RESEARCH WORK

Evaluation of the trees Sambucus nigra and Acacia decurrens as supplement for dairy cows in the Bogotá Savanna, Colombia

Teresa Carvajal1, L. Lamela2 y Aurora Cuesta1

1 Universidad de Ciencias Aplicadas y Ambientales (UDCA). Calle 222 N° 54 37, Bogotá, Colombia
E-mail: teresita.carvajal@gmail.com
2Estación Experimental de Pastos y Forrajes «Indio Hatuey», Matanzas, Cuba

ABSTRACT

A study was conducted in the high tropic of Colombia, Bogotá savanna, in order to achieve partial substitution of the concentrate feed by tree foliage from green wattle (Acacia decurrens) and elder (Sambucus nigra), through agreement with the farmers, without affecting milk production. The evaluations allowed characterizing the nutritional composition of the species selected by the farmers. Likewise, the concentrate feed protein was partially substituted by the tree protein of these species (in percentages of 10, 20 and 40); a switch back design was used to determine milk production in the evaluated diets. The best results with A. decurrens were obtained when 10% of the crude protein of the concentrate was substituted, and with S. nigra when including 20%. With these inclusions 96 and 94% of the production potential was achieved, respectively. It is concluded that the forage from the tree species A. decurrens and S. nigra has adequate nutritional value and it can be used as partial replacement of concentrate in 10 and 20%, respectively. Also, such forage has the acceptance of dairy farmers in the Bogotá savanna; therefore, it is suggested to use the developed methodology for similar studies in other regions of Colombia.

Key words: Acacia decurrens, Sambucus nigra, dairy cows, nutritional value.

INTRODUCTION

In the high Colombian tropic (Cundinamarca), and especially in the high plateau of the Bogotá Savanna, one of the main livestock production activities of the country is milk production with specialized cattle breeds (Holstein, Jersey, among others), which are fed with biomass produced by grasslands. One of the most important characteristics of these systems is the high cost, due to the supplementation with concentrate feed; this contributes to enhance the crisis that affects livestock production and the national dairy sector, which in recent years has been a cushioning factor because it contributes 10% within the agricultural GDP (Cárdenas, 2003a).

These production systems face other constraining factors, like the prices of agricultural inputs (fertilizers, herbicides, pesticides, seeds) that influence the final price for the consumer, which keeps away some resource-poor sectors from milk consumption and reduces the sector competitiveness before other producer countries. The feeding costs in cattle production systems represent up to 45% of total costs and the expenditures of concentrate feeding constitute the largest part of this percentage (Holmann et al., 2003).

Fresh milk production has variability in the produced volumes throughout the year, as a consequence -among other factors of the climate seasonality (rainy and dry season), which affects pasture availability and causes a variation of almost 10% in milk production (Ministerio de la Agricultura y Desarrollo Rural, 1999).

In this context, silvopastoral systems could be a choice to reduce the feeding problem of cows. Their main objective is the increase of production, productivity and sustainability (Sánchez et al., 2010), taking advantage of the contribution of perennial ligneous plants that is represented mainly by nutrient recycling (Rusch et al., 2009), soil protection from the wind and rain effects (Arias, 2009), increase of species diversity and forage and nutrient contribution for the animals (Franco, 2008). Nevertheless, in Colombia the current use of these systems is limited.

In order to mitigate these constraints the inclusion of tree forage was suggested as replacement for concentrate feed protein to reduce feeding costs, which was the objective of this study.

MATERIALS AND METHODS

Geographic location, climate and soil

The experiments were conducted at the Roncesvalles farm, in the Bogotá Savanna, Cundinamarca; this is located in the Colombian Andean region, known as zone of cold climate, corresponding to one of the branches of the Eastern Mountain Range, with a height of approximately 2 600 masl. It belongs to the characterization of Andean low mountain climate zones, with an average temperature of 12ºC, a bimodal rainfall regime and a relative humidity of 75% (IDEAM, 2008).The soils have good organic matter content and profiles of up to 1,20 m; they show acidity, with low levels of P, Ca, and Mg and average to high values of K and they are defined as having moderate fertility.

Experimental units, design and experimental procedure

Three experiments were conducted, in which three diets were tested with the inclusion of forage from the species Acacia decurrens and Sambucus nigra. In experiments one and two, 12 Holstein cows were used, in their third month of lactation, with approximately 550 kg of live weight; and in the third experiment, 12 Jersey cows of 400 kg of live weight were evaluated. The animals were distributed at a rate of four cows per treatment, according to a switch back design.

The production was daily recorded and it was processed through a multiplicative model with adjustment of lactation curve (Menchaca, 1978). For the analysis of milk production the following model was used:

Yij= a + b log n + cn + Pk+ Tl + eijk

Where:

Yij: log Yij

a: log A, constant common to all observations.

b, c: parameters of the lactation curve, according to the algebraic representation proposed by Wood (1969)

n: n-th lactation day corresponding to the Yij-th observation.

Pk: k-th experimental period.

