Livestock Research for Rural Development 27 (2) 2015 Guide for preparation of papers LRRD Newsletter

Citation of this paper

Influence of ngitili management on vegetation and soil characteristics in semi-arid Sukumaland, Tanzania

Ismail Saidi Selemani

Department of Animal Science and Production, Sokoine University of Agriculture,
P. O. Box 3004, Morogoro, Tanzania.
suma02seleman@yahoo.co.uk

Abstract

Establishment of temporary traditional exclosures known as ngitili has been used as coping strategy for rehabilitation of degraded rangelands in the Sukumaland. Ngitili involve conservation of standing hay during rainy season and open up for grazing at the peak of dry season. The present study evaluated the effects of ngitili management on vegetation attributes (herbaceous species composition and basal cover) and soil physical and chemical properties. Three existing grazing managements namely; private ngitili, communal ngitili and continuous grazing land were compared. From each grazing management, herbaceous species compositions and herbaceous vegetation cover were investigated by using a square quadrat (0.5 x 0.5 m) and the line intercept method respectively. Top soil samples were collected randomly from each grazing management for physical and chemical analysis.

The mean basal cover was found to be lowest for continuous grazing land and was highest for private ngitili. An increase in soil bulk density in continuous grazing land was an indication of soil compaction due to heavy grazing. The low values for soil organic carbon (SOC) and total nitrogen (TN) across all grazing managements could be due to poor vegetation decomposition caused by lack of moisture contents and unfavorable temperature in the study area. Moreover, the non-significant variation in the mean exchangeable cations and cation exchange capacity (CEC) across the grazing managements could be due to slow recovering of soil chemical functions.

Key words: grazing management, exclosures, species composition, basal cover


Introduction

Grazing is considered as the most economic way of utilizing rangeland vegetation where non-edible plant materials are converted to high quality food (meat and milk) for human consumption. However, uncontrolled grazing usually reduces plant covers that protect soil and thus lead to soil erosion and compaction (Oztas et al 2003). Rangeland degradation associated with overgrazing can result into substantial loss of herbaceous vegetation cover through destruction of plants by repeated defoliation and trampling. According to Hill et al (2006), the loss of herbaceous vegetation cover and shrubs exposes soil to erosion and oxidation of soil carbon. Overgrazing is considered the most important cause of reduction in palatable perennial plants in favor of less palatable and undesirable vegetation (Snyman and Du-Preez 2005). Poor grazing management practices usually lead to increased soil compaction, reduced soil aggregate ability and soil organic matter content. According to Snyman and Du-Preez (2005), soil organic matter is important in improving soil structure and thereby enhancing water infiltration.

 

On the other hand, a well managed grazing land stimulates growth of herbaceous species and improves nutrient cycling in rangelands (Schuman et al 2002). Currently establishment of exclosures has become an important tool used to restore degraded rangeland (Verdoodt et al 2009). Such restoration technique has positive effect on vegetation biodiversity, reduces soil erosion and increases water infiltration. Despite the fact that, soil physical, chemical and biological properties are important indicators of rangeland health, no previous studies were conducted in the north-western semi arid region of Tanzania to evaluate changes in soil properties in relation to grazing management. Understanding how soil functioning is critical for predicting rangeland trends and vegetation structure in semi-arid rangelands (Verdoodt et al 2010).

 

Historically, rangeland degradation in the north-western region of Tanzania dated back to colonial era. Vegetation clearance programme as the way of eradicating tsetse flies was environmentally catastrophe that led to severe vegetation loss in these regions (Barrow and Mlenge 2003). In 1980’s rehabilitation of these rangelands has been fostered through establishment of traditional exclosures known as ngitili (Barrow and Shah 2011). This traditional practice involves conservation of standing hay during rainy season and open up for grazing at the pick of drying season when feed is scarce. Managements of such temporary exclosures are under private and communal ownership. Despite the decisive role of exclosures in rehabilitation of degraded rangelands, yet few attempts have been done to evaluate effects of ngitili management on vegetation recovery in terms of herbaceous species composition and coverage.  Moreover, the effect of ngitili management on soil properties in the study area remains unknown. Biotic factors such as vegetation and abiotic factors (soil properties) are two most useful indicators of rehabilitation success that could provide the picture of rangeland health. The present study presented results on herbaceous vegetation composition, basal cover and soil physical and chemical properties in relation to grazing management (private ngitili, communal ngitili and continuous grazing lands).


