General characterization and organization of the surveyed farms
The farms surveyed are distributed across the different agroecological zones of the Setif region. 28% of the farms are located in the north and the same percentage in the south; however, 44% of the farms studied are located in the center. The cattle population on the farms studied ranges from 11 to 93 head, with an average of 34±21 head per farm. This average is higher than the average (25 head) obtained by
Bir et al., (2015) and
Guedjal et al., (2023) in the same region, who recorded average herd sizes of 25 and 7.16 head per farm, respectively, as well as those obtained by obtained by
Eulmi et al., (2023), in the Souk-Ahras and Biskra region (≤15 head) and
Cherfaoui-Yami (2021), in the Tizi-Ouzou region (17.72 head). Dairy cows represent 48.43% of cattle herds.
The Montbeliarde and Holstein breeds are found on most farms. 50% of the farms studied have both breeds and 44.44% have only the Montbeliarde breed. However, a low proportion of 5.56% is observed on farms that have the local breed.
Technical supervision is only available on pilot farms, compared to private farms.
Eating behavior
On the pilot farms, during the stabling phase (winter and fall), when grazing is very rare, dairy cows are fed a diet based on hay and silage, with supplemental concentrate, which is distributed throughout the year in varying quantities depending on the cows’ physiological condition. However, during the spring and summer, the cows are fed straw and grazed on natural meadows and on cereal stubble and fallow land.
By examining the forage systems of various private farms, three types of forage calendars can be identified, varying in their diversity depending on feed availability. Concentrate supplementation is provided throughout the year on all farms, with quantities adjusted according to the cows’ physiological condition.
• The first type (14.29%): In winter and autumn (stall period), the cows are fed exclusively on hay and concentrate, without grazing. In spring, they graze on natural meadows, cereal stubble and fallow land.
• The second type (14.29%): This type has a varied diet depending on the season. In winter and autumn, the cows consume mainly oat hay and straw in the stable, without grazing. In spring, they graze on natural meadows and in summer on stubble, while receiving a little alfalfa in the stable.
• The third type (71.42%): In spring, the cows graze on natural meadows. In winter and autumn, they consume hay (meadow and oats). In summer, their diet is based on straw with supplemental grazing.
For watering, the quantity of water drunk varies according to the season from one farm to another from 20 to 70 liters/day in winter and autumn, from 50 to 100 liters/day in summer and spring.
Breeding management
The criteria for female breeding are weight and growth in 22.22% of farms and age and the onset of heat in 44.45% and 11.11% of units, respectively. Heat detection is performed by daily observation on all farms.
Artificial insemination is practiced on 5.56% of farms, while natural breeding is practiced by 77.78% of farms. However, there are farms that apply both methods, at a rate of 16.66%. These results are consistent with those reported by
Benidir et al., (2017), in the same region and
Meskini et al., (2023), in the Mostaganem region, where heat detection was practiced occasionally by breeders and natural mating was the main insemination method, with a rate of 63% (80.33% for the first region and 63% for the latter). Similarly,
Cherfaoui-Yami (2021), in the Tizi-Ouzou region, recorded that 36% of farms use artificial insemination and 57% of farms practice natural mating and only 7% of farms practice both methods of reproduction.
Milking
Milk is extracted twice a day (morning and evening) using a milking machine on 38.89% of farms, with an average production of 15.5 liters/cow/day. Milking is done manually on 61.11% of farms, with an average production of 13.78 liters/cow/day.
Milk production on the study farms
The results recorded in Fig 3, show the variability in average milk production (10.00 to 17.81 liters/cow/day) across different farms. This difference in production is largely related to feeding techniques and the presence of technical supervision.
The average daily milk production per cow in the farms surveyed is 14.34±3.63 l/cow/day, this average is lower than that obtained by
Bir et al., (2015), in the region of Setif (15.01±3.70 l/cow/day) and that obtained by
Cherfaoui-Yami (2021), in the region of Tizi-Ouzou (17.4 l/cow/day),
Kechroud et al., (2024), in the region of Souk-Ahras (16.1 kg/Cow/day) and
Meskini et al., (2023), in the region of Mostaganem (18.19 l/cow/day).
