Snyman, Du Preez. 2005.

ARTICLE IN PRESS
Journal of
Arid
Environments
Journal of Arid Environments 60 (2005) 483–507
www.elsevier.com/locate/jnlabr/yjare
Rangeland degradation in a semi-arid South

Africa—II: influence on soil quality
H.A. Snymana,, C.C. du Preezb
a
Department of Animal, Wildlife and Grassland Sciences, P.O. Box 339, University of the Free State,
Bloemfontein 9300, South Africa
b
Department of Soil, Crop and Climate Sciences, P.O. Box 339, University of the Free State,
Bloemfontein 9300, South Africa
Received 28 March 2003; received in revised form 16 April 2004; accepted 15 June 2004
Available online 11 September 2004
Abstract
The impact of rangeland degradation on soil characteristics (compaction, temperature,

soil–water content, infiltrability, root and litter turnover, and the organic matter content), was
determined for a semi-arid rangeland. Sampling was from rangeland artificially maintained in
three different rangeland conditions, viz. good, moderate and poor. Due to the lower basal
cover of rangeland in poor condition, soil compaction increased (pp0.01) and temperature,
water-content, infiltrability and organic matter content decreased (pp0.01) with rangeland
degradation. The mean soil compaction of rangeland in good, moderate and poor conditions
was 6.37, 11.51 and 18.34 kg cm 2, respectively. The highest temperatures on top of the soil of
55, 49 and 46 1C for rangeland in poor, moderate and good conditions respectively, occurred
during December. Where rainfall is the biggest determining factor for production in rangeland
in poor condition, under higher soil–water conditions, nitrogen is for rangeland in good
condition. After only 5 years following degradation, organic C was significantly lower
(22.15%) over the first 50 mm soil layer and total N significantly lower (12.91%) over the first
100 mm in rangeland in poor condition than that of good condition rangeland. Rangeland
degradation lengthened the replacement of total root system with about a year and
decomposition time of litter with 8 months. The importance of maintaining rangeland in good
Corresponding author. Tel.: +27-51-401-2221; fax: +27-51-401-2608.

E-mail address: snymanha.sci@mail.uovs.ac.za (H.A. Snyman).
0140-1963/$ - see front matter r 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jaridenv.2004.06.005
ARTICLE IN PRESS
484 H.A. Snyman, C.C. du Preez / Journal of Arid Environments 60 (2005) 483–507
condition and soil quality to help ensure sustainable utilization of the grassland ecosystem was
stressed.
r 2004 Elsevier Ltd. All rights reserved.
Keywords: Infiltrability; Litter; Organic matter; Rangeland condition; Root turnover; Soil compaction;
Soil temperature; Soil–water content
1. Introduction
Concerns about the effects of agricultural management practices on the

environment have initiated interest in soil quality/health. Gregorich et al. (1994)
defined soil quality as ‘‘a composite measure of both a soil’s ability to function and
how well it functions, relative to a specific use’’. Rangeland soils represent an unique
challenge for soil quality assessment due to their high spatial and temporal
variability and multiple use demands (Manley et al., 1995). As rangeland managers
become increasingly concerned about the sustainability of rangeland resources, soil
responses to overgrazing and rangeland degradation need to be quantified to develop
suitable grazing practices. Overgrazing is considered the most important cause of
rangeland degradation. When the production potential of rangeland is over-
estimated the resultant overgrazing will cause a decrease of the palatable perennial
plants in favour of less palatable, undesirable vegetation (Van der Westhuizen et al.,
1999; Snyman, 2004). Such changes in rangeland condition usually lead to increased
soil compaction (Warren et al., 1986a, b; Thurow et al., 1988; Chanasyk and Naeth,
1995), reduced soil aggregate stability (Warren et al., 1986a, b; Russel et al., 2001;
Lal and Elliot, 1994), soil fertility (Dormaar and Willms, 1998; Ingram, 2002) and
soil organic matter content (Du Preez and Snyman, 1993, 2003; Whitford, 1996;
Snyman, 1999a). Soil organic matter is extremely important in rangeland ecosystem
functioning since it improves soil structure (Thurow et al., 1986) and thereby
enhances water infiltration (Smith et al., 1990) and reduction of soil erosion through
aggregate stabilization (Chevallier et al., 2001). The net effect of this is a better
water-use efficiency by rangeland (Reicosky et al., 1995; Williams et al., 1998;
Okatan and Reis, 1999; Snyman, 2005). Soil organic matter is also a critical factor in
soil fertility (Teague et al., 1999; Whitford, 1996). Few studies have evaluated the
effects of grazing or rangeland degradation on soil organic matter and its
relationship to water and nutrient cycling and related plant productivity (Milchunas
and Lauenroth, 1993; Manley et al., 1995; Schuman et al., 1999; Dormaar and
Willms, 1998; Emmerich and Heitschmidt, 2002). In general, the available data do
not indicate any single or consistent response of soil organic carbon (C) and nitrogen
(N) to grazing (Manley et al., 1995). Since a key component of overall ecosystem
sustainability occurs belowground, recovery is tied to the soils physical, chemical and
biological functions and processes (Singh and Coleman, 1973; Neary et al., 1999).
Prevailing soil–water potential, soil temperature (Drew, 1979, pp. 373–598; Distel
and Fernandez, 1988) and reproductive development (Troughton, 1978), are also
factors which greatly affect root growth and therefore influence soil quality. As
ARTICLE IN PRESS
H.A. Snyman, C.C. du Preez / Journal of Arid Environments 60 (2005) 483–507 485
growth limitations are imposed by low availability of water, low rates of net primary
productivity in arid and semi-arid ecosystems are most often attributed to low
availability of nutrients (Ingram, 2002; Schenk and Jackson, 2002; Holm, 2000). As
in nearly all terrestrial environments, most of the nutrients taken up from the soil by
plants in arid ecosystems arise from nutrient cycling rather than from parent
material per se (Charley and Cowling, 1968). As with many other biological
processes in arid systems, like plant uptake of nutrients and growth, decomposition
and mineralization are closely related to climate (Montana et al., 1988) and many
such systems can be characterized as producing ‘‘pulses’’ or ‘‘flushes’’ of nutrients
from mineralization during wet periods (Sparling and Ross, 1988; West et al., 1989;
Singh and Coleman, 1973). Factors affecting decomposition of plant material affect
nutrient cycling and therefore primary productivity (Ekaya and Kinyamario, 2001).
Nitrogen is considered the most limiting nutrient for plant growth in arid and semi-
arid ecosystems (Mazzarino and Bertiller, 1999).
Over the world, vegetation change is the most commonly described effect of
livestock production on rangelands. Numerous studies also described effects of
grazing on soils (Milchunas and Lauenroth, 1993; Emmerich and Heitschmidt,
2002), but conceptual understanding of responses of soil ecosystems still lags behind
that of above-ground systems (Allsopp, 1999; Neary et al., 1999; Snyman, 2004). In
arid and semi-arid rangelands, as soil and plant processes are controlled by rainfall
(Ludwig and Tongway, 1998, pp. 1–12; Ingram, 2002), one of the most important
principles in sustainable utilization of these areas is efficient soil–water management
(Snyman, 1998; Oesterheld et al., 2001). Understanding changes in hydrological
characteristics of the ecosystem under different rangeland conditions (Wiegand et al.,
2004) is therefore essential when making rangeland management decisions in these
areas to ensure sustainable animal production. Therefore, short- and long-term
studies are required to test interactions between rangeland degradation, climate and
vegetation change as affecting soil characteristics. The objective of this study was
therefore to determine the effect of rangeland degradation on different soil
characteristics to get an idea of soil quality range in a semi-arid climate.
2. Materials and methods
2.1. Site description
The research was conducted in Bloemfontein (28150’S; 26115;E, altitude 1350 m),
which is situated in the semi-arid (summer mean average 560 mm) region of South
Africa. Rain falls almost exclusively during summer (October–April), with a mean of
78 rainy days per year. Mean monthly maximum temperatures range from 17 1C in
July to 33 1C in January and a mean of 119 frost days per annum (Schulze, 1979).
The study area is situated in the Dry Sandy Highveld Grassland (Grassland
Biome) (Bredenkamp and Van Rooyen, 1996, p. 4) with a slope of 3.5%. The
botanical composition of the study site was determined from the 1995/96 to the 1998/
99 seasons by Snyman (1999b) and Snyman (2000). A mean rangeland condition
ARTICLE IN PRESS
score was expressed as a percentage of that in a benchmark site. Soils in the study
area are mostly fine sandy loams of the Bloemdal Form (Roodepoort family—3200)
(Soil Classification Working Group, 1991). Clay content increases with soil depth
from 10% in the A-horizon (0–300 mm, to 24% in the B1-horizon (300–600 mm) and
42% in the B2-horizon (600–1200 mm). Bulk densities were 1484 kg m 3 for horizon
A, 1563 kg m 3 for horizon B1 and 1758 kg m 3 for horizon B2, while upper limits for
soil–water holding capacity were 69, 73 and 82 mm, respectively (Snyman, 2000).
2.2. Treatments
Rangeland in 3 different conditions (good, moderate and poor) were studied from
2000/01 to 2001/02 seasons (July–June). The treatments were intended to reflect
states which could arise as a result of different grazing histories in this vegetation
type as been in good, moderate and poor conditions (Snyman, 2005). In this
rangeland, grazing is not random, but involves certain functional groups of plants
being advantaged and disadvantaged in time leading to fundamental changes in
community organization. A distinct composition and basal cover therefore
characterizes each treatment. The good condition rangeland was dominated by the
perennial bunchgrass Themeda triandra and had the highest basal cover, the
moderate condition rangeland was dominated by perennial bunchgrasses of
Eragrostis species, and the poor condition rangeland was dominated by the
stoloniferous perennial Tragus koelerioides and the short-lived perennial bunchgrass
Aristida congesta and had the lowest basal cover.
There were 3 randomly assigned replicates per compositional state (synonymous
with treatment). Each experimental unit was 2 15 m, with a mean slope of 3.5%.
Soil was uniform across plots. The determination and results of basal cover and
botanical composition obtained during the study period is fully discussed by Snyman
(2005). From 1995 the rangeland has been artificially kept in the above-mentioned
rangeland conditions (Snyman, 2005). This was achieved by actively removing, with
minimum disturbance, all undesirable or invader species (with respect to each
specific condition class) after determining the basal cover and botanical composition.
The rangeland was never grazed over the trial period.
2.3. Data collection
An estimate of soil compaction or soil penetration resistance was obtained from 30

