Bibliography: Land Degradation in South Africa project

environmental variables. It is similarly anticipated that the optima and shape

of one side of the response curve will be identified.

Ref ID : 1080

1538. O'Connor, T.G. Patch colonisation in a savanna grassland. Journal of

Vegetation Science 2:245-254, 1991.

Reprint : In File,

Keywords : GRAZING EFFECTS; SAVANNA; GRASSLAND; VEGETATION DYNAMICS; RAINFALL

Abstract : Patches (1 m diameter) were cleared in a heavily grazed and lightly

grazed savanna in South Africa; half of them were covered once with grass

litter. The greatest colonisation over three years occurred on heavily grazed

patches with litter, the least on open, lightly grazed patches. Annual rainfall

affected colonisation rate. Basal cover was lower on patches than in the

surrounding vegetation. Patch composition was weakly affected by treatment, and

was partly related to composition of adjacent vegetation, but the proportional

representation of species on patches differed from the surrounding vegetation.

The stoloniferous Digitaria eriantha and the obligate seed reproducers Aristida

bipartita and Heteropogon contortus were major components of patch communities

but were uncommon in the surrounding vegetation and in the seed bank. Setaria

incrassata and Themeda established well only on heavily grazed patches with

litter. The pattern of seedling establishment was the same as that of

colonisation. Most seedlings emerged at the beginning of the wet season, with

ca. 50% mortality soon thereafter. The colonising species can be partly

predicted from the availability of propagules (vegetative or seed), emergence

and establishment success, and subsequent growth.

Ref ID : 1064

1539. O'Connor, T.G. Influence of rainfall and grazing on the compositional

change of the herbaceous layer of a sandveld savanna. Journal of the Grassland

Society of southern Africa 8(3):103-109, 1991.

Reprint : Not in File,

Keywords : SAVANNA; RAINFALL; GRAZING EFFECTS; VEGETATION CHANGE; GAZANKULU

Notes : Only copied p.103.

Abstract : The influence of heavy grazing by cattle for three years, and of

rainfall interception for one year, on the herbaceous composition of a sandveld

savanna in Gazankulu was investigated. The composition and the abundance of the

predominant species changed substantially, owing mainly to rainfall.

Pogonarthria squarrosa, considered to increase under grazing, was initially the

most abundant species, but decline most rapidly on the heavily grazed, rainfall

interception treatment, owing to a high mortality. Digitaria eriantha increased

markely on the heavily-grazed treatment, despite up to o24% tuft mortality,

because of extensive vegetative recruitment from stoloniferous growth. Aristida

congesta subsp. barbicollis and Aristida stipitata, considered to increase under

heavy grazing, increased on the lightly-grazed treatment. There was an

explosion of annual forbs, predominantly Tephrosia purpurea, Indigofera costata

and Cassia absus, in the third year. The trend of rainfall during the study,

initially wet and successively drier, apparently had a greater influence on

compositional trends than the imposed treatments.

Ref ID : 1076

1540. O'Connor, T.G. Local extinction in perennial grasslands: a life-history

approach. The American Naturalist 137(6):753-773, 1991.

Reprint : In File,

Keywords : GRASSLAND; GRAZING EFFECTS; VEGETATION DYNAMICS; RAINFALL; DROUGHT;

POPULATION; MODELS

Abstract : A common element of many perennial grasslands, defined as an

extinction-prone perennial grass, is a palatable obligate seed reproducer,

producing low numbers of larger diaspores that are poorly dispersed. Such a

species can become locally extinct on a regional or landscape scale in graslands

that experience both a variable rainfall, in which drought is characteristic,

and sustained heavy grazing. The changes in the temporal and spatial patterns

of grazing in human-influenced systems has created the possibility of local

extinction. Catastrophic mortality of established populations occurs as a

consequence of droughts in combination with sustained grazing, which eventually

leads to the elimination of the population. Elimination of the established

population is more likely to occur in semarid than moist areas because of the

greater frequency and duration of droughts. The maintencance or reestablishment

of the population depends on successful recruitment, but seedling mortality is

high because of intraseasonal drought and trampling. The seed bank of an

extinction-prone species is rapidly eliminated if the input of seed is

curtailed. These species do not form persistent seed banks because the seeds

have a limited (2 - 3 yrs) intrinsic capacity for survival when in a state of

induced or enforced dormancy. Innate dormancy lasts for only a few months.

