Livestock Research for Rural Development 35 (4) 2023 LRRD Search LRRD Misssion Guide for preparation of papers LRRD Newsletter

Citation of this paper

Effect of Sambucus peruviana and Tithonia diversifolia silage on methane emissions by Holstein cows fed Cenchrus clandestinus

N Arias-Ortiz, D M Bolívar-Vergara and R Barahona-Rosales

Departamento de Producción Animal, Facultad de Ciencias Agrarias, Universidad Nacional de Colombia - Sede Medellín, Colombia


A study was carried out to evaluate methane (CH4) emissions by lactating cows fed a diet that included 10% of a 50:50 Sambucus peruviana and Tithonia diversifolia silage in a diet based on Cenchrus clandestinus grass and concentrate. Four Holstein cows with an average weight of 509 ± 42.9 Kg and with 182 ± 21 days of lactation were used, in an over-change design. Animals were housed in polytunnels, they received grass ad libitum, while the silo and the concentrate were supplied twice a day. The two diets evaluated were 79%C. clandestinus, 21% concentrated (control) and 71%C. clandestinus, 19% concentrate and 10% silage (50:50 S. peruviana and T. diversifolia). Gas samples were obtained every hour for 24 h, in which the methane concentration was determined by gas chromatography. The silage diet had higher concentrations of protein, NDF and ADF and intake of DM and fermentable DM (Kg/d) was higher (P <0.05) with this diet (17.91 and 11.36) than with the control diet (15.64 and 9.64), respectively. There were no differences (P> 0.05) in methane production, being 13.64 and 14.78 g CH4/ Kg of DM consumed for the silage and control diets, respectively. Enteric methane production (ppm) was 519.3 for the silage diet and 487.6 for the control diet (P = 0.352), while the estimated Ym was 4.2 and 4.7, respectively. These results suggest that while the inclusion of S. peruviana and T. diversifolia silage leads to an increase in dry matter intake, it does not increase methane emissions per animal.

Keywords: GHG, in vivo, methane, polytunnel, silage, tree marigold


In the Colombian highland tropics, ruminant diets are based on relatively few forage species, generally grasses such as kikuyu (Cenchrus clandestinus) and ryegrass (Lolium multiflorum) grasses, of low dry matter (DM) availability at times of low rainfall and when frosts occur. In addition, management factors such as inadequate fertilization programs and poor grazing rotations, reduce the nutritional value of forages and their biomass availability, which interferes with the optimal performance of grazing cattle (Barahona and Sánchez 2005). The reduction in fodder availability during the summer, which can be 25% of the total forage availability during the rainy season, can be solved through the supply of silage (Ojeda 2000). However, in most cases, dietary silage inclusion does not result in improved nutritional value of the diet, since the silage is usually made from grasses with high fiber and low soluble carbohydrate concentrations. Additionally, grass silage elaboration in the north of Antioquia, is commonly hampered due to difficulties in chopping and because its moisture percentages generate leachates that reduce the nutrient content of the final product, together with higher ammonia emissions (<3%) than recommended at the time of silage opening (Villalobos and Arce 2016). The dietary inclusion of tree and shrub forages can lead to greater cattle productivity, as the numbers and activity of ruminal microorganisms are increased due to the greater rumen availability of ammoniacal nitrogen, amino acids and peptides (Preston and Leng 1990), greater biomass supply and lower dietary fiber concentration (Chará et al 2017, Rivera et al 2017). Currently, one of the major concerns regarding animal production is their emission of greenhouse gases (GHG) (Cuartas et al 2014; Molina et al 2016a), which has led to the search of practices that result in reduced farm CO2 footprint (Chirinda et al 2017). Thus, the productive and environmental impact of all feeding alternatives for cattle in tropical production systems must be evaluated. The present study was carried out to evaluate the impact of including a mixed silage based on two non-traditional forages in the diet of lactating cows in a specialized dairy system, with special attention to their production and quality of milk and their CH4 emissions.

Materials and methods


The study was carried out at Hacienda La Montaña farm of the University of Antioquia, San Pedro de los Milagros, Antioquia, Colombia, at 2350 m.a.s.l., with an average temperature of 15 °C and located in a low-mountain humid forest life zone according to the classification of Espinal (1997).

Animal description

Four Holstein, two to four parity, second-third of lactation cows with 182 ± 21 days in lactation, a daily milk production of 20.7 ± 2.2 liters and 509 ± 42.9 Kg of live weight were used. Cows were weighed both at the beginning and at the end of the experiment. Before the start of the trial, animals received the same diet and were mechanically milked twice daily (4:30 a.m. and 2:00 p.m.). A veterinarian was always available to guarantee the adequate health status of the animals. Additionally, throughout the experiment, to determine their degree of well-being the temperature and humidity index (ITH) and milk production of the cows was monitored daily.


