Livestock Research for Rural Development 35 (10) 2023 LRRD Search LRRD Misssion Guide for preparation of papers LRRD Newsletter

Citation of this paper

Economic and ecological constraints on sheep management in the Algerian steppe

Harkat S1,2, Laoun A3, Belabdi I4, Yabrir B3 and Lafri M1,2

1 Institute of Veterinary Sciences, Saad Dahlab University, Blida, Algeria
2 Laboratory of Biotechnologies related to Animal Reproduction, Saad Dahlab University and Blida, Algeria
3 Faculty of Nature and Life Sciences, Ziane Achour University of Djelfa, Algeria
4 Faculty of Nature and Life Sciences, Hassiba Benbouali University of Chlef, Algeria


The aim of this work is to compare the structure of sheep herds in terms of sex ratio and lamb production in contrasting ecological and economic contexts. An 11-year survey conducted in the wilayas of Djelfa, Msila and Biskra reveals two contrasting flock management strategies adopted by Ouled Djellal sheep breeders. Breeders in Djelfa raise an average of 12 ewes per ram, while those in Msila and Biskra raise an average of 28 ewes per ram. This has led to a higher production of lambs/ewe/year in Djelfa than in the other two wilayas. According to the ecological context, the pasture productivity was higher in Djelfa compared to those of Biskra and Msila because of the higher annual rainfall and pluviothermic coefficient Q3. The Q3 and stocking rates were higher in Djelfa. However, the overall degradation of the steppe was more pronounced in Djelfa than in the other two wilayas. In addition, GLM models have shown the impact of climatic parameters on the survival of lambs and adults. According to the economical context, the higher lamb/ewe production in Djelfa is underpinned by increased marketing facilities. The market facilities encouraged breeders to maximize lamb production per ewe, by increasing the number of rams by flock, giving a higher degradation of the steppe. This study highlights the impact of environmental variables on the management practices of livestock farmers, and points the way for further studies of this type, which are essential as practices will have to adapt to climate change.

Keywords: arid climate, climate change, overgrazing


The sheep population in Algeria is estimated at more than 26 million heads (MADR 2018). Among the eight main sheep breeds identified, the Ouled-Djellal breed predominates with approximately 60% of the sheep livestock. It is particularly located in the steppe (Chellig 1992, Harkat et al 2015, Lafri et al 2016). The farming system is traditionally extensive based on the exploitation of free rangelands with seasonal migrations. Transhumance, a strategy for exploiting rangelands by moving over long distances, takes place from the end of April towards the north (Achaba) to October towards the south (Azzaba). With the recent increase in sedentary human populations and sheep flocks, this system has gradually become semi-extensive. This livestock system, as a result of overgrazing and consecutive droughts, has led to the degradation of steppe rangelands. Feed supplementation based on barley and wheat bran has proven to be a necessity for breeders to preserve the structure of their herds (Bister et al 2005, Elloumi et al 2006). Pasture production represents approximately 5 to 10% of the annual needs of sheep herds (Bneder 2005).

The steppe ecosystem faces strong constraints (Nedjraoui and Bédrani 2008): (i) a gradual decrease in precipitation over the past 40 years, more pronounced in the west than in the east (Hirche et al 2007) and (ii) a growing sheep herd that multiplied more than three times between 1968 and 2006 (Kanoun et al 2007). This animal growth corresponds to that of the human population, which increased eightfold between 1954 and 2008 (Lafri et al 2016). All these factors have dramatic consequences on soil quality. In particular, compaction leads to water runoff over pastures with decreasing levels of organic matter, clay and fine silt, resulting in decreased biomass production (Aidoud et al 1999).

The Ouled Djellal breed extends from the center to the east of the country. Currently, this breed is found between Laghouat and Tébessa and it is the only one in the wilayas of Biskra and Msila according to Chellig (1992). This breed is now expanding due to its ability to walk, its rapid growth and its conformation (Lafri et al 2016). As a result, Ouled Djellal sheep from Djelfa are often crossed with Rembi sheep, whose range is located in the west between Tiaret and Laghouat (Harkat et al 2017). According to a mitochondrial DNA study, the Ouled Djellal breed went through a migration event from the Middle East to North Africa around 1000 BC, while the Rembi breed had populated it about 2000 years ago (Ghernouti et al 2017).