points measurements per plot with a simple rod penetrometer (ELE pocket
penetrometer) (Friedel, 1987). Compaction readings were randomly placed and
taken to a depth of 6 mm. Soil compaction from an undisturbed bare soil surface
nearby (Snyman, 2000), was also taken. These readings were taken every third
month at about 18 h after at least 25 mm of rain had fallen (Donaldson et al., 1984;
Donaldson, 1986).
Soil temperature was recorded with mercury thermometers once a week in each
plot at 14 h00 at 50, 100 and 200 mm depths for all treatments. Although the
ARTICLE IN PRESS
thermometers were not properly ventilated, they were shielded. Soil temperature was
also recorded in each treatment on top of the soil.
The soil–water content was monitored with the aid of a neutron hydroprobe
(Model CPN 503, Denmark). The calibration of the hydroprobe and placement of
the neutron probe access tubes are fully described by Snyman (2005).
A double-ring infiltrometer, similar to that described by Knight et al., (1983) and
Wood (1987) was used to determine infiltration rate of the soil in the different
rangeland conditions. Three replications randomly distributed over the plots were
conducted over a soil water gradient (upper limit soil–water holding capacity to
lower limit plant-available water—Snyman, 2000). Plants falling within the double-
ring were defoliated, just before the observations, at a height of 30 mm with the least
disturbance of the soil surface itself.
The organic carbon (C) and total nitrogen (N) content (as measures of organic
matter content) were determined only 5 years (2000/01 season) after the creation of
the different rangeland conditions. Soil samples were taken at 50 mm depth intervals
to a depth of 300 mm at 15 randomly selected points in each plot. Composite samples
for each depth interval were prepared by combining and thoroughly mixing the 15
cores of 20 mm diameter. The composite samples were air-dried at room
temperature, sieved through a 2 mm sieve and stored until analysed for total N.
Subsamples were ground to completely pass through a 0.2 mm sieve and stored until
analysed for organic C. Total N was determined by a modified Kjeldahl method
(Bremmer and Mulvaney, 1982). Since the soil in question contains no free
carbonates the Mebius procedure (Nelson and Sommers, 1982) was used to
determine organic C. Organic C and total N were calculated on a gravimetric air-
dried soil basis and converted to tons per hectare using the relevant bulk densities.
These values were used to calculate C:N ratios.
The above- and belowground phytomass and litter productions used in the
estimation of root and litter turnover, are fully presented by Snyman (2005). Despite
the many sampling problems that could be encountered in the determination of root
turnover (Sims and Singh, 1971; Shackleton et al., 1988), the ratio of annual
increment to peak root phytomass (Dahlman and Kucera, 1965) are being used in
this study (Snyman, 2005). The annual increment was taken as the difference
between the maximum and minimum root phytomass production recorded during
any one year. Turnover times calculated for this study were calculated using the
annual increment over the 2000/01 and 2001/02 seasons. The same approach used for
root turnover, was also applied to the estimation of litter turnover.
A plant was defined as being wilted when it showed visible signs of wilting between
11 h00 and 16 h00 due to a loss of turgidity (Snyman, 1993). In this way, the
temporary and not the permanent wilting point was observed.
2.4. Statistical analysis
The experimental layout was a fully randomized design consisting of three

treatments with three replications. Two-way analysis of variance (ANOVA)
(rangeland condition soil layer) at the 95% confidence level were computed for
ARTICLE IN PRESS
organic C, total N, C:N ratio and soil–water content. All other data on soil
compaction and soil temperature were analysed using a one-way analysis of
variance technique (Winer, 1974). Data were collected through time on the same
experimental unit and therefore analysis also account for non-independence
between sampling times. Data for different months and/or years were analysed
separately. The Number Cruncher Statistical System (2000) software package was
mainly used.
3. Results
3.1. Soil compaction
Rangeland degradation significantly (pp0.01) increased soil compaction

over all the months (Fig. 1). Soil compaction remained relatively constant
in rangeland in good and moderate conditions, over the two growing seasons
(Fig. 1). In rangeland in poor condition, soil compaction even showed a slight
increase, though non-significant (p40.05) as the season progressed. Mean soil
compaction of an adjoining undisturbed bare area over the same period was on
average 19.23 cm 2, which is not much (4.9%) higher than that of rangeland in poor
condition (Fig. 1).
Fig. 1. Average soil compaction (kg cm 2) for the different rangeland conditions over the 2000/01 and
2001/02 growing seasons, measured every fourth month (n=90). Least significant difference (LSD) is
calculated at the 1% level.
ARTICLE IN PRESS
Fig. 2. Monthly average soil temperature (1C) taken at approximately 14:00 at 50 mm (A), 100 mm (B)
and 200 mm (C) depths, for the different rangeland conditions over the 2000/01 and 2001/02 growing
seasons (n=3). Least significant difference (LSD) is calculated at the 1% level.
3.2. Soil temperature
In all rangeland condition treatments the highest soil temperatures occurred as

expected at a depth of 50 mm and decreased with depth (Fig. 2). The highest soil
temperature occurred between January and February for all depths regardless of
rangeland condition. For almost the whole growing season, soil temperatures were
significantly (pp0.01) higher over all depths with rangeland degradation (Fig. 2).
There was increasingly less variation in soil temperatures with depth between
different rangeland conditions, while the difference was most marked between the
warmer months, namely December to February. Minimum soil temperatures were
almost similar (p40.05) regardless of depth and rangeland degradation.
In spring (September and October) the temperature measured on top of the soil
was already significantly (pp0.01) higher in rangeland in poor (50 1C) versus good
condition (40 1C). As the season continued and vegetation afforded better protection
ARTICLE IN PRESS
to the soil surface, especially rangeland in good condition, these differences became
much greater with rangeland degradation. The highest temperatures on top of the
soil of 55 1C, 49 1C and 46 1C, respectively for rangeland in poor, moderate and good
conditions, occurred during December.
3.3. Soil–water content and infiltrability
The soil–water content for the different rangeland conditions, determined

