Tree leaves in the diet of free-ranging ruminants in three areas of Burkina Faso

H H Hansen, L Sanou* and B M I Nacoulma**

The Royal Veterinary and Agricultural University, Denmark
*The National Forest Seed Center CNSF, Ougadougou, Burkina Faso
**Laboratoire de Biologie et Ecologie Végétales, Unité de Formation et de Recherche en Science de la Vie et de la Terre, Université de Ouagadougou, Burkina Faso
han@life.ku.dk

Abstract

A study was undertaken to document the botanical composition of free ranging cattle and small ruminants in three separate areas of Burkina Faso, as the first step towards improved use and management of trees.  Faecal samples were taken from cattle at three sites and from small ruminants at one site throughout a year and plant particles in the faeces identified in order to determine what species the animals ate.   

Differences in climate and vegetation availability were reflected in the ingested diets. Small ruminants in Kalenga ingested the largest number of all plant species and the largest number of browse species in the three areas. Cattle in Nobéré and small ruminants in Kalenga ingested similar proportions of browse species compared to cattle in Peni and Kalenga that ingested a smaller proportion of browse species. None of the animals in any of the areas ingested an average of more than 25% browse in their diets. Feretia apondanthera and Crossopteryx febrifuga were the most persistently consumed browse species by cattle and small ruminants respectively and may represent an unexplored nutritional potential. The relatively small amount of browse in the diets may be due to palatability factors, vegetation availability, lack of knowledge about nutritional content, or conscious restriction by villagers.   

Keywords: Browse, goats, sheep, cattle, ruminants, botanical composition

Introduction

Trees play a role in animal husbandry as a source of fodder and shade where grazing is only available in short seasons and browsing may be the most efficient way of utilising the natural vegetation (Ewel 1999). In many semi-arid areas the lack of fodder is identified as a major problem, yet recommended exotic species have not been readily adopted. Lack of trees and land degradation is often attributed to grazing animals. Lands are often depicted as being degraded because of grazing beyond the  vegetation “carrying capacity” (Bosch and Theunissen 1992). However, carrying capacity is a concept that can mean at least 4 distinctly different things (Behnke and Abel 1996) with widely varying consequences to production and the environment. Whether the free ranging animals are the primary cause of tree loss is debatable and there exists evidence to the contrary (Oba 1998). Environmental factors or management choices may exert a far greater effect on species composition than grazing (Oba et al 2003).  Nonetheless, knowledge of the browse species ingested and relative quantities can be helpful when determining the use of trees and planning for interventions based on tree use.  

Identification of the botanical composition of free-ranging livestock may be obtained through individual or group interviews or by animal based identification methods (Dove and Farrell 1993), though only one study was found that compared actual and perceived grazing choices (Munthali 2002) Selectivity, when defined as a mathematical comparison of available and eaten vegetation is different then selecting food or feed for survival. That an animal ‘likes’ a given species may be due to many factors, one of which is availability (Forbes and Kyriazakis 1995).  While fodder trees are promoted to increase animal production, fruit trees to increase cash crops and improve human nutrition and other trees for firewood. However, information on the perceived or actual ‘usefulness’ of trees to free grazing ruminants is scarce. This can be done by documenting the actual botanical composition of the diet as determined by plant particle identification in the faeces. 

The botanical composition of the diets of free ranging ruminants in the villages of  Nobéré and Peni and around the Kalenga dam, Burkina Faso was determined. The work was undertaken as a part of the DANIDA funded PETREA (People Trees and Agriculture in Africa) programme. The objective of the PETREA programme is to identify indigenous tree species for improved use and management in animal husbandry. 

Methods 

Nobéré is a village in the southern part of Burkina Faso, close to the border of Ghana and neighbour to a protected forest. The area is extremely dry and most of the fields surrounding the village are continuously cropped. The Kalenga dam is a water catchment that lies approx. 40 km west from Nobéré and the area has been given to Fulani people who are encouraged to become sedentary. Most of the area surrounding the dam is rangeland and only small plots are cropped. A detailed description of the area can be found in Nikiema (2005). The village of Peni lies at the eastern end of the country close to the border with Cote D’Ivoire. The vegetation is lusher than that in either Nobéré or Kalenga with mango and cashew groves as well as cropped and fallow fields. A detailed description of the area can be found in Ræbild et al (2007).

