Contribution of browse species to the pastoral Value of the southern Tunisian steppes**

C. Floret And E. Le Floc'h

Ecologists at the Centre d'Etudes Phytosociologiques et Ecologiques

Louis Emberger, CNRS, Montpellier, France

1.Background

2. The importance of rangeland and livestock in southern Tunisia

3. The importance of browse vegetation

4. Study methods

5. Results and discussion

References

1.Background

An overall study of the steppe rangelands of Tunisia has been carried out by Le Houérou (1959, 1969): The principal plant types were characterized, using relevé analysis techniques, according to the composition of the flora and their ecological characteristics. This author also provided estimates of standing biomass and of the production for this zone. Estimates of this kind also exist for other North African countries, such as Algeria (Le Houérou et al, 1937), and Morocco (Sébillotte et al, 1972).

However, few measurements of phytomass on the rangeland (and still fewer measurements of production) have been carried out in the arid Mediterranean area. In Algeria, Rodin et al (1970) have studied the "rtemisia" steppes from this point of view (Artemisia herba-alba) and also the Alfa steppes (Stipa tenacissima). In Tunisia, Novikoff (1976) measured the biomass and productivity of a Rhantherium suaveolens steppe. A summary of the state of knowledge on natural plant production in the arid zone was presented by Le Houérou (1973) and Le Houérou and Hoste (1977).

The aim of this paper is to present the result of a series of phytomass measurements carried out on the steppe rangelands, especially those with a basic browse component as in the Djeffara and Basses Plaines Meridionales Orientates regions (the bioclimate is lower arid Mediterranean: 100 < p < 200 mm). The measurements were earned out as part of a multidisciplinary, inter-organizational project launched by the Tunisian Government, the aim of which was to put forward a proposal for integrated agropastoral improvement schemes on an ecological basis. Most of the results have been discussed in a series of publications, (Floret, Le Flo'ch, 1973; Floret et al, 1978; and Pontanier, 1978), and interpreted in order to determine the role of ligneous species in the pastoral value of the southern Tunisian steppes. Introduced browse species will be discussed here (Acacias, Atriplex, Cactus, etc) since they play only a small role in the zone under consideration.

2. The importance of rangeland and livestock in southern Tunisia

The rangelands (i.e., areas occupied by natural vegetation which can provide a feed source for direct consumption by animals, but excluding fallow land still occupy in Tunisia, and in the southern and central areas of that country, approximately 80% of the arable land area, or 45% of the total area, which consists of arable land area, desert areas, forested zones and the alfa grass steppes, i.e. 7 million ha in all. For the most part the areas have an average annual rainfall below 300 mm, and either a Saharan or an arid Mediterranean bioclimate. Although the land ownership status may change, the rangelands are mostly subject to the system known as collective ownership.

For the areas concerned, the herd numbers are evaluated at 3.5 million sheep and 0.6 million goats. Although the latter are less plentiful in the central zone they account for approximately one third of the herd in the south. Rainfall variability may subject the herd to a sharp fall in numbers during drought years, since the animals obtain over 80% of their feed from the rangelands. Owing to the expansion of cropping (cereals and in some areas tree crops), the part played by animal production in the very low income of rural populations varies considerably and is on average estimated at 25%. It is thought that the rangeland area has receded by 30% during the last 30 years.

3. The importance of browse vegetation

Approximately 80% of the plant formations used rangeland grazing (excluding fallows) in central and southern Tunisia are dominated by low ligneous plants, and more particularly by chamaephytes, while nanophanerophytes are frequently present in some residual groups found in the talwegs and on inaccessible escarpments. The proportion varies according to local situations, as shown in Table 1.

Table 1. Extent of crops and rangeland, and degree to which they are dominated by browse plants in two test areas of southern Tunisia (lower arid bioclimate), 1975 results.

	Oglat Merteba 100,000 ha		Zougrata 81,000 ha
Crop %		10	38
Rangeland %		90	62
Proportion of rangeland %	dominated by low ligneous plants	70	87
	dominated by Grasses	30	13

Hence although there are some variations, formations with low ligneous strata, are predominant in the natural vegetation areas of the arid zones, although the forest and "rnatorrals" of the mountain areas have been gradually replaced as a result of degradation, by communities dominated by a grass, Stipatena-cissima. At highly specific localities traces of a woodland savanna with spiny acacias (Acacia raddiana) may also be noted, resembling those found to the south Sahara.

It is clear that perennial species and especially browse species are highly important in terms of offsetting the effects of soil erosion. Moreover, they provide the means for pastoral production on a more regular basis than do annuals during unfavourable rainfall years, allowing a certain standing stock to be maintained. Perennial species also allow the rangeland utilization period to be extended further into the year, since in the case of annuals it does not exceed two to three months. Browse vegetation, aside from its pastoral use, can also be used as fuel by families.

