Livestock Research for Rural Development 20 (2) 2008 Guide for preparation of papers LRRD News

Citation of this paper

Practices consistency in dairy farms of the dry Andes: analysis in a Peruvian peasants community

C Aubron and G Brunschwig*

Montpellier SupAgro, 2 Place Pierre Viala, 34060 Montpellier Cedex 1, France
*Ecole Nationale d’Ingénieurs des Travaux Agricoles Clermont-Ferrand (ENITACF), BP 35, 63370 Lempdes, France


The existing Andean dairy production patterns are scarcely known and yet they play a major role in the household food security of many families. This study examines the consistency of forage management through the monitoring of 8 dairy farmers of a community living in the harsh conditions of the dry Peruvian Andes and the analysis of their production practices. The Sinto farmers are able to produce milk throughout the year despite the very dry climatic conditions by adjusting cattle groups, grazing rotationaly on irrigated alfalfa fields, farming large rangelands zones and exchanging standing forage with other farmers.


Although the milk production per cow is low, the milk yield per hectare of alfalfa field is relatively high. This can be interpreted both as an optimisation of the system with regards to the restrictive factor which is the irrigated land and as a risk management, two components that may be considered to evaluate the sustainability of these livestock farming systems and to propose development options.

Key words: Andes, dairy farming, dairy production, forage management, forage system, irrigated alfalfa, livestock farming system, practices, Peru, risks


Dairy farming in the Andes has greatly increased in the past 15 years: raw milk and milk processed into cottage cheese provide the farmers with a regular income which is a key element to household food security (Bernet et al 2001a; Aubron 2006a). Furthermore, the national market for dairy products has been expanding in the last decades: on the one hand, the price of milk paid to the farmers has decreased less than that of other agricultural products and on the other hand, consumption of dairy products is suported by the growing urbanisation which induces new eating habits (Aubron 2006b).


The livestock practices in the dairy farming systems of marginal regions of the Andes are not well known (Hervé 1992) and often considered unproductive and inadequate compared to those of large farms of the Pacific Coast, where the cows are stall-fed (Bernet et al 2001b). Based on the monitoring of 8 farming households of a Peruvian community, this study contributes to improve knowledge about dairy farming practices in a harsh environment of the dry Andes. 


Study region 


The Sinto community (13°20’ S, 75°26’ W) is located at an altitude ranging from 3000 to 5000 meters above sea level in the high valley that ends on the Pacific Coast near the town of Pisco (Figure 1).

Figure 1
.  General map of Peru and location of the study region

On this western side of the Andes, the climate is very dry and the relief abrupt. The 8-month period without rainfall (Figure 2) is a major constraint for dairy farming as daily milk production requires daily forage. This difficulty can be partially overcome with irrigation.

Figure 2.  Precipitations and temperatures, Sinto Station, average values for the period 1972-1981

In the dry season the river water is collected and conveyed hillside through clay and concrete irrigation canals to the cultivated fields. Located downstream where the slope and the rocky masses do not prevent men and animals to circulate, the irrigated fields are generally small (100 m² to 1 ha) and sown with alfalfa. This pulse fodder crop is sown after an early potato development cycle and remains cultivated for 5 to 20 years. It is gradually overgrown with other herbaceous plants, the main one being the kikuyu (Pennisetum clandestinum), creeping grasses with asphyxiating effects on the alfalfa. The alfalfa fields constitute the main fodder resource for the grazing cows. In addition, the cows can graze the rangelands: these large zones of several hundreds of hectares, are often located at a high altitude (3800 to 4200 masl) where frost is frequent, and present a discontinuous plant cover with low forage yield (Dollfus 1982).


The status of Sinto has evolved over the years. The hacienda was replaced, after the agrarian reform by the cooperative, destroyed in 1985. Sinto is since then a peasants community. The 350 families of the community share a body of rights, defined by the community (Mayer 2002), regarding the management of the resources used in dairy farming. The rangelands and irrigation water are managed collectively and each crop field is cultivated by one family. The fields can be transmitted by inheritance but cannot be sold. The herds belong to the families and each farmer owns 1 to 40 cows.


In Sinto, milk and cottage cheese provide in Sinto a weekly cash income which covers a week’s food expenses for the household (Aubron 2006a). Thus the importance of regular production. Like in other livestock farming systems (Siegmund-Schultze et al 2007), cull cows and bulls are sold live when there’s an urgent cash-flow need. Small ruminant farming – sheep for the meat and goats, both for meat and milk (mixed to cow milk) – is often combined to dairy farming in the production systems. Potatoes and barley crops are solely grown for home consumption.