Tl : l-th treatment.

eij: log Eij, normal and independently distributed residual error with zero mean and variance s2

According to the management established in the farms, the cows in this stage of lactation should receive 3 kg of concentrate/animal/day. In experiments one and two, part of this concentrate was substituted (in its equivalent in protein) by forage from green wattle and elder at a rate of 10, 20 and 40%. In experiment three, the same substitution form was used, but the best inclusion percentages in the previous trials were considered, and an additional treatment with commercial concentrate was included.

The animals were separated only before milking, to receive the scheduled supplementation; this was manually done at 6:00 am and 5:00 pm. In all cases, the supplement was supplied distributed in the milking times.

Table 1 shows the experimental treatments.

During the experimental stage the animals grazed in a kikuyu grassland (Pennisetum clandestinum), divided by means of electrical fences, in this sense, a strip-plot design was established, capable of guaranteeing a stocking rate of one cow per hectare. The time of pasture recovery was 45 days and the strip occupation lasted seven days, with a grazing time of approximately 18 hours per day.

The foliage of the species A. decurrens and S. nigra (leaves and petioles) was taken from arrangements of living fences, which had been established for five years, at a rate of one tree per every 2 lineal meters.

Every design lasted 63 days, with three evaluation periods. Every period consisted in fourteen days of adaptation and seven days of data collection.

Table 2 shows the bromatological composition of the concentrate feed.

Sample collection. Pasture samples were manually collected, from the top of the plants, simulating the selection made by the animal when grazing, and those corresponding to the trees were taken from the foliage supplied to the cows, which included only the leaves and petioles.

To do the analyses of nutritional quality, the material was dried (AOAC, 2003) and it was ground in a Wile mill with sieve of 1mm. The following were determined: total crude protein (CP) and the five nitrogen fractions of crude protein (Fox et al., 1992): 1) non- protein nitrogen (NPN: fraction A); 2) true soluble nitrogen (fraction B1); 3) true insoluble nitrogen not bound to the NDF (fraction B2); 4) true nitrogen bound to the neutral detergent fiber (NNDF) (fraction B3); and 5) true nitrogen bound to the acid detergent fiber (NADF) (fraction C).

The neutral detergent fiber (NDF) and acid detergent fiber (ADF) were determined through the method described by Van Soest and Robertson (1985), through the ANKOM equipment.

Economic appraisal: An economic appraisal was made to determine the production cost of one milk liter in a farm of 100 animals, taking into account the values established in the dairy basin of the Cundiboyacense plateau, where the study area is located.

The basic study was aimed at testing the effects of tree forage inclusion, as partial replacement of the crude protein contributed by the concentrate feedstuff, on the production costs of one milk liter in the farm. The direct and indirect costs of a typical exploitation were taking into account, as well as the feeding costs, assuming the values of 2009 in Colombia, but converted to U.S. dollars to facilitate a better comprehension of the results.

RESULTS AND DISCUSSIONS

The nutritional analyses of the tree forage indicated that S. nigra showed higher CP values with respect to A. decurrens, and that both possess suitable nitrogen potential to be used in ruminant feeding.

S. nigra also showed higher percentages of the nitrogen soluble fractions, while A. decurrens had higher values of insoluble fractions and tannin contents. The presence of saponins was not detected (table 3).

The contents of crude protein in A. decurrens were similar to the ones obtained by Nova, Chamorro and Carulla (2005), while the S. nigra contents were equivalent to the ones found by Blanco, Chamorro and Arreaza (2005) in the Bogotá Savanna. These values are equivalent to a nutritional quality comparable to that of a commercial concentrate and, in the case of S. nigra, crude protein was higher in five percentage units with regards to A. decurrens.

In the evaluation of secondary metabolites, the tannin content showed a marked difference between the tree species: 0,08 for S. nigra and 8,1% for A. decurrens. Tannins are astringent substances that adhere to the protein and form a complex which is impossible to be solubilized by the enzymes of ruminant bacteria, hence it goes directly to the abomasum; there it is degraded to its initial components, which allows incorporating the so called bypass protein to the diet and this contributes to increase milk production (Frutos et al., 2000; Hervás et al., 2000).

The cell wall (NDF) and the lignocellulose fraction (ADF) were higher in A. decurrens and the lignin contents were similar in both species (table 4).

The NDF values for S. nigra and A. decurrens indicate that it has potential for animal feeding, and coincide with the ones reported by Nova et al. (2005).

The production systems of the high Colombian tropic guarantee animal feeding in the native pastures (Cárdenas, 2003b). Among them is kikuyu grass, which has become the most widely spread and used foliage, since its introduction in 1927.

One of its qualities is the adaptability, because it is found in climate zones that oscillate between 1 650 and 2 800 masl, with excellent yields in forage and acceptable quality, although with some water and fertilizer demand. However, its persistence and high biomass production have been limited by its susceptibility to frosts, which commonly appear in this region during January and February and, in a lesser proportion, during July and August.

