Rangeland condition and livestock carrying capacity under the traditional rotational grazing system in northern Tanzania

Anthony Z Sangeda and David D Maleko

Department of Animal, Aquaculture and Range Science, College of Agriculture, Sokoine University of Agriculture P O Box 3004, Morogoro, Tanzania
sangedaaz@gmail.com

Abstract

The rationale behind this study was to contribute into a better understanding of how traditional rotational grazing is effective in overcoming vegetation loss and pasture inadequacy in communal rangelands in the semi-arid areas. The grazing sites were categorized into lowlands, midlands and uplands. Systematic random sampling technique through use of transects and quadrat was employed to collect vegetation and soil data. The data were analyzed for descriptive statistics using Microsoft Excel Spreadsheet.

The herbaceous vegetation cover was highest (34.1%) in the upland, followed by midland (21%) and the lowland had least (3.7%). Average woody plants canopy cover was 5287±78 m²/ha in which Acacia tortilis , Acacia nilotica and Commiphora africana had highest canopy cover of 1293±12, 974±19 and 780±9m²/ha, respectively. The soils were slightly alkaline with mean pH of 8.18 ± 0.18 and mean soil organic carbon was 9.6±0.97 t/ha. The lowland areas had the lowest livestock carrying capacity of 7 TLU (ha/LU/yr), while both midland and upland sites had 3 TLU (ha/LU/yr). This study demonstrates that northern Tanzania semi-arid rangelands are characterized with bush encroachment and low livestock carrying capacity of around 3 TLU (ha/LU/yr).

Key words: bush encroachment, Ngorika-Manyara, pasture, stocking rate, vegetation cover

Introduction

Extensive grazing that involves herders and animal mobility, for centuries has enabled pastoralists to survive the risks and uncertainties such as drought and diseases in semi-arid rangelands (Lund 2007). Low and erratic rainfalls often less than 500mm/annum coupled with high evapotranspiration rates are known to be the major limiting factors for primary productivity in the semi-arid areas. In addition, the current climate change and variability has exacerbated the situation by making dry season feed shortage a more prominent problem in arid and semi-arid rangelands (Lohmann et al 2012; Martin et al 2014). A number of studies have indicated that rotational grazing commonly known as transhumance (movement between wet and dry pasture resources) plays an important role of reducing the problems of vegetation and soil loss in semi-arid areas. Degradation of semi-arid rangeland resources is mainly driven by overgrazing and excessive vegetation burning (McGranahan and Kirkman 2013). Nevertheless, pastoral mobility in various forms including nomadism and transhumance has enabled herders to access and utilize limited pasture and water resources (Fernandez-Gimenez and Le Febre 2006).

Despite the discerned advantages of the pastoral mobility, range lands have been declining in many parts of the world for a diverse of reasons including agricultural expansion into grazing lands and extension or establishment of wildlife protected areas (Niamir-Fuller 1999; Hobbs et al 2008; Turner et al 2016). Sedentarization policies that foster development of services such as waterholes, schools and health services as incentive or tools for forceful pastoral settlement is among other factors for shrinkage of pastoral mobility (Fernandez-Gimenez et al 2006; Butt 2010). Roba and Oba (2008) found out that pastoral settlement and decrease in grazing radii led to degradation of watering points and replacement of palatable herbaceous species by invasive shrubs and weeds in the semi-arid rangelands of Kenya. Indeed, disruption of the traditional grazing systems of Kajiado, Kenya through introduction of group ranches led to decline in livestock productivity with eventual negative effects to pastoral livelihoods (Thornton et al 2006).

In addition, to the aforementioned external pressures to pastoral production system in the East African semi-arid rangelands, also internal/local pressures including human and livestock population increase which lead to resource competition and degradation do operate concurrently (Talbot 1986). De Leeuw and Tothill (1990) defined livestock carrying capacity (K) as “the maximum number of animals, usually expressed as a standardized ‘Livestock Unit’ that an area of land can support on a sustainable basis”. In this case, a 250 kg live bodyweight animal is assumed to require daily dry feed of between 2.5 and 3.0% its bodyweight for effective maintenance and production. Pratt and Gwynne (1977) approximated 4 ha land (K) as optimal area for supporting one tropical livestock unit annually in the East African semi-arid rangelands. Nevertheless, it is necessary to estimate K for a given grazing site because K is not a static factor. Thus, it is necessary to assess and monitor rangeland condition and K for a given grazing site in order to understand whether the current range management practices of a particular site is appropriate or requires interventions for sustainability of rangeland resources.

