

2 W6 W2 W1

1.5

0.5 Species richness (log number ofspecies)

0 0 0.5 1 1.5 2 2.5 3 3.5

-0.5 Area (log m²)

2.5

2 W4

1.5

0.5 Species richness (log number of species)

0 0 0.5 1 1.5 2 2.5 3 3.5

-0.5 Area (log m²)

2.5

W14 1.5 W19

W16 1

W18 0.5 Species richness (log number ofspecies)

0 0 0.5 1 1.5 2 2.5 3 3.5

-0.5 Area (log m²)

Figure 6.3 (continued)

The transect begins in the west with two sites located in the Tanqua Karoo, survey plots W16 and W18, falling within the Stipagrostis obtusa Central Tanqua Grassy Plains association (Van der Merwe et al. 2008b). Species richness of survey plot W16 on level terrain near Bo Stompiesfontein, at an altitude of 374 m above sea level (Figure 6.2), with a low vegetation cover of 40% but a high rock cover of 99%, was found to be 12 species per 1000 m² (Table 6.1). Survey plot W18 had a low species richness of only 9 species per 1000 m² (Table 6.1) and was surveyed at an altitude of 405 m above sea level (Figure 6.2), on level terrain with a very low vegetation cover of approximately 10% but a high rock cover of 99%. These two plots are characterised by the lowest species richness along the transect (and also in the entire study area) and species-area curves with the shallowest slopes (Figure 6.3c, 6.3f and 3i). The first plot (W16) was located in a relatively homogeneous succulent patch and was dominated by three succulent species with only a few individuals of the other nine species present. In the case of plot W18, only one species, the grass Stipagrostis obtusa, dominated with the other eight species represented by only one individual or a cover of <1%. The dominance of a few species and small species richness values resulted in the very shallow species-area curve slopes. There was no sudden increase in species numbers at a certain plot size, indicating a large patch size and homogeneous vegetation. Slope values of the exponential function did not differ significantly (p = 0.4320) between the two sites although the intercept did differ significantly (p = 0.0078), (Figure 6.3f).

The next two plots along the transect were also surveyed in the Tanqua Karoo but in the Aridaria noctiflora Tanqua Karoo association (Van der Merwe et al. 2008b). Survey plot W19 and W14 were surveyed at a higher altitude (Figure 6.2) and vegetation cover (50 – 75%), but lower rock cover (70 – 95%) than the previous two plots. Although the species richness at these two sites was higher (26 species per 1000 m², Table 6.1) than at the previous two sites and the slopes of the species-area curves steeper (Figure 6.3c, 6.3f and 6.3i), these values still represent the lower end of the scale for the study area. In both of these plots the greatest contribution to the vegetation cover was made by five species while, the other 21 species did not contribute notably. The species-area curves remained shallow due to the very gradual increase in number of species encountered as the plots were surveyed and remained shallow to the 1000 m² where species richness remained small. No sudden increase in species numbers at any plot size was found within the plot indicating that the vegetation surveyed was homogeneous. The exponential function slopes of the two plots did not differ significantly (p = 0.6747) but the intercepts did differ significantly (p = 0.0277). A comparison between the two associations in the Tanqua Karoo revealed that the slopes of the exponential species-area curve function differed significantly between the associations (p = 0.0001), (Figure 6.3f).

Further eastwards the next two plots were located on the Roggeveld Escarpment in the Pteronia glauca – Euphorbia decussata Roggeveld Escarpment Karoo association (Van der Merwe et al. 2008b). Plot W6 and W5 were surveyed at intermediate altitudes (Figure 6.2),

the vegetation cover on both sites was 75% and rock cover ranged from 80 to 95%. The species richness of the plots was 100 species per 1000 m² and 90 species per 1000 m² (Table 6.1) respectively. The high species richness can possibly be ascribed to the vegetation being transitional from the Tanqua Karoo (Succulent Karoo Biome) to the Mountain Renosterveld (Fynbos Biome). Transition zones are hypothesised to have a greater species richness than adjacent areas (Shmida & Wilson 1985, Scheiner & Rey-Benayas 1994). The slopes of the species-area curves of these two plots were the steepest of all plots along the transect (Figure 6.3a, 6.3d and 6.3g). In plot W6 one species was more abundant and in plot W5 two species dominated, however, the bulk of the species in these two plots were common occurring frequently throughout the plot. Various geophyte and annual species with low abundances were continuously added as the plot size was enlarged. It was clear that an asymptote in species richness had not yet been reached at the 1000 m2 scale. The slopes of the exponential function species-area curves of the two sites sampled did not differ significantly (p = 0.3770) neither did the intercepts (p = 0.1125).

From the Roggeveld Escarpment the transect runs through the Mountain Renosterveld of the Dicerothamnus rhinocerotis Mountain Renosterveld association (Van der Merwe et al. 2008a). Plot W3 and W4 were the two highest altitude plots surveyed (Figure 6.2). Both plots had a 60% vegetation cover and a 75% rock cover and species richness of 75 and 72 species per 1000 m² respectively (Table 6.1). The species-area curves produced much steeper slopes (Figure 6.3b, 6.3e and 6.3h) than those produced for the Tanqua Karoo. These plots were relatively homogeneous and dominated by either Dicerothamnus rhinocerotis or Merxmuellera stricta. The bulk of the species were common throughout all the smaller plots surveyed, but many geophyte species with a low abundance were encountered. Species richness had not yet reached an asymptote at the 1000 m2 scale. Species-area curve slopes for the exponential function of the two sampled sites were not significantly different (p = 0.9057), neither were the intercepts (p = 0.2595). The exponential function slopes of these two Mountain Renosterveld plots did not differ significantly from the Roggeveld Escarpment plots (p = 0.1870) or Roggeveld Karoo plots (p = 0.6256).

The last two plots (W2 and W1) were surveyed in the Winter Rainfall Karoo on the eastern fringes of the study area in the Pteronia glomerata Roggeveld Karoo association (Van der Merwe et al. 2008b) at a slightly lower altitude than the preceding two plots (Figure 6.2). Vegetation cover varied between 50 and 60% and rock cover was low (<1 – 10%). Species richness ranged from 60 to 66 species per 1000 m² (Table 6.1). These relatively homogeneous plots were dominated by one (plot W2) or two (plot W1) species and produced species-area slopes less steep (Figure 6.3a, 6.3d and 6.3g) than those for plots W3 and W4 of the Mountain Renosterveld vegetation group. The bulk of the species were common and found throughout the various plot sizes within the 1000 m² plot. A large number of different annual and geophyte species contributed to the species richness. Once again, species

richness had not yet reached an asymptote within the Whittaker plot dimensions. The exponential function slopes of the species-area curves of the two sampled sites did not differ significantly (p = 0.8886), but the intercepts did differ significantly (p = 0.0183).

The year 2005, in which the data were collected, was a very poor rainfall year. This is expected to have had a significant effect on the number of species encountered, especially with respect to annuals and geophytes which are responsible for a large part of the species diversity in the region. Aronson and Shmida (1992) confirm this stating that, a general distinction can be drawn between ‘good’, ‘bad’ and ‘medium’ rainfall years and this distinction is immediately reflected in changes in species diversity that are largely dependent on annual, and in the case of the Hantam-Tanqua-Roggeveld subregion also geophytic, species.

At present, biodiversity data for the Succulent Karoo are limited to species richness and a literature search revealed no information pertaining to species-area curves in the region. Correspondence with various other scientists in the region, however, indicated that more data should become available in the near future. Species-area curves are most useful in comparing diversities between geographical regions, habitats, or taxa over a range of sample sizes (Connor & McCoy 1979) as done in this study. The information gained from the study can be used to guide conservation authorities on which areas to target for conservation. Ironically, the largest formal conservation area in the subregion, the Tankwa Karoo National Park, for the most part covers the Tanqua Karoo, which the current study showed to be the area with the lowest diversity. However, the park is expanding along the steep escarpment into the Roggeveld Mountains. These areas were found to be of high diversity value and should be a conservation priority.

Although the logistic model is often applied to calculate the total species richness for a community (He & Legendre 1996, Scheiner 2003) this practice has not been applied in this study. Caution is necessary when extrapolating from these curves to estimate the number of species in a large region (Connor & McCoy 1979, Palmer 1990, Palmer & White 1994, Procheş et al. 2003, Wilson & Shmida 1984). The failure of the species-area curve methods to predict total species richness could be due to species-area curves being functionally and fundamentally different at different scales (Williams 1964, Palmer 1990). Also, when considering that communities are patchy and samples of fixed size are drawn from a homogeneous population, when, in fact, populations are heterogeneous in time and space (Heltshe & Forrester 1983), the indiscriminate use of the species-area curve in conservation biology is not advocated.

6.5 Conclusions

Forty Whittaker plots were sampled throughout the Hantam, Tanqua and Roggeveld areas. Analyses of these data revealed an array of species-area curves for each of the functions used (untransformed linear, exponential and power functions), each with its own characteristics. The significance levels of the three functions varied greatly, however, the exponential and power functions produced better r-values and p-values.

Each association produced its own species-area curve characteristics as well as species richness values. Slope values for the samples within an association did not differ significantly, although the intercept value often did. Comparisons between associations along a west to east transect through the study area revealed significant differences in the slope value between the associations, except for the Dicerothamnus rhinocerotis Mountain Renosterveld which did not differ significantly from the associations bordering it on either side.

Generally, across the ten plots along the transect, the Tanqua Karoo had a low vegetation cover and both a low species richness and species-area curves with shallow slopes. The Roggeveld Escarpment Karoo vegetation, of the Winter Rainfall Karoo vegetation group, which is transitional between the Tanqua Karoo and the Mountain Renosterveld of the Roggeveld Mountains produced the highest species richness and steepest slopes for the species-area curves.

6.6 Acknowledgements

The authors would like to thank the Critical Ecosystem Partnership Fund (CEPF) through the Succulent Karoo Ecosystem Plan/Program (SKEP) initiative for funding the project. The Critical Ecosystem Partnership Fund is a joint initiative of Conservation International, the Global Environmental Facility, the Government of Japan, the MacArthur Foundation and the World Bank. A fundamental goal is to ensure civil society is engaged in biodiversity conservation. The various people who assisted with the field work are gratefully acknowledged. CapeNature, the Department of Tourism, Environment and Conservation (Northern Cape) and SANParks are thanked for the necessary permits and permission to conduct this research. This research was supported by the National Research Foundation under grant number 61277.

6.7 References

ACOCKS, J.P.H., 1953. Veld types of South Africa. Memoirs of the Botanical Survey of South Africa 28, 1-192.

ACOCKS, J.P.H., 1988. Veld types of South Africa. 3rd edn. Memoirs of the Botanical Survey of South Africa 57, 1-146. ARONSON, J., SHMIDA, A., 1992. Plant species diversity along a Mediterranean-desert gradient and its correlation with interannual rainfall fluctuations. Journal of Arid Environments 23, 235-247. BOND, W.J., RICHARDSON, D.M., 1990. What can we learn from extinctions and invasions about the effects of climate change? South African Journal of Science 86, 429-433. BURKE, A., ESLER, K.J., PIENAAR, E., BARNARD, P., 2003. Species richness and floristic relationships between mesas and their surroundings in southern African Nama Karoo. Diversity and Distributions 9, 43-53. CILLIERS, C., THERON, H., RÖSCH, H., LE ROUX, A., 2002. Succulent Karoo Ecosystem Plan, Sub-regional report, Hantam/Tanqua/Roggeveld. Succulent Karoo Ecosystem Plan report. CONNOR, E.F., McCOY, E.D., 1979. The statistics and biology of the species-area relationship. The American Naturalist 113, 791-833. COWLING, R.M., GIBBS RUSSEL, G.E., HOFFMAN, M.T., HILTON-TAYLOR, C., 1989. Patterns of plant species diversity in southern Africa. In: B.J. Huntley (Ed.). Biotic diversity in southern Africa. Concepts and Conservation, pp. 19-50. Oxford University Press, Cape Town. COWLING, R.M., HILTON-TAYLOR, C., 1994. Patterns of plant diversity and endemism in southern Africa: an overview. In: B.J. Huntley (Ed.). Botanical diversity in southern Africa. Strelitzia 1, pp. 31-52. National Botanical Institute, Pretoria. CRITICAL ECOSYSTEM PARTNERSHIP FUND, 2003. Ecosystem Profile: The Succulent Karoo hotspot, Namibia and South Africa. Critical Ecosystem Partnership Fund report. FRANCIS, M.L., FEY, M.V., PRINSLOO, H.P., ELLIS, F., MILLS, A.J., MEDINSKI, T.V., 2007. Soils of Namaqualand: Compensations for aridity. Journal of Arid Environments 70, 588-603. HE, F., LEGENDRE, P., 1996. On species-area relations. The American Naturalist 4, 719- 737. HE, F., LEGENDRE, P., 2002. Species diversity patterns derived from species-area models. Ecology 83, 1185-1198. HELLMANN, J.J., FOWLER, G.W., 1999. Bias, precision, and accuracy of four measures of species richness. Ecological Applications 9, 824-834. HELTSHE, J.F., FORRESTER, N.E., 1983. Estimating species richness using the Jackknife Procedure. Biometrics 39, 1-11. HILTON-TAYLOR, C., 1994. Western Cape Domain (Succulent Karoo). In: S.D. Davis, V.H. Heywood, A.C. Hamilton (Eds). Centres of plant diversity. A guide and strategy for their conservation, pp. 201-203. IUCN Publications Unit, Cambridge.

KEELEY, J.E., 2003. Relating species abundance distributions to species-area curves in two Mediterranean-type shrublands. Diversity and Distributions 9, 253-259. KÖRNER, C., 2000. Why are there global gradients in species richness? Mountains might hold the answer. Trends in Ecology and Evolution 15, 513-514. LOW, A.B., REBELO, A.G., 1996. Vegetation of South Africa, Lesotho and Swaziland. Department of Environmental Affairs and Tourism, Pretoria. MAGURRAN, A.E., 1988. Ecological Diversity and its measurement. Cambridge University Press, Cambridge. MIDGLEY, G.F., HANNAH, L., ROBERTS, R., MacDONALD, D.J., ALLSOPP, J., 2000. Have pleistocene climatic cycles influenced species richness patterns in the greater Cape Mediterranean Region? Journal of Mediterranean Ecology 2, 137-144. MUCINA, L., RUTHERFORD, M.C., POWRIE, L.W. (Eds), 2005. Vegetation map of South Africa, Lesotho and Swaziland, 1 : 1 000 000 scale sheet maps. South African Biodiversity Institute, Pretoria. MUCINA, L., JÜRGENS, N., LE ROUX, A., RUTHERFORD, M.C., SCHMIEDEL, U., ESLER, K.J., POWRIE, L.W., DESMET, P.G., MILTON, S.J., 2006. Succulent Karoo Biome. In: L. Mucina, M.C. Rutherford (Eds) 2006. The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19, pp. 220-299. South African National Biodiversity Institute, Pretoria. PALMER, M.W., 1990. The estimation of species richness by extrapolation. Ecology 71, 1195-1198. PALMER, M.W., WHITE, P.S., 1994. Scale dependence and the species-area relationship. The American Naturalist 144, 717-740. PIMM, S.L., RUSSEL, G.J., GITTLEMAN, J.L., BROOKS, T.M., 1995. The future of biodiversity. Science 269, 347-350. PODANI, J., CZÁRÁN, T., BARTHA, S., 1993. Pattern, area and diversity: the importance of spatial scale in species assemblage. Abstracta Botanica 17, 37-51. PROCHEŞ, Ş., COWLING, R.M., MUCINA, L., 2003. Species-area curves based on relevé data for the Cape Floristic Region. South African Journal of Science 99, 474-476. QUINN, G.P., KEOUGH, M.J., 2002. Experimental design and data analysis for biologists. Cambridge University Press, Cambridge. RUTHERFORD, M.C., MUCINA, L., POWRIE, L.W. 2003. Nama-Karoo veld types revisited: a numerical analysis of original Acocks’s field data. South African Journal of Botany 69, 52-61. RUTHERFORD, M.C., WESTFALL, R.H., 1994. Biomes of Southern Africa. An objective characterisation. Memoirs of the Botanical Survey of South Africa 63, 1-94. SCHEINER, S.M., 2003. Six types of species-area curves. Global Ecology and Biogeography 12, 441-447. SCHEINER, S.M., 2004. A mélange of curves – further dialogue about species-area relationships. Global Ecology and Biogeography 13, 479-484.

SCHEINER, S.M., COX, S.B., WILLIG, M., MITTELBACH, G.G., OSENBERG, C., KASPARI, M., 2000. Species richness, species-area curves and Simpson’s paradox. Evolutionary Ecology Research 2, 791-802. SCHEINER, S.M., JONES, S., 2002. Diversity, productivity and scale in Wisconsin vegetation. Evolutionary Ecology Research 4, 1097-1117. SCHEINER, S.M., REY-BENAYAS, J.M., 1994. Global patterns of plant diversity. Evolutionary Ecology 8, 331-347. SCHULZE, R.E., 1997. South African Atlas of Agrohydrology and – Climatology. Water Research Commission, Pretoria, Report TT82/96. SHMIDA, A., 1984. Whittaker’s plant diversity sampling method. Israel Journal of Botany 33, 41-46. SHMIDA, A., WILSON, M.V., 1985. Biological determinants of species diversity. Journal of Biogeography 12, 1-20. SMITH, G.F., HOBSON, S.R., MEYER, N.L., CHESSELET, P., ARCHER, R.H., BURGOYNE, P.M., GLEN, H.F., HERMAN, P.P.J., RETIEF, E., SMITHIES S.J., VAN JAARSVELD, E.J., WELMAN, W.G., 1993. Southern African succulent plants – an updated synopsis. Aloe 30, 32-74. STIRLING, G., WILSEY, B., 2001. Empirical relationships between species richness, evenness, and proportional diversity. The American Naturalist 158, 286-299. STOHLGREN, T.J., FALKNER, M.B., SCHELL, L.D., 1995. A modified-Whittaker nested vegetation sampling method. Vegetatio 117, 113-121. TJØRVE, E., 2003. Shapes and functions of species-area curves: a review of possible models. Journal of Biogeography 30, 827-835. VAN DER MERWE, H., VAN ROOYEN, M.W., VAN ROOYEN, N., 2008a. Vegetation of the Hantam-Tanqua-Roggeveld subregion, South Africa. Part 1. Fynbos Biome related vegetation. Koedoe 50, 61-71. VAN DER MERWE, H., VAN ROOYEN M.W., VAN ROOYEN, N., 2008b. Vegetation of the Hantam-Tanqua-Roggeveld subregion, South Africa. Part 2. Succulent Karoo Biome related vegetation. Koedoe 50, 161-183. VAN JAARSVELD, E., 1987. The succulent riches of South Africa and Namibia. Aloe 24, 45- 92. VAN WYK, A.E., SMITH, G.F. (Eds), 2001. Regions of Floristic Endemism in Southern Africa: A review with emphasis on succulents, pp. 1-199. Umdaus Press, Pretoria. VEECH, J.A., 2000. Choice of species-area function affects identification of hotspots. Conservation Biology 14, 140-147. VITOUSEK, P.M., D’ANTONIO, C.M., LOOPE, L.L., WESTBROOKS, R., 1996. Biological invasions as global environmental change. American Scientist 84, 468-478. WEATHER BUREAU, 1998. Climate of South Africa. Climate statistics up to 1990. WB 42. Government Printer, Pretoria.

WHITTAKER, R.H., 1977. Evolution of species diversity on land communities. Evolutionary Biology 10, 1-67. WHITTAKER, R.J. WILLIS, K.J., FIELD, R., 2001. Scale and species richness: towards a general, hierarchical theory of species diversity. Journal of Biogeography 28, 453- 470. WILLIAMS, C.B., 1964. Patterns in the balance of nature. Academic Press, New York. WILSON, M.V., SHMIDA, A., 1984. Measuring Beta diversity with presence-absence data. Journal of Ecology 72, 1055-1064.

Chapter 7

Plant diversity in the Hantam-Tanqua-Roggeveld, Succulent Karoo, South Africa: Diversity parameters

Abstract

Forty Whittaker plots were surveyed to gather plant diversity data in the Hantam-Tanqua- Roggeveld subregion of the Succulent Karoo. Species richness, evenness, Shannon’s index and Simpson’s index of diversity were calculated. Species richness ranged from nine to 100 species per 1000 m² (0.1 ha) with species richness for the Mountain Renosterveld being significantly higher than for the Winter Rainfall Karoo, which in turn was significantly higher than for the Tanqua Karoo. Evenness, Shannon and Simpson indices were found not to differ significantly between the Mountain Renosterveld and Winter Rainfall Karoo, however, these values were significantly higher than for the Tanqua Karoo. Species richness for all plot sizes <0.1 ha were significantly lower for the Tanqua Karoo than for the other two vegetation groups, which did not differ significantly from each other. Only at the 1000 m² scale did species richness differ significantly on the vegetation group level between the Mountain Renosterveld and the Winter Rainfall Karoo.

A Principal Co-ordinate Analysis (PCoA) of species richness data at seven plot sizes produced three distinct clusters in the ordination. One cluster represented the sparsely vegetated, extremely arid Tanqua Karoo which has a low species richness, low evenness values and low Shannon and Simpson indices. Another cluster represented the bulk of the Mountain Renosterveld vegetation with a high vegetation cover, high species richness, high evenness values and high Shannon and Simpson indices. The third cluster was formed by the remaining Mountain Renosterveld plots as well as the Winter Rainfall Karoo plots with intermediate values for the diversity parameters.

Keywords: Fynbos Biome, species evenness, Shannon index, Simpson index, species richness, Succulent Karoo Biome

7.1 Introduction

Despite changing fashions and preoccupations, diversity has remained a central theme to ecology (Magurran 1988). Diversity has two components: species richness, or the number of plant species in a given area, and species evenness, or how well abundance or biomass is distributed among species within a community (Wilsey & Potvin 2000). Numerous indices exist which use either species richness or evenness as well as a combination of these two components. In spite of various criticisms, these indices have sparked renewed interest in handling problems associated with the conservation of natural heritage or the changes in global ecology (Mouillot & Leprêtre 1999).

The Hantam-Tanqua-Roggeveld constitutes one of the four planning domains into which the Succulent Karoo was subdivided during the Succulent Karoo Ecosystem Plan (SKEP) initiative. In the course of the SKEP programme it soon became apparent that future planning of conservation and development in the subregion were hampered by a paucity of information available on the plant diversity (Cilliers et al. 2002, Critical Ecosystem Partnership Fund 2003). A study was therefore initiated to provide baseline information on patterns of plant diversity in this unique arid area. At the landscape level the diversity of plant associations have been described (Van der Merwe et al. 2008a, 2008b) and the species-area curves characterising the associations have been presented in Chapter 6.

The Succulent Karoo Biome is a predominantly winter rainfall arid region that occupies 111 000 km² on the fringes of South Africa’s Cape Floristic Region (Mucina et al. 2006). The species richness of the Succulent Karoo flora is exceptional in terms of established hotspots, but especially in comparison with similar arid environments (Cowling & Hilton-Taylor 1994). Although the Roggeveld mountain range was included in the SKEP planning domain, there is some controversy as to whether it is actually part of the Fynbos Biome or the Succulent Karoo Biome (Marloth 1908, Diels 1909, Weimarck 1941, Hilton-Taylor 1994, Low & Rebelo 1996, Jürgens 1997, Van Wyk & Smith 2001, Mucina & Rutherford 2006) or even Nama Karoo Biome (Rutherford & Westfall 1994). The Fynbos Biome as delineated by Rutherford and Westfall (1994) constitutes the major portion of the Cape Floristic Region (CFR). The CFR is recognised as one of the world’s plant kingdoms (Good 1947) and is additionally recognised as a global hotspot of diversity (Cowling & Hilton-Taylor 1994). There is a close affiliation between the Succulent Karoo and Fynbos Biomes and Jürgens (1997) proposed the recognition of the Floristic Kingdom of the Greater Cape Flora including at least two partial areas, the Cape Floristic Region and the Succulent Karoo Region. Similarly, Born et al. (2007) investigated the concept of the Greater Cape Floristic Region, which includes the whole winter rainfall area of southern Africa (arid and mesic climates).

The present paper aims to analyse the patterns in the plant diversity of the Hantam, Tanqua and Roggeveld areas in terms of species richness at different scales, species evenness and the Shannon and Simpson indices of diversity.

7.2 Study area

The Succulent Karoo occurs in the western regions of Namibia and South Africa (Milton et al. 1997). It is an area of high plant species diversity at both the local and regional scales (Cowling et al. 1989) and high levels of plant endemism. Within the Hantam-Tanqua- Roggeveld study area Hilton-Taylor (1994) identified three centres of endemism and Van Wyk and Smith (2001) recognised the Hantam-Roggeveld as one of the 13 principal regions and centres of plant endemism in southern Africa.

The Hantam, Tanqua and Roggeveld as defined in this study, lie in the predominantly winter rainfall region of the Northern and Western Cape Provinces of South Africa and cover an area of approximately three million hectares (Figure 7.1). Although the rain falls mainly in winter it does include a few summer thunderstorms with the mean annual precipitation in the subregion ranging from <100 mm to 467 mm per year, the maximum measured for Sutherland (Schulze 1997, Weather Bureau 1998).

Figure 7.1 Location of the Hantam-Tanqua-Roggeveld subregion in South Africa.

Topographically, the three regions differ vastly with the Hantam characterised by a gently undulating to a steeply rolling topography, with the exception of the flat-topped Hantam Mountain rising steeply above the flat surroundings of the town of Calvinia. The Tanqua Karoo covers vast plains from where the landscape rises steeply to the escarpment formed by the Roggeveld, Komsberg and Nuweveld Mountains. The Roggeveld plateau stretches eastwards from the escarpment into the interior of South Africa. Altitude varies from

approximately 290 m above sea level in the Tanqua Karoo to about 1800 m above sea level on the high-lying areas along the Roggeveld Mountain Range.

Shallow lithosols and duplex soils characterise the Hantam Karoo, while scattered dolerite intrusions produce red structural and red vertic clays. The Tanqua Karoo soils are comprised of shallow lithosols that often include deep unconsolidated deposits in the alluvial parts or desert pavement. Soils of the Hantam Mountains and the mountains of the great escarpment are shallow stony lithosols. The occasional lowlands contain duplex soils (Francis et al. 2007).

The study area includes three of Acocks’s (1953, 1988) veld types namely: Mountain Renosterveld, Succulent Karoo and Western Mountain Karoo. The Mountain Renosterveld, as defined by Acocks is equivalent to the Escarpment Mountain Renosterveld of Low and Rebelo (1996). While Acocks’s Succulent Karoo and Western Mountain Karoo form part of the Lowland Succulent Karoo and Upland Succulent Karoo of Low and Rebelo (1996) respectively. The latest vegetation map of South Africa (Mucina et al. 2005, Mucina & Rutherford 2006), identified 12 vegetation types within the area. A recent phytosociological classification of the area recognised eight major plant associations and 25 subassociations in the study area (Van der Merwe et al. 2008a, 2008b).

7.3 Materials and methods

Forty sample plots were surveyed across the Hantam, Tanqua and Roggeveld regions in each of the eight vegetation associations described in the area (Van der Merwe et al. 2008a, 2008b). Altitude, aspect, slope, position on the slope, geology, topography, percentage stone and stone size, soil type and colour, drainage, erosion, trampling and soil compaction are some of the environmental data noted at each plot.

Field surveys were conducted using Whittaker’s plant diversity plot technique (Shmida 1984) with the only modification being the field form and notations used on the field form (see Chapter 6 for more detail). The presence of each species encountered in a quadrat was noted within a column and a cover value was assigned to each species for the 1000 m² quadrat. Seven different plot sizes were used for comparisons of species richness. The total species number for each of the seven plot sizes (1 m², 5 m², 10 m², 20 m², 50 m², 100 m² and 1000 m²) were determined by using the mean of the ten 1 m² plots for the 1 m² plot, a mean of the two 5 m² plots for the 5 m² plot, mean of the total of ten 1 m² plots and the total of two 5 m² plots for the 10 m² plot, total of the ten 1 m² and the two 5 m² plots for a 20 m² plot, a mean of the two 50 m² plots for a 50 m² plot, the total of the two 50 m² plots for a 100 m² plot and the total for the 1000 m² plot.

Species richness (S), Shannon’s index of diversity (H’), Simpson’s index (D) and a measure of evenness (E) were calculated for each sampled plot at the 1000 m² (0.1 ha) size, using the PC-ORD computer program (PC-ORD Version 4 for Windows, MjM Software design) which calculates these four diversity measures as follows:

S = richness = number of non-zero elements in a sampling unit.

H’ = Shannon diversity

H’ = - ∑ pi log pi i

Where pi = importance probability in column i.

E = Evenness (equitability) = H’ / ln (richness)

D = Simpson’s index of diversity for an infinite population. This is the complement of Simpson’s original index and represents the likelihood that two randomly chosen individuals will be different species.

S 2 D = 1 - ∑ pi i i

Comparisons are made throughout the paper with respect to the eight main vegetation associations as described in Van der Merwe et al. (2008a, 2008b). Additionally, comparisons are made at a higher hierarchical level for which purpose some of the associations were grouped into vegetation groups. The three Mountain Renosterveld associations are grouped together and called Mountain Renosterveld, the Escarpment Karoo, Hantam Karoo and Roggeveld Karoo are collectively referred to as the Winter Rainfall Karoo and the Tanqua and Loeriesfontein Karoo together with the Central Tanqua Grassy Plains are termed Tanqua Karoo.

The STATISTICA computer package was used to conduct the statistical analyses (StaSoft, Inc. Version 7 and 8, 2300 East 14th Street, Tulsa, OK 74104). One-way Analysis of Variance (ANOVA) (Bonferroni’s post-hoc test), were performed to determine significant differences in diversity parameters between the vegetation associations and between species richness at different plot sizes (1 m², 5 m², 10 m², 20 m², 50 m², 100 m², 1000 m²). All ANOVAs were preceded by a test for normality. The total number of species for all seven plot sizes for the forty survey plots were ordinated using Principal Co-ordinate Analysis (PCoA) in the SYN- TAX computer program (Podani 2001).

7.4 Results and discussion

Species richness, or the number of species, remains the most widely used diversity measure (Stirling & Wilsey 2001) because it is relatively easy to measure, is comparable across communities, and is well understood by researchers, managers, and the public (Hellman & Fowler 1999). Richness at the 1000 m² level is regarded as a measure of alpha or within- habitat diversity (Whittaker 1977, Cowling et al. 1989, 1992) although significant turnover or internal beta diversity may be associated with habitat heterogeneity at this scale (Cowling et al. 1989).

In this study, the species richness values ranged from nine to 100 species per sampled 1000 m² (Table 7.1). The species richness of Mountain Renosterveld (mean: 79, range: 62-99 species, Table 1) was found to be significantly higher than that of the Winter Rainfall Karoo (mean: 64.6, range: 31-100 species), which in turn was significantly higher than that of the Tanqua Karoo (mean: 18.6, range: 9-30 species, Table 7.1) (Table 7.2). An analysis of the data at 1000 m² for each vegetation association did not produce significant differences (p > 0.05).

Table 7.1 Vegetation group, plant association number, survey plot number, species richness, species evenness, Shannon’s and Simpson’s index of diversity for 40 plots surveyed in the Hantam-Tanqua-Roggeveld subregion in 2005

Vegetation group Plant Survey Species Evenness Shannon Simpson asso- plot no. richness (E) index (H’) index (D) ciation no. Mountain Renosterveld 1 W23 99 0.794 3.647 0.9431 Mountain Renosterveld 1 W24 72 0.759 3.248 0.9265 Mountain Renosterveld 1 W25 65 0.533 2.225 0.6770 Mountain Renosterveld 1 W26 82 0.635 2.797 0.7877 Mountain Renosterveld 1 W27 70 0.563 2.392 0.7267 Mountain Renosterveld 2 W3 75 0.587 2.533 0.7833 Mountain Renosterveld 2 W4 72 0.745 3.186 0.9158 Mountain Renosterveld 2 W11 62 0.572 2.360 0.7327 Mountain Renosterveld 2 W12 65 0.501 2.091 0.6042 Mountain Renosterveld 2 W20 79 0.706 3.087 0.8965 Mountain Renosterveld 2 W21 84 0.630 2.791 0.7798 Mountain Renosterveld 2 W28 96 0.820 3.743 0.9439 Mountain Renosterveld 2 W29 85 0.641 2.849 0.8403 Mountain Renosterveld 2 W30 93 0.544 2.464 0.6668 Mountain Renosterveld 2 W40 66 0.595 2.493 0.7644

100

Table 7.1 (continued)

Mountain Renosterveld 3 W7 79 0.656 2.868 0.8432 Mountain Renosterveld 3 W8 99 0.779 3.582 0.9160 Winter Rainfall Karoo 4 W5 90 0.669 3.009 0.8170 Winter Rainfall Karoo 4 W6 100 0.585 2.693 0.6967 Winter Rainfall Karoo 4 W38 70 0.506 2.150 0.7244 Winter Rainfall Karoo 4 W39 68 0.554 2.338 0.7185 Winter Rainfall Karoo 5 W9 63 0.505 2.092 0.6629 Winter Rainfall Karoo 5 W10 74 0.612 2.633 0.8033 Winter Rainfall Karoo 5 W31 57 0.513 2.074 0.7108 Winter Rainfall Karoo 5 W32 47 0.539 2.076 0.7493 Winter Rainfall Karoo 5 W33 61 0.503 2.066 0.5813 Winter Rainfall Karoo 5 W34 62 0.674 2.783 0.8367 Winter Rainfall Karoo 5 W35 35 0.639 2.273 0.8525 Winter Rainfall Karoo 5 W36 71 0.622 2.650 0.8303 Winter Rainfall Karoo 5 W37 79 0.723 3.158 0.9001 Winter Rainfall Karoo 6 W1 60 0.560 2.294 0.7401 Winter Rainfall Karoo 6 W2 66 0.563 2.358 0.6542 Winter Rainfall Karoo 6 W22 31 0.606 2.081 0.7279 Tanqua Karoo 7 W13 30 0.415 1.143 0.4485 Tanqua Karoo 7 W14 26 0.668 2.178 0.8042 Tanqua Karoo 7 W17 13 0.742 1.902 0.7348 Tanqua Karoo 7 W19 26 0.493 1.607 0.6616 Tanqua Karoo 8 W15 14 0.170 0.450 0.1448 Tanqua Karoo 8 W16 12 0.473 1.176 0.5105 Tanqua Karoo 8 W18 9 0.065 0.144 0.0429

On a biome scale, Cowling et al. (1989) reported the highest mean species richness values per 1000 m2 for Renosterveld (86), Grassland (82) and Succulent Karoo (74), whereas the biomes with the most species-poor communities were forest (51) and Nama Karoo (47). Within the Fynbos Biome richness in 1000 m² plots ranged from 21 in the southern Fynbos to 142 in the renosterveld (Cowling et al. 1989). Since the Tanqua Karoo is located within the Succulent Karoo Biome, the mean species richness value of 18.6 found in the current study is very low compared to the mean value of 74 species per 1000 m² (Cowling et al. 1989) and mean value of between 42.5 and 49.8 (Anderson & Hoffman 2007), and the maximum value of 113 species per 1000 m² found by Cowling et al. (1989). The mean value determined in this study for the Winter Rainfall Karoo which is also part of the Succulent Karoo, lies between mean values cited by Cowling et al. (1989) and Anderson & Hoffman (2007). However, the 2005 year in which the data in the current study were collected was a very poor rainfall year that would have had a severely negative impact on the species richness (see Aronson & Shmida 1992) since annual and geophytic species constitute a large component of the Succulent Karoo diversity.

101

According to Low and Rebelo (1996) and Mucina et al. (2005), the Mountain Renosterveld is located in the Fynbos Biome even though it has been included in SKEP’s Succulent Karoo delineation. Born et al. (2007) demarcated the area as part of a proposed Greater Cape Floristic Region and mentioned its transitional nature. One 1000 m² plot in the Cape Floristic Region normally holds approximately 65 to 68 species (Cowling et al. 1992, Cowling & Holmes 1992, Procheş et al. 2003). A mean of 66 (range: 31-126) species per 1000 m² was found at 40 fynbos sites distributed throughout the fynbos region (Richardson et al. 1995). In the renosterveld vegetation mean values of 66 (Tilman et al. 1983), 84 (Cowling & Holmes 1992) and 60 (Kongor 2009) have been recorded. The mean species richness of 79 (range: 62-99) determined in this study compares well with the mean values reported for renosterveld. Mean species richness for sclerophyllus shrublands and woodlands from the South West Botanical Province in Western Australia is about 69 species which is comparable with fynbos values, whereas chaparral of the California Floristic Province is poorer in species with a mean of 30 species per 1000 m² (Cowling et al. 1992).

Table 7.2 Mean values and significance for species richness, species evenness, Shannon index and Simpson index within the Mountain Renosterveld, Winter Rainfall Karoo and Tanqua Karoo vegetation groups. Within a column values with the same superscript do not differ significantly at α = 0.05

Vegetation group Species Evenness (E) Shannon Simpson richness index (H’) index (D) Mountain Renosterveld 79.0a 0.65a 2.84a 0.81a Tanqua Karoo 18.6b 0.43b 1.27b 0.48b Winter Rainfall Karoo 64.6c 0.59a 2.42a 0.75a

Using seven different plots sizes (1 m², 5 m², 10 m², 20 m², 50 m², 100 m², 1000 m²) comparisons in species richness were made across the range of vegetation groups (Mountain Renosterveld, Winter Rainfall Karoo and Tanqua Karoo). A one-way ANOVA, indicated that there was a significant difference between Winter Rainfall Karoo and Mountain Renosterveld only at the 1000 m² plot size (Table 7.3). Significant differences were found between the Tanqua Karoo and Winter Rainfall Karoo as well as Mountain Renosterveld and Tanqua Karoo for all smaller plot sizes (Table 7.3). The fact that a significant difference between the Winter Rainfall Karoo and Mountain Renosterveld was only encountered at the 1000 m² (0.1 ha) plot size, indicates the importance of surveying relatively large plots to pick up differences in species richness. Therefore, most phytosociological data collected in South Africa at plot sizes of 200 m² or smaller, would be insufficient to pick up differences in species richness between vegetation groups. Although a study of 0.1 ha data from local inventories each sampled intensively on single occasions at a suite of scattered locations, provides data

102

for comparisons of variation in species richness across geographical space (Clinebell et al. 1995) it remains a local (alpha) scale study, likely to retain signal factors that vary measurably on local scales, such as soil nutrient status (Whittaker et al. 2001).

Table 7.3 Mean species richness values for a range of plot sizes within three vegetation groups in the Hantam-Tanqua-Roggeveld subregion. Within a column values with the same superscript do not differ significantly at α = 0.05

Vegetation group 1 m² 5 m² 10 m² 20 m² 50 m² 100 m² 1000 m² plot plot plot plot plot plot plot Mountain Renosterveld 12.9a 21.6 a 31.1 a 38.3 a 39.9 a 49.7 a 79.1 a Tanqua Karoo 2.2b 3.5 b 5.4 b 6.6 b 7.3 b 8.9 b 18.6 b Winter Rainfall Karoo 9.5 a 16.2 a 25.0 a 30.9 a 32.5 a 41.1 a 64.6 c

Evenness (E) is a measure of the ratio of observed diversity to maximum diversity (Magurran 1988). A 13-fold range of evenness values was obtained for the 1000 m² (0.1 ha) plots across the entire study area. The values ranged from 0.065 in the Tanqua Karoo (plot W18) to 0.820 in the Mountain Renosterveld (plot W28) (Table 1). Evenness is constrained between 0 and 1.0 with 1.0 representing a situation in which all species are equally abundant (Magurran 1988). In all except two cases, evenness was found to be larger than 0.4 (Table 7.1). Species evenness in the Mountain Renosterveld and Winter Rainfall Karoo was significantly higher than in the Tanqua Karoo (Table 7.2). The evenness values for the Tanqua Karoo plots were generally less than 0.5 except for survey plot W14 (0.668) and plot W17 (0.742) (Table 7.1), whereas evenness values for the Winter Rainfall Karoo and Mountain Renosterveld ranged from 0.501 (plot W12) to 0.820 (plot W28) (Table 7.1).

Both the Shannon index of diversity (H’) and the Simpson index (D) take evenness and species richness into account. The rankings of these two measures of diversity were fairly similar and the relationship between these parameters in the study area could best be described by a logarithmic equation (y = 0.3048 ln(x) + 0.4915; r2 = 0.8816). The Shannon index assumes that individuals are randomly sampled from an indefinitely large (that is an effectively infinite) population and that all species in the community are accounted for in the sample (Magurran 1988). However, it is likely that several species in the community will have been missed in the sampling effort and thus, the number of species found in the sample must be regarded as the lower bound to the number of species in the population (Heltshe & Forrester 1983). In general, the Shannon diversity index ranges between 1.5 and 3.5 and only rarely surpasses 4.5 (Magurran 1988). The original Simpson index (D) was referred to as a dominance measure since it was weighted towards the abundances of the commonest species rather than providing a measure of species diversity (Magurran 1988). In this study the complement of the original dominance measure (1 – D) was used to provide an estimate of diversity.

103

The indices derived in this study are believed to be underestimates of the potential values because of poor rainfall conditions in 2005 when the data were collected. The values for the Shannon index ranged 26-fold across the study area from 0.144 (plot W18, Tanqua Karoo) to 3.743 (plot W28, Mountain Renosterveld) (Table 7.1), however, all values, except for two survey plots (W15 and W18, both Tanqua Karoo), were > 1.0. The mean Shannon index value for the Tanqua Karoo was significantly lower than that for the Winter Rainfall Karoo and Mountain Renosterveld (Table 7.2). No significant difference was found for the Shannon index between the latter two vegetation groups.

The range of Simpson index values found for this study varied from 0.0429 (plot W18, Tanqua Karoo) to 0.9439 (plot W28, Mountain Renosterveld) (Table 7.1). The mean Simpson index values of the Mountain Renosterveld and Winter Rainfall Karoo were significantly higher than those of the Tanqua Karoo, however, there was no significant difference in the Simpson index values between the Mountain Renosterveld and Winter Rainfall Karoo (Table 7.2).

To analyse the patterns with respect to the data of the seven plot sizes for the 40 survey plots simultaneously, a Principal Co-ordinate Analysis (PCoA) was done. The resulting PCoA shows three distinct clusters (Figure 7.2). The first axis on the ordination diagram can be interpreted as an aridity gradient, which largely coincides with an altitudinal gradient. The cluster formed on the left hand side of the ordination represents a selection of Mountain Renosterveld plots and includes plots W5 and W6 which were surveyed on the Gannaga Mountain Pass (Pteronia glauca – Euphorbia decussata Escarpment Karoo) in the transition between the Mountain Renosterveld and Tanqua Karoo vegetation groups. This cluster of plots represents the high-lying vegetation, generally between 1020 m and 1421 m above sea level, with a high plant cover of 75 - 98% and sandstone of the Waterford Formation (Johnson et al. 2006) present in the environment. The analysis of the species-area curves for these plots revealed a combination of steep slopes and high y-intercept values for most plots (Chapter 6). The cluster formed on the right hand side of the ordination scatterplot represents the Tanqua Karoo survey plots (Figure 7.2). The low species richness values with respect to all seven plot sizes clearly distinguish this cluster from the other clusters. These survey plots are characterised by a low plant cover, generally between 10 and 60%, found on shales at an altitude of 374 to 647 m above sea level. Slopes of species-area curves for these plots were shallow and the values for the y-intercept low (Chapter 6). With the exception of the two plots on the Roggeveld escarpment all other plots surveyed in the Winter Rainfall Karoo belonged to the central cluster. However, several Mountain Renosterveld plots also fell within this cluster.

104

Figure 7.2 A Principal Co-ordinate Analysis (PCoA) scatterplot of species richness values for seven plot sizes on 40 plots surveyed in 2005 in the Hantam, Tanqua and Roggeveld areas. Numbers denote the survey plot numbers.

The ordination confirms that the Tanqua Karoo vegetation is very different from the vegetation in the rest of the study area and provides for interesting observations with respect to the Mountain Renosterveld of the subregion. The left hand cluster is comprised of the ‘true’ Mountain Renosterveld plots separating them from the centre cluster. This ‘true’ Mountain Renosterveld occurs on sandstones of the Waterford Formation (Johnson et al. 2006) and is assumed to be more closely related to the renosterveld of the Fynbos Biome. The centre cluster comprised of the Winter Rainfall Karoo plots, however also includes Mountain Renosterveld plots, indicating a strong link with the Succulent Karoo Biome.

7.5 Conclusions

Forty Whittaker sample plots were surveyed throughout the Mountain Renosterveld, Winter Rainfall Karoo and Tanqua Karoo vegetation groups of the Hantam, Tanqua and Roggeveld regions.

Species richness in the 1000 m² plots ranged from nine to 100 species per sampled 1000 m² among the 40 plots. Mean species richness was significantly lowest in the Tanqua Karoo and significantly highest in the Mountain Renosterveld vegetation. Mean species richness values of the Tanqua Karoo compared poorly with Succulent Karoo values cited in the literature. Species richness values for the Mountain Renosterveld compared well with values obtained for the renosterveld of the Fynbos Biome. However, the 2005 year in which the survey was conducted was a very poor rainfall year which would have had a negative impact on annual and geophyte species that make up a large component of Succulent Karoo and Fynbos Biome diversity.

Evenness values for the Tanqua Karoo were generally less than 0.5 and significantly less than the 0.501 to 0.820 for the Winter Rainfall Karoo and Mountain Renosterveld vegetation.

105

Similarly, the Shannon indices for the Tanqua Karoo (0.144 to 1.902) were significantly lower than those for the Winter Rainfall Karoo and Mountain Renosterveld (2.066 to 3.743). The Simpson indices varied from 0.0429 to 0.9439 across the study area and also indicated that the values for the Tanqua Karoo were significantly lower than those for the other two vegetation groups. No significant differences could be found between the Winter Rainfall Karoo and Mountain Renosterveld regarding either evenness, Shannon’s or Simpson’s index. In contrast, the species richness in 1000 m2 plots of the Mountain Renosterveld was significantly higher than that of the Winter Rainfall Karoo, which in turn was significantly higher than that of the Tanqua Karoo. At all other plot sizes (1 m², 5 m², 10 m², 20 m², 50 m², and 100 m²) significant differences were found in species richness between each plot size for the Winter Rainfall Karoo and Tanqua Karoo and the Mountain Renosterveld and Tanqua Karoo vegetation groups, but not between the Winter Rainfall Karoo and Mountain Renosterveld.

Species richness is still commonly used as a measure of diversity, however, species richness provides relatively little information and is a relatively coarse measure of diversity. If used on its own, and only measured at one plot size then care should be taken to determine the size of the plot needed to detect differences. Information gained from species counts is far more valuable if determined at different plot sizes because this allows the construction of species- area curves or the application of ordination techniques.

A Principal Co-ordinate Analysis (PCoA) of the sample plots, using species richness data for the seven plot sizes produced three distinct clusters. The one cluster was formed by the sparsely vegetated, arid, low-altitude Tanqua Karoo plots, while another cluster was formed by the Mountain Renosterveld plots found predominantly at high altitudes on sandstone of the Waterford Formation with a high vegetation cover. The centre cluster was formed by a combination of Winter Rainfall Karoo and Mountain Renosterveld vegetation groups. This ordination indicates that the way in which the species accumulate with an increase in size sampled of some of the Mountain Renosterveld plots is similar to the species accumulation pattern of the Succulent Karoo plots. However, there is a group of Mountain Renosterveld plots in the Hantam-Tanqua-Roggeveld subregion where the species accumulation pattern differs and these plots could possibly show affinities to the Fynbos Biome.

Information on species richness values and diversity indices gained in the study can be used to guide conservation efforts in the Hantam-Tanqua-Roggeveld subregion. For example, associations 7 and 8 are two of the most species poor associations, with the lowest levels of diversity, yet, they dominate the area conserved within the Tankwa Karoo National Park. The Tankwa Karoo National Park also includes, to a very limited extent, a small area of associations 2 and 4 which both have a high level of diversity. These areas could be notably expanded and the goal of the Tankwa Karoo National Park could be to include areas within

106

associations 1 and 6. This would then actively conserve six of the eight associations in the Hantam-Tanqua-Roggeveld subregion, including some of the most diverse associations in the region.