4. Study methods

Measurements were thus carried out to characterize the above ground phytomass of the phyto-ecological map units which had been simultaneously plotted to a scale of 1/50,000. On the map units selected for the phytomass measurement, 30 or so basic plots were outlined at random and then marked out in the field with stake and twine fences. The size of the plots was usually 32 m², but varied with the type and quantity of vegetation.

For every third plot the green weight of each clump of the main perennial species present was estimated. After its weight had been estimated each clump was cut at ground level and weighed green. In the remaining two plots the weight of each clump of the same perennial species was estimated, with satisfactory precision, since on the first plot the observers had staggered the estimates (a method similar to that of Pechanec and Pickford, 1937). The smaller perennial plants and litter were sampled in two subplots of 4 m² each, i.e. 60 smaller plots per map unit. The phytomass of each unit was subsequently expressed, for each species in weight of dry matter per ha (kg DM/ha). Generally the measurements were carried out during a relative rest period of the vegetation (before the annual full growth period).

Approximately 200 sites were thus measured and estimated for the above-ground phytomass of perennial plants. On about 20 of them phytomass measurements for annual plants at the moment of maximum vegetation were also carried out for several consecutive years. In addition, on approximately 10 sites representative of the most widespread types of rangeland in the area, measurements of both annual and perennial vegetation production (without grazing) were made, over periods ranging from three to eight years according to the site.

For each of the main plant species of southern Tunisia a palatability index was established on the basis of the initial proposals made by Le Houérou and Ionesco (1973), adapted according to the personal observations made by the authors¹. For each species the percentage which could be consumed in relation to the weight of the above-ground part of the plant was also evaluated ("proper use factor"), on the basis of bibliographical data (Israel, Soil Conservation Division, Ministry of Agriculture, 1957).

¹It should be noted that a study on the feeding behaviour of animals in this region is currently under way (Waechter, in preparation).

The combination of these two indices, attributed the phytomass of perennial plants on a species-by-species basis, enables a figure for the pastoral value of each site to be obtained by summation: above-ground phytomass of perennial species palatability index percentage consumable. The figure is expressed as a relative value in relation to optimum rangeland, which is considered to have a value of 100. For annual species a further relative value was calculated, taking into account only the production, whether measured or estimated. The programme and calculations were effected by A.M. Bacou at the CNRS Ecothéque Méditerranéenne.

5. Results and discussion

The main results are shown in Table 2. In one and the same situation (substratum and state of cover) the perennial species were listed in order of descending above-ground phytomass. The phytomass values of the perennial species given in this table represent an average for each site calculated from the measurements on 8 or 10 sites, except for the Atriplex halimus group, which was measured only once.

It was difficulty to classify the sites according to the categories of state of plant cover (good, average and poor), for which it is often difficult to define the exact limits. The table must therefore be studied taking into consideration, for example, the fact that the phytomass of the sandy plain steppes ranges on average from 400 to 1125 kg DM/ha, according to the state of plant cover. Salty areas with ground water, physiognomically dominated by Atriplex halimus or Nitraria retusa, together with the bottoms of wadis and other well-watered areas where further species such as Calycotome villosa and Ziziphus lotus are present, have ligneous phytomasses of over 10 tonnes of DM/ha, but are always of limited extent. Situations where the ligneous phytomass is over 1500 kg DM/ha are therefore rare, and steppe vegetation is usually degraded, so that in these areas plants groups usually have a phytomass ranging from 200 to 700 kg DM/ha.

It emerges clearly that on the vegetation units reviewed the contribution of ligneous plants to the perennial phytomass is generally higher than 85%. Only the grass steppes with Stipa tenacissima and Lygeum spartum or Aristida pungens, which were not studied, present a different picture.

Table 2. Phytomass of perennial plants and pastoral value of rangeland

^a Above-ground phytomass of perennials palatibility index % consumable.
^b Expressed as % in relation to optimum rangeland.

On each site, and even in cases where the species present are numerous, it emerges that a very small number of species (one, two or three) account for over 90% of the phytomass. The ligneous species which contribute most to the phytomass in the natural areas studied² (excluding salty and mountain areas) are the following: Rantherium suaveolens, Gymnocarpos decander, Helianthemum lippii var. intricatum, Helianthemum kahiricum, Arthrophytum scoparium, Arthrophytum schittianum, Echiochilon fruticosum.