Information with regard to rangeland condition and K in semi-arid pastoral areas that are under the traditional rotational grazing system is scarce. In this case, it is important to assess rangeland condition and livestock carrying capacity for semi-arid rangelands so as to generate new information with regard to status of livestock population, vegetation, soil and water resources that are necessary for grazing plans. Such information is a key to developing strategies that would ensure sustainable rangeland productivity and improved pastoral livelihoods.

Materials and methods

Study site and sampling design

The study was conducted in a semi arid area in Ngorika ward located in Simanjiro District, Manyara Region, Northern Tanzania. Simanjiro district is located between latitude 4° 24′ 36″ S and longitude 37° 10′ 48″ E. The lowland grazing areas namely Lorugum are located at an altitude of 659 m a.s.l along the Pangani River basin. The upland grazing areas namely Lolosokwan are located at an altitude of 981m a.s.l on the Lalatema Hills (Figure 1). Average annual rainfall ranges from 400 to 600mm, thus falling within the semi-arid ecological zone. The rain is bimodal in which the short rains fall between earlier November and mid-January and the long rains falls between earlier March to end of May. However, the rains are highly unpredictable and short rains often fail (Msoffe et al 2011).

Figure 1. A map for Ngorika ward showing the study villages, in Simanjiro District, Manyara, Tanzania

A total of 14 line transects ranging between 1-2 km long were laid in the study villages namely Lemkuna, Ngorika and Nyumba ya Mungu (N/Mungu) in which there were four transects in each village, except Lemkuna which had six. The study sites were categorized into three groups: lowland (wet season grazing areas), midland (late dry season grazing areas) and upland (grazing earlier to mid-dry season grazing area). Lowland areas included Lorugum in Lemkuna village, Kambi ya Swala in N/Mungu village and both sites were located close to Pangani River (less than 2 km from the River bank). The rest of the sites were established in Engusero in Ngorika village which is located close to N/Mungu Dam. Whilst, the midland sites included Endiyamtu, Elutoto and Olalili in Lemkuna, N/Mungu and Ngorika village respectively and they were located at the foot of Lalatema Hills. These areas were mainly reserved for late dry season grazing pasture in form of standing hay “Olalili”. The upland grazing site named Lolosokwan was located in Lemkuna village at the plateau area on Lalatema hills and is used for earlier to mid-dry grazing by cattle from all the three villages in Ngorika ward. At the beginning of dry season there exist seasonal watering points at Lolosokwan but during late dry season cattle should walk about 27 km to access water in Pangani River.

In each grazing site, two transects were established. Each transect was 1-2 km long depending on the size of the particular grazing site were established. Within each transect five (5) sampling sites each having 50 m long were established whilst avoiding areas where kraals were newly located. The inter-sampling site’s distance was 500 m. Within each sampling site a 50 m long tape measure was stretched and the rangeland condition was used to assess the rangeland health through line interception technique (Cummings and Smith 2000). Within the line transect, the distance covered by grass, bare, forb, litter and browse cover was recorded separately.

Moreover, within the sampling site three quadrats each measuring 50 cm x 50 cm were systematically placed at 0 m, 25 m and 50 m marks of the tape measure (systematic random sampling). Within each quadrant unit; herbaceous vegetation were clipped at a height of about 5 cm above the ground. Then, livestock uneatable materials such as dry sticks and unpalatable weeds were sorted of clipped vegetation sample, thereafter the sample was appropriately packed and sent to laboratory where it was dried to constant weight at 80⁰C. Thereafter, dry matter (DM) biomass and livestock carrying capacity was determined following procedures described by De Leeuw and Tothill (1990).