The aim of this work is to compare the structure of sheep herds in terms of sex ratio and lamb production in contrasting ecological and economic contexts. To highlight ecological constraints on sheep herds, we studied the link between climatic factors and pasture productivity, through the distribution of sheep stocking density. We also explored the impact of several climatic parameters on the survival of lambs and adult sheep during critical months. To understand the economic constraint, we compared the sheep trade situation between the different regions. The consequences of herders' strategies was analyzed in terms of sustainable activity and herd sensitivity to climatic constraints.

Materials and methods

Study area

The study area covers three wilayas: Djelfa, Msila (“M'sila" or "M'Sila" refer to Msila wilaya) and Biskra. This area is bordered by the Tellian mountain range on the north and the Hamadas on the south. The northern part of this area is at an altitude of 750 m and extends from the South-West to the North-East in the High Plains; it contains a succession of chotts. In the South, we find the mountain ranges of the Saharan Atlas with an altitude of 1200 m then the foothills with 500 to 800 m of altitude. The Hamadas is a flat area with an altitude of less than 400 m.

Environmental data

Steppe vegetation occupies about 375,000 km² in North Africa, most of which is in Algeria. In this country, according to Djebaili et al (1989), the steppe vegetation is classified as Lygeo-Stipetean and is found in the arid and semi-arid levels.

Two major groups of vegetation are distinguished: (i) Artemisia herba-alba group, covering 3 million hectares and receiving annual rainfall between 100 and 300 mm; (ii) alpha grass Stipa tenacissima group, found on drained soils and covering 4 million hectares with altitudes between 400 and 1800 m AMSL (Above Mean Sea Level: above mean sea level). Other groups are Lygeum spartum on calcareous crusts (2 million hectares) and Arthrophytum scoparium on rocky regs or deserts, with annual rainfall of less than 200 mm.

To assess the impact of climatic parameters on sheep flocks, it was first required to provide a precise distribution of flock densities within each territory. As climate depends mostly on altitudes, we had to detail the more or less heterogeneity of altitudes in the three wilayas. Using Google earth, we draw a physical map including contour elevation curves by 200 m-classes for each wilaya. Then we calculated the percentages of the surfaces corresponding to each class. We took into account the map of sheep stocking densities provided by Amaouche (2010) work, and two densities were considered: between 0.167 and 0.25 (and rarely beyond) heads per ha on one hand, and between 0.167 and 0.125 (and rarely below) on the other. The limits of the areas, corresponding to these two class densities, were reported on our physical map to calculate their surfaces for each altitude class.

Climatic data were provided by ONM (Office National de la Météorologie, in Algeria). It concerned 11 years between 1999 and 2009 for the wilayas of Biskra, Djelfa, and M’Sila (Table 1). In this work, three parameters were considered: m (average of the minima of the coldest month), M (average of the maxima of the hottest month) and P (monthly rainfall, in mm).

Table 1. Climatological data for the 3 stations (range of data between 1999 and 2009). The values are given for means and standard errors.



Altitude, m

Rainfall, mm

m, °C

M, °C



05° 44’ E

34° 48’ N


132 ± 23

6.12 ± 0.32

41.3 ± 0.3






193 ± 22

3.19 ± 0.49

39.4 ± 0.4



03° 20’ E

34° 41’ N


284 ± 21

-0.35 ± 0.48

34.6 ± 0.3


Q3 ; the pluviothermic coefficient of Emberger-Sauvage

The values for each capital city of wilayas were used to estimate the climatic parameters for each altitude class, using the following regression equations (p<0.05):

m= - 0.0057 altitude + 6.227

M = -0.0062 altitude + 41.94

P = monthly rainfall = 0.1362 altitude + 125.7

These three parameters were used to calculate Q3, the pluviothermic coefficient of Emberger-Sauvage (Sauvage 1963) modified by Stewart (1969, 1975) with Q3 = 3.43 P/(M-m) for each altitude class. This parameter is useful to describe the climatological constraints for animals and plants in Mediterranean countries (Petit & Boujenane 2018); it is also known to vary with precipitation and potential physical evapotranspiration above the sea level (Daget 1977). According to the Q3 values, the wilayas of Djelfa and Msila are located in an arid climatic stage, while Biskra is located between the arid and Saharan stages.

The quality of pastures in the different wilayas has been precisely described by Bneder (2005). They provided information on classes of plant productivity per hectare and of sheep stocking densities, indicating that around 40 to 48 % of surfaces are degraded in Biskra and M’Sila, whilst up to 70 % in Djelfa. Conversely, the very good quality pastures occupy less than 2% of surface (Table 2).

