Screening of potential indigenous browse species in semi-arid central Tanzania. A case of Gairo division

E J Mtengeti and A Mhelela

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

Abstract

The potential indigenous browse species at Gairo Division in central semi arid Tanzania were screened by using local goats, livestock keeper's indigenous knowledge and laboratory analysis of their nutritive value. Screening covered three grazing sites. The sites were; bush lands, two-year fallow lands and seasonal sand riverbanks. Local goats were followed in the grazing sites to see what they browsed. Samples of the browsed shrubs and trees species were collected in gunny bags and then shown to local livestock keepers and requested to give their indigenous knowledge on these browse species and then rank them according to their importance as livestock feed. The top ten ranked browse species were used for acceptability test with local goats so as to compare the ranking of livestock keepers and livestock preference. Crude protein (CP) and in-vitro dry matter digestibility (IVDMD) of the top ten ranked browse species were also determined. A total of 26 browse plants were observed being browsed by goats but local livestock keepers identified only 24 as potential browse species.

Most of the browse plants had several other uses. In general the ranking of the browse species by livestock keepers was close to the preference by local goats. Albizia petersiana had the highest intake rate followed by Grewia bicolor and the least was Lannea tomentosa (13.2, 11.2 and 3.6 g DM/min., respectively). The crude protein content ranged from 12.6 % in Lannea schweinfurthii to 28.1% in Acacia polyacantha. Maerua trychophylla had the highest digestibility followed by Lannea shweinfurthii (64.4 and 61.3 %, respectively).

It was concluded that in order to identify the right indigenous browse plants for conservation; the browsing animals, local livestock keepers' indigenous knowledge and laboratory analysis of their nutritive value must be included in the screening work.

Key words: Indigenous browse species, indigenous knowledge, screening, semi-arid Tanzania

 

Introduction

In Tanzania, about 70% of the national ruminant herd of  15, 9 and 4 million heads of cattle, goats and sheep respectively, are kept in semi-arid areas mainly under agro-pastoral systems (BDAS 2001). The area covers about 80 million ha that is about 40% of the Tanzania landmass (BDAS 2001). The area is therefore, heavily overstocked and signs of overgrazing, that include loss of vegetative cover, gullies and bush encroachment, are rather visible in various parts of the region. The main grazing fields are few isolated bushed grassland in abandoned shift cultivated lands and the seasonal river valleys. The crop fields are also grazed after harvest.

Due to overgrazing the area is mainly covered by short-lived grass species such as Urochloa, Dactoctylenum, Aristida, Chloris and Rhynchelytrum. Perennial grass species such as Hyperhenia, Themeda, Heteropogon, Cynodon and Cenchrus are found in isolated sites. These forages are severely overgrazed and nearly depleted within the first half of the 7 to 8 month of dry season. In the dry season therefore, as in other semi-arid areas in Africa (Le Houérou 1980), browse plants become an important component of the grazing animal diet.

The factors limiting introduction of exotic browse plant species in semi-arid areas include shortage of rain, communal uncontrolled grazing, wild fires and plant diseases. Indigenous browse species have high persistence in their respective areas (Woodward and Coppock 1995). They are, however, in danger of extinction due to over use (eg: browsing by ruminants, charcoal making and medicinal uses), shifting cultivation and other land use and deforestation related activities. Further, not all indigenous browse species are liked by browsing animals (Kuria et al 2005), are free from anti-nutritional factors and easy to establish. Moreover, the local livestock keepers may have their own preferences for browse plant species to preserve, depending on  multiple uses. Alemu et al (1998) found that Afar and Tigrai of Ethiopia prioritize the uses of indigenous shrubs and trees differently. Afars considered livestock fodder to be the best use of shrubs and trees while Tigrians chose construction as their best use. Therefore, any method used to screen potential indigenous browse plant species must take into consideration both the preference of browsing animals and local livestock keepers, as well as laboratory analysis to verify the nutritional quality of the browse species. By using these three aspects, the screening work may be rapid and a short list of potential indigenous multipurpose shrubs and trees to preserve or domesticate can be provided as early as possible to the farming systems in the respective sites.