A 50:50 Tithonia diversifolia and Sambucus peruviana silage was elaborated using forages harvested in the Cien Años de Soledad farm, Rionegro, Antioquia, Colombia, located in the Lower Mountain Humid Forest life zone (Espinal 1997), at 2,160 m.a.s.l., with an average annual temperature of 18 °C and an average relative humidity of 80%. Forages (T. diversifolia) and (S. peruviana) were harvested at a regrowth age of 90 days, ground to a particle size greater than 5 mm. During silage elaboration, a commercial inoculant with three strains of lactic acid bacteria and molasses diluted in water, in a 1:10 ratio was added to improve the fermentation process. Silage was opened after 30 days, repacked in bags and compacted again for its conservation until fed to the animals.

Two diets were offered: (1) 79% kikuyu grass plus 21% concentrate (Control) and, (2) 71% kikuyu grass, 19% concentrate and 10% of the Tithonia-Sambucus silage (Silage). Water was available ad libitum, and salt was offered as part of the concentrate. Two experimental periods were used, with diets exchanged between animals. Each experimental period consisted of 13 days, with 10 days of adaptation to the diet, two days of adaptation to the polytunnel and one day for sampling. The rest time between the two periods was five days.

Kikuyu grass was harvested daily at a regrowth age of 42 days from prairies fertilized with 50 Kg of N per hectare per year and was offered ad libitum using metal feeders with containers for the collection of refusals. The Tithonia-Sambucus silage was offered twice daily after each milking (8:00 and 16:00 hours). The offer of all diet components was calculated according to the nutrient requirements of the animals estimated by the Large Ruminant Nutrition System 1.0.33 (LRNS) program. To ensure a silage offer of at least 10% of the diet, 8.7 Kg of silage were offered daily. Likewise, to ensure an ad libitum grass intake, 80 Kg animal/day (considering a 10% refusal) of fresh kikuyu grass were offered.

Forage intake

Forage intake was measured daily throughout the experimental period as the difference between forage offered and rejected, taking a subsample of each forage for later analysis.

Laboratory analyses

At the end of the adaptation period, representative samples of offered forages and orts, previously ground through a 1 mm sieve in a Romer RAS mill (Romer Labs, Mexico), were analyzed in the bromatology laboratory of the Universidad Nacional de Colombia, Medellín campus. The analyzes included: crude protein (CP; Kjeldahl method, NTC 4657), NDF and ADF (Van Soest et al 1991 sequential technique) and ether extract by Soxhlet extraction (NTC 668). Ashes were determined by direct incineration (AOAC 942.05) and calcium (Ca) and phosphorus (P) concentration was determined by AA and U.V-VIS spectrophotometry (NTC 5151 and 4981), respectively.

Tannin concentration

Condensed tannins were determined in the Animal Nutrition and Feed Research Laboratory (NUTRILAB) (University Research Headquarters – SIU, University of Antioquia, Medellín, Colombia) using the butanol-HCl method (Terrill et al 1992, modified by Barahona 1999). Samples were extracted in triplicate by the addition of 70% acetone and diethyl ether. The supernatant, containing the soluble tannins, was obtained by centrifuging at 1800 RPM for 5 minutes. The solid residue was preserved for subsequent quantification of insoluble tannins. Tannins in both fractions were hydrolyzed with a 95-5% butanol-HCl solution, heating in a water bath at 95 °C for one hour, and the tannin concentration was determined in a spectrophotometer at 550 nm.

Milk production and composition

Total (morning and afternoon) milk production was measured daily throughout the experimental period using a Tru-Test milk meter. Milk composition was determined on the thirteenth day of each experimental period in samples collected with the same milk meter at the Milk Quality Laboratory of the Faculty of Agricultural Sciences of the University of Antioquia. Analyses included milk chemical composition (fat, protein, lactose, total solids and urea nitrogen in milk) determined by infrared spectroscopy and microbiological quality (somatic cell count) determined by flow cytometry.

Enteric methane emissions

This measurement consisted of a two-period crossover experiment each of 13 days. Two polytunnels, each with two separate compartments, so that animals were housed individually in a space with a total volume of 44.27 m3. Measured methane recovery rate was > 94%. Each structure had an entrance for animals and staff and a 12” hood equipped with an extractor with an extraction speed of 0.9 m2/s through which gas samples were obtained.

The recommendations of Murray et al (2001) were followed in these experiments. In each period, on days 11th and 12th, animals were adapted to the polytunnel enclosure by closing the tent for one hour and then leaving it open for two hours. On day 13th, gas samples were obtained at 70-minute intervals for a period of 24 hours, using a three-way valve connected to a 12 mL plastic syringe with which the air expelled by the extractor was captured after 10 seconds of the start of the extraction. Said valve was also attached to a hypodermic needle that was used to transfer the gas samples to 7 mL vacuum tubes. The air outside the tunnel (ambient) was sampled simultaneously. The collected gas samples were stored in a cool and dry place until they were taken to the laboratory for gas chromatography analysis. The liters of methane produced were calculated through the application of the ideal gas law as described by López and Newbold (2007), based on the concentration of methane (millimoles) and the total volume of each chamber of the polytunnel. To calculate methane emissions per Kg of degraded dry matter (DDM), in vitro dry matter digestibility values of 62% and 66% were used for the Control and Silage diets (Arias, unpublished data).