7.6 Acknowledgements

7.7 References

ACOCKS, J.P.H., 1953. Veld types of South Africa. Memoirs of the Botanical Survey of South Africa 28, 1-192. ACOCKS, J.P.H., 1988. Veld types of South Africa. 3rd edn. Memoirs of the Botanical Survey of South Africa 57, 1-146. ANDERSON, P.M.L. AND HOFFMAN, M.T., 2007. 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 70, 686-700. ARONSON, J. AND SHMIDA, A., 1992. Plant species diversity along a Mediterranean-desert gradient and its correlation with interannual rainfall fluctuations. Journal of Arid Environments 23, 235-247. BORN, J., LINDER, H.P. AND DESMET, P., 2007. The Greater Cape Floristic Region. Journal of Biogeography 34, 147-162. CILLIERS, C., THERON, H., RÖSCH, H. AND LE ROUX, A., 2002. Succulent Karoo Ecosystem Plan, Sub-regional report, Hantam/Tanqua/Roggeveld. Succulent Karoo Ecosystem Plan report. CLINEBELL, H.R.R., PHILLIPS, O.L., GENTRY, A.H., STARK, N. AND ZUURING, H., 1995. Prediction of neotropic tree and liana richness from soil and climatic data. Biodiversity and Conservation 4, 56-90. COWLING, R.M., GIBBS RUSSEL, G.E., HOFFMAN, M.T. AND HILTON-TAYLOR, C., 1989. Patterns of plant species diversity in southern Africa. In: B.J. Huntley (Ed.). Biotic

107

diversity in southern Africa. Concepts and Conservation, pp. 19-50. Cape Town, Oxford University Press. COWLING, R.M. AND HILTON-TAYLOR, C., 1994. Patterns of plant diversity and endemism in southern Africa: an overview. In: B.J. Huntley (Ed.). Botanical diversity in Southern Africa. Strelitzia 1, pp. 31-52. National Botanical Institute, Pretoria. COWLING, R.M. AND HOLMES, P.M., 1992. Endemism and speciation in a lowland flora from the Cape Floristic Region. Biological Journal of the Linnean Society 47, 367- 383. COWLING, R.M., HOLMES, P.M. AND REBELO, A.G., 1992. Plant diversity and endemism. In: R.M. Cowling (Ed.). The Ecology of Fynbos: Nutrients, fire and diversity, pp. 62- 112. Oxford University Press, Cape Town. CRITICAL ECOSYSTEM PARTNERSHIP FUND, 2003. Ecosystem Profile: The Succulent Karoo hotspot, Namibia and South Africa. Critical Ecosystem Partnership Fund report. DIELS, L., 1909. Formationen und Florenelemente im nordwestlichen Kapland. Botanische Jahrbücher 44, 91-124. FRANCIS, M.L., FEY, M.V., PRINSLOO, H.P., ELLIS, F., MILLS, A.J. AND MEDINSKI, T.V., 2007. Soils of Namaqualand: Compensations for aridity. Journal of Arid Environments 70, 588-603. GOOD, R., 1947. The geography of flowering plants. Longmans, Green & Co., New York. HELLMANN, J.J. AND FOWLER, G.W., 1999. Bias, precision, and accuracy of four measures of species richness. Ecological Applications 9, 824-834. HELTSHE, J.F. AND FORRESTER, N.E., 1983. Estimating species richness using the Jackknife Procedure. Biometrics 39, 1-11. HILTON-TAYLOR, C., 1994. Western Cape Domain (Succulent Karoo). In: S.D. Davis, V.H. Heywood and A.C. Hamilton (Eds). Centres of plant diversity. A guide and strategy for their conservation, pp. 201-203. IUCN Publications Unit, Cambridge. JOHNSON, M.R., VAN VUUREN, C.J., VISSER, J.N.J., COLE, D.I., WICKENS, H. DE.V., CHRISTIE, A.D.M., ROBERTS, D.L. AND BRANDL, G., 2006. Sedimentary rocks of the Karoo Supergroup. In: M.R. Johnson, C.R. Anhaeusser and R.J. Thomas (Eds). The geology of South Africa, pp. 461-500. The Geological Society of South Africa, Johannesburg/Council for Geoscience, Pretoria. JÜRGENS, N., 1997. Floristic biodiversity and history of African arid regions. Biodiversity and Conservation 6, 495-514. KONGOR, R.Y., 2009. Plant response to habitat fragmentation: clues from species and functional diversity in three Cape lowland vegetation types of South Africa. PhD thesis, University of Stellenbosch, Stellenbosch. LOW, A.B. AND REBELO, A.G., 1996. Vegetation of South Africa, Lesotho and Swaziland. Department of Environmental Affairs and Tourism, Pretoria.

108

MAGURRAN, A.E., 1988. Ecological Diversity and its measurement. University Press, Cambridge. MARLOTH, R., 1908. Das Kapland, insonderheit das Reich der Kapflora, das Waldgebiet und die Karoo, pflanzengeografisch dargestellt. Wissenschaftliche Ergebnisse der Deutscher Tiefsee-Expedition ‘Waldivia’, 1898 – 1899. 2, T. 3, Fischer, Jena. MILTON, S.J., YEATON, R.I., DEAN, W.R.J. AND VLOK, J.H.J., 1997. Succulent Karoo. In: R.M. Cowling, D.M. Richardson and S.M. Pierce (Eds). Vegetation of southern Africa, pp. 99-129. Cambrige University Press, Cambridge. MOUILLOT, D. AND LEPRÊTRE, A., 1999. A comparison of species diversity estimators. Research on Population Ecology 41, 203-215. MUCINA, L., JÜRGENS, N., LE ROUX, A., RUTHERFORD, M.C., SCHMIEDEL, U., ESLER, K.J., POWRIE, L.W., DESMET, P.G. AND MILTON, S.J., 2006. Succulent Karoo Biome. In: L. Mucina and M.C. Rutherford (Eds). The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19, pp. 220-299. South African National Biodiversity Institute, Pretoria. MUCINA, L. AND RUTHERFORD, M.C. (Eds), 2006. The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19. South African National Biodiversity Institute, Pretoria. MUCINA, L., RUTHERFORD, M.C. AND POWRIE, L.W. (Eds), 2005. Vegetation map of South Africa, Lesotho and Swaziland, 1 : 1 000 000 scale sheet maps. South African Biodiversity Institute, Pretoria. PODANI, J., 2001. SYN-TAX 2000 Computer programs for data analysis in ecology and systematics. Scientia publishing, Budapest. PROCHEŞ, Ş., COWLING, R.M. AND MUCINA, L., 2003. Species-area curves based on relevé data for the Cape Floristic Region. South African Journal of Science 99, 74- 476. RICHARDSON, D.M., COWLING, R.M., BOND, W.J., STOCK, W.D. AND DAVIS, G.W., 1995. Links between biodiversity and ecosystem function in the Cape Floristic Region. In: G.W. Davis and D.M. Richardson (Eds). 1995. Mediterranean-type ecosystems: The function of biodiversity, pp. 285-333. Springer-Verlag, Berlin Heidelberg. RUTHERFORD, M.C. AND R.H. WESTFALL., 1994. Biomes of Southern Africa. An objective characterisation. Memoirs of the Botanical Survey of South Africa 63, 1-94. SCHULZE, R.E., 1997. South African Atlas of Agrohydrology and – Climatology. Water Research Commission, Pretoria, Report TT82/96. SHMIDA, A., 1984. Whittaker’s plant diversity sampling method. Israel Journal of Botany 33, 41-46. STIRLING, G. AND WILSEY, B., 2001. Empirical relationships between species richness, evenness, and proportional diversity. The American Naturalist 158, 286-299. TILMAN, D., BOND, W.J., CAMPBELL, B.M., KRUGER, F.J., LINDER, H.P., SCHOLZ, A., TAYLOR, H.C. AND WITTER, M., 1983. Origin and maintenance of plant species

109

diversity. In: J.D. Day (Ed.). Mineral nutrients in Mediterranean ecosystems. South African National Scientific Programmes. Report No. 71. Council for Scientific and Industrial Research, Pretoria. VAN DER MERWE, H., VAN ROOYEN, M.W., AND VAN ROOYEN, N., 2008a. Vegetation of the Hantam-Tanqua-Roggeveld subregion, South Africa. Part 1. Fynbos Biome related vegetation. Koedoe 50, 61-71. VAN DER MERWE, H., VAN ROOYEN M.W., AND VAN ROOYEN, N., 2008b. Vegetation of the Hantam-Tanqua-Roggeveld subregion, South Africa. Part 2. Succulent Karoo Biome related vegetation. Koedoe 50, 160-183. VAN WYK, A.E. AND SMITH, G.F. (Eds), 2001. Regions of Floristic Endemism in Southern Africa: A review with emphasis on succulents, pp. 1- 199. Umdaus Press, Pretoria. WEATHER BUREAU, 1998. Climate of South Africa. Climate statistics up to 1990. WB 42. Government Printer, Pretoria. WEIMARCK, H., 1941. Phytogeographical groups, centres and intervals within the Cape flora. Lunds Universitets Årsskrif Avd. 2. 37, 1-143. WHITTAKER, R.H., 1977. Evolution of species diversity on land communities. Evolutionary Biology 10, 1-67. WHITTAKER, R.J. WILLIS, K.J. AND FIELD, R., 2001. Scale and species richness: towards a general, hierarchical theory of species diversity. Journal of Biogeography 28, 453- 470. WILSEY, B.J. AND POTVIN, C., 2000. Biodiversity and ecosystem functioning: importance of species evenness in an old field. Ecology 81, 887-892.

110

Chapter 8

Plant diversity in the Hantam-Tanqua-Roggeveld, Succulent Karoo, South Africa: Life form spectra

Abstract

The Hantam-Tanqua-Roggeveld subregion is situated in a predominately winter rainfall area of the Succulent Karoo and Fynbos Biomes of South Africa. These biomes are both recognised as global hotspots of diversity with high species diversities at local and regional scales.

Life forms are closely related to climate and habitat and are often used as a trait to group species growing under certain environmental conditions together. In general, the eight plant associations found in the study area were dominated by chamaephyte, cryptophyte (geophyte) and therophyte species.

At a species level, phanerophyte contributions were lower in the Winter Rainfall Karoo and Tanqua Karoo than in the Mountain Renosterveld. Cryptophyte contributions were higher in the Mountain Renosterveld than in the Tanqua Karoo. Similar hemicryptophyte contributions were found in the Mountain Renosterveld and Winter Rainfall Karoo vegetation with slightly higher values obtained for the Tanqua Karoo. Therophyte contributions were similar in the Mountain Renosterveld and the Winter Rainfall Karoo and significantly less in the Tanqua Karoo.

On a vegetation cover level, the phanerophyte contributions were significantly highest in the Mountain Renosterveld and lowest in the Tanqua Karoo. Chamaephyte cover varied greatly with the cover in the Mountain Renosterveld and Winter Rainfall Karoo significantly higher than in the Tanqua Karoo. Hemicryptophyte, liana and parasite cover contributions were the least. Cryptophyte cover was high for Mountain Renosterveld and Winter Rainfall Karoo and lower for the Tanqua Karoo as was therophyte cover yet no statistical significance was found.

Since the Succulent Karoo Biome is known for its succulents, the degree of succulence was compared among the vegetation associations and broad vegetation groups. Succulent species were found mostly in the chamaephyte, hemicryptophyte and therophyte life form categories. The contribution of succulents on a species level was significantly lowest in the Mountain Renosterveld and highest in the Winter Rainfall Karoo and Tanqua Karoo vegetation, with a higher than expected vegetation cover for the succulent life form in a strongly Mountain Renosterveld (Fynbos Biome) association, indicating its transitional nature between the two biomes.

111

A comparison of life form spectra with another site in the Succulent Karoo, Goegap Nature Reserve, produced more similar patterns than a comparison with another winter rainfall desert, the Mojave Desert. The Mountain Renosterveld life form spectra compared poorly to another Fynbos Biome site at Swartboskloof.

Keywords: Fynbos Biome, Mountain Renosterveld, succulence, Succulent Karoo Biome, Tanqua Karoo, Whittaker plots, Winter Rainfall Karoo

112

8.1 Introduction

The Hantam-Tanqua-Roggeveld subregion was one of the subregions delineated for management purposes during the Succulent Karoo Ecosystem Plan (SKEP) initiative in 2002. This initiative was launched to identify and generate consensus for a 20-year conservation and sustainable land use strategy for the Succulent Karoo hotspot of biodiversity (Conservation International – website).

Three biomes namely: the Fynbos, Succulent Karoo and Nama Karoo Biomes (Rutherford and Westfall 1994), meet in the vicinity of the Hantam-Tanqua-Roggeveld subregion. The transitional nature of the area has lead to some controversy as to whether the vegetation, especially that of the Roggeveld, should be classified as belonging to the Fynbos (Low & Rebelo 1996, Mucina et al. 2005, Rebelo et al. 2006, Rutherford et al. 2006) or Succulent Karoo Biome (Hilton-Taylor 1994, Jürgens 1997, Van Wyk & Smith 2001) or even the Nama Karoo Biome (Acocks 1988, Rutherford & Westfall 1994). A recent paper by Born et al. (2007) on the validity of the recognition of a Greater Cape Floristic Region also highlights the area’s transitional nature.

Globally, there are few other areas that can claim to be as biologically distinct as the Succulent Karoo Biome (Milton et al. 1997, Cowling & Pierce 1999, Mucina et al. 2006), with this biome being recognised by the IUCN as one of the global hotspots of diversity (Myers et al. 2000, Critical Ecosystem Partnership Fund 2003). The Succulent Karoo Biome is an arid area and plant species diversity at both local and regional scales is reported to be the highest recorded for any arid region in the world (Cowling et al. 1989).

For the most part the Cape Floristic Region (CFR) comprises the Fynbos Biome. The Cape flora’s unique species composition has led it to be recognised as one of the world’s floristic kingdoms (Good 1947), on par with much larger regions (Rebelo et al. 2006). This region is also recognised as a global hotspot of biodiversity with one of the highest species densities and levels of endemism, at both local and regional scales, for temperate or tropical continental regions (Cowling et al. 1989, 1992, Cowling & Hilton-Taylor 1994).

The most common, parsimonious and accepted plant life form classification is Raunkiaer’s (1934) (Van Rooyen et al. 1990, Pavón et al. 2000) who suggested that the location of a plant’s renewal buds, as differentiated in various life forms, best expresses its adaptation to the unfavourable season for plant life (Danin & Orshan 1990). Because plant architecture and physiognomy can often be linked to the different climatic or environmental conditions under which a species grows and certain life forms are restricted to growing in particular habitats (Barbour et al. 1999), a life form spectrum of a region can give preliminary information concerning the habitat and the adaptive suite of plant traits occurring there.

113

Although life form spectra based on the total flora of a sufficiently large area are useful in the indication of the general or prevailing phytoclimate within such an area, the effects of local environments (microclimates and edaphic conditions) are best revealed when the spectra are modified by the use of quantitative data on the roles of various species in the respective local communities (Cain 1950, Danin & Orshan 1990). Life form studies have shown that the distribution and abundance of different plant life forms have well defined limits along altitudinal gradients and suggest that environmental heterogeneity in semi-arid environments contributes to the diversity of life forms, thus affecting community physiognomy and structure (Pavón et al. 2000). Additionally, assessing effects of climate change by life form or plant functional types (PFTs) is increasingly being applied to identify future trends in ecosystem structure (Broennimann et al. 2006).

The aim of this study was to compare life form spectra at a species as well as a vegetation cover level at the scale of three broad vegetation groups and eight plant associations found in the Hantam-Tanqua-Roggeveld study area. A degree of succulence was calculated and also used in an attempt to provide clarity on the Succulent Karoo vs. Fynbos Biome affinities of the vegetation in the Hantam-Tanqua-Roggeveld subregion.

8.2 Study area

The Hantam-Tanqua-Roggeveld subregion is situated in the predominantly winter rainfall area of the Northern and Western Cape Provinces of the Republic of South Africa (Figure 8.1). Although the rains fall mainly in winter, a few summer thunderstorms do occur. Rainfall ranges from 25 mm in parts of the Tanqua Karoo to 467 mm per year, the maximum recorded for Sutherland in the Roggeveld area (Weather Bureau 1998).

Rocks of the Ecca Group cover most of the area and include sediments of the Koedoesberg Formation (sandstone and shale) and the Tierberg Formation (shale), (Council for Geoscience 2008). The Dwyka Group consisting of tillite, sandstone, mudstone and shale crops out in the west of the study area with the mudstones of the Beaufort Group found on the eastern side of the study area (Council for Geoscience 2008). Igneous rock intrusions of dolerite occur throughout the region.

The Hantam is characterised by shallow lithosols and duplex soils, while red structural and vertic clays are produced by the scattered dolerite intrusions. Soils of the Tanqua Karoo are shallow lithosols that often include deep unconsolidated deposits in the alluvial parts or desert pavement. The soils of the great escarpment and Hantam Mountains are shallow stony lithosols with the occasional lowlands comprised of duplex soils (Francis et al. 2007).

114

Figure 8.1 The eight plant associations found in the Hantam-Tanqua-Roggeveld

subregion (after Van der Merwe et al. 2008a, 2008b).

The vegetation of the Succulent Biome is dominated by dwarf shrubs, many of them being leaf succulents (Werger 1986, Van Rooyen et al. 1990, Rutherford & Westfall 1994, Milton et al. 1997, Mucina et al. 2006). Spectacular autumn and spring floral displays are known to a greater or lesser degree in the Hantam-Tanqua-Roggeveld. A wide range of geophytes (e.g. Iridaceae and Hyacinthaceae), succulents (e.g. Aizoaceae, Crassulaceae and Euphorbiaceae) and annual plants (e.g. Asteraceae, Brassicaceae and Scrophulariaceae) flower on fallow lands but also in the undisturbed natural vegetation (Van Wyk & Smith 2001, Van Rooyen 2002).

The vegetation of the Fynbos Biome is characterised by the co-dominance of usually fine- leaved, sclerophyllous, evergreen shrubs, dwarf shrubs and hemicryptophytes (Rutherford & Westfall 1994). Fynbos, renosterveld and strandveld are the three main vegetation groups of the Fynbos Biome, with only the renosterveld group occurring in the study area. Fynbos vegetation is confined to sandy infertile soils and characterised by the universal presence of

115

restioids, a high cover of ericoid shrubs and the common occurrence of overstorey proteoid shrubs (Cowling et al. 1997). Renosterveld is described as an evergreen, fire-prone shrubland/grassland occurring on relatively fertile clay rich shale and granite derived soils (Boucher & Moll 1981, Cowling et al. 1997).

Three of Acocks’s (1953, 1988) veld types are included in the study area namely: Mountain Renosterveld, Succulent Karoo and Western Mountain Karoo. These veld types generally fall into what Low and Rebelo (1996) named the Escarpment Mountain Renosterveld, Lowland Succulent Karoo and Upland Succulent Karoo respectively. The latest vegetation map of South Africa (Mucina et al. 2005, Mucina & Rutherford 2006) identified 12 vegetation types within the area. The eight major plant associations recognised by Van der Merwe et al. (2008a, 2008b) in a recent phytosociological classification and mapping study in the study area form the basis of the present investigation. The eight associations were grouped into three vegetation groups of which two, i.e. the Winter Rainfall Karoo and Tanqua Karoo, form part of the Succulent Karoo Biome and one, i.e. the Mountain Renosterveld, is part of the renosterveld of the Fynbos Biome (Table 8.1).

8.3 Materials and methods

Field surveys were conducted in 2005 using Whittaker’s plant diversity plot technique (Shmida 1984) with the methodology described by Shmida (1984) being slightly modified (see Chapter 6 for more detail). The 40 Whittaker sample plots surveyed were selected to cover all eight plant associations distinguished in the area (Van der Merwe et al. 2008a, 2008b) (Figure 8.1). At each site environmental data such as altitude, aspect, slope, position on the slope, geology, topography, percentage stone and stone size, soil type and colour, drainage, erosion, trampling and soil compaction were noted.

All the species encountered in the 1000 m² survey plot were noted and a cover value assigned to each species. These species were then classified into broad life form categories following Raunkiaer’s (1934) classification as modified by Mueller-Dombois and Ellenberg (1974, see Appendix 1).

The relative contribution of each life form, at a species level and at a vegetation cover level, was calculated for each plot. Comparisons of the life forms were made across the eight plant associations as well as for the three broad vegetation groups (Mountain Renosterveld, Winter Rainfall Karoo and Tanqua Karoo) present in the region. An analysis of variance was performed using the GLM (General Linear Model) Procedure in SAS (SAS® Version 8.2 running on an IBM z9 mainframe computer under z/VM 5.3.0 at the University of Pretoria). The assumption that the variances among treatment levels were constant was violated and thus the data were transformed. A power transformation test indicated that the appropriate

116

transformation would be of the form: log10 (life form + 1). These transformed life form values were then used in the statistical analysis. The complete SAS outputs are included in Appendix 2 for the association level output and Appendix 3 for the vegetation group level output.

Statistical analyses of the data to investigate a degree of succulence were conducted using the STATISTICA computer package (StaSoft, Inc. Version 8, 2300 East 14th Street, Tulsa, OK 74104) (ANOVA’s – Kruskal-Wallis test) since the data were not normally distributed.

8.4 Results and discussion

Forty Whittaker plots were surveyed throughout the eight plant associations present in the study area (Table 8.1) (Van der Merwe et al. 2008a, 2008b). A varying number of plots were surveyed in each association due to differences in size and environmental heterogeneity of the associations (Table 8.1). Plant association 1 was co-dominated by cryptophyte, therophyte and chamaephyte species (Table 8.1). A similar dominance structure was found for all other associations of Mountain Renosterveld as well as the associations of the Winter Rainfall Karoo (associations 2, 3, 4, 5 and 6, Table 8.1). The associations in the Tanqua Karoo group had a different dominance structure. Plant association 7 was dominated by chamaephyte species, while association 8 was co-dominated by chamaephyte and cryptophyte species (Table 8.1).

Table 8.1 Mean percentage contribution per life form on a species level in eight plant associations, belonging to three broad vegetation groups, in the Hantam-Tanqua-Roggeveld subregion. The mean number of species per life form is indicated in brackets

Plant association Broad No. Mean percentage contribution by species vegetation of (Mean number of species per life form) group plots P Ch H C T L Par 1 Rosenia Mountain 5 3.0 24.7 10.7 33.2 27.0 1.2 0.2 oppositifolia Renosterveld (2.2) (19.0) (8.4) (25.6) (20.4) (1.0) (0.2) Mountain abc a ab b c ab a Renosterveld 2 Dicerothamnus Mountain 10 5.4 26.3 8.2 30.6 26.8 2.5 0.3 rhinocerotis Renosterveld (4.0) (20.4) (6.3) (23.9) (20.9) (1.9) (0.2) Mountain c a ab b c c a Renosterveld 3 Passerina truncata Mountain 2 3.5 34.6 7.7 27.0 26.1 1.1 0.0 Mountain Renosterveld (3.0) (31.0) (7.0) (24.5) (23.0) (1.0) (0.0) Renosterveld abc a ab b c abc a 4 Pteronia glauca – Winter Rainfall 4 6.1 34.6 8.1 25.4 22.3 3.5 0.0 Euphorbia Karoo (5.3) (27.8) (7.0) (21.5) (18.0) (3.0) (0.0) decussata c a ab b c c a Escarpment Karoo

117

Table 8.1 (continued)

5 Eriocephalus Winter Rainfall 9 1.8 29.7 7.1 27.3 32.0 2.2 0.0 purpureus Hantam Karoo (1.2) (19.0) (4.2) (16.7) (18.6) (1.3) (0.0) Karoo ab a a b c bc a 6 Pteronia glomerata Winter Rainfall 3 1.1 30.4 10.6 24.9 31.4 1.6 0.0 Roggeveld Karoo Karoo (0.7) (15.0) (5.3) (14.3) (16.0) (1.0) (0.0) a a ab b c abc a 7 Aridaria noctiflora Tanqua Karoo 4 3.7 47.6 13.9 18.2 15.6 1.0 0.0 Tanqua and (0.8) (10.3) (3.3) (5.0) (4.3) (0.3) (0.0) Loeriesfontein abc a b a b a a Karoo 8 Stipagrostis obtusa Tanqua Karoo 3 3.7 39.0 14.0 32.4 4.8 6.1 0.0 Central Tanqua (0.3) (4.3) (1.7) (4.0) (0.7) (0.7) (0.0) Grassy Plains ab a ab b a c a P = phanerophyte, Ch = chamaephyte, H = hemicryptophyte, C = cryptophyte, T = therophyte, L = liana and Par = parasite. Within a column, values with the same letters do not differ significantly at α = 0.05. Letters should only be compared within a life form.

An analysis of variance using the GLM (General Linear Model) Procedure in SAS on the transformed data indicated that the interaction between the main factors was significant and thus the interpretation of the results has to done on the interaction level (Table 8.2).

Table 8.2 Summary output of the analysis of variance (GLM) with association and life form (species level) as main factors

Source of variation Degrees of Sum of Mean square F value P value freedom squares Association 7 0.8205 0.1172 2.31 0.0273 Life form (species 6 58.1393 9.6899 190.83 < 0.0001 level) Association x Life 42 6.4168 0.1528 3.01 < 0.0001 form (species level)

The total number of species per life form encountered in the 1000 m² Whittaker plots expressed as a percentage of the total number of species indicated that the contributions of phanerophyte species were low (1.1 – 6.1%) with a significant difference (p < 0.01 in all instances) found between associations 2 and 4 and associations 5, 6 and 8. In contrast, contributions by chamaephyte species were generally high (24.7 – 47.6%) with no significant difference found between them at an association level. The higher number of chamaephyte species found relative to phanerophyte species in this study could be a case of the succulent species mostly being found among the chamaephyte life form. Hemicryptophyte species’ contributions were similar yet produced a significant difference between association 5 and association 7 (p < 0.05).

Contributions by cryptophyte (geophyte) species were relatively constant throughout and ranged from 18.2% to 33.2%. These high values confirm the high diversity of bulbous plants in

118

both the Fynbos, especially renosterveld, and Succulent Karoo which is a striking feature shared by these two areas (Esler et al. 1999a, Procheş et al. 2005, 2006). Association 7 was marked by a lower cryptophyte contribution which was found significantly different (p < 0.05) from all other associations. It should be noted however, that the data used in this study were collected in 2005, which was a very poor rainfall year. This is expected to have had a marked effect on the number of geophyte and annual species encountered and their contributions to the flora could have been underestimated.

Therophyte (annual) contributions were lowest in the two associations from the Tanqua Karoo and differed significantly (p < 0.05) between association 7 and all other associations as well as between association 8 and all other associations. In general, therophyte dominance indicates the desert nature of the climate in a study area (Withrow 1932, Raunkiaer 1934, Van Rooyen et al. 1990, Fox 1992, Van Rooyen 1999). It is therefore surprising that the therophytes made a significantly smaller contribution in the two Tanqua Karoo associations (associations 7 and 8) which are located in the most arid part of the study area but this is probably a result of the below normal rainfall of the survey year. It has been suggested that in desert and semi-desert areas a relatively predictable seasonal rainfall favours the development of a therophyte flora (Westoby 1980, Cowling & Pierce 1999). Therophytes are believed to be more resistant to summer drought than the hemicryptophytes and geophytes, since the former spend the summer in the form of seeds and the latter in the form of vegetative organs (Danin & Orshan 1990). Raunkiaer (1934) suggested that the biological spectra of Mediterranean-type regions are also characterised by high percentages of therophytes that survive the dry summer in the form of seeds (Danin & Orshan 1990). However, life form spectra with a relatively low percentage of therophytes are known from Mediterranean climates e.g. South Australia and South Africa.

Liana species were present in low numbers in all the plant associations however, a significant difference was found for lianas between association 1 and associations 2, 4 and 8 as well as between association 7 and associations 2, 4, 5 and 8 (p < 0.05). Parasite species were only encountered in plant associations 1 and 2 (Table 8.1).

The GLM Procedure on the transformed life form (species level) and vegetation group data also produced a significant difference at the interaction level (Table 8.3). At the species level, the phanerophyte contribution was significantly lower (p < 0.05) in the Winter Rainfall Karoo and Tanqua Karoo groups than in the Mountain Renosterveld group. Hemicryptophytes made a significantly smaller (p < 0.05) contribution in the Winter Rainfall Karoo than in the Tanqua Karoo, whereas the cryptophyte contribution was significantly (p < 0.05) higher for the Mountain Renosterveld than for the Tanqua Karoo. Therophytes contributed significantly (p < 0.0001) less in the Tanqua Karoo group than in both the Mountain Renosterveld and Winter

119

Rainfall Karoo groups. No significant difference was found at the vegetation group level for chamaephytes, lianas or parasites.

Table 8.3 Summary output of the analysis of variance (GLM) with vegetation group and life form (species level) as main factors

Source of variation Degrees of Sum of Mean square F value P value freedom squares Vegetation group 2 0.5706 0.2853 5.15 0.0064 Life form (species 6 62.1143 10.3524 187.05 < 0.0001 level) Vegetation group x 12 3.7066 0.3089 5.58 < 0.0001 Life form (species level)

The contribution of each life form to the vegetation cover in the various plant associations (Table 8.4) produced different results to those found at a species level (Table 8.1) and once again the statistical analysis indicated a significant interaction between the life forms and associations (Table 8.5).

Table 8.4 Mean percentage cover contribution per life form in eight plant associations, belonging to the three broad vegetation groups, in the Hantam-Tanqua-Roggeveld subregion. The mean cover per life form is indicated in brackets

Plant association Broad No. of Mean percentage contribution by cover vegetation plots (Mean cover per life form) group P Ch H C T L Par 1 Rosenia Mountain 5 16.6 47.1 3.5 14.7 17.8 0.3 0.1 oppositifolia Renosterveld (18.4) (53.1) (3.9) (16.5) (20.0) (0.3) (0.1) Mountain cd abc a a ab a a Renosterveld 2 Dicerothamnus Mountain 10 30.8 33.9 2.91 18.2 13.3 0.8 0.1 rhinocerotis Renosterveld (33.1) (34.7) (3.2) (21.7) (14.5) (0.9) (0.1) Mountain de ab a a ab a a Renosterveld 3 Passerina truncata Mountain 2 38.3 32.1 3.7 11.8 13.6 0.5 0.0 Mountain Renosterveld (38.0) (30.5) (3.5) (11.3) (13.1) (0.5) (0.0) Renosterveld e abc a a ab a a 4 Pteronia glauca – Winter Rainfall 4 37.1 40.0 3.3 9.4 8.8 1.3 0.0 Euphorbia Karoo (42.1) (43.2) (3.7) (10.3) (9.8) (1.4) (0.0) decussata e abc a a a a a Escarpment Karoo 5 Eriocephalus Winter Rainfall 9 0.9 51.4 2.5 8.3 36.5 0.5 0.0 purpureus Hantam Karoo (0.9) (48.1) (2.5) (8.2) (40.9) (0.4) (0.0) Karoo a a a a b a a

120

Table 8.4 (continued)

6 Pteronia glomerata Winter Rainfall 3 0.3 49.2 2.9 11.5 12.3 0.5 0.0 Roggeveld Karoo Karoo (0.2) (51.3) (1.4) (6.2) (8.3) (0.4) (0.0) ab c a a ab a a 7 Aridaria noctiflora Tanqua Karoo 4 1.2 79.0 12.5 3.8 3.6 0.3 0.0 Tanqua and (0.3) (38.0) (5.9) (2.1) (2.1) (0.0) (0.0) Loeriesfontein ab c a a ab a a Karoo 8 Stipagrostis obtusa Tanqua Karoo 3 0.1 25.7 71.2 2.4 0.5 0.1 0.0 Central Tanqua (0.0) (11.6) (53.5) (1.5) (0.3) (0.1) (0.0) Grassy Plains abc c a a ab a A P = phanerophyte, Ch = chamaephyte, H = hemicryptophyte, C = cryptophyte, T = therophyte, L = liana and Par = parasite. Within a column, values with the same letters do not differ significantly at α = 0.05. Letters should only be compared within a life form.

Phanerophytes made a significantly larger contribution to the vegetation cover of associations 1 to 4 than in associations 5 to 8 (p < 0.05), with a significant difference (p < 0.05) between association 1 and associations 3 and 4 also found. Chamaephytes were abundant in all the associations but their contribution to the vegetation cover varied greatly. A significant difference in chamaephyte cover was found between associations 2 and 5 and associations 6, 7 and 8 (p < 0.05).

Hemicryptophytes, lianas and parasites contributed the least to the vegetation cover in all the associations except for association 8 where the hemicryptophyte cover, as a result of the dominance and high cover of the grass Stipagrostis obtusa, was higher than in all other associations yet no significant differences were found. The cryptophytes’ contribution to the vegetation cover varied from 2.4% to 18.2% with no significant differences found. On a therophyte cover level the only significant difference (p < 0.05) was found between association 4 and association 5.

Table 8.5 Summary output of the analysis of variance (GLM) with association and life form (cover level) as main factors

Source of variation Degrees of Sum of Mean square F value P value freedom squares Association 7 1.7944 0.2563 3.69 0.0009 Life form (cover 6 61.7956 10.2992 148.26 < 0.0001 level) Association x Life 42 11.8453 0.2820 4.06 < 0.0001 form (cover level)

The interaction between the life forms (cover level) and vegetation groups also produced a significant difference (Table 8.6) and thus this interaction is used to further interpret the cover level data. Chamaephyte cover was found to be significantly lower (p < 0.05) in the Tanqua Karoo than in the Mountain Renosterveld and Winter Rainfall Karoo. Phanerophyte cover was

121

found to be significantly higher (p < 0.0001) in the Mountain Renosterveld than the Winter Rainfall Karoo as well as significantly higher (p < 0.05) in these two groups than the Tanqua Karoo. No significant differences were found for the hemicryptophyte, cryptophyte, therophyte, liana or parasite life forms at the cover level.

Table 8.6 Summary output of the analysis of variance (GLM) with vegetation group and life form (cover level) as main factors

Source of variation Degrees of Sum of Mean square F value P value freedom squares Vegetation group 2 1.0385 0.5192 6.08 0.0026 Life form (cover 6 68.7823 11.4637 134.16 < 0.0001 level) Vegetation group x 12 6.0313 0.5026 5.88 < 0.0001 Life form (cover level)

Because of the controversy as to whether the vegetation in the study area falls within the Succulent Karoo or the Fynbos Biomes, a degree of succulence was calculated from the data at a species as well as vegetation cover level. It was found that succulence usually occurred among the chamaephyte, hemicryptophyte and therophyte species for associations 1 to 8, however, for association 4, one phanerophyte species, Tylecodon paniculata, fell within the succulent category in two of the surveyed plots.

At a species level the succulent species’ percentage contribution to the Mountain Renosterveld was very low and ranged from 2.9% to 4.1% among the associations, with the percentage contribution to the Winter Rainfall Karoo higher at 13.5 to 16.9%, while the percentage contribution to the Tanqua Karoo was highest at 30.3 and 31.1% (Figure 8.2). Statistically (Kruskal-Wallis test), Winter Rainfall Karoo and Tanqua Karoo succulent species contributions were significantly higher than Mountain Renosterveld species contributions (p < 0.05).

Succulent species’ cover ranged from 1.4% to 58.2% throughout the study area (Figure 8.3). Values for the Mountain Renosterveld ranged from 1.4% (association 2) to 3.5% (association 3) and 11.6% (association 1) even although the contributions of succulents to the species level analysis in these three associations were similar (Figure 8.2). The contribution of succulents to the vegetation cover in the Winter Rainfall Karoo vegetation group also varied greatly from 28.8% in association 4, 35.4% in association 5 and was lowest for association 6 with only 9.1%. The ‘true’ succulent vygieveld of the Tanqua Karoo (association 7) had a high succulent cover of 58.2%, while the grassy plains (association 8) had a lower succulent cover contribution of 32.6%. Although the percentage cover contribution between associations 7 and 8 differed greatly, the percentage contribution by species was very similar (Figure 8.2). An ANOVA (Kruskal-Wallis test) comparing the succulent life form between the eight plant associations found a significant difference (p < 0.05) between the associations, as well as a

122

significant difference between the three broad vegetation groups where the Mountain Renosterveld vegetation group was found to be significantly different from the other two vegetation groups (p < 0.001).

) 35

Contribution by succulents (% species (% succulents by Contribution 0 12345678 Associa tions

Figure 8.2 Number of succulent species expressed as a percentage of the total number of species in eight plant associations in the Hantam-Tanqua-Roggeveld subregion.

The classification of plants and vegetation into major types on the basis of plant form is based on the observation that the capacity to survive different geographic, climatic and ecological conditions is often linked to plant architecture and physiognomy (Vandvik & Birks 2002). Succulence is a determining factor for defining the Succulent Karoo Biome with regression models indicating that rainfall evenness is an important factor explaining succulent richness per site (Cowling et al. 1994). Thus, centres of succulent diversity occur outside the strongly winter rainfall zones (Cowling et al. 1994). A number of environmental variables have been found to be correlated to succulence. According to Werger (1986) succulents are common in the winter rainfall area, where night frosts below -4°C are rare. The incidence of succulence is also correlated with soil salinity (Barkman 1979) and possibly with levels of soil phosphorus, potassium, calcium and magnesium (Hoffman & Cowling 1987).

123

Contribution by succulents (% cover) (% succulents by Contribution 0 12345678 Associa tions

Figure 8.3 Percentage succulent vegetation cover expressed as a percentage of the total vegetation cover in eight plant associations in the Hantam-Tanqua-Roggeveld subregion.

The data collected in this study highlights the importance of the succulent life form relative to the other life forms at a species level as well as at a vegetation cover level. At a species level there was a clear difference between values for the Mountain Renosterveld (lowest), Winter Rainfall Karoo (intermediate) and Tanqua Karoo (highest). However, at a vegetation cover level, this was not as clear, with Mountain Renosterveld values low and Tanqua Karoo values high but Winter Rainfall Karoo values straddling these two extremes. The high succulent and species cover levels for the Tanqua Karoo confirm the statement by Cowling et al. (1994) that centres of succulent diversity occur outside the strongly winter rainfall zones.

The low presence of the succulent life form in the Mountain Renosterveld vegetation group at a species and vegetation cover level and high cover contribution of phanerophytes re-enforces the delineation by Mucina and Rutherford (2006) and Van der Merwe et al. (2008a) who place this vegetation type within the Fynbos Biome and not the Succulent Karoo Biome. However, the succulent life form percentage vegetation cover contribution of plant association 1, with a value of 12.5%, is higher than what would be expected for Mountain Renosterveld vegetation indicating that the escarpment types of renosterveld show strong karroid affiliations (Mucina & Rutherford 2006). This association (Rosenia oppositifolia Mountain Renosterveld) was also the association that showed some floristic links between the Fynbos Biome related vegetation and Succulent Karoo Biome related vegetation in the Hantam-Tanqua-Roggeveld area (Van der Merwe et al. 2008a, 2008b). The presence of such transitional units re-enforces various authors’ contentions that there is a relationship between the Karoo and Cape Flora (Bayer 1984, Gibbs Russell 1987, Jürgens 1997) and supports Jürgens (1997) who proposed the recognition of the Floristic Kingdom of the Greater Cape Flora including at least two separate

124

regions, the Cape Floristic Region and the Succulent Karoo Region. This concept was recently supported by Born et al. (2007), who investigated an area named the Greater Cape Floristic Region.

In this study, the percentage contribution of succulent species to the Winter Rainfall Karoo was intermediate to the values found for the Mountain Renosterveld and Tanqua Karoo. However, succulent vegetation cover for associations 4 and 5 was significantly higher than for association 6. This could be as a result of association 6 (Roggeveld Karoo) having a strong transitional nature as it is located between the Mountain Renosterveld vegetation of the Roggeveld Mountains and the summer rainfall Nama Karoo Biome, with boundaries between the Succulent and Nama Karoo fluid and blurred (Milton et al. 1997). The low cover of succulents in association 6 (Roggeveld Karoo) supports the view of Rutherford and Westfall (1994) who incorporated the area in the Nama Karoo Biome. A high percentage contribution of succulent species was encountered in the Tanqua Karoo, yet the succulent contribution to the vegetation cover was much higher in association 7 than in association 8. The larger contribution by succulent species to the life form spectra at a species and vegetation cover level in the Winter Rainfall Karoo and Tanqua Karoo than in the Mountain Renosterveld confirms that the former two vegetation groups belong within the Succulent Karoo Biome.

Overall there is a clear difference between the life form spectra of the different broad vegetation groups with the Mountain Renosterveld of the Fynbos Biome differing most from the Tanqua Karoo, a strong Succulent Karoo part of the Succulent Karoo Biome. However, such spectra differ depending on whether only the presence of species is taken into account or also their cover or other quantitative vegetational parameters.

A comparison of the life form spectra found in the Hantam-Tanqua-Roggeveld subregion with another site in the Succulent Karoo (Goegap Nature Reserve), the winter rainfall Mojave Desert and a site in the Fynbos Biome (Swartboskloof) is presented in Table 8.7. The patterns found between associations 1 to 8 and Goegap Nature Reserve are more similar than the pattern found in the Mojave Desert. In general, the hemicryptophyte percentage was higher and the cryptophyte percentage was lower at the Goegap Nature Reserve than in the study area. None of the Mountain Renosterveld life form spectra (associations 1 to 3) compared well with the Fynbos Biome spectrum at Swartboskloof, where the phanerophyte contribution was much higher and therophyte contribution much lower than in the study area. This is supported by Oliver et al. (1983, in Rutherford & Westfall 1994) who state that the vegetation of the Roggeveld is only marginally similar to the vegetation structure of the Fynbos Biome but that it shows some floristic affinities to the Fynbos Biome.

125

Table 8.7 A comparison of life form spectra between the eight plant associations (Assoc. 1-8) in the study area, Goegap Nature Reserve (Succulent Karoo), Mojave Desert (Desert) and Swartboskloof (Fynbos Biome)

Life forms Vegetation Phanerophyte Chamaephyte Hemicryptophyte Cryptophyte Therophyte Other

Assoc. 1 3.0 24.7 10.7 33.2 27.0 4 Assoc. 2 5.4 26.3 8.2 30.6 26.8 2.8 Assoc. 3 3.5 34.6 7.7 27.0 26.1 1.1 Assoc. 4 6.1 34.6 8.1 25.4 22.2 3.5 Assoc. 5 1.8 29.7 7.1 27.3 32.0 2.2 Assoc. 6 1.1 30.4 10.6 24.9 31.4 1.6 Assoc. 7 3.7 47.6 13.9 18.2 15.6 1.0 Assoc. 8 3.7 39.0 14.0 32.4 4.8 6.0 Goegap Nature 0.0 6.0 32.0 17.0 17.0 28.0 Reserve Mojave desert 11.3 6.0 50.0 0.7 32.0 0.0 0.0 Swartboskloof 34.0 31.0 16.0 15.0 4.0 Data extracted from Beatley (1976) (in Esler et al. 1999b) and Van Rooyen et al. (1990).

8.5 Conclusions

A comparison of various life forms in the eight plant associations identified in the Hantam- Tanqua-Roggeveld subregion (Van der Merwe et al. 2008a, 2008b) produced different results when considered at a species level or at a vegetation cover level. Additionally, patterns between the three broad vegetation groups of the Mountain Renosterveld, Winter Rainfall Karoo and Tanqua Karoo differed at the different levels.

Six of the eight associations (Mountain Renosterveld and Winter Rainfall Karoo associations) were dominated by chamaephyte, cryptophyte (geophyte) and therophyte species with one association (Tanqua Karoo) dominated by only chamaephyte species and another association (Tanqua Karoo) by chamaephyte and cryptophyte species. Phanerophyte species contributions were low and a significantly higher contribution was found in the Mountain Renosterveld than both the Winter Rainfall Karoo and Tanqua Karoo. Chamaephyte species contributions were generally high and no significant differences were found between the three vegetation groups. Hemicryptophyte contributions were significantly smaller in the Winter Rainfall Karoo than the Tanqua Karoo. Cryptophyte species contributions were relatively constant across the three vegetation groups but were significantly higher for the Mountain Renosterveld than the Tanqua Karoo. Therophyte contributions were similar in the Mountain Renosterveld and Winter Rainfall Karoo and significantly less in the Tanqua Karoo. No significant differences were found at the vegetation group level for lianas or parasites.

126

On a vegetation cover level, phanerophyte contributions were found to be significantly higher in the Mountain Renosterveld than the Winter Rainfall Karoo as well as significantly higher between these two groups and the Tanqua Karoo. Chamaephyte cover varied greatly but was significantly higher in the Mountain Renosterveld and Winter Rainfall Karoo groups than in the Tanqua Karoo vegetation group. No significant differences were found for hemicryptophytes, cryptophytes, therophytes, lianas and parasites at the cover level.

The Succulent Karoo Biome is largely defined by the presence of succulents and thus this trait was extracted from the life form categories. Most of the succulent species were found to originate from the chamaephyte, hemicryptophyte and therophyte species and one species, was a phanerophyte. Succulent cover was generally lowest for the Mountain Renosterveld, with higher cover values found in the Winter Rainfall Karoo and the Tanqua Karoo. This confirms the Succulent Karoo Biome affinities of the Winter Rainfall Karoo and Tanqua Karoo. Yet, a higher than expected vegetation cover of the succulent life form was found for association 1, which is a strong Mountain Renosterveld (Fynbos Biome) vegetation group, indicating its transitional nature between the two biomes.

A comparison of life form spectra with another site in the Succulent Karoo produced similar patterns, however, these patterns are very different from the pattern encountered in the winter rainfall Mojave Desert. The Mountain Renosterveld life form spectra determined in this study did not compare well with a Fynbos Biome site where phanerophyte contribution was much higher and therophyte contribution much lower.

8.6 Acknowledgements

The authors would like to thank the Critical Ecosystem Partnership Fund (CEPF) through the Succulent Karoo Ecosystem Plan/Program (SKEP) initiative for funding the project. The Critical Ecosystem Partnership Fund is a joint initiative of Conservation International, the Global Environmental Facility, the Government of Japan, the MacArthur Foundation and the World Bank. A fundamental goal is to ensure civil society is engaged in biodiversity conservation. The various people who assisted with the field work are gratefully acknowledged. Dr M. van der Linde and Dr L. Debusso of the Statistics Department at the University of Pretoria are thanked for their assistance with the statistical analysis. This research was supported by the National Research Foundation under grant number 61277.

8.7 References

ACOCKS, J.P.H. 1953. Veld types of South Africa. Memoirs of the Botanical Survey of South Africa 28, 1-192.

127

ACOCKS, J.P.H. 1988. Veld types of South Africa. 3rd edn. Memoirs of the Botanical Survey of South Africa 57, 1-146. BAYER, M.B. 1984. The Cape and the Karoo – a winter rainfall biome versus a fynbos biome. Veld & Flora 70, 17-19. BARBOUR, M.G., BURK, J.H. PITTS, W.D., GILLIAM, F.S. AND SCHWARTZ, M.W. (Eds) 1999. Terrestrial plant ecology (3rd edition). Benjamin/Cummings Publishing Company Inc., California. BARKMAN, J.J. 1979. The investigation of vegetation texture and structure. In: M.J.A. Werger, (Ed.) The study of vegetation, pp. 125-160. Junk, London. BEATLEY, J.C. 1976. Vascular plants of the Nevada Test Site and central and southern Nevada: Ecological and geographic distributions. Energy Research and Development Administration, Washington DC. BORN, J., LINDER, H.P. AND DESMET, P. 2007. The Greater Cape Floristic Region. Journal of Biogeography 34, 147-162. BOUCHER, C. AND MOLL, E.J. 1981. South African Mediterranean shrublands. In: F. Di Castri, D.W. Goodall and R.L. Specht (Eds). Ecosystems of the world, pp. 233-248. Mediterranean type shrublands. Elsevier, Amsterdam. BROENNIMANN, O., THUILLER, W., HUGHES, G., MIDGLEY, G.F., ALKEMADE, J.M.R. AND GUISAN, A. 2006. Do geographic distribution, niche property and life form explain plants’ vulnerability to global change? Global Change Biology 12, 1079-1093. CAIN, S.A. 1950. Life-forms and phytoclimate. The Botanical Review 16, 1-32. CONSERVATION INTERNATIONAL WEBSITE: http://www.biodiversityhotspots.org (Accessed 2006) COUNCIL FOR GEOSCIENCE 2008. Geological data 1: 250 000. Electronic data provided by the Council for Geoscience, Silverton, Pretoria. COWLING, R.M., ESLER, K.J., MIDGLEY, G.F. AND HONIG, M.A. 1994. Plant functional diversity, species diversity and climate in arid and semi-arid southern Africa. Journal of Arid Environments 27, 141-158. COWLING, R.M., GIBBS RUSSEL, G.E., HOFFMAN, M.T. AND HILTON-TAYLOR, C. 1989. Patterns of plant species diversity in southern Africa. In: B.J. Huntley (Ed.). Biotic diversity in southern Africa. Concepts and Conservation, pp. 19-50. Oxford University Press, Cape Town. COWLING, R.M. AND HILTON-TAYLOR, C. 1994. Patterns of plant diversity and endemism in southern Africa: an overview. In: B.J. Huntley (Ed.). Botanical diversity in southern Africa. Strelitzia 1, pp. 31-52. National Botanical Institute, Pretoria. COWLING, R.M., HOLMES, P.M. AND REBELO, A.G. 1992. Plant diversity and endemism. In: R.M. Cowling (Ed.). The Ecology of Fynbos: Nutrients, fire and diversity, pp. 62-112. Oxford University Press, Cape Town. COWLING, R.M. AND PIERCE, S. 1999. Namaqualand a succulent desert. Fernwood Press, Vlaeberg.