Table 2. Variations in pasture quality in the three wilayas

Pasture quality

Phytomass classes
kg dm.ha-1

densities ha-1

Surface %

Surface %

Surface %

Very good


0.5 to 1






0.25 to 0.5






0.167 to 0.25






0.125 to 0.167




Very degradated






From the same source, it was mentioned that the percentages of breeders practicing transhumance are 90, 80 and 65% for Djelfa, M’Sila and Biskra, respectively.

Animal material

The headcounts of the different sheep age groups were provided by the Ministry of Agriculture and Rural Development (MADR 2018): male and female lambs of the year (< 6 months), male and female yearlings, aged of one year, and adults (rams and ewes).

The ewes can lamb once or twice a year, according to rearing conditions. Few breeders synchronize the estrus in order to increase the possibility to get two lambings per year. Artificial insemination is only used in reproduction in the field of scientific research. The litter size averaged around 1.4, but the ewes do not lamb every year. Three lambing seasons were observed, one in February-April corresponding to around the ¾ of lambings, one in May to August with less than 5%, and one in September to November with around ¼ (Boulkaboual and Moulaye 2006, Kanoun et al 2013, Benteboula et al 2014). The lambs born in autumn are heavier and more appreciated for the meat market than those of spring and summer (Kanoun et al 2008).

The age classes chosen for slaughtering are mainly the yearlings and old ewes and rams. However, female yearlings and ewes aged of less than 5 years can be concerned through illicit actions. Numerous animals (yearlings and adults of both sexes) are sacrificed for religious feasts, but the dates change from year to year in the Gregorian calendar.

Statistical analyses

To compare the ratios between ewes over rams, and lambs over adults (rams or ewes), we conducted Kruskal Wallis tests due to non-normality of data distribution. To compare the Q3 coefficient of flocks at both stocking densities for each wilayas, we conducted a series of one-way ANOVA.

To test the impact of climatic parameters on age class sizes, we used a GLM (general linear model) in which individuals of the previous year and monthly values of climatic parameters (P, m, and M) were introduced. As the dataset was limited by the number of years, it was impossible to consider all the months in a same analysis, and we tried numerous combinations of 4 months to select the most significant ones step by step.

As for lambs, the general equation was tested:

Lamb n = (ewes n-1 + ram n-1 + climate (month i) + climate (month i+1) + …. + constant,

with climate (month i) being the value of one climatic parameter on the month i; n designs the considered year.

For the adult females (and the same for rams), the following equation was applied:

Ewe number n = ewes n-1 + female yearling n-1 + climate (month i) + climate (month i+1) + … + constant,

The analyses were conducted using Excel 2013 and PAST v. 2.17 (Hammer et al 2001).


Comparison of sheep headcounts in the three wilayas

The annual variations in the number of sheep (all age groups and sexes included) are shown in figure. 1. The populations’ size were fairly constant, the highest numbers being recorded in the wilaya of Djelfa and the lowest in the wilaya of Biskra, with a slight drop from 1999 to 2001 in Djelfa and Msila.

Figure 1. Annual variations in sheep headcounts in the three wilayas. The blue line corresponds to the region of Biskra,
the orange line corresponds to the region of Msila and the grey one is for the Djelfa region

The sex-ratio within adults showed significantly contrasting values (Kruskal Wallis, p-value < 10-4). Indeed, there were in average 14 ewes per ram in the wilaya of Djelfa, but 28 and 35 in the other two wilayas (Figure 2a). The ratio between lambs and rams was about 6 in Djelfa, which was significantly lower than in the other two wilayas (N=30, Kruskal Wallis, p-value < 0.05), around 10 (Figure 2b). Annual lamb production per ewe was also contrasted (Figure 2c). The ratio was about 0.35 in Biskra and M’Sila wilayas, but significantly higher in Djelfa with a value of 0.45 (N = 30, Kruskal Wallis, p-value < 0.001).

Figure 2. Sheep ratios according to the region of Biskra, Msila and Djelfa; A) ewes/rams ratio; B) lambs/rams ratio and C) lambs/ewes ratio.
Different letters (a or b) above error bars (i.e. ±SE) indicate significant differences (p<0.05) among groups

 We then tested for a relationship between annual lamb production per ewe and adult sex ratio. In the wilaya of Biskra, Figure 3a shows variations of the two ratios during the study period, which were negatively correlated (N=11, Spearman correlation, p-value=0.0016). This negative relation was also verified when collecting data from the three wilayas (figure 3b, N = 33, Spearman correlation, p = 0.017).