Photo 1. Schematic view and dimensions of the polytunnels used to
measure enteric methane emissions from Holstein cows
Temperature and relative humidity

When methane samples were collected, the temperature and relative humidity inside and outside the polytunnels were recorded with the help of a thermo hygrometer. Using these data, the temperature and humidity index (THI) (Thom 1959) modified by Kibler (1964) was calculated.

THI = 1.8*T – (1- (RH/100)) * (T-14.3) + 32

Where: THI: temperature and relative humidity index; T: temperature in °C; RH: relative humidity in %.

For the interpretation of this index, the Thom (1959) scale was used, which classifies the THI values as follows: 74 or less = normal, 75 - 78 = alert, 79 - 83 = danger and 83 or more = emergency.

Methane analysis

Methane concentration was determined in the Greenhouse Gas Analysis Laboratory of the International Center for Tropical Agriculture (CIAT), Palmira, Valle del Cauca. A Shimadzu GC-2014 gas chromatograph (Shimadzu, Japan) was used, equipped with a flame ionization detector and an electron capture detector (ECD). The chromatographic conditions were as follows: Hayesep N column 3 meters long; mobile phase: high purity nitrogen at a flow of 35 ml per minute. The oven, injector and detector temperatures were 250°C, 100°C and 325°C, respectively. The Scott methane standard used was mixed in nitrogen.

Statistical analysis

The data obtained for intake, milk production and composition and methane emissions were analyzed following a crossover design. The PROC GLM procedure of SAS® software, version 9.2 (SAS Institute Inc., Cary, NC, USA, 1989) was used. The model used was:

Yijk = μ + δ i + Aj + Pk + ҽijk

Where: Yijk: observations of the i-th diet for the j-th animal in the k-th period; μ: general average of the population;δ i: effect of the i-th diet; Aj: j-th animal effect; Pk: effect of the k-th period; ҽij: experimental error.


Diet nutritional quality

Although there was a similar NDF concentration among the two experimental diets (Table 1), ADF and lignin concentration was numerically greater in the Silage diet.

Table 1. Chemical composition of the feeds evaluated in this study




Protein, g/Kg




NDF g/Kg




ADF g/Kg




Lignin g/Kg




Fat g/Kg








Ash g/Kg




Calcium g/Kg




Phosphorus, g/Kg




Gross energy, mcal/Kg




The CP concentration was lower in the diet control, but the concentration of other nutrients was similar between both diets (Table 2). In turn, the silage tannin presence was 0.31 ± 0.1% and 1.82 ± 0.1% for soluble and bound tannins, respectively (Table 2).

Table 2. Chemical composition of a diet based on kikuyu grass (Cenchrus clandestinus) and commercial concentrate without (Control) and with (Silage) inclusion of T. diversifolia and S. peruviana silage



Control + Silage

Protein, %



NDF, %



ADF, %



Lignin, %



Fat, %









Calcium, %



Phosphorus, %



nergy, Mcal/Kg



Soluble tannins, %

0.31 ± 0.1

Bound tannins, %

1.82 ± 0.1

* Control: Kikuyu grass: concentrate (79:21);
Silage: Kikuyu grass: concentrate: silage (71: 19: 10)


The average as is kikuyu grass intake (Kg/day) of the control diet was 67.95 (Table 3), similar to what was observed for silage diet (66.9; P> 0.05). On the other hand, due to differences in silage intake, the total DMI was 1.18 times higher in the silage diet ( P <0.05). Thus, when offered the silage diet, cows DMI was 3.34% of live weight, compared to only 2.83% when cows were offered the control diet. Likewise, intake (g DM/Kg metabolic weight) was higher in the silage than in the control diet (158 and 134, respectively), with differences (P <0.05) in all DMI parameters. Regarding the intake of nutrients, lignin and energy (Table 4), there were differences for NDF, ADF, lignin, fat, ash, calcium and gross energy intakes, which were greater in the silage diet (P <0.05).