Faecal samples were collected from cattle in 3 areas and from small ruminants in one of the three areas. Cattle faecal samples were collected from the villages of Peni and Nobéré. Five  farmer families volunteered to allow collection of faeces from their holding pens around the family compound in each of these 2 villages.  The same amount of faeces was taken from at least 3 separate fresh faecal piles at each of the 5 farms, dried in a forced air dryer at about 50° C for 24 hours and sent to Denmark for plant particle analyses. Seven sampling periods were defined of either one or two days over a 13 month period in order to include as much botanical variation as possible. In Peni the sampling was repeated on two consecutive days and samples were identified by farmer and date. A composite sample with equal ground dry weight from each of the 5 farms at each date was made and analysed. In Nobéré, samples were collected on a single day from each farm for 4 of the 7 sampling dates. For these samples, two separate samples were constructed from the ground, dry faeces.  The remaining dates were treated as in Peni.  In the area of Kalenga, single day sampling was planned because of the distance of the collection sites from nearest village. The same amount of faeces was sampled from 5 fresh faecal piles of either small ruminants or cattle and the procedure repeated with new fresh faecal heaps for a replicate sample that was then dried and sent to Denmark. Sample dates are shown in table 1.

Samples of 280 plant species were identified by an experienced forest botanist from the National Forest Seed Agency (CNSF) during field collection visits in April and August 2002.  Herbarium samples of unnamed grasses and forbs were collected. Samples were dried, ground and single species reference slides made. This included species not identified by name, but collected and identified by number. There were 4 unnamed non-woody forbs and 5 unnamed grasses.  The reference slides were used when identifying plant particles in the faeces using the microhistology technique (Sparks and Malechek 1968; Holechek and Vavra 1981).  Plant particles in the faeces were identified to species using the reference slides. Plant particles were identified in 20 fields on 5 microscope slides at 100 magnification.  Relative frequency is the number of times a species is identification as a proportion of the total plant identifications for a sample (100 fields; 5 slides) as described by Sparks and Malechek (1968). Species are ranked by ordering the average relative frequency of all identified species in a given area from greatest to least.  

Results and discussion 

Diet botanical composition 

Table 2 and 3 show the woody species ingested, the average relative % of particles found in the faeces, the rank of each species of all species identified and number of samples in which the species was found for cattle and small ruminants. 

A total of 50 species were identified in the faeces of cattle in Nobéré of which 15 are tree or shrub species.

In Peni, 43 species were identified in the faeces of cattle, of which 11 are tree or bush species. Thirty eight (38) species, of which 11 are tree or shrub species were found in the faeces of cattle in Kalenga and 63 species, of which 20 were trees or shrub species were found in the faeces of small ruminants in Kalenga.  This means that between 31 and 33% of the variety of species identified in the faeces from cattle in Nobére and small ruminants in Kalenga were browse or woody species whereas 27-28% of the species identified in the faeces from cattle in Peni and Kalenga were from browse species.  With only four exceptions out of 28, browse species were not in the top ten ranked species eaten by cattle or small ruminants. This may contradict the classification of cattle as bulk roughage grazers and small ruminants as concentrate selectors (Van Soest 1990). 

Seasonality of diet is suggested by the number of times a species was found in the faeces throughout the sampling period. The more samples a species is found in, the more spread out during the year a species is eaten.  Only the species Feretia apondanthera was found in more than 4 samples for cattle in both Peni and Nobéré, suggesting that the other browse species are only eaten during limited periods of time.  Feretia apondanthera and Crossopteryx febrifuga were found in 9 and 6 faeces samples from cattle in Kalenga, suggesting that these species are consumed over longer periods of time.  Vitellaria paradoxa and Crossopteryx febrifuga were found in 6 and 12 samples from small ruminant faeces in Kalenga. The persistent presence of shea nut leaves (Vitellaria paradoxa) in the faeces was not expected as shea nut is a commercial tree and there weren’t many shea nut trees in the vicinity around the dam.  While cattle are carefully herded the small ruminants are left to wander over a greater area and this may be the source of the leaves. 