Further, within the same sampling points undisturbed soil core samples were taken for determination of soil bulk density. In addition, soil samples were collected at 10 cm deep from each point and then bulked and sent to laboratory for analysis of pH, organic carbon and soil textural classes. The amount of soil organic carbon per hectare was estimated by formula provided by GRDC (Undated): Organic carbon/hectare (Tonnes carbon/hectare) = 10,000 m² x soil depth (m) x bulk density (g/cm³) x (% organic carbon/100).

Within each sampling site Point Centered Quarter (PCQ) method was employed in which a cross was placed at 25 m mark of tape measure for locating the nearest four trees falling under the 4 cross quarters (Mitchell 2010). Parameters recorded under PCQ method included quarter distance, tree/shrub canopy diameter and height that are necessary for estimation of tree density, cover and available browse for foraging animals. Air dried samples were packed for analysis. The samples were analysed at the Department of Animal, Aquaculture and Range Sciences, Sokoine University of Agriculture in Morogoro, Tanzania. Analysis of dry matter (DM), ash, crude fiber (CF) and ether extract was conducted according to procedures described in AOAC (1990).

In addition, forage samples for the most dominant forage grass species and the browse species that were identified by the experienced local herders as of high value feed were collected. Further, interviews with key informants were carried out. These included livestock, veterinary and agricultural officers, knowledgeable livestock keepers and traditional elders. These interviews provided information on grazing and livestock management practices.

Data analysis

The data were analyzed by means of Microsoft Excel Spreadsheet computer program to generate descriptive statistics including percentage, mean and standard error. Content analysis was used to extract useful information obtained in the in-depth interviews.

Results and discussion

Livestock population and grazing patterns

Pastoralism based on seasonal movement between wet and dry season pastures (transhumance) in a defined pattern is the predominant land use in Ngorika ward. There were about 9,373 cattle, 7,963 goats, 3,227 sheep and 114 donkeys (Table 1). Crop farming is minimal and concentrated at Lemkuna village floodplains and valleys adjacent to the Pangani River basin locally called Ngage. The main crops that are grown at Ngage include paddy, maize and various vegetables such as water melons, onions and tomatoes. The area under irrigation was approximately 5% of Ngorika ward’s total land. During wet seasons livestock were grazed in the lowlands near home. It was further revealed that most goats, sheep and donkeys graze in the lowland throughout the year. However, with the advance of dry season usually from July to August cattle are moved to uplands (Lolosokwan). Grazing at Lolosokwan involve encamping due to its distant location (about 27 km) from settlements where the permanent watering points are also located (Pangani River and N/Mungu Dam). Late dry season begins from October spanning to late February and during this period livestock grazes in the reserved grazing areas “Olalili”. It was further reported that livestock will only start grazing on the Olalili after the elders have met and agreed to open the reserved grazing areas. The grazing itself has pattern to follow in which calves and weak cows will start grazing first for some few days before the entire herd is allowed to graze. Nevertheless, there are norms that discourage grazing of reserved pastures before the right time which is normally when unreserved areas are depleted of pasture and animals start to loose condition, specifically at the peak of drought condition.

Selected soil properties

Lolosokwan was occupied by red clay soils with some sandy areas. The lowlands had light sandy soils along the offshores of N/Mungu Dam and close to Pangani River basin. These are areas that were providing forage for dry season but soils were deemed to dry out rapidly due to higher coarseness. In general, midlands were dominated by sandy loam soils and with some patches having vertisols (mbuga soils) that were important for dry season grazing (Table 2). The soils were slightly alkaline with mean pH of 8.18 ± 0.18 and visually the lowland soils were calcareous with chalky color. The mean soil organic carbon was 9.6 ± 0.97 t/ha which implies that rangeland soils has some potential for sequestering carbon (Table 2). However, the level of soil carbon storage observed in this study is very low compared to that of tropical submontane natural rainforests reserved as catchments in Eastern Arc Mountains of Tanzania. In which, Munishi et al. (2004) reported soil organic carbons at 0 -15 cm depth of about 164 and 246 t/ha for Usambara and Uluguru natural forest reserves, respectively. Nonetheless, Smith (2014) evidenced that legacy effects of grassland use and management have strong effects on the soil carbon contents and cautioned that before generalizing the measurements, history of the grassland sites should be known. Higher soil organic carbon in some lowland grazing areas can be attributed to their close location to the settlements where night kraals are also located. In addition, the lowland areas were under continuous grazing and it was further noticed that goats and donkeys forage on the lowlands throughout the year. Thus, probably higher manure droppings might have resulted to the higher organic soil carbon in most lowland grazing sites. Low soil organic carbon content in the midland areas might be attributed to the short grazing time and the predominant sandy loam soils.