This study is part of PhD research work in comparative agriculture, at the frontier of social and agronomical sciences (Cochet 2005), based on a 2-year field experience within the peasant community of Sinto in the Peruvian Andes. The first step consisted in establishing a diagnostic assessment of the agriculture by the means of surveys and a thorough observation of the landscape. Then we carried out a monthly monitoring of 8 peasant families over a one-year period. This enabled to understand the tecnico-economic functioning of the livestock farming systems (Lhoste 1984) and land management, through the analysis of the practices (Landais and Balent 1993). Comprehensive rather than quantitative, this analysis is based on two aspects: the assessment of the agronomic and zootechnical consequences of these practices and the understanding of the conditions and deciding factors leading to their implementation by the farmers (Milleville 1987).


The 8 families were selected according to their geographical location, their forage management method, the herd size and composition and the marketing channel of their dairy products. The monthly monitoring consisted in observing throughout the day all livestock production activities and in collecting farm characteristics (herd composition, field size, location and rights of usage, labor force) and all the relevant data since our previous visit: transfer of the animals to other fields and rangelands, variations in the herd size and in milk production, labor planning, marketing channels of the dairy products, etc.


We describe the forage management practices identified during the monitoring visits in a grazing schedule which showed the evolution of the following: (i) the forage resources required to feed the animals, (ii) the herd composition, (iii) the milk yield per cow and per household. The analysis of this data highlights the critical periods for each farmer due to fodder deficit and the practices implemented consequently.


Results: forage management practices 

Cattle grouping


Livestock farmers comonly divide their animals into groups for various reasons (Dedieu et al 1997). In Sinto, the purpose is to provide the dairy cows, with the best fodder and with most of the household labor force. The lactating cows and their calves (which presence is necessary during milking) make up the first group that grazes alfalfa fields under the watchful eye of their owners 3 to 6 hours a day. Referred to as “lazy cows”, the non-dairy cattle – heifers, young bulls, dry cows – forms the second group that grazes freely on rangelands (Figure 3).

Figure 3.   Spatial functioning of the livestock farming system. Example of a family during one or two months.

Rotational grazing system on irrigated alfalfa fields


Each household has an access to one to fifteen alfalfa fields, which are scattered and sometimes located at over an hour walking-distance from the house. Under Andean conditions and with irrigation during the dry season, alfalfa can complete several growth cycles per year, with a relatively regular regrowth interval, more or less 75 days in Sinto. The yield of the alfalfa fields is affected by several factors (water supply, altitude, orientation, soil, age, grazing practices, etc) thus it varies over time and according to the field location. The alfalfa crops being the main fodder for the dairy group, the farmers try as far as possible to take into acount this variability for the organisation of grazing rotation. 


During the rest period between two grazing times, alfalfa collar and root reserves can be stockpiled and used for the following growth. Despite sometimes long grazing times on the same field (over a week), the close watch of the animals prevents them from removing the regrowing plants and ensures regular animal performance (Voisin 1963). The farmer checks that the cows eat every day “some old, some new”, similary to strip grazing.


Occasional rangeland grazing to decrease stocking rate on alfalfa fields.


In order to destock the alfalfa fields and to extend the regrowth period, the farmers sometimes graze the dairy group on the lands boarding the paths, the irrigation canals, the fields or on rangelands (Figure 4).

Figure 4. Grazing schedule of the dairy group for one monitored family.
Alfalfa fields are reprensented in black color and rangelands in grey.

During the rainy season, this practice enables the farmers who live in remote valleys and who therefore have access to large rangelands to feed herds of up to 20 lactating cows. During the dry season, this practice applies to the rangelands excluded from grazing during the rainy season. Most of them are managed collectively. They are called “potreros” and ensure stockpiled forage, which is consumed dry and headed out during the dry season when the grazing is opened by the group controlling its utilisation. The diet change – alfalfa to natural pasture – generally leads to a decrease in milk production. This unfavourable effect is amplified as the dry season goes on. Moreover, parasitism problems may appear.


Practices of exchange forage: guaranteeing flexibility


In Sinto, like in other regions of the Andes, there is a market for standing alfalfa (Roman 1988). The price depends on the quantity and quality of available forage for the on-going cycle. The price is estimated by the farmers, in number of grazing days and indirectly in quantity of milk likely to be produced. If a farmer wants to maintain the same level of milk production, which implies providing the cows with constant quantities and quality of feed, he might buy alfalfa crops at a higher price than the estimated value. Alfalfa prices also fluctuate with the supply and demand of forage within the community and according to the amount irrigation work required by alfalfa cropping. Additionally, prices depend on the relationship between the persons making the deal, which leads to a strong fluctuation of the prices of a grazing day per cow. Alfalfa is usually supplied by a few farmers, who often do not own livestock and for whom selling planted alfalfa crops has become their main agricultural activity. Nevertheless, any family can buy and sell alfalfa crops, hence giving flexibility to forage management.