Eulmi et al., (2023), in the region of Souk-Ahras and Biskra (15 l/cow/day). This quantity of milk produced is slightly higher than that obtained (14.2±4.73 kg of milk/cow/day) by
Boukhechem et al., (2019), in different farms in northern Algeria.
Factors affecting variation in milk production
Variation in milk production is influenced by several factors.
Technical support
The presence of technical expertise on farms plays a very important role in the success of cattle management (feeding, watering, hygiene and health). This factor explains why dairy cows on pilot farms are the highest milk producers, with an average of 17.01±0.57 l/cow/day, higher than that obtained on private farms (13.58±1.8 l/cow/day).
The effect of feeding
The basic ration
The nature of the basic ration (preserved fodder (dry, silage), green fodder) determines the variation in production between pilot farms and private farms and even between the latter.
The distribution of the ration is linked to the season of the year, the region and the cultivated areas used; the latter are on average different (1708±611.10) ha and (67.75± 103.57) ha on pilot and private farms, respectively. According to
Ghozlane et al., (2010), large farms produce and deliver the largest quantities of milk.
At the pilot farm level, feeding cows silage allows them to produce 3.43 l/cow/day more milk than cows on private farms.
Brocard et al. (2019), reported that early grass silage increased milk production by +1.9 kg of milk/cow/day. Similarly,
Coulon et al., (1995), reported that cows fed grass silage produced 0.4 and 1.9 kg/d more milk, respectively, than those fed corn silage or hay. According to
Cuvelier and Dufrasne (2005), when grass is available in sufficient quality and quantity, grazing can indeed allow an average production of 20 kg of milk/day (around 25 kg in spring and 15 kg in the late season). Silage provides a higher energy content than hay, making it ideal for increasing milk yields. Cows fed silage generally produce more milk due to its higher digestibility and nutrient availability. Hay, while beneficial for maintaining fiber intake, may not support peak milk production as efficiently as silage (
Kimd Group of Company, 2025).
The effect of concentrate feeding
All farmers use supplemental feed. It varies depending on the season and market prices. Concentrate supplementation can be simple (wheat bran) or composed of a mixture of wheat bran plus barley, corn, or soybean meal, phosphate, mineral salts and vitamins.
Table 1 shows the significant effect (P<0.01) of concentrate feed on milk production. The amount of concentrate distributed varies from one farm to another. In class one, a significant amount of milk is produced compared to other classes due to the consumption of a good amount of concentrate feed, averaging 10.5±1.08 kg/cow/day, with a milk production of 17.01±0.57 kg/cow/day and good quality (barley, bran, CMV).
These results are close to those obtained by
Cherfaoui-Yami (2021), in the Tizi-Ouzou region (17.4 l/cow/day) with an average concentrate intake of 9.98±2.32 l/cow/day. and higher than those reported by
Eulmi et al., (2023), in the Souk-Ahras and Biskra region, the average production is 15 l/cow/day with an average concentrate intake of 8 Kg/cow/day and lower than those obtained by
Meskini et al., (2023), in the Mostaganem region, an average concentrate intake of 9.85±0.32 Kg/cow/day and a milk production of 18.19±0.45 l/cow/day.
According to
Tranvoiz et al., (2018), the addition of concentrate leads to a rapid response on milk production. Furthermore,
Vergonjeanne (2014) indicates that increasing the concentrate can produce up to 15% more milk with a maximum of 4 kg of milk/cow/day depending on the initial feeding situation and provided that the risks of acidosis and metabolism are controlled.
The effect of season
Cow milk production varies according to the seasons. Table 2, illustrates the significant effect of season on milk production (p<0.001).
The highest milk production (18±1.71 l/cow/day) is recorded in spring, this increase being due to feed availability. Furthermore, the lowest milk production (10.92±3.34 l/cow/day) is recorded in winter. These results indicate that the highest production coincides with the period of green fodder availability and they are lower than those obtained by
Habibi et al. (2021), with an average of 18.45±33 and 17.06±34.5 l/cow/day in the spring and summer seasons, respectively. While the lowest milk production obtained in winter 16.39±31.5 L.