Trees and shrubs’ density and cover

A total number of 217 woody plants including trees and shrubs were measured using PCQ technique. Mean tree density was 328±3trees/ha (Table 3). The canopy cover for woody plants was very high and to some extent limited grass development (Table 5). The average woody canopy cover was 5287±78 m ²/ha, and ranged from 4781 to 6831 m²/ha (Table 3). Moreover, Acacia tortilis, Acacia nilotica and Commiphora africana had highest canopy cover of 1293±12, 974±19 and 780± 9m²/ha respectively. This implied that the grazing sites are highly encroached with woody vegetation (shrubs, bushes and trees) that was occupying almost 50% of a hectare. Informal discussion with the pastoralists revealed that there was a fire and woody vegetation cutting ban in Ngorika grazing areas that was imposed by government officials in close collaboration with community members in early 2000s. Since then woody plants cover has been increasing drastically although the areas are close to Moshi Municipality where the demand for wood and charcoal is high. This observation concurs with Eldridge et al (2011) findings which elucidated that bush encroachment in rangelands is driven by a multitude of factors including overgrazing, reduced fire frequency and intensity, and reduction of anthropogenic disturbances such as vegetation clearance.

Rangeland condition

The uplands and midlands were found to have an average grass cover of 31.13 and 21% respectively (Table 4 and 5). The low grass cover in the reserved midland grazing sites was unexpected but presence of cattle droppings and grazing signs indicated that there was some grazing taking place. Informal discussion revealed that there was illegal grazing in which cattle from nearby wards were reported to be secretly grazing in the reserved sites while the cattle belonging to pastoralists in Ngorika ward were in the uplands.

The lowland range sites had the least mean grass cover (3.7%) and some sites were found not be having potential for supporting cattle, owing to the presence of large bare soil areas. Such sites fit the description of poor range condition at the time of this study (Table 4). However, the lowland sites were found to have a potential for goat browsing due to the presence of adequate shrubs including Acacia trees which some were still rich of edible pods. The most dominant browse species wereGrewia spp., Acacia spp., Commiphora spp., Salvadora spp. and Balanites spp. All the midland and upland range sites were moderately grazed and still had some wilt-dry grasses and green forbs cover. The most dominant grass species wereAristida spp., Cenchrus spp., Chloris rhoxybagiyana and Eragrostis pyramidalis (endiyamtu). Moreover, both the midland and upland sites had few meters wide soil bare patches that were rarely connected, implying that the site was of fair range condition (Table 4 and 5). Browse resources were ample and all range sites showed good conditions for supporting browsers including small ruminants such as goats, as well as wild browsers as the percent canopy cover was higher throughout the range sites.

When the rangeland condition survey was repeated during the end of wet season in June 2013, it was found that the condition has improved in some sites. Especially in the midlands where percentage grass cover increased but for the upland there was no improvement. Also, the lowland grazing area for Lemkuna village (Lorugum) had negligible grass cover percent but with higher browse percent (Table 5).

Van Wijngaarden (1985) as cited in De Leeuw and Tothill (1990), reported that 10% increase in woody cover was resulting to decrease in perennial herbaceous cover by 7%, whereby perennial herbaceous cover became depleted when woody cover was over 90%. Therefore, the higher woody cover in the Ngorika ward grazing sites might have reduced grass cover.

Livestock carrying capacity

The lowland range sites were found unable to support cattle during the dry season due to very low availability of pasture in particular grasses. The midland and upland areas were suitable for cattle but with low capacity (inadequate grasses) (Table 6 and 7). At Lolosokwan, the forage yield was moderate, approximately 1.26 tonnes of dry matter (DM) per hectare. The observed grazing cattle at Lolosokwan were still in good health and no significant malnutrition (emaciation) signs were noticed by mid-August. However, the longer walking distances (27 km) to the watering points and the low content of crude protein (2.85%) in most of the dominant grass species in particular Aristida spp. (Table 9), implied exhaustion with eventual loss of animal condition. Nonetheless, it was reported that supplementary feeding is uncommon practice which implied decline in milk yield and body weight with advance of drought condition.