Share dairy farming is an agreement between cattle owners and alfalfa fields owners. During a given period of time, the field owner graze on his alfalfa the lactating cows and their calves of the cattle owner. The alfalfa owner keeps the milk produced for payment. Like the previous practice, this system has the advantage of keeping the same diet and clearing the household labor from livestock production activities, which can be useful for the cattle owner in case of travelling, illness or competition with other farming activities. The advantage for the alfalfa field owner is to make profitable use of available labor force and forage by producing milk. Implemented by the farmers to deal with the tide-over periods of their grazing schedules (Figure 4), these exchange practices secure feeding management of the dairy group and guarantee a better dispatch of the forage resources, at individual and collective levels.


Secondary feeding practices to increase and balance global ration.


Due to fresh alfalfa feed, bloat is the main cause of death of the dairy cattle in Sinto. In addition to increasing the ration, the purpose of several practices called “secondary feeding practices” is to reduce the bloat-risk with a more balanced diet raising fibre intake. The farmers often graze the cows free on rangelands near the alfalfa field morning and evening during the rainy season. Before taking their cows to the alfalfa fields in the morning, some farmers let them feed one or two hours on nearby rangelands or wheat and barley stubble. They refer to this practice under the Quechua name for “breakfast”. The morning distribution either of ichu, a siliceous grass, cut with a sickle, common in Peruvian highland pastures or of barney hay or wheat straw completes at times alfalfa grazing. Finally other farmers buy concentrates made of residues of the coastal crops (maize, molasses, cotton oilseeds) and feed the cows small quantities (approximately a kilo) every morning. 


Discussion: three levels of consistency in the livestock farming practices 

A regular daily milk production


The first consistent level is the economic purpose of milk production. It must provide a weekly income essential to cover the household’s weekly food expense. Because of its key-role in the food security of the peasant families, milk production must be regular throughout the year and the volume slightly above break-even point to ensure the purchase of weekly goods. The achievement of this objective is made possible by combining the various practices here described which guarantee the best possible feed intake for lactating cows.


The monthly milk production chart of the 8 families shows a relatively regular trend: the global volume of milk produced by the 8 families only decreases by 30% during the dry season (Figure 5). In a region where the dry season lasts 8 months, this is quite a performance, which is achieved thanks to the irrigation of the alfalfa fields and to the individual or collective forage management practices (rotation, stockpiling of some rangelands, distribution of concentrate, hay or straw feeds, etc).

Figure 5.
  Global milk production change during the year for the 8 monitored families (liters per month).

However, there is a strong link between the quantity of milk produced by a family and its stability and the size of irrigated fields, the expansion of which is now limited within the studied community. These livestock farming systems will need to investigate new forage production methods requiring less irrigation water in order to strengthen their ecological sustainability (Brunschwig and Aubron 2005). Dry cultivation of barley and oats, their conservation as hay or silage, are possible options.


Optimised milk production with regards to the restrictive factor: irrigated land


With the data collected during the monitoring, our average milk production estimates are: 980 L per cow and per lactation period and 640 L per cow per annum. If compared to the production of other cows, three facts must be taken into account: some of the milk goes to the calves, milking takes place only once a day and Sinto cows are very small-sized. Nevertheless, these individual results are low for cows fed with highly nutritious forage such as alfalfa. It can be explained by underfeeding and an unbalanced diet (Bernet and León-Velarde 2000).


With the average values of land-use for 40 fields and the data collected during the monitoring, the average yearly yield of the alfalfa fields is estimated at 3900 L of milk produced per hectare. This figure must be interpreted cautiously as forage production as well as milk production of the herds fluctuates (Aubron 2006a). To compare this result with performances of other dairy systems, it would be necessary to evaluate precisely the feed supply of other forage and the amount of alfalfa forage used by non dairy production such as calf growth. Notwithstanding, these figures are still very high compared to those obtained in France in all-grass systems (3000 L/ha in Normandy, 1500-1800 L/ha in the Vosges) and even in systems combining grass and cereal concentrates (4000 L/ha in Rhône-Alpes with 1600 kg of concentrates per cow, Réseaux d’Elevage 2005).


Hence, in these dairy farms good milk yields per hectare of alfalfa can be achieved despite a low milk yield per cow. A similar combination of results was obtained in other Andean livestock farming systems for meat production (Brunschwig 1996). This highlights a second level of consistency: irrigated land rather than the size of the herds is a restrictive factor. This conforms with the strong link existing between access to irrigated land and the combination quantity / stability of dairy production crucial to the food security of the households.