Relative intake can be directly related to relative particle frequency in the faeces, if differential digestibility doesn’t destroy all identifiable particles (Johnson et al 1983).   Given the expected low quality of diets available to the free ranging ruminants in Burkina Faso, differential digestibility is not expected to skew the results. It is therefore concluded that Feretia apondanthera accounted for between 2 and 3.5% of the cattle diets in Peni, Nobéré and Kalenga.  Feretia apondanthera, while ingested more continuously by the cattle in Kalenga was an average of only 1.4% of the diet of small ruminants and was only found in two samples suggesting that it was not as important for small ruminants as for cattle. It may even be cut and given only to cattle.  Crossopteryx febrifuga was found in 4 and 6 samples in Nobéré and Kalenga suggesting a relative importance of this species as well.

The persistence of Crossopteryx febrifuga in the faeces of small ruminants at 5% suggests an importance either due to availability, nutritional quality or both. Small ruminants consumed more individual browse species than did cattle. Capparis corymbosa, Combretum glutinosum, Cussonia barteri, and Stereospermum kunthianum were all found in only 2-3 samples, but in quantities from 3-9% suggesting the short term importance of theses species as fodder.

With the exception of Acacia sieberiana, individual browse species were less than 5% of the yearly average cattle diet.  Acacia sieberiana was, however, found to be 10% of the identified particles from Nobéré on the 30^th of May 2002 but not in other samples, suggesting that this tree might be an important browse species during a relatively short period of time. Unfortunately, samples were not taken on two consecutive days and so daily differences and/or individual choices may weigh heavily in the results.  Depending on the digestibility of the plant cell wall, and the amount ingested, particles from the leaves could be expected to be found in the faeces from 2-10 days after ingestion and the sampling techniques were not designed to reflect a single meal, but an average of many meals taken from 2-9 days before sampling.  

The proportion of the average frequencies for browse when compared to grass and forbs was greater and similar for cattle in Nobéré and small ruminants in Kalenga than for cattle in Peni and Kalenga (figure 1).  

The large quantities of browse in the faeces of cattle in Nobéré may be explained in the absolute scarcity of available grass or forbs for grazing in this area forcing the animals or herders to find browse in neighbouring forested areas.  Likewise, the fact that only 10% of the cattle diets were browse species in Peni can be explained by the relative abundance of grasses and forbs in the area.  However, whatever the abundance of vegetation, animals are herded to areas deemed suitable by herders each day or browse is cut to feed the animals.  

The species identified in the faeces may reflect grazing ‘choices’, of the animals (Provenza 1996) or the total quantity of available vegetation for grazing.   Selectivity is an objective comparison of eaten with available diet and, as such, both elements can contribute to why little browse was found in the diets. Animals must be moved during the cropping season or occasionally browse is cut and carried to the animals in the compound, which may explain variation not due to availability.  Some of the farmers in this area of Burkina Faso do not practice cutting and carrying fodder to ruminants and may therefore not value browsed fodder trees as highly as farmers who must cut and carry browse to their livestock. Communities with a long tradition of cut and carry fodder management are often very knowledgeable on fodder plants as it seems important to only harvest truly useful fodder (Thapa et al 1997).  The scarcity of browse in the diets of ruminants in Burkina Faso may be due to the relative lack of availability to the animals, lack of knowledge by herders and farmers of browse as a nutritious fodder or a conscious priority choice of villagers by not cutting or allowing the herded animals access to the browse species.  

A 6 week interval between faeces collections was planned and may have been too long to identify species of importance that only were ingested in short periods.  However, the year long average of browse species suggests that browse does not comprise the majority of ingested species and is not consumed in quantities greater than grass and forbs. Whether this is due to selectivity can only be determined after the actual diet is compared with what is available forage for the animals.  Climatic effects may have a far more profound impact on the development of vegetation than does grazing and the scale of effects should be considered (Oba et al 2003).   The animals in Burkina Faso appear not to rely heavily on browse species for the bulk of their intake on a yearly basis. Nonetheless, knowledge of actual animal intake, as addressed in this research, together with availability and nutritional value of fodder plants and thereafter dissemination of such knowledge is useful before a change from free ranging to restriction is considered.  