128

COWLING, R.M., RICHARDSON, D.M. AND MUSTARD, P.J. 1997. Fynbos. In: R.M. Cowling, D.M. Richardson and S.M. Pierce (Eds). Vegetation of southern Africa, pp. 99-129. Cambridge University Press, Cambridge. CRITICAL ECOSYSTEM PARTNERSHIP FUND, 2003. Ecosystem Profile: The Succulent Karoo hotspot, Namibia and South Africa. Critical Ecosystem Partnership Fund report. DANIN, A. AND ORSHAN, G. 1990. The distribution of Raunkiaer life forms in Israel in relation to the environment. Journal of Vegetation Science 1, 41-48. ESLER, K.J., RUNDEL, P.W. AND COWLING, R.M. 1999a. The Succulent Karoo in a global context: plant structural and functional comparison with North America winter-rainfall deserts. In: W.R. Dean and S.J. Milton (Eds). The Karoo: Ecological patterns and processes, pp. 123-144. Cambridge University Press, Cambridge. ESLER, K.J., RUNDEL, P.W. AND VOSTER, P. 1999b. Biogeography of prostrate-leaved geophytes in semi-arid South Africa: hypotheses on functionality. Plant Ecology 142, 105-120. FRANCIS, M.L., FEY, M.V., PRINSLOO, H.P., ELLIS, F., MILLS, A.J. AND MEDINSKI, T.V. 2007. Soils of Namaqualand: Compensations for aridity. Journal of Arid Environments 70, 588-603. FOX, G.A. 1992. The evolution of life history traits in desert annuals. American Journal of Botany 77, 1508-1518. GIBBS RUSSELL, G.E. 1987. Preliminary floristic analysis of the major biomes in Southern Africa. Bothalia 17, 213-227. GOOD, R. 1947. The geography of flowering plants. Longmans, Green & Co., New York. HILTON-TAYLOR, C. 1994. Western Cape Domain (Succulent Karoo). In: S.D. Davis, V.H. Heywood and A.C. Hamilton (Eds). Centres of plant diversity. A guide and strategy for their conservation, pp. 201-203. IUCN Publications Unit, Cambridge. HOFFMAN, M.T. AND COWLING, R.M. 1987. Plant physiognomy, phenology and demography. In: R.M. Cowling and P.W. Roux (Eds). The Karoo Biome: a preliminary synthesis. Part 2 – Vegetation and history. South African National Scientific Programmes Report 142, CSIR, Pretoria. JÜRGENS, N. 1997. Floristic biodiversity and history of African arid regions. Biodiversity and Conservation 6, 495-514. LOW, A.B. AND REBELO A.G. 1996. Vegetation of South Africa, Lesotho and Swaziland. Department of Environmental Affairs and Tourism, Pretoria. MILTON, S.J., YEATON, R.I., DEAN, W.R.J. AND VLOK, J.H.J. 1997. Succulent Karoo. In: R.M. Cowling, D.M. Richardson and S.M. Pierce (Eds). Vegetation of southern Africa, pp. 99-129. Cambridge University Press, Cambridge. MUCINA, L., JÜRGENS, N., LE ROUX, A., RUTHERFORD, M.C., SCHMIEDEL, U., ESLER, K.J., POWRIE, L.W., DESMET, P.G. AND MILTON, S.J. 2006. Succulent Karoo Biome. In: L. Mucina and M.C. Rutherford (Eds). The vegetation of South Africa,

129

Lesotho and Swaziland. Strelitzia 19, pp. 220-299. South African National Biodiversity Institute, Pretoria. MUCINA, L. AND RUTHERFORD, M.C. (Eds) 2006. The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19. South African National Biodiversity Institute, Pretoria. MUCINA, L., RUTHERFORD, M.C. AND POWRIE, L.W. (Eds) 2005. Vegetation map of South Africa, Lesotho and Swaziland, 1 : 1 000 000 scale sheet maps. South African National Biodiversity Institute, Pretoria. MUELLER-DOMBOIS, D. AND ELLENBERG, H. (Eds) 1974. Aims and methods of vegetation ecology. Wiley, New York. MYERS, N., MITTERMEIR, R.A., MITTERMEIR, C.G., DE FONSECA, G.A.B. AND KENT, J. 2000. Biodiversity hotspots for conservation priorities. Nature 403, 853-858. PAVÓN, N.P., HERNÁNDEZ-TREJO, H. AND RICO-GRAY, V. 2000. Distribution of plant life forms along an altitudinal gradient in the semi-arid valley of Zapotitlán, Mexico. Journal of Vegetation Science 11, 39-42. PROCHEŞ, Ş., COWLING, R.M. AND DU PREEZ, D.R. 2005. Patterns of geophyte diversity and storage organ size in the winter-rainfall region of southern Africa. Diversity and Distributions 11, 101-109. PROCHEŞ, Ş., COWLING, R.M., GOLDBLATT, P., MANNING, J.C. AND SNIJMAN, D.A. 2006. An overview of the Cape geophytes. Biological Journal of the Linnean Society 87, 27-43. RAUNKIAER, C. 1934. The life forms of plants and statistical plant geography. Oxford University Press, Oxford. REBELO, A.G., BOUCHER, C., HELME, N., MUCINA, L. AND RUTHERFORD, M.C. 2006. Fynbos Biome. In: L. Mucina and M.C. Rutherford (Eds). The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19, pp. 52-219. South African National Biodiversity Institute, Pretoria. RUTHERFORD, M.C. AND WESTFALL, R.H. 1994. Biomes of Southern Africa. An objective characterisation. Memoirs of the Botanical Survey of South Africa 63, 1-94. RUTHERFORD, M.C., MUCINA, L. AND POWRIE, L.W. 2006. Biomes and bioregions of southern Africa. In: L. Mucina and M.C. Rutherford (Eds). The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19, pp. 30-51. South African National Biodiversity Institute, Pretoria. SHMIDA, A. 1984. Whittaker’s plant diversity sampling method. Israel Journal of Botany 33, 41-46. VAN DER MERWE, H., VAN ROOYEN, M.W. AND VAN ROOYEN, N. 2008a. Vegetation of the Hantam-Tanqua-Roggeveld subregion, South Africa. Part 1. Fynbos Biome related vegetation. Koedoe 50, 61-71. VAN DER MERWE, H., VAN ROOYEN, M.W. AND VAN ROOYEN, N. 2008b. Vegetation of the Hantam-Tanqua-Roggeveld subregion, South Africa. Part 2. Succulent Karoo Biome related vegetation. Koedoe 50, 160-183.

130

VANDVIK, V. AND BIRKS, H.J.B. 2002. Pattern and process in Norwegian upland grasslands: a functional analysis. Journal of Vegetation Science 13, 123-134. VAN ROOYEN, M.W. 1999. Functional aspects of short-lived plants. In: W.R.J. Dean and S.J. Milton (Eds), The Karoo, ecological patterns and processes, pp. 107-122. Cambridge University Press, Cambridge. VAN ROOYEN, M.W. 2002. Management of the old field vegetation in the Namaqua National Park, South Africa: conflicting demands of conservation and tourism. Geographical Journal 168: 211-223. VAN ROOYEN, M.W., THERON, G.K. AND GROBBELAAR, N. 1990. Life form and dispersal spectra of the flora of Namaqualand, South Africa. Journal of Arid Environments 19, 133-145. VAN WYK, A.E. AND SMITH, G.F. (Eds) 2001. Regions of Floristic Endemism in Southern Africa: A review with emphasis on succulents, pp. 1-199. Umdaus Press, Pretoria. WEATHER BUREAU 1998. Climate of South Africa. Climate statistics up to 1990. WB 42. Government Printer, Pretoria. WERGER, M.J.A. 1986. The Karoo and southern Kalahari. In: M. Evenari, I. Noy-Meir and D.W. Goodall (Eds). Hot deserts and arid shrublands, pp. 283-359. Elsevier, Amsterdam. WESTOBY, M. 1980. Elements of a theory of vegetation dynamics in arid rangelands. Israel Journal of Botany 28, 169-194. WITHROW, A.P. 1932. Life forms and leaf size classes of certain plant communities of the Cincinnati region. Ecology 13, 12-35.

131

Chapter 9

Life form and species diversity on abandoned croplands in the Roggeveld, South Africa

Abstract

The Roggeveld mountain range forms part of the Fynbos Biome and consists of an island of Mountain Renosterveld vegetation surrounded by Succulent Karoo Biome vegetation. In order to improve our understanding of the vegetation recovery on abandoned croplands in the Mountain Renosterveld vegetation of the Roggeveld, a variety of species and life form diversity parameters were studied on abandoned croplands of different ages and compared with each other and to the natural vegetation.

Therophytes and chamaephytes were found to be the most abundant life forms on abandoned croplands, while hemicryptophytes, phanerophytes, lianas and parasites were the least abundant. Chamaephytes made an overwhelming contribution to the relative cover on abandoned croplands with phanerophytes, cryptophytes and therophytes contributing significantly less, while hemicryptophyte and liana contributions were negligible.

Species-area curves using the exponential function, differed significantly between the abandoned croplands of all ages and the natural vegetation. Species richness increased with the time since abandonment but no similar increase in species evenness, Shannon or Simpson indices were found. A regression using species richness values on age since abandonment predicted that an abandoned cropland of approximately 33-years should be as species rich as the natural vegetation, however a Principal Co-ordinate Analysis of floristic data indicated that all the abandoned croplands were floristically still extremely different from the natural vegetation. Across all nine survey plots only 15 species contributed to a high cover on the plots.

Vegetation recovery of abandoned croplands in the Mountain Renosterveld of the Roggeveld occurs naturally, yet the rate of recovery varies among the life forms and an important component of the flora, the geophytes (cryptophytes), still remains greatly underrepresented after 20 years of abandonment.

Keywords: Diversity indices, Fynbos Biome, old field, renosterveld, species richness, Succulent Karoo, Whittaker plots

132

9.1 Introduction

The Cape Floristic Region (CFR) of South Africa has one of the highest species densities and levels of endemism, at both local and regional scales, for any temperate or tropical continental region (Cowling et al. 1989, 1992) and is recognised as a global hotspot of biodiversity (Cowling & Hilton-Taylor 1994). The Fynbos Biome, as delineated by Rutherford and Westfall (1994), constitutes the major part of the CFR. This biome comprises three quite different, naturally fragmented vegetation types, namely fynbos, renosterveld and strandveld, that occur in winter- and summer-rainfall areas and are dominated by small-leaved, evergreen shrubs whose regeneration is intimately linked to fire (Rebelo et al. 2006). Renosterveld is one of the most threatened vegetation types in South Africa as a result of transformation by agriculture, alien invasive plants and urbanisation (Rebelo 2001, Rouget et al. 2006). The degree of transformation is strongly linked to topography and geographical location (Rebelo et al. 2006) with the lowlands usually showing a higher degree of transformation than the uplands mainly because the lowlands are more favourable for agriculture and urbanisation. Certain high-lying areas such as the Outeniqua and Tsitsikamma Mountains are, however vulnerable to transformation due to afforestation (Rebelo et al. 2006).

The vegetation of the Roggeveld mountain range was classified as renosterveld vegetation by Low and Rebelo (1996), Mucina et al. (2005) and Van der Merwe et al. (2008). This Mountain Renosterveld on the Roggeveld mountain range therefore constitutes an island of Fynbos Biome vegetation surrounded by the Succulent Karoo Biome. Although it is classified as part of the renosterveld vegetation (Fynbos Biome), it lies on the arid extreme of this vegetation type and differs from other renosterveld vegetation types further south in exhibiting many karroid properties.

Among the earliest references to the botanical wealth of the Mountain Renosterveld date from the early 1900s when Diels (1909) mentioned the presence of the Cape element in the flora and the high levels of endemism on the Hantam Mountain. Weimarck (1941) who proposed a classification of the Cape species into five phytogeographical groups treated the Hantam- Roggeveld as a subcentre of his North-Western Centre and stated that the subcentre constituted the last outlier of the Cape element in the inner parts of western South Africa. The Roggeveld was also one of the three centres of endemism that Hilton-Taylor (1994) identified within the Western Cape Domain, however, he considered the Western Cape Domain as part of the Succulent Karoo Biome.

The dominant shrub, Dicerothamnus rhinocerotis, renosterbos, after which the renosterveld is named, is an indigenous species with encroaching properties. Disturbed areas are quickly colonised and dominated by D. rhinocerotis, which is considered disadvantageous from an agricultural point of view since it is not grazed (Shearing 1997). Historical records of early

133

explorers indicate that renosterveld had abundant grasses and that as a result of high grazing pressure renosterbos was increasing in abundance and grasses becoming scarcer (transcriptions and translation of R.J. Gordon’s travels, Cullinan 2003). The dominance of D. rhinocerotis was also mentioned by Marloth in 1908 who suggested that the frequent burning of vegetation by the early settlers to increase palatable vegetation available for their livestock actually increased the dominance of D. rhinocerotis. Marloth (1908) also mentions the monocultures of renosterbos on abandoned croplands and that within the first winter of land abandonment D. rhinocerotis seedlings colonise abandoned croplands and that the plants grow prolifically and become denser.

The first European farmers settled along the northern slopes of the Roggeveld mountains in the 1740s (Van der Merwe 1938). These settlements were restricted to the valleys and mountainous regions where permanent water could be found (Van der Merwe 1938, 1988). These first farmers cultivated crops on a small scale to be self-sustainable. During his travels to the region in 1778 Gordon (Cullinan 2003) noted that he saw planted trees, vegetables and garden fruit.

The higher rainfall of the Roggeveld in comparison to the surrounding Succulent Karoo areas has allowed farmers to plough large tracts of land to cultivate crops. Numerous utilised and abandoned croplands lie scattered throughout the Roggeveld landscape. Production costs have increased substantially over the years and farmers have been forced to cultivate fewer fields on their lands. This worldwide trend of the increased level of land abandonment is primarily as a result of environmental and socio-economic changes (Cramer et al. 2007).

In semi-arid to arid regions studies on secondary succession are rather scarce, possibly because succession proceeds very slowly under these harsh conditions (Otto et al. 2006). However, plant community succession is one of the most important aspects of vegetation ecology (Zhang 2005) since successional plant communities provide a model system for testing a variety of ecological hypotheses regarding the controls on biodiversity that could be applied to the management and restoration of plant communities (Huberty et al. 1998). Additionally, with the current predictions of climate change the study of plant succession and vegetation recovery take on an even stronger urgency (Bazzaz 2000). Climate change will add another layer of complexity to the restoration of old fields and could exacerbate the ecological thresholds to plant community assembly (Cramer et al. 2007).

The aim of the current study was to use a space-for-time approach to follow the recovery of the vegetation on abandoned croplands in the Roggeveld and to evaluate the rate of recovery in terms of the species composition and various parameters of species and life form diversity. Biodiversity parameters such as species richness, evenness and the Shannon and Simpson

134

indices were used to gain a better understanding of the process of recovery and how the abandoned croplands compare with the original natural vegetation.

9.2 Study area

The study was carried out on the farm Soekop (32° 02’ 10.1” S and 20° 07’ 06.9” E) in the Mountain Renosterveld vegetation of the Roggeveld (Figure 9.1). The Roggeveld mountain range forms the steep escarpment which separates the low-lying Tanqua Karoo basin from the interior plateau of South Africa. Rocks of the Ecca group cover most of the Roggeveld Mountains (Rubidge & Hancox 1999) with the study area belonging to the Waterford Formation which consists of sandstone, rhythmite, shale and mudstone with wave marks and slumping being common features (Council for Geoscience 2008). Shallow stony lithosol soils are characteristic of the Roggeveld (Francis et al. 2007).

∗

Figure 9.1 The location of the study site (∗) within the Roggeveld mountain range.

Rainfall in the study area ranges from 132 mm to 467 mm per year (Weather Bureau 1998) and although it falls mainly in winter it does include a few summer thunderstorms. In the winter, snowfalls occur with a mean of six snow days recorded per year over a 24-year period by the Weather Bureau (1998). At Sutherland, the mean daily minimum temperature for the

135

coldest month, July, is –2.4ºC, while the extreme minimum, -13.6ºC, was recorded in July 1970 and August 1978 (Weather Bureau 1998). The mean daily maximum for the warmest month, January, is 27.1ºC, while the extreme maximum recorded was 35.5ºC in January 1980 (Weather Bureau 1998).

The study area falls within the Roggeveld Shale Renosterveld vegetation type of Mucina et al. (2005, Rebelo et al. 2006). A recent finer scale classification and mapping of the vegetation associations of the area indicates that the study area falls within the Dicerothamnus rhinocerotis Mountain Renosterveld vegetation association (Van der Merwe et al. 2008). This vegetation association is dominated by a high cover of D. rhinocerotis while Merxmuellera stricta and Dimorphotheca cuneata also characterise the vegetation unit. Strong annual and geophyte components in spring are usually present following good winter rains.

9.3 Materials and Methods

Nine sample plots using Whittaker’s plant diversity plot technique (Shmida 1984) were surveyed. Eight of these were surveyed on abandoned croplands of various ages (3-, 4-, 8-, 10-, 15- and 20-years old). Additionally, a plot in the undisturbed natural vegetation close to the abandoned cropland of 20-years old was surveyed. All surveys were conducted on one farm in the same vegetation type and on the same geological substrate. Furthermore, the surveys were all conducted in one season. The only modification to the methodology as described by Shmida (1984) was the modification of the field form and notations used on the field form (see Chapter 6 for a full description of the methodology).

Species encountered in the surveys were separated into broad life form categories following Raunkiaer (1934) as modified in Mueller-Dombois and Ellenberg (1974 see Appendix 1). The relative contributions of each life form, in terms of species as well as plant cover, to the 1000 m² sample plots were calculated.

The total species number for seven plot sizes (1 m², 5 m², 10 m², 20 m², 50 m², 100 m² and 1000 m²) were determined. These seven plot sizes were used to construct Type II species- area curves (Scheiner 2003, 2004) for each of the nine plots sampled using the exponential function since this function produced the best results in a study across the entire Hantam- Tanqua-Roggeveld subregion (Chapter 6). The exponential function is expressed as a semilog function: S = z x log A + c (Veech 2000), Where: S = species richness A = area of survey plot z and c are constants.

136

H’ = Shannon diversity

H’ = - ∑ pi log pi i

Where pi = importance probability in column i.

E = Evenness (equitability) = H’ / ln (richness).

S 2 D = 1 - ∑ pi i The Shannon index was also used to calculate a life form diversity index using frequencies of life forms instead of species.

The Chi-square test of the STATISTICA computer package (StaSoft, Inc. Version 8, 2300 East 14th Street, Tulsa, OK 74104) was used to compare the life-form distributions. Floristic data for all nine plots surveyed were ordinated using Principal Co-ordinate Analysis (PCoA) in the SYN-TAX computer program (Podani 2001) because the use of a wide array of distance measures in Principal Co-ordinate Analysis (PCoA) can give a marked improvement over Principal Component Analysis (PCA) (McCune & Grace 2002). The statistical significance of the differences between slope values and intercepts of the exponential function curves were analysed by an Analysis of Covariance (Quinn & Keough 2002) linear regression with GraphPad Prism 4.03 for Windows (GraphPad software, San Diego, California, USA, www.graphpad.com.).

9.4 Results and discussion

Whittaker’s plant diversity plot technique has proved to be an efficient method of sampling used around the world, especially in semi-arid environments (Shmida 1984, Chapter 6, Chapter 7, Chapter 8). The data derived by this technique were used to determine various life form and species diversity parameters for abandoned croplands of various ages in the

137

Roggeveld. These data were compared to baseline data of the various plant diversity parameters compiled for the Roggeveld area derived with the same technique (Chapter 6, Chapter 7, Chapter 8)

The tradition of classifying plants and vegetation into major types on the basis of plant form has a long history (Vandvik & Birks 2002). The life form of a species refers to the vegetative form of the plant body and it is assumed to be a result of morphological adjustment to the climate and environment (Cain 1950, Barkman 1979, Van Rooyen et al. 1990, Semenova & van der Maarel 2000). Cain (1950) stated that the spectra for successional communities of various sorts might reflect edaphic conditions and give a good measure of the changing environment as succession proceeds. When considering the abundance of each life form at 1000 m2 (Figure 9.2) and the contribution per life form expressed as a percentage of the total number of species (Figure 9.3), trends along the successional sequence are observed.

Generally, therophyte and chamaephyte species were the most abundant species throughout the various stages of secondary succession in the Roggeveld (Figure 9.2b, 9.2e, 9.3) while, liana species were scarce and only recorded on fields abandoned for 10 years or more (Figure 9.2f, 9.3). No parasite species were encountered in any of the plots surveyed. Therophyte species increased in number with age of the abandoned croplands up to croplands last ploughed 8-years ago and with a noticeable decrease in number found on the abandoned croplands of 10-years and older (Figure 9.2e, 9.3). The natural vegetation held less therophyte species than the 4- and 8-year old abandoned croplands (Figure 9.2e), with the relative contribution of therophyte species to the life form distribution demonstrating the same trend except for the 3-year old abandoned croplands where they constituted almost half of the species (Figure 9.3). Chamaephyte species generally increased in number as the age of the abandoned croplands increased, with the natural vegetation holding the highest number of chamaephyte species (Figure 9.2b). However, the relative contribution of chamaephyte species showed an initial decrease whereafter it increased to the 15-year old abandoned cropland and then decreased in the 20-year old abandoned cropland and in the natural vegetation (Figure 9.3).

138

a) b)

25 25

20 20

15 15

10 10

5 5 Number of chamaephyte species Number of phanerophyte species 0 0 3 4 8 10 15 20 Natural 3 4 8 10 15 20 Natural Age of abandoned cropland (years) Age of abandoned cropland (years)

c) d)

s 25

20 20

15 15

10 10

5 5 Number of cryptophyte species of cryptophyte Number 0 0 Number of hemicryptophyte specie 3 4 8 10 15 20 Natural 3 4 8 10 15 20 Natural Age of abandoned cropland (years) Age of abandoned cropland (years)

e) f)

25 25

20 20

15 15

10 10

5 Number of liana species Numberof therophyte species 0 0 3 4 8 10 15 20 Natural 3 4 8 10 15 20 Natural Age of abandoned cropland (years) Age of abandoned cropland (years)

Figure 9.2 Total number of species per life form in 1000 m² plots on abandoned croplands of various ages and the natural vegetation in the Roggeveld for the following life forms: a) phanerophytes, b) chamaephytes, c) hemicryptophytes, d) cryptophytes, e) therophytes and f) lianas.

Cryptophyte (geophyte) species generally increased gradually along the time scale since abandonment (Figure 9.2d) however, the relative contribution of cryptophyte species to the 20-year old abandoned cropland was still less than half the relative contribution to the natural vegetation (Figure 9.3). Different life forms have different sensitivities to soil disturbance (McIntyre et al. 1995) and cryptophyte species are severely depleted by continuous ploughing to produce crops. These effects of cultivation still prevail in the species composition on the abandoned croplands (Stromberg & Griffin 1996) and was particularly evident in the current study when considering the rich geophytic component of the Roggeveld (Van Wyk & Smith 2001).

139

Figure 9.3 Number of species per life form per 1000 m2 expressed as a percentage of the total number of species across the abandoned croplands of various ages and the natural vegetation in the Roggeveld.

The number and relative abundance of hemicryptophyte species increased with age since land abandonment (Figure 9.2c, 9.3) yet, the number of species remained small (≤ 5 species per 1000 m²). Phanerophyte numbers remained relatively constant throughout all the plots of varying ages (Figure 9.2a, 9.3) and therefore the relative contribution of this life form showed a slight decrease. This could be ascribed to the ability of Dicerothamnus rhinocerotis to quickly colonise disturbed areas throughout the Roggeveld region. Liana species were only present in abandoned croplands of 10-years and older (Figure 9.2f, 9.3).

Therophyte species were initially replaced predominantly by chamaephyte species and in later successional ages by a combination of chamaephyte and cryptophyte species. This replacement of therophyte species with mainly chamaephyte species differs from the ‘classical’ sequence of life forms on old fields where the annual species are replaced by biennials and herbaceous perennials (i.e. hemicryptophytes) before these are replaced by chamaephytes and eventually phanerophytes (Debussche et al. 1996, Huberty et al. 1998, Otto et al. 2006). However, which mechanisms of species replacement operate at a particular time and place in succession depends largely on the growth form of the dominant species of the site (e.g. annual vs. perennial life cycle, canopy-forming vs. understorey species, clonal vs. non-clonal). Species replacement also depends on the life history characteristics, dispersal mode and growth form of the potential occupants of the site and the nature of natural disturbances (Armesto & Pickett 1986). Considering that the Roggeveld vegetation is generally rich in chamaephyte and cryptophyte species and poor in phanerophyte and

140

hemicryptophyte species, the successional transition from therophyte species to chamaephyte species and later a combination of chamaephyte and cryptophyte species can be understood.

The Shannon life form diversity index ranged from 0.628 (OL2) to 1.115 (OL6) along the successional sequence with no evidence of increasing complexity with increasing age since abandonment. Yet, when considering Figure 9.3, it seems as though there is a better spread of species amongst the life forms as succession proceeds.

By weighting every species by its relative cover and expressing this value as a percentage of the total relative cover, a very different life form distribution is obtained (Figure 9.4). Phanerophyte contribution decreased from the 3- to 8-year old abandoned croplands, then increased substantially on the 10-year old abandoned cropland and decreased again to the 20-year old abandoned cropland, thereafter remaining at a slightly increased lower level (Figure 9.4). The dominant life form, contributing to the highest cover, is the chamaephyte life form. Chamaephytes comprise a third of the vegetation cover on 3-year old abandoned croplands and continually increased in cover to a value of 80.3% on 20-year old abandoned croplands, with the natural vegetation holding 70.9% chamaephyte cover (Figure 9.4). Hemicryptophytes increased after three years of abandonment thereafter remaining relatively constant throughout all the abandoned croplands at approximately half the relative cover found in the natural vegetation (Figure 9.4). The relative contribution of cryptophytes (geophytes) to vegetation cover increased from the third year of abandonment and varied within a narrow band throughout the abandoned croplands, with the cryptophytes contributing to a higher cover (8.2%) in the natural vegetation (Figure 9.4). Therophytes’ contribution to the total cover was highest during the first 8-years of cropland abandonment and decreased noticeably on the 10-year old abandoned cropland and remained at this lower level, which was similar to its contribution in the natural vegetation (Figure 9.4). The smallest contribution to total cover of any life form was that of the lianas which remained ≤ 0.6% of the total cover (Figure 9.4).

141

Figure 9.4. Relative cover per life form expressed as a percentage of the total cover across the abandoned croplands of various ages and the natural vegetation in the Roggeveld.

The Chi-squared test comparing the life form spectra on a species basis of the natural vegetation with the 3-, 4- and 8-year old abandoned croplands produced highly significant differences (p< 0.001). No significant differences were found between the natural vegetation and abandoned croplands of 10- and 20-years old however, a significant difference (p< 0.05) was found between the natural vegetation and the 15-year old abandoned cropland.

Species-area curves were constructed using the exponential function because this function generally performs well (Connor & McCoy 1979, Tjørve 2003) and was also found to produce the best fit in a study of the natural vegetation of the Hantam-Tanqua-Roggeveld (Chapter 6). Contrary to general theoretical expectations about species/area relationships, Bazzaz (2000) found that species number in successional habitats rarely rises smoothly and asymptotically with an increase in area because patchiness in the distribution of resources and species can generate abrupt changes in species/area relations. However, in this study no sudden increases in species number were evident and in all cases an asymptote seemed to be reached. A comparison of the slope and intercept values of the exponential curves between the abandoned croplands and the natural vegetation found that the curve for the natural vegetation was always significantly different from that of the abandoned croplands (Table 9.1). In most cases the slopes were different, but whenever the slope values were not different the intercept values were found to be significantly different.

Species richness, or the number of species, is currently the most widely used diversity measure (Stirling & Wilsey 2001) and is used extensively in secondary succession literature

142

(Squiers & Wistendahl 1976, Inouye et al. 1987, Lavorel 1994, Prieur-Richard et al. 2000, Tunnell et al. 2004, Gibson et al. 2005, Zhan et al. 2007) in comparisons between abandoned croplands of various ages and the natural vegetation to indicate the degree of recovery.

Table 9.1 Age of the abandoned cropland, survey plot number, exponential function, r-value and p-value (significance of fit) for the species-area curve for the plots surveyed on abandoned croplands of various ages and in the natural vegetation. The statistical significance of the comparisons of slope and intercept values between each abandoned cropland with the natural vegetation is also provided

Age of Survey Exponential (Semi-log) function Significance of comparison of abandoned plot slope and intercept values cropland number between abandoned croplands and the natural vegetation Linear equation r-value p-value Slope value Intercept value 3 years OL1 y=1.3686+8.2400x 0.9621 0.0005*** 0.0002*** 3 years OL2 y=0.4980+5.4866x 0.9310 0.0023** 0.0001*** 4 years OL3 y=0.0107+8.5086x 0.9596 0.0006*** 0.0004*** 4 years OL4 y=5.1820+9.6119x 0.9887 0.0000*** 0.0004*** 8 years OL5 y=2.1718+13.9909x 0.9888 0.0000*** 0.0269* 10 years OL6 y=-4.9583+13.2600x 0.8841 0.0082** 0.1386 0.0002*** 15 years OL7 y=1.1083+11.4901x 0.9651 0.0004*** 0.0064** 20 years OL8 y=-4.4219+13.6977x 0.8716 0.0106* 0.1980 0.0005*** Natural OL9 y=5.3234+18.9851x 0.9810 0.0001*** vegetation ns Not significant, * p < 0.05 Significant, ** p < 0.01 Highly significant, *** p < 0.001 Very highly significant

Generally, the species richness values acquired in this study for the abandoned croplands indicate the expected low species richness on abandoned croplands of three to four years in age (20 to 33 species per 1000 m², Table 9.2). There is a marked increase in species richness at the ages between eight and 20 years (40 to 48 species per 1000 m², Table 9.2). The undisturbed vegetation close to the 20-year old abandoned cropland had a species richness of 66 per 1 000 m2 (Table 9.2). The latter value compares well with values obtained for other natural vegetation in the Roggeveld in which species richness varied from 62 to 99 species per 1000 m² with a mean value of 79 species per 1 000 m² (Chapter 7). A regression of the age of the abandoned cropland against the number of species found per 1000 m² suggests that theoretically an abandoned cropland of approximately 33 years after abandonment should be as species rich as the natural vegetation (Figure 9.5). However, such an extrapolation of the regression beyond the actual data points must be viewed with caution.

143

Table 9.2 Age of abandoned cropland, survey plot number, species richness, species evenness, Shannon’s index of diversity and Simpson’s index for abandoned croplands of different ages and the natural vegetation in the Roggeveld

Age of Survey Species Species Shannon Simpson abandoned plot richness evenness index (H’) index (D) cropland number (E) 3 years OL1 28 0.613 2.044 0.744 3 years OL2 20 0.404 1.211 0.435 4 years OL3 29 0.734 2.471 0.856 4 years OL4 33 0.613 2.144 0.743 8 years OL5 45 0.797 3.014 0.884 10 years OL6 45 0.594 2.260 0.770 15 years OL7 40 0.579 2.138 0.760 20 years OL8 48 0.573 2.218 0.781 Natural vegetation OL9 66 0.592 2.480 0.762

A measure of species evenness (E) was calculated for each plot sampled. Evenness is constrained between zero and 1.0 with 1.0 representing a situation in which all species are equally abundant (Magurran 1988). Evenness values ranged from 0.573 (OL8, 20-year old) to 0.797 (OL5, 8-year old) with one 3-year old abandoned cropland (OL2, 3-year old) having an evenness value of 0.404 (Table 9.2). Species evenness values showed a marked increase up to the 8-year old abandoned cropland thereafter it dropped again and remained relatively steady at a lower level which is very close to the evenness value (0.592) of the natural vegetation in the vicinity of the 20-year old abandoned cropland (Table 9.2). Evenness values on the abandoned croplands were therefore comparable with previously determined values for the Roggeveld, which ranged from 0.501 to 0.820 (Chapter 7).

The Shannon indices calculated for the nine plots in this study ranged from 1.211 (OL2, 3- year old) to 3.014 (OL5, 8-year old) (Table 9.2), whereas values found in a study conducted in the natural vegetation ranged from 2.225 to 3.743 for the Mountain Renosterveld vegetation of the Roggeveld (Chapter 7). The Shannon and Simpson indices take both evenness and species richness into account (Magurran 1988). Shannon indices showed the same pattern as the evenness values with an increase up to eight years since abandonment and thereafter a sudden decline and remaining at that lower level. Simpson indices found for all nine plots surveyed showed little variation and ranged from 0.743 (OL4, 4-year old) to 0.884 (OL5, 8- year old) with the exception of OL2 (3-year old) with a Simpson index of 0.435 (Table 9.2). Simpson indices for the natural vegetation of the Roggeveld showed a greater variation and ranged from 0.604 to 0.944 (Chapter 7).

144

None of the diversity parameters, species evenness, Shannon or Simpson indices, reflected the trend in species richness, which showed a continuous increase in the number of species. This seems to indicate that although species richness increases, the dominance of a few species also increases disproportionately.

40 y = 1.2596x + 25.451 r2 = 0.6298 30 p = 0.0187

Number of species per 0.1 ha 10

0 0 5 10 15 20 25 30 35 Age of abandoned cropland (years)

Figure 9.5 Regression of the number of species per 1000 m² (0.1 ha) on the age of abandoned cropland.

Across all the survey plots 15 species contributed most to the cover found within a plot (Table 9.3). Eight of these species (Chrysocoma ciliata, Dicerothamnus rhinocerotis, Dimorphotheca cuneata, Euryops laterifolius, Helichrysum hamulosum, Merxmuellera stricta, Oedera genistifolia and Poa bulbosa) also contributed to the high vegetation cover of the natural vegetation. A maximum of five (Chrysocoma ciliata, Dicerothamnus rhinocerotis, Dimorphotheca cuneata, Medicago polymorpha and Selago cf. rigida) of the 15 species contributed a high cover, at any one time, on the abandoned croplands. Four of the seven annual species with the highest abundances were abundant only on the abandoned croplands up to an age of 8-years, these species were Cotula nudicaulis, Hordeum murinum, Karoochloa tenella and Medicago polymorpha. Thus, no annuals at high abundances were found on abandoned croplands of 10-years and older. Furthermore, the two introduced weedy species (Hordeum murinum and Medicago polymorpha) were virtually absent on croplands of 10-years and older. The perennial component cover varied greatly across all the surveyed plots. All eight perennial species with the highest abundances were present in the natural vegetation, which confirms the statement by Van der Putten et al. (2000) that in succession the identity of the local species matter. The fact that three of the perennial

145

species were present from the very early stages of succession seems to support the tolerance and not facilitation model of Connell and Slatyer (1977, Gurevitch et al. 2002).

Table 9.3 The percentage cover of the 15 most abundant species in abandoned croplands of various ages and in the natural vegetation in the Roggeveld

Species Annual/ OL1 OL2 OL3 OL4 OL5 OL6 OL7 OL8 OL9 Perennial Chrysocoma ciliata Perennial 10.0 35.0 0.5 1.0 10.0 5.0 40.0 25.0 5.0 Cotula nudicaulis Annual 2.0 0.5 0.5 0.5 0.5 Dicerothamnus Perennial 25.0 2.0 5.0 0.5 0.5 15.0 15.0 5.0 2.0 rhinocerotis Dimorphotheca Perennial 3.0 0.5 0.5 1.0 5.0 25.0 10.0 2.0 10.0 cuneata Erodium Annual 0.5 0.5 0.5 0.5 0.5 <0.1 0.5 0.5 moschatum Euryops laterifolius Perennial 0.5 0.5 0.5 0.5 0.5 10.0 Helichrysum Perennial 0.5 <0.1 0.5 0.5 0.5 25.0 2.0 hamulosum Hordeum murinum Annual 0.5 3.0 10.0 0.5 Karoochloa tenella Annual 0.5 5.0 0.5 1.0 0.5 0.5 Medicago Annual 1.0 3.0 10.0 20.0 0.5 <0.1 polymorpha Merxmuellera stricta Perennial <0.1 0.5 50.0 Oedera genistifolia Perennial 0.5 0.5 5.0 0.5 2.0 Pentaschistus Annual 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 0.5 aristifolia Poa bulbosa Annual 0.5 5.0 2.0 Selago cf. rigida Perennial 5.0 0.5 0.5 0.5 0.5 5.0 0.5 0.5

To analyse floristic patterns with respect to the data of the nine survey plots a Principal Co- ordinate Analysis (PCoA) was done. The resulting PCoA shows the relatively close proximity of the younger abandoned croplands (3-, 4- and 8-years old) to the older abandoned croplands (10-, 15- and 20-years old) and the large floristic gap between these abandoned croplands and the natural vegetation (OL9) (Figure 9.6). Thus, even although the regression of the age of abandoned cropland and number of species per 1000 m² indicates that after approximately 33-years an abandoned cropland will be as species rich as the natural vegetation (Figure 9.5) floristically, this abandoned cropland will still be appreciably different from the natural vegetation (Figure 9.6).

146

Time

Figure 9.6 A Principal Co-ordinate Analysis (PCoA) of the floristic data of the abandoned croplands (1-8) and the natural vegetation (9).

When comparing the position of the natural vegetation with the abandoned croplands it is implied that succession proceeds on a single set pathway and that it leads to a predictable stable community. However, large variability in the recovery of the vegetation is likely to be found. Differences among fields may be due to pre-abandonment treatments, such as the type of crop or cultivation practice, climatic conditions after abandonment or grazing management. Cramer and Hobbs (2007) stated that within limitations imposed by climate and soil characteristics, it does appear that the combination of past land use type and land use intensity explains much of the difference in patterns within similar ecosystem types.

The 2005 year in which these data were collected was a poor rainfall year. This is expected to have underestimated the number of species, especially with respect to annuals and geophytes, which comprise a large part of the species diversity in the region. Thus the species richness values for both the abandoned croplands and the undisturbed vegetation are expected to be higher than the values reported here. Additional annuals would have affected both the abandoned croplands and the natural vegetation species richness values, whereas the geophytic component would have had a larger impact on the natural vegetation survey data.

147

The pattern of recovery of Roggeveld Mountain Renosterveld seems to differ from the West Coast Renosterveld of the Cape Floristic Region. Studies conducted on Elandsberg Private Nature Reserve on remnant renosterveld vegetation and abandoned croplands indicated the apparent slow return of indigenous renosterveld vegetation on abandoned croplands (Midoko- Iponga et al. 2005). The main difference between the two areas is the dominance by introduced alien annual grasses on abandoned croplands in the West Coast Renosterveld. These weedy grasses arrest the whole recovery process. Current restoration efforts aim to reduce the cover of the introduced grasses while at the same time maintaining or even increasing species richness and diversity of indigenous target species (Krug & Krug 2007). In spite of the Roggeveld probably having harsher environmental conditions, recovery on abandoned croplands is occurring and seems to be continuing with a steady increase in species richness occurring and values of evenness, Shannon and Simpson indices of diversity being similar to those of the natural vegetation from approximately 10 years after abandonment.

All life forms are well represented within 20-years of cropland abandonment, however cryptophyte species are still underrepresented. This is to be expected since these species are eradicated by ploughing and reproduction, by vegetative means or seed set, occurs slowly. Geophyte species are an important component of the Roggeveld flora and from a diversity point of view, the return of these species on abandoned croplands is important.

No fires were experienced on any of the surveyed abandoned croplands, but it is expected that fire would simply slow down secondary succession with early successional species such as Dicerothamnus rhinocerotis, Chrysocoma ciliata and Dimorphotheca cuneata re- establishing within the next growing season. This is supported by data collected on post-fire monitoring plots in the Roggeveld that show D. rhinocerotis seedling established in plots within nine months following a fire (Chapter 10).

9.5 Conclusions

The Roggeveld, located within the Fynbos Biome, has a higher rainfall than the surrounding areas and has been used to cultivate wheat and other fodder crops for hundreds of years. Primarily due to an increase in production costs, many croplands are no longer utilised and now lie barren.

Chamaephyte and therophyte species were the most abundant life forms on abandoned croplands of all ages while hemicryptophyte, phanerophyte, liana and parasite species were the least abundant life forms. However, liana and parasite species were also seldom found in the natural vegetation. Therophyte species were the most abundant on young abandoned croplands and decreased in number from 10-years of age to values similar to those

148

encountered in the natural vegetation. Chamaephyte, cryptophyte and hemicryptophyte species increased in number with an increase in age of the abandoned croplands, the highest values being found for the natural vegetation. Phanerophyte species were few and did not differ much across the age range of the surveyed plots.

The relative cover of each life form expressed as a percentage of the total cover indicates the overwhelming contribution made by chamaephytes which increase along the successional sequence and remained at a high cover in the natural vegetation. Phanerophyte contributions fluctuated throughout the abandoned cropland succession, while therophyte cover was high on the 3- to 8-year old sites, thereafter decreasing substantially and remaining at a low value, similar to the natural vegetation. Hemicryptophytes contributed negligibly to cover on the abandoned croplands and in the natural vegetation. Cryptophyte cover fluctuated on abandoned croplands but at all times remained approximately a quarter lower than in the natural vegetation.

Comparison of slope and intercept values of the exponential function species-area curves across all the abandoned croplands and the natural vegetation found that the curves of the abandoned croplands differed significantly from those of the natural vegetation in all instances.

Species richness increased with the age of the abandoned cropland, with the species richness for the natural vegetation being the highest at 66 species per 1000 m², however, a similar increase in the values of evenness, Shannon and Simpson indices were not found with increasing age of the abandoned cropland. A regression of species richness against age of abandoned croplands predicted that an abandoned cropland of approximately 33-years should be as species rich as the natural vegetation. Yet, a Principal Co-ordinate Analysis of the floristic data indicated that floristically all the abandoned croplands were still extremely different from the natural vegetation.

Across all the survey plots, 15 species contributed most to the high cover with eight of these also contributing to the high cover of the natural vegetation. Four annual species were present in high abundances on abandoned croplands of ≤ 8-years of age, with no indigenous or alien invasive annual species at high abundances on older abandoned croplands. The perennial component cover varied greatly across all the surveyed plots.

The results of this study contribute to our understanding of restoration efforts on abandoned croplands in the Roggeveld. The Roggeveld Mountain Renosterveld abandoned croplands surveyed were not negatively influenced by alien annual grasses as in the West Coast Renosterveld and species richness increased continually. Recovery of abandoned croplands seems to occur naturally, however after 20-years, cryptophyte species were still largely

149

underrepresented in comparison to the natural vegetation. It is expected that the influence of fire would simply slow down secondary succession with early colonisers establishing again by the next growing season.

9.6 Acknowledgements

The authors would like to thank the Critical Ecosystem Partnership Fund (CEPF) through the Succulent Karoo Ecosystem Plan/Program (SKEP) initiative for funding the project. The Critical Ecosystem Partnership Fund is a joint initiative of Conservation International, the Global Environmental Facility, the Government of Japan, the MacArthur Foundation and the World Bank. A fundamental goal is to ensure civil society is engaged in biodiversity conservation. The various people who assisted with the field work are gratefully acknowledged. This research was supported by the National Research Foundation under grant number 61277.

9.7 References

ARMESTO, J.J. AND PICKETT, S.T.A. 1986. Removal experiments to test mechanisms of plant succession in old fields. Vegetatio 66, 85-93. BARKMAN, J.J. 1979. The investigation of vegetation texture and structure. In: M.J.A. Werger (Ed.). The study of vegetation, pp. 125-160. Junk, London. BAZZAZ, F.A. 2000. Plants in changing environments. Linking physiological, population and community ecology. Cambridge University Press, Cambridge. CAIN, S.A. 1950. Life-forms and phytoclimate. The Botanical Review 16, 1-32. CONNELL, J.H. AND SLATYER, R.O. 1977. Mechanisms of succession in natural communities and their role in community stability and organisation. The American Naturalist 111, 1119-1144. CONNOR, E.F. AND McCOY, E.D. 1979. The statistics and biology of the species-area relationship. The American Naturalist 113, 791-833. COUNCIL FOR GEOSCIENCE 2008. Geological data 1: 250 000. CD data provided by the Council for Geoscience, Silverton, Pretoria. COWLING, R.M., GIBBS RUSSEL, G.E., HOFFMAN, M.T. AND HILTON-TAYLOR, C. 1989. Patterns of plant species diversity in southern Africa. In: B.J. Huntley (Ed.). Biotic diversity in southern Africa. Concepts and Conservation, pp. 19-50. Oxford University Press, Cape Town. COWLING, R.M. AND HILTON-TAYLOR, C. 1994. Patterns of plant diversity and endemism in southern Africa: an overview. In: B.J. Huntley (Ed.). Botanical diversity in southern Africa. Strelitzia 1, pp. 31-52. National Botanical Institute, Pretoria.

150

COWLING, R.M., HOLMES, P.M. AND REBELO, A.G. 1992. Plant diversity and endemism. In: R.M. Cowling (Ed.). The Ecology of Fynbos: Nutrients, fire and diversity, pp. 62- 112. Oxford University Press, Cape Town. CRAMER, V.A. AND HOBBS, R.J. (Eds) 2007. Old fields. Dynamics and restoration of abandoned farmland. Island Press, Washington. CRAMER, V.A., HOBBS, R.J. AND STANDISH, R.J. 2007. What’s new about old fields? Land abandonment and ecosystem assembly. Trends in Ecology and Evolution 23, 104- 112. CULLINAN, P. 2003. Robert Jacob Gordon 1743 – 1795. The man and his travels at the Cape. http:// web.uct.ac.za/depts/age/people/Gordon/frameset.htm DEBUSSCHE, M., JOSÉ, E., LEPART, J., HOUSSARD, C. AND LAVOREL, S. 1996. Changes in Mediterranean plant succession: old-fields revisited. Journal of Vegetation Science 7, 519-526. DIELS, L. 1909. Formationen und florenelemente im nordwestlichen Kapland. Botanische Jahrbücher. 44, 91-124. FRANCIS, M.L., FEY, M.V., PRINSLOO, H.P., ELLIS, F., MILLS, A.J. AND MEDINSKI, T.V. 2007. Soils of Namaqualand: Compensations for aridity. Journal of Arid Environments 70, 588-603. GIBSON, D.J., MIDDLETON, B.A., FOSTER, K., HONU, Y.A.K., HOYER, E.W. AND MATHIS, M. 2005. Species frequency dynamics in an old-field succession: effects of disturbance, fertilization and scale. Journal of Vegetation Science 16, 415-422. GUREVITCH, J., SCHEINER, S.M. AND FOX, G.A. 2002. The Ecology of Plants. Chapter 13. Disturbance and Succession, pp. 253-274. Sinauer, Massachusetts, U.S.A. HILTON-TAYLOR, C. 1994. Western Cape Domain (Succulent Karoo). In: S.D. Davis, V.H. Heywood and A.C. Hamilton (Eds). Centres of plant diversity. A guide and strategy for their conservation, pp. 201-203. IUCN Publications Unit, Cambridge. HUBERTY, L.E., GROSS, K.L. AND MILLER, C.J. 1998. Effects of nitrogen addition on successional dynamics and species diversity in Michigan old-fields. Journal of Ecology 86, 794-803. INOUYE, R.S., HUNTLY, N.J., TILMAN, D., TESTER, J.R., STILLWELL, M. AND ZINNEL, K.C. 1987. Old-field succession on a Minnesota sand plain. Ecology 68, 12-26. KRUG, C.B. AND KRUG, R.M. 2007. Restoration of old fields in Renosterveld: a case study in a Mediterranean-type shrubland of South Africa. In: V.A. Cramer and R.J. Hobbs (Eds). Old fields. Dynamics and restoration of abandoned farmland, pp. 265-285. Island Press, Washington. LAVOREL, S., LEPART, J., DEBUSSCHE, M., LEBRETON, J-D. AND BEFFY, J-L. 1994. Small scale disturbances and the maintenance of species diversity in Mediterranean old fields. OIKOS 70, 455-473. LOW, A.B. AND REBELO, A.G.. 1996. Vegetation of South Africa, Lesotho and Swaziland. Department of Environmental Affairs and Tourism, Pretoria.