Material and methods

Study area

 

This study was carried out in Shinyanga and Simiyu regions, Tanzania. Historically the Simuyu region was a part of the Shinyanga region (Figure 1) before it was officially gazetted by the government as a new administrative region in 2012. These two regions are located at 2-3oS; and 31-31.5oE and their altitude is between 1000 to 1500 m above sea level (Rubanza et al 2007). Ecologically, most of the study areas are semi-arid with mean annual rainfall of 600 mm (Wiskerke 2008). The rainfall is normally unimodal, falls from November to April. The study area had large variation in rainfall pattern and distribution. For example, in most of years, the dry season mean precipitation is less than 50 mm per year (Kamwenda 2002). The landscape is mostly flat, covered with isolated stones hills and scattered acacia trees. Historically the region was heavily covered with miombo and acacia woodlands before introduction of a massive clearing programme for eradication of tsetse flies (Pye-Smith 2010).

Figure 1: The map of Tanzania showing study area (Shinyanga). The map adopted from Wiskerke (2008)

Sampling procedures

 

The vegetation survey was carried out in Shinyanga rural (Shinyanga region) and Meatu district (Simiyu region) from October to November in 2011. The study was base on existing traditional conserved forages (ngitili). Three grazing management namely; private ngitili, communal ngitili, and continuously grazing land were compared in terms of vegetation composition, basal cover and soil physical and chemical properties. Continuous grazing in this context is defined as yearlong grazing where animals are on a range unit for at least the whole growing season (Heady 1961). Data were collected in four different sub-plots established from each grazing management.  Out of four sub-plots from each grazing management, two sub-plots were from Shinyanga rural district and the other two sub-plots were from Meatu district. The size of each sub-plot was approximately one acre. A careful pre-survey was done one week before experiment to layout the appropriate grazing managements, establishing sub-plots within grazing management and marking the transect lines in each sub-plot.  The GPS device was used to mark the coordinates for each sub-plot. 

 

In each sub-plot, two parallel transect lines of 400 m long were established. The distance between the transect lines were 200 m. However the design of transect lines was adjusted accordingly depend on shape and size of particular sub-plot. A wooden square quadrat (0.25 x 0.25 m) was used to estimate herbaceous species composition from each sub-plot. The quadrat was dropped along the transect line at the interval of 20 m apart. All species found within the quadrat were identified and the numbers of individual species were recorded. Species that were not identified in the field were sampled and taken to the Sokoine University of Agriculture for further identification. Herbaceous vegetation cover was estimated by using the line intercept method developed by Canfield (1941) and modified by Cumming and Smith (2000). In this study a 50 m long tape measure was used as one sampling unit of which four sampling units were established per transect line making eight sampling units per sub-plot and a total of 16 sampling units per grazing management. The distances of tape measure intercepted by plant species were recorded. Furthermore, distances of tape measure intercepted by grasses, forbs, shrubs and trees were also recorded in a separate recording sheet.

 

For soil sampling four holes were excavated (to a depth of 40 cm) randomly in each sub-plot. Eight soil samples were taken from each sub-plot whereas four of them were top soil (to a depth of 20 cm) and other four deep soil (to a depth 40 cm). All samples were thoroughly mixed, air dried and passed through 2 mm mesh screen to be ready for chemical analysis. A set of eight samples from each sub-plot were bulked separately (four top soil samples and four deep soil samples). From the bulked soil, four sub-samples (two top samples and two deep samples) were chemically analyzed in the Department of Soil Science at Sokoine University of Agriculture. Soil chemical analyses were done based on procedures described by Jackson (1970). For determination of soil bulk density four replicates of soil cores were taken from each grazing management. The cores made up of metallic rings were hammered into the top soil, labeled and taken to the Department of Soil Science at Sokoine University of Agriculture. The bulk density was determined by oven-drying at 105oC for 48 h a method described by Tefera et al (2007).