fortnightly over a drying out gradient are presented in Fig. 3. The first observation
was conducted about 3 days after the last rainfall of the above-average rainfall
characterizing the first half of the 2001/02 growing season, while the successive
observations took place over a further two month period of almost no rainfall up to
30 March 2002. When comparing the soil–water contents of the different rangeland
conditions (Fig. 3) with that obtained by Snyman (2000) on the same soil form, the
conclusion can be made that the desiccation pattern stretched from upper limit soil
Fig. 3. Soil–water content (mm), measured every second week, over different depths for rangeland in good
(A), moderate (B) and poor (C) condition, over the period 2 February 2002 to 30 March 2002 (n=6).
ARTICLE IN PRESS
water holding capacity for this soil form to lower limit plant-available water for this
rangeland ecotope.
Fig. 3 clearly shows in all soil layers that rangeland degradation was accompanied by
a notable decrease (po0.01) in soil–water content. Over this period of desiccation, for
example, the water-content over the first 900 mm soil layer was 49% higher in
rangeland in good than in poor condition. Most variation in soil–water content
occurred over the first 0–300 mm soil layer in all rangeland conditions. After the ninth
week (30 March) of the desiccation pattern, all the grass in rangeland in poor condition
shown visible signs of temporary wilted, while only a few grasses wilted in rangeland in
good condition. This point of desiccation can therefore by viewed as the lower limit of
plant-available water for the layers 0–300, 300–600 and 600–900 mm for this rangeland
ecotope of respectively 18, 43 and 30 mm.
The total soil layers in rangeland in good condition was almost filled with water
before the soil water monitoring (Fig. 3) due to the above-average rainfall
characterizing the pre-season and high infiltration rate. This point, viz. week 1 (2
February) in rangeland in good condition can therefore be seen as nearly the upper
limit (drained upper limit) plant-available water for the different layers 0–300,
300–600 and 600–900 mm for this soil form of respectively 65, 72 and 62 mm.The
lower soil–water content in rangeland in poor condition, already at the inception of
the drying out period (Fig. 3) can also be attributed to higher runoff losses in the pre-
season due to lower plant cover (Snyman, 2005). On average over the 2000/01 and
2001/02 growing seasons 1.6%, 4.8% and 10% of the annual rainfall run off from
rangeland in good, moderate and poor conditions respectively.
The infiltrability of the soil determined for each rangeland condition during weeks
1, 3 and 5 (2 February, 16 February and 2 March), as indicated in Fig. 3, is
graphically presented in Fig. 4. Despite the soil–water content (Figs. 3a–c), the initial
infiltration rates were the highest in rangeland in good condition and did decrease
with rangeland degradation. As expected, the initial infiltration rates were higher as
the soil dried out in all rangeland conditions regardless of soil compaction.
Rangeland in good condition stored almost twice as much water over the drying
out period (2 February, 2002–30 March, 2002) in the soil profile than rangeland in
poor condition (Fig. 3). The more water filling of soil pores can be seen as an
important factor for the converse finding of lower infiltrability in rangeland in good
condition than that of the other rangeland conditions after about 5 min of
monitoring, when soil–water content was close to field water capacity (Fig. 4a).
Fig. 4c clearly shows that rangeland degradation throughout the full 50 min of
infiltration monitoring, had a significantly lower infiltrability in the drier soil profile.
Due to the slight variation in soil–water content, between high and low in rangeland
in poor condition (Fig. 3) the infiltration rate is relatively stable regardless of the
soil–water content (Fig. 4).
3.4. Root and litter turnover
Root and litter turnover rates, as well as calculated times for decomposition, are
presented in Table 1 for the different rangeland conditions. The rate of turnover for
ARTICLE IN PRESS
Fig. 4. Infiltration rate (mm h 1) measured every fourth week, for the different rangeland conditions,
measured with soil at approximately field water capacity (Week 1=A), between field water capacity and
permanent wilting point (Week 5=B) and near permanent wilting point (Week 9=C), over the period 2
February 2002–30 March 2002 (n=9).
Table 1
Calculate root and litter turnover rates (year 1) and time for decomposition (months) for the different
rangeland conditions
Rangeland condition Turnover rate (year 1) Replacement roots Decomposition (aboveground)
Roots Litter Months Months
Good 0.59 0.52 20.34 23.08