Therefore, the size of the seed bank is determined primarily by the annual input

of seed. But even under favourable rainfall and light grazing, 100% of seed

produced can be lost to predation. Seed production can fail from drought and

from the direct consumption of inflorescences. An extinction-prone species is

self-dispersed and therefore poorly dispersed. The rate and scale of incidental

animal dispersal are unlikely to assist in the maintenance or reestablishment

of a population because of the low probability of seedling survival. The

decline of an extinction-prone species under heavy grazing is compounded by the

concomitant competitive influences of invasive species and unpalatable species,

which both benefit from heavy grazing. Irreversible changes in the

physicochemical environment may also result in local extinction, but these

processes are not mutually exclusive. The autogenic model of secondary

succession is inadequate for these grasslands if the present concept of local

extinction is upheld.

Ref ID : 1048

1541. O'Connor, T.G. Patterns of plant selection by grazing cattle in two

savanna grasslands: a plant's eye view. Journal of the Grassland Society of

southern Africa 9(3):97-104, 1992.

Reprint : In File,

Keywords : SAVANNA; GRASSLAND; GRAZING EFFECTS; VELD CONDITION

Abstract : The amount of grazing experienced by individual tufts of the most

abundant perennial grasses of two Transvaal savannas was monitored at regular

intervals for two growing seasons. In general, Setaria incrassata and Themeda

triandra were heavily grazed, Bothriochloa insculpta and Aristida bipartita were

lightly grazed, and Heteropogon contortus and Digitaria eriantha were

intermediate. Tufts with a greater amount of moribund material were less

heavily grazed. The smallest tufts of all species were the most lightly grazed,

although the most heavily-grazed tufts were on occasion intermediate in size

rather than being the largest. A tuft had an increased likelihood of being

heavily grazed if it had previously been heavily grazed, both within a growing

season and for two successive growing seasons. A spatial pattern of grazing was

evident, as individual species were less heavily grazed at different locations

within the paddocks. The influence of species identity on the amount a tuft was

grazed was not absolute but rather quantitative and conditional on the level of

other variables. Tuft size and the amount a tuft had been previously grazed

generally had a greater influence on the amount a tuft was grazed during an

individual grazing event than species identity, location or moribund material.

Species selection is possibly, in part, a consequence of selection for plant

structure which optimizes intake.

Ref ID : 1825

1542. O'Connor, T.G. Composition and population responses of an African savanna

grassland to rainfall and grazing. Journal of Applied Ecology 31:155-171, 1994.