The midland range sites were conserved for late dry season grazing and the yield was estimated at 0.85 and 1.0 tonne DM/ha at Elutoto and Engusero sites respectively. Lack of watering points in the upland and midlands grazing areas was another constraint towards optimization of their grazing potential, and thus lowering their carrying capacity. Nevertheless, bush removal should be well planned and should be done selectively whilst leaving trees of higher forage values as browse is important in sub-Saharan rangelands (Bayer 1990).

The total number of cattle in Ngorika was small compared to the land size and vegetation potential in which there were 9,359 cattle (Table 5) although the range land has had potential to support 35,857 (Table 8). The low number of cattle population in Ngorika ward was reported to be caused by the drought of 2009/2010 that was reported to kill over 60% of the cattle in Ngorika ward. This implied that stocking rate per household could be increased if range and water infrastructures are improved; this is deemed important for optimizing the use of the Ngorika ward grazing land.

The lowland areas had the lowest livestock carrying capacity 7 TLU (ha/LU/yr), while both midland and upland sites had 3 TLU (ha/LU/yr. In general, Ngorika grazing areas were found to have livestock carrying capacity of around 3 TLU (ha/LU/yr), see Tables 6 and 7. In generally, this indicates that continues grazing reduces the TLU of semi-arid grazing areas due to grass loss.

Land tenure and informal protection of communal grazing lands

The grazing areas within the village lands were communally owned and with unsecured land tenure. There were no long-term legally binding agreements to protect the village livestock grazing lands conversion to other land uses such as agriculture, wildlife reserves, mining or settlements. However, it was revealed that there are several institutions that deal with the problems of encroachment and livestock keepers who are illegally grazing in reserved sites at a particular time. In the event that land is invaded, the case is reported to respective family elders and if not resolved they are then taken to the maasai pastoralist traditional leaders (Olegwanani) for further mediation. If there are signs of reluctance to the directives of these institutions, the case is reported to the village government for further conventional actions. This finding is in agreement with Maleko and Koipap (2015) who asserted that strong and trustworthy local institutions are vital for proper management of communal pastoral resources.

Conclusions

• The rangeland condition ranged from very poor in areas under continuous grazing to fair in the reserved dry season grazing sites (Olalili ). This implies that continuous grazing is unhealthy to rangelands as it accelerates loss of grass cover and enhances bush encroachment.
  
  
• Livestock carrying capacity in semi-arid areas is low at around 3 TLU (ha/LU/yr)). This implies that for meeting year round animal requirements for pasture a large tract of land is required otherwise risks of overgrazing are high.
  
  
• Semi-arid rangeland soils have potential for carbon storage (climate change mitigation), approximated between 4.8 and 9.6 SOC (tonnes/ha) for the case of Ngorika ward grazing sites.

Acknowledgements

The following institutions and persons are acknowledged: VECO Tanzania for financial support to undertake a study for assessing livestock land carrying capacity in Ngorika Ward through Synergy Pilot Project (SPP). Kayombo Canisius and Protus Mihalu (the late) for identification of plant species. Local communities, village and Ngorika ward leaders, Simanjiro District for guiding us in field whilst providing enormous information. Sokoine University of Agriculture for granting permission to the researchers to undertake this study.

References

AOAC 1990 Official Methods of Analysis. 15th ed. Association of Official Analytical Chemists, Washington, DC.

Bayer W 1990 Use of native browse by Fulani cattle in central Nigeria. Agroforestry Systems, 12, 217-228.

Butt B 2010 Pastoral resource access and utilization: Quantifying the spatial and temporal relationships between livestock mobility, density and biomass availability in southern Kenya. Land Degradation & Devt. 21(6), 520-539.

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.

De Leeuw P N and Tothill J C 1990 The concept of rangeland carrying capacity in sub-Saharan Africa: Myth or reality. London: Overseas Development Institute, Pastoral Development Network.