Conclusions 

• Based on faeces analysis, the intake of browse species by cattle and small ruminants was low (less than 25% browse of their diets).

  
  

• Feretia apondanthera and Crossopteryx febrifuga were the most persistently consumed browse species by cattle and small ruminants respectively and may therefore represent an unexplored potential.

  
  

• The lesser amounts of browse species found in the ruminant diets may be due to palatability factors, relative availability, lack of knowledge of nutritional value of browse or conscious choice by villagers. 

  
  

• These results should be considered and coupled with biomass availability analyses of all fodder vegetation before changes are suggested to limit free ranging livestock or if promotion of trees for fodder is considered.

References 

Behnke R and Abel N 1996 Revisited: The overstocking controversy in semi-arid Africa. World Animal Review 87: 4-27  http://www.fao.org/AG/aGa/agap/FRG/FEEDback/War/W2650T/W2650t02.htm#P17_5939

Bosch O J H and Theunissen J D 1992 Differences in the responses of species on the degradation of gradient in the semi-arid grasslands of southern Africa and the role of ecotypic variation. In: Chapman, G (Editor)., Desertified Grasslands: Their Biology and Management. The Linnean Society of London, London, pp. 95-108

Dove H and Farrell D J 1993 Advances in the estimation of intake and diet selection in the grazing animal. Recent advances in animal nutrition in Australia 1993: 34-44

Ewel J J 1999 Natural systems as models for the design of sustainable systems of land use. Agroforestry Systems 45: 1-21

Forbes J M and Kyriazakis I 1995 Food preferences in farm animals: Why don’t they always choose wisely? Proceedings of the Nutrition Society 54: 429-440

Holechek J L and Vavra M 1981 The effect of slide and frequency observation numbers on the precision of microhistological analysis. Journal of Range Management 34: 337-338

Johnson M K, Wofford H and Pearson H A 1983 Digestion and fragmentation: influence on herbivore diet analysis. Journal of Wildlife Management 47: 877-879 

Munthali D 2002 Grazing behaviour of free ranging oxen in the Sanyati communal areas of Zimbabwe. M.Ph. dissertation, University of Zimbabwe 2002 

Nikiema A  2005 Agroforestry Parkland Species Diversity: Uses and management in semi-arid West Africa (Burkina Faso). Ph.D thesis, Wageningen University, Wageningen, 102 pp http://library.wur.nl/wda/dissertations/dis3706.pdf

Oba G 1998 Effects of excluding goat herbivory on Acacia tortilis woodland around pastoralist settlements in northwest Kenya. Acta Oecologica-International Journal of Ecology 19: 395-404

Oba G, Weladji R B, Lusigi W J and Stenseth N C 2003 Scale-dependent effects of grazing on rangeland degradation in northern Kenya: A test of equilibrium and non-equilibrium hypotheses. Land Degradation and Development 14: 83-94.

Provenza F 1996 A functional explanation for palatability. Rangelands in a sustainable biosphere Salt LakeUtah, USA, 23-28 July 1995. 5th. International Rangeland Congress. Volume 2: 123-125

Raebild A, Hansen H H, Dartell J, Ky J M K and Sanou L 2007 Ethnicity, land use and woody vegetation: a case study from south-western Burkina Faso. Agroforestry Systems 70: 157-167

Sparks D R and Malechek J C 1968 Estimating percentage dry weight in diets using a microscopic technique. Journal of Range Management 21: 264-265

Thapa B, Walker D H and Sinclair F L 1997 Indigenous knowledge of the feeding value of tree fodder. Animal Feed Science and Technology 67: 97-114

Van Soest P J 1990 Comparative digestion among herbivores. Proceedings of a one-day symposium held on May 26, 1989 at the Kungsangen Research Station, Uppsala. Rapport—Institutionen-for-Husdjurens-Utfodring-och-Vard. 1990, No. 197: 86-101; 25

Received 24 October 2007; Accepted 5 February 2008; Published 1 March 2008

Go to top