Moderate 0.56 0.44 21.43 27.27
Poor 0.40 0.39 30.00 30.77
both roots and litter are lower with rangeland degradation. Rangeland degradation
lengthened the replacement of the total root system with about a year while the
decomposition period of the litter will be eight months longer (Table 1). If ratios are
ARTICLE IN PRESS
calculated from average root phytomass or litter production (Sims and Singh, 1971)
instead of peak values (Dahlman and Kucera, 1965), the average turnover rates in all
treatments are only 5.2–7.0 months shorter.
3.5. Organic matter content
The soil organic matter content (organic C and total N) following rangeland
degradation after 5 years, is graphically presented in Fig. 5. Both organic C
and total N of the soil, only 5 years following degradation, decreased
significantly (pp 0.01) with depth over the first 75 mm. A significant interaction
(pp0.01) was obtained between organic matter of different rangeland conditions
and soil depth. Deeper than 75 mm both organic C and total N were influenced
non-significantly (p40.05) by depth in all rangeland conditions (data not
shown). Fig. 5 clearly shows that organic C over the first 50 mm soil layer was
significantly (pp0.01) lower in rangeland in poor than in good condition,
while rangeland degradation had no significant (p40.05) influence on it deeper
than 50 mm.
Rangeland degradation significantly (pp0.01) lowered total N of the soil up to a
depth of 100 mm.The most differences in organic matter content in the soil were
measured in the 0–50 mm layer where rangeland in poor condition contained 25%
and 16% less organic C and total N, respectively. These differences declined with
depth.
Fig 5. Soil organic matter content (kg ha 1) (A-organic C and B=total N), following rangeland
degradation after 5 years, over the 0 to 100 mm soil depth (n=45). Least significant difference (LSD) is for
the 1% level.
ARTICLE IN PRESS
The ratio of C:N varied notably between the different rangeland conditions and
over the depths (data not shown). Only over the first 0–25 mm depth, the C:N ratio
in rangeland in good condition was higher than that in the other two conditions. The
ratio of rangeland in moderate condition was the highest over most of the soil
depths. As the C:N ratio did not differ much (p40.05) from each other for all the
rangeland conditions and varied between 8.15 and 10.15, it can be concluded that
mineralization mainly took place in the soil.
4. Discussion
4.1. Soil compaction
Most researchers support the findings of this study that rangeland degradation
usually leads to increased soil compaction due to a decrease in plant cover
(Donaldson et al., 1984; Du Toit, 1986; Warren et al., 1986a; Friedel, 1987;
Chanasyk and Naeth, 1995; Mworia et al., 1997; Broersma et al., 1999; Snyman,
2005), reduced soil aggregate stability (Warren et al., 1986b) and reduced soil fertility
(Dormaar and Willms, 1998). The loss of vegetative and litter cover with degradation
(Warren et al., 1986a,b; Thurow et al., 1988; Hoffman and Ashwell, 2001; Holm,
2000; Snyman, 2005) allows direct impact of raindrops on soils (Lal and Elliot, 1994;
Russel et al., 2001), and may also produce hydrophobic substances that can reduce
infiltration (DeBano et al., 1970, 1976; Emmerich and Cox, 1992; Snyman, 1999a,
pp. 355–386, 1999b). Thurow et al. (1986) also argued that the function of above-
ground biomass is to protect the surface soil from the disaggregation effect of direct
raindrop impact. Therefore the poor above-ground production with rangeland
degradation (Snyman, 2005) directly contributed to an increase in soil compaction.
The lesser soil compaction at the start of the growing season in all rangeland
conditions can possibly be ascribed to a slight lifting of it by the frost occurring
during the winter months (Du Toit, 1986). With more exposure of the soil to the
elements of nature, it increased until December, after which better protection of the
soil resulting from an increase in aboveground phytomass (Snyman, 2005) in the
post-season, slightly decreased it in rangeland in moderate condition. The mean soil
compaction of 6.37 kg cm 2 obtained in rangeland in good condition in this study,
compared well with the 8.75 kg cm 2 obtained on the same soil form also on
rangeland in good condition (Snyman, 2001).
4.2. Soil temperature
The higher soil temperature with rangeland degradation in this study (Fig. 2) is
agreement with previous work (Du Preez and Snyman, 1993; Snyman, 1999a), which
can largely be attributed to lower plant and litter cover with rangeland degradation
(Snyman, 2004,2005). These higher soil temperatures recorded from rangeland in
poor condition over all depths, could potentially restrict root growth (Snyman,
2005). The soil temperatures differed less between the rangeland conditions with
ARTICLE IN PRESS
depth. To the contrary, in the North American prairies was found that root growth
is associated with increases in soil temperature (in areas with winter snow) and water
in spring and summer (Dahlman and Kucera, 1965; Bartos and Sims, 1974).
Unfortunately little is known about the way in which root systems integrate the
effects of wide ranges of temperature between different zones (Drew, 1979; Distel
and Fernandez, 1988). However, Drew (1979) pointed out that even slight deviations
of soil temperature from the optimum for root growth in perennial plants can have a
marked effect on the growth process. Although temperatures as high as 45 1C could
decrease root extension rate in pioneer grasses, there is considerable variation
between species and genera (Bowen, 1991, pp. 309–330) and it is likely that all of
these species have roots adapted to high soil temperatures.
4.3. Soil–water content and infiltrability
This study clearly showed that rangeland degradation decreased the soil–water
content in all soil layers. The main reasons could be the higher runoff (Wright et al.
1976, 1982; Snyman and Opperman, 1984; O’Connor, et al., 2001; Snyman, 2005)
and evaporation (Van de Vijver, 1999) due to lower plant cover accompanying
rangeland degradation (Snyman, 2004). The runoff in this study of 1.6%, 4.8% and
10% of the annual rainfall, agrees with the 3.5%, 5.6% and 8.7% from rangeland in
good, moderate and poor condition respectively, obtained on the same soil over a 19
year period (Snyman, 1998, 1999a). Infiltration capacity or runoff is linearly related
to basal cover of perennial grasses (Van den Berg et al., 1975; Slack and Larson,
1981), which can also be affected by litter cover (Gifford and Hawkins, 1978;
Stroosnijder, 1996; Van de Vijver, 1999; Snyman, 2005) and aboveground biomass
(Kelly and Walker, 1974; Scholes and Walker, 1993; Snyman, 2002).
From 2 February 2002 to 30 March 2002, the water withdrawal (during which
hardly any rain occurred) from rangeland in good condition was, for example,
24 mm over the first 300 mm depth vs. the water loss of 30 mm for deteriorated
rangeland. As rangeland in good condition had significantly (pp0.01) more roots
than rangeland in poor condition (Snyman, 2004), the above withdrawal pattern
should have been the other way round. The lower plant cover and litter production
of rangeland in poor condition (Snyman, 2001) explain the anomaly by more water
loss occurring due to evaporation from the soil surface. The contribution of
evaporation may be much more with rangeland degradation than previously believed
(Richardson and Hole, 1978; Emmerich, 1999; Ekaya et al., 2001). This is supported
by Snyman (1998) who states that the largest percentage of soil drying immediately
after wetting, in semi-arid areas, can be ascribed to evaporation (Es) from the soil
surface. The magnitude of direct Es from the soil surface in arid and semi-arid
rangelands might range from 20% to 70% of the infiltrated rain (Le Houèrou, 1984).
Also interesting from this study is that in rangeland in good condition, soil–water
content decreased more in soil layers deeper than the 300 mm, than in other
rangeland conditions. The curtailing of growth due to water limitations depends
therefore not only on climatic conditions and soil properties but also on species
differences e.g. rooting depth (Snyman, 2005). The water withdrawal over the
ARTICLE IN PRESS
300–600 mm and 600–900 layers were 17 and 20 mm, respectively, for rangeland in
good condition versus that of rangeland in poor condition of only 6 and 2 mm.This
tendency can largely be ascribed to the greater root mass of rangeland in good
condition over these depths (Snyman, 2005). Root production in response to
increased soil water has also been recorded for African (Shackleton et al., 1988;
Rethman et al., 1997; McNaughton et al., 1998), Australian grasslands (Mott et al.,
1992; Ingram, 2002) and semi-arid Argentina (Distel and Fernandez, 1988).
Increased soil water and subsequent increase in root proliferation results in
increased rate of mass flow and diffusion of nutrients into roots (Drew, 1979).
In all rangeland conditions, the soil–water content to a depth of 900 mm was only
slightly supplemented over the growing season. Due to the variable and low rainfall
of semi-arid regions the general tendency is that the perennial herbaceous layer
withdraws water from throughout the whole soil profile during the growing season
and can store much water for a following growing season or period (Snyman, 2000).
Grasses in these areas generally grow optimally for relatively short periods before
temporary wilting takes place (Snyman, 1993, 1998; Ingram, 2002). This study
showed that this point was reached faster with rangeland degradation. It appears
that in the arid and semi-arid areas, deep percolation (exceeding rooting zone) only
occurs under extremely high rainfall conditions (Fischer and Turner, 1978; Scholes
and Walker, 1993; Bennie et al., 1994; Snyman and Oosthuizen, 1999). The quantity
of water, which can be stored by the soil profile, varies mainly with the silt plus clay
content and depth (Bennie, 1991), while the type of clay minerals and organic matter
content (Reicosky et al., 1995; Okatan and Reis, 1999; Van de Vijver, 1999) can also
make a contribution.
One way of regarding the ‘‘strategy’’ of root distribution in semi-arid areas,
recognizing that water is the major factor controlling root growth, is to consider the
length of time water remains available at different depths, the types of root systems
and life cycles best suited to intercept this efficiently (Noy-Meir, 1973; Drew, 1979).
During this study the roots of rangeland in good condition, based on water
withdrawal pattern and root mass (Snyman, 2005) were largely concentrated over
the first 600 mm depth, while occurring more shallow with rangeland degradation.
The roots of rangeland in poor condition (pioneer species) are located at a depth of
not more than 300 mm even in deep soil and therefore benefit from the lightest rains
(Snyman, 2005). The large percentage lateral roots in rangeland in good condition
(climax plants) over the first 300 mm (Snyman, 2005) is due to utilization of the more
frequent rains, largely responsible for above-ground production. The thicker and
deeper roots contrarily (Snyman, 2005) is largely responsible for survival during
water stress periods. It can generally be concluded that in arid and semi-arid
environments, many grasses do not have a deep enough root system to access
groundwater and are reliant on surface water after rainfall events (Drew, 1979; Distel
and Fernandez, 1998; Ingram, 2002) leading to a short growing season (Troughton,
1978; Sala et al., 1991; Snyman, 2005).
The increase in soil compaction and decrease in litter (Gifford and Hawkins, 1978;
Knight et al., 1983; Stroosnijder, 1996; Mwendera and Saleem, 1997; Snyman, 2005)
with rangeland degradation are of the most important reasons for the slower
ARTICLE IN PRESS
infiltration rate (Fig. 4) in the soil after a rain shower obtained in this study in
rangeland in poor condition (Allison, 1973; Du Preez and Snyman, 1993; Holm et
al., 2002). According to Simanton et al. (1991) and Emmerich and Cox (1992), there
could be little difference in infiltration rate on rangeland just after removing the
standing biomass by clipping or burning. Therefore the more plant material due to
denser cover in rangeland in good condition (Snyman, 2004, 2005), though cut,
could not have contributed much to hindering of infiltration in rangeland in good
condition in this study. As the soil dried out, the above tendency is reversed by
rangeland degradation, causing a decrease in infiltration rate after a few minutes of
monitoring. The significantly higher root production occurring in all soil layers in
rangeland in good condition (Snyman, 2005), could possibly have created larger pore
spaces in the soil for the more rapid infiltration rate than that of rangeland in poor
condition with its lower root production (DeBano et al., 1970; Richardson and Hole,
1978; Du Preez and Snyman, 1993; Warren et al., 1986a; Snyman, 2005). Another
possibility is that the release of entrapped air due to larger soil pores in rangeland in
good condition (Richardson and Hole, 1978) could cause higher infiltrability (Jarrett
and Fritton, 1978; Graetz and Tongway, 1986). As expected it took longer before the
infiltration rates started stabilizing in al the rangeland conditions as the soil dried out
(Fig. 4). Data on frost action or soil biological activity were not collected in this
study, which also could influence infiltration rates over the season (Achouri and
Gifford, 1984; Rostagno, 1989). The general conclusion can be made that retention
of water within a rangeland depends on infiltration capacity (Slack and Larson,
1981), which depending on soil type and slope, can be affected by degradation
through reduction of plant or litter cover (Gifford and Hawkins, 1978; Abel, 1993,
pp. 173–195; Stroosnijder, 1996; Mwendera and Saleem, 1997).
4.4. Root and litter turnover
It is not surprising that with the poor root development and low litter production
(Snyman, 2005) of rangeland in poor condition that the rate of turnover for both
these components decreased with rangeland degradation. Although there are
controversies about the calculation of root and litter turnover, various researchers
found shorter turnover ratios if the average values (Sims and Singh, 1971), instead of
peak values (Dahlman and Kucera, 1965) are used (Kumar and Joshi, 1972; Singh
and Yadava, 1974; Strugnell and Piggott, 1978; Smith, 1985; Shackleton et al., 1988).
Obviously, certain portions of the root system are more active than other (the fine
root system for instance) and therefore turnover times will not be uniform for the
whole system (Shackleton et al., 1989). In Western Australia the root longevity of
Themeda triandra was measured as 1.9 years, compared to 3.6 years for Astrebla
pectinata and 6.3 years for Eragrostis xerophila rangelands (Ingram, 2002). The roots
of some grass species may be relatively short-living or they may be more perennial
(Wolfson and Tainton, 1999, pp. 54–90). It seems logical to assume that there is a
close relation between the life-expectancy of a tiller and of the roots that develop
from it. As rangeland in poor condition is dominated by pioneer grass species which
are poorly perennial, it can be expected that their root system should also show a
ARTICLE IN PRESS
short life cycle (Chapin, 1980), in contrast with this study the reverse was determined
with rangeland degradation. This tendency is supported by Shackleton et al. (1989)
that grass species, typical of rangelands in poor condition, may reduce nutrient
cycling rates since they decompose slower than more palatable species. Moreover, in
arid environments where diffusion is limited by low soil water, a more extensive root
system aids water and nutrient uptake (Chapin, 1991) and this is partially achieved
through roots with increased longevity (Chapin, 1980).
In these arid environments, production and decomposition of roots and litter are
pulsed in nature and trends are closely related to rainfall occurrence. Soil water is
thus a limiting factor both to the production and decomposition of litter (Wiltshire,
1990; Ekaya and Kinyamario, 2001; Ingram, 2002). Belowground material
consistently decomposed faster than above-ground material (Ekaya and Kinyamar-
io, 2001). Peaks in both above-ground and belowground material decomposition
rates coincided with rainfall peaks. High summer temperatures coupled with heavy
rainfall and accompanying high humidity, provide ideal conditions for decomposi-
tion. Reasons for the faster turnover in good rangeland are therefore associated with
the microclimate, especially the higher soil–water content (Snyman, 2005). In most
arid and semi-arid rangelands litter and root turnovers are very slow (Whitford et
al., 1988). In warm, high-rainfall areas, breakdown of grass litter is rapid (50–53% of
mass in 3 months), but the rate depends on moisture availability and species (West,
1979, pp. 647–659; Mott et al., 1984, pp. 56–82; Shackleton et al., 1989).
The turnover rate for roots of 0.59 obtained in this study for rangeland in good
condition (Table 1), compares well with that obtained in other rangeland areas
worldwide. Turnover times are generally higher on grazed sites (Sims and Singh,
1971; Strugnell and Piggott, 1978; Shackleton et al., 1988). Root turnover results
obtained over Africa, North America, India and Marion Island vary for grazed
rangeland between 0.19 and 0.68 and for ungrazed areas between 0.22 and 0.77
(Dahlman and Kucera, 1965; Kumar and Joshi, 1972; Singh and Yadava, 1974;
Smith, 1985).
4.5. Organic matter content
The drastical decrease in organic C and total N of 25% and 16%, respectively,
after only 5 years of rangeland degradation, obtained over the first 50 mm soil layer
(Fig. 5), is a cause of concern. Making these figures more astonishing is that Du
Preez and Snyman (1993), on the same soil form and the same soil depth, obtained a
loss in organic C of 33% and in total N of 23% over a 15 year period with rangeland
degradation. This data exhibits the importance of the surface few mm of soil as it
relates to C and N dynamics in a rangeland. Other researchers also reported on most
changes in organic matter in rangeland over the first 200 mm soil depth (Manley et
al., 1995; Chevallier et al., 2001; Montani et al., 1996). Smoliak et al. (1972) and
Manley et al. (1995) also found that species composition shifts within a plant
community can result in C changes. If accepted that organic matter content was the
same for the different rangeland conditions with the inception of this study, the loss
in organic C over the first 300 mm soil depth for rangeland in moderate and poor
ARTICLE IN PRESS
conditions were 333 and 1660 kg ha 1 respectively over the 5 year period only. Over
the same period the loss in total N was 85 and 174 kg ha 1. Over a 15 year period of
rangeland degradation, Du Preez and Snyman (1993) on the same soil form, found
that about 4200 kg C ha 1 and 300 kg N ha 1 more were lost from soil in rangeland
in poor condition than rangeland in good condition.
Already after 5 years, a decrease in organic matter content in the soil with
rangeland degradation was expected though not as drastic, as less plant litter was
replaced in the soil due to a decrease in above- (Dormaar and Willms, 1998; Snyman,
1998) and belowground biomass production in rangeland in poor and moderate
conditions (Rosswall, 1976; Detling et al., 1979; Holland and Detling, 1990; Manley
et al., 1995; Snyman, 2005). The average aboveground phytomass productions for
rangeland in good, moderate and poor conditions for this study were, respectively,
2341, 1648 and 695 kg ha 1, versus the belowground phytomass of 3666, 3016 and
1430 kg ha 1 (Snyman, 2005). This is supported by the findings of Troughton (1957)
that root phytomass of grasses is a more important contributor towards soil organic
matter than above-ground phytomass. This must be viewed against the background
that soil of climax rangeland soils are low in available nitrogen with most nitrogen
found in organic matter (Roux, 1969). As the major part of root mass is present in
the surface 100 mm of soil in this rangeland system (Tainton, 1981, pp. 27–56;
Dormaar et al., 1984; Moore, 1989; Snyman, 2005), nutrient availability in this zone
is very important. Although surface runoff was higher on the poor rangeland
condition (Snyman, 2005) erosion accounted for only a small proportion (3.5%) of
lost nutrients (5.1% of nitrogen) which is supported by Du Preez and Snyman
(1993). The potential losses of organic C and total N due to erosion over the study
period were calculated using the measured values of organic C and total N in the
0–50 mm soil layer of the good rangeland. Although grass roots are the primary
source of organic matter in rangeland soil (Neary et al., 1999; Balesident and
Balabane, 1996) and can also stimulate soil organic matter mineralisation (Ladd et
al., 1994), aboveground litter provides a secondary source (Aandahl, 1981; Manley et
al., 1995). The lower litter production accompanying rangeland in poor condition
(Snyman, 2005) definitely contributed towards lower organic matter with rangeland
degradation.
Deeper than 100 mm the differences in total N between different rangeland
conditions were non-significant (p40.05). Especially in rangeland in moderate and
poor conditions, only a small percentage of the root system is present over these
deeper soil depths (Snyman, 2005). Therefore it is not surprising that rangeland
degradation over the 5 years had no effect on the lower portion of the soil profile.
The rate of soil organic matter decomposition can increase to a maximum at about
37 1C and then declines (Jenkinson, 1981). Therefore, decomposition of soil organic
matter under moderate and poor rangeland conditions may be faster than that under
good conditions as a result of higher temperatures (Du Preez and Snyman, 1993),
which is contradictory with the estimations of this study. According to Wiltshire
(1990), decomposition of soil organic matter in these semi-arid regions is restricted
both spatially and temporally to the layer moistened by rainfall. The higher
soil–water content in all the soil layers in rangeland in good condition, increased
ARTICLE IN PRESS
decomposition in this rangeland condition. Where rainfall might be largely the