 

Statistical analysis

 

The general linear model of SAS (2004) was applied to analyze the effect of grazing management on vegetation cover and soil properties. The following model was used in this experiment: Yij= µ + αi + €ij, where Yij is general responses, µ is overall mean, αi effects of grazing management (private ngitili, communal ngitili and continuous grazing land) and €ij is standard error. The means were compared using the Duncan multiple range test as described by Montgomery (2001).


Results

Herbaceous vegetation composition and basal cover

 

The herbaceous vegetation composition in all three grazing management was dominated by the most two common grass species namely; Sweet pitted grass (Bothriochloa insculpta) and black-spear grass (Heteropogon contortus). The most common species in this context are those recorded in almost all grazing managements (Table 1). Accordingly, other common distributed herbaceous species include, Indigofera species, tridax daisy (Tridax procumbens), and shrub stylo (Stylosanthes scabra). Some species such as star grass (Cynodon nlemfuensis), Indian bluegrass (Bothriochloa pertusa), digitgrass (Digitaria decumbens) and Thatching grass (Hyparrhenia rufa) were skewed distributed, found to dominate highly on private ngitili than other grazing managements. On the other hand, Tridax procumbens, Indigofera spicata, Stylosanthes scabra and Commelina benghalensis were relatively abundant in communal ngitili and continuous grazing land than the private ngitili.

Table 1: Herbaceous species composition from three grazing managements

Herbaceous species

Frequency of occurrence (in %) per quadrat (0.25 x 0.25 m)

 

Private ngitili

Communal ngitili

Continuous grazing land

Amaranthus muricatus

0.37

0.28

0.00

Aristida spp

1.87

1.95

0.38

Bothriochloa pertusa

1.50

0.00

0.00

Bothriochloa insculpta

25.5

30.70

37.30

Cenchrus ciliaris

4.12

1.56

2.66

Commelina benghalensis

0.00

1.11

3.04

Cynodon nlemfuensis

21.70

0.84

0.38

Cyperus rotundus

0.75

1.39

0.00

Digitaria decumbens

4.87

0.5

0.00

Heteropogon contortus

15.40

23.4

13.70

Hyparhenia rufa

4.12

1.11

0.76

Indigofera spicata

4.12

13.10

9.89

Allium crispum

1.12

1.28

1.53

Sida gromerata

1.12

0.28

2.28

Stylosanthes scabra

1.50

8.36

7.22

Tridax procumbens

12.00

14.20

20.90

Table 2 presented the herbaceous vegetation coverage across the grazing managements. All ngitili (both private and communal) were found to have significantly higher vegetation cover compared to continuous grazing land (P < 0.05).  More than 70 % of the land in continuous grazing management was bare soil. Grass cover was almost 2.5 times higher in ngitili than in continuous grazing lands. Moreover, tree cover was found to be 3 times higher in private ngitili than communal ngitili and 2 times higher in private ngitili than in continuous grazing land.  Shrubs and forbs coverage were found to be independent of grazing management.

Table 2: Vegetation cover  from three grazing management systems

 

Private ngitili

Communal ngitili

Continuous grazed land

SEM

P value

Grasses (%)

36.5a

36.2a

14.4b

3.6

0.01

Forbs (%)

1.8

2.1

0.5

0.7

0.22

Shrubs (%)

3.9

4.1

3.4

0.4

0.92

Trees (%)

19.9a

6.4b

9.1b

2.6

0.01

Total basal cover (%)

62.2a

48.8b

27.3c

8.2

0.01

Influence of grazing management on Soil properties

 

The mean exchangeable cations and cation exchange capacity from study areas have been summarized in tTble 3. Results show non-significant different in both exchangeable cations and CEC between protected exclosures and continuous grazing lands. Neither private ngitili nor communal ngitili found to differ significantly in terms of exchangeable cations and cation exchange capacity (P > 0.05). Despite the non-significant difference in CEC, continuous grazing lands appeared to have slightly higher value of CEC against protected exclosures.