²In neighbouring drier or wetter areas, and in salty areas, other species should be added to this list. But measurements to obtain phytomass would have been too numerous and could not be included here. The species would include: a) for the Saharan bioclimatic formations: Anthyllis henoniana, Traganum nudatum, Rectama retam, Heliantherum brachypodum, Suaeda mollis, Colligonum sp.

b) for the upper arid bioclimatic zone: Ziziphus lotus, Rosmarinus officinalis.

c) for salty areas: Limoniastrum guyonianum, Salicornia arabica, Anthrocnemun indicum, Salsola sieberi, Salsola tetranda, Suaeda fruticosa, Halocnemum strobilacuem.

As regards relative pastoral value, obviously the lowlands and water distribution areas have the highest value, when they are not under cultivation, in terms of the sum total of figures attributed for perennial and annual plants. Next come the steppes of the sandy areas, while the white artemisia steppes (with Artemisia herba-alba) on slopes with loamy soils have low values. The latter are highly degraded owing to long-term cultivation and overgrazing. These relative values are subject to considerable fluctuation according to the type of year in terms of rainfall. Thus in a humid year the annual vegetation of loamy areas can be more plentiful than in sandy areas.

As shown in Table 2, it should be noted that there is no correlation between the ligneous phytomass present and the pastoral value. This situation is due to the fact that the phytomass largely consists of plants which are more or less unpalatable to animals, or which make poor fuel. However, whatever the plants may be they play an important anti-erosion role. They also have a synergic effect on the germination of annuals and, as has been seen, enable the period of consumption on the rangeland to be lengthened. In addition, most of the ligneous species, even those refused by sheep and goats, are consumed by camels, giving rise to tender shoots which can then be browsed by other animals.

It is therefore appropriate to manage vegetation properly and to think in terms of encouraging or propagating on a local basis some of these species, using techniques which still have to be established. In particular the following might be considered: Artemisia herba-alba, Echiochilon fruticosum, Agryrolobium and Rhantherium suaveolens. This does not rule out the continued planting of forage shrubs of different origins, ash is currently been carried out in southern Tunisia.

References

Israel, Soil Conservation Division, Ministry of Agriculture (1957). Range Survey Plant List.

Floret, Ch., Le Floc'h, E. et al (1973). 'Production sensibilité et évolution de la végetation et du milieu en Tunisie présaharienne. Conséquences pour la planification de 1'aménagement regional de la zone — test d'Oglat Merteba'. Inst. Nat. Rech. Agron. Tunisie et CEPT/CNRS, Montpellier, France. Doc. CEPE n° 71.

Floret, C., Pontanier, R. (1978). 'Relations climat-sol-végétation dans quelques formations végétales naturelles du sud tunisien. (Production - bilan hydrique des sols)'. Inst. Rég. Aride Médenine, Dir. Ress. Eau et Sols Tunis, CEPE/CNRS Montpellier, Paris, ORSTOM.

Floret, C., Le Floc'h, E. et al (1978). 'Modéle écologique regional en vue de la planification et de 1'aménagement agropastoral des régions arides — Application à la région de Zougrata'. Inst. Régions arides Médenine, Dir. Ress. Eau et Sols Tunis, CEPE/CNRS Montpellier, Paris, ORSTOM.

Le Houérou, H.N. (1959). 'Recherces écologiques et floristiques sur la végétation de la Tunisie méridionale'. Mémoire H.S. Inst. Rech. Sah., Alger.

Le Houérou, H.N. (1969). 'La végétation de la Tunisie stepping (avec références au Maroc, à 1'Algérie et a la Libye)' Annales Inst. nat. Rech Agron. Tunisie 42; 5.

Le Houérou, H.N. (1972). 'An assessment of the primary and secondary production of the arid grazing lands ecosystems of North Africa'. In International symposium USSR on eco-physiological foundation of ecosystems productivity in arid-zone, Leningrad.

Le Houérou, H.N., et Hoste, C.H. (1977). 'Rangeland production and annual rainfall relations in the Mediterranean basin and in the African Sahelo-Sudanian Zone'. J. of range management, 30 (3), pp. 181-189.

Ionesco, T., Le Houérou, H.N. (1973). 'Appétabilité des espéces végétales de la Tunisie steppique. Doc. Projet FAO TUN 69:00 et Projet FAO TUN 71/525, 68 p. ronéo.

Novikoff, G. (1976). 'Tuniian Présaharian project'. Progress Report n° 5, US IBP Desert Biome Logan USA.

Pechanec, J.F. and Pickford, G.D. (1973). 'A weight estimate method for the determination of range or pasture production'.

Quezel, P.; Santa, S. (1962). 'Nouvelle flore de 1'Algérie et des régions désertiques mésridionales. Paris, Centre Nat. Rech. Scient.