Reprint : In File,

Keywords : POPULATION; SAVANNA; GRASSLAND; RAINFALL; DROUGHT

Notes : (1) The influence of inter-annual rainfall variability and cattle

grazing on the abundance (density, basal area), population dynamics

(recruitment, growth, survival) and population structure of the dominant

perennial grasses of an African savanna was investigated from 1984 to 1989. The

unpalatable Aristida bipartita and the palatable Heteropogon contortus and

Themeda triandra reproduce by seed; Bothriochloa insculpta reproduces mainly by

seed, but sometimes by stolons; Digitaria eriantha can be stoloniferous; Setaria

incrassata has rhizomatous growth. (2) Species abundance was more responsive

to rainfall variability than to grazing. Density increased in all species in

response to several successive dry years because of recruitment and tuft

fragmentation, but eventually decreased in a severe final drought year because

of plant mortality. The response of basal cover differed between species and

sites, although all species (except Digitaria) declined in the final drought

year. (3) Seedling recruitment, plant basal area, and mortality were likewise

more strongly influenced by rainfall variability than by grazing. Differences

in the population dynamics of species were related to their life history

attributes. Aristida, a prolific seed producer, was favoured by grazing because

high levels of seedling recruitment could be maintained. Seedling recruitment of

palatable species (Heteropogon, Themeda) and of Bothriochloa was initially

increased under heavy grazing because openings were produced within the

vegegation, but this was reversed by the end of the study because of a decline

in available seed. There was limited seedling recruitment of Digitaria and

Setaria, but recruitment of Digitaria from stoloniferous growth increased

population size, although there was no expansion of Setaria clones under the

conditions of this study. Themeda and Aristida seedlings attained greater sizes

under heavy than under light grazing. (4) Mortality of all species increased

over time, and was generally highest in the final severe drought year. Almost

all mortality occurred in small individuals (<2.5 cm diameter) which had

recently recruited or which had fragmented from existing tufts. Fragments of

Aristida or Heteropogon were more likely to die than young tufts of the same

size. The direction of the effect of grazing on mortality depended on year.

Mean tuft size of most species decreased over time due to seedling recruitment

and tuft fragmentation. (5) The fraction of basal area after 5 years comprising

individuals present at the start of the study, an index of population turnover,

was greater than half for the seed reproducing species and Bothriochloa, and

almost unity for Setaria (>98%) and Digitaria (94%). This fraction for seed-

reproducing species was slightly less under heavy grazing. (6) Thus, the

pattern of community change was determined by the unique responses of

recruitment, growth and mortality to rainfall variability and grazing among

species. Heteropogon was almost eliminated and Themeda was markedly reduced by

the combination of heavy grazing and drought. This supports the idea that

palatable, seed-reproducing species may be prone to local extinction under such

conditions because of increased mortality of mature plants and the diminishing

availability of seed leading to diminished recruitment. The unpalatable

Aristida showed irruptive behaviour under the same conditions because it could

maintain seedling recruitment. Long-term experiments are required to separate

the effects of rainfall and grazing on the demography of savanna grasses.

Ref ID : 239

1543. O'Connor, T.G. Acacia karroo invasion of grassland: effects of cattle

dung, shading, moisture availability and the grass sward on seedling emergence

and establishment. 1994.

Reprint : In File,

Keywords : GRASSLAND; BUSH ENCROACHMENT; VEGETATION DYNAMICS; EASTERN CAPE;

RAINFALL

Notes : Probably submitted somewhere already.

Abstract : Accacia karroo Hayne is the most important woody invader of grassland

in SA, and can greatly reduce the productivity of the grassland. A field

experiment was conducted to test the hypotheses that emergence, growth and the

1st year's survival of Acadia karroo would be enhanced by (1) defoliation of the

grass sward, (2) increased irradiance, (3) increased moisture availability and

(4) its germination within cattle dung pats. The study was conducted on one

site above and one below the natural altitudinal treeline of this species in

grassland of the eastern Cape, SA. Not one seedling emerged from dung pats.

Neigher location nor the other treatments affected the density of emerging

seedlings, although only 40.4 seedlings m-2 emerged of the 200 seeds m-2

planted. Shading dramatically increased the density of surviving seedlings. In

the open, only 3 and 1.5 seedlings m-2 remained respectively at the end of the

growing season or the beginning of the next, compared to 23.3 and 19,5 seedlings

m-2 under shading for these respective times. This was attributed to the effect

of shade on moisture availability in a season which recieved only 54% of average

rainfall. Seedling survival until the end of the growing season was enhanced

(30%) by shade at both sites, but also by supplemental water (24%) and

defoliation of the sward (7%) at the site above the treeline. Across sites and

treatments, seedling survival was related to moisture availability, with no or

poor survival for <500 mm rainfall, indicating this species can only establish

in certain rainfall years. Seedling survival over winter was not influenced by

treatment, but was greater for larger seedlings. Treatments affected seedling

size, in particular seedlings growing under shade and within a dense grass

canopy were etiolated. A.karroo seedlings are capable of establishing and

surviving within a dense grass sward for at least a year, tolerant of low

irradiance and of interference, which, because most seeds do not persist for

much longer than a year, suggests this species forms predominantly a seedling

bank. This has implications for the invasion of grassland by woody species.