Table 3. Average as is and dry matter forage intake by lactating Holsteins cows receiving a diet based on kikuyu grass (Cenchrus clandestinus) and commercial concentrate without (Control) and with (Silage) inclusion of T. diversifolia and S. peruviana silage





Forage as is, Kg/d




Concentrate, Kg/d



Silage, Kg/d



Dry matter, Kg/d




Dry matter digested, Kg




Dry matter intake, % of live weight




Dry matter intake, g /Kg metabolic weight




Control: Kikuyu grass: concentrate (79:21); Silage: Kikuyu grass: concentrate: silage (71: 19: 10). a,b Means in the same column with different letters are statistically different according to Tukey's test (P < 0.05)

Temperature and relative humidity inside and outside the polytunnels

The highest temperature within the polytunnels (28.6°C) occurred at 14:20 hours, while the lowest temperature (11.9°C) occurred at 1:10 hours. Throughout the experiment, the polytunnel temperature was higher than that of the ambient and the greatest differences were observed between 8:25 and 16:45 hours, which on average were 5.6ºC higher in the polytunnels. There were marked contrasts in the internal and external polytunnel relative humidity, in nine of the 20 measurements, specifically between 8:00 am and 7:00 pm. The external relative humidity was lower (40.5%) at 3:35 pm and between 7:00 p.m. and 8:00 a.m., the relative humidity within the polytunnels was close to 98%, while the ambient relative humidity was 86%. According to the THI scale, 90.3% of the time in which the animals remained in the polytunnels, they faced a THI between “normal” and “alert”. Between 10:30 and 14:05 hours, the THI suggested that the animals were in caloric stress. However, diet intake and milk production data remained constant throughout the entire experimental period (data not shown), indicating that the animals did not suffer from thermal stress. Additionally, the animals did not lose weight, on the contrary, a daily gain of 571 g/d was observed during the experimental period.

Figure 1. Average (two experimental periods) of the temperature and humidity index (THI, %) observed
inside and outside (ambient) of the polytunnels during the 24-hour gas collection period
Milk production and composition

As can be seen in Table 4, none of the variables related to milk production and composition were affected by silage supply (P> 0.05), despite increases in diet DMI.

Table 4. Milk production and composition by lactating Holsteins cows receiving a diet based on kikuyu grass (Cenchrus clandestinus) and commercial concentrate without (Control) and with (Silage) inclusion of T. diversifolia and S. peruviana silage





Milk production, Kg/d


20.62 a


Milkfat, %

3.05 a

3.09 a


Protein, %

2.76 a

2.80 a


Total solids, %

11.70 a

11.22 a


Milkfat, Kg/d

0.63 a

0.63 a


Protein, Kg/d

0.57 a

0.57 a


Total solids, Kg/d


2.30 a


Control: Kikuyu grass: concentrate (79:21); Silage: Kikuyu grass: concentrate: silage (71: 19: 10).
a,b Means in the same column with different letters are statistically different according to Tukey's test ( p> 0.05)

Methane emissions

There were no significant differences in CH4 emissions (P> 0.05) among diets evaluated. On average, the daily emissions were 228 and 212.5 g for the silage and control diets, respectively (Table 5). When cows consumed silage, produced an average of 7% more methane per day, which was due to higher DMI. However, there were no significant difference for the daily methane emissions (Figure 2).

Table 5. Daily production of methane by lactating Holsteins cows receiving a diet based on kikuyu grass (Cenchrus clandestinus) and commercial concentrate without (Control) and with (Silage) inclusion of T. diversifolia and S. peruviana silage





Methane, g/d




Methane, g/Kg of DMI




Methane, g/Kg of digested DM




Methane, g/liter milk




Methane, g/Kg of milkfat




Methane, g/Kg of milk protein




Methane, g/Kg of milk total solids




Ym, %




Control: Kikuyu grass: concentrate (79:21); Silage: Kikuyu grass: concentrate: silage (71: 19: 10). a,b Means in the same column with different letters are statistically different according to Tukey's test (p < 0.05). Ym = Enteric methane emission as a percentage of the gross energy intake

Figure 2. Methane production accumulated in 24 h from lactating Holsteins cows receiving a diet based on kikuyu grass (Cenchrus clandestinus) and commercial concentrate without (Control) and with (Silage) inclusion of T. diversifolia and S. peruviana silage

The estimation of methane emissions per Kg of degraded dry matter (DDM) showed that the animals that consumed the silage diet emitted 20.04 g CH 4/Kg DDM, emitting 3.94 g CH4/Kg DDM less than those consuming the control diet, but without significant differences (P = 0.16). When methane production was expressed in g per liter of milk produced (Table 6), values were 10.95 and 10.24 for the silage and control diets, respectively. The percentage of gross energy lost in the form of methane (Ym) was 4.7 and 4.2% for the control and silage diet, respectively, without significant differences.


Diet nutritional quality

There was high protein concentration in both experimental diets. Excesses in dietary protein can compromise herd fertility and generate energy losses when excess nitrogen is excreted as (Butler 1998). Correa et al (2008) indicated that high (in excess) concentration of crude protein and NDF are among the most relevant nutritional limitations in kikuyu grass in Colombia. The NDF concentration of the silage diet (Table 1), although lower than that of the control diet, continued to be high compared to the recommended dietary NDF concentrations. The degradability of the NDF is very variable, due to changes in composition and structure, yet high dietary NDF leads to limited energy availability for ruminants (Buxton and Redfearn 1997), as more than 50% of the fiber may not be degraded (Cherney et al 1991). Apráez et al (2012) reported concentrations of 19.6% DM, 21.1% CP, 23.4% NDF and 15.8% ADF in S. nigra and in vitro, the forage of T. diversifolia and S. peruviana has been reported to show high degradability, which could lead to increased voluntary intake or animal productivity.

