Introduction

Small-scale dairy systems in Mexico are a rural development option since they enable farming families to overcome poverty indices¹. These systems are important in many areas of the world with common features to the highlands of Mexico², like in other Latin American countries as in the Andean highland regions of Peru^3,4 and Uganda⁵.

In Mexico, they are defined by small farms with herds between three and 35 cows plus replacements, and rely on their family labour⁶. Their sustainability is jeopardised by high feeding costs in the face of stagnated prices for milk, mainly due to their reliance on external inputs of which bought-in commercial compound concentrates represent the highest proportion of costs, since farmers believe that high levels of concentrate supplementation are essential for milk production, even at the moderate milk yields in these systems. Therefore, the economic scale limits their sustainability⁶.

A higher use of quality home-grown forages may increase the profitability and hence sustainability of these systems, as is the case for intensive grazing of temperate pastures in farms with access to irrigation, that has been shown to reduce feeding costs in these systems⁷.

However, one limitation of grazing dairy cows is the low intake of dry matter⁸, particularly under high grazing pressure, so that concentrates may be required in these conditions to sustain milk yields. Small-scale dairy systems traditionally have more cattle than the carrying capacity of their small farms with stocking rates over 3.0 cows/ha of agricultural land⁶, so that a high concentrate use may be justified in these systems.

Therefore, the objective of this work was to assess the productive and economic response of grazing dairy cows at a high stocking rate to increased levels of compound concentrate supplementation, as well as their effect on feed intake.

Material and methods

The work took place in Ejido San Cristóbal, a smallholder campesino village where most families are small-scale dairy farmers, located in the highlands of central Mexico at 19º 24’ N and 99º 51’ W, at an altitude of 2,650 m. The region has a sub-humid temperate climate with a distinct rainy season (May - October) and dry season (November - April), and average annual rainfall of 800 - 1,000 mm and a mean annual temperature of 13°C. The experiment took place during the rainy season, from June 26^th to August 27^th of 2000.

A demonstration module in feeding strategies for small dairy herds was established in consultation with the community on a 1.5 ha^-1 plot of the local school. A local participating farmer managed the module with a herd of six local milking cows following a participatory livestock technology research approach⁹ so that results were applicable by farmers in the region and other areas with similar systems.

Results

Average temperature was 13.6^oC, with a maximum and minimum of 20.5 and 6.8 ^oC, respectively. Total rain fall during the experiment was 332 mm, distributed as follows 139, 122 and 61 mm in EP1, EP2 and EP3, respectively.

Table 1 shows results for net herbage accumulation (NHA) per period and per day, as well as mean sward height. Net herbage accumulation and DHA in EP3 were significantly higher than in EP one and two (P<0.01).

Table 2 shows the chemical composition of the pasture herbage. Crude protein and digestibility were not different across EP (P>0.05). Crude protein ranged from 122 to 162 (EP1 an EP3, respectively), with a mean of 147 g/kg DM. Digestibility (Dig) mean digestibility was 581 (g/kg DM). Dry mater, ash, OM, NDF and ADF were significantly different across EP (P<0.05). Estimated metabolizable energy (eME) was different (P<0.001) among EP, the lowest value was in EP1 (10.1), whereas EP2 and 3, were not different among each other (11.2 and 11.2 MJ ME/kg DM).

Table 3 shows the results for feed intake, with significant differences (P<0.05) among treatments. There were no differences between T1 and T2 in herbage intake but lower intake in T6. Due to concentrate supplementation, total feed intake was not significantly different (P>0.05) between T3 and T6, but total intake was significantly lower (P<0.05) for T1. In time, there was a reduction in herbage intake, in Period 2, in spite of improved grazing conditions; this, lead to a significantly lower total intake (P<0.05) in Period 2, compare to periods one and three.

There were significant differences (P<0.05) among treatments for milk yields (Table 4); with no differences between T1 and T3, which were significantly different (P<0.05) from T6 which had the highest yield.

Table 5 shows results for feeding costs (in US dollars). Increased concentrate supplementation increased feeding costs. Total feeding costs in T6 are almost three times the feeding cost in T1, whilst milk income was only 28 % higher, which results in a feeding cost per kilo of milk 2.2 times more expensive in T6 than in T1; with figures for T3 intermediate. Profit per kilogram of milk was therefore 33 % higher in T1 than in T6.