Table 3: The exchangeable cations and CEC of top soil under different grazing management

Grazing system

Exchangeable cations (ppm)

C.E.C

Ca

Mg

K

Mn

(meq/100gm)

Private ngitili

4077.4

270.9

249.6

212.7

13.6

Communal ngitili

3750

218.3

189.3

62.4

11.3

Continuous grazing

4755.1

232.6

180.6

149.7

17.9

SEM

1269.1

29.6

33.8

68.5

4.8

P value

0.9

0.5

0.6

0.3

0.4

The soil texture composition, its bulk density and pH of soil sampled from study areas are summarized under table 4. There is variation in texture composition (silty loam and sandy loam) across study sites. Private ngitili found to have 18 units higher percentage of sandy loam compared to communal ngitili and continuous grazing lands. The communal ngitili and continuous grazing lands were found to have more or less similar textural composition. The bulk densities of soil vary slightly across grazing management, the value being higher in continuous grazing land followed by communal ngitili and private ngitili. However, these values were not statistically significance (P > 0.05). The pH range of soil is slightly acidic to slightly alkaline from communal ngitili, continuous grazing lands, and private ngitili respectively (Table 4). However, the variation in pH was also not statistically significance (P > 0.05).

Table 4: Texture composition, Bulky density, and pH of soil sampled from different grazing management

Grazing system

Silty Loam (%)

Sandy Loam (%)

Bulk Density (gcm-3)

pH (mean ± SE)

Private ngitili

37.5b

62.5a

1.43

7.35

Communal ngitili

56.3a

43.8b

1.52

6.38

Continuous grazing

56.3a

43.8b

1.65

6.53

SEM

10.8

10.8

0.11

0.44

P value

0.01

0.01

0.08

0.55

The organic carbon, total Nitrogen and C:N ratio are summarized in table 5. Communal ngitili found to have low percentage organic carbon and total Nitrogen. Private ngitili and continuous grazing sites had more or less similar values of organic carbon and total Nitrogen. Private ngitili found to have slightly higher value of C:N ratio followed by continuous grazing sites and communal ngitili in respective order. However, these variations presented in table 5 are statistically non-significant.

Table 5: Organic Carbon and total Nitrogen  of soil  from three grazing managements

Grazing management

Organic carbon (ppm)

Total Nitrogen (ppm)

C:N ratio

Private ngitili

1.04

0.11

9.59

Communal ngitili

0.77

0.08

9.02

Continuous grazing

1.00

0.11

9.20

SEM

0.15

0.01

0.18

P value

0.23

0.09

0.09


Discussion

The main objective for the establishment of ngitili in the semi-arid region of the north-western Tanzania was to allow regeneration of native vegetation as the way to provide forage for livestock and reduce soil erosion. In semi-arid Shinyanga and Simuyu regions, the establishment of seasonal exclosures (ngitili) has been fostered by the Environmental and Soil Conservation Programme, in swahili known as “Hifadhi Ardhi Shinyanga” (HASHI) since 1980’s. The significant variation in herbaceous plant cover between protected land and non-protected ones is an indication of vegetation response following grazing management. Reduction in herbaceous vegetation coverage in continuous grazing land as compared to protected exclosures (both private and communal ngitili) is most likely due to heavy grazing pressure. According to Snyman (1999), the most significant factor responsible for loss of vegetation cover is repeated defoliation of palatable plant species following poor grazing practices and stocking animals beyond acceptable ecological carrying capacity. In the present study, more than 70 % of continuous grazing land was bare soil, the findings that were almost agreed with those reported by Verdoodt et al (2010) who found that about 63-67 % of continuously grazed rangelands in Kenya were bare grounds. According to Yong-Zhong et al (2005), continuous grazing is detrimental to rangeland vegetation and soil. It allow less vegetation cover and litter accumulation that consequently lead to poor microbial activities and hence low organic carbon and nitrogen concentration in the soil.