151

MAGURRAN, A.E. 1988. Ecological Diversity and its measurement. Cambridge University Press, Cambridge. MARLOTH, R. 1908. Das Kapland, insonderheit das Reich der Kapflora, das Waldgebiet und die Karoo, pflanzengeografisch dargestellt. Wissenschaftliche Ergebnisse der Deutscher Tiefsee-Expedition ‘Waldivia’, 1898 – 1899. 2, T. 3, Fischer, Jena. McCUNE, B. AND GRACE, J.B. 2002. Analysis of Ecological Communities. MjM Stoftware Design, Gleneden Beach, Oregon. McINTYRE, S., LAVOREL, S. AND TREMONT, R.M. 1995. Plant life-history attributes: their relationship to disturbance response in herbaceous vegetation. Journal of Ecology 83, 31-44. MIDOKO-IPONGA, D., KRUG, C.B. AND MILTON, S.J. 2005. Competition and herbivory influence growth and survival of shrubs on old fields: implications for restoration of renosterveld shrubland. Journal of Vegetation Science 16, 685-692. MUCINA, L. RUTHERFORD, M.C. AND POWRIE, L.W. (Eds) 2005. Vegetation map of South Africa, Lesotho and Swaziland, 1 : 1 000 000 scale sheet maps. South African National Biodiversity Institute, Pretoria. MUELLER-DOMBOIS, D. AND ELLENBERG, H. (Eds) 1974. Aims and methods of vegetation ecology. Wiley, New York. OTTO, R., KRÜSI, B.O., BURGA, C.A. AND FERNÁNDEZ-PALACIOS, J.M. 2006. Old-field succession along a precipitation gradient in the semi-arid coastal region of Tenerife. Journal of Arid Environments 65, 156-178. PODANI, J. 2001. SYN-TAX 2000 Computer programs for data analysis in ecology and systematics. Scientia publishing, Budapest. PRIEUR-RICHARD, A-H., LAVOREL, S., GRIGULIS, K. AND DOS SANTOS, A. 2000. Plant community diversity and invasibility by exotics: invasion of Mediterranean old fields by Conyza bonariensis and Conyza canadensis. Ecology Letters 3, 412-422. QUINN, G.P. AND KEOUGH, M.J. 2002. Experimental design and data analysis for biologists. Cambridge University Press, Cambridge. RAUNKIAER, C. 1934. The life forms of plants and statistical plant geography. Oxford University Press, Oxford. REBELO A.G. 2001. The state of plants in the Cape flora. In: G.H. Verdoorn and J. Le Roux (Eds). The state of South Africa’s species, pp. 18-43. Endangered Wildlife Trust, Pretoria. REBELO, A.G., BOUCHER, C., HELME, N., MUCINA, L. AND RUTHERFORD, M.C. 2006. Fynbos Biome. In: L. Mucina and M.C. Rutherford (Eds). The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19, pp. 52-219. South African National Biodiversity Institute, Pretoria. ROUGET, M., JONAS, Z., COWLING, R.M., DESMET, P.G., DRIVER, A., MOHAMED, B., MUCINA, L., RUTHERFORD, M.C. AND POWRIE, L.W. 2006. Ecosystem status and protection levels of vegetation types. In: L. Mucina and M.C. Rutherford (Eds). The

152

vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19, pp. 724-737. South African National Biodiversity Institute, Pretoria. RUBIDGE, B.S. AND HANCOX, P.J. 1999. The Karoo – a palaeontological wonderland. In: Viljoen, M.J. and Reimold, W.U. (Eds). An introduction to South Africa’s geological and mining heritage, pp. 83-91. Published by the Geological Society of South Africa and Mintek. RUTHERFORD, M.C. AND WESTFALL, R.H. 1994. Biomes of Southern Africa. An objective characterisation. Memoirs of the Botanical Survey of South Africa 63, 1-94. SCHEINER, S.M. 2003. Six types of species-area curves. Global Ecology and Biogeography 12, 441-447. SCHEINER, S.M. 2004. A mélange of curves – further dialogue about species-area relationships. Global Ecology and Biogeography 13, 479-484. SEMENOVA, G.V. AND VAN DER MAAREL, E. 2000. Plant functional types – a strategic perspective. Journal of Vegetation Science 11, 917-922. SHEARING, D. 1997. Karoo: South African Wildflower Guide 6. National Book Printers, Drukkery street, Goodwood, Western Cape. SHMIDA, A. 1984. Whittaker’s plant diversity sampling method. Israel Journal of Botany 33, 41-46. SQUIERS, E.R. AND WISTENDAHL, W.A. 1976. Changes in plant species diversity during early secondary succession in an experimental old-field system. The American Midland Naturalist 98, 11-21. STIRLING, G. AND WILSEY, B. 2001. Empirical relationships between species richness, evenness, and proportional diversity. The American Naturalist 158, 286-299. STROMBERG, M.R. AND GRIFFIN, J.R. 1996. Long-term patterns in coastal California grasslands in relation to cultivation, gophers, and grazing. Ecological applications 6, 1189-1211. TJØRVE, E. 2003. Shapes and functions of species-area curves: a review of possible models. Journal of Biogeography 30, 827-835. TUNNELL, S.J., ENGLE, D.M. AND JORGENSEN, E.E. 2004. Old-field grassland successional dynamics following cessation of chronic disturbance. Journal of Vegetation Science 15, 431-436. VAN DER MERWE, H., VAN ROOYEN, M.W. AND VAN ROOYEN, N. 2008. Vegetation of the Hantam-Tanqua-Roggeveld subregion, South Africa. Part 1. Fynbos Biome related vegetation. Koedoe 50, 160-183. VAN DER MERWE, P.J. 1938. Die trekboer in die geskiedenis van die Kaapkolonie (1657- 1842). Nasionale Pers Beperk, Kaapstad. VAN DER MERWE, P.J. 1988. Die noordwaartse beweging van die boere voor die Groottrek (1770-1842). The State Library, Pretoria. VAN DER PUTTEN, W.H., MORTIMER, S.R., HEDLUND, K., VAN DIJK, C., BROWN, V.K., LEPŠ, J., RODRIGUES-BARRUECO, C., ROY, J., DIAZ LEN, T.A., GROMSEN, D.,

153

KORTHALS, G.W., LAVOREL, S., SANTA REGINA, I. AND SMILAUER, P. 2000. Plant species diversity as a driver of early succession in abandoned fields: a multi- site approach. Oecologia 124, 91-99. VANDVIK, V. AND BIRKS, H.J.B. 2002. Pattern and process in Norwegian upland grasslands: a functional analysis. Journal of Vegetation Science 13, 123-134. VAN ROOYEN, M.W., THERON, G.K. AND GROBBELAAR, N. 1990. Life form and dispersal spectra of the flora of Namaqualand, South Africa. Journal of Arid Environments 19, 133-145. VAN WYK, A.E. AND SMITH, G.F. (Eds) 2001. Regions of Floristic Endemism in Southern Africa: A review with emphasis on succulents, pp. 1-199. Umdaus Press, Pretoria. VEECH, J.A. 2000. Choice of species-area function affects identification of hotspots. Conservation Biology 14, 140-147. WEATHER BUREAU 1998. Climate of South Africa. Climate statistics up to 1990. WB 42. Government Printer, Pretoria. WEIMARCK, H. 1941. Phytogeographical groups, centres and intervals within the Cape flora. Lunds Universitets Årsskrif Avd. 2. 37, 1-143. ZHAN, X., LI, L. AND CHENG, W. 2007. Restoration of Stipa kryloviisteppes in inner Mongolia of China: assessment of seed banks and vegetation composition. Journal of Arid Environments 68, 298-307. ZHANG, J-T. 2005. Succession analysis of plant communities in abandoned croplands in the eastern Loess plateau of China. Journal of Arid Environments 63, 458-474.

154

Chapter 10

Vegetation trends following fire in the Roggeveld, South Africa

Abstract

The Mountain Renosterveld vegetation of the Roggeveld is an escarpment type renosterveld showing strong karroid affinities with fire playing an important role as a landscape scale disturbance that shapes these plant communities. Line transect data accumulated over ten years were analysed and this paper reports on the post-fire vegetation trends with respect to changes in vegetation cover, species numbers and life form categories over this timeframe.

Within the first nine months following fire, vegetation began to establish with the vegetation cover remaining at a higher level from year 3 to year 10. Species richness varied between 13 and 17 species at the first survey with the highest species richness generally encountered after three years at each transect. Shannon index values varied greatly between the five localities with four of the localities showing similar trends after the fire. The highest Shannon index values were generally found within the first three years and lowest Shannon index values found in year 9 and year 10. Species composition data for each plot over the ten year period were ordinated using Principal Co-ordinate Analysis. In all cases, these ordinations indicated a clear separation in species composition between the first two years (year 1 and year 2) following the fire and the remaining years (year 3 to year 10).

This study seems to lend support to the ‘initial floristic composition’ model of Egler (1954) with all or the majority of species encountered during succession already present at the beginning of the recovery phase and a rapid re-establishment of the initial plant community.

Keywords: Fynbos Biome, ‘initial floristic composition’ model, life forms, long-term monitoring, Mountain Renosterveld, post-fire vegetation recovery, vegetation cover

155

10.1 Introduction

The Mountain Renosterveld vegetation covering the Roggeveld mountain range is an island of Fynbos Biome vegetation surrounded by Succulent Karoo Biome vegetation (Van der Merwe et al. 2008). The Fynbos Biome coincides more or less with the area covered by the Cape Floral Kingdom which is recognised as one of the world’s six floristic kingdoms (Good 1947), on par with much larger regions worldwide (Rebelo et al. 2006). The Fynbos Biome also constitutes the largest portion of the Cape Floristic Region (CFR) a region which is internationally renowned for its exceptional species diversity and which is recognised as one of 34 global hotspots of biodiversity (www.biodiversityhotspot.org – accessed 22 February 2009). This biome comprises three quite different, naturally fragmented vegetation types, namely fynbos, renosterveld and strandveld, that occur in winter- and summer-rainfall areas and are dominated by small-leaved, evergreen shrubs whose regeneration is intimately related to fire (Rebelo et al. 2006).

Renosterveld is an evergreen, fire-prone shrubland or grassland dominated by small, cupressoid-leaved, evergreen asteraceous shrubs (principally Dicerothamnus rhinocerotis, renosterbos, rhino bush) with an understory of grasses (Poaceae) and a high biomass and diversity of geophytes (Moll et al. 1984, McDowell & Moll 1992, Cowling et al. 1997, Rebelo et al. 2006). A major feature of renosterveld, at least the coastal units, is the extensive transformation that has taken place over the last 100 years (Rebelo et al. 2006). The escarpment renosterveld types, such as on the Roggeveld Mountains, show strong karroid affinities.

Dicerothamnus rhinocerotis, after which the renosterveld is named, is an indigenous perennial species with encroaching properties and is not grazed (Shearing 1997). Early explorers to the Roggeveld indicated the abundance of grasses in renosterveld and that as a result of high grazing pressure renosterbos was increasing in abundance and grasses were becoming scarcer (transcriptions and translation of R.J. Gordon’s travels, Cullinan 2003). Marloth (1908) also mentioned D. rhinocerotis dominance and suggested that frequent burning of vegetation by the early settlers to increase palatable vegetation for their livestock was actually increasing the dominance of D. rhinocerotis.

Fire is a landscape scale disturbance that creates gaps in plant communities which provide space for plant establishment (Carson & Pickett 1990). Disturbance by fire can contribute to the maintenance of diversity in two manners: firstly, fire contributes to the maintenance of species richness by avoided competitive exclusion and secondly, fire can increase spatial heterogeneity (Lavorel et al. 1994).

156

The Mediterranean-type vegetation is one of the world’s major fire-prone biomes (Capitanio & Carcaillet 2008). In areas in which this type of vegetation occurs, as for example the Mountain Renosterveld of the Roggeveld, fire is a crucial process controlling vegetation dynamics and structure (Capitanio & Carcaillet 2008), with the post-fire regeneration process highly dependent on the pre-fire vegetation (Hanes 1971, Trabaud & Lepart 1980, Pausas 1999, Lloret & Vilá 2003). Various studies have found that in such ecosystems species composition and structure rapidly recover after fire (Hanes 1971, Lloret & Vilá 2003). In fynbos, the effect of fire on species composition, vegetation structure and successional patterns depends on the frequency, intensity and season of the fire and it was found that there is also considerable variation in response between sites (Kruger & Bigalke 1984, Cowling et al. 1997).

One of the traits frequently used in attempts to classify species according to their response to disturbance, is growth form (McIntyre et al. 1995). A description of growth form, expressed in terms of dormant bud position, as developed by Raunkiaer (1934), encapsulates sets of correlated traits relating to persistence and architecture that are relevant to disturbance response (McIntyre et al. 1995, 1999). Guo (2001) found that using plant groups provided a useful framework for describing post-fire chaparral succession because they affect ecosystem processes in some predictable ways and may also reflect the underlying environmental changes after fire.

On 26 January 1999 more than 10 000 ha burnt in a lightning induced fire in the Roggeveld. This event created an opportunity to investigate the post-fire recovery of the vegetation in the Mountain Renosterveld. The aim of the current paper is to report on post-fire vegetation trends in the Mountain Renosterveld vegetation of the Roggeveld with reference to the changes in species numbers, vegetation cover and life form categories over a time-span of 10 years (1999-2008). Additionally, a Principal Co-ordinate Analysis was used to investigate species compositional changes over the ten-year period for each survey plot.

10.2 Study area

The study was conducted on three adjacent farms (Botuin, Klawervlei and Droëkloof) in the Mountain Renosterveld vegetation of the Roggeveld (Figure 10.1, Table 10.1). According to Mucina et al. (2005) and Rebelo et al. (2006), the study area falls within the Roggeveld Shale Renosterveld vegetation type, whereas the finer scale classification and mapping by Van der Merwe et al. (2008) indicated that the study area was part of the Dicerothamnus rhinocerotis Mountain Renosterveld plant association (association 2). Plant association 2 is dominated by D. rhinocerotis with Merxmuellera stricta and Chrysocoma ciliata further characterising the vegetation unit. Following good winter rains the annual and geophyte component make up a large part of the vegetation.

157

Figure 10.1 Location of the post-fire monitoring transects on three adjacent farms in the Roggeveld.

Although the rains within this region fall mainly in winter, a few summer thunderstorms contribute to the total annual precipitation of 132 mm to 467 mm per year (Weather Bureau 1998). Winter snowfalls occur with a mean of 6 snow days recorded per year over a 24-year period for Sutherland (Weather Bureau 1998). The mean daily maximum temperature for January (the warmest month) measured at the Sutherland is 27.1°C, while the mean daily minimum temperature for July (the coldest month) is -2.4°C.

Rocks of the Ecca Group cover most of the Roggeveld Mountains (Rubidge & Hancox 1999) with shallow stony lithosols characteristic (Francis et al. 2007). Three of the five post-fire monitoring sites were located on Land Type Da 69 with red B horizons, while the other two sites were located on Land Type Fa 650 with lime rare or absent in the entire landscape (Council for Geoscience 1989) (Table 10.1).

158

Table 10.1 Prominent features of the five post-fire monitoring sites in the Roggeveld, indicating GPS starting point co-ordinates, altitude, aspect, land type and dominant perennial plant species ten years after a fire

Site 1 Site 2 Site 3 Site 4 Site 5 GPS co- 31°58’35.6”S 31°59’15.2”S 31°58’39.6”S 31°55’49.7”S 31°55’19.8”S ordinates: 20°00’27.3”E 20°01’11.5”E 20°01’26.2”E 20°01’34.7”E 20°01’43.9”E Starting point Altitude 1440 m 1371 m 1296 m 1334 m 1385 m Aspect East North West South South Land type Da 69 Da 69 Da 69 Fa 650 Fa 650 Dominant Dicerothamnus Merxmuellera Merxmuellera Dicerothamnus Leysera perennial plant rhinocerotis, stricta, stricta, rhinocerotis, gnaphalodes, species after Merxmuellera Chrysocoma Chrysocoma Merxmuellera Lotononis 10 years stricta, ciliata, ciliata stricta hirsuta following a fire Muraltia Dicerothamnus vulnerans rhinocerotis

10.3 Materials and methods

To acquire objective quantitative data on the vegetation changes that occurred following a fire in 1999, a point or plotless method was used. Due to the steep slopes and rock-strewn areas, the descending point method (Roux 1963, Novellie & Strydom 1987) was deemed the most appropriate method to track post-fire vegetation trends at the five selected sites in the Roggeveld. A canopy strike was recorded when the descending point touched any plant material or fell within the canopy spread of the individual. When considering only basal strikes the perennial grasses are favoured above the karoo bush component, because perennial grasses have a small base to canopy spread ratio (Vorster 1982).

The transects were permanently marked with iron poles (‘droppers’) indicating the beginning and end points of a 50 m rope which was marked at 1 m intervals. Four lines, 1 m apart and parallel to one another were surveyed in order to limit the chance of surveying transitional areas between different vegetation types or habitats. The transects were monitored yearly in the last week of September or the first week of October. Ten years of data were collected between 1999 and 2008.

The number of strikes on a species were calculated as a percentage of the total number of point observations made and were not expressed as a percentage of all the strikes (Du Toit 1998a). Thus, in this study, the number of strikes per species were expressed as a percentage of the 200 points surveyed and these totals added to determine the percentage vegetation cover. The sum of the individual plant species percentages obtained rarely totals

159

one hundred because the number of strikes observed are fewer than the total number of point observations made (Du Toit 1998b).

Additionally, the species were classified into the dominant life forms as defined by Raunkiaer (1934) and modified by Mueller-Dombois and Ellenberg (1974, Appendix 1). Temporal changes in the post-fire vegetation were investigated in terms of (a) total vegetation cover (b) total species richness, (c) Shannon index of diversity, (d) vegetation cover per life form, (e) species richness per life form, and (f) changes in species composition across a 10 year period.

Shannon’s index of diversity (H’), often referred to as Shannon-Wiener index, was calculated for each sampled plot using the PC-ORD computer program (PC-ORD Version 4 for Windows, MjM Software design) which calculates this diversity measure as follows: H’ = Shannon diversity

H’ = - ∑ pi log pi i

Where pi = importance probability of species i.

The species compositional data for each of the five post-fire monitoring transects over the ten year period were ordinated using Principal Co-ordinate Analysis (PCoA) in the SYN-TAX computer program (Podani 2001). Principal Co-ordinate Analysis generally performs well with species data because it allows the use of a wide array of distance measures and therefore gives a marked improvement over Principal Component Analysis (McCune & Grace 2002).

10.4 Results and discussion

The vegetation cover increased within the first nine months following the fire (Figure 10.2). Vegetation cover at sites 1 and 5 increased steadily from 1999 to 2001 (Figure 10.2a, 10.2e), while the vegetation cover at sites 2 and 3 decreased between 1999 and 2000 and then increased from 2000 to 2001 (Figure 10.2b, 10.2c). Post-fire transect 4 had no change in vegetation cover between 1999 and 2000 but vegetation cover increased in 2001 (Figure 10.2d). The high vegetation cover recorded in the 2001 season could be attributed to the region receiving good rains in that year. From 2001 until 2008 vegetation cover in all the post- fire monitoring transects remained higher than the vegetation cover found immediately following the fire in 1999 and 2000. In most instances there was a decrease in vegetation cover in 2003, 2004 and 2005. These three years were characterised by a low rainfall and drought conditions resulting in a low therophyte vegetation cover (Figure 10.2). The vegetation on Land Type Da had a far higher cover of perennial species (59.5% - 71%) than Land Type Fa (35.5% - 47%).

160

a) b) 100 650 100 650 90 90 550 550 80 80 70 450 70 450 60 350 60 350 50 50 40 250 40 250 30 150 30 150 Rainfall (mm) Rainfall (mm) Rainfall 20 20

Vegetation cover (%) 50 Vegetation cover (%) 50 10 10 0 -50 0 -50 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Year Year

Phanerophytes Chamaephytes Hemicryptophytes Cryptophytes Phanerophytes Chamaephytes Hemicryptophytes Cryptophytes Therophytes Lianas Rainf all Therophytes Lianas Rainf all

c) d) 100 650 100 650 90 90 550 550 80 80 70 450 70 450 60 350 60 350 50 50 250 40 40 250 30 150 30 Rainfall (mm) Rainfall 150 Rainfall (mm) Rainfall 20 20

Vegetation cover (%) cover Vegetation 50

Vegetation cover (%) cover Vegetation 50 10 10 0 -50 0 -50 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Year Year

Phanerophytes Chamaephytes Hemicryptophytes Cryptophytes Phanerophytes Chamaephytes Hemicryptophytes Cryptophytes Therophytes Lianas Rainf all Therophytes Lianas Rainf all

e) 100 650 90 550 80 70 450 60 350 50 40 250 30 150 Rainfall (mm) Rainfall 20

Vegetation cover (%) cover Vegetation 50 10 0 -50 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Year

Phanerophytes Chamaephytes Hemicryptophytes Cryptophytes Therophytes Lianas Rainf all Figure 10.2 Vegetation cover changes per life form (columns) determined from a 200-point survey and annual rainfall (line) for five post-fire monitoring transects: a) site 1, b) site 2, c) site 3, d) site 4; and e) site 5.

Within nine months of the fire, the total number of species surveyed in the 200-point line transects ranged between 13 and 17 species per transect (Figure 10.3). At all sites on Land Type Da the highest species richness was reached within 3 years (2001) after the fire, whereafter species richness declined or remained more or less constant. A similar increase in the second and third post-fire year and subsequent decrease in species richness was reported in Californian chaparral (Guo 2001). In the case of the sites on Land Type Fa, species richness initially declined after the high value in 2001 but increased again to reach a maximum in 2006. Therophyte species contributed more to species numbers on Land Type Fa (Figure 10.3d & 10.3c) than on Land Type Da (Figure 10.3a-c. In all cases, chamaephyte species made a large contribution to the total species numbers.

161

a) b) 40 40 35 35 30 30 25 25 20 20 15 15 10 10 5 5 Total number of species of number Total 0 species of number Total 0 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Year Year

Phanerophytes Chamaephytes Hemicryptophytes Phanerophytes Chamaephytes Hemicryptophytes Cryptophytes Therophytes Lianas Cryptophytes Therophytes Lianas

c) d) 40 40 35 35 30 30 25 25 20 20 15 15 10 10 5 5 Total number of species of number Total 0 species of number Total 0 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Year Year

Phanerophytes Chamaephytes Hemicryptophytes Phanerophytes Chamaephytes Hemicryptophytes Cryptophytes Therophytes Lianas Cryptophytes Therophytes Lianas

e) 40 35 30 25 20 15 10 5 Total number of species of number Total 0 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Year

Phanerophytes Chamaephytes Hemicryptophytes Cryptophytes Therophytes Lianas

Figure 10.3 Total number of species per life form determined from a 200-point survey for the five post-fire monitoring transects: a) site 1, b) site 2, c) site 3, d) site 4; and e) site 5.

Various studies in Mediterranean-type ecosystems have indicated quick community recovery after fire and concluded that species richness was at its highest soon after fire (Capitanio and Carcaillet 2008). Species diversity has been found to peak within the first year (Keeley et al. 1981, Schwilk et al. 1997) or second year (Potts et al. 2003) with Guo (2001) reporting the highest species richness in the second post-fire year on north-facing slopes and in the third year on south-facing slopes. In all instances species richness was reported to decline in the following years. Additionally, when studying the species composition soon after the fire it becomes evident that all, or the majority of, species present during the succession are in place from the beginning of the recovery phase. It appears that species richness in Fynbos vegetation is higher soon after fire and succession proceeds by successive elimination of species (Bond & Van Wilgen 1996). Since data collected in this study support the fact that nearly all of the species were already present in the early recovery phase and that the initial plant community established rapidly, the ‘initial floristic composition’ model of Egler (1954) seems to apply to this ecosystem, however, further research is needed to resolve which, if any, model of vegetation dynamics most accurately reflects post-fire dynamics in Mediterranean ecosystems (Capitanio & Carcaillet 2008).

162

Shannon’s index values ranged from 1.107 (site 2) to 3.030 (site 5) (Table 10.2) across the ten year period at the five monitoring sites. The highest two values for site 1 to site 4 were found within the first three years following fire, yet, the highest two values for site 5 were found in year 3 (2001) and year 10 (2008). Four of the sites (sites 1 to 4) showed similar trends, with the lowest Shannon values ranging between 1.107 (site 2) and 1.521 (site 1) and highest Shannon values varying from 2.093 (site 2) to 2.258 (site 1) (Table 10.2). In general, the lowest Shannon values in these sites were found in year 9 and year 10. The lowest Shannon value for site 5 was 2.564, which is higher than the highest values encountered in site 1 to site 4 (Table 10.2). This initial high Shannon value following fire results from the presence of a high number of geophytes (cryptophyte species) and annuals (therophyte species), which react quickly after fire.

Table 10.2 Shannon index values for five post-fire monitoring transects from 1999 to 2008

Year Site 1 Site 2 Site 3 Site 4 Site 5 1999 2.258 2.093 2.151 2.043 2.564 2000 1.769 2.009 1.781 1.874 2.694 2001 2.078 1.973 2.058 2.195 2.979 2002 1.971 1.570 1.775 1.971 2.913 2003 1.445 1.147 1.454 1.751 2.613 2004 1.727 1.219 1.635 1.905 2.660 2005 1.604 1.340 1.466 1.983 2.927 2006 1.548 1.171 1.369 2.015 2.739 2007 1.568 1.107 1.296 1.410 2.939 2008 1.521 1.143 1.475 1.580 3.030

Upon classifying the species encountered in the field surveys into Raunkiaer (1934) classic life form categories (Appendix 1), two different patterns emerged across the 10 years of survey results. The three post-fire monitoring transects located on Land Type Da (site 1, site 2 and site 3) produced different life form spectra from the two post-fire monitoring sites located on Land Type Fa (Site 4 and Site 5) (Figure 10.4 and 10.5). In general, the contribution of phanerophyte and chamaephyte species was higher on Land Type Da (Figure 10.4a-c) than on Land Type Fa (Figure 10.4d, 10.4e), whereas the contribution of therophyte species was larger on Land Type Fa than Land Type Da (Figure 10.4a-e).

163

b) a) 100% 100%

80% 80%

60% 60%

40% 40% Species per life form life per Species Species per form life 20% 20%

0% 0% 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Year Year

Phanerophytes Chamaephytes Hem icryptophytes Phanerophytes Chamaephytes Hemicryptophytes Cryptophytes Therophytes Liana Cryptophytes Therophytes Liana

c) d) 100% 100%

80% 80%

60% 60%

40% 40%

Species per form life 20% Species per form life 20%

0% 0% 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Year Year

Phanerophytes Chamaephytes Hem icryptophytes Phanerophytes Chamaephytes Hem icryptophytes Cryptophytes Therophytes Liana Cryptophytes Therophytes Liana

e) 100%

80%

60%

40%

Species per form life 20%

0% 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Year

Phanerophytes Chamaephytes Hem icryptophytes Cryptophytes Therophytes Liana

Figure 10.4 Number of species per life form expressed as a percentage of the total number of species for post-fire monitoring transects: a) site 1, b) site 2, c) site 3, d) site 4; and e) site 5.

The percentage contribution per life form to the vegetation cover produced different results (Figure 10.5) from the percentage contribution at a species level (Figure 10.4). Phanerophyte species contributed to a larger proportion of the vegetation cover than their species numbers suggested at site 1, 2 and 4 (Figure 10.5a, 10.5b and 10.5d). This was primarily as a result of the dominance by D. rhinocerotis. Site 3 was a Merxmuellera stricta grass dominated site with the grass limiting D. rhinocerotis establishment, hence, the lower phanerophyte species vegetation cover contribution (Figure 10.5c). Chamaephyte species cover dominated in the three Land Type Da post-fire monitoring transects (Figure 10.5a-c), whereas the contribution of therophyte species to the overall vegetation cover of these transects was limited to a small percentage in the first four years following the fire (Figure 10.5a-c). In contrast, therophyte species cover contribution on Land Type Fa (Figure 10.5d, 10.5e) remained relatively high throughout the first 10 years following the fire.

164

a) b) 100% 100%

80% 80%

60% 60%

40% 40%

20% 20% Percentage cover per life form life per cover Percentage Percentage cover per life form life per cover Percentage 0% 0% 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Year Year

Phanerophytes Chamaephytes Hemicryptophytes Phanerophytes Chamaephytes Hemicryptophytes Cryptophytes Therophytes Liana Cryptophytes Therophytes Liana

c) d) 100% 100% 90% 80% 80% 70% 60% 60% 50% 40% 40% 30% 20% 20% 10% Percentage cover per life form life per cover Percentage Percentage cover per form life 0% 0% 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Year Year

Phanerophytes Chamaephytes Hem icryptophytes Phanerophytes Chamaephytes Hemicryptophytes Cryptophytes Therophytes Liana Cryptophytes Therophytes Liana

e) 100%

80%

60%

40%

20% Percentage cover per life form life per cover Percentage 0% 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Year

Phanerophytes Chamaephytes Hem icryptophytes Cryptophytes Therophytes Liana

Figure 10.5 Percentage vegetation cover per life form expressed as a percentage of the total vegetation cover for post-fire monitoring transects: a) site 1, b) site 2, c) site 3, d) site 4; and e) site 5.

The prominent difference between site 4 and site 5 on Land Type Fa was the high cover contribution of phanerophyte species in site 4 and their near absence in site 5 (Figure 10.5d, 10.5e). This difference was due to the farmer removing all the D. rhinocerotis plants in site 5 in 2000 in order to limit their establishment and later dominance of the area. This intervention by the farmer has limited D. rhinocerotis to a few individuals and the percentage contribution of therophyte species to the vegetation cover is higher than in site 4 (Figure 10.5d, 10.5e). Also, species richness in site 5 was more than twice as high where the D. rhinocerotis individuals were removed (Figure 10.3). When considering only the perennial species, sites 1 to 4 had between five and ten perennial species in 2008, whereas 22 perennial species were encountered in site 5. Yet, at a vegetation cover level site 1 to site 4 had a perennial vegetation cover of between 47% and 71% while site 5 only had a perennial vegetation cover of 35.5%. In 2008, vegetation cover at site 5 was 79% of which 35.5% was contributed by

165

perennial species. In contrast, in 2008 at site 4 the total vegetation cover was 54.5%, but 47% was contributed by perennial species. Whether this annual dominated vegetation cover resulting from the removal of D. rhinocerotis is more desirable than the perennial D. rhinocerotis vegetation remains to be seen.

A Principal Co-ordinate Analysis of species compositional data for ten years following the fire at each post-fire monitoring transect produced varying results (Figure 10.6). There was, however, in all cases a clear spread along the two axes separating the ten years, with year 1 and year 2 found on the right hand side of the ordination diagram and subsequent years found towards the left of the ordination space. The three Da land type transects (site 1, site 2 and site 3) produced a similar ordination diagram in that year 1, year 2 and year 3 were spread progressively from the right hand side of the ordination diagram towards the left hand side of the ordination diagram (Figure 10.6a-c). Year 4 to year 10 occurred in various sequences however, in all cases towards the left hand side of the ordination space (Figure 10.6a-c).

a) b)

c) d)

Figure 10.6 Principal Co-ordinate Analysis of species composition data for ten years (1-10) following a fire in the Roggeveld at post-fire monitoring transects: a) site 1, b) site 2, c) site 3, d) site 4 and e) site 5.

166

The two Land Type Fa transects (site 4 and site 5) were similar in that year 1 and year 2 occurred close together on the extreme right hand side of the ordination diagram (Figure 10.6d, 10.6e). The order of the subsequent years (year 3 to year 10) were chronologically spread out towards the left hand side of the ordination for site 4, but were rather haphazardly arranged for site 5 (Figure 10.6d, 10.6e). This haphazard arrangement in the ordination space of site 5 could be as a result of the removal of D. rhinocerotis seedlings in 2000.

These ordination results clearly indicated that in the first two years following fire, the species composition of all the survey transects were very different from the species composition in the subsequent years and that generally from year 3 to year 10, the species composition within a transect was more similar.

10.5 Conclusions

Vegetation began to establish within the first few months following fire in the Mountain Renosterveld vegetation of the Roggeveld. At all the post-fire monitoring sites, the vegetation cover remained higher from year 3 to year 10 than the vegetation cover found immediately following the fire (year 1 and year 2). Additionally, a single high rainfall year and three consecutive drought years had a large influence on the vegetation cover, increasing especially therophyte cover under good rainfall conditions and decreasing cover under poor rainfall conditions.

The total number of species per transect varied between 13 and 17 species within the first nine months following the fire. Species richness in most instances was highest (14 – 31 species) three years after the fire with chamaephyte species contributing the most throughout all the transects but, on the two Land Type Fa sites, therophyte species often contributed more to the total species richness of the transect.

Shannon index values generally ranged from 1.107 to 2.258 with one study transect an exception where the lowest Shannon value was 2.564 and the highest value 3.030. This transect also differed from the others in that the lowest Shannon value was found in the first year following fire and the highest value was found in the tenth year following the fire, all the other transects had the lowest Shannon values in year 9 and year 10 and the highest Shannon values in year 1 to year 3.

Temporal changes in life form composition were found at a species and vegetation cover level and depended on land type. Generally, at a species level, the phanerophyte and chamaephyte species contribution was higher on Land Type Da than on Land Type Fa, while therophyte species percentage contributions were higher on Land Type Fa than Land Type Da. Between the five post-fire monitoring transects, phanerophyte vegetation cover

167

contribution was highest on two transects on Land Type Da and one transect on Land Type Fa due to the dominance of D. rhinocerotis. Chamaephyte vegetation cover was highest on Land Type Da while therophyte or phanerophyte vegetation cover was highest on Land Type Fa.

A Principal Co-ordinate Analysis of species data for each survey plot, in all cases, placed year 1 and year 2 on the right hand side of the ordination diagram with the subsequent years spread along both axes towards the left hand side of the ordination space. This indicates the large difference in species composition between the first two years and subsequent years following fire.

These data seems to support Egler’s (1954) ‘initial floristic composition’ model which states that all, or the majority of, species present during the succession are in place at the beginning of the recovery phase and re-establishment of the initial plant community is a rapid phenomenon (Capitanio & Carcaillet 2008).

10.6 Acknowledgements

The Department of Tourism, Environment and Conservation (Northern Cape) are thanked for the initial permission and financial support from 1999 to 2003 to conduct the research. The authors would also like to thank the Critical Ecosystem Partnership Fund (CEPF) through the Succulent Karoo Ecosystem Plan/Program (SKEP) initiative for supporting the project from 2003 to 2007. The Critical Ecosystem Partnership Fund is a joint initiative of Conservation International, the Global Environmental Facility, the Government of Japan, the MacArthur Foundation and the World Bank. A fundamental goal is to ensure civil society is engaged in biodiversity conservation. The various people who assisted with the field work over the years are gratefully acknowledged. This research was supported by the National Research Foundation under grant number 61277.

10.7 References

BOND, W.J. AND VAN WILGEN, B.W. 1996. Fire and plants. Chapman & Hall, London. CAPITANIO, R. AND CARCAILLET, C. 2008. Post-fire Mediterranean vegetation dynamics and diversity: A discussion of succession models. Forest Ecology and Management 255, 431-439. CARSON, W.P. AND PICKETT, S.T.A. 1990. Role of resources and disturbance in the organization of an old-field plant community. Ecology 71, 226-238. COUNCIL FOR GEOSCIENCE 1989. Geological map 3120 Williston. Council for Geoscience, Pretoria.

168

COWLING, R.M., RICHARDSON, D.M. AND MUSTARD, P.J. 1997. Fynbos. In: R.M. Cowling, D.M. Richardson and S.M. Pierce (Eds). Vegetation of southern Africa, pp. 99-129. Cambridge University Press, Cambridge. CULLINAN, P. 2003. Robert Jacob Gordon 1743 – 1795. The man and his travels at the Cape. http:// web.uct.ac.za/depts/age/people/Gordon/frameset.htm DU TOIT, P.C.V. 1998a. Research note: Grazing-index method procedures of vegetation surveys. African Journal of Range and Forage Science 14, 107-110. DU TOIT, P.C.V. 1998b. Description of a method for assessing veld condition in the Karoo. African Journal of Range and Forage Science 14, 90-93. EGLER, F.E. 1954. Vegetation science concepts. I. Initial floristic composition – a factor in old-field vegetation development. Vegetatio 4, 412-418. FRANCIS, M.L., FEY, M.V., PRINSLOO, H.P., ELLIS, F., MILLS, A.J. AND MEDINSKI, T.V. 2007. Soils of Namaqualand: Compensations for aridity. Journal of Arid Environments 70, 588-603. GOOD, R. 1947. The geography of flowering plants. Longmans, Green & Co., New York. GUO, Q. 2001. Early post-fire succession in California chaparral: changes in diversity, density, cover and biomass. Ecological research 16, 471-485. HANES, T.L. 1971. Succession after fire in the chaparral of southern California. Ecological Monographs 41, 27-52. KEELEY, S.C., KEELEY, J.E., HUTCHINSON, S.M. AND JOHNSON, A.W. 1981. Post-fire succession of herbaceous flora in southern California chaparral. Ecology 62, 1608- 1621. KRUGER, F.J., AND BIGALKE, R.C. 1984. Fire in fynbos. In: P. De V. Booysen and N.M. Tainton (Eds). Ecological effects of fire in South African Ecosystems, pp. 69-114. Springer-Verlag, Berlin. LAVOREL, S., LEPART, J., DEBUSSCHE, M., LEBRETON, J-D. AND BEFFY, J-L. 1994. Small scale disturbances and the maintenance of species diversity in Mediterranean old fields. OIKOS 70, 455-473. LLORET, F. AND VILÁ, M. 2003. Diversity patterns of plant functional types in relation to fire regime and previous land use in Mediterranean woodlands. Journal of Vegetation Science 14, 387-398. MARLOTH, R. 1908. Das Kapland, insonderheit das Reich der Kapflora, das Waldgebiet und die Karoo, pflanzengeografisch dargestellt. Wissenschaftliche Ergebnisse der Deutscher Tiefsee-Expedition ‘Waldivia’, 1898 – 1899. 2, T. 3, Fischer, Jena. McCUNE, B. AND GRACE, J.B. 2002. Analysis of Ecological Communities. MjM Stoftware Design, Gleneden Beach, Oregon. McDOWELL, C.R. AND MOLL, E.J. 1992. The influence of agriculture on the decline of West Coast Renosterveld, south-western Cape, South Africa. Journal of Environmental Management 35, 173-192.

169

McINTYRE, S., LAVOREL, S., LANDSBERG, J. AND FORBES, T.D.A. 1999. Disturbance response in vegetation – towards a global perspective on functional traits. Journal of Vegetation Science 10, 621-630. McINTYRE, S., LAVOREL, S. AND TREMONT, R.M. 1995. Plant life-history attributes: their relationship to disturbance response in herbaceous vegetation. Journal of Ecology 83, 31-44. MOLL, E.J., CAMPBELL, B.M., COWLING, R.M., BOSSI, L., JARMAN, M.L. AND BOUCHER, C. 1984. A description of major vegetation categories in and adjacent to the Fynbos Biome. South African National Scientific Programmes Report 83, 1-29. MUCINA, L. RUTHERFORD, M.C. AND POWRIE, L.W. (Eds) 2005. Vegetation map of South Africa, Lesotho and Swaziland, 1 : 1 000 000 scale sheet maps. South African National Biodiversity Institute, Pretoria. MUELLER-DOMBOIS, D. AND ELLENBERG, H. (Eds) 1974. Aims and methods of vegetation ecology. Wiley, New York. NOVELLIE, P. AND STRYDOM, G. 1987. Monitoring the response of vegetation to use by large herbivores: an assessment of some techniques. South African Journal of Wildlife Research 17, 109-117. PAUSAS, J.G. 1999. Response of plant functional types to changes in the fire regime in Mediterranean ecosystems: A simulation approach. Journal of Vegetation Science 10, 717-711. PODANI, J. 2001. SYN-TAX 2000 Computer programs for data analysis in ecology and systematics. Scientia publishing, Budapest. POTTS, S.G., VULLIAMY, B., DAFNI, A., NE’EMAN, G., O’TOOLE, C., ROBERTS, S. AND WILLMER, P. 2003. Response of plant-pollinator communities to fire: changes in diversity, abundance and floral reward structure. OIKOS 101, 103-112. RAUNKIAER, C. 1934. The life forms of plants and statistical plant geography. Oxford University Press, Oxford. REBELO, A.G., BOUCHER, C., HELME, N., MUCINA, L. AND RUTHERFORD, M.C. 2006. Fynbos Biome. In: L. Mucina and M.C. Rutherford (Eds). The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19, pp. 52-219. South African National Biodiversity Institute, Pretoria. ROUX, P.W. 1963. The descending point method of vegetation survey. A point-sampling method for the measurement of semi-open grasslands and Karoo vegetation in South Africa. South African Journal of Agricultural Science 6, 273-288. RUBIDGE, B.S. AND HANCOX, P.J. 1999. The Karoo – a palaeontological wonderland. In: M.J. Viljoen and W.U. Reimold (Eds). An introduction to South Africa’s geological and mining heritage, pp. 83-91. Published by the Geological Society of South Africa and Mintek.

170

SCHWILK, D.W., KEELEY, J.E. AND BOND, W.J. 1997. The intermediate disturbance hypothesis does not explain fire and diversity pattern in fynbos. Plant Ecology 132, 77-84. SHEARING, D. 1997. Karoo: South African Wildflower Guide 6. National Book Printers, Drukkery street, Goodwood, Western Cape. TRABAUD, L. AND LEPART, J. 1980. Diversity and stability in garrigue ecosystems after fire. Vegetatio 43, 49-57. VAN DER MERWE, H., VAN ROOYEN, M.W. AND VAN ROOYEN, N. 2008. Vegetation of the Hantam-Tanqua-Roggeveld subregion, South Africa. Part 1. Fynbos Biome related vegetation. Koedoe 50, 61-71. VORSTER, M. 1982. The development of the ecological index method for assessing veld condition in the Karoo. Proceedings of the Grassland Society of Southern Africa 17, 84-89. WEATHER BUREAU 1998. Climate of South Africa. Climate statistics up to 1990. WB 42. Government Printer, Pretoria.

171

Chapter 11

General discussion and synthesis

11.1 Introduction

This chapter provides a discussion and synthesis on the findings of the study and aims to provide insight on the Succulent Karoo and Fynbos Biome affinities of the Hantam-Tanqua- Roggeveld subregion.

11.2 Vegetation mapping of the Hantam-Tanqua-Roggeveld subregion

Although the entire Hantam-Tanqua-Roggeveld subregion was included in the SKEP (Succulent Karoo Ecosystem Plan) planning domain, such a delineation is not universally accepted. According to Low and Rebelo (1996) and Mucina et al. (2005) the vegetation of the subregion falls within both the Succulent Karoo and Fynbos Biomes. The Two Way Indicator Species Analysis (TWINSPAN) on the entire data set of 390 relevés resulted in two phytosociological tables supporting the clear difference in species composition of the Fynbos Biome related vegetation and the Succulent Karoo Biome related vegetation. The first data set of 107 relevés, characterised the vegetation of the predominantly Mountain Renosterveld with Fynbos Biome affinities, while the second data set of 283 relevés, characterised the Succulent Karoo Biome vegetation. Within the Succulent Karoo Biome vegetation there was another split at a lower level separating the Tanqua Karoo and Loeriesfontein plots (association 7 and association 8) from associations 4, 5 and 6. These two large vegetation groups within the Succulent Karoo vegetation were referred to as the Tanqua Karoo and Winter Rainfall Karoo respectively.

The phytosociological tables were used to describe the eight major associations and 25 subassociations found in the Hantam-Tanqua-Roggeveld subregion in terms of their species composition, environmental parameters and relationship to one another as well as to map their geographical distribution. Various map outputs were generated including: a) a vegetation map of the Fynbos Biome related vegetation (Van der Merwe et al. 2008a), b) a vegetation map of the Succulent Karoo Biome related vegetation (Van der Merwe et al. 2008b), c) a detailed vegetation map of the entire subregion including vegetation mosaics (Figure 11.1), and d) a simplified vegetation map of the entire subregion (Chapter 6, Figure 6.1).

172

Figure 11.1 A detailed vegetation map of the entire Hantam-Tanqua-Roggeveld study area showing both the Fynbos Biome related and Succulent Karoo Biome related vegetation.

173

Figure 11.1 (continued)

Previous maps published by Acocks (1953, 1988), Low and Rebelo (1996) and Mucina et al. (2005) show interesting similarities and some dissimilarities to the map generated in this study (Figure 11.1). Since the scale of these maps and the current map differ and the reasons for compiling the maps also differ, only generalisations can be made when comparing the maps. The current study found the Sutherland area, a region classified by Acocks (1953, 1988) as Western Mountain Karoo and Low and Rebelo (1996) as Upland Succulent Karoo, was more closely related to the Mountain Renosterveld vegetation of the Fynbos Biome than

174

to the Succulent Karoo vegetation. Mucina et al. (2005) identifies this area as Roggeveld Karoo (SKt3) whereas this study incorporated the area into the Mountain Renosterveld vegetation of plant association 1.

The Succulent Karoo Biome related vegetation included two of Acocks’s (1953, 1988) veld types. Associations 4 (Escarpment Karoo), 5 (Hantam Karoo) and 6 (Roggeveld Karoo) combined form part of the Western Mountain Karoo (Veld Type 28) which is incorporated into Low and Rebelo’s (1996) Upland Succulent Karoo (Unit 56) while associations 7 (Tanqua and Loeriesfontein Karoo) and 8 (Central Tanqua Grassy Plains) are both included into Acocks’s (1953, 1988) Succulent Karoo (Veld Type 31) which relates to Low and Rebelo’s (1996) Lowland Succulent Karoo (Unit 57). Approximately ten vegetation units were mapped by Mucina et al. (2005) within the Succulent Karoo Biome related vegetation as identified by this study. There was a good agreement between the Tanqua Escarpment Shrubland (SKv4) of Mucina et al. (2005) and association 4 (Escarpment Karoo). Generally, the delineation of the Hantam Karoo (association 5), Roggeveld Karoo (association 6) and Tanqua Karoo (association 7) in this study was more restricted than that of the Hantam Karoo (SKt2), Roggeveld Karoo (SKt3) and Tanqua Karoo (SKv5) of Mucina et al. (2005). The maps of Acocks (1953, 1988) and Low and Rebelo (1996) differ appreciably in the van Rhynsdorp/Doorn River region, however, the Mucina et al. (2005) map was more closely related to the map presented in this thesis. This region is situated in the transition between the Succulent Karoo Biome (Tanqua Karoo and the van Rhynsdorp Succulent Karoo) and the Fynbos Biome, and requires additional study to determine the boundaries of each different vegetation type.

Grazing and cropping are the main land uses in the Mountain Renosterveld vegetation of the Fynbos Biome related vegetation, while grazing is the main land use in the Succulent Karoo Biome related vegetation since this area receives less rainfall for growing crops. A major threat to the vegetation of the study area, identified by the farming community, Northern Cape Department of Agriculture and through personal observations during field surveys was the less than ideal farming practises due to a lack of infrastructure, especially fencing, for optimal farm management. Although damage happens fast, recovery in the Karoo is very slow, as it depends mainly upon unpredictable rainfall events (Esler et al. 2006). Sustainable farm management planning is critical for ensuring a productive, profitable future in the region (Esler et al. 2002, Millennium Ecosystem Assessment 2005).

The second threat is that of invasive alien species, especially Prosopis species. Prosopis glandulosa was introduced to the Karoo to aid in fodder production for small stock in the drier months of the year when little natural vegetation is available (Zimmerman 1991). Unfortunately, more than one species of Prosopis was introduced resulting in hybridisation between the species. These hybrids have invaded large tracts of land and are a serious threat

175

especially in the Hantam and Tanqua Karoo areas. Other invasive alien species include Nerium oleander and Nicotiana glauca which are also found mostly in the Hantam and Tanqua Karoo areas and are usually restricted to drainage lines. Invasive annual grasses such as Hordeum murinum and Bromus pectinatus, commonly occurring in the Hantam and Roggeveld areas, have a naturalised status.

Large scale transformation of natural vegetation, especially along drainage lines, into flood irrigated lands has taken place in the past. Few of these lands, especially in the Tanqua Karoo, are still utilised for cropping and now lie barren with little vegetation cover to combat erosion or the lands are infested by invasive species, usually Prosopis species.

Large tracts of the Mountain Renosterveld vegetation of the Roggeveld are dominated by the unpalatable renosterbos, Dicerothamnus rhinocerotis, which is considered an encroacher by most farmers with its dominance being blamed on centuries of incorrect farming practices in the region (Marloth 1908, Acocks 1988, Cullinan 2003). Overgrazing and incorrect burning practices are thought to have had a substantial effect on the grassy component of the vegetation, reducing it to the current state (Marloth 1908, Acocks 1988, Cullinan 2003), with grazing pressure given as a reason for the near extinction of the indigenous rye grass (Secale strictum subsp. africanum) (Van Wyk & Smith 2001, Rebelo et al. 2006).

Formal conservation of the Hantam-Tanqua-Roggeveld subregion is limited. Two local municipal reserves, the Akkerendam Nature Reserve (230 ha) and the Nieuwoudtville Wildflower Reserve (115 ha) are located within the study area. Akkerendam Nature Reserve protects a part of the plateau and slope of the Hantam Mountain, while the Nieuwoudtville Wildflower Reserve protects a combination of the Fynbos and Succulent Karoo Biome related vegetation on the western extreme of the study area. The Tankwa Karoo National Park has expanded its boundaries considerably over the last few years protecting large parts of the Tanqua Karoo and small areas of the Roggeveld escarpment and Roggeveld plateau, however, large tracts of land especially in the Tanqua Karoo are highly degraded. The latest addition (12 December 2008) to the protected area network is the 6200 ha of the Hantam National Botanical Gardens situated just south and east of Nieuwoudtville. This area protects Succulent Karoo and Fynbos Biome related vegetation. Private nature reserves and conservancies are few and are generally found within the Tanqua Karoo. Most of them are located at the southern end of the Tanqua Basin, close to the town of Ceres.

Formal conservation on a plant association level in the Hantam-Tanqua-Roggeveld subregion is highly inadequate. Large areas of association 7 (Tanqua Karoo) and association 8 (Central Tanqua Grassy Plains) are protected within the borders of the Tankwa Karoo National Park, with small areas of association 4 (Escarpment Karoo) and association 2 (Mountain Renosterveld) being conserved along the Roggeveld escarpment into the Mountain

176

Renosterveld vegetation of the Roggeveld plateau. An extremely small area of association 3 (Mountain Renosterveld) is conserved within the Akkerendam Nature Reserve since it forms part of the town of Calvinia’s water catchment area. Association 2, located as an outlier mosaic vegetation unit close to Nieuwoudtville, and association 5 have recently been conserved with the proclamation of the Hantam National Botanical Gardens. The Roggeveld Karoo (association 6) is not formally conserved in any way. Currently, the largest threats to the unprotected associations within the Hantam-Tanqua-Roggeveld subregion are incorrect management of grazing land as well as alien invasive trees, especially Prosopis species. This reinforces the general view that livestock grazing has been identified as a major threat to biodiversity in the Succulent Karoo (Mucina et al. 2006).

11.3 Plant diversity studies

11.3.1 Species-area relationships

Analysis of the data collected using 40 Whittaker plots scattered throughout the Hantam, Tanqua and Roggeveld areas revealed an array of species-area curves. Significance levels of the untransformed linear function, exponential function and power function varied greatly, however, better r–values and p-values were obtained for the exponential and power functions with the exponential function performing marginally better.

A comparison of plant associations along a west to east transect from the Tanqua Karoo across the escarpment into the Roggeveld crossing five different associations, generally revealed significant differences in slope values between the associations, except for the Dicerothamnus rhinocerotis Mountain Renosterveld which did not differ significantly from associations bordering it on either side. Generally, the Tanqua Karoo had a low vegetation cover, low species richness values and species-area curves with shallow slopes. The Roggeveld Escarpment Karoo vegetation, which falls within the Winter Rainfall Karoo vegetation group, is transitional between the Tanqua Karoo and the Mountain Renosterveld of the Roggeveld Mountains and produced the highest species richness values and steepest species-area curves.