Figure 3. Trends of animal ratios from 1999 to 2008. A) Trend of ewes/rams ratio (in orange) and lambs/ewes ratio in grey,
in Biskra region; B) overall trend of ewes/rams (orange line) and lambs/ewes ratio (in grey)

The stocking densities of sheep also varied significantly according to altitude classes and wilayas (Table 3).

Table 3. Surface percentages of sheep flocks according to altitude classes and stocking densities per hectare




Stocking densities







<100 m



100-300 m



300-500 m





500-700 m







700-900 m





900-1100 m





>1100 m


In Djelfa and M’Sila, the highest surface percentages of stocking densities were observed in the 700-900 m altitude class. In Biskra, the higher stocking density was mainly found in the 300-500 m class. From these data, we calculated the mean Q3 coefficient for each stocking density in each wilaya (Figure 4). There were significant differences in Q3 coefficients with lower values in Biskra.

Figure 4. Variations in the mean pluviothermic coefficients (Q3) of sheep flocks in the three wilayas, according to their stocking densities.
Different letters (a, b or c) above error bars (i.e. ±SE) indicate significant differences (N = 600, Tukey tests, p < 10-5).
The orange color shows stock density of 0.167 to 0.125 and the blue color shows stock density of 0.26 to 0.167
Analysis of critical months

The results concerning lamb and adult population sizes for Biskra, Djelfa and M’Sila wilayas are gathered in Supplementary data. Main results were summarized in Figure 5, without distinction of sex. We indicated the sign of the regression coefficients, and a symbol corresponding to top-values significance. There was a shift in the dates of lambings in the three wilayas, because of altitudes, between 1180 m at Djelfa and 82 m at Biskra. Moreover, the GLM analyses indicates that in M’sila, the climatic constraints were weaker than in the two other wilayas for lambs (Figure 5) and adults (Figure 6).

Figure 5. Analysis of critical months on the survival of lambs.

The parameters m, M and P on either side of the axis are indicated with the mention of the positive (+) or negative (-) significant effect on lambs. The dark grey correspond to the February-April lambings, the light grey to May-June lambings, and the medium grey to September to November lambings. The shade of gray represents lambing intensity: dark gray means high intensity lambing, medium gray means medium intensity lambing and light gray means low intensity lambing. Details of the GLM models are indicated in the supplementary data.

Figure 6. Analysis of critical months on the survival of adults (below the line)

As for lambs, the numbers recorded in Biskra was negatively related to the ewes of the previous year for the three climatic parameters m, M and P (Supplementary data, Table A1). A comparable situation was observed in Msila for most cases (Supplementary data, Table A3). On the other hand, the number of Djelfa lambs depended positively on the size of the ewe populations of the previous year (Supplementary data, Table A5). Lamb numbers were positively related to rams of the previous year only in Biskra for the climatic parameter P (Supplementary data, Table A1).

The lambs were sensitive to monthly climatic parameters. They were favored by June rainfalls in Biskra and Djelfa wilayas, and also in September in Biskra, but were significantly affected by the July rains in the same region. The minima means had also consequences on their survival. Higher values of February and June had positive effects in Biskra, of March in M’sila, and of December in Djelfa. However, it was the opposite for the months of March in Biskra, January and February in Djelfa. In turn, the maxima means had negative effects during July and October in Biskra, but very positive in June. In M’sila, M had a positive effect in March. In Djelfa, the consequences of this parameter had negative consequences from March to October.

In the three wilayas, the numbers of ewes were generally positively correlated to the ewes of the previous year. Likewise, the ram numbers of Biskra and M’sila were often positively linked to the rams of the previous year, but not at all in Djelfa.

The adults showed different responses from their progenies as for climatic influences. In Biskra, May and July rainfalls were benefic for both sexes. In M’sila, the influence of this factor was negative in February and July. In Djelfa, there was a positive effect from February to June, but in this wilaya, as well as in M’sila, the rainfalls in July negatively impacted the survival of rams. Higher minima means had only an effect, a negative one, in Biskra, in September and November. Maxima means decreased adult survival in June in Biskra, March and November in M’sila, and July, October and December in Djelfa. The only positive effect of M was recorded in January in M’sila.