 

Uncontrolled grazing in continuous grazing lands causes decline in soil physical properties, resulting in changes in vegetation and nutrient cycling (Pei et al 2008). An increase in value of bulk density in continuous grazing sites is an indication of increased in soil compaction. Oztas et al (2003) described that reduced in plant coverage due to heavy grazing pressure normally resulted into soil erosion and compaction. According to Synman and Du-Preez (2005), loss of vegetation cover allows direct impact of raindrops on soil and thus directly contributes to soil compaction. In general soils of high bulk density have unfavorable physical condition than those of low values. An increase in soil compaction may reduce rate of water infiltration as described by Synman and Du-Preez (2005).

 

Despite the fact that, exclosures enhance soil organic carbon (SOC) and total nitrogen (TN) accumulation and decreased soil pH (Pei et al 2008), the present study could not found significance differences in these variables between exclosures and continuous grazing lands.  The non-variation in SOC, TN and pH between exclosures and continuous grazing land could probably because of poor activities of micro-organisms in the soil in all researched plots. Poor microbial activities in the soil are most likely because of insufficient moisture contents in the soil that probably resulted in poor root systems (Su et al 2004). Kirschbaum et al (1995) reported that, amount of SOC and TN found to be positively correlated to precipitation and negatively correlated to temperature. In semi-arid rangeland, precipitation is very low and temperature is too high to favor vegetation decomposition through microbial activities. The mean annual rainfall in the study area is around 600 mm per annum and minimum and maximum temperatures vary from 16.1 to 33.5oC respectively (Rubanza et al 2007).

 

The non-significant difference in the C:N ratio cannot be taken as a clear-cut indication of none significant variation in soil fertility in all three grazing managements. According to Tefera et al 2007), SOC and TN are most sensitive soil quality indicator, thus within a narrow range deviation it may serve as a suitable indicator of variation in soil fertility. Following the above argument, a slight reduction in SOC and TN in communal ngitili may be an evidence of poor soil condition. Management of communal grazing lands across semi-arid regions is constrained with number of factors that make soil conservation difficult (Takar et al 1990). Heavy grazing pressure due to unclear right of entrance and use of common resources is one of the challenges facing communal rangelands. Most of these communal rangelands are not fenced and their resources are freely access to all communities (Selemani 2014).

 

On the other hand, the observed relative low herbaceous vegetation coverage in communal ngitili compared to private ngitili probably suggests an existence of heavy grazing pressure in communal lands than privately owned lands. Similar result was also reported by Tefera et al (2007), who found that bare ground was by far more common on communal land in semi-arid Borana rangeland than other land uses. The heavy grazing pressure in communal ngitili could be due to poor grazing management in communal rangelands. Quinn et al (2007) associated poor condition of most communal rangelands in Tanzania with poor monitoring and lack of clear boundaries. According to Selemani et al (2013), poor vegetation productivity in communal ngitili is a result of lack of community responsibility in the management of common resources. The authors associated lack of community responsibility with un-equal utilization of resources from communal ngitili and lack of participatory in decision making. For sustainable rehabilitation of degraded communal rangelands, I recommend these communal grazing areas to be managed like the wildlife management areas (WMAs)in Tanzania, where the decision making powers are participatory and resources accrued from communal rangelands are shared equally.

 

The results from present study also show that the means for exchangeable cations and cation exchange capacity were not significant difference for all grazing managements. These findings are supported by those of Tefera et al (2007), who reported that exchangeable cations in communal land, traditional grazing reserve and government ranch in southern Ethiopia was not statistically different. Non-significant variation in exchangeable cations and CEC may be attributed to slow recovery process of soil chemical functions. Verdoodt et al (2009) described that soil functions recover much more slowly than vegetation structure and diversity thus makes difficult to quantify the variation in CEC. Alternative explanation to this could be because of poor organic matter (SOC and TN) in the soil in all grazing managements, since one of the effect of organic matter is to retain and protect cations from leaching and from removal by runoff (Mekuria et al 2007).