The objectives of this study were therefore to use the browsing animal and local livestock preferences and thereafter using the laboratory analysis technique to identify and prioritize the potential indigenous browse plant species in different grazing sites of the semi-arid central Tanzania.

Materials and Methods

Study area

The study was conducted at Kilimani, Sanga-njelu, Chigage and Chakwale villages located in Gairo division-Kilosa district. The four villages cover an area of 35 km². The area lies in between 36^o 45' E and 6^o 30' S with an altitude of 1200 m above sea level. It receives rainfall ranging from 400 to 500 mm per year falling between December and April with a short dry spell from mid-January to mid-February. The dry season is normally seven months (ie: between May and November). The rainfall is normally erratic and falling in storms resulting in high run-offs and intensive soil erosion especially in cultivated and overgrazed areas. The area is inhabited mainly by the agro-pastoral Kaguru tribe. The major land use system is subsistence crop farming and keeping indigenous ruminant  livestock breeds comprised of cattle, goats and sheep.

Identification of the browse plant species

Three types of grazing sites in each study village were included in this study. The types of grazing sites were bushed grassland, two years fallow land and seasonal sand riverbank. The shrub or tree, that was observed being browsed by goats, was recorded and a sample (a twig) collected and brought to the local livestock keepers to get more information on the potential of the plant as livestock feed and other uses.

Local livestock keeper's knowledge on the browse plant species

A total of 40 local livestock keepers, 10 from each village (age ≥ 50 years old), were requested to assemble in one site and observe the displayed samples of the shrubs and trees that were noted to be browsed by goats in the different grazing sites. Their knowledge on these shrubs and trees as potential feed for livestock and other uses was recorded. The displayed samples were then pressed and preserved for scientific identification at Sokoine University of Agriculture (SUA) Morogoro.

Assessing browse plant species distribution in different grazing sites

Only plants that were considered as being important as livestock feed, or having other outstanding uses, were considered for assessment of their distribution in different grazing sites. Twenty-five plots, each 10 m x 10 m and 50 m apart, arranged along a transect line, were used to score the frequency of occurrence of the browse plants in each grazing site in each village. The frequency of occurrence of each plant was determined by considering the number of plots in which it was recorded, as a percentage of the total plots in the respective grazing site.

Acceptability test of the browse plant species to the goats

Following the discussion with livestock keepers, ten browse plant species were found to be very important as livestock feed. The ten plants were used for animal acceptability test and analysis of the nutritive value.  Ten one-year-old male goats were used in this test. The goats were obtained from one livestock keeper. Every morning at 7.00 a.m. for seven days before going to graze the ten goats were locked in a kraal made of branches of thorn bushes within the livestock keeper's homestead. In the kraal the goats were allowed to browse twigs (with stems ≤ 5 mm wood diameters) of the ten browse plant species, in a cafeteria fashion for 15 minutes.  The browse twigs were tied in a bundle of 1 kg and hung by a twine rope to a horizontal wooden pole supported at both ends by two poles stuck in the ground. The twigs were hanging at 40 cm above the ground surface so as to give easy reach to the browsing goats. The space between the twig bundles was 1.5 m. The position of the twigs of each browse plant was changed every day so as to remove bias of browse species position. The goats were not allowed to browse the twigs any more after 15 min. They were allowed to go out of the kraal and were led to the grazing field. The fresh weight eaten from each browse species was obtained by subtracting the weight of the remaining twigs from the 1 kg given. Before feeding on each day a fresh sample of the twigs of each browse species weighing about 300 g was collected, air-dried and sent to SUA for determination of the dry matter content,  chemical composition analysis and in-vitro dry matter digestibility.  The crude protein and mineral contents were analysis according to AOAC (1990) procedures. In vitro dry matter digestibility was determined according to Tilley and Terry (1963). The neutral detergent fibre and acid detergent fibre contents were analysed according to Van Soest et al (1991). The reason of extending the test for seven days was to make sure that the animals had enough time to get used to the feed material and thus show their real preference for the browse plant of their choice (Grovum 1988).