As milk production is associated to high dietary Ca requirements, it is important to highlight the higher concentrations of Ca in the silage diet associated to the dietary inclusion of T. diversifolía. Both dietary Ca and P levels were acceptable for dairy cows producing 20 liters of milk/day (Mahecha et al 2007). The tannin concentration of the silage was perhaps low to affect ruminal fermentation parameters. However, that depends on the type of tannin (molecular weight, anthocyanidin composition) (Barahona 1999), with some tannins being extremely efficient to bind to hydrolytic enzymes, even at low concentrations (Barahona et al 2006). Depending on their size and composition, tannins are bound to fiber or proteins, making them undetectable with conventional measurement techniques (Sepúlveda et al 2003) which could explain the low tannin concentrations observed in the silage.

Voluntary intake

The intake of DM was 1.18 times greater in the silage diet (Table 3). This agrees with the study of Mojica et al (2009), where the addition of silage to mid lactation cows increased their DMI by 1.11, being 21.9 ± 1.5 without silage and 24.3 ± 0.5 with the addition of oat silage. Also, Barahona et al (2003) reported increases in DMI of 1.04 and 1.07 times in diets with 25 and 50% oat silage, compared to diets with no silage. The intake of DM is closely related to the fiber concentration of the diet (Mertens 1994; NRC 2001; Forbes 2005), therefore by including T. diversifolia and S. peruviana silage, greater DMI would be expected, given the low fiber concentration in the silage. High dietary NDF favors rumen filling, increasing diet retention time, limiting DMI and energy availability. Physical factors, such as plant cell wall content, reduce DMI as fiber has less solubility, occupies more space in the digestive tract and slows down rumen degradation (Barahona and Sánchez 2005). There are reports of high degradability of T. diversifolia forage (Rosales 1996), which increases the rate of passage and voluntary DMI.

Different reports state that DMI, which is the equivalent to IDM (incubated DM) in in vitro studies, is the main factor that influences the production of enteric CH4 (Buddle et al 2011). Blaxter and Clapperton (1965) when evaluating the relationship between DMI and CH 4 emissions, reported a close relationship between DMI and diet digestibility, with greater CH4 production in low digestibility diets when DMI was at a maintenance level. On the contrary, in diets with high digestibility and DMI at three times the maintenance level, there was a 30% reduction in the production of CH4 per Kg of DMI (Blaxter and Clapperton, 1965). Greater DMI is related to higher rate of passage and to lower diet rumen degradability, which results in lower methane production (Pinares-Patiño et al 2003). Recent research has suggested that an increase in DMI in fodder-rich diets decreases methane emissions per unit of dry matter consumed (11 to 14%) (Sun et al 2011, Hammond et al 2011). However, this reduction is not as large as that reported by Blaxter and Clapperton (1965), probably due to the low variability in chemical composition and digestibility of the forages evaluated. The relationship between voluntary intake and emission of methane per unit of intake, suggests that all those practices associated with the increase in forage DMI would potentially be conducive to lower methane production per unit of DMI (Vargas et al 2012). When expressing DMI in relation to live weight, there was significant difference (P <0.05) between treatments, with silage addition increasing DMI 1.18 times. Similar results were reported by Molina et al (2016b), who found that the inclusion of T. diversifolia increased DMI 1.26 times in lactating cows, a response also reported by Mahecha et al (2007) for Holstein cows.

Temperature y relative humidity

Overall, both the temperature and the relative humidity within the polytunnels were within normal parameters, allowing adequate welfare for the animals and ensuring the reliability in these in vivo methane measurements, as animals were in their comfort zone. This was ratified with the DMI and milk production data. Similar observations were reported by Molina et al 2016a in polytunnel experiments. The THI allows to measure the caloric stress of cattle (Hahn 1999). Armstrong (1994) suggested that cattle are in comfort zone with a THI below 71, with mild stress between 72 and 79, with moderate stress between 80 and 89 and with severe stress with a THI above 90. Huhnke et al (2001) suggested that a THI between 79 and 83 is a dangerous situation, and when it exceeds 84 there is an emergency. Other authors reported lower thresholds of THI in which there is stress for dairy cows (<72; Collier et al 2011).

Milk production and composition

Milk production and composition was not affected by silage intake (Table 5), nor by the greater DMI in the silage diet (Table 3). Mahecha et al (2007) reported that in Holstein x Cebu animals supplemented with forage of T. diversifolia as a partial replacement of 35% of the concentrate feed, there were no changes in milk production and composition. However, in interpreting the current results, it is necessary to consider that short window of observation limited observing other variables reliably, such as changes in body weight or in reproductive responses, so it is advisable to include these considerations in future studies with silage of T. diversifolia and S. peruviana.