limiting environmental factor to production in rangeland in poor condition, it seems
that under higher soil–water conditions, N may be the case with an improvement in
rangeland condition. In contrast, Neary et al. (1999) argued that N is not the most
important driving factor for decomposition in a semi-arid ecosystem, but rather the
water availability and C quality of the litter play a much greater role in controlling
decomposition. Year-to-year fluctuations of environmental conditions in arid areas
may affect differences in biomass production which may again influence organic
matter content stronger in especially degraded rangeland. Lowered productivity is
exacerbated by the poor water-use efficiency at all rainfall levels of poor condition
rangeland (Snyman and Fouché, 1993; Williams et al., 1998; Snyman, 1988, 1998,
1999c, 2005; Le Houérou, 1984). Rangeland and savanne are the most fire-adapted
ecosystems, with 83–85% of their C belowground (Neary et al., 1999).
5. Conclusions
A prerequisite for sustainable land use is that soil quality is maintained or

improved. Therefore the drastic decrease in soil quality with rangeland degradation
is a cause of concern, because once it is lost recuperation is a slow process or could be
irreversible. In the past there was largely concentrated on the above- and
belowground productivity decreased due to rangeland degradation. In this study
the effect of degradation is also shown on various soil characteristics like soil
temperature which was increased, soil–water content and infiltrability which were
decreased and further lead to a lower organic matter content of the soil, eventually
causing increased drought risks. Year-to-year fluctuations of environmental
conditions in arid areas affect differences in biomass production which again more
intensively influence the organic matter content specifically with rangeland
degradation. Where rainfall might be the largest determining environmental factor
influencing production in rangeland in poor condition, it is N under higher
soil–water conditions with rangeland condition improvement. As a decrease in plant
cover usually accompanies rangeland degradation in arid and semi-arid areas,
together with lower plant production and litter, it forms a slower supplementation of
plant material towards the improvement and stabilizing of the ecosystem. Since roots
are the major C source in soil, the poor root mass with rangeland degradation
contributed also to slow building up of organic matter in the soil. It can rightly be
concluded that semi-arid rangeland, which have retrogressed beyond a threshold of
drought resilience, will only be restored by mechanical inputs. The answer is clearly
one of guarding against degradation in the first instance, rather than in relying on the
regeneration of rangeland.
In the past the lower basal cover accompanying rangeland degradation as
determinative and most important factor in the poorer soil–water balance in arid and
semi-arid ecosystems, was mostly emphasized (Snyman, 1999a, 2000). This study
clearly showed that the contribution of increased soil compaction, soil temperature
and poorer infiltrability, as well as the lower organic matter content of the soils of
ARTICLE IN PRESS
degraded rangeland on an impoverished soil–water balance must not be under