 

Results on soil texture revealed that communal ngitili and continuous grazing sites have more or less equal proportion of silty loam and sandy loam. On the other hand, private ngitili found to have more composition of sandy loam (62.50 %) than silty loam (37.50 %). These texture composition is too complex to be explain by single factor such as impact of grazing alone since soil formation develop over time as a result of interactions of several factors including climate, topography, parental materials and vegetation type. Nevertheless, the variation in herbaceous species composition across the three grazing managements could be contributed by variation in textural composition. For example, the relatively low occurrence of Bothriochloa insculpta in private ngitili could be because of high proportion of sandy soil. Generally this species is not suited to sandy soil (Skerman and Riveros 1989). On the other hand, the relatively high occurrence of Heteropogon contortus in the private ngitili is most likely because of high proportion of sandy soil. Generally, Heteropogon contortus grows well in wide variety of well-drained soil including sandy loam soil (Skernam and Riveros 1989). Alternative explanation could also be that, heavy grazing pressure in the continuous grazing land and communal ngitili might have limited H. contortus to become dominant. According to Bielfelt (2013), heavy grazing pressure may prevent H. contortus from becoming dominant, whereas reduced grazing may allow increased dominance. However, the textural compositions of soil in all three study areas are in the favorable range for pasture growth given other conditions are not limiting.


Conclusion


References

Barrow E and Mlenge W 2003 Trees as key to pastoralist risk management in semi-arid landscapes in Shinyanga, Tanzania and Turkana, Kenya. The International Conference on Rural Livelihoods, Forests and Biodiversity, Bonn, Germany. 15 pp. 

Barrow E and Shah A 2011 Restoring Woodlands, Sequestering Carbon and Benefiting Livelihoods in Shinyanga, Tanzania. Economic of ecosystem and biodiversity. (http://www.teebweb.org) accessed on 15th December, 2014. 

Bielfelt B J 2013 Invasion by a native grass: Implications of increased dominance of Heteropogon contortus for grassland birds. Master thesis, University-Kingsville, Texa. 117pp

Canfield R H 1941  Application of the line interception method in sample range vegetation. Journal of Forestry, 39: 388–394. 

Cummings J and Smith D 2000 The line-intercept method: A tool for introductory plant ecology laboratories. Proceedings of the 22nd Workshop/Conference of the Association for Biology Laboratory Education (ABLE). 234-246 pp. 

Heady H F 1961 Continuous vs. Specialized Grazing Systems: A Review and Application to the California Annual Type. Journal of Range Management, 14 (4): 182-193. 

Hill M J, Roxburgh S H, McKeon G M, Carter J O and Barrett D J 2006 Analysis of soil carbon outcomes from interaction between climate and grazing pressure in Australian rangelands using Range-ASSESS. Environmental Modelling and Software, 21 (6): 779-801. 

Jackson M L  1970 Soil Chemical Analysis: Prentica-Hall, Inc, Englewood cliffs, NJ.

Kamwenda G J 2002 Ngitili agrosilvipastoral systems in the United Republic of Tanzania. Unasylva, 53: 46-50. 

Kirschbaum M U F 1995 The temperature and dependence of soil organic matter decomposition and the effect of global warming on soil organic C storage. Soil Biology and Biochemistry, 27: 753-760. 

Mekuria W Veldkamp E, Haile M, Nyssen J, Muys B and Gebrehiwot K 2007 Effectiveness of exclosures to restore degraded soils as a result of overgrazing in Tigray, Ethiopia. Journal of Arid Environments, 69 (2): 270-284. 

Montgomery D C 2001 Design and Analysis of Experiments (5th ed). John Wiley & Sons, York. 

Oztas T, Koc A and Comakli B 2003 Changes in vegetation and soil properties along a slope on overgrazed and eroded rangelands. Journal of Arid Environments, 55 (1): 93-100. 