Discussion

The effect of weather condition is reflected in the chemical composition of the herbage throughout the different growing seasons. This grazing season was characterized by low herbage growth rate, short regrowth, low herbage mass availability and low herbage intake. In addition, grass development was based on vegetative growth, characterized with higher proportions of grass leaf, lower proportions of grass stem and dead material, and more digestible than other vegetative state. Temperate herbages used for dairy cows are described as high quality when chemical composition is around 180-240 g DM kg-¹, 180 to 250 g CP kg-¹ DM, 400 to 500 g NDF kg-¹ DM, and 10.47 to 12.14 MJ ME kg-¹ DM¹⁹. Under the conditions of this experiment, herbage was characterized by low concentrations of crude protein, low energy and low amounts of non-structural carbohydrates and DM.

Crude protein content of pasture herbage was lower than a report in southern Brazil²⁰, and lower than reports of work undertaken in the same study area^21,22; but sufficient to meet protein requirements for moderate yielding dairy cows²³.

Structural carbohydrate content determines digestibility, intake, and the nutritional value of forages. Average values of NDF and ADF of pasture herbage were 521.37 and 327.57 g/kg DM, respectively; lower from reports in a previous work in the same area, but during the dry season²⁴.

Estimated metabolizable energy was 10.8 MJ/kg DM on average (Table 2) representing an herbage of good quality. However, this value is lower than reports of 12.3 MJ EM/kg DM in the highlands of Mexico¹⁰, or 11.2 MJ EM/kg DM reported in New Zealand²⁵. The interaction of lower energy content and herbage availability may explain low intakes and low milk yields observed (Tables 3 and 4).

Net herbage accumulation was low during periods 1 and 2, which given the high stocking rate, resulted in a high grazing pressure and low herbage availability, with very low sward metre heights. Improved grazing conditions in terms of herbage growth and availability for Period 3, enabled cows to recover milk yields similar to those of Period 1, overcoming the loss of almost 3 kg/cow/d^-1 from Period 1 to Period 2 as grazing conditions deteriorated.

Difficult grazing conditions with low herbage availability and moderate milk yields resulted in low herbage intakes, which were significantly decreased (P<0.05) by the high supplementation rate in T6, with high substitution rates.

It has been reported that 0.31 kg/d^-1 of concentrate supplemented to grazing dairy cows results in a 1.0 kg DM reduction in herbage intake²⁶. Nonetheless, herbage intake in the experiment herein reported was similar to reports by^27,28 with grazing dairy cows in low herbage mass pastures during winter in France, reporting a mean daily intake of 7.2 kg DM/cow.

Observed milk yields were lower that results for grazing cows reported by in the USA²⁶, in the UK²⁹, in southeast USA³⁰, and in Mexico²⁴. However, observed milk yields were similar to reports by others²⁷ from cows under difficult grazing conditions in western France, illustrating the fact that difficult grazing conditions limit intake, and yields, particularly during late lactation.

Milk fat contents of milk are similar to results reported in France²⁷, in Spain³¹, and in Chile³². Although there were no statistical differences in milk fat content (P>0.05) there was a trend towards lower content in the high supplementation treatment T6.

Protein content of milk was higher than reports from works in Mexico with small-scale dairy farmers^33,34, as well as higher than reports from Chile³²; but similar to milk protein content from a study in the UK²⁹.

In terms of feeding costs and returns, in spite of lower yields for T1 and T3 compared to T6, the margin over feeding costs is similar for the three treatments. However, in terms of feeding cost per kg of milk and profit over feeding costs, T1 has lower cost and higher profit. Feeding cost per kilo of milk was 2.5 times higher in T6 compared to T1, and margin was 25 % higher in T1 than T6, with T3 showing intermediate economic results. Low feeding costs and similar margins among treatments result in a very high returns / feeding costs ratio for T1 compared to T6.

Conclusions and implications

In conclusion, there is no economic benefit of increased concentrate supplementation since total margins over feeding costs are similar; and farmers need a greater cash flow to cover the increased costs of higher amounts of concentrate used. However, there is a need for long-term experiments since supplementation alleviates the grazing pressure due to high stocking rates. NHA was low resulting in difficult grazing conditions limiting the intake of cows. Forage or by-products of lower cost than commercial compound feeds may also be an alternative to sustain cow and pasture productivity. An optimal compromise in productive, economic and in the soundness of the pasture will ensure the long term sustainability of these small-scale farming systems.