Data analysis

The villages were rather similar in terms of the frequency of occurrence of the browse species in each grazing site. The data of the same type of grazing site in each village were therefore pooled together and expressed as a percentage of occurrence in all plots observed in the four villages. The results of intake rate and chemical composition of the browse species were subjected to the analysis of variance according to Steel and Torrie (1980). After the analysis of variance, the means which showed significant differences were compared using the Least Significant Difference.

Results and discussion

A total of 26 browse plant species was identified as being browsed by livestock (Table 1).

The local livestock keepers, however, singled out Catunaregam spenosa and Pteleopsis myrtifolia that livestock do not browse. Such a disagreement between the observation of browsing livestock and livestock keepers' knowledge was somehow perplexing because the same livestock keepers indicated that the leaves of the two plant can be used as vegetables and their fruits are eaten by human. Therefore, the disagreement requires further clarification.

Except in one case, the rest of the plants had several other uses besides being browsed by goats. Beside browsing, 20 plant species were used for fuel wood, 9 for building poles, 8 for medicine, 8 for making herding sticks and tool handles, 5 for shade, 4 for fruits and leaves as vegetable, 2 for making beehives, 1 for providing fuel sticks used as lighter in the household and 1 for an indicator of good soil fertility. Most of these uses may lead to disappearance of some plant species. Therefore, a fast screening method to identify the potential indigenous browse plant species for domestication should be developed. The livestock keepers ranked Lannea shweinfurthii the highest and Dalbergia melanoxylon as the least browsed plant. However, the latter browse species is much liked for wood sculptures due to its hard black wood and it is highly used in the famous Makonde carving in Tanzania (Mbuya et al 1994) and thus requires to be conserved from extinction. The use of Albizia antihelmintica as a medicinal plant for livestock is in agreement with what has been reported elsewhere among the Maasai in Kajiado, Kenya (Farah et al 1996).

Most browse plants had lower frequency of occurrence along the riverbanks than in bush and fallow lands (Table 2).

Only 9 out of the 26 plants had mean frequency of > 50% along the riverbanks. This could be due to plant ecological preference (Alemu et al 1998) and/or being over-browsed by livestock as they go down to the river for water. Lannea tomentosa and Justicia sp. were spotted in all bush plots possibly because they are neither used as fuel wood nor to make tool handles (Table 1). Also Lannea tomentosa was rather unpalatable to the browsing animals (Table 3) while Justicia sp. was very palatable but its runners protected most of the growing points from browsing animals. The differences of occurrence between the fallow and the bush lands for most of the identified browse species were narrow. This shows how resting of land in these areas can rejuvenate the vegetation species which were nearly lost through cultivation. This was the strategy of shift cultivation system that at present cannot be practiced in many areas due to shortage of land because of increased human population.

Unexpectedly, plants with few uses showed lower recovery than those with many uses after two years fallow, possibly due to genetic differences or because the farmers tend to destroy them completely when they open a new farm. Generally, high recovery rates of these shrubs is in agreement with the concept that overgrazing and shift cultivation in dry rangelands deplete the herbaceous under-storey layer and promote bush encroachment which reduce grazing lands carrying capacity (Baumer 1990; Strange 1980). The large ruminants are unable to penetrate dense bushes and cannot subsist well on shrubs and trees as the small ruminants especially goats. One reason of an increasing interest in small ruminants production among agro-pastoralists and settled pastoralists (Legesse 1984; Wienpahl 1985) could be due to the fact that goats with their small pointed mouth can utilize efficiently the bushes and trees; moreover, as they are smaller in size than cattle they are able to pass easily through the dense bushes. The results of an earlier study conducted at the same area are in agreement with this concept because the livestock keepers were found keeping one and a half times more goats than cattle (Kavia 1997).

On average Albizia petersiana had the highest intake rate and Lannea tomentosa the least (Table 3).