Methane emissions

The methane emissions observed in this study were like those reported by Noguera and Posada (2017), who using respirometry chambers, reported emissions of 287g CH4/d in Holstein cows weighing 541 ± 30 Kg, with 141 ± 34 days in milk and producing 26.9 ± 2 liters of milk/day. However, several authors have reported greater daily methane emissions (g/d) in dairy cows than those reported in the present study. Grainger et al (2007) reported methane productions of 318 ± 54 g/d in lactating cows producing 22.3 ± 3.78 L of milk and consuming 18 ± 3.2 Kg of DM of a diet of 70% grass-30% concentrate. Brask et al (2015) reported a methane production of 382 ± 72 g/d in lactating cows consuming 18 ± 2.5 Kg DM of grass or corn silage. In turn, Bell et al. (2017), reported that in lactating cows consuming a diet composed of 60-80% grass and 20-40% concentrate, methane production of 379, 366, 378 and 363 g/d for animals consuming 17.1, 15.7, 17.2 and 15.6 Kg of DM. In the present study, the increase in DMI observed with the inclusion of silage was not associated with increases in methane emissions (Table 6). However, a positive correlation (between 37 and 77%) between DMI and methane emission animal/d has been reported (Lassey et al 2001, Pinares-Patiño and Clark 2008). The current results suggest that, although there is a positive relationship between DMI and methane emissions animal/d, that depends on the diet consumed. Thus, although animals receiving the silage diet consumed 1.18 times more DM (Table 3), they produced only 0.89 times the methane per Kg of DMI (Table 6) than animals consuming the control diet. This may be associated with changes in the rumen microbial population. Galindo et al (2011) reported that the dietary inclusion of 10 and 20% T. diversifolia reduced rumen methanogen populations. On the other hand, with high degradability forages, DMI can be increased while at the same time, reducing CH4 emission per unit of ingested DM (Chaves et al 2008).

The presence of secondary metabolites in S. peruviana could help reduce enteric methane emissions. However, the presence of secondary metabolite contents in T. diversifolia is not high (Mahecha and Rosales 2005), which is beneficial, as neither DMI nor DM digestibility is limited (Rivera et al 2018). Given the nutritional characteristics of S. peruviana, and its chemical composition, this appears as a promising forage resource. However, there is little nutritional information in the literature for this forage. For comparison purposes, studies with S. nigra (elderberry) are included, which includes the subspecies S. peruviana and S. mexicana (Díaz 2003). In S. nigra, the presence of volatile oils and phytosterols (Cruz et al 2011), mucilage, tannins, vitamins A and C, glycosides, cyanogens, viburic acid and alkaloids has been reported (Ody 1993). The loss of gross energy in the form of methane (Ym) found in this work is lower than that reported in several Latin American countries. In Brazil, Pedreira et al (2009) reported values of 5.8 and 6.6% for Holstein cows and their crosses, respectively. In Uruguay, Dini et al (2012) reported values of 6.4 and 6.7% in Holstein dairy cows grazing in pastures rich in legumes and in grasses, respectively. However, in the case of North America, Mangino et al (2010), reported values between 4.8 and 5.8% for dairy cows in different regions of the United States. Likewise, Noguera and Posada (2017) reported Ym values of 4.9% for Holstein cows in northern Antioquia. A lower value of Ym is related to greater efficiency in the use of gross energy and lower methane emissions.


There was an increase in DM and nutrient intake associated with the inclusion of T. diversifolia and S. peruviana silage. However, there were no increases in milk production, nor changes in its composition. The inclusion of silage did not affect methane emissions, despite increased DMI. It is recommended to further evaluate the use of different proportions of T. diversifolia and S. peruviana in the preparation of silage and of different levels of inclusion of such silages in the diet.


AOAC Official Method 2005 Determination of Ash in Animal Feed. In Official Methods of Analysis of AOAC International. AOAC International. Gaithersburg, MD (USA).

Apráez, J E Delgado J M and Narváez J P 2012 Composición nutricional, degradación in vitro y potencial de producción de gas, de herbáceas y arbóreas Livestock Research for Rural Development 24(3): 1-11.

Armstrong D 1994 Heat stress interaction with shade and cooling Journal of dairy science 77(7): 2044-2050.

Barahona R 1999 Condensed tannins in tropical forage legumes: their characterization and study of their nutritional impact from the standpoint of structure-activity relationships PhD Diss University of Reading

Barahona R Cuesta P and Báez F 2003 Producción y evaluación del uso de ensilajes en sistemas de producción de lechería especializada de Nariño. Renovación y manejo de pradera y utilización de ensilajes en el trópico alto Plan de modernización de la ganadería bovina Colombiana 49-56

Barahona Rosales R and Sánchez Pinzón S 2005 Limitaciones físicas y químicas de la digestibilidad de pastos tropicales y estrategias para aumentarla Ciencia y Tecnología Agropecuaria 6(1): 69-82.