By considering the Algerian ovine stock of the steppe, we could show that the variations observed in terms of flock structure, lamb production rate and probability of survival, at different age classes, were largely impacted by human management and climatic variations. The case of yearlings will not be discussed because they constitute the major part of individuals that are sold or slaughtered.

Sheep population structure and farmers' strategy

Two opposite situations were observed when considering the herds of Djelfa on the one hand and those of Biskra and Msila on the other. In the first one, the ewe/ram ratio was around 14, i.e. around two times less than in M’Sila and Biskra. A clear link between ewe/ram ratios and lamb/ewe ratios was found. This finding was supported by the GLM model, for the wilayas of Biskra and Msila, where the production of lambs was negatively correlated with the number of ewes and positively with rams. This means that the number of rams represented the limiting factor. In contrast, in Djelfa the number of lambs depended largely on the number of ewes of the previous year, and not on the ram population. It means that the limiting factor was the ewes, and not the rams. As a result, the production of lambs per ram was around 5 in Djelfa and 9 and 11 in Biskra and M’sila, respectively. Among the hypotheses that could explain the effect of the number of adult males on the fertility of the ewes, the most probable is the presence of several males, which could promote the onset of estrus in the ewes. According Benyounes et al (2015), the low fertility of Ouled Djellal ewes is partly due to an insufficient body condition score, non-optimal pasture production and the number of rams involved in breeding.

We have to understand why the sheep holders from Djelfa have adopted a different strategy than those of Biskra and M’sila. Two considerations can be developped: ecological and economic considerations.

For the ecological viewpoint, two points can be considered to explain the differences in lamb production per ewe: (i) the variability of pasture productivity and (ii) the heterogeneity intra-breed (i.e. phenotypic variations within the Ouled Djellal breed). In Djelfa, the breeders can afford to produce numerous lambs as their pasture productivity is higher than in the other wilayas. We found that annual rainfall and hence the pluviothermic coefficient Q3 corresponding to the higher stocking densities was obviously higher than in the two other wilayas. This difference was even more pronounced with Biskra, situated in the limit between arid and Saharian stages. As the productivity of pasture follows the level of Q3 coefficient, it means that the available amount of animal food was greater in Djelfa than in M’sila and of course than in Biskra. However, the data provided by Bneder (2005) indicated that around 70% of pastures of Djelfa were degraded, much more than in Biskra and M’sila wilayas (48 and 42 % respectively). Regarding the heterogeneity of the Ouled Djellal breed, we could expect that the animals of Djelfa are less heavy than in the other wilayas, hence a lower need for food necessary to obtain a satisfactory physical condition. The study of Harkat et al (2015) showed, indeed that animals from Biskra were significantly heavier (70.5 kg) than those from Msila and Djelfa (59.5 and 58.5 kg respectively), the latter not being different from each other.

For the economic viewpoint, the wilaya of Djelfa is crossed by the RN1: one of the main roads of the country which connects Algiers, located in the north, to Tamanrasset, located in the south. Moreover, the second more important small ruminant center in Algeria is settled in Djelfa with more than 1 million sheep heads marketed annually (Kanoun et al 2016). As the breeders have a great confidence to be able to sell their production, we propose that they have adopted the solution of keeping more rams in their flocks to increase the fertility of the ewes. With a sex-ratio of 30 ewes per ram, the fertility of ewes was higher than 90%. Moreover, it was reported that sometimes the high number of rams in the flock might decrease the fertility of ewes because of the concurrence between mating rams. The dramatic consequence of this non-sustainable policy is a severe degradation of pasture quality, resulting in bare ground with poor water retention capacity, soil compaction and loss of plant productivity (Nedjraoui and Bédrani 2008).

Effects of monthly climatic variations on sheep categories

As mentioned in the material and methods section, the majority of breeders practice transhumance from April to October to reach fallow land and mown crops in the northern wilayas. As a result, the influence of climatic parameters of each wilaya was limited in this period. However, this influence could be not negligible, because (i) a part of flocks stays in the wilaya, and (ii) climatic parameters could vary in parallel, in transhumance target wilayas. Thus, the results concerning the transhumance period should be taken with caution.