Table 5 shows the average nutritional value of the kikuyu grass available for animals during the experiments, this is in correspondence with the information given by Sánchez and Villaneda (2009), who consider that the potential for milk production from this pasture has a limit close to 12 L/cow/day, although some data show that this may increase up to 29 L/cow/day, due to its net energy content of lactation.

Regarding biomass production, it was higher than 16 t DM/ha/year, with a resting time of 45 days, which indicated that the animals could choose a pasture that, complemented with the forage of the tree species, can produce more than 12 kg of milk/cow/day.

The A. decurrens and S. nigra forages obtained in the experimental farms showed a high nutritional value when they were cut at 45 and 60 days , respectively, and their production guaranteed the amounts needed to be included in the diets (table 6).

A. decurrens showed higher forage production in six cuttings (4,5 t DM/year), while S. nigra produced 25 kg per tree in eight cuttings every 45 days, with a total of 4,2 t DM/year. These results are similar to the ones reported by Giraldo (1999), when these species where used in similar arrangements of living fences, typical of the farmers in the zone, and, they were similar to the ones obtained in preliminary tests in Piedras Blancas (Medellín), where 4 t DM/ha were produced in two years (Escobar, 1993).

Regarding production per tree, it was higher than the one found by Giraldo (1999), who obtained 0,686 kg in plants that were 17 months old.

On the other hand, these productions were higher than the ones reported by Giraldo and Bolívar (1999), when analyzing A. decurrens with two planting densities; these authors found that the plant biomass produced was 1,4 and 1,0 t DM/ha for the low and high density, respectively.

Figure 1 shows milk production according to the inclusion percentage of A. decurrens forage (experiment 1); the best results were obtained when 10% of the crude protein of the diet concentrate was substituted, and the lowest production was obtained with 40%. Fernández, Zapata, and Giraldo (1999) obtained milk productions of 13 kg/cow/day, when they included A. decurrens at 15% in a diet that had as feeding base the forage from kikuyu grass, result that is close to the one obtained in this study with 10% of inclusion.

However, figure 2 shows that the higher production with S. nigra (experiment 2) was achieved with the inclusion of 20% (14 kg daily); this value is very close to the ones reported by Blanco, Chamorro and Arreaza (2005), who obtained between 16 and 18 kg per day in milk production estimated by the CNPS model.

These results proved the need to compare them with those of a diet that included commercial concentrate, similar to the one used by the farmers of the zone with animals of an equivalent dairy potential (Jersey). In this sense, to conduct the studies, the best treatments of the previous experiments were included: A. decurrens at 10% and S. nigra at 20%. Figure 3 shows the performance of milk production when it was compared with the concentrate as supplement.

The best results were achieved in the diet in which the commercial concentrate was used as supplement; the productions were close to 20 kg per day, with significant differences (p<0,01) for the rest of the treatments. In the case of the A. decurrens inclusion at 10%, milk production was equivalent to 96% of the control treatment or traditional diet (19,1 kg), followed by S. nigra, in which it was 94% (18,7 kg).

González and Chamorro (2005), when evaluating an integral diet with A. decurrens, Chusquea scadens, Solanum tuberosum, palm kernel and corn silage, and a diet supplemented with commercial concentrate (control), obtained a milk production of 19,5 kg in Holstein cows, which was higher in 21,8% than the control treatment, with a lower cost; this showed the benefits of including A. decurrens in integral diets.

The productions obtained in this experiment were higher than the ones reported by Chamorro (2005), when he used the ligneous forage tree Leucaena leucocephala in grazing associated to African star grass (Cynodon nlemfuensis) and the dairy cows were supplemented with 2 kg of rice meal, daily.

On the other hand, they are similar to the ones reported by Rodríguez et al. (2010) when they supplemented Holstein cows with meal from mulberry (Morus alba) and elder (Sambucus peruviana) forage mixed with oat meal (Avena sativa). The milk production with the inclusion of the ligneous forage exceeded the control of oat meal alone (21,5; 18,7 and 18,0 for mulberry, elder and oat, respectively). In general, the productions using tree forages prove the potential that these feedstuffs have to be used in cows of average potential, because this indicator oscillates between 10 and 20 kg of milk/cow/day.

From the economic point of view, the inclusion of the trees in the diets allowed a significant saving due to concentrate purchase expenditure reduction. With A. decurrens at 10% US $0, 12/cow/day were saved, and with S. nigra at 20%, the reduction was US $0, 24/cow/day, with the additional advantage that these foliages contribute to decrease dependence on external inputs, by being produced in the farm itself (table 7).

If a lactation of 305 days and a herd of 100 cows are considered, the inclusion of tree forage from A. decurrens would allow an annual saving of $3 660,00, which is more evident when the forage from the tree S. nigra is included to replace 20% of the concentrate protein, therefore in a lactation US $7 320,00 would be saved.

It is concluded that the species A. decurrens and S. nigra have an adequate nutritional value to partially replace concentrate feed in the diets of dairy cows, therefore their inclusion in 10 and 20% allows, in average potential cows, milk productions of 96 and 94%, respectively, with regards to the milk production obtained with supplementation based on concentrate, saving US $0,12-0,24/cow/day.