Barahona R Sánchez S Lascano C E Owen E Morris P and Theodorou M K 2006 Effect of condensed tannins from tropical legumes on the activity of fibrolytic enzymes from the rumen fungus Neocallimastyx hurleyensis Enzyme and microbial technology 39(2): 281-288.

Bell R A Darling W G Ward R S Basava-Reddi L Halwa L Manamsa K and Dochartaigh B Ó 2017 A baseline survey of dissolved methane in aquifers of Great Britain Science of the Total Environment 601, 1803-1813.

Brask M Weisbjerg M R Hellwing A L Bannink A and Lund P 2015 Methane production and diurnal variation measured in dairy cows and predicted from fermentation pattern and nutrient or carbon flow Animal 9(11): 1795-1806.

Blaxter K L and Clapperton J L 1965 Prediction of the amount of methane produced by ruminants British Journal of nutrition 19(1): 511-522.

Buddle B M Denis M Attwood G T Altermann E Janssen P H Ronimus R S and Wedlock D N 2011 Strategies to reduce methane emissions from farmed ruminants grazing on pasture The Veterinary Journal 188(1): 11-17.

Butler W R 1998 Effect of protein nutrition on ovarian and uterine physiology Journal of Dairy Science 81: 2533-2539

Buxton D R and Redfearn D D 1997 Plant limitations to fiber digestion and utilization The Journal of Nutrition, 127(5): 814S-818S.

Chará J Rivera J Barahona R Murgueitio E Deblitz C Reyes E and Zuluaga A 2017 Intensive silvopastoral systems: economics and contribution to climate change mitigation and public policies. In Integrating Landscapes: Agroforestry for Biodiversity Conservation and Food Sovereignty (pp. 395-416). Springer, Cham.

Chaves A V He M L Yang W Z Hristov A N McAllister T A and Benchaar C 2008 Effects of essential oils on proteolytic, deaminative and methanogenic activities of mixed ruminal bacteria Canadian Journal of Animal Science 88(1): 117-122.

Cherney J H Cherney D J R Akin D E and Axtell J D 1991 Potential of brown-midrib, low-lignin mutants for improving forage quality In Advances in Agronomy 46:157-198

Chirinda N Arenas, L Loaiza S Trujillo C Katto M Chaparro P and Barahona R 2017 Novel technological and management options for accelerating transformational changes in rice and livestock systems Sustainability 9(11): 1891.

Collier R J Zimbelman R B Rhoads R P Rhoads M L and Baumgard L H 2011 A re-evaluation of the impact of temperature humidity index (THI) and black globe humidity index (BGHI) on milk production in high producing dairy cows Western Dairy Management Conference 113-125

Correa H J Pabón M L and Carulla J E 2008 Valor nutricional del pasto kikuyo (Pennisetum clandestinum Hoechst Ex Chiov.) para la producción de leche en Colombia (Una revisión): I-Composición química y digestibilidad ruminal y posruminal Livestock Research for Rural Development 20(4)

Cruz A Rodríguez C and Ortiz C 2011 Efecto insecticida in vitro del extracto etanólico de algunas plantas sobre la mosca adulta Haematobia irritans Revista Cubana de Planta Medicinales 16(3):216-226

Cuartas Cardona C A Naranjo Ramírez J F Tarazona Morales A M Murgueitio Restrepo E Chará Orozco J D Ku Vera J and Barahona Rosales R 2014 Contribution of intensive silvopastoral systems to animal performance and to adaptation and mitigation of climate change Revista Colombiana de Ciencias Pecuarias 27(2):76-94.

Diaz J 2003 Informe Técnico. Caracterización del mercado Colombiano de plantas medicinales y aromáticas. Instituto Alexander Von Humboldt Ministerio de Ambiente Vivienda y Desarrollo Territorial

Dini Y Gere J Briano C Manetti M Juliarena P Picasso V... Astigarraga L 2012 Methane emission and milk production of dairy cows grazing pastures rich in legumes or rich in grasses in Uruguay. Animals, 2(2), 288-300.

Espinal L S 1997 Zonas de vida o formación vegetal de Colombia: Memoria explicativa sobre el mapa ecológico (Vol. 13). Bogotá: Instituto Geografico" Agustin Codazzi.

Forbes J M 2005 Voluntary Feed Intake and Diet Selection. In: Dijkstra J, Forbes JM, France J. 2nd ed. Quantitative aspects of ruminant digestion and metabolism. CABI Publishing. p. 607-627

Galindo J González N Sosa A Ruíz T Torres V Aldana A I and Noda A C 2011 Efecto de Tithonia diversifolia (Hemsl.) Gray (Botón de oro) en la población de protozoos y metanógenos ruminales en condiciones in vitro Revista Cubana de Ciencia Agrícola 45(1)

Grainger C Clarke T McGinn S M Auldist M J Beauchemin K A Hannah M C and Eckard R J 2007 Methane emissions from dairy cows measured using the sulfur hexafluoride (SF6) tracer and chamber techniques Journal of dairy science 90(6): 2755-2766.