emphasized. Especially lower root mass and poor root establishment with rangeland
degradation do not lead to sufficient pore spaces for a better infiltration capacity in
the soil. Rangeland in poor condition therefore utilizes water much more inefficiently
than the case with rangeland in good condition. With water being the prime
determinant of plant growth in semi-arid rangelands, reduced vegetation production
with degradation, can also be explained by the reduction of soil–water content as
result of vegetation litter loss, which increases loss of water through runoff and
evaporation.
The fact that the quality of natural resources must be maintained to ensure
sustainable land-use, brought about that soil quality is contemplated intensively
these days (Miller and Wali, 1995). The results obtained in this study can serve as
guidelines or norms for soil and rangeland degradation in the semi-arid central
rangeland of South Africa where rainfall and nitrogen are limiting environmental
factors. Indicators for identifying deterioration of natural resources are an aspect
justifying future attention. These indicators must include the continual monitoring
of different rangeland types, where its implementation must definitely take place
with the determination of threshold values to be at all successful within a production
system.
References
Aandahl, A.R., 1981. Soils of the Great Plains: Land use, Crops and Grasses. University of Nebraska
Press, Lincoln, NE.
Abel, N.O.J., 1993. Reducing cattle numbers in Southern African Communal Range: Is it worth it? In:
Behnke, R.H., Scoones I., Kerven,C. (Eds.), Range Ecology at Disequilibrium. Overseas Development
Institute, London, England.
Achouri, M., Gifford, G.F., 1984. Spatial and seasonal variability of field measured infiltration rates on a
rangeland site in Utah. Journal of Range Management 37, 451–455.
Allison, F.E., 1973. Soil Organic Matter and its Role in Crop Production. Elsevier, Amsterdam, 150pp.
Allsopp, N., 1999. Consequences to soil patterns and processes in Southern Africa following vegetation
change associated with livestock management. Proceedings of the VI International Rangeland
Congress, Townsville, Australia, pp. 96–101.
Balesident, J., Balabane, M., 1996. Major contribution of roots to soil carbon storage inferred from maize
cultivation soils. Soils Biological Biochemic 28, 1261–1263.
Bartos, D.L., Sims, P.L., 1974. Root dynamics of a shortgrass ecosystem. Journal of Range Management
27, 33–36.
Bennie, A.T.P., 1991. Sustainable agriculture in dry areas of South Africa. Inaugural address, University
of the Free State, Bloemfontein, South Africa, 26pp.
Bennie, A.T.P., Hoffman, J.E., Coetzee, M.J., Very, H.S., 1994. Opgaring en benutting van reënwater in
grond vir die stabilisering van plantproduksie in halfdroë gebiede. Water Research Commission
Report 227/1/94, Pretoria, South Africa, 159pp.
Bowen, G.D., 1991. Soil temperature, root growth and plant function. In: Weisel, Y., Eshel, A., Kafkafi,
W. (Eds.), Plant Roots: The Hidden Half. Marcel Dekker Institute, New York 410pp.
Bredenkamp, G., Van Rooyen, N., 1996. Dry Sandy Highveld Grassland. In: Low, A.B., Rebelo, A.G.
(Eds.), Vegetation of South Africa, Lesotho and Swaziland. Dept Environmental Affairs and Tourism,
Pretoria, South Africa, 85pp.
Bremmer, J.M., Mulvaney, C.S., 1982. Total nitrogen. In: Page, A.L. (Ed.), Methods of Soil Analysis. Part
2. Chemical and Microbiological Properties. American Society of Agronomy, Madison, WI, 198pp.
ARTICLE IN PRESS
Broersma, K., Krysic, M., Thompson, D.J., Bornke, A.A., 1999. Effect of long-term grazing on soil
quality in southern British Columbia. Proceedings of the VI International Rangeland Congress, vol. 1,
Townsville, Australia, pp. 114–115.
Chanasyk, D.S., Naeth, M.A., 1995. Grazing impacts of bulk density and soil strength in the foothill
fescue grasslands of Alberta Canada. Canadian Journal of Soil Science 75, 551–557.
Chapin, F.S., 1980. The mineral nutrition of wild plants. Annual Review of Ecological Systems 11,
233–260.
Chapin, F.S., 1991. Effects of multiple environmental stresses on nutrient availability and use. In:
Mooney, H.A., Winner, W.E., Pell, E.J. (Eds.), Response of Plants to Multiple Stresses. Academic
Press, Inc., San Diego, CA.
Charley, J.L., Cowling, S.W., 1968. Changes in soil nutrient status resulting from overgrazing and their
consequences in plant communities of semi-arid areas. Proceedings of the Ecological Society of
Australia 3, 28–38.
Chevallier, T., Blanchart, E., Girardin, C., Mariotti, A., Albrecht, A., Eeller, C., 2001. The role of
biological activity (roots, earthworms) in medium-term C dynamics in vertisol under a Digitaria
decumbens (Granimae) pasture. Applied Soil Ecology 16, 11–21.
Dahlman, R.C., Kucera, C.L., 1965. Root productivity and turnover in native prairie. Ecology 46,
84–89.
DeBano, L.F., Mann, L.D., Hamilton, D.A., 1970. Translocation of hydrophobic substances in soil by
burning. Soil Science American Proceedings 34, 130–133.
DeBano, L.F., Savage, S.U., Hamilton, D.A., 1976. The transfer of leaf and hydrophobic substances
during burning. Soil Science American Journal 40, 779–787.
Detling, J.K., Dyer, M.I., Winn, D.T., 1979. Nett photosynthesis, root respiration and regrowth of
Bouteloua gracilis following simulated grazing. Oecologia 41, 127–134.
Distel, R.A., Fernandez, O.A., 1988. Dynamic of root growth and decay in two grasses native to semi-arid
Argentina. Australian Journal of Ecology 13, 327–336.
Donaldson, C.J., 1986. Simple technique for estimating soil compaction. Karoo Agric 3, 55–58.
Donaldson, C.H., Rootman, G., Grossman, D., 1984. Long-term nitrogen and phosphorus application on
veld. Journal of the Grassland Society of South Africa 1, 27–32.
Dormaar, J.F., Willms, W.D., 1998. Effect of forty-four years of grazing onf fescue grassland soils.
Journal of Range Management 51, 122.
Dormaar, J.F., Johnston, A., Smaliak, S., 1984. Seasonal changes in carbon content and dehydrogenase,
phosphatase and urease activities in mixed prairies and fescue grassland Ah horizons. Journal of Range
Management 37, 31–35.
Drew, M.C., 1979. Root development and activities. In: Goodall, R.A., Perry, R.A. (Eds.), Arid-land
Ecosystems. Cambridge University Press, London, 870pp.
Du Preez, C.C., Snyman, H.A., 1993. Organic matter content of a soil in a semi-arid climate with three
long-standing veld conditions. African Journal of Range and Forage Science 19, 108–110.
Du Preez, C.C., Snyman, H.A., 2003. Soil organic matter changes following rangeland degradation in a
semi-arid South Africa. Proceedings of the Vll International Rangeland Congress, Durban, South
Africa, pp. 476–478.
Du Toit, G.V.N., 1986. Die fisiese uitwerking van beweiding deur skape op Karooveld. M.Sc. Agriculture
Thesis, University of the Free State, Bloemfontein, South Africa, 132pp.
Ekaya, W.N., Kinyamario, J.J., 2001. Production and decomposition of plant tiller in an arid rangeland of
Kenya. African Journal of Range and Forage Science 18, 125–129.
Ekaya, W.N., Kinyamario, J.I., Kurue, C.N., 2001. Abiotic and herbaceous vegetational characteristics of
a arid rangeland in Kenya. African Journal of Range and Forage Science 18, 117–124.
Emmerich, W.E., 1999. Nutrient dynamics of rangeland burns in South-eastern Arizona. Journal of Range
Management 52, 606–614.
Emmerich, W.E., Cox, J.R., 1992. Hydrologic characteristics immediately after seasonal burning on
introduced and native grasslands. Journal of Range Management 45, 476–479.
Emmerich, W.E., Heitschmidt, R.K., 2002. Drought and grazing: II Effects on runoff and water quality.
Journal of Range Management 55, 229–234.
ARTICLE IN PRESS
Fischer, R.A., Turner, N.C., 1978. Plant productivity in the arid and semi-arid zone. Annual Review Plant
Physiology 29, 277–317.
Friedel, M.H., 1987. A preliminary investigation of woody plant increase in the western Transvaal and
implications for veld assessment. Journal of the Grassland Society Southern Africa 4, 25–30.
Gifford, G.F., Hawkins, R.H., 1978. Hydrologic impact of grazing on infiltration: a critical review. Water
Resource Research 14, 305–313.
Graetz, R.D., Tongway, D.J., 1986. Influence of grazing management on vegetation, soil structure and
nutrient distribution and the infiltration of applied rainfall in a semi-arid chenopod shrubland.
Australian Journal of Ecology 34, 123–134.
Gregorich, E.G., Carter, M.R., Angers, D.A., Monreal, C.M., Ellet, B.H., 1994. Towards a minimum data
set to assess soil organic matter quality in agricultural soil. Canadian Journal of Soil Science 74,
367–385.
Hoffman, M.T., Ashwell, A., 2001. Nature divided land degradation in South Africa. University of Cape
Town Press, Cape Town, South Africa, 168pp.
Holland, E.A., Detling, J.K., 1990. Plant response to herbivory and belowground nitrogen cycling.
Ecology 71, 1040–1049.
Holm, A.M., 2000. A study of degradation processes at patch to landscape scale within the arid shrubland
of Western Australia. Ph.D. Thesis, University of Western Australia, Perth, 192pp.
Holm, A.M., Bennet, L.T., Loneragan, W.A., Adams, M.A., 2002. Relationships between empirical and
nominal indices of landscape function in the arid shrubland of Western Australia. Journal of Arid
Environments 50, 1–21.
Ingram, L.J., 2002. Growth, nutrient cycling and grazing of three perennial tussock grasses of the
Pilbara region of NW Australia. Ph.D. Thesis, Department of Botany, University of Western
Australia, 279pp.
Jarrett, A.R., Fritton, D.D., 1978. Effects of entrapped soil air on infiltration. Transaction of the ASAE
21, 901–906.
Jenkinson, D.S., 1981. The fate of plant and animal residues in soil. In: Greenland, D.J., Hayes, M.H.B.
(Eds.), The Chemistry of Soil Processes. Wiley, New York 126pp.
Kelly, R.D., Walker, B.H., 1974. The effects of different forms of land use on the ecology of a semi-arid
region in south-eastern Rhodesia. Journal of Ecology 62, 553–574.
Knight, R.W., Blackburn, W.H., Scifres, C.J., 1983. Infiltration rates and sediment production following
herbicide/fire brush treatments. Journal of Range Management 36, 153–157.
Kumar, A., Joshi, M.C., 1972. The effects of grazing on the structure and productivity of the vegetation
near Pilani, Rajasthan, India. Journal of Ecology 60, 665–674.
Ladd, J.N., Amato, M., Zhou, L.K., Schultz, J.E., 1994. Differential effects of rotation, plant residue and
nitrogen fertilizer on microbal biomass and organic matter in an Australian alfisol. Soil Biological
Biochem 26, 821–831.
Lal, R., Elliot, W., 1994. Erodibility and erosivity. In: Lal, R. (Ed.), Soil Erosion Research Methods,
second ed. Soil and Water Conservation Society, vol. 1a, Ankeny, pp. 181–208.
Le Houérou, H.N., 1984. Rain use efficiency: A unifying concept in arid-land ecology. Journal of Arid
Ludwig, J., Tongway, D., 1998. A landscape approach to rangeland ecology. In: Ludwig, L., Tongway,
D., Freudenberger, D., Noble, J., Hodgkinson, K. (Eds.), Landscape and Ecology. Function and
Management. CSIRO Publishing, Melbourne, 210pp.
Manley, J.T., Schuman, G.E., Reeder, J.D., Hart, R.H., 1995. Rangeland soil carbon and nitrogen
responses to grazing. Journal of Soil and Water Conservation 50, 294–298.
Mazzarino, M.J., Bertiller, M.B., 1999. Soil N pools and processes as indicators of desertification in semi-
arid woodlands and semi-arid steppes of Argentina, Proceedings of the VV International Rangeland
Congress. Townsville, Australia, 101–105pp.
McNaughton, S.J., Banjikiva, F.F., McNaughton, M.M., 1998. Root biomass and productivity in a
grazing ecosystem: the Serengeti. Ecology 79, 587–592.
Milchunas, D.G., Lauenroth, W.K., 1993. Quantitive effects of grazing on vegetation and soils over a
global range of environments. Ecological Monographs 63, 327–366.
ARTICLE IN PRESS
Miller, F.P., Wali, M.K., 1995. Soil, land use and sustainable agriculture: a review. Canadian Journal of
Soil Science 75, 413–422.
Montana, C., Ezcurra, B., Carillo, A., Delhoume, J.P., 1988. The decomposition of litter in grasslands of
northern Mexico: a comparison between arid and non-arid environments. Journal of Arid
Montani, T., Busso, C.A., Fernandez, O.A., Brevedan, R.E., 1996. Production and nitrogen cycling in an
ecosystem of Eragrostis curvula in semi-arid Argentina. 1. Plant biomass and productivity. Acta
Oecologia 17, 151–162.
Moore, A.C., 1989. Sekere fenologiese en fisiologiese reaksies van Themeda triandra op verskillende
ontblaringskedules. M.Sc. Thesis, University for CHE, Potchefstroom, South Africa, 77pp.
Mott, J.J., Ludlow, M.M., Richards, J.H., Parsons, A.D., 1992. Effects of moisture supply in the dry
season and subsequent defoliations on persistence of the savanna grasses, Themeda triandra,
Heteropogon contortus and Panicum maximum. Australian Journal of Agricultural Research 43,
241–260.
Mott, J.J., Williams, M.H., Andrew, M.H., Gillison, A.N., 1984. Australian Savanna Ecosystems. In:
Tothill, J.C., Mott, J.J. (Eds.), International Savanna Symposium. Commonwealth Agricultural
Bureaux, Brisbane, 314pp.