Risk management


Some of the practices described are part of a global risk-management strategy: the forage exchange and the conservation of the crop residues secure cow feeding; the stockpiling of some rangelands and the flexibility of the collective rules organising its use are an answer to the climatic risks of the dry season.


Furthermore, even if milk production per cow is weak, owning a big-size cattle herd is a better asset than just farming a few very productive dairy cows. Owning a large rustic herd is part of a risk-management strategy; it prevents the farmer from a big loss in case of the death of an animal, which is frequent for non-lactating animals grazing free on rangelands. Rangelands supply with fodder calves growth, cows pregnancy and bull fattening. They are essential for meat production which provides additional non-dairy income. They don’t require much investment neither financial nor in terms of labor. This combination of the alfalfa fields intensive farming for milk production and the rangelands extensive farming for meat production is similar to the dynamic of intensification/desintensification observed in the neighbouring valley of Cañete (Wiegers et al 1999).





Aubron C 2006a Le lait des Andes vaut-il de l’or ? Logiques paysannes et insertion marchande de la production fromagère andine. PhD dissertation. Institut National Agronomique Paris-Grignon (INA P-G), Paris, France


Aubron C 2006b Productores andinos de queso artesanal y liberalización del mercado de los lácteos en el Perú. Debate Agrario 40/41: 119-140


Bernet T and León-Velarde C 2000 Income effects of fodder and herd management on small-scale milk producers in the northern Peruvian Andes. Livestock Research For Rural Development. (12) 3:


Bernet T, Pradel W and Walker T S 2001a The importance of farmers' income security for enhancing regional development: Evidence from southern Peru. Livestock Research For Rural Development. (13) 3:


Bernet T, Staal S and Walker T S 2001b Changing milk production trends in Peru: Small-scale highland farming versus coastal agrobusiness. Moutain Research and Development 21: 268–275


Brunschwig G 1996 El alto valle de Cañete: el matorral y la puna. In: Morlon P (coord.), Comprender la agricultura campesina en los Andes centrales: Perú-Bolivia. IFEA - CBC, Lima, Peru, pp 374-398


Brunschwig G and Aubron C 2005 Durabilité des systèmes d’élevage bovins laitiers dans les Andes centrales péruviennes : cas des communautés de Sinto et de Huancaya. Rencontres autour des Recherches sur les Ruminants 12: 214


Cochet H 2005 L’agriculture comparée. Genèse et formalisation d’une discipline. INA P-G, Paris, France, 88 p


Dedieu B, Chabanet G, Josien E and Bécherel F 1997 Organisation du pâturage et situations contraignantes en travail : démarche d'étude et exemples en élevage bovin viande. Fourrages 149: 21-36


Dollfus O 1982 Development of land-use patterns in the central Andes. Moutain Research and Development 2(1): 39-48


Hervé D 1992 Adaptación a los sistemas agrarios andinos de un modelo de intensificación lechera. SEPIA IV: 347-369


Landais E and Balent G 1993 Introduction à l'étude des systèmes d'élevage extensifs. In: Landais E (editeur), Pratiques d'élevage extensif. Identifier, modéliser, évaluer. INRA, Paris, France, pp 13-34


Lhoste P 1984 Le diagnostic sur le système d'élevage. Les Cahiers de la Recherche Développement 3-4: 84-88


Mayer E 2002 The Articulated Peasant: Household Economies in the Andes. Westwiew Press, Boulder, 390 p


Milleville P 1987 Recherches sur les pratiques des agriculteurs. Les Cahiers de la Recherche Développement 16: 3-7


Roman C 1988 La ganadería lechera en la Cuenca alta del Cañete, su rol en el desarrollo de los sistemas agrarios y su articulación al mercado nacional. In: Eresue M and Brougère A M (editores), Políticas agrarias y estrategias campesinas en la cuenca del Cañete. UNALM – IFEA, Lima, Peru, pp 85-97


Réseaux d’Elevage 2005 Cas-type des exploitations d’élevage. Institut de l’Elevage et Chambres d’Agriculture, Paris, France


Siegmund-Schultze M, Rischkowsky B, de Veiga J B and King J M 2007 Cattle are cash generating assets for mixed smallholder farms in the Eastern Amazon. Agricultural Systems 94: 738–749.


Voisin A 1963 Productividad de la hierba. Tecnos, Madrid, Spain, 554 p


Wiegers E S, Hijmans R J, Hervé D and Fresco L O 1999 Land Use Intensification and Desintensification in the Upper Cañete Valley, Peru. Human Ecology 27: 319-339

Received 30 July 2007; Accepted 27 November 2007; Published 1 February 2008

Go to top