Eldridge D J, Bowker M A, Maestre F T, Roger E, Reynolds J F and Whitford W G 2011 Impacts of shrub encroachment on ecosystem structure and functioning: towards a global synthesis. Ecology Letters, 14(7), 709-722.

Fernandez-Gimenez M E and Le Febre S 2006 Mobility in pastoral systems: Dynamic flux or downward trend ? The International Journal of Sustainable Development and World Ecology , 13(5), 341-362.

Grains Research and Development Corporation (GRDC) -Undated How to measure and interpret results in relation to soil organic carbon. Pp 55-58. Available at < www.grdc.com.au/Downloads.ashx?q=/~/media/...pdf ‎>. Accessed 28^th May 2017.

Hobbs N T, Galvin K A, Stokes C J, Lackett J M, Ash A J, Boone R B, Robin SR and Thornton P K 2008 Fragmentation of rangelands: implications for humans, animals, and landscapes. Global Environmental Change, 18(4), 776-785.

Lohmann D, Tietjen B, Blaum N, Joubert D F and Jeltsch F 2012 Shifting thresholds and changing degradation patterns: climate change effects on the simulated long‐term response of a semi‐arid savanna to grazing. Journal of Applied Ecology, 49(4), 814-823.

Lund H G 2007 Accounting for the world's rangelands. Rangelands, 29(1), 3-10.

Maleko D D and Koipap M 2015 Opportunities and constraints for overcoming dry season livestock feed shortages in communal semi-arid rangelands of Northern Tanzania: A case of Longido District. Livestock Research for Rural Development 27(4).

Martin R, Müller B, Linstädter A and Frank K 2014 How much climate change can pastoral livelihoods tolerate? Modeling rangeland use and evaluating risk. Global Environmental Change, 24, 183-192.

McGranahan D A and Kirkman K P 2013 Multifunctional rangeland in Southern Africa: Managing for production, conservation, and resilience with fire and grazing. Land, 2(2), 176-193.

Mitchell K 2010 Quantitative analysis by the point-centered quarter method. Available at <http://arxiv.org/abs/1010.3303v2> Accessed 12 July 2012.

Msoffe F U, Kifugo S C, Said M Y, Neselle M O, Van Gardingen P, Reid R S, Ogutu J O, Herero M and De Leeuw J 2011 Drivers and impacts of land-use change in the Maasai Steppe of northern Tanzania: an ecological, social and political analysis. Journal of Land Use Science 6, 261-281.

Munishi, P K T and Shear T H 2004 Carbon storage in Afromontane rain forests of the Eastern Arc mountains of Tanzania: their net contribution to atmospheric carbon. Journal of Tropical Forest Science, 16 (1), 78-93.

Niamir-Fuller M 1999 Managing mobility in African rangelands: The legitimization of transhumance. London: Intermediate Technology Publications.

Oba G, Post E, Syvertsen P O and Stenseth N C 2000 Bush cover and range condition assessments in relation to landscape and grazing in southern Ethiopia. Landscape Ecology, 15(6), 535-546.

Pratt D J and Gwynne M D 1977 Rangeland management and ecology in East Africa. Hodder and Stoughton Ltd. pp 42-43.

Roba H G and Oba G 2008 Integration of herder knowledge and ecological methods for land degradation assessment around sedentary settlements in a sub-humid zone in northern Kenya. The International Journal of Sustainable Development and World Ecology , 15(3), 251-264.

Talbot L M 1986 Demographic factors in resource depletion and environmental degradation in East African rangeland. Population and Development Review, 441-451.

Thornton P K, Burn Silver S B, Boone R B and Galvin K A 2006 Modelling the impacts of group ranch subdivision on agro-pastoral households in Kajiado, Kenya. Agricultural Systems, 87(3), 331-356.

Turner M D, McPeak J G, Gillin K, Kitchell E and Kimambo N 2016 Reconciling flexibility and tenure security for pastoral resources: The geography of transhumance networks in eastern Senegal. Human ecology, 44(2), 199-215.

Received 20 January 2018; Accepted 27 March 2018; Published 1 May 2018

Go to top