Mworia, J.K., Mnene, W.M., Musimbi, D.K., Reid, R.S., 1997. Resilience of soil and vegetation subjected
to different grazing intensities in a semi-arid rangeland of Kenya. African Journal of Range and
Forage Science 14, 26–31.
Mwendera, E.J., Saleem, M.A.M., 1997. Infiltration rates, surface runoff, and soil loss as influenced by
grazing pressure in the Ethiopian highlands. Soil Use & Management 13, 29–35.
Neary, D.G., Klopatek, C.C., DeBano, L.F., Efolliot, P.F., 1999. Fire effects on belowground
sustainability: a review and synthesis. Forest Ecology and Management 122, 51–71.
Nelson, D.W., Sommers, L.E., 1982. Total carbon, organic carbon and organic matter. In: Page, A.L.
(Ed.), Methods of Soil Analysis Part 2. Chemical and Microbiological Properties. American Society of
Agronomy, Madison WI, 168pp.
Noy-Meir, I., 1973. Desert ecosystems: environment and producers. Annual Review of Ecological Systems
4, 25–51.
O’Connor, T.G., Haines, L.M., Snyman, H.A., 2001. Influence of precipitation and species composition
on phytomass of a semi-arid African grassland. Journal of Ecology 89, 850–860.
Oesterheld, M., Loreti, J., Semmartin, M., Sala, O.E., 2001. Inter-annual variation in primary production
of a semi-arid grassland related to previous-year production. Journal of Vegetation Science 12,
137–142.
Okatan, A., Reis, M., 1999. Relationships between soil hydrophysical properties and range condition in
the forest range in Trabzon Meryemana Creek water shed Turkey. Proceedings of the VIth
International Rangeland Congress vol. 1, Townsville, Australia. Vol. 1. pp. 326–327.
Reicosky, D.C., Kemper, W.D., Langdale, G.W., Douglas, C.L., Rasmussen, P.E., 1995. Soil organic
matter changes resulting from tillage and biomass production. Journal of Soil and Water Conservation
35, 253–261.
Rethman, N.F.G., Venter, P.S., Lindeque, J.P., 1997. Influence of soil water availability on the above and
belowground phytomass of five sub-tropical grass species. Applied Plant Sciences 11, 29–30.
Richardson, J.L., Hole, F.D., 1978. Influence of vegetation on water repellency in selected western
Wiscousin soils. Soil Science American Journal 42, 465–467.
Rosswall, T., 1976. The internal nitrogen cycle between micro-organisms, vegetation and soil. Ecological
Bulletin 22, 157–167.
Rostagno, C.M., 1989. Infiltration and sediment production as affected by soil surface condition in a
shrubland of Partagonia, Argentina. Journal of Range Management 42, 382–385.
Roux, E.R., 1969. Grass: The Story of Frankenwald. Oxford University Press, Cape Town, South Africa.
Russell, J.R., Betteridge, K., Costall., D.A., Mackay, A.D., 2001. Cattle treading effects on sediment loss
and water infiltration. Journal of Range Management 54, 184–190.
Sala, O.E., Lauenroth, W.K., Parton, W.J., 1991. Long-term soil water dynamics in the shortgrass steppe.
Ecology 73, 1175–1181.
ARTICLE IN PRESS
Schenk, H.J., Jackson, R.B., 2002. Rooting depths lateral root spreads and below-ground/aboveground
allometries of plants in water-limited ecosystems. Journal of Ecology 90, 480–494.
Scholes, R.J., Walker, B.H., 1993. An African Savanna: Synthesis of the Nylsvlei Study. Cambridge
University Press, Cambridge 168pp.
Schulze, E.R., 1979. Climate of South Africa. Part 8. General Survey. Weather Bureau, Pretoria,
158pp.
Schuman, G.E., Reeder, J.D., Morgan, J.A., Lecain, D.R., Hart, R.H., 1999. Influence of grazing on the
carbon and nitrogen balance of a mixed-grass prairie. Proceedings of the VI International Rangeland
Congress, Townsville, Australia, pp. 230–232.
Shackleton, C.M., McKenzie, B., Granger, J.E., 1988. Seasonal changes in root biomass, root/shoot ratios
and turnover in two coastal grassland communities in Transkei. South African Journal of Botany 54,
465–471.
Shackleton, C.M., McKenzie, B., Granger, J.E., 1989. Breakdown and decomposition in three coastal
grassland communities in Transkei. South African Journal of Botany 55, 551–559.
Simanton, J.R., Weitz, M.A., Larson, H.D., 1991. Rangeland experiments to parameterize the water
erosion prediction project model: vegetation canopy cover effects. Journal of Range Management 44,
276–282.
Sims, P.L., Singh, J.S., 1971. Herbage dynamics and net primary production in certain ungrazed and
grazed grasslands in North America. In: French, N.R. (Ed.), Preliminary Analysis of Structure and
Function of Grasslands, Department of Science, Colorado State University, 314 pp.
Singh, J.S., Coleman, D.C., 1973. A technique for evaluating functional root biomass in grassland
ecosystems. Canadian Journal of Botany 51, 1867–1870.
Singh, J.S., Yadava, P.S., 1974. Seasonal variation in composition, plant biomass and net primary
production of a tropical grassland of Kurukshetra. India Ecological Monograph 44, 351–376.
Slack, D.C., Larson, C.L., 1981. Modelling infiltration, the key process in water management, runoff and
erosion. In: Lal, R., Russel, A.W. (Eds.), Tropical Agricultural Hydrology. Wiley, Chichester,
pp. 433–450.
Smith, V.R., 1985. Seasonal dynamics of standing crop and chemical composition in Marion Island (sub-
antartic) vegetation. Ph.D. Thesis, University of the Free State, Bloemfontein, 111pp.
Smith, H.J.C., Levy, G.L., Shainberg, L., 1990. Water-droplet energy and soil amendments: effect on
infiltration and erosion. Soil Science American Journal 54, 1084–1087.
Smoliak, S., Dormaar, J.F., Johnston, A., 1972. Long-term grazing effects of Stipa-Bouteloua prairie soils.
Journal of Range Management 25, 246–250.
Snyman, H.A., 1988. Measuring water-use efficiency from rangeland in the Central Grassveld from
evapotranspiration. Water South Africa 14, 153–158.
Snyman, H.A., 1993. The effect of defoliation during wilting on the production of Themeda triandra and
Eragrostis lehmanniana in semi-arid grassland. African Journal Range and Forage Science 10, 113–117.
Snyman, H.A., 1998. Dynamics and sustainable utilization of the rangeland ecosystem in arid and semi-
arid climates of southern Africa. Journal of Arid Environments 39, 645–666.
Snyman, H.A., 1999a. Soil erosion and conservation. In: Tainton, N.M. (Ed.), Veld Management in South
Africa. University of Natal Press, Scottsville, South Africa, 472pp.
Snyman, H.A., 1999b. Quantification of the soil–water balance under different veld condition classes in a
semi-arid climate. African Journal of Range & Forage Science 16, 108–117.
Snyman, H.A., 1999c. Short-term effect of soil–water, defoliation and rangeland condition on productivity
of a semi-arid rangeland in South Africa. Journal of Arid Environments 43, 47–62.
Snyman, H.A., 2000. Soil–water utilisation and sustainability in a semi-arid grassland. Water South Africa
26, 331–341.
Snyman, H.A., 2001. Water-use efficiency and infiltration under different rangeland conditions and
cultivation in a semi-arid climate of South Africa. Proceedings of the XIX International Grassland
Congress, Sao Paulo, Brazil, pp. 965–966.
Snyman, H.A., 2002. Short-term response of rangeland botanical composition and productivity to
fertilization (N and P) in a semi-arid climate of South Africa. Journal of Arid Environments 50,
167–183.
ARTICLE IN PRESS
Snyman, H.A., 2004. Soil seed bank evaluation and seedling establishment along a degradation gradient in
a semi-arid rangeland. African Journal of Range and Forage Science 21, 37–47.
Snyman, H.A., 2005. Rangeland degradation in a semi-arid South Africa—I: influence on seasonal root
distribution, root/shoot ratios and water-use efficiency. Journal of Arid Environments, doi: 10.1016/
j.jaridenv.2004.06.006, in press.
Snyman, H.A., Fouché, H.J., 1993. Estimating seasonal herbage production of a semi-arid grassland
based on veld condition, rainfall and evapotranspiration. African Journal of Range & Forage Science
10, 21–24.
Snyman, H.A., Opperman, D.P.J., 1984. Afloopstudies vanaf natuurlike veld in verskillende
suksessiestadia van die sentrale Oranje-Vrystaat. Journal of the Grassland Society of Southern Africa
1, 11–15.
Snyman H.A., Oosthuizen, I.B., 1999. Rangeland and soil condition: Their effects on productivity in a
semi-arid climate in South Africa. In: Proceedings of the VI International Rangeland Congress, vol. 1,
Townsville, Australia, pp. 211–212.
Soil Classification Working Group, 1991. Soil Classification: A Taxonomic System for South Africa.
Department of Agriculture Development, Pretoria South Africa, 262pp.
Sparling, G.P., Ross, D.J., 1988. Microbial contributions to the increased nitrogen mineralization after
air-drying of soils. Plant and Soil 105, 163–168.
Stroosnijder, L., 1996. Modelling the effect of grazing on infiltration, runoff and primary production in the
Sahel. Ecological Modelling 92, 79–88.
Strugnell, R.G., Piggott, C.D., 1978. Biomass shoot production and grazing of two grasslands in the
Rwenzori National Park, Uganda. Journal of Ecology 66, 73–96.
Tainton, N.M., 1981. The ecology of the main grazing lands of South Africa. In: Tainton, N.M. (Ed.),
Veld and Pasture Management in South Africa. Shuter & Shooter: University of Natal Press,
Pietermaritzburg, South Africa, 481pp.
Teague, W.R., Foy, J.K., Cross, B.T., Dowhower, S.L., 1999. Soil carbon and nitrogen changes following
root-plowing of rangeland. Journal of Range Management 52, 666–670.
Thurow, T.L., Blackburn, W.H., Taylor Jr., C.A., 1986. Hydrologic characteristics of vegetation types as
affected by livestock grazing systems. Journal of Range Management 39, 505–509.
Thurow, T.L., Blackburn, W.H., Taylor, C.A., 1988. Some vegetation responses to selected livestock
grazing strategies. Journal of Range Management 41, 108–114.
Troughton, A., 1957. The underground organs of herbage grasses. Commonwealth Bureau of Pastures
and Field Crops. Bulletin 44.
Troughton, A., 1978. The influence of reproductive development upon the root system of perennial
ryegrass and some effects upon herbage production. Journal of Agriculture Science 91, 427–431.
Van de Vijver, C.A.D.M., 1999. Fire and life in Tarangire: effects of burning and herbivory on an East
African savanna system. Ph.D. Thesis, Wageningen University, The Netherlands, 177pp.
Van den Berg, J.A., Roberts, B.R., Vorster, L.F., 1975. The effect of seasonal grazing on the cover and
composition of Cymbopogon-Themeda veld. Proceedings of the Grassland Society of Southern Africa
10, 111–117.
Van der Westhuizen, H.C., Van Rensburg, W.L.J., Snyman, H.A., 1999. The quantification of rangeland
condition in a semi-arid grassland of South Africa. African Journal of Range and Forage Science 16,
49–61.
Warren, S.D., Blackburn, W.H., Foulor, C.A., 1986a. Soil hydrologic response to a number of pastures
and stocking density under intensive rotation grazing. Journal Range Management 39, 500–504.
Warren, S.D., Blackburn, W.H., Taylor, C.A., 1986b. Effects of season and stage of rotation cycle on
hydrologic conditions on rangeland under intensive rotation grazing. Journal of Range Management
39, 486–491.
West, N.E., 1979. Formation distribution and function of plant litter. In: Goodall, D.W., Perry, R.A.,
Howes, K.M.W. (Eds.), Arid Land Ecosystems: Function Structure and Management, vol. 1.
Cambridge University Press, Cambridge, 866pp.
West, A.W., Sparling, G.P., Speir, T.W., 1989. Microbial activity in gradually dried or rewetted soil as
governed by water and substrate availability. Australian Journal of Soil Research 27, 747–757.
ARTICLE IN PRESS
Whitford, W.J., 1996. The importance of the biodiversity of soil biota in arid ecosystems. Biodiversity and
Conservation 5, 185–195.
Whitford, W.G., Stinnett, J., Anderson, J., 1988. De-composition of roots in a Chihuahuan Desert
Ecosystem. Oecologia 75, 8–11.
Wiegand, T., Snyman, H.A., Kellner, K., Paruelo, J.M., 2004. Do grassland have a memory: Modeling
phytomass production of a semi-arid South African grassland. Ecosystems 7, 243–258.
Williams, K.J., Wilsey, B.J., McNaughton, S.J., Banyikwa, F.F., 1998. Temporally variable rain does not
limit yields of Serengeti grasses. Oikos 81, 463–470.
Wiltshire, G.H., 1990. Nitrogen mineralization, nutrification and nitrogen balance in laboratory
incubation of soil from natural grassland in the central Orange Free State. South African Journal
of Plant and Soil 7, 45–49.
Winer, B.J., 1974. Statistical Principles in Experimental Design. McGraw-Hill, London, 218pp.
Wolfson, M.M., Tainton, N.M., 1999. The morphology and physiology of the major forage plants. In:
Tainton, N.M. (Ed.), Veld Management in South Africa. University of Natal Press, Scottsville, South
Africa, 472pp.
Wood, M.K., 1987. Plot numbers required to determine infiltration rate and sediment production on
rangelands in south central New Mexico. Journal of Range Management 40, 259–263.
Wright, H.A., Churchill, F.M., Stevens, W.C., 1976. Effects of prescribed burning on sediment, water
yield and water quality from dozed juniper lands in central Texas. Journal Range Management 29,
294–298.
Wright, H.A., Churchill, F.M., Stevens, W.C., 1982. Soil loss, runoff and water quality of seeded and
unseeded steep watersheds following prescribed burning. Journal Range Management 35, 382–385.