The rainfalls have positive effects on the growth of plants consumed by adults and thus their health but at a lesser extent on “old” lambs. It was consitent to observe a positive effect of May and July precipitations on adults in Biskra, of February, May and June in Djelfa. As productivity was weak in Biskra, we found a benefic effect of the rainfalls in June and September for the lambs. Presumably, this is an indirect influence on their mothers, and perhaps a more direct influence on older lambs that begin to graze. However, it is more difficult to explain the negative influence of July rainfalls on lambs in Biskra and on ewes in Djelfa. We could evoke the development of intestinal diseases or endoparasites. According to Boulkaboual and Moulaye (2006), the Ouled Djellal breed was subject to several classes of internal parasites, of which the Strongylida Nematodes showed a peak in November and at a lesser extent in March. Interestingly, these authors pointed out Coccidia that occurred mainly from July to September in lambs and ewes. These parasites are favored by temperature and humidity and lead to diarrheic syndrome. This last category could be a candidate for July over-mortality. Another survey dealing with several breeds across Algeria confirms the monthly fluctuations of internal parasites (Triki-Yamani and Bachir-Pacha 2010). These authors also described the seasonal variations of ectoparasites which showed a peak in summer. It would be interesting to test their link with the rains of July.

It could be expected that high temperatures of winter, expressed by the minima means m, would have a benefic effect on sheep survival. It was verified for the lambs of Biskra in February and of Djelfa in December, but not for adults in both wilayas. However, higher values of this parameter had negative effects on the lambs in march and on the adults in september in Biskra, and on the lambs of Djelfa in January and February. It is difficult to interpret these observations, but the role of Strongylida could be invoked as the periods match. Other investigations should be necessary to test this hypothesize. Likewise, it was expected that the highest temperature values, expressed by M, should have a negative impact during summer. It was verified in the case of lambs in Biskra in July, in spite of marginal significance of statistics of adults in June of the same wilaya, and of ewes in July at Djelfa. However again, hot temperatures of October in Biskra were harmful for the lambs but we don’t know if it concerns the new born ones (more fragile) or those born in spring (more resistant). The same observation was made for the male lambs of Djelfa in September and October. We have no explanation about the sensitivity of ewes of Djelfa for the months of October and December.


In conclusion, it seems that economic constraints had the major influence on the two opposite strategies of holders. In Djelfa wilaya, the strong demand for sheep yearlings encouraged breeders to maximize lamb production per ewe, by increasing the number of rams by flock. This policy lead to a dramatic degradation of pastures. On the other hand, in the two other wilayas, the system was more “classic”, with few rams by flock, resulting in lower lamb production per ewe, and less degraded pastures. We suspect that the high degradation of pastures in Djelfa had consequences on the strong sensitivity of flocks to climatic parameters, very contrasted with M’sila wilaya, in spite of comparable climatic conditions. However, the overall quality of pastures was not satisfactory, as a result of an overgrazing non-adapted to low productivities. There is an urgent need to reduce the pressure of grazing in pastures, and to frame grazing on the steppe.


We thank Anne Da Silva and Daniel Petit for their contribution to the work.


Aidoud A, Slimani, H., Aidoud-Lounis, F et Touffet, J 1999 Changements édaphiques le long d’un gradient d’intensité de pâturage dans une steppe d’Algérie. Ecologia Mediterranea, 25(2), 163‑171.

Amaouche I 2010 Les potentialités agropastorales de la steppe algérienne. Requêtes cartographiques, analyse et interprétation de l’information géographique. Haut-Commissariat au Développement de la Steppe (HCDS), Djelfa et Bureau National d’Etudes pour le Développement Rural, BNEDER, Djelfa, Algérie : HCDS et BNEDER, 80 p.

Benyounes A, Fakhet S et Lamrani F 2015 Réponse des brebis Ouled Djellal à l’effet mâle après isolement physique simple des béliers. Nat. Technol., 36, 37‑44.

Bister J-L, Bouchaib Band Paquay R 2005 L’élevage du mouton et ses systèmes de production au Maroc. Institut National de la Recherche Agronomique.

Bneder 2005 Potentiel de l’espace steppique et agropastoral (Biskra, M’sila et Djelfa) : Rapport provisoire. [Rapport provisoire]. Djelfa, Algérie : HCDS et BNEDER (Bureau National d’Etudes pour le Développement Rural), 80 p.

Boulkaboul A et Moulaye K 2006 Parasitisme interne du mouton de race Ouled Djellal en zone semi-aride d’Algérie.