11.3.2 Diversity parameters

The species richness in the 1000 m² (0.1 ha) plots ranged from nine to 100 species per sampled 1000 m² among the 40 Whittaker plots. Mean species richness was significantly higher in the Mountain Renosterveld than in the Winter Rainfall Karoo, which in turn was significantly higher than in the Tanqua Karoo. However, the low mean species richness values for the Tanqua Karoo compared poorly to Succulent Karoo values cited in the literature (Cowling et al. 1989, Cowling & Hilton-Taylor 1994, Anderson & Hoffman 2007). Species

177

richness values obtained for the Fynbos Biome, and in particular for the renosterveld vegetation type within this biome, compared well with Mountain Renosterveld vegetation values found in this study (Cowling et al. 1992, Cowling & Holmes 1992, Richardson 1995, Procheş et al. 2003, Kongor 2009). However, the year in which the survey was conducted was a poor rainfall year, which would have had a negative influence on the annual and geophyte species that make up a large component of the Succulent Karoo and Fynbos diversity and the values presented in this study probably underestimated the potential diversity in the region. Evenness, Shannon and Simpson indices were found not to differ significantly between the Mountain Renosterveld and Winter Rainfall Karoo but these values were significantly higher than for the Tanqua Karoo.

Comparisons of species richness at seven different plot sizes (1 m², 5 m², 10 m², 20 m², 50 m², 100 m², 1000 m²) produced significant differences between the Winter Rainfall Karoo and Tanqua Karoo and between the Mountain Renosterveld and Tanqua Karoo at each plot size. A significant difference between the Winter Rainfall Karoo and Mountain Renosterveld was only obtained at the 1000 m² plot size indicating the importance of surveying relatively large plots that enable researchers to pick up differences in species richness between different regions not evident at smaller plot sizes.

This relationship between the broad vegetation groups was visually summarised and illustrated when a Principal Co-ordinate Analysis was conducted with the diversity data for the seven plot sizes producing three ordination clusters. The Tanqua Karoo cluster was characterised by a low plant cover, shallow slopes and low y-intercept values. At the other extreme a cluster of a selection of high-lying Mountain Renosterveld plots was produced, generally possessing steep species-area curve slopes and high y-intercept values, and representing a ‘true’ Mountain Renosterveld vegetation. The central cluster consisted of Winter Rainfall Karoo plots as well as several transitional Mountain Renosterveld plots indicating that the Mountain Renosterveld vegetation of the Hantam-Tanqua-Roggeveld consists of ‘true’ Mountain Renosterveld as well as a transitional form between the Mountain Renosterveld and Winter Rainfall Karoo vegetation groupings.

Despite changing fashions and preoccupations, diversity has remained a central theme to ecology (Magurran 1988). In spite of various criticisms, diversity indices have sparked renewed interest in handling problems associated with the conservation of natural heritage or the changes in global ecology (Mouillot & Leprêtre 1999). Diversity indices determined in this study indicate that ironically, the two most species poor associations (association 7 and association 8), with the lowest diversity values, are well conserved within the Tankwa Karoo National Park and that two associations (association 2 and association 4) with high levels of diversity are included to a very limited extent. The importance of conserving mountain communities, such as associations 2 and 4 in this study, has been emphasised (Burke et al.

178

2003) since altitude is an important contributor to explaining species diversity (Körner 2000). Altitudinal gradients have also been recognised as an important buffer in the event of climate change (Bond & Richardson 1990) because mountain habitats could provide refugia for species during climate change (Midgley et al. 2000). This further emphasises the importance of conserving especially association 4 (Escarpment Karoo) and association 3 (Mountain Renosterveld). The Tankwa Karoo National Park boundaries could therefore be extended to include a larger area of association 2 and association 4 and should even, potentially, be extended to include areas within association 1 and association 6, two additional associations with relatively high diversity values.

Diversity measures should, however, only be used as an indicator of which areas should be conserved but these potential areas of conservation importance must be considered together with information on the floristic relationships in the region in order to ensure a functioning ecosystem is conserved. Low diversity areas should also be considered and conserved since in many cases species or species combinations in these areas are unique to the region.

11.3.3 Life form spectra

Comparisons of the various life forms, between the associations and vegetation groups, produced different results when studied at a species or vegetation cover level. Generally, the eight plant associations were dominated by chamaephyte, cryptophyte (geophyte) and therophyte (annual) species.

At the species level, phanerophyte contributions were significantly higher in the Mountain Renosterveld than both the Winter Rainfall Karoo and Tanqua Karoo. Hemicryptophyte contributions were significantly higher in the Tanqua Karoo than the Winter Rainfall Karoo, while cryptophyte and therophyte contributions were significantly lower in the Tanqua Karoo than the Mountain Renosterveld. Therophyte contributions also differed significantly between the Winter Rainfall Karoo and the Tanqua Karoo. No significant differences were found between the vegetation groups for chamaephyte, liana and parasite species. At a vegetation cover level, the phanerophyte cover contribution was found to be significantly higher in the Mountain Renosterveld than the Winter Rainfall Karoo and the Tanqua Karoo. Chamaephyte contributions were significantly higher in the Mountain Renosterveld and Winter Rainfall Karoo groups than the Tanqua Karoo vegetation group. No significant differences were found for hemicryptophytes, cryptophytes, therophytes, lianas or parasites.

It was found that, generally, the succulent species were chamaephyte, hemicryptophyte and therophyte species. The percentage contribution of the succulent species was low in Mountain Renosterveld, intermediate in Winter Rainfall Karoo and highest in the Tanqua Karoo while, succulent vegetation cover was generally lowest for the Mountain Renosterveld,

179

with higher cover values found for the Winter Rainfall Karoo and Tanqua Karoo. This confirms the inclusion of the Winter Rainfall Karoo and Tanqua Karoo into the Succulent Karoo Biome. But, the higher than expected succulent cover for one strongly Mountain Renosterveld (Fynbos Biome) association (association 1), indicates its transitional nature between the Succulent Karoo and Fynbos Biomes. Another Succulent Karoo site, Goegap Nature Reserve, produced similar patterns to this study, yet, these patterns were found to be very different from the pattern encountered in the winter rainfall Mojave Desert. In this study, life form spectra for Mountain Renosterveld did not compare well with a Fynbos Biome site where therophyte contributions were much lower and phanerophyte contributions much higher. The low degree of succulence in the Winter Rainfall Karoo association, association 6, could be due to its transitional nature since it is situated between the Mountain Renosterveld of the Fynbos Biome and the Nama Karoo Biome and is probably more closely related to the Nama Karoo Biome vegetation.

11.4 Life forms and species diversity on abandoned croplands in the Roggeveld

The Roggeveld, located within the Fynbos Biome (Low & Rebelo 1996, Mucina et al. 2005, Van der Merwe et al 2008a), has a higher rainfall than the surrounding areas and has been used to cultivate wheat and fodder crops for hundreds of years. Many of these lands now lie barren as a result of an increase in production costs.

The most abundant life forms on abandoned croplands of all ages were chamaephytes and therophytes. Therophyte species were most abundant on young abandoned croplands and decreased in number from 10-years of age to values similar to those encountered in the natural vegetation. Phanerophyte species did not differ much across the age range of the surveyed plots and numbers remained low. Chamaephyte, cryptophyte and hemicryptophyte species increased in number with an increase in age of abandoned croplands with the highest values found for the natural vegetation. Chamaephytes made an overwhelming contribution to the relative cover on abandoned croplands while, phanerophytes, cryptophytes and therophytes contributed significantly less. Hemicryptophyte and liana cover contributions were negligible.

In all instances, slope or intercept values of the exponential function species-area curves differed significantly between the abandoned croplands and the natural vegetation. An increase in age of abandoned cropland leads to an increase in species richness, with species richness values highest for the natural vegetation. Evenness, Shannon and Simpson indices did, however, not reflect the same increase. Only 15 species contributed to the highest cover on the abandoned croplands with eight of these also contributing to the high cover of the natural vegetation.

180

A regression of species richness against age of abandoned croplands predicted that an abandoned cropland of about 33-years in age should be as species rich as the natural vegetation, however, a Principal Co-ordinate Analysis of the floristic data indicated that floristically there is an extremely large gap between the abandoned croplands and the natural vegetation. The recovery rate of the various life forms varies and an important component of the flora, the geophytes (cryptophytes), still remain greatly underrepresented after 20-years of abandonment.

The pattern of recovery on abandoned croplands in the Mountain Renosterveld of the Roggeveld seems to differ from those of the West Coast Renosterveld of the Cape Floristic Region (CFR) where remnant renosterveld vegetation and abandoned croplands were studied on the Elandsberg Private Nature Reserve. These studies indicated the apparent slow return of indigenous renosterveld vegetation on abandoned croplands (Midoko-Iponga et al. 2005). Annual weedy alien grasses arrest the whole recovery process in the West Coast Renosterveld, however, this was not the case in this study in the Mountain Renosterveld of the Roggeveld. Current restoration efforts in the West Coast Renosterveld aim to reduce the cover of the introduced grasses while, at the same time, maintaining or even increasing species richness and diversity of indigenous target species (Krug & Krug 2007). The recovery in the Roggeveld probably occurs under harsher environmental conditions yet, species richness steadily increases and evenness values, Shannon and Simpson indices of diversity on abandoned croplands from approximately 10 years after abandonment are similar to those of the natural vegetation.

The recovery of vegetation on abandoned croplands in the Roggeveld could be left to continue at its natural pace with no management interventions. This will quickly lead to some form of perennial vegetation cover to primarily combat soil erosion, which is a major problem on abandoned croplands due to the mountainous topography of the area and occasional summer thunderstorms that produce large volumes of runoff. This is, obviously, the cheapest option for the land owner. A second, more costly option, would be to establish perennial shrubs and grasses that are of a higher forage value than those that initially colonise and later dominate abandoned croplands. However, it is suggested that this be initiated on abandoned croplands that are in the initial stages of vegetation recovery. The pioneering species will be able to provide some sort of protection against the harsh environmental conditions on an abandoned cropland by providing windbreaks, shade, humus, breakage of the soil crust or collection of water runoff. These advantages to a newly sown seed or planted seedling should outweigh the negative influences of competition in the initially hard and barren soil. Old abandoned croplands with high densities of Dicerothamnus rhinocerotis will have to be thinned before attempting to establish other vegetation since these dense stands will out- compete other vegetation.

181

11.5 Vegetation trends following fire in the Roggeveld

Establishment of vegetation began within nine months following a fire in the Roggeveld (Mountain Renosterveld vegetation) with only the first two years after the fire being marked by a low vegetation cover. The study found different vegetation trends on the two land types present at the sites (Land Type Da and Land Type Fa). Species richness was highest within three years after the fire with chamaephyte species contributions the most except on Land Type Fa where therophyte species often contributed more. Generally, all except one plot, had the lowest Shannon values in year 9 and year 10 and the highest Shannon values in year 1 to year 3.

Phanerophyte and chamaephyte species contributions were generally higher on Land Type Da than Land Type Fa, while therophyte species contributions were higher on Land Type Fa than Land Type Da. On a vegetation cover level, phanerophyte cover was highest on the two Land Type Da plots and one Land Type Fa plot due to the dominance of Dicerothamnus rhinocerotis. Chamaephyte vegetation cover was highest on Land Type Da and therophyte vegetation cover highest on Land Type Fa. A large difference in species composition between the first two years following fire and the subsequent years was illustrated using a Principal Co-ordinate Analysis.

Quick community recovery after fire in Mediterranean-type ecosystems has been reported in various studies with species richness highest soon after fire (Keeley et al. 1981, Bond & Van Wilgen 1996, Schwilk et al. 1997, Guo 2001, Potts et al. 2001, Capitanio & Carcaillet 2008). In all instances species richness was reported to decline in the following years. When studying the species composition after a fire it becomes evident that all, or the majority of, species present during the succession are in place at the beginning of the recovery phase. These data seem to support Egler’s (1954) ‘initial floristic composition’ model.

The removal of D. rhinocerotis by one of the farmers in order to limit their establishment and later dominance of the area produced interesting results, namely: 1) therophyte species percentage contribution to the vegetation cover was higher than in a comparable site; 2) species richness was more than twice as high where D. rhinocerotis was removed compared to a similar site; 3) perennial vegetation cover contributed to 35.5% of the total vegetation cover of 79% while at the comparable site, perennial vegetation cover contributed to 54.5% of the total vegetation cover of 47%; 4) the biomass of the site which is primarily composed of small annual vegetation is probably much lower than in the comparable site, however, the biomass at the second site is comprised mainly of D. rhinocerotis, which is of limited value to the farmer since it is not a palatable species; and 5) the fact that D. rhinocerotis did not establish in large numbers after being initially removed by the farmer seems to indicate that this species is unable to compete with other species once the other species have established,

182

even if comprised of predominantly annual species. This has important management implications, indicating that if the farmer can establish plant species of forage value D. rhinocerotis will not be able to establish and dominate at a location.

11.6 Succulent Karoo and Fynbos Biome affinities: a synthesis of results

A synthesis of all the data provides a clearer understanding of the broad vegetation groups and how they are related to one another and provide some clarity on how they are affiliated to the Succulent Karoo and Fynbos Biomes. A summary table is presented as Table 11.1.

11.6.1 Phytosociology

The Mountain Renosterveld vegetation group consists of three associations (1. Rosenia oppositifolia Mountain Renosterveld, 2. Dicerothamnus rhinocerotis Mountain Renosterveld, 3. Passerina truncata Mountain Renosterveld), similarly the Winter Rainfall Karoo consists of three associations (4. Pteronia glauca Escarpment Karoo, 5. Eriocephalus purpureus Hantam Karoo, 6. Pteronia glomerata Roggeveld Karoo) and the Tanqua Karoo of two associations (7. Aridaria noctiflora Tanqua and Loeriesfontein Karoo, 8. Stipagrostis obtusa Central Tanqua Grassy Plains).

11.6.2 Environmental parameters

The mean annual precipitation for the Mountain Renosterveld vegetation ranges between 200 mm to 400 mm per year with a coefficient of variation of between 25% and 40%. Winter Rainfall Karoo has a slightly lower mean annual precipitation and a higher coefficient of variation while the Tanqua Karoo has the same coefficient of variation for annual precipitation as the Winter Rainfall Karoo but a much lower mean annual precipitation. Mean maximum and minimum temperatures for the warmest and coldest months of the year are cooler for Mountain Renosterveld than for Winter Rainfall Karoo or Tanqua Karoo, with Winter Rainfall Karoo temperatures intermediate and Tanqua Karoo temperatures the highest. Snow is six times more common in the Mountain Renosterveld than Winter Rainfall Karoo, with no snowfall occurring in the Tanqua Karoo.

The concept of the heat unit (or degree day) revolves around the development of a plant or organism being dependent upon the total heat to which it was subjected during its lifetime, or else during a certain developmental stage (Schulze 1997). Heat units are expressed as degree days, where these are an accumulation of mean temperatures above a certain lower threshold value (below which active development is considered not to take place) and below an upper limit (above which growth is considered to remain static or even decline) over a period of time (Schulze 1997). The degree days above 10 °C for April to September are the

183

lowest in the Mountain Renosterveld, some of the lowest values for the entire South Africa, intermediate for the Winter Rainfall Karoo and highest for the Tanqua Karoo.

Some plants, which have a dormant season during winter, may have to accumulate chill units with temperatures below a threshold in order to stimulate growth, develop leaves, flowers or set fruit (Steyn et al. 1996, Schulze 1997). The required amount of chilling for completion of the rest period varies between species, cultivars and different locations. Chill units have been derived from models using threshold temperatures. The accumulated positive chill units from May to September for Mountain Renosterveld are some of the highest values for the entire South Africa, these values are much lower in the Winter Rainfall Karoo and even lower for the Tanqua Karoo.

Soils underlying Mountain Renosterveld are shallow stony lithosols and duplex soils in the occasional lowlands. The soils of the Winter Rainfall Karoo are shallow lithosols and duplex soils but where dolerite occurs soils are red structured and red vertic clays. Tanqua Karoo soils are shallow lithosols that often include a desert pavement and deep unconsolidated deposits in the alluvial parts. Generally, Mountain Renosterveld is found at high altitudes, Winter Rainfall Karoo at lower altitudes and Tanqua Karoo vegetation at the lowest altitudes. Fire is an important disturbance in Mountain Renosterveld vegetation and not in Winter Rainfall Karoo and Tanqua Karoo vegetation.

11.6.3 Diversity parameters

All the diversity parameters and species area relationships could separate the Tanqua Karoo from the Mountain Renosterveld and the Winter Rainfall Karoo, however only the species richness at 1000 m2 could show a significant difference between the Mountain Renosterveld and the Winter Rainfall Karoo. The ordination diagram partially separated the Mountain Renosterveld from the Winter Rainfall Karoo.

11.6.4 Life forms

When analysing the classic life form spectra at a species level, phanerophyte contributions for the Mountain Renosterveld were significantly higher than for the Winter Rainfall Karoo and Tanqua Karoo. No significant difference was found between chamaephytes, lianas or parasites at the species level. Contributions by hemicryptophyte were significantly higher for the Tanqua Karoo than the Winter Rainfall Karoo whereas those of cryptophytes were significantly higher in the Mountain Renosterveld than in the Tanqua Karoo. Significantly higher therophyte contributions were found in Mountain Renosterveld vegetation compared to the Tanqua Karoo vegetation. Upon extraction of succulent species, Mountain Renosterveld

184

vegetation held significantly lower succulent species contributions than Winter Rainfall Karoo or Tanqua Karoo vegetation.

At a cover level, phanerophyte cover was significantly higher in Mountain Renosterveld than Tanqua Karoo or Winter Rainfall Karoo, with Winter Rainfall Karoo values intermediate and significantly different from the low Tanqua Karoo values. Chamaephyte cover was significantly higher in the Mountain Renosterveld and Winter Rainfall Karoo than in the Tanqua Karoo. No significant differences in cover were found between the vegetation groups of the hemicryptophyte, cryptophyte, therophyte, liana and parasite life forms. Succulent species cover was significantly higher for the Winter Rainfall Karoo and Tanqua Karoo than for the Mountain Renosterveld.

The above-mentioned results clearly indicate that in the majority of cases the vegetation of the Tanqua Karoo is different from the vegetation of the Mountain Renosterveld and Winter Rainfall Karoo. These results also indicate a close relationship between the Mountain Renosterveld and Winter Rainfall Karoo vegetation in that clear differences between them are few, yet, differences were found on a species richness level and especially at the 1000 m² plot size.

In the early 1900s, Marloth (1908) and Diels (1909) both recognised the Tanqua Karoo as an area where vegetation was sparse and characterised by succulents. Marloth (1908) provided a detailed description of the Dicerothamnus rhinocerotis, renosterbos, dominated Roggeveld while Diels (1909) suggested that the Hantam Mountain, with its Mountain Renosterveld, was an outlier of the Cape flora but also linked it to Marloth’s (1909) description of the Roggeveld. Weimarck (1941) treated the Hantam-Roggeveld as a subcentre of his North-Western Centre however, hesitantly stated that the subcentre constituted the last outlier of the Cape element in the inner parts of the western South Africa.

Acocks’s (1988) map of the vegetation in A.D. 1950 places the Hantam (Winter Rainfall Karoo) and Roggeveld (Mountain Renosterveld) within the Karoo veld type and the Tanqua Karoo in the Succulent Karoo and Desert veld types. Rutherford and Westfall (1994), however, place both the Hantam and Tanqua Karoo within the Succulent Karoo Biome but agree on the Roggeveld as part of the Nama Karoo Biome. They state that the Roggeveld shows some floristic affinities to the Fynbos Biome, but the life form combination precludes it from being considered as part of the Fynbos Biome. Low and Rebelo (1996) also place the Hantam and Tanqua Karoo in the Succulent Karoo Biome however include the Roggeveld within their renosterveld group of the Fynbos Biome stating that the Cape Floral Kingdom traditionally does not include the fynbos and renosterveld outliers to the north and east. Jürgens (1997) placed the entire study area within the Succulent Karoo Biome while, Mucina et al. (2005) and Mucina et al. (2006) place the Hantam and Tanqua Karoo within the

185

Succulent Karoo Biome and the Roggeveld within the Fynbos Biome (Rebelo et al. 2006) clearly stating that they did not apply the explicit and globally derived definition of a biome and only considered botanical elements (Rutherford et al. 2006). Born et al. (2007) placed the entire region within the Succulent Karoo Biome using the Succulent Karoo Ecosystem Plan (SKEP) defined subregions as the unit of their study. This study however, indicated stronger links to the Fynbos Biome than the Succulent Karoo Biome for some of the subregions.

This study of the Hantam-Tanqua-Roggeveld subregion clearly indicates that the Tanqua Karoo is very different from the Winter Rainfall Karoo and the Mountain Renosterveld and that it belongs to the Succulent Karoo Biome. There is a relationship between the Mountain Renosterveld and Winter Rainfall Karoo vegetation within the study area, however, these two groups differ in various aspects. The floristic data support a closer relationship between the Winter Rainfall Karoo and Tanqua Karoo, yet, the Roggeveld Karoo does seem to differ the most and is likely to be more closely related to the Nama Karoo Biome than the Succulent Karoo Biome, but this will have to be researched in more detail. The Mountain Renosterveld is clearly different from the Winter Rainfall Karoo and Tanqua Karoo but does not fit the classic Fynbos Biome delimitations. How this arid Mountain Renosterveld links to other renosterveld vegetation types in the ‘true’ Fynbos Biome will have to be investigated in more detail on a study of renosterveld vegetation.

186

Table 11.1 A summary of the prominent relationships between the three main vegetation groups Attribute Mountain Renosterveld Winter Rainfall Karoo Tanqua Karoo 1. Phytosociology (floristics) Associations 1, 2 and 3 Associations 4, 5 and 6 Associations 7 and 8 2. Environmental parameters Mean annual precipitation 200 mm – 400 mm 100 mm – 400 mm <100 mm – 200 mm Coefficient of variation for mean 35% to 40%, higher-lying areas 25% to 35% to 40% 35% to 40% annual precipitation 35% Mean daily minimum and maximum Minimum: -2°C to 2°C Minimum: 2°C to 4°C Minimum: 4°C to 6°C for the coldest months (June and Maximum: 12°C to 14 °C Maximum: 16°C to 18°C Maximum: 18°C to 20°C July) Mean daily minimum and maximum Minimum: 10°C to 14°C Minimum: 12°C to 14°C Minimum: 14°C to 18°C for the warmest months (January Maximum: 28°C to 30 °C Maximum: 30°C to 32 °C Maximum: 32°C to >34 °C and February) Snow 6 snow days per year over a 24-year 1 snow day per year over a 20-year No snow period period Heat units (degree days) <200 - 400 400 – 800 600 - 1000 Accumulated positive chill units (May 1250 - >1750 750 – 1000 250 - 500 to September) Soils Shallow stony lithosols and duplex soils Shallow lithosols and duplex soils, but Shallow lithosols often including desert in the occasional lowlands where dolerite occurs soils are red pavement and deep unconsolidated structured and red vertic clays deposits in the alluvial parts Altitude High-lying, generally 700 – 1600 m above 300 – 1400 m above seal level Low-lying, generally 200 – 800 m above sea level sea level Fire Fire No fire No fire

187

Table 11.1 (continued) 3. Diversity parameters Species-area curves No significant difference found between the slopes of the Mountain Renosterveld Significantly shallower slopes than for the and Winter Rainfall Karoo Mountain Renosterveld and Winter Rainfall Karoo Species richness in 1000 m2 Significantly higher values than for the Significantly higher values than for the Significantly lower values than for the Winter Rainfall Karoo and Tanqua Karoo Tanqua Karoo Mountain Renosterveld and Winter Rainfall Karoo Species richness across six plot No significant difference between the Mountain Renosterveld and Winter Rainfall Significantly different from the Mountain sizes (1 m², 5 m², 10 m², 20 m², Karoo at plot sizes less than 1000 m² Renosterveld and Winter Rainfall Karoo for 50 m², 100 m²) all plot sizes less than 1000 m² Evenness Evenness values significantly higher than for the Tanqua Karoo Evenness values significantly lower than for the Mountain Renosterveld and Winter Rainfall Karoo Shannon index of diversity No significant difference found between the Shannon index values of the Mountain Significantly lower Shannon index values Renosterveld and Winter Rainfall Karoo but values significantly different from the than for the Mountain Renosterveld and Tanqua Karoo Winter Rainfall Karoo Simpson index of diversity Simpson index values generally higher than for the Tanqua Karoo but, no Significantly lower Simpson index values significant difference between the Mountain Renosterveld and Winter Rainfall Karoo than for the Mountain Renosterveld and Winter Rainfall Karoo Principal Co-ordinate Analysis of A cluster representing a selection of high- Central cluster of Winter Rainfall Karoo A cluster representing Tanqua Karoo floristic data for seven plot sizes lying plots generally with steep species- vegetation plots and several vegetation with a low plant cover, shallow area curve slopes and high y-intercept transitional Mountain Renosterveld species-area curve slopes and low y- values = ‘True’ Mountain Renosterveld plots intercept values vegetation

188

Table 11.1 (continued) 4. Life forms Classic life forms at a species level expressed as a percentage of the total number of species Significantly higher values than for the Significantly lower values than for the Mountain Renosterveld Phanerophytes Winter Rainfall Karoo and Tanqua Karoo Chamaephytes No significant difference between the three vegetation groups Low species values Significantly lower values than for the Significantly higher values than for the Hemicryptophytes Tanqua Karoo Winter Rainfall Karoo Significantly higher values than for the Intermediate cryptophyte species Significantly lower values than for the Cryptophytes Tanqua Karoo numbers Mountain Renosterveld Therophytes Significantly higher therophyte species numbers Significantly lower therophyte numbers Lianas No significant difference between the three vegetation groups Parasites No significant difference between the three vegetation groups Significantly lower succulent species Significantly higher succulent species contributions than for the Mountain Succulents (combined) contributions than for the Winter Rainfall Renosterveld Karoo and Tanqua Karoo

189

Table 11.1 (continued) Classic life forms at a vegetation cover level expressed as a percentage of the total cover Phanerophytes Significantly higher phanerophyte cover Intermediate phanerophyte cover Lowest phanerophyte cover Significantly higher cover than the Tanqua Karoo Significantly lower cover than the Chamaephytes Mountain Renosterveld and Winter Rainfall Karoo Hemicryptophytes No significant difference between the three vegetation groups Cryptophytes No significant difference between the three vegetation groups Therophytes No significant difference between the three vegetation groups Lianas No significant difference between the three vegetation groups Parasites No significant difference between the three vegetation groups Significantly lower succulent cover than Significantly higher cover than for the Mountain Renosterveld Succulents (combined) for the Winter Rainfall Karoo and the Tanqua Karoo

190

References

ACOCKS, J.P.H. 1953. Veld types of South Africa. Memoirs of the Botanical Survey of South Africa 28, 1-192. ACOCKS, J.P.H. 1988. Veld types of South Africa. 3rd edn. Memoirs of the Botanical Survey of South Africa 57, 1-146. ANDERSON, P.M.L. AND HOFFMAN, M.T. 2007. 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 70, 686-700. BOND, W.J. AND RICHARDSON, D.M. 1990. What can we learn from extinctions and invasions about the effects of climate change? South African Journal of Science 86, 429-433. BOND, W.J. AND VAN WILGEN, B.W. 1996. Fire and plants. Chapman & Hall, London. BORN, J., LINDER, H.P. AND DESMET, P. 2007. The Greater Cape Floristic Region. Journal of Biogeography 34, 147-162. BURKE, A., ESLER, K.J., PIENAAR, E. AND BARNARD, P. 2003. Species richness and floristic relationships between mesas and their surroundings in southern African Nama Karoo. Diversity and Distributions 9, 43-53. CAPITANIO, R. AND CARCAILLET, C. 2008. Post-fire Mediterranean vegetation dynamics and diversity: A discussion of succession models. Forest Ecology and Management 255, 431- 439. COWLING, R.M., GIBBS RUSSEL, G.E., HOFFMAN, M.T. AND HILTON-TAYLOR, C. 1989. Patterns of plant species diversity in southern Africa. In: B.J. Huntley (Ed.). Biotic diversity in southern Africa. Concepts and Conservation, pp. 19-50. Oxford University Press, Cape Town. COWLING, R.M. AND HILTON-TAYLOR, C. 1994. Patterns of plant diversity and endemism in southern Africa: an overview. In: B.J. Huntley (Ed.). Botanical diversity in southern Africa. Strelitzia 1, pp. 31-52. National Botanical Institute, Pretoria. COWLING, R.M. AND HOLMES, P.M. 1992. Endemism and speciation in a lowland flora from the Cape Floristic Region. Biological Journal of the Linnean Society 47, 367-383. COWLING, R.M., HOLMES, P.M. AND REBELO, A.G. 1992. Plant diversity and endemism. In: R.M. Cowling (Ed.). The Ecology of Fynbos: Nutrients, fire and diversity, pp. 62-112. Oxford University Press, Cape Town. CULLINAN, P. 2003. Robert Jacob Gordon 1743 – 1795. The man and his travels at the Cape. http:// web.uct.ac.za/depts/age/people/Gordon/frameset.htm DIELS, L. 1909. Formationen und Florenelemente im nordwestlichen Kapland. Botanische Jahrbücher 44, 91-124.

191

EGLER, F.E. 1954. Vegetation science concepts. I. Initial floristic composition – a factor in old- field vegetation development. Vegetatio 4, 412-418. ESLER , K.J., JONES, F.E., BURKE, A., SAMWAYS, M., BARNARD, P., GILBERT, F. AND WISSEL C. 2002. Guidelines for sustainable rangeland management: recommendations on ecological restoration of rangelands in the semi-arid Nama Karoo Biome. Magenta media, Cape Town. ESLER, K.J., MILTON, S.J. AND DEAN, W.R.J. 2006. Karoo veld Ecology and Management. Briza publications, Pretoria. GUO, Q. 2001. Early post-fire succession in California chaparral: changes in diversity, density, cover and biomass. Ecological research 16, 471-485. JÜRGENS, N. 1997. Floristic biodiversity and history of African arid regions. Biodiversity and Conservation 6, 495-514. KEELEY, S.C., KEELEY, J.E., HUTCHINSON, S.M. AND JOHNSON, A.W. 1981. Post-fire succession of herbaceous flora in southern California chaparral. Ecology 62, 1608-1621. KONGOR, R.Y., 2009. Plant response to habitat fragmentation: clues from species and functional diversity in three Cape lowland vegetation types of South Africa. PhD thesis, University of Stellenbosch, Stellenbosch. KÖRNER, C. 2000. Why are there global gradients in species richness? Mountains might hold the answer. Trends in Ecology and Evolution 15, 513-514. KRUG, C.B. AND KRUG, R.M. 2007. Restoration of old fields in Renosterveld: a case study in a Mediterranean-type shrubland of South Africa. In: V.A. Cramer and R.J. Hobbs (Eds). Old fields. Dynamics and restoration of abandoned farmland, pp. 265-285. Island Press, Washington. LOW, A.B. AND REBELO A.G. 1996. Vegetation of South Africa, Lesotho and Swaziland. Department of Environmental Affairs and Tourism, Pretoria. MAGURRAN, A.E. 1988. Ecological diversity and its measurement. University Press, Cambridge. MARLOTH, R. 1908. Das Kapland, insonderheit das Reich der Kapflora, das Waldgebiet und die Karoo, pflanzengeografisch dargestellt. Wiss. Ergebn. Deutsch. Tiefsee-Exped. ‘Waldivia’, 1898 – 1899. 2, T. 3, Fischer, Jena. MIDGLEY, G.F., HANNAH, L., ROBERTS, R., MacDONALD, D.J. AND ALLSOPP, J. 2000. Have pleistocene climatic cycles influenced species richness patterns in the greater Cape Mediterranean Region? Journal of Mediterranean Ecology 2, 137-144. MIDOKO-IPONGA, D., KRUG, C.B. AND MILTON, S.J. 2005. Competition and herbivory influence growth and survival of shrubs on old fields: implications for restoration of renosterveld shrubland. Journal of Vegetation Science 16, 685-692. MILLENNIUM ECOSYSTEM ASSESSMENT 2005. Ecosystems and human well-being: Desertification Synthesis, Washington DC.

192

MOUILLOT, D. AND LEPRÊTRE, A. 1999. A comparison of species diversity estimators. Research on Population Ecology 41, 203-215. MUCINA, L., JÜRGENS, N., LE ROUX, A., RUTHERFORD, M.C., SCHMIEDEL, U., ESLER, K.J., POWRIE, L.W., DESMET, P.G. AND MILTON, S.J. 2006. Succulent Karoo Biome. In: L. Mucina and M.C. Rutherford (Eds). The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19, pp. 220-299. South African National Biodiversity Institute, Pretoria. MUCINA, L., RUTHERFORD, M.C. AND POWRIE, L.W. (Eds) 2005. Vegetation map of South Africa, Lesotho and Swaziland, 1 : 1 000 000 scale sheet maps. South African National Biodiversity Institute, Pretoria. POTTS, S.G., VULLIAMY, B., DAFNI, A., NE’EMAN, G., O’TOOLE, C., ROBERTS, S. AND WILLMER, P. 2003. Response of plant-pollinator communities to fire: changes in diversity, abundance and floral reward structure. OIKOS 101, 103-112. PROCHEŞ, Ş., COWLING, R.M. AND MUCINA, L. 2003. Species-area curves based on relevé data for the Cape Floristic Region. South African Journal of Science 99, 74-476. REBELO, A.G., BOUCHER, C., HELME, N., MUCINA, L. AND RUTHERFORD, M.C. 2006. Fynbos Biome. In: L. Mucina and M.C. Rutherford (Eds). The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19, pp. 52-219. South African National Biodiversity Institute, Pretoria. RICHARDSON, D.M., COWLING, R.M., BOND, W.J., STOCK, W.D. AND DAVIS, G.W. 1995. Links between biodiversity and ecosystem function in the Cape Floristic Region. In: G.W. Davis and D.M. Richardson (Eds). Mediterranean-type ecosystems: The function of biodiversity, pp. 285-333. Springer-Verlag, Berlin, Heidelberg. RUTHERFORD, M.C., MUCINA, L. AND POWRIE, L.W. 2006. Biomes and bioregions of southern Africa. In: L. Mucina and M.C. Rutherford (Eds). The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19, pp. 30-51. South African National Biodiversity Institute, Pretoria. RUTHERFORD, M.C. AND WESTFALL., R.H. 1994. Biomes of Southern Africa. An objective characterisation. Memoirs of the Botanical Survey of South Africa 63, 1-94. SCHULZE, R.E. 1997. South African Atlas of Agrohydrology and – Climatology. Water Research Commission, Pretoria, Report TT82/96. SCHWILK, D.W., KEELEY, J.E. AND BOND, W.J. 1997. The intermediate disturbance hypothesis does not explain fire and diversity pattern in fynbos. Plant Ecology 132, 77-84. STEYN, H.M., VAN ROOYEN, N. AND VAN ROOYEN, M.W. 1996. The prediction of phenological stages in four Namaqualand ephemeral species using thermal unit indices. Israel Journal of Plant Sciences 44, 147-160.

193

194

Chapter 12

Summary

Located within the Succulent Karoo and Fynbos Biomes, the Hantam-Tanqua-Roggeveld lies within the Northern and Western Cape Provinces of the Republic of South Africa. The aim of this study was to gain information on the floristics, diversity and dynamics of the vegetation to improve our understanding, conservation and management of this unique, botanically unexplored, arid area.

After undertaking a phytosociological study, which identified eight associations and 25 subassociations 40 Whittaker plots were surveyed across the eight associations to collect diversity data. Comparisons of the slopes of the species-area curves showed that the slopes of the Tanqua Karoo associations were significantly shallower than those of the other associations.

Species richness values for the Mountain Renosterveld were highest, Winter Rainfall Karoo values intermediate and Tanqua Karoo values lowest. The evenness, Shannon and Simpson indices did not differ significantly between the Mountain Renosterveld and Winter Rainfall Karoo, however, these values were significantly higher than for the Tanqua Karoo. An ordination of species richness data confirmed a clear Tanqua Karoo cluster, but the Mountain Renosterveld could only be partially separated from the Winter Rainfall Karoo.

Chamaephyte, cryptophyte and therophyte species dominated the study area. Comparisons of life form spectra among associations showed clear differences in the percentage contribution of life forms at a species and vegetation cover level. The percentage contribution of succulent species was low in Mountain Renosterveld, intermediate in Winter Rainfall Karoo and highest in the Tanqua Karoo. Results confirmed the Tanqua Karoo and Winter Rainfall Karoo inclusion into the Succulent Karoo Biome and the strong karooid affinities of the Mountain Renosterveld.

Various species and life form diversity parameters were studied on abandoned croplands of different ages and in the natural vegetation of the Roggeveld (Mountain Renosterveld vegetation). Therophyte and chamaephyte species at a species level were most abundant. Although therophytes dominated early successional croplands they were quickly replaced by chamaephytes which made an overwhelming contribution to vegetation cover. Species-area curves using the exponential function differed significantly for all the abandoned croplands and the natural vegetation. Species richness increased with an increase in age of the abandoned croplands but evenness, Shannon and Simpson indices did not show a similar increase. It was found that after 33-years, an abandoned cropland should be as species rich

195

as the natural vegetation, however, a Principal Co-ordinate Analysis indicated that floristically these abandoned croplands were still extremely different from the natural vegetation. The rate of recovery varied among life forms and geophytes (cryptophytes) still remain greatly under represented after 20 years of abandonment.

A ten year post-fire study in the Roggeveld indicated that the vegetation cover began to establish within nine months following fire. Generally, species richness and Shannon index values reached a maximum within three years after the fire and then declined in the subsequent years. A Principal Co-ordinate Analysis of species compositional data clearly separated the first two years from the following years. This study seems to lend support for the ‘initial floristic composition’ model of Egler (1954).

196

Chapter 14

References

198

BEATLEY, J.C. 1976. Vascular plants of the Nevada Test Site and central and southern Nevada: Ecological and geographic distributions. Energy Research and Development Administration, Washington DC. BOND, W.J. AND RICHARDSON, D.M., 1990. What can we learn from extinctions and invasions about the effects of climate change? South African Journal of Science 86, 429-433. BOND, W.J. AND VAN WILGEN, B.W. 1996. Fire and plants. Chapman & Hall, London. BORN, J., LINDER, H.P. AND DESMET, P. 2007. The Greater Cape Floristic Region. Journal of Biogeography 34, 147-162. BOUCHER, C. AND MOLL, E.J. 1981. South African Mediterranean shrublands. In: F. Di Castri, D.W. Goodaal and R.L. Specht (Eds) Ecosystems of the world, pp. 233-248. Mediterranean type shrublands. Elsevier, Amsterdam. BROENNIMANN, O., THUILLER, W., HUGHES, G., MIDGLEY, G.F., ALKEMADE, J.M.R. AND GUISAN, A. 2006. Do geographic distribution, niche property and life form explain plants’ vulnerability to global change? Global Change Biology 12, 1079-1093. BURKE, A., ESLER, K.J., PIENAAR, E. AND BARNARD, P. 2003. Species richness and floristic relationships between mesas and their surroundings in southern African Nama Karoo. Diversity and Distributions 9, 43-53. CAIN, S.A. 1950. Life-forms and phytoclimate. The Botanical Review 16, 1-32. CAM, E., NICHOLS, J.D., HINES, J.E., SAUER, J.R., ALPIZAR-JARA, R. AND FLATHER,C.H. 2002. Disentangling sampling and ecological explanations underlying species-area relationships. Ecology 83, 1118-1130. CAPITANIO, R. AND CARCAILLET, C. 2008. Post-fire Mediterranean vegetation dynamics and diversity: A discussion of succession models. Forest Ecology and Management 255, 431-439. CARSON, W.P. AND PICKETT, S.T.A. 1990. Role of resources and disturbance in the organization of an old-field plant community. Ecology 71, 226-238. CILLIERS, C., THERON, H., RÖSCH, H. AND LE ROUX, A. 2002. Succulent Karoo Ecosystem Plan, Sub-regional report, Hantam/Tanqua/Roggeveld. Succulent Karoo Ecosystem Plan report. CLINEBELL, H.R.R., PHILLIPS, O.L., GENTRY, A.H., STARK, N. AND ZUURING, H. 1995. Prediction of neotropic tree and liana richness from soil and climatic data. Biodiversity and Conservation 4, 56-90. COLWELL, R.K., MAO, C.X. AND CHANG, J. 2004. Interpolating, extrapolating, and comparing incidence-based species accumulation curves. Ecology 85, 2717-2727. CONNELL, J.H. AND SLATYER, R.O. 1977. Mechanisms of succession in natural communities and their role in community stability and organisation. The American Naturalist 111, 1119-1144. CONNOR, E.F. AND McCOY, E.D. 1979. The statistics and biology of the species-area relationship. The American Naturalist 113, 791-833.

199

CONSERVATION INTERNATIONAL WEBSITE: http://www.biodiversityhotspots.org (accessed 2006). COUNCIL FOR GEOSCIENCE 1973. Geological map 3218 Clanwillliam. Council for Geoscience, Pretoria. COUNCIL FOR GEOSCIENCE 1983. Geological map 3220 Sutherland. Council for Geoscience, Pretoria. COUNCIL FOR GEOSCIENCE 1989. Geological map 3120 Williston. Council for Geoscience, Pretoria. COUNCIL FOR GEOSCIENCE 1991. Geological map 3220 Ladismith. Council for Geoscience, Pretoria. COUNCIL FOR GEOSCIENCE 1997. Geological map 3319 Worcester. Council for Geoscience, Pretoria. COUNCIL FOR GEOSCIENCE 2001. Geological map 3118 Calvinia. Council for Geoscience, Pretoria. COUNCIL FOR GEOSCIENCE 2008. Geological data 1: 250 000. CD data provided by the Council for Geoscience, Silverton, Pretoria. COWLING, R.M. 1999. Planning for persistence – systematic reserve design in southern Africa’s Succulent Karoo desert. Parks 9 ,17-30. COWLING, R.M., ESLER, K.J., MIDGLEY, G.F. AND HONIG, M.A. 1994. Plant functional diversity, species diversity and climate in arid and semi-arid southern Africa. Journal of Arid Environments 27, 141-158. COWLING, R.M., ESLER, K.J. AND RUNDEL, P.W. 1999. Namaqualand, South Africa – an overview of a unique winter-rainfall desert ecosystem. Plant Ecology 142, 3-21. COWLING, R.M., GIBBS RUSSEL, G.E., HOFFMAN, M.T. AND HILTON-TAYLOR, C. 1989. Patterns of plant species diversity in southern Africa. In: B.J. Huntley (Ed.). Biotic diversity in southern Africa. Concepts and Conservation, pp. 19-50. Oxford University Press, Cape Town. COWLING, R.M. AND HILTON-TAYLOR, C. 1994. Patterns of plant diversity and endemism in southern Africa: an overview. In: B.J. Huntley (Ed.). Botanical diversity in southern Africa. Strelitzia 1, pp. 31-52. National Botanical Institute, Pretoria. COWLING, R.M., HOLMES, P.M. 1992. Endemism and speciation in a lowland flora from the Cape Floristic Region. Biological Journal of the Linnean Society 47, 367-383. COWLING, R.M., HOLMES, P.M. AND REBELO, A.G. 1992. Plant diversity and endemism. In: R.M. Cowling (Ed.). The Ecology of Fynbos: Nutrients, fire and diversity, pp. 62- 112. Oxford University Press, Cape Town. COWLING, R.M. AND PIERCE, S. 1999. Namaqualand a succulent desert. Fernwood Press, Vlaeberg. COWLING, R.M., RICHARDSON, D.M. AND MUSTARD, P.J. 1997. Fynbos. In: R.M. Cowling, D.M. Richardson and S.M. Pierce (Eds). Vegetation of southern Africa, pp. 99-129. University Press, Cambridge.

200

COX, C.B. AND MOORE, P.D. 1994. Biogeography: an ecological and evolutionary approach. 5th Edition. Blackwell Scientific Publishing, Oxford. CRAMER, V.A. AND HOBBS, R.J. (Eds) 2007. Old fields. Dynamics and restoration of abandoned farmland. Island Press, Washington. CRAMER, V.A., HOBBS, R.J. AND STANDISH, R.J. 2007. What’s new about old fields? Land abandonment and ecosystem assembly. Trends in Ecology and Evolution 23, 104- 112. CRITICAL ECOSYSTEM PARTNERSHIP FUND, 2003. Ecosystem Profile: The Succulent Karoo hotspot, Namibia and South Africa. Critical Ecosystem Partnership Fund report. CULLINAN, P. 2003. Robert Jacob Gordon 1743 – 1795. The man and his travels at the Cape. http:// web.uct.ac.za/depts/age/people/Gordon/frameset.htm DANIN, A. AND ORSHAN, G. 1990. The distribution of Raunkiaer life forms in Israel in relation to the environment. Journal of Vegetation Science 1, 41-48. DEBUSSCHE, M., JOSÉ, E., LEPART, J., HOUSSARD, C. AND LAVOREL, S. 1996. Changes in Mediterranean plant succession: old-fields revisited. Journal of Vegetation Science 7, 519-526. DESMET, P.G. AND COWLING, R.M. 1999. The climate of the Karoo. A functional approach. In: W.R. Dean and S.J. Milton (Eds). The Karoo: Ecological patterns and processes, pp. 3-16. Cambridge University Press, Cambridge. DESMET, P. AND COWLING, R.M. 2004. Using the species-area relationship to set baseline targets for conservation. Ecology and Society 9, 11 [online] URL: http://www.ecologyandsociety.org/vol9/iss2/art11 DIELS, L. 1909. Formationen und Florenelemente im nordwestlichen Kapland. Botanische Jahrbücher 44, 91-124. DRAKARE, S., LENNON, J.J. AND HILLEBRAND, H. 2006. The imprint of geographical, evolutionary and ecological context on species-area relationships. Ecology Letters 9, 215-227. DU PLESSIS, H.M. 1987. Land Types of the maps 2816 Alexander Bay, 2818 Warmbad, 2916 Springbok, 2918 Pofadder, 3017 Garies, 3018 Loeriesfontein. Memoirs on the Agricultural Natural Resources of South Africa 9, 1-538. DU TOIT, P.C.V. 1998a. Research note: Grazing-index method procedures of vegetation surveys. African Journal of Range and Forage Science 14, 107-110. DU TOIT, P.C.V. 1998b. Description of a method for assessing veld condition in the Karoo. African Journal of Range and Forage Science 14, 90-93. EGLER, F.E. 1954. Vegetation science concepts. I. Initial floristic composition – a factor in old-field vegetation development. Vegetatio 4, 412-418. EPSTEIN, H.E., GILL, R.A., PARUELO, J.M., LAUENROTH, W.K., JIA, G.J. & BURKE, I.C. 2002. The relative abundance of three plant functional types in temperate grasslands

201

and shrublands of North and South America: effects of projected climate change. Journal of Biogeography 29, 875-888. ESLER , K.J., JONES, F.E., BURKE, A., SAMWAYS, M., BARNARD, P., GILBERT, F. AND WISSEL C. 2002. Guidelines for sustainable rangeland management: recommendations on ecological restoration of rangelands in the semi-arid Nama Karoo Biome. Magenta media, Cape Town. ESLER, K.J., MILTON, S.J. AND DEAN, W.R.J. (Eds) 2006. Karoo veld Ecology and Management. Briza publications, Pretoria. ESLER, K.J., RUNDEL, P.W. AND COWLING, R.M. 1999a. The Succulent Karoo in a global context: plant structural and functional comparison with North America winter-rainfall deserts. In: W.R. Dean and S.J. Milton (Eds). The Karoo: Ecological patterns and processes, pp. 123-144. Cambridge University Press, United Kingdom. ESLER, K.J., RUNDEL, P.W. AND VOSTER, P. 1999b. Biogeography of prostrate-leaved geophytes in semi-arid South Africa: hypotheses on functionality. Plant Ecology 142, 105-120. FOX, G.A. 1992. The evolution of life history traits in desert annuals. American Journal of Botany 77, 1508-1518. FRANCIS, M.L., FEY, M.V., PRINSLOO, H.P., ELLIS, F., MILLS, A.J. AND MEDINSKI, T.V. 2007. Soils of Namaqualand: Compensations for aridity. Journal of Arid Environments 70, 588-603. GERMISHUIZEN, G. AND MEYER, N.L. (Eds) 2003. Plants of southern Africa: an annotated checklist. Strelitzia 14. National Botanical Institute, Pretoria. GIBBS RUSSELL, G.E. 1987. Preliminary floristic analysis of the major biomes in Southern Africa. Bothalia 17, 213-227. GIBSON, D.J., MIDDLETON, B.A., FOSTER, K., HONU, Y.A.K., HOYER, E.W. AND MATHIS, M. 2005. Species frequency dynamics in an old-field succession: effects of disturbance, fertilization and scale. Journal of Vegetation Science 16, 415-422. GOOD, R. 1947. The geography of flowering plants. Longmans, Green & Co., New York. GOTELLI, N.J. AND COLWELL, R.K. 2001. Quantifying biodiversity: procedures and pitfalls in the measurement and comparison of species richness. Ecology Letter 4, 379-391. GUO, Q. 2001. Early post-fire succession in California chaparral: changes in diversity, density, cover and biomass. Ecological research 16, 471-485. GUREVITCH, J., SCHEINER, S.M. AND FOX, G.A. 2002. The Ecology of Plants. Sinauer, Massachusetts. HANES, T.L. 1971. Succession after fire in the chaparral of southern California. Ecological Monographs 41, 27-52. HE, F. AND LEGENDRE, P. 1996. On species-area relations. The American Naturalist 4, 719-737. HE, F. AND LEGENDRE, P. 2002. Species diversity patterns derived from species-area models. Ecology 83, 1185-1198.