Pei S, Fu H and Wan C 2008 Changes in soil properties and vegetation following exclosure and grazing in degraded Alxa desert steppe of Inner Mongolia, China. Agriculture, Ecosystems and application Environment, 124 (1–2): 33-39. 

Pye-Smith C (ed.) 2010  A Rural Revival in Tanzania. How agroforestry is helping farmers to restore the woodlands in Shinyanga Region. The World Agroforestry Centre, Nairobi. 37 pp. 

Quinn C H, Huby M, Kiwasila H and Lovett J C 2007 Design principles and common pool resource management: An institutional approach to evaluating community management in semi-arid Tanzania. Journal of Environmental Management, 84: 100-113. 

Rubanza C D K, Shem M N, Bakengesa S S, Ichinohe T and Fujihara T 2007 Effects of Acacia nilotica, A. polyacantha and Leucaena leucocephala leaf meal supplementation on performance of Small East African goats fed native pasture hay basal forages. Small Ruminant Research, 70 (2-3): 165-173. 

Schuman G E, Janzen H H and Herrick J E 2002 Soil carbon dynamics and potential carbon sequestration by rangelands. Environmental Pollution, 116 (3): 391-396. 

Selemani I S, Eik L O, Holand Ø, Ådnøy T, Mtengeti E and Mushi D 2013 The effects of a deferred grazing system on rangeland vegetation in a north-western, semi-arid region of Tanzania. African Journal of Range & Forage Science, 30 (3): 141-148. 

Selemani I S 2014 Communal rangelands management and challenges underpinning pastoral mobility in Tanzania: a review. Livestock Research for Rural Development, Volume 26, Article #78. http://www.lrrd.org/lrrd26/5/sele26078.html

Skerman P J and Riveros F 1989 Tropical grasses. Food and Agriculture Organization of United Nations. FAO, Rome, Italy. 832 pp.

Snyman H A 1999 Soil erosion and conservation, In: Veld Management in South Africa, Tainton N M (ed).University of Natal Press: Pietermaritzburg. 355-380 pp. 

Snyman H A and Du Preez C C 2005 Rangeland degradation in a semi-arid South Africa II: influence on soil quality. Journal of Arid Environments, 60 (3): 483-507. 

Su Y Z, Zhao H L, Zhang T H and Zhao X Y 2004 Soil properties following cultivation and non-grazing of a semi-arid sandy grassland in northern China. Soil and Tillage Research, 75 (1): 27-36. 

Takar A A, Dobrowolski J P and Thurow T L 1990 Influence of Grazing, Vegetation Life-Form, and Soil Type on Infiltration Rates and Interrill Erosion on a Somalion Rangeland. Journal of Range Management, 43 (6): 486-490. 

Tefera S, Snyman H A and Smit G N 2007 Rangeland dynamics in southern Ethiopia: (1) Botanical composition of grasses and soil characteristics in relation to land-use and distance from water in semi-arid Borana rangelands. Journal of Environmental Management, 85 (2): 429-442. 

Verdoodt A, Mureithi S M, Ye L and Van Ranst E 2009 Chronosequence analysis of two enclosure management strategies in degraded rangeland of semi-arid Kenya. Agriculture, Ecosystems and Environment, 129 (1-3): 332-339. 

Verdoodt A, Mureithi S M and Van Ranst E 2010 Impacts of management and enclosure age on recovery of the herbaceous rangeland vegetation in semi-arid Kenya. Journal of Arid Environments, 30: 1-8. 

Wiskerke W 2008 Towards a sustainable biomass energy supply for rural households in semi-arid Shinyanga, Tanzania. Thesis in Master of Science. Utrecht: Utrecht University, Department of Science, Technology and Society. 112 pp. 

Yong-Zhong S, Yu-Lin L, Jian-Yuan C and Wen-Zhi Z 2005. Influences of continuous grazing and livestock exclusion on soil properties in a degraded sandy grassland, Inner Mongolia, northern China. CATENA, 59 (3): 267-278 


Received 20 November 2014; Accepted 20 January 2015; Published 4 February 2015

Go to top