Ref ID : 1827

1544. O'Connor, T.G. Transformation of a savanna grassland by drought and

grazing. African Journal of Range and Forage Science 12(2):53-60, 1995.

Reprint : In File,

Keywords : SAVANNA; GRASSLAND; DROUGHT; POPULATION

Abstract : The relative effects of drought and heavy grazing on the floristic

composition, population size and structure, and basal cover of an African

savanna grassland were differentiated by comparing changes over eight years,

which included a severe drought year, across a gradient of grazing history.

Drought had an overriding effect on community change, but grazing history had an

additional effect. Severe drought in combination with a history of severe

grazing transformed grassland of predominantly palatable, perennial grasses

(Themeda triandra, Setaria incrassata, Heteropogon contortus) to grassland

dominated by the unpalatable perennial Aristida bipartita, annual grasses and

forbs. Palatable species were almost eliminated from the sward between tree

canopies, but residual populations were protected beneath certain woody species.

The most lightly grazed grassland maintained its character of palatable,

perennial grass species but was changed considerably in the relative proportion

of these species. After the drought, tuft size of palatable species was smaller

than any previously recorded but recovered thereafter. A number of annual and

perennial grass species were recorded for the first time subsequent to the

drought. Consideration of the life history attributes of species was useful for

predicting species' responses. Grazing management during drought episodes would

seem critical for determining the direction of community change. The notion

that savanna grasslands are insensitive to grazing because of their

disequilibrium behaviour was rejected.

Ref ID : 246

1545. O'Connor, T.G. and Bredenkamp, G.J. Grassland. In: Vegetation of Southern

Africa. edited by Cowling, R.M., Richardson, D.M., and Pierce,

S.M.Cambridge:Cambridge University Press, 1997,p. 215-257.

Reprint : In File,

Keywords : GRASSLAND; VEGETATION DYNAMICS; BOTANICAL SURVEY; RAINFALL; FIRE

Notes : The authors conclude the following. They have sought to illustrate that

the rainfall gradient across the grassland biome is a main determinant of

community composition, primary production, foliage nutrient content, nutrient

cycling and attributes of species such as photosynthetic pathway, secondary

chemicals and phenology. Rainfall in semi-arid regions, and hence production

and nutrient cycling, is more variable than in moister regions. Indeed,

rainfall regime seems to determine the distribution of the biome both directly

(i.e. water balance) and indirectly through fire regime, although biotic effects

of grazing can influence biome boundaries. A temperature gradient is also

undoubtedly important, and is partly independent of rainfall, although this

relationship has not been well investigated. Soil type is a critical modifier

of the influence of rainfall regime at a local or regional scale. Although all

grasslands of the biome comprise mainly tufted perennials, it is tentatively

suggested that semi-arid grassland has faster turnover of individual tufts,

because of the increased frequency of drought-related mortality, and therefore

has the potential for rapid compositional change. In contrast, tuft turnover

and change in high-rainfall patterns, grazing has a more immediate effect on

community change in semi-arid than moist grassland. Changes in community

composition can dramatically influence water balance, production, nutrient

cycling, foliage quality, soil loss and fire behaviour. Community change

depends on the influence of communities on the abiotic environment and on

species attributes, but the response of species to environment is contextual

rather than absolute.

Ref ID : 990

1546. O'Connor, T.G. and Pickett, G.A. The influence of grazing on seed

production and seed banks of some African savanna grasslands. Journal of Applied

Ecology 29:247-260, 1992.