Chellig R 1992 Les races ovines algériennes. O.P.U. Alger, 80p.

Daget P 1977 Le bioclimat méditerranéen : Analyse des formes climatiques par le système d’Emberger. Vegetatio, 34(2), 87‑103.

Djebaili S, Djellouli Y et Daget P 1989 Les steppes pâturées des Hauts Plateaux algériens. Fourrages, 120 : 393-400.

Elloumi M, Alary V and Selmi S 2006 Politiques et stratégies des éleveurs dans le gouvernorat de Sidi Bouzid (Tunisie centrale). Afrique contemporaine, n° 219(3), 63‑79.

Ghernouti N, Bodinier M, Ranebi D, Maftah A, Petit D and Gaouar, S B S 2017 Control Region of mtDNA identifies three migration events of sheep breeds in Algeria. Small Ruminant Research, 155, 66‑71.

Hammer D, Harper P and Ryan P 2001 Past: Paleontological statistics software package for education and data analysis. Electr.4 (9).

Harkat S, Laoun A, Belabdi I, Benali R, Outayeb D, Payet-Duprat N, Blanquet V, Lafri M and Da Silva A 2017 Assessing patterns of genetic admixture between sheep breeds : Case study in Algeria. Ecology and Evolution, 7(16), 6404‑6412.

Harkat S, Laoun A, Benali R, Outayeb D, Ferrouk M, Maftah A A, Da Silva A-L and Lafri M 2015 Phenotypic characterization of the major sheep breed in Algeria. Revue de Médecine Vétérinaire, 166(5‑6), 138‑147.

Hirche A, Boughani A et Salamani M 2007 Évolution de la pluviosité annuelle dans quelques stations arides algériennes. Science et changements planétaires / Sécheresse, 18(4), 314‑320.

Kanoun A, Kanoun M, Yakhlef H et Cherfaoui M A 2007 Pastoralisme en Algérie : Systèmes d’élevage et stratégies d’adaptation des (...)—3R - Rencontres autour des Recherches sur les Ruminants. Rencontres Recherches Ruminants (3R).

Kanoun M, Huguenin J et Kanoun Meguellati A 2016 Savoir-faire des agropasteurs ovins de Djelfa (Algérie) en milieux steppiques en matière d’engraissement des produits d’élevages ovins [Conference_item]. The value chains of Mediterranean sheep and goat products. Organisation of the industry, marketing strategies, feeding and production systems; CIHEAM.

Kanoun M, Huguenin J, Kanoun Meguellati A et Zaki B 2013 Facultés d’adaptation des agropasteurs à un contexte d’incertitudes dans la région steppique d’El-Guedid-Djelfa en Algérie [Conference_item]. 20èmes Rencontres autour des recherches sur les ruminants, Paris, France, 4 et 5 décembre 2013; s.n.

Kanoun M, Kanoun A, Baira M and Ziki B 2008 Les produits de l’élevage ovins steppiques : cas des agneaux de Djelfa. Options Méditerranéennes. Série A : Séminaires Méditerranéens. Seminar of the Scientific-Professional Network on Mediterranean Livestock Farming, 2, Zaragoza (Spain), 18-20 May 2006.

Lafri M, Harkat S, Ferrouk M, Brouri L et Da Silava A 2016 La diversité génétique des races ovines algériennes : Etat des lieux et perspectives. Options Méditerranéennes. Séries A: Mediterranean Seminars.

MADR 2018 Statistiques Agricoles. Ministère de l’Agriculture et du Développement Rural, Alger.

Nedjraoui D et Bédrani S 2008 La désertification dans les steppes algériennes : Causes, impacts et actions de lutte.

Petit D et Boujenane I 2018 Importance of determining the climatic domains of sheep breeds. Animal: An International Journal of Animal Bioscience, 12(7), 1501‑1507.

Sauvage C 1963 Étages bioclimatiques. Atlas du Maroc. Notices, applications, physiol. Météo. 1963 ; 6 : 31 p.

Stewart P 1975 In Bulletin de la Société d’histoire naturelle de l’Afrique du Nord [...] Tome 15ème (p. 239‑345).

Triki-Yamani R et Bachir-Pacha M 2010 Cinétique mensuelle du parasitisme ovin en Algérie : Résultats de trois années d’enquêtes sur le terrain (2004-2006). Revue Méditerranéenne Vétérinaire, 161(4), 193‑200.