Snyman, Du Preez. 2005.

Uploaded by

Copyright:

Available Formats

Snyman, Du Preez. 2005.

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Snyman, Du Preez. 2005.

Uploaded by

Copyright:

Available Formats

ARTICLE IN PRESS

Rangeland degradation in a semi-arid South

The impact of rangeland degradation on soil characteristics (compaction, temperature,

Corresponding author. Tel.: +27-51-401-2221; fax: +27-51-401-2608.

Concerns about the effects of agricultural management practices on the

2. Materials and methods

2.1. Site description

2.3. Data collection

An estimate of soil compaction or soil penetration resistance was obtained from 30

2.4. Statistical analysis

The experimental layout was a fully randomized design consisting of three

3.1. Soil compaction

Rangeland degradation signiﬁcantly (pp0.01) increased soil compaction

3.2. Soil temperature

In all rangeland condition treatments the highest soil temperatures occurred as

3.3. Soil–water content and infiltrability

The soil–water content for the different rangeland conditions, determined

3.4. Root and litter turnover

Rangeland condition Turnover rate (year 1) Replacement roots Decomposition (aboveground)

Roots Litter Months Months

Good 0.59 0.52 20.34 23.08

3.5. Organic matter content

4.1. Soil compaction

4.2. Soil temperature

4.3. Soil–water content and infiltrability

4.4. Root and litter turnover

4.5. Organic matter content

decomposition in this rangeland condition. Where rainfall might be largely the

A prerequisite for sustainable land use is that soil quality is maintained or

degraded rangeland on an impoverished soil–water balance must not be under

You might also like