Reprint : In File,

Keywords : VEGETATION DYNAMICS; GRAZING EFFECTS; VEGETATION CHANGE; SAVANNA;

GRASSLAND; POPULATION; RAINFALL; DROUGHT; MODELS

Abstract : (1) Species composition, population structure and the seed banks of

the perennial grasses Aristida bipartita, Bothriochloa insculpta, Heteropogon

contortus, Themeda triandra and Urochloa mosambicensis, were described for nine

sites in mesic savanna grasslands (seed banks for five) on a similar soil type

ranging in grazing history from sustained heavy grazing (> 20 years) to light

grazing. The influence of two seasons of grazing on the seed production and the

seed bank of Themeda, Setaria incrassata, Heteropogon, Aristida, Bothriochloa

and Digitaria eriantha, was estimated at two other sites. (2) Compositionnal

variation reflected grazing history. Lightly grazed sites were characterized by

the longer-lived, palatable perennials Themeda, Bothriochloa, Heteropogon and

Digitaria, and heavily grazed sites by the shorter-lived perennials Urochloa,

Sporobolus nitens, Chloris virgata, the unpalatable Aristida, and some forb

species. Grass tuft size declined with increasing grazing pressure. Sustained

heavy grazing reduced or eliminated Themeda, Bothriochloa and Heteropogon from

the seed bank, but increased the abundance of Aristida and Urochloa spp. (3)

Rainfall influenced the seasonal pattern of seed production, which decreased

with intra-seasonal drought. The number of seeds produced m-2 of tuft basal

area varied from 21 000 for Setaria to 1.9 million for Aristida. Annual seed

production m-2 of ground with light grazing was <400 for Themeda, for

Heteropogon, 1 300 - 4 000 for Setaria, > 10 000 for Aristida, and 300 for

Bothriochloa. (4) Seed viability varied from zero for Digitaria to 68% for

Setaria, and varied within a season for Themeda (24 - 49%) and Setaria (26 -

68%). The seeds of all species tested possessed a period of innate dormancy of

<1 year. (5) Seven to ten grass and six to seven forb species were found in

the seed banks at two sites. Perennial grasses constituted 85 - 95% of the seed

bank, with >90% of all seeds occurring in the top 2 cm of soil. Seed density

differed markedly between sites and grazing treatments. (6) A simple model of

the seed bank dynamics of the perennial grasses showed that because their seeds

germinate readily and have a limited survival when in secnodary dormancy, the

seed bank size is determined primarily by the abundance of the species in the

vegetation. Thus, the seed banks of Themeda and Heteropogon, which produce

small numbers of seeds, can easily be eliminated by sustained grazing. This was

also indicated by the distribution of these two species along the gradient of

grazing history.

Ref ID : 157

1547. O'Connor, T.G. and Roux, P.W. Vegetation changes (1949-71) in a semi-arid,

grassy dwarf shrubland in the Karoo, South Africa: influence of rainfall

variability and grazing by sheep. Journal of Applied Ecology 32:612-626, 1995.

Reprint : In File,

Keywords : VEGETATION CHANGE; NAMA KAROO; KAROO DESERTIFICATION; VEGETATION

DYNAMICS; STOCKING RATE; DROUGHT; MONITORING; RAINFALL; AID

Notes : (1) The influence of annual and seasonal (summer versus winter) rainfall

variability and of different seasons and systems of sheep grazing, including

complete protection, on the botanical species composition of a community

characterized by dwarf shrubs, annual and perennial grasses, and ephemeral

dicotyledons, was investigated for 23 years in the Karoo, South Africa. (2)

Ordination and bivariate techniques revealed dramatic changes in response to

interannual rainfall variability, mostly associated with annuals or short-lived

perennials. Annual rainfall showed a pronounced temporal pattern being wetter

in the 1950s and drier in the 1960s, and with annual rainfall correlated with

summer rainfall. (3) Annual grasses and ephemeral dicotyledons errupted in

certain years of suitable rainfall to an extent that was influenced by grazing