202

HELLMANN, J.J. AND FOWLER, G.W. 1999. Bias, precision, and accuracy of four measures of species richness. Ecological Applications 9, 824-834. HELTSHE, J.F. AND FORRESTER, N.E. 1983. Estimating species richness using the Jackknife Procedure. Biometrics 39, 1-11. HENNEKENS, S.M. AND SCHAMINÉE, J.H.J. 2001. TURBOVEG, a comprehensive data base management system for vegetation data. Journal of Vegetation Science 12, 589-591. HILL, M.O. 1979. TWINSPAN – A FORTRAN program for arranging multivariate data in an ordered two-way table by classification of the individuals and attributes. Ecology & Systematics, Cornell University Ithaca, New York. HILTON-TAYLOR, C. 1994. Western Cape Domain (Succulent Karoo). In: S.D. Davis, V.H. Heywood and A.C. Hamilton (Eds). Centres of plant diversity. A guide and strategy for their conservation, pp. 201-203. IUCN Publications Unit, Cambridge. HOFFMAN, M.T. AND COWLING, R.M. 1987. Plant physiognomy, phenology and demography. In: R.M. Cowling and P.W. Roux (Eds) The Karoo Biome: a preliminary synthesis. Part 2 – Vegetation and history. South African National Scientific Programmes Report 142, CSIR, Pretoria. HUBERTY, L.E., GROSS, K.L. AND MILLER, C.J. 1998. Effects of nitrogen addition on successional dynamics and species diversity in Michigan old-fields. Journal of Ecology 86, 794-803. INOUYE, R.S., HUNTLY, N.J., TILMAN, D., TESTER, J.R., STILLWELL, M. AND ZINNEL, K.C. 1987. Old-field succession on a Minnesota sand plain. Ecology 68, 12-26. JOHNSON, M.R., VAN VUUREN, C.J., VISSER, J.N.J., COLE, D.I., WICKENS, H. DE.V., CHRISTIE, A.D.M., ROBERTS, D.L. AND BRANDL, G. 2006. Sedimentary rocks of the Karoo Supergroup. In: M.R. Johnson, C.R. Anhaeusser and R.J. Thomas (Eds). The geology of South Africa, pp. 461-500. The Geological Society of South Africa, Johannesburg/Council for Geoscience, Pretoria. JÜRGENS, N. 1997. Floristic biodiversity and history of African arid regions. Biodiversity and Conservation 6, 495-514. JÜRGENS, N., GOTZMANN, I.H. AND COWLING, R.M. 1999. Remarkable medium-term dynamics of leaf succulent Mesembryanthemaceae shrubs in the winter-rainfall desert of northwestern Namaqualand, South Africa. Plant Ecology 142, 87-96. KEELEY, J.E. 2003. Relating species abundance distributions to species-area curves in two Mediterranean-type shrublands. Diversity and Distributions 9, 253-259. KEELEY, S.C., KEELEY, J.E., HUTCHINSON, S.M. AND JOHNSON, A.W. 1981. Post-fire succession of herbaceous flora in southern California chaparral. Ecology 62, 1608- 1621. KONGOR, R.Y., 2009. Plant response to habitat fragmentation: clues from species and functional diversity in three Cape lowland vegetation types of South Africa. PhD thesis, University of Stellenbosch, Stellenbosch.

203

KÖRNER, C. 2000. Why are there global gradients in species richness? Mountains might hold the answer. Trends in Ecology and Evolution 15, 513-514. KRUG, C.B. AND KRUG, R.M. 2007. Restoration of old fields in Renosterveld: a case study in a Mediterranean-type shrubland of South Africa. In: V.A. Cramer and R.J. Hobbs (Eds). Old fields. Dynamics and restoration of abandoned farmland, pp. 265-285. Island Press, Washington. KRUGER, F.J. AND BIGALKE, R.C. 1984. Fire in fynbos. In: P. De V. Booysen and N.M. Tainton (Eds). Ecological effects of fire in South African Ecosystems. pp 69-114. Springer-Verlag, Berlin. LAVOREL, S., LEPART, J., DEBUSSCHE, M., LEBRETON, J-D. AND BEFFY, J-L. 1994. Small scale disturbances and the maintenance of species diversity in Mediterranean old fields. OIKOS 70, 455-473. LLORET, F. AND VILÁ, M. 2003. Diversity patterns of plant functional types in relation to fire regime and previous land use in Mediterranean woodlands. Journal of Vegetation Science 14, 387-398. LOMOLINO, M.V. 2000. Ecology’s most general, yet protean pattern: the species-area relationship. Journal of Biogeography 27, 17-26. LOW, A.B. AND REBELO, A.G.. 1996. Vegetation of South Africa, Lesotho and Swaziland. Department of Environmental Affairs and Tourism, Pretoria. MAGURRAN, A.E. 1988. Ecological diversity and its measurement. Cambridge University Press, Cambridge. MARLOTH, R. 1908. Das Kapland, insonderheit das Reich der Kapflora, das Waldgebiet und die Karoo, pflanzengeografisch dargestellt. Wissenschaftliche Ergebnisse der Deutscher Tiefsee-Expedition ‘Waldivia’, 1898 – 1899. 2, T. 3, Fischer, Jena. McCUNE, B. AND GRACE, J.B. 2002. Analysis of Ecological Communities. MjM Stoftware Design, Gleneden Beach, Oregon. McDOWELL, C.R. AND MOLL, E.J. 1992. The influence of agriculture on the decline of West Coast Renosterveld, south-western Cape, South Africa. Journal of Environmental Management 35, 173 – 192. McINTYRE, S., LAVOREL, S., LANDSBERG, J. AND FORBES, T.D.A. 1999. Disturbance response in vegetation – towards a global perspective on functional traits. Journal of Vegetation Science 10, 621-630. McINTYRE, S., LAVOREL, S. AND TREMONT, R.M. 1995. Plant life-history attributes: their relationship to disturbance response in herbaceous vegetation. Journal of Ecology 83, 31-44. MIDGLEY, G.F., HANNAH, L., ROBERTS, R., MacDONALD, D.J. AND ALLSOPP, J. 2000. Have pleistocene climatic cycles influenced species richness patterns in the greater Cape Mediterranean Region? Journal of Mediterranean Ecology 2, 137-144.

204

MIDOKO-IPONGA, D., KRUG, C.B. AND MILTON, S.J. 2005. Competition and herbivory influence growth and survival of shrubs on old fields: implications for restoration of renosterveld shrubland. Journal of Vegetation Science 16, 685-692. MILLENNIUM ECOSYSTEM ASSESSMENT 2005. Ecosystems and human well-being: Desertification Synthesis, Washington DC. MILTON, S.J., YEATON, R.I., DEAN, W.R.J. AND VLOK, J.H.J. 1997. Succulent Karoo. In: R.M. Cowling, D.M. Richardson and S.M. Pierce (Eds). Vegetation of southern Africa, pp. 99-129. University Press, Cambridge. MOLL, E.J. & BOSSI, L. 1984. Assessment of the extent of the natural vegetation of the Fynbos Biome of South Africa. South African Journal of Science 80, 355-358. MOLL, E.J., CAMPBELL, B.M., COWLING, R.M., BOSSI, L., JARMAN, M.L. AND BOUCHER, C. 1984. A description of major vegetation categories in and adjacent to the Fynbos Biome. South African National Scientific Programmes Report 83, 1-29. MOUILLOT, D. AND LEPRÊTRE, A. 1999. A comparison of species diversity estimators. Research on Population Ecology 41, 203-215. MUCINA, L., JÜRGENS, N., LE ROUX, A., RUTHERFORD, M.C., SCHMIEDEL, U., ESLER, K.J., POWRIE, L.W., DESMET, P.G. AND MILTON, S.J. 2006. Succulent Karoo Biome. In: L. Mucina and M.C. Rutherford (Eds). The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19, pp. 220-299. South African National Biodiversity Institute, Pretoria. MUCINA, L. AND RUTHERFORD, M.C. (Eds) 2006. The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19. South African National Biodiversity Institute, Pretoria. MUCINA, L., RUTHERFORD, M.C. AND POWRIE, L.W. (Eds) 2005. Vegetation map of South Africa, Lesotho and Swaziland, 1 : 1 000 000 scale sheet maps. South African National Biodiversity Institute, Pretoria. MUELLER-DOMBOIS, D. AND ELLENBERG, H. (Eds) 1974. Aims and methods of vegetation ecology. Wiley, New York. MYERS, N., MITTERMEIR, R.A., MITTERMEIR, C.G., DE FONSECA, G.A.B. AND KENT, J. 2000. Biodiversity hotspots for conservation priorities. Nature 403, 853-858. NELDER V.J., WILSON, B.A., THOMPSON, E.J. AND DILLEWAARD, H.A.. 2005. Methodology for Survey and Mapping of Regional Ecosystems and Vegetation Communities in Queensland. Version 3.1. Updated September 2005. Queensland Herbarium, Environmental Protection Agency, Brisbane. NOVELLIE, P. AND STRYDOM, G. 1987. Monitoring the response of vegetation to use by large herbivores: an assessment of some techniques. South African Journal of Wildlife Research 17, 109-117. OTTO, R., KRÜSI, B.O., BURGA, C.A. AND FERNÁNDEZ-PALACIOS, J.M. 2006. Old-field succession along a precipitation gradient in the semi-arid coastal region of Tenerife. Journal of Arid Environments 65, 156-178.

205

PALMER, M.W. 1990. The estimation of species richness by extrapolation. Ecology 71, 1195-1198. PALMER, M.W. AND WHITE, P.S. 1994. Scale dependence and the species-area relationship. The American Naturalist 144, 717-740. PAUSAS, J.G. 1999. Response of plant functional types to changes in the fire regime in Mediterranean ecosystems: A simulation approach. Journal of Vegetation Science 10, 717-711. PAVÓN, N.P., HERNÁNDEZ-TREJO, H. AND RICO-GRAY, V. 2000. Distribution of plant life forms along an altitudinal gradient in the semi-arid valley of Zapotitlán, Mexico. Journal of Vegetation Science 11, 39-42. PIMM, S.L., RUSSEL, G.J., GITTLEMAN, J.L. AND BROOKS, T.M. 1995. The future of biodiversity. Science 269, 347-350. PODANI, J. 2001. SYN-TAX 2000 Computer programs for data analysis in ecology and systematics. Scientia publishing, Budapest. PODANI, J., CZÁRÁN, T. AND BARTHA, S. 1993. Pattern, area and diversity: the importance of spatial scale in species assemblage. Abstracta Botanica 17, 37-51. POTTS, S.G., VULLIAMY, B., DAFNI, A., NE’EMAN, G., O’TOOLE, C., ROBERTS, S. AND WILLMER, P. 2003. Response of plant-pollinator communities to fire: changes in diversity, abundance and floral reward structure. OIKOS 101, 103-112. PRIEUR-RICHARD, A-H, LAVOREL, S., GRIGULIS, K. AND DOS SANTOS, A. 2000. Plant community diversity and invasibility by exotics: invasion of Mediterranean old fields by Conyza bonariensis and Conyza canadensis. Ecology Letters 3, 412-422. PROCHEŞ, Ş., COWLING, R.M. AND DU PREEZ, D.R. 2005. Patterns of geophyte diversity and storage organ size in the winter-rainfall region of southern Africa. Diversity and Distributions 11, 101-109. PROCHEŞ, Ş., COWLING, R.M., GOLDBLATT, P., MANNING, J.C. AND SNIJMAN, D.A. 2006. An overview of the Cape geophytes. Biological Journal of the Linnean Society. 87, 27-43. PROCHEŞ, Ş., COWLING, R.M. AND MUCINA, L. 2003. Species-area curves based on relevé data for the Cape Floristic Region. South African Journal of Science 99, 74- 476. QUINN, G.P., KEOUGH, M.J. 2002. Experimental design and data analysis for biologists. Cambridge University Press, Cambridge. RAUNKIAER, C. 1934. The life forms of plants and statistical plant geography. Oxford University Press, Oxford. REBELO A.G. 2001. The state of plants in the Cape flora. In: Verdoorn, G.H. and Le Roux, J. (Eds). The state of South Africa’s species, pp. 18-43. Endangered Wildlife Trust, Pretoria. REBELO, A.G., BOUCHER, C., HELME, N., MUCINA, L. AND RUTHERFORD, M.C. 2006. Fynbos Biome. In: L. Mucina and M.C. Rutherford (Eds). The vegetation of South

206

Africa, Lesotho and Swaziland. Strelitzia 19, pp. 52-219. South African National Biodiversity Institute, Pretoria. RICHARDSON, D.M., COWLING, R.M., BOND, W.J., STOCK, W.D. AND DAVIS, G.W. 1995. Links between biodiversity and ecosystem function in the Cape Floristic Region. In: G.W. Davis and D.M. Richardson (Eds). Mediterranean-type ecosystems: The function of biodiversity, pp. 285-333. Springer-Verlag, Berlin, Heidelberg. ROUGET, M., JONAS, Z., COWLING, R.M., DESMET, P.G., DRIVER, A., MOHAMED, B., MUCINA, L., RUTHERFORD, M.C. AND POWRIE, L.W. 2006. Ecosystem status and protection levels of vegetation types. In: L. Mucina and M.C. Rutherford (Eds). The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19, pp. 724-737. South African National Biodiversity Institute, Pretoria. ROUX, P.W. 1963. The descending point method of vegetation survey. A point-sampling method for the measurement of semi-open grasslands and Karoo vegetation in South Africa. South African Journal of Agricultural Science 6, 273-288. RUBIN, F. 1998. The physical environment and major plant communities of the Tankwa Karoo National Park. Koedoe 41, 61-94. RUBIDGE, B.S. AND HANCOX, P.J. 1999. The Karoo – a palaeontological wonderland. In: M.J. Viljoen and W.U. Reimold (Eds). An introduction to South Africa’s geological and mining heritage, pp. 83-91. Published by the Geological Society of South Africa and Mintek. RUTHERFORD, M.C. AND WESTFALL, R.H. 1986. Biomes of Southern Africa. An objective characterisation. Memoirs of the Botanical Survey of South Africa 54, 1-98. RUTHERFORD, M.C. AND WESTFALL, R.H. 1994. Biomes of Southern Africa. An objective characterisation. Memoirs of the Botanical Survey of South Africa 63, 1-94. RUTHERFORD, M.C., MUCINA, L. AND POWRIE, L.W. 2003. Nama-Karoo veld types revisited: a numerical analysis of original Acocks’ field data. South African Journal of Botany 69, 52-61. RUTHERFORD, M.C., MUCINA, L. AND POWRIE, L.W. 2006. Biomes and bioregions of southern Africa. In: L. Mucina and M.C. Rutherford (Eds). The vegetation of South Africa, Lesotho and Swaziland. Strelitzia 19, pp. 30-51. South African National Biodiversity Institute, Pretoria. SEMENOVA, G.V. AND VAN DER MAAREL, E. 2000. Plant functional types – a strategic perspective. Journal of Vegetation Science 11, 917-922. SCHEINER, S.M. 2003. Six types of species-area curves. Global Ecology and Biogeography 12, 441-447. SCHEINER, S.M. 2004. A mélange of curves – further dialogue about species-area relationships. Global Ecology and Biogeography 13, 479-484. SCHEINER, S.M., COX, S.B., WILLIG, M., MITTELBACH, G.G., OSENBERG, C. AND KASPARI, M. 2000. Species richness, species-area curves and Simpson’s paradox. Evolutionary Ecology Research 2, 791-802.

207

SCHEINER, S.M. AND JONES, S. 2002. Diversity, productivity and scale in Wisconsin vegetation. Evolutionary Ecology Research 4, 1097-1117. SCHEINER, S.M. AND REY-BENAYAS, J.M. 1994. Global patterns of plant diversity. Evolutionary Ecology 8, 331-347. SCHULZE, R.E. 1997. South African Atlas of Agrohydrology and – Climatology. Water Research Commission, Pretoria, Report TT82/96. SCHWILK, D.W., KEELEY, J.E. AND BOND, W.J. 1997. The intermediate disturbance hypothesis does not explain fire and diversity pattern in fynbos. Plant Ecology 132, 77-84. SEMENOVA, G.V. AND VAN DER MAAREL, E. 2000. Plant functional types – a strategic perspective. Journal of Vegetation Science 11, 917-922. SHEARING, D. 1997. Karoo: South African Wildflower Guide 6. National Book Printers, Drukkery street, Goodwood, Western Cape. SHMIDA, A. 1984. Whittaker’s plant diversity sampling method. Israel Journal of Botany 33, 41-46. SHMIDA, A. AND WILSON, M.V. 1985. Biological determinants of species diversity. Journal of Biogeography 12, 1-20. SMITH, G.F., HOBSON, S.R., MEYER, N.L., CHESSELET, P., ARCHER, R.H., BURGOYNE, P.M., GLEN, H.F., HERMAN, P.P.J., RETIEF, E., SMITHIES S.J., VAN JAARSVELD, E.J. AND WELMAN, W.G. 1993. Southern African succulent plants – an updated synopsis. Aloe 30, 32-74. SMITH, T.M., SHUGART, H.H. AND WOODWARD, F.I. (Eds). Plant functional types: their relevance to ecosystem properties and global change. Cambridge University Press, Cambridge. SNIJMAN, D. AND PERRY, P. 1987. A floristic analysis of the Nieuwoudtville Wildflower Reserve, north-western Cape. South African Journal of Botany 53, 445-454. SOUTH AFRICAN NATION BIODIVERSITY INSTITUTE, 2006. hppt://www.sanbi.org/consfarm/ (accessed October 2006). SQUIERS, E.R. AND WISTENDAHL, W.A. 1976. Changes in plant species diversity during early secondary succession in an experimental old-field system. The American Midland Naturalist 98, 11-21. STEYN, H.M., VAN ROOYEN, N., VAN ROOYEN, M.W. 1996. The prediction of phenological stages in four Namaqualand ephemeral species using thermal unit indices. Israel Journal of Plant Sciences 44, 147-160. STIRLING, G. AND WILSEY, B. 2001. Empirical relationships between species richness, evenness, and proportional diversity. The American Naturalist 158, 286-299. STOHLGREN, T.J., FALKNER, M.B. AND SCHELL, L.D. 1995. A modified-Whittaker nested vegetation sampling method. Vegetatio 117, 113-121.

208

STROMBERG, M.R. AND GRIFFIN, J.R. 1996. Long-term patterns in coastal California grasslands in relation to cultivation, gophers, and grazing. Ecological applications 6, 1189-1211. TILMAN, D., BOND, W.J., CAMPBELL, B.M., KRUGER, F.J., LINDER, H.P., SCHOLZ, A., TAYLOR, H.C. AND WITTER, M. 1983. Origin and maintenance of plant species diversity. In: J.D. Day (Ed.) Mineral nutrients in Mediterranean ecosystems. South African National Scientific Programmes. Report No. 71. Council for Scientific and Industrial Research, Pretoria. TJØRVE, E. 2003. Shapes and functions of species-area curves: a review of possible models. Journal of Biogeography 30, 827-835. TRABAUD, L. AND LEPART, J. 1980. Diversity and stability in garrigue ecosystems after fire. Vegetatio 43, 49-57. TUNNELL, S.J., ENGLE, D.M. AND JORGENSEN, E.E. 2004. Old-field grassland successional dynamics following cessation of chronic disturbance. Journal of Vegetation Science 15, 431-436. UGLAND, K.I., GRAY, J.S. AND ELLINGSEN, K.E. 2003. The species-accumulation curve and estimation of species richness. Journal of Animal Ecology 72, 888-897. VAN DER MERWE, H., VAN ROOYEN, M.W. AND VAN ROOYEN, N. 2008a. Vegetation of the Hantam-Tanqua-Roggeveld subregion, South Africa. Part 1. Fynbos Biome related vegetation. Koedoe 50, 61-71. VAN DER MERWE, H., VAN ROOYEN, M.W. AND VAN ROOYEN, N. 2008b. Vegetation of the Hantam-Tanqua-Roggeveld subregion, South Africa. Part 2. Succulent Karoo Biome related vegetation. Koedoe 50, 160-183. VAN DER MERWE, P.J. 1938. Die trekboer in die geskiedenis van die Kaapkolonie (1657- 1842). Nasionale Pers Beperk, Kaapstad. VAN DER MERWE, P.J. 1988. Die noordwaartse beweging van die boere voor die Groottrek (1770-1842). The State Library, Pretoria. VAN DER PUTTEN, W.H., MORTIMER, S.R., HEDLUND, K., VAN DIJK, C., BROWN, V.K., LEPŠ, J., RODRIGUES-BARRUECO, C., ROY, J., DIAZ LEN, T.A., GROMSEN, D., KORTHALS, G.W., LAVOREL, S., SANTA REGINA, I. AND SMILAUER, P. 2000. Plant species diversity as a driver of early succession in abandoned fields: a multi- site approach. Oecologia 124, 91-99. VANDVIK, V. AND BIRKS, H.J.B. 2002. Pattern and process in Norwegian upland grasslands: a functional analysis. Journal of Vegetation Science 13, 123-134. VAN JAARSVELD, E. 1987. The succulent riches of South Africa and Namibia. Aloe 24,45- 92. VAN ROOYEN, M.W. 1999. Functional aspects of short-lived plants. In: W.R. DEAN AND S.J. MILTON (Eds). The Karoo: Ecological patterns and processes, pp. 107-122. Cambridge University Press, Cambridge.

209

VAN ROOYEN, M.W. 2002. Management of the old field vegetation in the Namaqua National Park, South Africa: conflicting demands of conservation and tourism. Geographical Journal 168: 211-223. VAN ROOYEN, M.W., THERON, G.K. AND GROBBELAAR, N. 1990. Life form and dispersal spectra of the flora of Namaqualand, South Africa. Journal of Arid Environments 19, 133-145. VAN WYK, A.E. AND SMITH, G.F. (Eds) 2001. Regions of Floristic Endemism in Southern Africa: A review with emphasis on succulents, pp. 1-199. Umdaus Press, Pretoria. VEECH, J.A. 2000. Choice of species-area function affects identification of hotspots. Conservation Biology 14, 140-147. VITOUSEK, P.M., D’ANTONIO, C.M., LOOPE, L.L. AND WESTBROOKS, R. 1996. Biological invasions as global environmental change. American Scientist 84, 468-478. VORSTER, M. 1982. The development of the ecological index method for assessing veld condition in the Karoo. Proceedings of the Grassland Society of Southern Africa 17, 84-89. WEATHER BUREAU 1998. Climate of South Africa. Climate statistics up to 1990. WB 42. Government Printer, Pretoria. WEIMARCK, H. 1941. Phytogeographical groups, centres and intervals within the Cape flora. Lunds Universitets Årsskrif Avd. 2. 37, 1-143. WERGER, M.J.A. 1974. On concepts and techniques applied in the Zürich-Montpellier method of vegetation survey. Bothalia 11, 309-323. WERGER, M.J.A. 1986. The Karoo and southern Kalahari. In: M. Evenari, I. Noy-Meir and D.W. Goodall (Eds). Hot deserts and arid shrublands, pp. 283-359. Elsevier, Amsterdam. WESTOBY, M. 1980. Elements of a theory of vegetation dynamics in arid rangelands. Israel Journal of Botany 28, 169-194. WHITTAKER, R.H. 1977. Evolution of species diversity on land communities. Evolutionary Biology 10, 1-67. WHITTAKER, R.J., WILLIS, K.J. AND FIELD, R. 2001. Scale and species richness: towards a general, hierarchical theory of species diversity. Journal of Biogeography 28, 453- 470. WILLIAMS, C.B. 1964. Patterns in the balance of nature. Academic Press, New York. WILSEY, B.J. AND POTVIN, C. 2000. Biodiversity and ecosystem functioning: importance of species evenness in an old field. Ecology 81, 887-892. WILSON, M.V. AND SHMIDA, A. 1984. Measuring Beta diversity with presence-absence data. Journal of Ecology 72, 1055-1064. WITHROW, A.P. 1932. Life forms and leaf size classes of certain plant communities of the Cincinnati region. Ecology 13, 12-35.

210

ZHAN, X., LI, L. AND CHENG, W. 2007. Restoration of Stipa kryloviisteppes in inner Mongolia of China: assessment of seed banks and vegetation composition. Journal of Arid Environments 68, 298-307. ZHANG, J-T. 2005. Succession analysis of plant communities in abandoned croplands in the eastern Loess plateau of China. Journal of Arid Environments 63, 458-474. ZIMMERMANN, H.G. 1991. Biological control of mesquite, Prosopis spp. (Fabaceae), in South Africa. Agriculture, Ecosystems and Environment 37, 175-186.

211

Appendix 1. Definitions of the life forms used in the study follows an adaptation of Raunkiaer’s life forms (1934), modified from Mueller-Dombois and Ellenberg (1974)

Life form General description Phanerophyte Plants that grow taller than 50 cm, or whose shoots die back periodically to that height limit. Chamaephyte Plants whose mature branch or shoot system remains perennially within 50 cm above ground surface, or plants that grow taller than 50 cm, but those shoots die back periodically to that height limit. Another typical habit is sprawling along the ground. Hemicryptophyte Perennial (including biennial) herbaceous plants with periodic shoot reduction to a remnant shoot system that lies relatively flat on the ground surface. Typically herbaceous throughout but can show some secondary lignification, particularly when standing as a dead remnant. Cryptophyte Perennial (including biennial) herbaceous plants with periodic shoot (Geophytes) reduction of the complete shoot system to storage organs that are imbedded in the soil. Therophyte Annuals. Plants whose shoot and root systems die after seed production and which complete their whole life cycle within one year. Liana Plants that grow by supporting themselves on others. Plants that germinate on the ground and maintain contact with the soil. Vascular semi-parasites Green vascular plants growing attached to other living autotrophic plants. Vascular parasites Heterotrophic plants, vascular plants growing on living plants.

212

Appendix 2. SAS output using the General Linear Model (GLM) Procedure on associations and life form (species level and cover level)

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 1 (G01-R2b) : PROC GLM with CLASS varbs of ASSOC & LFORM and MODEL of TLFORMS TLFORMC = ASSOC¦LFORM from data set MWVRA 13:00 Monday, June 8, 2009

The GLM Procedure

Class Level Information

Class Levels Values

ASSOC 8 Assoc-1 Assoc-2 Assoc-3 Assoc-4 Assoc-5 Assoc-6 Assoc-7 Assoc-8

LFORM 7 Chamaeph Cryptoph Hemicryp Liana Parasite Phanerop Therophy

Number of observations 280

213

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 2 (G01-R2b) : PROC GLM with CLASS varbs of ASSOC & LFORM and MODEL of TLFORMS TLFORMC = ASSOC¦LFORM from data set MWVRA 13:00 Monday, June 8, 2009

The GLM Procedure

Dependent Variable: TLFORMS

Sum of Source DF Squares Mean Square F Value Pr > F

Model 55 82.33434636 1.49698812 29.48 <.0001

Error 224 11.37415544 0.05077748

Corrected Total 279 93.70850180

R-Square Coeff Var Root MSE TLFORMS Mean

0.878622 25.26565 0.225339 0.891877

Source DF Type I SS Mean Square F Value Pr > F

ASSOC 7 0.82047551 0.11721079 2.31 0.0273 LFORM 6 75.09709099 12.51618183 246.49 <.0001 ASSOC*LFORM 42 6.41677986 0.15278047 3.01 <.0001

Source DF Type II SS Mean Square F Value Pr > F

ASSOC 7 0.82047551 0.11721079 2.31 0.0273 LFORM 6 75.09709099 12.51618183 246.49 <.0001 ASSOC*LFORM 42 6.41677986 0.15278047 3.01 <.0001

Source DF Type III SS Mean Square F Value Pr > F

ASSOC 7 0.82047551 0.11721079 2.31 0.0273 LFORM 6 58.13928735 9.68988122 190.83 <.0001 ASSOC*LFORM 42 6.41677986 0.15278047 3.01 <.0001

214

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 3 (G01-R2b) : PROC GLM with CLASS varbs of ASSOC & LFORM and MODEL of TLFORMS TLFORMC = ASSOC¦LFORM from data set MWVRA 13:00 Monday, June 8, 2009

The GLM Procedure

Dependent Variable: TLFORMC

Sum of Source DF Squares Mean Square F Value Pr > F

Model 55 89.5475298 1.6281369 23.44 <.0001

Error 224 15.5604760 0.0694664

Corrected Total 279 105.1080057

R-Square Coeff Var Root MSE TLFORMC Mean

0.851957 34.22121 0.263565 0.770180

Source DF Type I SS Mean Square F Value Pr > F

ASSOC 7 1.79438757 0.25634108 3.69 0.0009 LFORM 6 75.90779762 12.65129960 182.12 <.0001 ASSOC*LFORM 42 11.84534459 0.28203201 4.06 <.0001

Source DF Type II SS Mean Square F Value Pr > F

ASSOC 7 1.79438757 0.25634108 3.69 0.0009 LFORM 6 75.90779762 12.65129960 182.12 <.0001 ASSOC*LFORM 42 11.84534459 0.28203201 4.06 <.0001

Source DF Type III SS Mean Square F Value Pr > F

ASSOC 7 1.79438757 0.25634108 3.69 0.0009 LFORM 6 61.79560939 10.29926823 148.26 <.0001 ASSOC*LFORM 42 11.84534459 0.28203201 4.06 <.0001

215

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 4 (G01-R2b) : PROC GLM with CLASS varbs of ASSOC & LFORM and MODEL of TLFORMS TLFORMC = ASSOC¦LFORM from data set MWVRA 13:00 Monday, June 8, 2009

The GLM Procedure Least Squares Means

TLFORMS Standard LSMEAN ASSOC LSMEAN Error Pr > ¦t¦ Number

Assoc-1 0.89868721 0.03808917 <.0001 1 Assoc-2 0.94954058 0.02693311 <.0001 2 Assoc-3 0.90234715 0.06022427 <.0001 3 Assoc-4 0.96121004 0.04258499 <.0001 4 Assoc-5 0.87039268 0.02838999 <.0001 5 Assoc-6 0.86581711 0.04917291 <.0001 6 Assoc-7 0.79858925 0.04258499 <.0001 7 Assoc-8 0.80378893 0.04917291 <.0001 8

Least Squares Means for effect ASSOC Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMS i/j 1 2 3 4 5 6 7 8

1 0.2768 0.9591 0.2750 0.5520 0.5977 0.0811 0.1285 2 0.2768 0.4751 0.8171 0.0443 0.1368 0.0030 0.0100 3 0.9591 0.4751 0.4257 0.6317 0.6389 0.1609 0.2062 4 0.2750 0.8171 0.4257 0.0773 0.1439 0.0075 0.0163 5 0.5520 0.0443 0.6317 0.0773 0.9358 0.1620 0.2420 6 0.5977 0.1368 0.6389 0.1439 0.9358 0.3025 0.3734 7 0.0811 0.0030 0.1609 0.0075 0.1620 0.3025 0.9364 8 0.1285 0.0100 0.2062 0.0163 0.2420 0.3734 0.9364

TLFORMC Standard LSMEAN ASSOC LSMEAN Error Pr > ¦t¦ Number

Assoc-1 0.83373591 0.04455058 <.0001 1 Assoc-2 0.83709578 0.03150202 <.0001 2 Assoc-3 0.87766656 0.07044066 <.0001 3 Assoc-4 0.85070622 0.04980907 <.0001 4 Assoc-5 0.65650312 0.03320605 <.0001 5

216

Assoc-6 0.70765762 0.05751456 <.0001 6 Assoc-7 0.72008634 0.04980907 <.0001 7 Assoc-8 0.73251506 0.05751456 <.0001 8

217

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 5 (G01-R2b) : PROC GLM with CLASS varbs of ASSOC & LFORM and MODEL of TLFORMS TLFORMC = ASSOC¦LFORM from data set MWVRA 13:00 Monday, June 8, 2009

The GLM Procedure Least Squares Means

Least Squares Means for effect ASSOC Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMC

i/j 1 2 3 4 5 6 7 8

1 0.9510 0.5987 0.7998 0.0016 0.0845 0.0904 0.1655 2 0.9510 0.5996 0.8176 0.0001 0.0496 0.0483 0.1122 3 0.5987 0.5996 0.7549 0.0049 0.0629 0.0691 0.1119 4 0.7998 0.8176 0.7549 0.0014 0.0614 0.0650 0.1217 5 0.0016 0.0001 0.0049 0.0014 0.4420 0.2893 0.2536 6 0.0845 0.0496 0.0629 0.0614 0.4420 0.8704 0.7602 7 0.0904 0.0483 0.0691 0.0650 0.2893 0.8704 0.8704 8 0.1655 0.1122 0.1119 0.1217 0.2536 0.7602 0.8704

NOTE: To ensure overall protection level, only probabilities associated with pre-planned comparisons should be used.

TLFORMS Standard LSMEAN LFORM LSMEAN Error Pr > ¦t¦ Number

Chamaeph 1.50522443 0.04060179 <.0001 1 Cryptoph 1.40205641 0.04060179 <.0001 2 Hemicryp 1.01781788 0.04060179 <.0001 3 Liana 0.42142122 0.04060179 <.0001 4 Parasite 0.01739224 0.04060179 0.6688 5 Phanerop 0.54926133 0.04060179 <.0001 6 Therophy 1.25590281 0.04060179 <.0001 7

Least Squares Means for effect LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMS

i/j 1 2 3 4 5 6 7

218

1 0.0737 <.0001 <.0001 <.0001 <.0001 <.0001 2 0.0737 <.0001 <.0001 <.0001 <.0001 0.0116 3 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 4 <.0001 <.0001 <.0001 <.0001 0.0270 <.0001 5 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 6 <.0001 <.0001 <.0001 0.0270 <.0001 <.0001 7 <.0001 0.0116 <.0001 <.0001 <.0001 <.0001

TLFORMC Standard LSMEAN LFORM LSMEAN Error Pr > ¦t¦ Number

Chamaeph 1.63414432 0.04748943 <.0001 1 Cryptoph 1.07586243 0.04748943 <.0001 2

219

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 6 (G01-R2b) : PROC GLM with CLASS varbs of ASSOC & LFORM and MODEL of TLFORMS TLFORMC = ASSOC¦LFORM from data set MWVRA 13:00 Monday, June 8, 2009

The GLM Procedure Least Squares Means

TLFORMC Standard LSMEAN LFORM LSMEAN Error Pr > ¦t¦ Number

Hemicryp 0.55127493 0.04748943 <.0001 3 Liana 0.17438228 0.04748943 0.0003 4 Parasite 0.00645210 0.04748943 0.8921 5 Phanerop 0.80311690 0.04748943 <.0001 6 Therophy 1.19373782 0.04748943 <.0001 7

Least Squares Means for effect LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMC

i/j 1 2 3 4 5 6 7

1 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 2 <.0001 <.0001 <.0001 <.0001 <.0001 0.0806 3 <.0001 <.0001 <.0001 <.0001 0.0002 <.0001 4 <.0001 <.0001 <.0001 0.0131 <.0001 <.0001 5 <.0001 <.0001 <.0001 0.0131 <.0001 <.0001 6 <.0001 <.0001 0.0002 <.0001 <.0001 <.0001 7 <.0001 0.0806 <.0001 <.0001 <.0001 <.0001

NOTE: To ensure overall protection level, only probabilities associated with pre-planned comparisons should be used.

TLFORMS Standard LSMEAN ASSOC LFORM LSMEAN Error Pr > ¦t¦ Number

Assoc-1 Chamaeph 1.40579972 0.10077448 <.0001 1 Assoc-1 Cryptoph 1.52910877 0.10077448 <.0001 2 Assoc-1 Hemicryp 1.02342810 0.10077448 <.0001 3 Assoc-1 Liana 0.24213508 0.10077448 0.0171 4 Assoc-1 Parasite 0.06925151 0.10077448 0.4927 5 Assoc-1 Phanerop 0.57687011 0.10077448 <.0001 6

220

Assoc-1 Therophy 1.44421721 0.10077448 <.0001 7 Assoc-2 Chamaeph 1.42305156 0.07125832 <.0001 8 Assoc-2 Cryptoph 1.49284872 0.07125832 <.0001 9 Assoc-2 Hemicryp 0.95332723 0.07125832 <.0001 10 Assoc-2 Liana 0.50271622 0.07125832 <.0001 11 Assoc-2 Parasite 0.06988639 0.07125832 0.3278 12 Assoc-2 Phanerop 0.76663244 0.07125832 <.0001 13 Assoc-2 Therophy 1.43832148 0.07125832 <.0001 14 Assoc-3 Chamaeph 1.55128198 0.15933844 <.0001 15 Assoc-3 Cryptoph 1.44509020 0.15933844 <.0001 16 Assoc-3 Hemicryp 0.93252574 0.15933844 <.0001 17 Assoc-3 Liana 0.32812797 0.15933844 0.0406 18 Assoc-3 Parasite 0.00000000 0.15933844 1.0000 19 Assoc-3 Phanerop 0.62991484 0.15933844 0.0001 20 Assoc-3 Therophy 1.42948930 0.15933844 <.0001 21

221

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 7 (G01-R2b) : PROC GLM with CLASS varbs of ASSOC & LFORM and MODEL of TLFORMS TLFORMC = ASSOC¦LFORM from data set MWVRA 13:00 Monday, June 8, 2009

The GLM Procedure Least Squares Means

TLFORMS Standard LSMEAN ASSOC LFORM LSMEAN Error Pr > ¦t¦ Number

Assoc-4 Chamaeph 1.54448545 0.11266929 <.0001 22 Assoc-4 Cryptoph 1.40983187 0.11266929 <.0001 23 Assoc-4 Hemicryp 0.93861361 0.11266929 <.0001 24 Assoc-4 Liana 0.65186376 0.11266929 <.0001 25 Assoc-4 Parasite 0.00000000 0.11266929 1.0000 26 Assoc-4 Phanerop 0.82446803 0.11266929 <.0001 27 Assoc-4 Therophy 1.35920757 0.11266929 <.0001 28 Assoc-5 Chamaeph 1.42540218 0.07511286 <.0001 29 Assoc-5 Cryptoph 1.43693234 0.07511286 <.0001 30 Assoc-5 Hemicryp 0.89296187 0.07511286 <.0001 31 Assoc-5 Liana 0.46675505 0.07511286 <.0001 32 Assoc-5 Parasite 0.00000000 0.07511286 1.0000 33 Assoc-5 Phanerop 0.39429436 0.07511286 <.0001 34 Assoc-5 Therophy 1.47640293 0.07511286 <.0001 35 Assoc-6 Chamaeph 1.48946295 0.13009929 <.0001 36 Assoc-6 Cryptoph 1.38353060 0.13009929 <.0001 37 Assoc-6 Hemicryp 1.05927906 0.13009929 <.0001 38 Assoc-6 Liana 0.34376881 0.13009929 0.0088 39 Assoc-6 Parasite 0.00000000 0.13009929 1.0000 40 Assoc-6 Phanerop 0.27551096 0.13009929 0.0353 41 Assoc-6 Therophy 1.50916739 0.13009929 <.0001 42 Assoc-7 Chamaeph 1.66395649 0.11266929 <.0001 43 Assoc-7 Cryptoph 1.02295968 0.11266929 <.0001 44 Assoc-7 Hemicryp 1.17086599 0.11266929 <.0001 45 Assoc-7 Liana 0.17134930 0.11266929 0.1297 46 Assoc-7 Parasite 0.00000000 0.11266929 1.0000 47 Assoc-7 Phanerop 0.56533860 0.11266929 <.0001 48 Assoc-7 Therophy 0.99565470 0.11266929 <.0001 49 Assoc-8 Chamaeph 1.53835509 0.13009929 <.0001 50 Assoc-8 Cryptoph 1.49614910 0.13009929 <.0001 51 Assoc-8 Hemicryp 1.17154145 0.13009929 <.0001 52 Assoc-8 Liana 0.66465360 0.13009929 <.0001 53 Assoc-8 Parasite 0.00000000 0.13009929 1.0000 54 Assoc-8 Phanerop 0.36106133 0.13009929 0.0060 55 Assoc-8 Therophy 0.39476191 0.13009929 0.0027 56

222

223

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 8 (G01-R2b) : PROC GLM with CLASS varbs of ASSOC & LFORM and MODEL of TLFORMS TLFORMC = ASSOC¦LFORM from data set MWVRA 13:00 Monday, June 8, 2009

The GLM Procedure Least Squares Means

Least Squares Means for effect ASSOC*LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMS i/j 1 2 3 4 5 6 7 8 9 10 11 12 13 14

1 0.3878 0.0078 <.0001 <.0001 <.0001 0.7877 0.8890 0.4814 0.0003 <.0001 <.0001 <.0001 0.7924 2 0.3878 0.0005 <.0001 <.0001 <.0001 0.5520 0.3911 0.7692 <.0001 <.0001 <.0001 <.0001 0.4628 3 0.0078 0.0005 <.0001 <.0001 0.0020 0.0035 0.0014 0.0002 0.5706 <.0001 <.0001 0.0386 0.0009 4 <.0001 <.0001 <.0001 0.2264 0.0197 <.0001 <.0001 <.0001 <.0001 0.0359 0.1642 <.0001 <.0001 5 <.0001 <.0001 <.0001 0.2264 0.0004 <.0001 <.0001 <.0001 <.0001 0.0005 0.9959 <.0001 <.0001 6 <.0001 <.0001 0.0020 0.0197 0.0004 <.0001 <.0001 <.0001 0.0026 0.5486 <.0001 0.1256 <.0001 7 0.7877 0.5520 0.0035 <.0001 <.0001 <.0001 0.8640 0.6939 <.0001 <.0001 <.0001 <.0001 0.9619 8 0.8890 0.3911 0.0014 <.0001 <.0001 <.0001 0.8640 0.4893 <.0001 <.0001 <.0001 <.0001 0.8797 9 0.4814 0.7692 0.0002 <.0001 <.0001 <.0001 0.6939 0.4893 <.0001 <.0001 <.0001 <.0001 0.5890 10 0.0003 <.0001 0.5706 <.0001 <.0001 0.0026 <.0001 <.0001 <.0001 <.0001 <.0001 0.0653 <.0001 11 <.0001 <.0001 <.0001 0.0359 0.0005 0.5486 <.0001 <.0001 <.0001 <.0001 <.0001 0.0094 <.0001 12 <.0001 <.0001 <.0001 0.1642 0.9959 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 13 <.0001 <.0001 0.0386 <.0001 <.0001 0.1256 <.0001 <.0001 <.0001 0.0653 0.0094 <.0001 <.0001 14 0.7924 0.4628 0.0009 <.0001 <.0001 <.0001 0.9619 0.8797 0.5890 <.0001 <.0001 <.0001 <.0001 15 0.4411 0.9065 0.0056 <.0001 <.0001 <.0001 0.5707 0.4633 0.7381 0.0007 <.0001 <.0001 <.0001 0.5182 16 0.8351 0.6563 0.0263 <.0001 <.0001 <.0001 0.9963 0.8996 0.7846 0.0053 <.0001 <.0001 0.0001 0.9691 17 0.0128 0.0018 0.6302 0.0003 <.0001 0.0605 0.0072 0.0054 0.0015 0.9052 0.0146 <.0001 0.3429 0.0041 18 <.0001 <.0001 0.0003 0.6487 0.1711 0.1884 <.0001 <.0001 <.0001 0.0004 0.3183 0.1404 0.0127 <.0001 19 <.0001 <.0001 <.0001 0.2004 0.7137 0.0025 <.0001 <.0001 <.0001 <.0001 0.0044 0.6893 <.0001 <.0001 20 <.0001 <.0001 0.0380 0.0409 0.0033 0.7787 <.0001 <.0001 <.0001 0.0652 0.4669 0.0015 0.4343 <.0001 21 0.9001 0.5977 0.0323 <.0001 <.0001 <.0001 0.9378 0.9706 0.7170 0.0069 <.0001 <.0001 0.0002 0.9597 22 0.3599 0.9191 0.0007 <.0001 <.0001 <.0001 0.5078 0.3633 0.6989 <.0001 <.0001 <.0001 <.0001 0.4267 23 0.9787 0.4309 0.0112 <.0001 <.0001 <.0001 0.8203 0.9211 0.5341 0.0007 <.0001 <.0001 <.0001 0.8310 24 0.0023 0.0001 0.5753 <.0001 <.0001 0.0175 0.0010 0.0003 <.0001 0.9122 0.0012 <.0001 0.1984 0.0002 25 <.0001 <.0001 0.0147 0.0072 0.0002 0.6203 <.0001 <.0001 <.0001 0.0247 0.2644 <.0001 0.3902 <.0001 26 <.0001 <.0001 <.0001 0.1106 0.6473 0.0002 <.0001 <.0001 <.0001 <.0001 0.0002 0.6006 <.0001 <.0001 27 0.0002 <.0001 0.1894 0.0002 <.0001 0.1028 <.0001 <.0001 <.0001 0.3348 0.0166 <.0001 0.6648 <.0001 28 0.7582 0.2622 0.0273 <.0001 <.0001 <.0001 0.5744 0.6325 0.3172 0.0026 <.0001 <.0001 <.0001 0.5535 29 0.8762 0.4102 0.0016 <.0001 <.0001 <.0001 0.8811 0.9819 0.5154 <.0001 <.0001 <.0001 <.0001 0.9008 30 0.8046 0.4641 0.0012 <.0001 <.0001 <.0001 0.9538 0.8935 0.5897 <.0001 <.0001 <.0001 <.0001 0.9893 31 <.0001 <.0001 0.3004 <.0001 <.0001 0.0126 <.0001 <.0001 <.0001 0.5605 0.0002 <.0001 0.2237 <.0001

224

32 <.0001 <.0001 <.0001 0.0753 0.0018 0.3819 <.0001 <.0001 <.0001 <.0001 0.7287 0.0002 0.0041 <.0001 33 <.0001 <.0001 <.0001 0.0553 0.5822 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 0.5004 <.0001 <.0001 34 <.0001 <.0001 <.0001 0.2273 0.0103 0.1477 <.0001 <.0001 <.0001 <.0001 0.2961 0.0020 0.0004 <.0001 35 0.5749 0.6754 0.0004 <.0001 <.0001 <.0001 0.7981 0.6069 0.8739 <.0001 <.0001 <.0001 <.0001 0.7134 36 0.6117 0.8098 0.0050 <.0001 <.0001 <.0001 0.7836 0.6548 0.9818 0.0004 <.0001 <.0001 <.0001 0.7306 37 0.8925 0.3773 0.0297 <.0001 <.0001 <.0001 0.7126 0.7902 0.4619 0.0041 <.0001 <.0001 <.0001 0.7122 38 0.0363 0.0047 0.8277 <.0001 <.0001 0.0037 0.0202 0.0150 0.0038 0.4758 0.0002 <.0001 0.0497 0.0113 39 <.0001 <.0001 <.0001 0.5375 0.0967 0.1580 <.0001 <.0001 <.0001 <.0001 0.2851 0.0662 0.0048 <.0001 40 <.0001 <.0001 <.0001 0.1426 0.6743 0.0006 <.0001 <.0001 <.0001 <.0001 0.0008 0.6380 <.0001 <.0001 41 <.0001 <.0001 <.0001 0.8395 0.2114 0.0684 <.0001 <.0001 <.0001 <.0001 0.1270 0.1671 0.0011 <.0001 42 0.5306 0.9037 0.0035 <.0001 <.0001 <.0001 0.6935 0.5621 0.9125 0.0002 <.0001 <.0001 <.0001 0.6334 43 0.0891 0.3733 <.0001 <.0001 <.0001 <.0001 0.1474 0.0721 0.2006 <.0001 <.0001 <.0001 <.0001 0.0919 44 0.0120 0.0010 0.9975 <.0001 <.0001 0.0035 0.0058 0.0030 0.0005 0.6020 0.0001 <.0001 0.0558 0.0021 45 0.1216 0.0186 0.3304 <.0001 <.0001 0.0001 0.0719 0.0598 0.0165 0.1041 <.0001 <.0001 0.0027 0.0460 46 <.0001 <.0001 <.0001 0.6400 0.5001 0.0078 <.0001 <.0001 <.0001 <.0001 0.0137 0.4474 <.0001 <.0001

225

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 9 (G01-R2b) : PROC GLM with CLASS varbs of ASSOC & LFORM and MODEL of TLFORMS TLFORMC = ASSOC¦LFORM from data set MWVRA 13:00 Monday, June 8, 2009

The GLM Procedure Least Squares Means

Least Squares Means for effect ASSOC*LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMS i/j 15 16 17 18 19 20 21 22 23 24 25 26 27 28