According to
Legarto et al., (2014), milk production is minimal from July to December and maximal from March to May, the difference between these two periods being 1.3 kg/day in Holstein, 1.7 kg/day in Normande and 2.8 kg/day in Montbeliarde. For their part,
Hachana and Bousselmi (2018), report that milk controls carried out in winter showed a difference in average milk production compared to controls carried out during the autumn season despite the quantities of milk produced being very similar. Thus, summer and spring controls showed comparable average milk productions of around 23.95 kg/day and 23.96 kg/day respectively. Furthermore,
Lim et al., (2021), indicate that milk production was significantly reduced (Holstein 29.02 kg/day and Jersey 19.75 kg/day) in autumn compared to other seasons (Holstein 30.14 kg/day and Jersey 20.96 kg/day).
Azimi et al. (2022), in their study recorded that the seasonal variation and milk performance of the three breeds (Holistan, Garai and Crossbreeds) were found to be very significant, as the production in other seasons was low. The average milk production was 2 470±37 liters in spring, followed by 2427±41 liters in summer and a minimum in winter, respectively 2 403±82 liters.
Breed effect
Milk production also varies by breed. According to our results (Table 3), the average daily milk production of Holstein cows is 14.62±2.06 l/cow/day, higher than that of Montbéliarde cows (13.02±1.96 L/cow/day), a difference of 1.6 kg/cow/day, representing 10.94% of production. Least squares analysis showed a significant effect of genetic group on milk production (p<0.05). Pie Noire cows had higher adjusted averages than Pie Rouge cows, confirming the superior milk production of the Pie Noire breed under the conditions studied.
Ce résultat est supérieur à celui obtenu par
Belhadia and Yakhlef (2013), who recorded a production of 11.37 liters/cow/day for the Montbéliarde and 11.90 liters/cow/day for the Holstein Friesian in the semi-arid region of Chlef and that indicated by
Boujenane and Aïssa (2008), qui ont constaté une différence de 9,98 % dans la production laitière entre les deux races. Par ailleurs, notre résultat est inférieur à ceux de
Balandraud et al., (2018), qui ont observé que les vaches Montbéliardes produisent environ 12 % de lait en moins que les vaches Holstein, avec des taux de matières grasses et de protéines plus élevés.
The observed milk yields remain below the genetic potential of the exploited breeds, mainly Montbéliarde and Holstein. This situation can be explained by the fact that the studied cattle are the result of several generations raised locally, although their ancestors were originally imported several decades ago. Despite a certain adaptation to local conditions, their milk performance remains lower than that of recently imported breeds, notably due to constraints related to local farming systems and to feeding that is often insufficient or inadequate with regard to their nutritional requirements.
Regional effect
The results presented in Table 4 show that milk production in the northern region (15.26±2.25 l/cow/day) is higher than in the southern region (14.60 ± 1.52 l/cow/day).
With a difference (4.32%) that is not significant at the 5% probability level. This variation can be explained by the effect of climate and forage availability (plant cover) in both regions.
The northern climate is characterized by harsh winters, which can begin as early as autumn and moderately warm summers. The cold stimulates the appetite of dairy cows, leading to increased feed consumption and, consequently, higher milk production. In the south, however, high temperatures and drought lead to low feed consumption, which results in lower production. According to
Charron (1988), an animal exposed to cold adjusts its heat resistance by consuming more available feed. Otherwise, it uses nutrients to the detriment of milk production, depleting its body reserves. As a result, milk production decreases with decreasing temperature while fat and protein levels increase.
Eulmi et al., (2023), in a study conducted in two distinct bioclimatic zones “Souk-Ahras (semi-arid region) and Biskra (arid region) “found that there was no significant difference (p>0.05) between the two regions in terms of milk production. However, a significant difference was noted in terms of lactation duration, favoring farms located in semi-arid zones.
For his part,
Dawod (2022) showed in his study that a high temperature-humidity index, combined with low rainfall, is associated with a decline in dairy performance, including a decrease in daily milk production and a reduction in protein and fat contents.
Furthermore, a 28% reduction in average monthly precipitation could lead to a decrease in milk production estimated at between 1 000 and 12 000 liters per year. In addition, a 4°C increase in average monthly temperature could also cause a decrease in production, estimated at between 8 000 and 12 000 liters per year. In contrast, a 4°C increase in maximum temperature could lead to a much more marked drop, reaching up to 14 000 liters per month
(Somoza et al., 2018).