1 0.4411 0.8351 0.0128 <.0001 <.0001 <.0001 0.9001 0.3599 0.9787 0.0023 <.0001 <.0001 0.0002 0.7582 2 0.9065 0.6563 0.0018 <.0001 <.0001 <.0001 0.5977 0.9191 0.4309 0.0001 <.0001 <.0001 <.0001 0.2622 3 0.0056 0.0263 0.6302 0.0003 <.0001 0.0380 0.0323 0.0007 0.0112 0.5753 0.0147 <.0001 0.1894 0.0273 4 <.0001 <.0001 0.0003 0.6487 0.2004 0.0409 <.0001 <.0001 <.0001 <.0001 0.0072 0.1106 0.0002 <.0001 5 <.0001 <.0001 <.0001 0.1711 0.7137 0.0033 <.0001 <.0001 <.0001 <.0001 0.0002 0.6473 <.0001 <.0001 6 <.0001 <.0001 0.0605 0.1884 0.0025 0.7787 <.0001 <.0001 <.0001 0.0175 0.6203 0.0002 0.1028 <.0001 7 0.5707 0.9963 0.0072 <.0001 <.0001 <.0001 0.9378 0.5078 0.8203 0.0010 <.0001 <.0001 <.0001 0.5744 8 0.4633 0.8996 0.0054 <.0001 <.0001 <.0001 0.9706 0.3633 0.9211 0.0003 <.0001 <.0001 <.0001 0.6325 9 0.7381 0.7846 0.0015 <.0001 <.0001 <.0001 0.7170 0.6989 0.5341 <.0001 <.0001 <.0001 <.0001 0.3172 10 0.0007 0.0053 0.9052 0.0004 <.0001 0.0652 0.0069 <.0001 0.0007 0.9122 0.0247 <.0001 0.3348 0.0026 11 <.0001 <.0001 0.0146 0.3183 0.0044 0.4669 <.0001 <.0001 <.0001 0.0012 0.2644 0.0002 0.0166 <.0001 12 <.0001 <.0001 <.0001 0.1404 0.6893 0.0015 <.0001 <.0001 <.0001 <.0001 <.0001 0.6006 <.0001 <.0001 13 <.0001 0.0001 0.3429 0.0127 <.0001 0.4343 0.0002 <.0001 <.0001 0.1984 0.3902 <.0001 0.6648 <.0001 14 0.5182 0.9691 0.0041 <.0001 <.0001 <.0001 0.9597 0.4267 0.8310 0.0002 <.0001 <.0001 <.0001 0.5535 15 0.6379 0.0065 <.0001 <.0001 <.0001 0.5894 0.9722 0.4693 0.0019 <.0001 <.0001 0.0002 0.3261 16 0.6379 0.0239 <.0001 <.0001 0.0004 0.9449 0.6110 0.8568 0.0101 <.0001 <.0001 0.0017 0.6603 17 0.0065 0.0239 0.0079 <.0001 0.1807 0.0284 0.0019 0.0152 0.9751 0.1518 <.0001 0.5803 0.0298 18 <.0001 <.0001 0.0079 0.1467 0.1818 <.0001 <.0001 <.0001 0.0020 0.0985 0.0941 0.0117 <.0001 19 <.0001 <.0001 <.0001 0.1467 0.0056 <.0001 <.0001 <.0001 <.0001 0.0010 1.0000 <.0001 <.0001 20 <.0001 0.0004 0.1807 0.1818 0.0056 0.0005 <.0001 <.0001 0.1151 0.9105 0.0014 0.3199 0.0002 21 0.5894 0.9449 0.0284 <.0001 <.0001 0.0005 0.5563 0.9199 0.0126 <.0001 <.0001 0.0022 0.7191 22 0.9722 0.6110 0.0019 <.0001 <.0001 <.0001 0.5563 0.3990 0.0002 <.0001 <.0001 <.0001 0.2462 23 0.4693 0.8568 0.0152 <.0001 <.0001 <.0001 0.9199 0.3990 0.0034 <.0001 <.0001 0.0003 0.7510 24 0.0019 0.0101 0.9751 0.0020 <.0001 0.1151 0.0126 0.0002 0.0034 0.0733 <.0001 0.4745 0.0089 25 <.0001 <.0001 0.1518 0.0985 0.0010 0.9105 <.0001 <.0001 <.0001 0.0733 <.0001 0.2799 <.0001 26 <.0001 <.0001 <.0001 0.0941 1.0000 0.0014 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 27 0.0002 0.0017 0.5803 0.0117 <.0001 0.3199 0.0022 <.0001 0.0003 0.4745 0.2799 <.0001 0.0009 28 0.3261 0.6603 0.0298 <.0001 <.0001 0.0002 0.7191 0.2462 0.7510 0.0089 <.0001 <.0001 0.0009 29 0.4756 0.9111 0.0056 <.0001 <.0001 <.0001 0.9815 0.3801 0.9086 0.0004 <.0001 <.0001 <.0001 0.6254 30 0.5169 0.9631 0.0046 <.0001 <.0001 <.0001 0.9663 0.4279 0.8416 0.0003 <.0001 <.0001 <.0001 0.5666 31 0.0002 0.0020 0.8225 0.0015 <.0001 0.1368 0.0026 <.0001 0.0002 0.7363 0.0764 <.0001 0.6135 0.0007

226

32 <.0001 <.0001 0.0088 0.4321 0.0086 0.3553 <.0001 <.0001 <.0001 0.0006 0.1730 0.0007 0.0088 <.0001 33 <.0001 <.0001 <.0001 0.0638 1.0000 0.0004 <.0001 <.0001 <.0001 <.0001 <.0001 1.0000 <.0001 <.0001 34 <.0001 <.0001 0.0025 0.7076 0.0262 0.1824 <.0001 <.0001 <.0001 <.0001 0.0584 0.0040 0.0017 <.0001 35 0.6712 0.8591 0.0023 <.0001 <.0001 <.0001 0.7902 0.6156 0.6235 <.0001 <.0001 <.0001 <.0001 0.3877 36 0.7641 0.8294 0.0073 <.0001 <.0001 <.0001 0.7709 0.7495 0.6440 0.0016 <.0001 <.0001 0.0001 0.4499 37 0.4157 0.7650 0.0294 <.0001 <.0001 0.0003 0.8234 0.3507 0.8787 0.0104 <.0001 <.0001 0.0013 0.8877 38 0.0176 0.0620 0.5384 0.0005 <.0001 0.0380 0.0733 0.0052 0.0428 0.4840 0.0188 <.0001 0.1738 0.0828 39 <.0001 <.0001 0.0046 0.9395 0.0961 0.1656 <.0001 <.0001 <.0001 0.0007 0.0748 0.0470 0.0057 <.0001 40 <.0001 <.0001 <.0001 0.1121 1.0000 0.0025 <.0001 <.0001 <.0001 <.0001 0.0002 1.0000 <.0001 <.0001 41 <.0001 <.0001 0.0016 0.7983 0.1818 0.0863 <.0001 <.0001 <.0001 0.0002 0.0298 0.1108 0.0016 <.0001 42 0.8380 0.7557 0.0055 <.0001 <.0001 <.0001 0.6989 0.8376 0.5644 0.0011 <.0001 <.0001 <.0001 0.3845 43 0.5643 0.2633 0.0002 <.0001 <.0001 <.0001 0.2308 0.4542 0.1122 <.0001 <.0001 <.0001 <.0001 0.0571 44 0.0073 0.0316 0.6435 0.0005 <.0001 0.0452 0.0384 0.0012 0.0160 0.5971 0.0208 <.0001 0.2142 0.0359 45 0.0525 0.1613 0.2232 <.0001 <.0001 0.0060 0.1864 0.0199 0.1351 0.1464 0.0013 <.0001 0.0308 0.2385 46 <.0001 <.0001 0.0001 0.4226 0.3809 0.0197 <.0001 <.0001 <.0001 <.0001 0.0029 0.2834 <.0001 <.0001

227

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 10 (G01-R2b) : PROC GLM with CLASS varbs of ASSOC & LFORM and MODEL of TLFORMS TLFORMC = ASSOC¦LFORM from data set MWVRA 13:00 Monday, June 8, 2009

The GLM Procedure Least Squares Means

Least Squares Means for effect ASSOC*LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMS i/j 29 30 31 32 33 34 35 36 37 38 39 40 41 42

1 0.8762 0.8046 <.0001 <.0001 <.0001 <.0001 0.5749 0.6117 0.8925 0.0363 <.0001 <.0001 <.0001 0.5306 2 0.4102 0.4641 <.0001 <.0001 <.0001 <.0001 0.6754 0.8098 0.3773 0.0047 <.0001 <.0001 <.0001 0.9037 3 0.0016 0.0012 0.3004 <.0001 <.0001 <.0001 0.0004 0.0050 0.0297 0.8277 <.0001 <.0001 <.0001 0.0035 4 <.0001 <.0001 <.0001 0.0753 0.0553 0.2273 <.0001 <.0001 <.0001 <.0001 0.5375 0.1426 0.8395 <.0001 5 <.0001 <.0001 <.0001 0.0018 0.5822 0.0103 <.0001 <.0001 <.0001 <.0001 0.0967 0.6743 0.2114 <.0001 6 <.0001 <.0001 0.0126 0.3819 <.0001 0.1477 <.0001 <.0001 <.0001 0.0037 0.1580 0.0006 0.0684 <.0001 7 0.8811 0.9538 <.0001 <.0001 <.0001 <.0001 0.7981 0.7836 0.7126 0.0202 <.0001 <.0001 <.0001 0.6935 8 0.9819 0.8935 <.0001 <.0001 <.0001 <.0001 0.6069 0.6548 0.7902 0.0150 <.0001 <.0001 <.0001 0.5621 9 0.5154 0.5897 <.0001 <.0001 <.0001 <.0001 0.8739 0.9818 0.4619 0.0038 <.0001 <.0001 <.0001 0.9125 10 <.0001 <.0001 0.5605 <.0001 <.0001 <.0001 <.0001 0.0004 0.0041 0.4758 <.0001 <.0001 <.0001 0.0002 11 <.0001 <.0001 0.0002 0.7287 <.0001 0.2961 <.0001 <.0001 <.0001 0.0002 0.2851 0.0008 0.1270 <.0001 12 <.0001 <.0001 <.0001 0.0002 0.5004 0.0020 <.0001 <.0001 <.0001 <.0001 0.0662 0.6380 0.1671 <.0001 13 <.0001 <.0001 0.2237 0.0041 <.0001 0.0004 <.0001 <.0001 <.0001 0.0497 0.0048 <.0001 0.0011 <.0001 14 0.9008 0.9893 <.0001 <.0001 <.0001 <.0001 0.7134 0.7306 0.7122 0.0113 <.0001 <.0001 <.0001 0.6334 15 0.4756 0.5169 0.0002 <.0001 <.0001 <.0001 0.6712 0.7641 0.4157 0.0176 <.0001 <.0001 <.0001 0.8380 16 0.9111 0.9631 0.0020 <.0001 <.0001 <.0001 0.8591 0.8294 0.7650 0.0620 <.0001 <.0001 <.0001 0.7557 17 0.0056 0.0046 0.8225 0.0088 <.0001 0.0025 0.0023 0.0073 0.0294 0.5384 0.0046 <.0001 0.0016 0.0055 18 <.0001 <.0001 0.0015 0.4321 0.0638 0.7076 <.0001 <.0001 <.0001 0.0005 0.9395 0.1121 0.7983 <.0001 19 <.0001 <.0001 <.0001 0.0086 1.0000 0.0262 <.0001 <.0001 <.0001 <.0001 0.0961 1.0000 0.1818 <.0001 20 <.0001 <.0001 0.1368 0.3553 0.0004 0.1824 <.0001 <.0001 0.0003 0.0380 0.1656 0.0025 0.0863 <.0001 21 0.9815 0.9663 0.0026 <.0001 <.0001 <.0001 0.7902 0.7709 0.8234 0.0733 <.0001 <.0001 <.0001 0.6989 22 0.3801 0.4279 <.0001 <.0001 <.0001 <.0001 0.6156 0.7495 0.3507 0.0052 <.0001 <.0001 <.0001 0.8376 23 0.9086 0.8416 0.0002 <.0001 <.0001 <.0001 0.6235 0.6440 0.8787 0.0428 <.0001 <.0001 <.0001 0.5644 24 0.0004 0.0003 0.7363 0.0006 <.0001 <.0001 <.0001 0.0016 0.0104 0.4840 0.0007 <.0001 0.0002 0.0011 25 <.0001 <.0001 0.0764 0.1730 <.0001 0.0584 <.0001 <.0001 <.0001 0.0188 0.0748 0.0002 0.0298 <.0001 26 <.0001 <.0001 <.0001 0.0007 1.0000 0.0040 <.0001 <.0001 <.0001 <.0001 0.0470 1.0000 0.1108 <.0001 27 <.0001 <.0001 0.6135 0.0088 <.0001 0.0017 <.0001 0.0001 0.0013 0.1738 0.0057 <.0001 0.0016 <.0001 28 0.6254 0.5666 0.0007 <.0001 <.0001 <.0001 0.3877 0.4499 0.8877 0.0828 <.0001 <.0001 <.0001 0.3845 29 0.9137 <.0001 <.0001 <.0001 <.0001 0.6316 0.6702 0.7807 0.0156 <.0001 <.0001 <.0001 0.5777 30 0.9137 <.0001 <.0001 <.0001 <.0001 0.7106 0.7269 0.7226 0.0126 <.0001 <.0001 <.0001 0.6311 31 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 0.0013 0.2694 0.0003 <.0001 <.0001 <.0001

228

32 <.0001 <.0001 <.0001 <.0001 0.4959 <.0001 <.0001 <.0001 0.0001 0.4138 0.0021 0.2043 <.0001 33 <.0001 <.0001 <.0001 <.0001 0.0003 <.0001 <.0001 <.0001 <.0001 0.0230 1.0000 0.0680 <.0001 34 <.0001 <.0001 <.0001 0.4959 0.0003 <.0001 <.0001 <.0001 <.0001 0.7369 0.0093 0.4300 <.0001 35 0.6316 0.7106 <.0001 <.0001 <.0001 <.0001 0.9308 0.5371 0.0060 <.0001 <.0001 <.0001 0.8275 36 0.6702 0.7269 <.0001 <.0001 <.0001 <.0001 0.9308 0.5654 0.0203 <.0001 <.0001 <.0001 0.9148 37 0.7807 0.7226 0.0013 <.0001 <.0001 <.0001 0.5371 0.5654 0.0794 <.0001 <.0001 <.0001 0.4954 38 0.0156 0.0126 0.2694 0.0001 <.0001 <.0001 0.0060 0.0203 0.0794 0.0001 <.0001 <.0001 0.0152 39 <.0001 <.0001 0.0003 0.4138 0.0230 0.7369 <.0001 <.0001 <.0001 0.0001 0.0630 0.7110 <.0001 40 <.0001 <.0001 <.0001 0.0021 1.0000 0.0093 <.0001 <.0001 <.0001 <.0001 0.0630 0.1357 <.0001 41 <.0001 <.0001 <.0001 0.2043 0.0680 0.4300 <.0001 <.0001 <.0001 <.0001 0.7110 0.1357 <.0001 42 0.5777 0.6311 <.0001 <.0001 <.0001 <.0001 0.8275 0.9148 0.4954 0.0152 <.0001 <.0001 <.0001 43 0.0795 0.0950 <.0001 <.0001 <.0001 <.0001 0.1674 0.3117 0.1046 0.0005 <.0001 <.0001 <.0001 0.3694 44 0.0033 0.0025 0.3381 <.0001 <.0001 <.0001 0.0010 0.0072 0.0373 0.8331 0.0001 <.0001 <.0001 0.0052 45 0.0614 0.0507 0.0413 <.0001 <.0001 <.0001 0.0250 0.0655 0.2179 0.5174 <.0001 <.0001 <.0001 0.0506 46 <.0001 <.0001 <.0001 0.0302 0.2070 0.1011 <.0001 <.0001 <.0001 <.0001 0.3175 0.3205 0.5456 <.0001

229

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 11 (G01-R2b) : PROC GLM with CLASS varbs of ASSOC & LFORM and MODEL of TLFORMS TLFORMC = ASSOC¦LFORM from data set MWVRA 13:00 Monday, June 8, 2009

The GLM Procedure Least Squares Means

Least Squares Means for effect ASSOC*LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMS i/j 43 44 45 46 47 48 49 50 51 52 53 54 55 56

1 0.0891 0.0120 0.1216 <.0001 <.0001 <.0001 0.0072 0.4214 0.5835 0.1560 <.0001 <.0001 <.0001 <.0001 2 0.3733 0.0010 0.0186 <.0001 <.0001 <.0001 0.0005 0.9552 0.8414 0.0308 <.0001 <.0001 <.0001 <.0001 3 <.0001 0.9975 0.3304 <.0001 <.0001 0.0027 0.8544 0.0020 0.0045 0.3691 0.0303 <.0001 <.0001 0.0002 4 <.0001 <.0001 <.0001 0.6400 0.1106 0.0336 <.0001 <.0001 <.0001 <.0001 0.0109 0.1426 0.4706 0.3547 5 <.0001 <.0001 <.0001 0.5001 0.6473 0.0012 <.0001 <.0001 <.0001 <.0001 0.0004 0.6743 0.0775 0.0492 6 <.0001 0.0035 0.0001 0.0078 0.0002 0.9393 0.0061 <.0001 <.0001 0.0004 0.5943 0.0006 0.1911 0.2696 7 0.1474 0.0058 0.0719 <.0001 <.0001 <.0001 0.0033 0.5679 0.7526 0.0989 <.0001 <.0001 <.0001 <.0001 8 0.0721 0.0030 0.0598 <.0001 <.0001 <.0001 0.0015 0.4378 0.6226 0.0914 <.0001 <.0001 <.0001 <.0001 9 0.2006 0.0005 0.0165 <.0001 <.0001 <.0001 0.0002 0.7593 0.9823 0.0314 <.0001 <.0001 <.0001 <.0001 10 <.0001 0.6020 0.1041 <.0001 <.0001 0.0040 0.7512 0.0001 0.0003 0.1427 0.0529 <.0001 <.0001 0.0002 11 <.0001 0.0001 <.0001 0.0137 0.0002 0.6390 0.0003 <.0001 <.0001 <.0001 0.2761 0.0008 0.3406 0.4675 12 <.0001 <.0001 <.0001 0.4474 0.6006 0.0003 <.0001 <.0001 <.0001 <.0001 <.0001 0.6380 0.0509 0.0295 13 <.0001 0.0558 0.0027 <.0001 <.0001 0.1325 0.0872 <.0001 <.0001 0.0068 0.4925 <.0001 0.0068 0.0129 14 0.0919 0.0021 0.0460 <.0001 <.0001 <.0001 0.0010 0.5008 0.6970 0.0734 <.0001 <.0001 <.0001 <.0001 15 0.5643 0.0073 0.0525 <.0001 <.0001 <.0001 0.0048 0.9499 0.7889 0.0662 <.0001 <.0001 <.0001 <.0001 16 0.2633 0.0316 0.1613 <.0001 <.0001 <.0001 0.0222 0.6507 0.8042 0.1849 0.0002 <.0001 <.0001 <.0001 17 0.0002 0.6435 0.2232 0.0001 <.0001 0.0612 0.7466 0.0036 0.0066 0.2465 0.1942 <.0001 0.0059 0.0095 18 <.0001 0.0005 <.0001 0.4226 0.0941 0.2254 0.0007 <.0001 <.0001 <.0001 0.1033 0.1121 0.8729 0.7463 19 <.0001 <.0001 <.0001 0.3809 1.0000 0.0041 <.0001 <.0001 <.0001 <.0001 0.0014 1.0000 0.0806 0.0562 20 <.0001 0.0452 0.0060 0.0197 0.0014 0.7410 0.0622 <.0001 <.0001 0.0091 0.8660 0.0025 0.1926 0.2542 21 0.2308 0.0384 0.1864 <.0001 <.0001 <.0001 0.0272 0.5972 0.7462 0.2112 0.0003 <.0001 <.0001 <.0001 22 0.4542 0.0012 0.0199 <.0001 <.0001 <.0001 0.0007 0.9716 0.7791 0.0313 <.0001 <.0001 <.0001 <.0001 23 0.1122 0.0160 0.1351 <.0001 <.0001 <.0001 0.0100 0.4560 0.6165 0.1676 <.0001 <.0001 <.0001 <.0001 24 <.0001 0.5971 0.1464 <.0001 <.0001 0.0200 0.7207 0.0006 0.0014 0.1773 0.1128 <.0001 0.0009 0.0018 25 <.0001 0.0208 0.0013 0.0029 <.0001 0.5877 0.0320 <.0001 <.0001 0.0028 0.9408 0.0002 0.0925 0.1366 26 <.0001 <.0001 <.0001 0.2834 1.0000 0.0005 <.0001 <.0001 <.0001 <.0001 0.0001 1.0000 0.0370 0.0227 27 <.0001 0.2142 0.0308 <.0001 <.0001 0.1053 0.2838 <.0001 0.0001 0.0449 0.3541 <.0001 0.0076 0.0133 28 0.0571 0.0359 0.2385 <.0001 <.0001 <.0001 0.0235 0.2990 0.4271 0.2767 <.0001 <.0001 <.0001 <.0001 29 0.0795 0.0033 0.0614 <.0001 <.0001 <.0001 0.0017 0.4529 0.6381 0.0924 <.0001 <.0001 <.0001 <.0001 30 0.0950 0.0025 0.0507 <.0001 <.0001 <.0001 0.0013 0.5003 0.6938 0.0787 <.0001 <.0001 <.0001 <.0001 31 <.0001 0.3381 0.0413 <.0001 <.0001 0.0163 0.4490 <.0001 <.0001 0.0650 0.1300 <.0001 0.0005 0.0011

230

32 <.0001 <.0001 <.0001 0.0302 0.0007 0.4674 0.0001 <.0001 <.0001 <.0001 0.1891 0.0021 0.4824 0.6322 33 <.0001 <.0001 <.0001 0.2070 1.0000 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 1.0000 0.0171 0.0092 34 <.0001 <.0001 <.0001 0.1011 0.0040 0.2079 <.0001 <.0001 <.0001 <.0001 0.0733 0.0093 0.8251 0.9975 35 0.1674 0.0010 0.0250 <.0001 <.0001 <.0001 0.0005 0.6804 0.8955 0.0436 <.0001 <.0001 <.0001 <.0001 36 0.3117 0.0072 0.0655 <.0001 <.0001 <.0001 0.0045 0.7907 0.9710 0.0854 <.0001 <.0001 <.0001 <.0001 37 0.1046 0.0373 0.2179 <.0001 <.0001 <.0001 0.0252 0.4010 0.5411 0.2505 0.0001 <.0001 <.0001 <.0001 38 0.0005 0.8331 0.5174 <.0001 <.0001 0.0045 0.7120 0.0098 0.0184 0.5424 0.0330 <.0001 0.0002 0.0004 39 <.0001 0.0001 <.0001 0.3175 0.0470 0.1993 0.0002 <.0001 <.0001 <.0001 0.0825 0.0630 0.9252 0.7819 40 <.0001 <.0001 <.0001 0.3205 1.0000 0.0012 <.0001 <.0001 <.0001 <.0001 0.0004 1.0000 0.0510 0.0330 41 <.0001 <.0001 <.0001 0.5456 0.1108 0.0936 <.0001 <.0001 <.0001 <.0001 0.0355 0.1357 0.6424 0.5176 42 0.3694 0.0052 0.0506 <.0001 <.0001 <.0001 0.0032 0.8741 0.9437 0.0678 <.0001 <.0001 <.0001 <.0001 43 <.0001 0.0022 <.0001 <.0001 <.0001 <.0001 0.4663 0.3306 0.0046 <.0001 <.0001 <.0001 <.0001 44 <.0001 0.3543 <.0001 <.0001 0.0045 0.8641 0.0031 0.0065 0.3889 0.0385 <.0001 0.0002 0.0003 45 0.0022 0.3543 <.0001 <.0001 0.0002 0.2727 0.0338 0.0600 0.9969 0.0036 <.0001 <.0001 <.0001 46 <.0001 <.0001 <.0001 0.2834 0.0142 <.0001 <.0001 <.0001 <.0001 0.0045 0.3205 0.2715 0.1956

231

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 12 (G01-R2b) : PROC GLM with CLASS varbs of ASSOC & LFORM and MODEL of TLFORMS TLFORMC = ASSOC¦LFORM from data set MWVRA 13:00 Monday, June 8, 2009

The GLM Procedure Least Squares Means

Least Squares Means for effect ASSOC*LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMS i/j 1 2 3 4 5 6 7 8 9 10 11 12 13 14

47 <.0001 <.0001 <.0001 0.1106 0.6473 0.0002 <.0001 <.0001 <.0001 <.0001 0.0002 0.6006 <.0001 <.0001 48 <.0001 <.0001 0.0027 0.0336 0.0012 0.9393 <.0001 <.0001 <.0001 0.0040 0.6390 0.0003 0.1325 <.0001 49 0.0072 0.0005 0.8544 <.0001 <.0001 0.0061 0.0033 0.0015 0.0002 0.7512 0.0003 <.0001 0.0872 0.0010 50 0.4214 0.9552 0.0020 <.0001 <.0001 <.0001 0.5679 0.4378 0.7593 0.0001 <.0001 <.0001 <.0001 0.5008 51 0.5835 0.8414 0.0045 <.0001 <.0001 <.0001 0.7526 0.6226 0.9823 0.0003 <.0001 <.0001 <.0001 0.6970 52 0.1560 0.0308 0.3691 <.0001 <.0001 0.0004 0.0989 0.0914 0.0314 0.1427 <.0001 <.0001 0.0068 0.0734 53 <.0001 <.0001 0.0303 0.0109 0.0004 0.5943 <.0001 <.0001 <.0001 0.0529 0.2761 <.0001 0.4925 <.0001 54 <.0001 <.0001 <.0001 0.1426 0.6743 0.0006 <.0001 <.0001 <.0001 <.0001 0.0008 0.6380 <.0001 <.0001 55 <.0001 <.0001 <.0001 0.4706 0.0775 0.1911 <.0001 <.0001 <.0001 <.0001 0.3406 0.0509 0.0068 <.0001 56 <.0001 <.0001 0.0002 0.3547 0.0492 0.2696 <.0001 <.0001 <.0001 0.0002 0.4675 0.0295 0.0129 <.0001

Least Squares Means for effect ASSOC*LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMS i/j 15 16 17 18 19 20 21 22 23 24 25 26 27 28

47 <.0001 <.0001 <.0001 0.0941 1.0000 0.0014 <.0001 <.0001 <.0001 <.0001 <.0001 1.0000 <.0001 <.0001 48 <.0001 <.0001 0.0612 0.2254 0.0041 0.7410 <.0001 <.0001 <.0001 0.0200 0.5877 0.0005 0.1053 <.0001 49 0.0048 0.0222 0.7466 0.0007 <.0001 0.0622 0.0272 0.0007 0.0100 0.7207 0.0320 <.0001 0.2838 0.0235 50 0.9499 0.6507 0.0036 <.0001 <.0001 <.0001 0.5972 0.9716 0.4560 0.0006 <.0001 <.0001 <.0001 0.2990 51 0.7889 0.8042 0.0066 <.0001 <.0001 <.0001 0.7462 0.7791 0.6165 0.0014 <.0001 <.0001 0.0001 0.4271 52 0.0662 0.1849 0.2465 <.0001 <.0001 0.0091 0.2112 0.0313 0.1676 0.1773 0.0028 <.0001 0.0449 0.2767 53 <.0001 0.0002 0.1942 0.1033 0.0014 0.8660 0.0003 <.0001 <.0001 0.1128 0.9408 0.0001 0.3541 <.0001 54 <.0001 <.0001 <.0001 0.1121 1.0000 0.0025 <.0001 <.0001 <.0001 <.0001 0.0002 1.0000 <.0001 <.0001 55 <.0001 <.0001 0.0059 0.8729 0.0806 0.1926 <.0001 <.0001 <.0001 0.0009 0.0925 0.0370 0.0076 <.0001 56 <.0001 <.0001 0.0095 0.7463 0.0562 0.2542 <.0001 <.0001 <.0001 0.0018 0.1366 0.0227 0.0133 <.0001

Least Squares Means for effect ASSOC*LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

232

Dependent Variable: TLFORMS i/j 29 30 31 32 33 34 35 36 37 38 39 40 41 42

47 <.0001 <.0001 <.0001 0.0007 1.0000 0.0040 <.0001 <.0001 <.0001 <.0001 0.0470 1.0000 0.1108 <.0001 48 <.0001 <.0001 0.0163 0.4674 <.0001 0.2079 <.0001 <.0001 <.0001 0.0045 0.1993 0.0012 0.0936 <.0001 49 0.0017 0.0013 0.4490 0.0001 <.0001 <.0001 0.0005 0.0045 0.0252 0.7120 0.0002 <.0001 <.0001 0.0032 50 0.4529 0.5003 <.0001 <.0001 <.0001 <.0001 0.6804 0.7907 0.4010 0.0098 <.0001 <.0001 <.0001 0.8741 51 0.6381 0.6938 <.0001 <.0001 <.0001 <.0001 0.8955 0.9710 0.5411 0.0184 <.0001 <.0001 <.0001 0.9437 52 0.0924 0.0787 0.0650 <.0001 <.0001 <.0001 0.0436 0.0854 0.2505 0.5424 <.0001 <.0001 <.0001 0.0678 53 <.0001 <.0001 0.1300 0.1891 <.0001 0.0733 <.0001 <.0001 0.0001 0.0330 0.0825 0.0004 0.0355 <.0001 54 <.0001 <.0001 <.0001 0.0021 1.0000 0.0093 <.0001 <.0001 <.0001 <.0001 0.0630 1.0000 0.1357 <.0001 55 <.0001 <.0001 0.0005 0.4824 0.0171 0.8251 <.0001 <.0001 <.0001 0.0002 0.9252 0.0510 0.6424 <.0001 56 <.0001 <.0001 0.0011 0.6322 0.0092 0.9975 <.0001 <.0001 <.0001 0.0004 0.7819 0.0330 0.5176 <.0001

233

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 13 (G01-R2b) : PROC GLM with CLASS varbs of ASSOC & LFORM and MODEL of TLFORMS TLFORMC = ASSOC¦LFORM from data set MWVRA 13:00 Monday, June 8, 2009

The GLM Procedure Least Squares Means

Least Squares Means for effect ASSOC*LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMS i/j 43 44 45 46 47 48 49 50 51 52 53 54 55 56

47 <.0001 <.0001 <.0001 0.2834 0.0005 <.0001 <.0001 <.0001 <.0001 0.0001 1.0000 0.0370 0.0227 48 <.0001 0.0045 0.0002 0.0142 0.0005 0.0074 <.0001 <.0001 0.0005 0.5645 0.0012 0.2365 0.3227 49 <.0001 0.8641 0.2727 <.0001 <.0001 0.0074 0.0018 0.0040 0.3079 0.0557 <.0001 0.0003 0.0006 50 0.4663 0.0031 0.0338 <.0001 <.0001 <.0001 0.0018 0.8188 0.0474 <.0001 <.0001 <.0001 <.0001 51 0.3306 0.0065 0.0600 <.0001 <.0001 <.0001 0.0040 0.8188 0.0790 <.0001 <.0001 <.0001 <.0001 52 0.0046 0.3889 0.9969 <.0001 <.0001 0.0005 0.3079 0.0474 0.0790 0.0064 <.0001 <.0001 <.0001 53 <.0001 0.0385 0.0036 0.0045 0.0001 0.5645 0.0557 <.0001 <.0001 0.0064 0.0004 0.1003 0.1438 54 <.0001 <.0001 <.0001 0.3205 1.0000 0.0012 <.0001 <.0001 <.0001 <.0001 0.0004 0.0510 0.0330 55 <.0001 0.0002 <.0001 0.2715 0.0370 0.2365 0.0003 <.0001 <.0001 <.0001 0.1003 0.0510 0.8548 56 <.0001 0.0003 <.0001 0.1956 0.0227 0.3227 0.0006 <.0001 <.0001 <.0001 0.1438 0.0330 0.8548

TLFORMC Standard LSMEAN ASSOC LFORM LSMEAN Error Pr > ¦t¦ Number

Assoc-1 Chamaeph 1.65608789 0.11786977 <.0001 1 Assoc-1 Cryptoph 1.14005255 0.11786977 <.0001 2 Assoc-1 Hemicryp 0.63681260 0.11786977 <.0001 3 Assoc-1 Liana 0.09767095 0.11786977 0.4082 4 Assoc-1 Parasite 0.03203548 0.11786977 0.7860 5 Assoc-1 Phanerop 1.01591302 0.11786977 <.0001 6 Assoc-1 Therophy 1.25757888 0.11786977 <.0001 7 Assoc-2 Chamaeph 1.45242262 0.08334651 <.0001 8 Assoc-2 Cryptoph 1.16836149 0.08334651 <.0001 9 Assoc-2 Hemicryp 0.58655792 0.08334651 <.0001 10 Assoc-2 Liana 0.24128729 0.08334651 0.0042 11 Assoc-2 Parasite 0.01958131 0.08334651 0.8145 12 Assoc-2 Phanerop 1.28630357 0.08334651 <.0001 13 Assoc-2 Therophy 1.10515626 0.08334651 <.0001 14 Assoc-3 Chamaeph 1.49928254 0.18636847 <.0001 15 Assoc-3 Cryptoph 1.09666789 0.18636847 <.0001 16

234

Assoc-3 Hemicryp 0.65485160 0.18636847 0.0005 17 Assoc-3 Liana 0.18117820 0.18636847 0.3320 18 Assoc-3 Parasite 0.00000000 0.18636847 1.0000 19 Assoc-3 Phanerop 1.55285164 0.18636847 <.0001 20 Assoc-3 Therophy 1.15883406 0.18636847 <.0001 21 Assoc-4 Chamaeph 1.54617168 0.13178241 <.0001 22 Assoc-4 Cryptoph 0.97728267 0.13178241 <.0001 23 Assoc-4 Hemicryp 0.60562319 0.13178241 <.0001 24 Assoc-4 Liana 0.31982489 0.13178241 0.0160 25 Assoc-4 Parasite 0.00000000 0.13178241 1.0000 26 Assoc-4 Phanerop 1.51836058 0.13178241 <.0001 27 Assoc-4 Therophy 0.98768051 0.13178241 <.0001 28 Assoc-5 Chamaeph 1.42205644 0.08785494 <.0001 29 Assoc-5 Cryptoph 0.94136556 0.08785494 <.0001 30 Assoc-5 Hemicryp 0.53160670 0.08785494 <.0001 31

235

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 14 (G01-R2b) : PROC GLM with CLASS varbs of ASSOC & LFORM and MODEL of TLFORMS TLFORMC = ASSOC¦LFORM from data set MWVRA 13:00 Monday, June 8, 2009

The GLM Procedure Least Squares Means

TLFORMC Standard LSMEAN ASSOC LFORM LSMEAN Error Pr > ¦t¦ Number

Assoc-5 Liana 0.15474097 0.08785494 0.0795 32 Assoc-5 Parasite 0.00000000 0.08785494 1.0000 33 Assoc-5 Phanerop 0.23455702 0.08785494 0.0081 34 Assoc-5 Therophy 1.31119511 0.08785494 <.0001 35 Assoc-6 Chamaeph 1.85177904 0.15216922 <.0001 36 Assoc-6 Cryptoph 1.06161151 0.15216922 <.0001 37 Assoc-6 Hemicryp 0.42387462 0.15216922 0.0058 38 Assoc-6 Liana 0.16276349 0.15216922 0.2859 39 Assoc-6 Parasite 0.00000000 0.15216922 1.0000 40 Assoc-6 Phanerop 0.25533979 0.15216922 0.0947 41 Assoc-6 Therophy 1.19823488 0.15216922 <.0001 42 Assoc-7 Chamaeph 1.83237780 0.13178241 <.0001 43 Assoc-7 Cryptoph 1.09438976 0.13178241 <.0001 44 Assoc-7 Hemicryp 0.46491582 0.13178241 0.0005 45 Assoc-7 Liana 0.13345199 0.13178241 0.3123 46 Assoc-7 Parasite 0.00000000 0.13178241 1.0000 47 Assoc-7 Phanerop 0.27231644 0.13178241 0.0399 48 Assoc-7 Therophy 1.24315258 0.13178241 <.0001 49 Assoc-8 Chamaeph 1.81297655 0.15216922 <.0001 50 Assoc-8 Cryptoph 1.12716801 0.15216922 <.0001 51 Assoc-8 Hemicryp 0.50595703 0.15216922 0.0010 52 Assoc-8 Liana 0.10414048 0.15216922 0.4944 53 Assoc-8 Parasite 0.00000000 0.15216922 1.0000 54 Assoc-8 Phanerop 0.28929310 0.15216922 0.0586 55 Assoc-8 Therophy 1.28807029 0.15216922 <.0001 56

236

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 15 (G01-R2b) : PROC GLM with CLASS varbs of ASSOC & LFORM and MODEL of TLFORMS TLFORMC = ASSOC¦LFORM from data set MWVRA 13:00 Monday, June 8, 2009

The GLM Procedure Least Squares Means

Least Squares Means for effect ASSOC*LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMC i/j 1 2 3 4 5 6 7 8 9 10 11 12 13 14

1 0.0022 <.0001 <.0001 <.0001 0.0002 0.0176 0.1597 0.0009 <.0001 <.0001 <.0001 0.0111 0.0002 2 0.0022 0.0028 <.0001 <.0001 0.4572 0.4815 0.0315 0.8447 0.0002 <.0001 <.0001 0.3121 0.8092 3 <.0001 0.0028 0.0014 0.0004 0.0239 0.0002 <.0001 0.0003 0.7281 0.0066 <.0001 <.0001 0.0014 4 <.0001 <.0001 0.0014 0.6941 <.0001 <.0001 <.0001 <.0001 0.0008 0.3209 0.5891 <.0001 <.0001 5 <.0001 <.0001 0.0004 0.6941 <.0001 <.0001 <.0001 <.0001 0.0002 0.1486 0.9313 <.0001 <.0001 6 0.0002 0.4572 0.0239 <.0001 <.0001 0.1485 0.0028 0.2921 0.0033 <.0001 <.0001 0.0624 0.5371 7 0.0176 0.4815 0.0002 <.0001 <.0001 0.1485 0.1785 0.5372 <.0001 <.0001 <.0001 0.8425 0.2922 8 0.1597 0.0315 <.0001 <.0001 <.0001 0.0028 0.1785 0.0168 <.0001 <.0001 <.0001 0.1601 0.0036 9 0.0009 0.8447 0.0003 <.0001 <.0001 0.2921 0.5372 0.0168 <.0001 <.0001 <.0001 0.3181 0.5923 10 <.0001 0.0002 0.7281 0.0008 0.0002 0.0033 <.0001 <.0001 <.0001 0.0037 <.0001 <.0001 <.0001 11 <.0001 <.0001 0.0066 0.3209 0.1486 <.0001 <.0001 <.0001 <.0001 0.0037 0.0613 <.0001 <.0001 12 <.0001 <.0001 <.0001 0.5891 0.9313 <.0001 <.0001 <.0001 <.0001 <.0001 0.0613 <.0001 <.0001 13 0.0111 0.3121 <.0001 <.0001 <.0001 0.0624 0.8425 0.1601 0.3181 <.0001 <.0001 <.0001 0.1257 14 0.0002 0.8092 0.0014 <.0001 <.0001 0.5371 0.2922 0.0036 0.5923 <.0001 <.0001 <.0001 0.1257 15 0.4778 0.1047 0.0001 <.0001 <.0001 0.0294 0.2742 0.8187 0.1064 <.0001 <.0001 <.0001 0.2980 0.0548 16 0.0119 0.8442 0.0382 <.0001 <.0001 0.7146 0.4663 0.0828 0.7258 0.0132 <.0001 <.0001 0.3540 0.9669 17 <.0001 0.0288 0.9349 0.0122 0.0052 0.1030 0.0068 0.0001 0.0126 0.7383 0.0440 0.0021 0.0022 0.0284 18 <.0001 <.0001 0.0400 0.7053 0.4995 0.0002 <.0001 <.0001 <.0001 0.0483 0.7687 0.4295 <.0001 <.0001 19 <.0001 <.0001 0.0043 0.6582 0.8846 <.0001 <.0001 <.0001 <.0001 0.0045 0.2385 0.9237 <.0001 <.0001 20 0.6401 0.0625 <.0001 <.0001 <.0001 0.0157 0.1819 0.6233 0.0610 <.0001 <.0001 <.0001 0.1930 0.0293 21 0.0251 0.9322 0.0188 <.0001 <.0001 0.5176 0.6547 0.1518 0.9628 0.0055 <.0001 <.0001 0.5330 0.7929 22 0.5348 0.0225 <.0001 <.0001 <.0001 0.0030 0.1040 0.5483 0.0162 <.0001 <.0001 <.0001 0.0970 0.0051 23 0.0002 0.3582 0.0554 <.0001 <.0001 0.8272 0.1143 0.0026 0.2217 0.0129 <.0001 <.0001 0.0487 0.4130 24 <.0001 0.0028 0.8601 0.0045 0.0014 0.0212 0.0003 <.0001 0.0004 0.9028 0.0203 0.0002 <.0001 0.0016 25 <.0001 <.0001 0.0743 0.2102 0.1050 0.0001 <.0001 <.0001 <.0001 0.0885 0.6150 0.0554 <.0001 <.0001 26 <.0001 <.0001 0.0004 0.5812 0.8564 <.0001 <.0001 <.0001 <.0001 0.0002 0.1232 0.9002 <.0001 <.0001 27 0.4368 0.0335 <.0001 <.0001 <.0001 0.0049 0.1416 0.6728 0.0258 <.0001 <.0001 <.0001 0.1381 0.0086 28 0.0002 0.3897 0.0484 <.0001 <.0001 0.8733 0.1283 0.0032 0.2478 0.0107 <.0001 <.0001 0.0567 0.4520 29 0.1128 0.0563 <.0001 <.0001 <.0001 0.0062 0.2644 0.8022 0.0373 <.0001 <.0001 <.0001 0.2635 0.0095 30 <.0001 0.1779 0.0394 <.0001 <.0001 0.6126 0.0325 <.0001 0.0622 0.0037 <.0001 <.0001 0.0048 0.1776 31 <.0001 <.0001 0.4750 0.0035 0.0008 0.0011 <.0001 <.0001 <.0001 0.6504 0.0173 <.0001 <.0001 <.0001

237

32 <.0001 <.0001 0.0012 0.6982 0.4048 <.0001 <.0001 <.0001 <.0001 0.0004 0.4756 0.2656 <.0001 <.0001 33 <.0001 <.0001 <.0001 0.5071 0.8277 <.0001 <.0001 <.0001 <.0001 <.0001 0.0475 0.8717 <.0001 <.0001 34 <.0001 <.0001 0.0067 0.3528 0.1697 <.0001 <.0001 <.0001 <.0001 0.0040 0.9557 0.0772 <.0001 <.0001 35 0.0198 0.2456 <.0001 <.0001 <.0001 0.0458 0.7157 0.2448 0.2395 <.0001 <.0001 <.0001 0.8373 0.0903 36 0.3104 0.0003 <.0001 <.0001 <.0001 <.0001 0.0023 0.0223 0.0001 <.0001 <.0001 <.0001 0.0013 <.0001 37 0.0023 0.6840 0.0283 <.0001 <.0001 0.8125 0.3097 0.0253 0.5390 0.0067 <.0001 <.0001 0.1966 0.8021 38 <.0001 0.0003 0.2698 0.0915 0.0430 0.0024 <.0001 <.0001 <.0001 0.3494 0.2938 0.0207 <.0001 0.0001 39 <.0001 <.0001 0.0145 0.7355 0.4977 <.0001 <.0001 <.0001 <.0001 0.0154 0.6513 0.4101 <.0001 <.0001 40 <.0001 <.0001 0.0011 0.6124 0.8680 <.0001 <.0001 <.0001 <.0001 0.0009 0.1657 0.9102 <.0001 <.0001 41 <.0001 <.0001 0.0487 0.4136 0.2472 0.0001 <.0001 <.0001 <.0001 0.0575 0.9355 0.1756 <.0001 <.0001 42 0.0182 0.7627 0.0039 <.0001 <.0001 0.3445 0.7581 0.1443 0.8635 0.0005 <.0001 <.0001 0.6122 0.5922 43 0.3198 0.0001 <.0001 <.0001 <.0001 <.0001 0.0013 0.0156 <.0001 <.0001 <.0001 <.0001 0.0006 <.0001 44 0.0017 0.7964 0.0103 <.0001 <.0001 0.6576 0.3570 0.0226 0.6357 0.0013 <.0001 <.0001 0.2197 0.9450 45 <.0001 0.0002 0.3320 0.0389 0.0151 0.0021 <.0001 <.0001 <.0001 0.4361 0.1529 0.0047 <.0001 <.0001 46 <.0001 <.0001 0.0048 0.8398 0.5668 <.0001 <.0001 <.0001 <.0001 0.0040 0.4899 0.4660 <.0001 <.0001

238

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 16 (G01-R2b) : PROC GLM with CLASS varbs of ASSOC & LFORM and MODEL of TLFORMS TLFORMC = ASSOC¦LFORM from data set MWVRA 13:00 Monday, June 8, 2009

The GLM Procedure Least Squares Means

Least Squares Means for effect ASSOC*LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMC i/j 15 16 17 18 19 20 21 22 23 24 25 26 27 28

1 0.4778 0.0119 <.0001 <.0001 <.0001 0.6401 0.0251 0.5348 0.0002 <.0001 <.0001 <.0001 0.4368 0.0002 2 0.1047 0.8442 0.0288 <.0001 <.0001 0.0625 0.9322 0.0225 0.3582 0.0028 <.0001 <.0001 0.0335 0.3897 3 0.0001 0.0382 0.9349 0.0400 0.0043 <.0001 0.0188 <.0001 0.0554 0.8601 0.0743 0.0004 <.0001 0.0484 4 <.0001 <.0001 0.0122 0.7053 0.6582 <.0001 <.0001 <.0001 <.0001 0.0045 0.2102 0.5812 <.0001 <.0001 5 <.0001 <.0001 0.0052 0.4995 0.8846 <.0001 <.0001 <.0001 <.0001 0.0014 0.1050 0.8564 <.0001 <.0001 6 0.0294 0.7146 0.1030 0.0002 <.0001 0.0157 0.5176 0.0030 0.8272 0.0212 0.0001 <.0001 0.0049 0.8733 7 0.2742 0.4663 0.0068 <.0001 <.0001 0.1819 0.6547 0.1040 0.1143 0.0003 <.0001 <.0001 0.1416 0.1283 8 0.8187 0.0828 0.0001 <.0001 <.0001 0.6233 0.1518 0.5483 0.0026 <.0001 <.0001 <.0001 0.6728 0.0032 9 0.1064 0.7258 0.0126 <.0001 <.0001 0.0610 0.9628 0.0162 0.2217 0.0004 <.0001 <.0001 0.0258 0.2478 10 <.0001 0.0132 0.7383 0.0483 0.0045 <.0001 0.0055 <.0001 0.0129 0.9028 0.0885 0.0002 <.0001 0.0107 11 <.0001 <.0001 0.0440 0.7687 0.2385 <.0001 <.0001 <.0001 <.0001 0.0203 0.6150 0.1232 <.0001 <.0001 12 <.0001 <.0001 0.0021 0.4295 0.9237 <.0001 <.0001 <.0001 <.0001 0.0002 0.0554 0.9002 <.0001 <.0001 13 0.2980 0.3540 0.0022 <.0001 <.0001 0.1930 0.5330 0.0970 0.0487 <.0001 <.0001 <.0001 0.1381 0.0567 14 0.0548 0.9669 0.0284 <.0001 <.0001 0.0293 0.7929 0.0051 0.4130 0.0016 <.0001 <.0001 0.0086 0.4520 15 0.1280 0.0016 <.0001 <.0001 0.8391 0.1978 0.8374 0.0231 0.0001 <.0001 <.0001 0.9335 0.0260 16 0.1280 0.0951 0.0006 <.0001 0.0849 0.8138 0.0501 0.6015 0.0325 0.0008 <.0001 0.0660 0.6335 17 0.0016 0.0951 0.0737 0.0137 0.0008 0.0571 0.0001 0.1592 0.8294 0.1436 0.0045 0.0002 0.1462 18 <.0001 0.0006 0.0737 0.4925 <.0001 0.0003 <.0001 0.0006 0.0643 0.5442 0.4282 <.0001 0.0005 19 <.0001 <.0001 0.0137 0.4925 <.0001 <.0001 <.0001 <.0001 0.0085 0.1625 1.0000 <.0001 <.0001 20 0.8391 0.0849 0.0008 <.0001 <.0001 0.1363 0.9767 0.0124 <.0001 <.0001 <.0001 0.8800 0.0140 21 0.1978 0.8138 0.0571 0.0003 <.0001 0.1363 0.0911 0.4272 0.0162 0.0003 <.0001 0.1166 0.4541 22 0.8374 0.0501 0.0001 <.0001 <.0001 0.9767 0.0911 0.0025 <.0001 <.0001 <.0001 0.8815 0.0030 23 0.0231 0.6015 0.1592 0.0006 <.0001 0.0124 0.4272 0.0025 0.0473 0.0005 <.0001 0.0041 0.9556 24 0.0001 0.0325 0.8294 0.0643 0.0085 <.0001 0.0162 <.0001 0.0473 0.1266 0.0013 <.0001 0.0415 25 <.0001 0.0008 0.1436 0.5442 0.1625 <.0001 0.0003 <.0001 0.0005 0.1266 0.0875 <.0001 0.0004 26 <.0001 <.0001 0.0045 0.4282 1.0000 <.0001 <.0001 <.0001 <.0001 0.0013 0.0875 <.0001 <.0001 27 0.9335 0.0660 0.0002 <.0001 <.0001 0.8800 0.1166 0.8815 0.0041 <.0001 <.0001 <.0001 0.0048 28 0.0260 0.6335 0.1462 0.0005 <.0001 0.0140 0.4541 0.0030 0.9556 0.0415 0.0004 <.0001 0.0048 29 0.7082 0.1157 0.0002 <.0001 <.0001 0.5262 0.2027 0.4341 0.0054 <.0001 <.0001 <.0001 0.5438 0.0066 30 0.0073 0.4518 0.1657 0.0003 <.0001 0.0033 0.2923 0.0002 0.8208 0.0351 0.0001 <.0001 0.0003 0.7702 31 <.0001 0.0066 0.5503 0.0904 0.0105 <.0001 0.0026 <.0001 0.0053 0.6407 0.1825 0.0009 <.0001 0.0044

239

32 <.0001 <.0001 0.0160 0.8980 0.4534 <.0001 <.0001 <.0001 <.0001 0.0048 0.2984 0.3296 <.0001 <.0001 33 <.0001 <.0001 0.0017 0.3802 1.0000 <.0001 <.0001 <.0001 <.0001 0.0002 0.0446 1.0000 <.0001 <.0001 34 <.0001 <.0001 0.0425 0.7958 0.2562 <.0001 <.0001 <.0001 <.0001 0.0200 0.5909 0.1400 <.0001 <.0001 35 0.3623 0.2989 0.0017 <.0001 <.0001 0.2421 0.4604 0.1393 0.0361 <.0001 <.0001 <.0001 0.1922 0.0423 36 0.1443 0.0019 <.0001 <.0001 <.0001 0.2154 0.0044 0.1304 <.0001 <.0001 <.0001 <.0001 0.0991 <.0001 37 0.0702 0.8843 0.0923 0.0003 <.0001 0.0424 0.6865 0.0169 0.6757 0.0245 0.0003 <.0001 0.0242 0.7138 38 <.0001 0.0056 0.3381 0.3142 0.0795 <.0001 0.0025 <.0001 0.0065 0.3676 0.6057 0.0363 <.0001 0.0055 39 <.0001 0.0001 0.0420 0.9391 0.4994 <.0001 <.0001 <.0001 <.0001 0.0288 0.4361 0.4196 <.0001 <.0001 40 <.0001 <.0001 0.0070 0.4522 1.0000 <.0001 <.0001 <.0001 <.0001 0.0029 0.1135 1.0000 <.0001 <.0001 41 <.0001 0.0006 0.0982 0.7582 0.2897 <.0001 0.0002 <.0001 0.0004 0.0832 0.7490 0.2060 <.0001 0.0003 42 0.2122 0.6733 0.0249 <.0001 <.0001 0.1419 0.8701 0.0853 0.2735 0.0036 <.0001 <.0001 0.1132 0.2967 43 0.1459 0.0015 <.0001 <.0001 <.0001 0.2220 0.0035 0.1260 <.0001 <.0001 <.0001 <.0001 0.0934 <.0001 44 0.0774 0.9920 0.0554 <.0001 <.0001 0.0458 0.7779 0.0161 0.5304 0.0093 <.0001 <.0001 0.0239 0.5675 45 <.0001 0.0061 0.4062 0.2151 0.0428 <.0001 0.0026 <.0001 0.0065 0.4510 0.4371 0.0133 <.0001 0.0055 46 <.0001 <.0001 0.0233 0.8346 0.5594 <.0001 <.0001 <.0001 <.0001 0.0120 0.3184 0.4747 <.0001 <.0001

240

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 17 (G01-R2b) : PROC GLM with CLASS varbs of ASSOC & LFORM and MODEL of TLFORMS TLFORMC = ASSOC¦LFORM from data set MWVRA 13:00 Monday, June 8, 2009

The GLM Procedure Least Squares Means

Least Squares Means for effect ASSOC*LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMC i/j 29 30 31 32 33 34 35 36 37 38 39 40 41 42

1 0.1128 <.0001 <.0001 <.0001 <.0001 <.0001 0.0198 0.3104 0.0023 <.0001 <.0001 <.0001 <.0001 0.0182 2 0.0563 0.1779 <.0001 <.0001 <.0001 <.0001 0.2456 0.0003 0.6840 0.0003 <.0001 <.0001 <.0001 0.7627 3 <.0001 0.0394 0.4750 0.0012 <.0001 0.0067 <.0001 <.0001 0.0283 0.2698 0.0145 0.0011 0.0487 0.0039 4 <.0001 <.0001 0.0035 0.6982 0.5071 0.3528 <.0001 <.0001 <.0001 0.0915 0.7355 0.6124 0.4136 <.0001 5 <.0001 <.0001 0.0008 0.4048 0.8277 0.1697 <.0001 <.0001 <.0001 0.0430 0.4977 0.8680 0.2472 <.0001 6 0.0062 0.6126 0.0011 <.0001 <.0001 <.0001 0.0458 <.0001 0.8125 0.0024 <.0001 <.0001 0.0001 0.3445 7 0.2644 0.0325 <.0001 <.0001 <.0001 <.0001 0.7157 0.0023 0.3097 <.0001 <.0001 <.0001 <.0001 0.7581 8 0.8022 <.0001 <.0001 <.0001 <.0001 <.0001 0.2448 0.0223 0.0253 <.0001 <.0001 <.0001 <.0001 0.1443 9 0.0373 0.0622 <.0001 <.0001 <.0001 <.0001 0.2395 0.0001 0.5390 <.0001 <.0001 <.0001 <.0001 0.8635 10 <.0001 0.0037 0.6504 0.0004 <.0001 0.0040 <.0001 <.0001 0.0067 0.3494 0.0154 0.0009 0.0575 0.0005 11 <.0001 <.0001 0.0173 0.4756 0.0475 0.9557 <.0001 <.0001 <.0001 0.2938 0.6513 0.1657 0.9355 <.0001 12 <.0001 <.0001 <.0001 0.2656 0.8717 0.0772 <.0001 <.0001 <.0001 0.0207 0.4101 0.9102 0.1756 <.0001 13 0.2635 0.0048 <.0001 <.0001 <.0001 <.0001 0.8373 0.0013 0.1966 <.0001 <.0001 <.0001 <.0001 0.6122 14 0.0095 0.1776 <.0001 <.0001 <.0001 <.0001 0.0903 <.0001 0.8021 0.0001 <.0001 <.0001 <.0001 0.5922 15 0.7082 0.0073 <.0001 <.0001 <.0001 <.0001 0.3623 0.1443 0.0702 <.0001 <.0001 <.0001 <.0001 0.2122 16 0.1157 0.4518 0.0066 <.0001 <.0001 <.0001 0.2989 0.0019 0.8843 0.0056 0.0001 <.0001 0.0006 0.6733 17 0.0002 0.1657 0.5503 0.0160 0.0017 0.0425 0.0017 <.0001 0.0923 0.3381 0.0420 0.0070 0.0982 0.0249 18 <.0001 0.0003 0.0904 0.8980 0.3802 0.7958 <.0001 <.0001 0.0003 0.3142 0.9391 0.4522 0.7582 <.0001 19 <.0001 <.0001 0.0105 0.4534 1.0000 0.2562 <.0001 <.0001 <.0001 0.0795 0.4994 1.0000 0.2897 <.0001 20 0.5262 0.0033 <.0001 <.0001 <.0001 <.0001 0.2421 0.2154 0.0424 <.0001 <.0001 <.0001 <.0001 0.1419 21 0.2027 0.2923 0.0026 <.0001 <.0001 <.0001 0.4604 0.0044 0.6865 0.0025 <.0001 <.0001 0.0002 0.8701 22 0.4341 0.0002 <.0001 <.0001 <.0001 <.0001 0.1393 0.1304 0.0169 <.0001 <.0001 <.0001 <.0001 0.0853 23 0.0054 0.8208 0.0053 <.0001 <.0001 <.0001 0.0361 <.0001 0.6757 0.0065 <.0001 <.0001 0.0004 0.2735 24 <.0001 0.0351 0.6407 0.0048 0.0002 0.0200 <.0001 <.0001 0.0245 0.3676 0.0288 0.0029 0.0832 0.0036 25 <.0001 0.0001 0.1825 0.2984 0.0446 0.5909 <.0001 <.0001 0.0003 0.6057 0.4361 0.1135 0.7490 <.0001 26 <.0001 <.0001 0.0009 0.3296 1.0000 0.1400 <.0001 <.0001 <.0001 0.0363 0.4196 1.0000 0.2060 <.0001 27 0.5438 0.0003 <.0001 <.0001 <.0001 <.0001 0.1922 0.0991 0.0242 <.0001 <.0001 <.0001 <.0001 0.1132 28 0.0066 0.7702 0.0044 <.0001 <.0001 <.0001 0.0423 <.0001 0.7138 0.0055 <.0001 <.0001 0.0003 0.2967 29 0.0001 <.0001 <.0001 <.0001 <.0001 0.3732 0.0152 0.0414 <.0001 <.0001 <.0001 <.0001 0.2040 30 0.0001 0.0011 <.0001 <.0001 <.0001 0.0032 <.0001 0.4945 0.0036 <.0001 <.0001 0.0001 0.1452 31 <.0001 0.0011 0.0027 <.0001 0.0176 <.0001 <.0001 0.0029 0.5404 0.0369 0.0028 0.1173 0.0002

241

32 <.0001 <.0001 0.0027 0.2143 0.5213 <.0001 <.0001 <.0001 0.1270 0.9636 0.3794 0.5675 <.0001 33 <.0001 <.0001 <.0001 0.2143 0.0603 <.0001 <.0001 <.0001 0.0167 0.3553 1.0000 0.1476 <.0001 34 <.0001 <.0001 0.0176 0.5213 0.0603 <.0001 <.0001 <.0001 0.2824 0.6832 0.1833 0.9060 <.0001 35 0.3732 0.0032 <.0001 <.0001 <.0001 <.0001 0.0024 0.1569 <.0001 <.0001 <.0001 <.0001 0.5210 36 0.0152 <.0001 <.0001 <.0001 <.0001 <.0001 0.0024 0.0003 <.0001 <.0001 <.0001 <.0001 0.0027 37 0.0414 0.4945 0.0029 <.0001 <.0001 <.0001 0.1569 0.0003 0.0034 <.0001 <.0001 0.0002 0.5262 38 <.0001 0.0036 0.5404 0.1270 0.0167 0.2824 <.0001 <.0001 0.0034 0.2263 0.0501 0.4344 0.0004 39 <.0001 <.0001 0.0369 0.9636 0.3553 0.6832 <.0001 <.0001 <.0001 0.2263 0.4502 0.6675 <.0001 40 <.0001 <.0001 0.0028 0.3794 1.0000 0.1833 <.0001 <.0001 <.0001 0.0501 0.4502 0.2367 <.0001 41 <.0001 0.0001 0.1173 0.5675 0.1476 0.9060 <.0001 <.0001 0.0002 0.4344 0.6675 0.2367 <.0001 42 0.2040 0.1452 0.0002 <.0001 <.0001 <.0001 0.5210 0.0027 0.5262 0.0004 <.0001 <.0001 <.0001 43 0.0102 <.0001 <.0001 <.0001 <.0001 <.0001 0.0012 0.9233 0.0002 <.0001 <.0001 <.0001 <.0001 0.0019 44 0.0397 0.3350 0.0005 <.0001 <.0001 <.0001 0.1724 0.0002 0.8708 0.0010 <.0001 <.0001 <.0001 0.6065 45 <.0001 0.0029 0.6741 0.0514 0.0037 0.1472 <.0001 <.0001 0.0034 0.8386 0.1348 0.0218 0.2989 0.0003 46 <.0001 <.0001 0.0126 0.8932 0.4004 0.5239 <.0001 <.0001 <.0001 0.1505 0.8844 0.5080 0.5455 <.0001

242

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 18 (G01-R2b) : PROC GLM with CLASS varbs of ASSOC & LFORM and MODEL of TLFORMS TLFORMC = ASSOC¦LFORM from data set MWVRA 13:00 Monday, June 8, 2009

The GLM Procedure Least Squares Means

Least Squares Means for effect ASSOC*LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMC i/j 43 44 45 46 47 48 49 50 51 52 53 54 55 56

1 0.3198 0.0017 <.0001 <.0001 <.0001 <.0001 0.0204 0.4159 0.0065 <.0001 <.0001 <.0001 <.0001 0.0572 2 0.0001 0.7964 0.0002 <.0001 <.0001 <.0001 0.5604 0.0006 0.9467 0.0011 <.0001 <.0001 <.0001 0.4427 3 <.0001 0.0103 0.3320 0.0048 0.0004 0.0404 0.0007 <.0001 0.0115 0.4973 0.0061 0.0011 0.0723 0.0008 4 <.0001 <.0001 0.0389 0.8398 0.5812 0.3243 <.0001 <.0001 <.0001 0.0350 0.9732 0.6124 0.3205 <.0001 5 <.0001 <.0001 0.0151 0.5668 0.8564 0.1755 <.0001 <.0001 <.0001 0.0146 0.7083 0.8680 0.1827 <.0001 6 <.0001 0.6576 0.0021 <.0001 <.0001 <.0001 0.2000 <.0001 0.5638 0.0086 <.0001 <.0001 0.0002 0.1588 7 0.0013 0.3570 <.0001 <.0001 <.0001 <.0001 0.9350 0.0043 0.4988 0.0001 <.0001 <.0001 <.0001 0.8743 8 0.0156 0.0226 <.0001 <.0001 <.0001 <.0001 0.1809 0.0388 0.0621 <.0001 <.0001 <.0001 <.0001 0.3445 9 <.0001 0.6357 <.0001 <.0001 <.0001 <.0001 0.6319 0.0003 0.8125 0.0002 <.0001 <.0001 <.0001 0.4909 10 <.0001 0.0013 0.4361 0.0040 0.0002 0.0451 <.0001 <.0001 0.0021 0.6427 0.0059 0.0009 0.0880 <.0001 11 <.0001 <.0001 0.1529 0.4899 0.1232 0.8424 <.0001 <.0001 <.0001 0.1286 0.4301 0.1657 0.7823 <.0001 12 <.0001 <.0001 0.0047 0.4660 0.9002 0.1065 <.0001 <.0001 <.0001 0.0055 0.6265 0.9102 0.1215 <.0001 13 0.0006 0.2197 <.0001 <.0001 <.0001 <.0001 0.7822 0.0027 0.3600 <.0001 <.0001 <.0001 <.0001 0.9919 14 <.0001 0.9450 <.0001 <.0001 <.0001 <.0001 0.3771 <.0001 0.8992 0.0007 <.0001 <.0001 <.0001 0.2929 15 0.1459 0.0774 <.0001 <.0001 <.0001 <.0001 0.2630 0.1936 0.1234 <.0001 <.0001 <.0001 <.0001 0.3810 16 0.0015 0.9920 0.0061 <.0001 <.0001 0.0004 0.5217 0.0032 0.8992 0.0148 <.0001 <.0001 0.0009 0.4272 17 <.0001 0.0554 0.4062 0.0233 0.0045 0.0951 0.0106 <.0001 0.0509 0.5366 0.0230 0.0070 0.1301 0.0091 18 <.0001 <.0001 0.2151 0.8346 0.4282 0.6901 <.0001 <.0001 0.0001 0.1784 0.7491 0.4522 0.6536 <.0001 19 <.0001 <.0001 0.0428 0.5594 1.0000 0.2341 <.0001 <.0001 <.0001 0.0366 0.6656 1.0000 0.2305 <.0001 20 0.2220 0.0458 <.0001 <.0001 <.0001 <.0001 0.1762 0.2808 0.0782 <.0001 <.0001 <.0001 <.0001 0.2723 21 0.0035 0.7779 0.0026 <.0001 <.0001 0.0001 0.7122 0.0071 0.8954 0.0072 <.0001 <.0001 0.0004 0.5917 22 0.1260 0.0161 <.0001 <.0001 <.0001 <.0001 0.1054 0.1864 0.0385 <.0001 <.0001 <.0001 <.0001 0.2011 23 <.0001 0.5304 0.0065 <.0001 <.0001 0.0002 0.1551 <.0001 0.4573 0.0201 <.0001 <.0001 0.0007 0.1240 24 <.0001 0.0093 0.4510 0.0120 0.0013 0.0751 0.0007 <.0001 0.0102 0.6210 0.0135 0.0029 0.1175 0.0008 25 <.0001 <.0001 0.4371 0.3184 0.0875 0.7990 <.0001 <.0001 <.0001 0.3561 0.2851 0.1135 0.8796 <.0001 26 <.0001 <.0001 0.0133 0.4747 1.0000 0.1454 <.0001 <.0001 <.0001 0.0127 0.6054 1.0000 0.1521 <.0001 27 0.0934 0.0239 <.0001 <.0001 <.0001 <.0001 0.1412 0.1447 0.0532 <.0001 <.0001 <.0001 <.0001 0.2538 28 <.0001 0.5675 0.0055 <.0001 <.0001 0.0002 0.1718 <.0001 0.4891 0.0175 <.0001 <.0001 0.0006 0.1370 29 0.0102 0.0397 <.0001 <.0001 <.0001 <.0001 0.2599 0.0271 0.0947 <.0001 <.0001 <.0001 <.0001 0.4465 30 <.0001 0.3350 0.0029 <.0001 <.0001 <.0001 0.0580 <.0001 0.2915 0.0140 <.0001 <.0001 0.0003 0.0497 31 <.0001 0.0005 0.6741 0.0126 0.0009 0.1030 <.0001 <.0001 0.0008 0.8841 0.0158 0.0028 0.1693 <.0001

243

32 <.0001 <.0001 0.0514 0.8932 0.3296 0.4587 <.0001 <.0001 <.0001 0.0468 0.7736 0.3794 0.4446 <.0001 33 <.0001 <.0001 0.0037 0.4004 1.0000 0.0869 <.0001 <.0001 <.0001 0.0044 0.5540 1.0000 0.1011 <.0001 34 <.0001 <.0001 0.1472 0.5239 0.1400 0.8118 <.0001 <.0001 <.0001 0.1239 0.4587 0.1833 0.7557 <.0001 35 0.0012 0.1724 <.0001 <.0001 <.0001 <.0001 0.6679 0.0047 0.2961 <.0001 <.0001 <.0001 <.0001 0.8954 36 0.9233 0.0002 <.0001 <.0001 <.0001 <.0001 0.0028 0.8571 0.0009 <.0001 <.0001 <.0001 <.0001 0.0094 37 0.0002 0.8708 0.0034 <.0001 <.0001 0.0001 0.3681 0.0006 0.7609 0.0105 <.0001 <.0001 0.0004 0.2938 38 <.0001 0.0010 0.8386 0.1505 0.0363 0.4523 <.0001 <.0001 0.0013 0.7033 0.1387 0.0501 0.5324 <.0001 39 <.0001 <.0001 0.1348 0.8844 0.4196 0.5868 <.0001 <.0001 <.0001 0.1122 0.7856 0.4502 0.5571 <.0001 40 <.0001 <.0001 0.0218 0.5080 1.0000 0.1775 <.0001 <.0001 <.0001 0.0196 0.6289 1.0000 0.1802 <.0001 41 <.0001 <.0001 0.2989 0.5455 0.2060 0.9329 <.0001 <.0001 <.0001 0.2454 0.4830 0.2367 0.8748 <.0001 42 0.0019 0.6065 0.0003 <.0001 <.0001 <.0001 0.8236 0.0047 0.7415 0.0015 <.0001 <.0001 <.0001 0.6767 43 0.0001 <.0001 <.0001 <.0001 <.0001 0.0018 0.9233 0.0006 <.0001 <.0001 <.0001 <.0001 0.0074 44 0.0001 0.0009 <.0001 <.0001 <.0001 0.4256 0.0004 0.8708 0.0038 <.0001 <.0001 <.0001 0.3370 45 <.0001 0.0009 0.0767 0.0133 0.3025 <.0001 <.0001 0.0012 0.8386 0.0744 0.0218 0.3839 <.0001 46 <.0001 <.0001 0.0767 0.4747 0.4570 <.0001 <.0001 <.0001 0.0656 0.8844 0.5080 0.4396 <.0001

244

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 19 (G01-R2b) : PROC GLM with CLASS varbs of ASSOC & LFORM and MODEL of TLFORMS TLFORMC = ASSOC¦LFORM from data set MWVRA 13:00 Monday, June 8, 2009

The GLM Procedure Least Squares Means

Least Squares Means for effect ASSOC*LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMC i/j 1 2 3 4 5 6 7 8 9 10 11 12 13 14

47 <.0001 <.0001 0.0004 0.5812 0.8564 <.0001 <.0001 <.0001 <.0001 0.0002 0.1232 0.9002 <.0001 <.0001 48 <.0001 <.0001 0.0404 0.3243 0.1755 <.0001 <.0001 <.0001 <.0001 0.0451 0.8424 0.1065 <.0001 <.0001 49 0.0204 0.5604 0.0007 <.0001 <.0001 0.2000 0.9350 0.1809 0.6319 <.0001 <.0001 <.0001 0.7822 0.3771 50 0.4159 0.0006 <.0001 <.0001 <.0001 <.0001 0.0043 0.0388 0.0003 <.0001 <.0001 <.0001 0.0027 <.0001 51 0.0065 0.9467 0.0115 <.0001 <.0001 0.5638 0.4988 0.0621 0.8125 0.0021 <.0001 <.0001 0.3600 0.8992 52 <.0001 0.0011 0.4973 0.0350 0.0146 0.0086 0.0001 <.0001 0.0002 0.6427 0.1286 0.0055 <.0001 0.0007 53 <.0001 <.0001 0.0061 0.9732 0.7083 <.0001 <.0001 <.0001 <.0001 0.0059 0.4301 0.6265 <.0001 <.0001 54 <.0001 <.0001 0.0011 0.6124 0.8680 <.0001 <.0001 <.0001 <.0001 0.0009 0.1657 0.9102 <.0001 <.0001 55 <.0001 <.0001 0.0723 0.3205 0.1827 0.0002 <.0001 <.0001 <.0001 0.0880 0.7823 0.1215 <.0001 <.0001 56 0.0572 0.4427 0.0008 <.0001 <.0001 0.1588 0.8743 0.3445 0.4909 <.0001 <.0001 <.0001 0.9919 0.2929

Least Squares Means for effect ASSOC*LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMC i/j 15 16 17 18 19 20 21 22 23 24 25 26 27 28

47 <.0001 <.0001 0.0045 0.4282 1.0000 <.0001 <.0001 <.0001 <.0001 0.0013 0.0875 1.0000 <.0001 <.0001 48 <.0001 0.0004 0.0951 0.6901 0.2341 <.0001 0.0001 <.0001 0.0002 0.0751 0.7990 0.1454 <.0001 0.0002 49 0.2630 0.5217 0.0106 <.0001 <.0001 0.1762 0.7122 0.1054 0.1551 0.0007 <.0001 <.0001 0.1412 0.1718 50 0.1936 0.0032 <.0001 <.0001 <.0001 0.2808 0.0071 0.1864 <.0001 <.0001 <.0001 <.0001 0.1447 <.0001 51 0.1234 0.8992 0.0509 0.0001 <.0001 0.0782 0.8954 0.0385 0.4573 0.0102 <.0001 <.0001 0.0532 0.4891 52 <.0001 0.0148 0.5366 0.1784 0.0366 <.0001 0.0072 <.0001 0.0201 0.6210 0.3561 0.0127 <.0001 0.0175 53 <.0001 <.0001 0.0230 0.7491 0.6656 <.0001 <.0001 <.0001 <.0001 0.0135 0.2851 0.6054 <.0001 <.0001 54 <.0001 <.0001 0.0070 0.4522 1.0000 <.0001 <.0001 <.0001 <.0001 0.0029 0.1135 1.0000 <.0001 <.0001 55 <.0001 0.0009 0.1301 0.6536 0.2305 <.0001 0.0004 <.0001 0.0007 0.1175 0.8796 0.1521 <.0001 0.0006 56 0.3810 0.4272 0.0091 <.0001 <.0001 0.2723 0.5917 0.2011 0.1240 0.0008 <.0001 <.0001 0.2538 0.1370

Least Squares Means for effect ASSOC*LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

245

Dependent Variable: TLFORMC i/j 29 30 31 32 33 34 35 36 37 38 39 40 41 42

47 <.0001 <.0001 0.0009 0.3296 1.0000 0.1400 <.0001 <.0001 <.0001 0.0363 0.4196 1.0000 0.2060 <.0001 48 <.0001 <.0001 0.1030 0.4587 0.0869 0.8118 <.0001 <.0001 0.0001 0.4523 0.5868 0.1775 0.9329 <.0001 49 0.2599 0.0580 <.0001 <.0001 <.0001 <.0001 0.6679 0.0028 0.3681 <.0001 <.0001 <.0001 <.0001 0.8236 50 0.0271 <.0001 <.0001 <.0001 <.0001 <.0001 0.0047 0.8571 0.0006 <.0001 <.0001 <.0001 <.0001 0.0047 51 0.0947 0.2915 0.0008 <.0001 <.0001 <.0001 0.2961 0.0009 0.7609 0.0013 <.0001 <.0001 <.0001 0.7415 52 <.0001 0.0140 0.8841 0.0468 0.0044 0.1239 <.0001 <.0001 0.0105 0.7033 0.1122 0.0196 0.2454 0.0015 53 <.0001 <.0001 0.0158 0.7736 0.5540 0.4587 <.0001 <.0001 <.0001 0.1387 0.7856 0.6289 0.4830 <.0001 54 <.0001 <.0001 0.0028 0.3794 1.0000 0.1833 <.0001 <.0001 <.0001 0.0501 0.4502 1.0000 0.2367 <.0001 55 <.0001 0.0003 0.1693 0.4446 0.1011 0.7557 <.0001 <.0001 0.0004 0.5324 0.5571 0.1802 0.8748 <.0001 56 0.4465 0.0497 <.0001 <.0001 <.0001 <.0001 0.8954 0.0094 0.2938 <.0001 <.0001 <.0001 <.0001 0.6767

246

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 20 (G01-R2b) : PROC GLM with CLASS varbs of ASSOC & LFORM and MODEL of TLFORMS TLFORMC = ASSOC¦LFORM from data set MWVRA 13:00 Monday, June 8, 2009

The GLM Procedure Least Squares Means

Least Squares Means for effect ASSOC*LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMC

i/j 43 44 45 46 47 48 49 50 51 52 53 54 55 56

47 <.0001 <.0001 0.0133 0.4747 0.1454 <.0001 <.0001 <.0001 0.0127 0.6054 1.0000 0.1521 <.0001 48 <.0001 <.0001 0.3025 0.4570 0.1454 <.0001 <.0001 <.0001 0.2470 0.4044 0.1775 0.9329 <.0001 49 0.0018 0.4256 <.0001 <.0001 <.0001 <.0001 0.0051 0.5651 0.0003 <.0001 <.0001 <.0001 0.8236 50 0.9233 0.0004 <.0001 <.0001 <.0001 <.0001 0.0051 0.0016 <.0001 <.0001 <.0001 <.0001 0.0155 51 0.0006 0.8708 0.0012 <.0001 <.0001 <.0001 0.5651 0.0016 0.0043 <.0001 <.0001 0.0001 0.4554 52 <.0001 0.0038 0.8386 0.0656 0.0127 0.2470 0.0003 <.0001 0.0043 0.0632 0.0196 0.3151 0.0003 53 <.0001 <.0001 0.0744 0.8844 0.6054 0.4044 <.0001 <.0001 <.0001 0.0632 0.6289 0.3905 <.0001 54 <.0001 <.0001 0.0218 0.5080 1.0000 0.1775 <.0001 <.0001 <.0001 0.0196 0.6289 0.1802 <.0001 55 <.0001 <.0001 0.3839 0.4396 0.1521 0.9329 <.0001 <.0001 0.0001 0.3151 0.3905 0.1802 <.0001 56 0.0074 0.3370 <.0001 <.0001 <.0001 <.0001 0.8236 0.0155 0.4554 0.0003 <.0001 <.0001 <.0001

NOTE: To ensure overall protection level, only probabilities associated with pre-planned comparisons should be used.

247

Appendix 3. SAS output using the General Linear Model (GLM) Procedure on associations and life form (species level and cover level)

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 1 (G01-R3a) : PROC GLM with CLASS varbs of GROUP & LFORM and MODEL of TLFORMS TLFORMC = GROUP¦LFORM from data set MWVRA 10:38 Thursday, June 18, 2009

The GLM Procedure

Class Level Information

Class Levels Values

GROUP 3 MR TK WRK

LFORM 7 Chamaeph Cryptoph Hemicryp Liana Parasite Phanerop Therophy

Number of observations 280

248

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 2 (G01-R3a) : PROC GLM with CLASS varbs of GROUP & LFORM and MODEL of TLFORMS TLFORMC = GROUP¦LFORM from data set MWVRA 10:38 Thursday, June 18, 2009

The GLM Procedure

Dependent Variable: TLFORMS

Sum of Source DF Squares Mean Square F Value Pr > F

Model 20 79.37425240 3.96871262 71.71 <.0001

Error 259 14.33424940 0.05534459

Corrected Total 279 93.70850180

R-Square Coeff Var Root MSE TLFORMS Mean

0.847034 26.37744 0.235254 0.891877

Source DF Type I SS Mean Square F Value Pr > F

GROUP 2 0.57058753 0.28529376 5.15 0.0064 LFORM 6 75.09709099 12.51618183 226.15 <.0001 GROUP*LFORM 12 3.70657388 0.30888116 5.58 <.0001

Source DF Type II SS Mean Square F Value Pr > F

GROUP 2 0.57058753 0.28529376 5.15 0.0064 LFORM 6 75.09709099 12.51618183 226.15 <.0001 GROUP*LFORM 12 3.70657388 0.30888116 5.58 <.0001

Source DF Type III SS Mean Square F Value Pr > F

GROUP 2 0.57058753 0.28529376 5.15 0.0064 LFORM 6 62.11432032 10.35238672 187.05 <.0001 GROUP*LFORM 12 3.70657388 0.30888116 5.58 <.0001

249

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 3 (G01-R3a) : PROC GLM with CLASS varbs of GROUP & LFORM and MODEL of TLFORMS TLFORMC = GROUP¦LFORM from data set MWVRA 10:38 Thursday, June 18, 2009

The GLM Procedure

Dependent Variable: TLFORMC

Sum of Source DF Squares Mean Square F Value Pr > F

Model 20 82.9775425 4.1488771 48.56 <.0001

Error 259 22.1304633 0.0854458

Corrected Total 279 105.1080057

R-Square Coeff Var Root MSE TLFORMC Mean

0.789450 37.95363 0.292311 0.770180

Source DF Type I SS Mean Square F Value Pr > F

GROUP 2 1.03844985 0.51922492 6.08 0.0026 LFORM 6 75.90779762 12.65129960 148.06 <.0001 GROUP*LFORM 12 6.03129499 0.50260792 5.88 <.0001

Source DF Type II SS Mean Square F Value Pr > F

GROUP 2 1.03844985 0.51922492 6.08 0.0026 LFORM 6 75.90779762 12.65129960 148.06 <.0001 GROUP*LFORM 12 6.03129499 0.50260792 5.88 <.0001

Source DF Type III SS Mean Square F Value Pr > F

GROUP 2 1.03844985 0.51922492 6.08 0.0026 LFORM 6 68.78232713 11.46372119 134.16 <.0001 GROUP*LFORM 12 6.03129499 0.50260792 5.88 <.0001

250

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 4 (G01-R3a) : PROC GLM with CLASS varbs of GROUP & LFORM and MODEL of TLFORMS TLFORMC = GROUP¦LFORM from data set MWVRA 10:38 Thursday, June 18, 2009

The GLM Procedure Least Squares Means

TLFORMS Standard LSMEAN GROUP LSMEAN Error Pr > ¦t¦ Number

MR 0.92903154 0.02156573 <.0001 1 TK 0.80081768 0.03360776 <.0001 2 WRK 0.89223910 0.02222944 <.0001 3

Least Squares Means for effect GROUP Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMS

i/j 1 2 3

1 0.0015 0.2359 2 0.0015 0.0241 3 0.2359 0.0241

TLFORMC Standard LSMEAN GROUP LSMEAN Error Pr > ¦t¦ Number

MR 0.84088062 0.02679612 <.0001 1 TK 0.72541294 0.04175873 <.0001 2 WRK 0.71464536 0.02762081 <.0001 3

Least Squares Means for effect GROUP Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMC

i/j 1 2 3

1 0.0207 0.0012 2 0.0207 0.8299 3 0.0012 0.8299

251

NOTE: To ensure overall protection level, only probabilities associated with pre-planned comparisons should be used.

TLFORMS Standard LSMEAN LFORM LSMEAN Error Pr > ¦t¦ Number

Chamaeph 1.50345838 0.04030574 <.0001 1 Cryptoph 1.38126478 0.04030574 <.0001 2 Hemicryp 1.02607088 0.04030574 <.0001 3 Liana 0.42609088 0.04030574 <.0001 4 Parasite 0.02049258 0.04030574 0.6116 5 Phanerop 0.55069755 0.04030574 <.0001 6 Therophy 1.21013103 0.04030574 <.0001 7

252

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 5 (G01-R3a) : PROC GLM with CLASS varbs of GROUP & LFORM and MODEL of TLFORMS TLFORMC = GROUP¦LFORM from data set MWVRA 10:38 Thursday, June 18, 2009

The GLM Procedure Least Squares Means

Least Squares Means for effect LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMS i/j 1 2 3 4 5 6 7

1 0.0330 <.0001 <.0001 <.0001 <.0001 <.0001 2 0.0330 <.0001 <.0001 <.0001 <.0001 0.0029 3 <.0001 <.0001 <.0001 <.0001 <.0001 0.0014 4 <.0001 <.0001 <.0001 <.0001 0.0297 <.0001 5 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 6 <.0001 <.0001 <.0001 0.0297 <.0001 <.0001 7 <.0001 0.0029 0.0014 <.0001 <.0001 <.0001

TLFORMC Standard LSMEAN LFORM LSMEAN Error Pr > ¦t¦ Number

Chamaeph 1.62518610 0.05008119 <.0001 1 Cryptoph 1.07764311 0.05008119 <.0001 2 Hemicryp 0.54059641 0.05008119 <.0001 3 Liana 0.17012721 0.05008119 0.0008 4 Parasite 0.00698021 0.05008119 0.8893 5 Phanerop 0.69237746 0.05008119 <.0001 6 Therophy 1.20928031 0.05008119 <.0001 7

Least Squares Means for effect LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMC i/j 1 2 3 4 5 6 7

1 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 2 <.0001 <.0001 <.0001 <.0001 <.0001 0.0642 3 <.0001 <.0001 <.0001 <.0001 0.0330 <.0001

253

4 <.0001 <.0001 <.0001 0.0220 <.0001 <.0001 5 <.0001 <.0001 <.0001 0.0220 <.0001 <.0001 6 <.0001 <.0001 0.0330 <.0001 <.0001 <.0001 7 <.0001 0.0642 <.0001 <.0001 <.0001 <.0001

NOTE: To ensure overall protection level, only probabilities associated with pre-planned comparisons should be used.

TLFORMS Standard LSMEAN GROUP LFORM LSMEAN Error Pr > ¦t¦ Number

MR Chamaeph 1.43306342 0.05705755 <.0001 1 MR Cryptoph 1.49789479 0.05705755 <.0001 2 MR Hemicryp 0.97149790 0.05705755 <.0001 3

254

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 6 (G01-R3a) : PROC GLM with CLASS varbs of GROUP & LFORM and MODEL of TLFORMS TLFORMC = GROUP¦LFORM from data set MWVRA 10:38 Thursday, June 18, 2009

The GLM Procedure Least Squares Means

TLFORMS Standard LSMEAN GROUP LFORM LSMEAN Error Pr > ¦t¦ Number

MR Liana 0.40553491 0.05705755 <.0001 4 MR Parasite 0.06147773 0.05705755 0.2823 5 MR Phanerop 0.69473557 0.05705755 <.0001 6 MR Therophy 1.43901643 0.05705755 <.0001 7 TK Chamaeph 1.61012732 0.08891777 <.0001 8 TK Cryptoph 1.22575515 0.08891777 <.0001 9 TK Hemicryp 1.17115547 0.08891777 <.0001 10 TK Liana 0.38276543 0.08891777 <.0001 11 TK Parasite 0.00000000 0.08891777 1.0000 12 TK Phanerop 0.47779120 0.08891777 <.0001 13 TK Therophy 0.73812922 0.08891777 <.0001 14 WRK Chamaeph 1.46718439 0.05881358 <.0001 15 WRK Cryptoph 1.42014439 0.05881358 <.0001 16 WRK Hemicryp 0.93555928 0.05881358 <.0001 17 WRK Liana 0.48997231 0.05881358 <.0001 18 WRK Parasite 0.00000000 0.05881358 1.0000 19 WRK Phanerop 0.47956589 0.05881358 <.0001 20 WRK Therophy 1.45324743 0.05881358 <.0001 21

Least Squares Means for effect GROUP*LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMS i/j 1 2 3 4 5 6 7 8 9 10 11

1 0.4225 <.0001 <.0001 <.0001 <.0001 0.9412 0.0950 0.0508 0.0138 <.0001 2 0.4225 <.0001 <.0001 <.0001 <.0001 0.4663 0.2891 0.0106 0.0022 <.0001 3 <.0001 <.0001 <.0001 <.0001 0.0007 <.0001 <.0001 0.0168 0.0599 <.0001 4 <.0001 <.0001 <.0001 <.0001 0.0004 <.0001 <.0001 <.0001 <.0001 0.8295 5 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 0.0026 6 <.0001 <.0001 0.0007 0.0004 <.0001 <.0001 <.0001 <.0001 <.0001 0.0034 7 0.9412 0.4663 <.0001 <.0001 <.0001 <.0001 0.1065 0.0446 0.0118 <.0001 8 0.0950 0.2891 <.0001 <.0001 <.0001 <.0001 0.1065 0.0025 0.0006 <.0001

255

9 0.0508 0.0106 0.0168 <.0001 <.0001 <.0001 0.0446 0.0025 0.6645 <.0001 10 0.0138 0.0022 0.0599 <.0001 <.0001 <.0001 0.0118 0.0006 0.6645 <.0001 11 <.0001 <.0001 <.0001 0.8295 0.0026 0.0034 <.0001 <.0001 <.0001 <.0001 12 <.0001 <.0001 <.0001 0.0002 0.5611 <.0001 <.0001 <.0001 <.0001 <.0001 0.0026 13 <.0001 <.0001 <.0001 0.4946 0.0001 0.0410 <.0001 <.0001 <.0001 <.0001 0.4505 14 <.0001 <.0001 0.0281 0.0018 <.0001 0.6816 <.0001 <.0001 0.0001 0.0007 0.0051 15 0.6775 0.7081 <.0001 <.0001 <.0001 <.0001 0.7313 0.1812 0.0244 0.0059 <.0001 16 0.8748 0.3436 <.0001 <.0001 <.0001 <.0001 0.8180 0.0759 0.0694 0.0203 <.0001 17 <.0001 <.0001 0.6613 <.0001 <.0001 0.0036 <.0001 <.0001 0.0069 0.0280 <.0001 18 <.0001 <.0001 <.0001 0.3038 <.0001 0.0131 <.0001 <.0001 <.0001 <.0001 0.3155 19 <.0001 <.0001 <.0001 <.0001 0.4538 <.0001 <.0001 <.0001 <.0001 <.0001 0.0004 20 <.0001 <.0001 <.0001 0.3671 <.0001 0.0092 <.0001 <.0001 <.0001 <.0001 0.3647 21 0.8056 0.5863 <.0001 <.0001 <.0001 <.0001 0.8623 0.1424 0.0338 0.0086 <.0001

256

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 7 (G01-R3a) : PROC GLM with CLASS varbs of GROUP & LFORM and MODEL of TLFORMS TLFORMC = GROUP¦LFORM from data set MWVRA 10:38 Thursday, June 18, 2009

The GLM Procedure Least Squares Means

Least Squares Means for effect GROUP*LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMS i/j 12 13 14 15 16 17 18 19 20 21

1 <.0001 <.0001 <.0001 0.6775 0.8748 <.0001 <.0001 <.0001 <.0001 0.8056 2 <.0001 <.0001 <.0001 0.7081 0.3436 <.0001 <.0001 <.0001 <.0001 0.5863 3 <.0001 <.0001 0.0281 <.0001 <.0001 0.6613 <.0001 <.0001 <.0001 <.0001 4 0.0002 0.4946 0.0018 <.0001 <.0001 <.0001 0.3038 <.0001 0.3671 <.0001 5 0.5611 0.0001 <.0001 <.0001 <.0001 <.0001 <.0001 0.4538 <.0001 <.0001 6 <.0001 0.0410 0.6816 <.0001 <.0001 0.0036 0.0131 <.0001 0.0092 <.0001 7 <.0001 <.0001 <.0001 0.7313 0.8180 <.0001 <.0001 <.0001 <.0001 0.8623 8 <.0001 <.0001 <.0001 0.1812 0.0759 <.0001 <.0001 <.0001 <.0001 0.1424 9 <.0001 <.0001 0.0001 0.0244 0.0694 0.0069 <.0001 <.0001 <.0001 0.0338 10 <.0001 <.0001 0.0007 0.0059 0.0203 0.0280 <.0001 <.0001 <.0001 0.0086 11 0.0026 0.4505 0.0051 <.0001 <.0001 <.0001 0.3155 0.0004 0.3647 <.0001 12 0.0002 <.0001 <.0001 <.0001 <.0001 <.0001 1.0000 <.0001 <.0001 13 0.0002 0.0394 <.0001 <.0001 <.0001 0.9091 <.0001 0.9867 <.0001 14 <.0001 0.0394 <.0001 <.0001 0.0652 0.0207 <.0001 0.0160 <.0001 15 <.0001 <.0001 <.0001 0.5722 <.0001 <.0001 <.0001 <.0001 0.8671 16 <.0001 <.0001 <.0001 0.5722 <.0001 <.0001 <.0001 <.0001 0.6910 17 <.0001 <.0001 0.0652 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 18 <.0001 0.9091 0.0207 <.0001 <.0001 <.0001 <.0001 0.9005 <.0001 19 1.0000 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 20 <.0001 0.9867 0.0160 <.0001 <.0001 <.0001 0.9005 <.0001 <.0001 21 <.0001 <.0001 <.0001 0.8671 0.6910 <.0001 <.0001 <.0001 <.0001

TLFORMC Standard LSMEAN GROUP LFORM LSMEAN Error Pr > ¦t¦ Number

MR Chamaeph 1.51783710 0.07089587 <.0001 1 MR Cryptoph 1.15160079 0.07089587 <.0001 2 MR Hemicryp 0.60937326 0.07089587 <.0001 3 MR Liana 0.19197553 0.07089587 0.0072 4 MR Parasite 0.02094062 0.07089587 0.7679 5

257

MR Phanerop 1.23813553 0.07089587 <.0001 6 MR Therophy 1.15630147 0.07089587 <.0001 7 TK Chamaeph 1.82406298 0.11048323 <.0001 8 TK Cryptoph 1.10843758 0.11048323 <.0001 9 TK Hemicryp 0.48250491 0.11048323 <.0001 10 TK Liana 0.12088991 0.11048323 0.2749 11 TK Parasite 0.00000000 0.11048323 1.0000 12 TK Phanerop 0.27959215 0.11048323 0.0120 13 TK Therophy 1.26240303 0.11048323 <.0001 14 WRK Chamaeph 1.53365824 0.07307779 <.0001 15 WRK Cryptoph 0.97289095 0.07307779 <.0001 16 WRK Hemicryp 0.52991106 0.07307779 <.0001 17 WRK Liana 0.19751617 0.07307779 0.0073 18 WRK Parasite 0.00000000 0.07307779 1.0000 19 WRK Phanerop 0.55940468 0.07307779 <.0001 20

258

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 8 (G01-R3a) : PROC GLM with CLASS varbs of GROUP & LFORM and MODEL of TLFORMS TLFORMC = GROUP¦LFORM from data set MWVRA 10:38 Thursday, June 18, 2009

The GLM Procedure Least Squares Means

TLFORMC Standard LSMEAN GROUP LFORM LSMEAN Error Pr > ¦t¦ Number

WRK Therophy 1.20913642 0.07307779 <.0001 21

Least Squares Means for effect GROUP*LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMC i/j 1 2 3 4 5 6 7 8 9 10 11

1 0.0003 <.0001 <.0001 <.0001 0.0057 0.0004 0.0204 0.0020 <.0001 <.0001 2 0.0003 <.0001 <.0001 <.0001 0.3889 0.9626 <.0001 0.7426 <.0001 <.0001 3 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 0.0002 0.3347 0.0002 4 <.0001 <.0001 <.0001 0.0892 <.0001 <.0001 <.0001 <.0001 0.0278 0.5886 5 <.0001 <.0001 <.0001 0.0892 <.0001 <.0001 <.0001 <.0001 0.0005 0.4471 6 0.0057 0.3889 <.0001 <.0001 <.0001 0.4151 <.0001 0.3241 <.0001 <.0001 7 0.0004 0.9626 <.0001 <.0001 <.0001 0.4151 <.0001 0.7157 <.0001 <.0001 8 0.0204 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 9 0.0020 0.7426 0.0002 <.0001 <.0001 0.3241 0.7157 <.0001 <.0001 <.0001 10 <.0001 <.0001 0.3347 0.0278 0.0005 <.0001 <.0001 <.0001 <.0001 0.0214 11 <.0001 <.0001 0.0002 0.5886 0.4471 <.0001 <.0001 <.0001 <.0001 0.0214 12 <.0001 <.0001 <.0001 0.1448 0.8734 <.0001 <.0001 <.0001 <.0001 0.0022 0.4398 13 <.0001 <.0001 0.0126 0.5051 0.0499 <.0001 <.0001 <.0001 <.0001 0.1952 0.3107 14 0.0528 0.3994 <.0001 <.0001 <.0001 0.8535 0.4197 0.0004 0.3253 <.0001 <.0001 15 0.8766 0.0002 <.0001 <.0001 <.0001 0.0040 0.0003 0.0292 0.0015 <.0001 <.0001

Least Squares Means for effect GROUP*LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMC

i/j 12 13 14 15 16 17 18 19 20 21

1 <.0001 <.0001 0.0528 0.8766 <.0001 <.0001 <.0001 <.0001 <.0001 0.0027 2 <.0001 <.0001 0.3994 0.0002 0.0804 <.0001 <.0001 <.0001 <.0001 0.5725

259

3 <.0001 0.0126 <.0001 <.0001 0.0004 0.4358 <.0001 <.0001 0.6240 <.0001 4 0.1448 0.5051 <.0001 <.0001 <.0001 0.0010 0.9566 0.0605 0.0004 <.0001 5 0.8734 0.0499 <.0001 <.0001 <.0001 <.0001 0.0841 0.8372 <.0001 <.0001 6 <.0001 <.0001 0.8535 0.0040 0.0097 <.0001 <.0001 <.0001 <.0001 0.7760 7 <.0001 <.0001 0.4197 0.0003 0.0728 <.0001 <.0001 <.0001 <.0001 0.6043 8 <.0001 <.0001 0.0004 0.0292 <.0001 <.0001 <.0001 <.0001 <.0001 <.0001 9 <.0001 <.0001 0.3253 0.0015 0.3071 <.0001 <.0001 <.0001 <.0001 0.4478 10 0.0022 0.1952 <.0001 <.0001 0.0003 0.7207 0.0324 0.0003 0.5621 <.0001 11 0.4398 0.3107 <.0001 <.0001 <.0001 0.0022 0.5635 0.3623 0.0011 <.0001 12 0.0747 <.0001 <.0001 <.0001 <.0001 0.1372 1.0000 <.0001 <.0001 13 0.0747 <.0001 <.0001 <.0001 0.0599 0.5361 0.0358 0.0356 <.0001 14 <.0001 <.0001 0.0416 0.0297 <.0001 <.0001 <.0001 <.0001 0.6879 15 <.0001 <.0001 0.0416 <.0001 <.0001 <.0001 <.0001 <.0001 0.0019

260

Prof MW van Rooyen - Research Project - T09053 - PLK0086 - PB971871 9 (G01-R3a) : PROC GLM with CLASS varbs of GROUP & LFORM and MODEL of TLFORMS TLFORMC = GROUP¦LFORM from data set MWVRA 10:38 Thursday, June 18, 2009

The GLM Procedure Least Squares Means

Least Squares Means for effect GROUP*LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMC

i/j 1 2 3 4 5 6 7 8 9 10 11

16 <.0001 0.0804 0.0004 <.0001 <.0001 0.0097 0.0728 <.0001 0.3071 0.0003 <.0001 17 <.0001 <.0001 0.4358 0.0010 <.0001 <.0001 <.0001 <.0001 <.0001 0.7207 0.0022 18 <.0001 <.0001 <.0001 0.9566 0.0841 <.0001 <.0001 <.0001 <.0001 0.0324 0.5635 19 <.0001 <.0001 <.0001 0.0605 0.8372 <.0001 <.0001 <.0001 <.0001 0.0003 0.3623 20 <.0001 <.0001 0.6240 0.0004 <.0001 <.0001 <.0001 <.0001 <.0001 0.5621 0.0011 21 0.0027 0.5725 <.0001 <.0001 <.0001 0.7760 0.6043 <.0001 0.4478 <.0001 <.0001

Least Squares Means for effect GROUP*LFORM Pr > ¦t¦ for H0: LSMean(i)=LSMean(j)

Dependent Variable: TLFORMC

i/j 12 13 14 15 16 17 18 19 20 21

16 <.0001 <.0001 0.0297 <.0001 <.0001 <.0001 <.0001 <.0001 0.0231 17 <.0001 0.0599 <.0001 <.0001 <.0001 0.0015 <.0001 0.7756 <.0001 18 0.1372 0.5361 <.0001 <.0001 <.0001 0.0015 0.0571 0.0005 <.0001 19 1.0000 0.0358 <.0001 <.0001 <.0001 <.0001 0.0571 <.0001 <.0001 20 <.0001 0.0356 <.0001 <.0001 <.0001 0.7756 0.0005 <.0001 <.0001 21 <.0001 <.0001 0.6879 0.0019 0.0231 <.0001 <.0001 <.0001 <.0001

NOTE: To ensure overall protection level, only probabilities associated with pre-planned comparisons should be used.

261