Hahn G L 1999 Dynamic responses of cattle to thermal heat loads Journal of animal science 77(1): 10-20.

Hammond K J Hoskin S O Burke J L Waghorn G C Koolaard J P and Muetzel S 2011 Effects of feeding fresh white clover (Trifolium repens) or perennial ryegrass (Lolium perenne) on enteric methane emissions from sheep Animal Feed Science and Technology 166, 398-404.

Huhnke R L McCowan L C Meraz L C Harp S L and Payton M E 2001 Determining the frequency and duration of elevated temperature-humidity index. ASAE Annual Internacional Meeting, Sacramento, CA. America Society of Agricultural and Biological Engineer, St. Joseph, MI.

Instituto Colombiano de Normas Técnicas y Certificación ICONTEC 1973 NTC 668. Alimentos y materias primas. Determinación de los contenidos de grasa y fibra cruda. Bogotá, Colombia.

Instituto Colombiano de Normas Técnicas y Certificación ICONTEC 2001 NTC 4981. Alimentos para animales. Determinación del contenido de fosforo. Método espectrofotométrico. Bogotá, Colombia.

Instituto Colombiano de Normas Técnicas y Certificación ICONTEC 2003 NTC 5151. Alimento para animales. Determinación de los contenidos de Calcio, Cobre, Hierro, Magnesio, Manganeso, Potasio, Sodio y Zinc. Método usando espectrometría de absorción atómica. Bogotá, Colombia.

Instituto Colombiano de Normas Técnicas y Certificación ICONTEC 1999 NTC 4657. Alimento para animales. Determinación del contenido de nitrógeno y cálculo del contenido de proteína cruda. Método Kjeldahl. Bogotá, Colombia.

Kibler H H 1964 Environmental physiology and shelter engineering. LXVII. Thermal effects of various temperature-humidity combinations on Holstein cattle as measured by eight physiological responses. Res. Bull. Missouri Agric. Exp. Station, 862.

Lassey K R Walker C F McMillan A M and Ulyatt M J 2001 On the performance of SF6 permeation tubes used in determining methane emission from grazing livestock Chemosphere-Global Change Science 3(4):367-376.

López S and Newbold C J 2007 Analysis of methane. Measuring methane production from ruminants 1-13

Mahecha L and Rosales M 2005 Valor nutricional del follaje de botón de oro Tithonia diversifolia (Hemsl.) Gray, en la producción animal en el trópico Livestock Research for Rural Development 17(9), 1

Mahecha L Escobar J P Suárez J F and Restrepo L F 2007 Tithonia diversifolia (hemsl.) Gray (botón de oro) como suplemento forrajero de vacas F1 (Holstein por Cebú) Livestock Research for Rural Development 19(2): 1-6.

Mangino M J Peterson M K and Jacobs H 2010 Development of an emissions model to estimate methane from enteric fermentation in cattle. In Proceedings of the 12th international emission inventory conference on emission inventories: applying new technologies (pp. 11-21)

Mertens D R 1994 Regulation of forage intake. Forage quality, evaluation, and utilization, (foragequalityev) 450-493.

Mojica J E Castro E León J M Cárdenas E A Pabón M and Carulla J E 2009 Efecto de la oferta de pasto kikuyo y ensilaje de avena sobre la producción y calidad composicional de la leche bovina Ciencia y Tecnología Agropecuaria 10(1):81-90.

Molina I C Angarita E A Mayorga O L Chará J and Barahona-Rosales R 2016a Effect of Leucaena leucocephala on methane production of Lucerna heifers fed a diet based on Cynodon plectostachyus Livestock Science 185: 24-29.

Molina I C Donney´s G Montoya S Villegas G Rivera J E Lopera J J Chará J and Barahona R 2016b Emisiones in vivo de metano en sistemas de producción con y sin inclusión de Tithonia diversifolia. 3° Congreso Nacional de Sistemas Silvopastoriles - VIII Congreso Internacional de Sistemas Agroforestales.

Murray P J Gill E Balsdon S L and Jarvis S C 2001 A comparison of methane emissions from sheep grazing pastures with differing management intensities. Nutrient Cycling in Agroecosystems 60(1-3): 93-97.

Noguera R R and Posada S L 2017 Factor de emisión de metano entérico para vacas Holstein lactantes en la zona norte de Antioquia-Colombia Livestock Research for Rural Development 29: 6.

NRC 2001 National Research Council. The nutrient requirement of dairy cattle. Seventh edition. National Academy Press, Washington. 381.

Ody P 1993 The Herb Society Las Plantas Medicinales. Guía práctica con remedios útiles para los trastornos más comunes. Buenos Aires, Argentina. Javier Vergara Editor s.a. 191p

Ojeda F 2000 La conservación como ensilaje en zonas tropicales. Estación experimental de Pastos y Forrajes Indio Hatuey. Cuba. Pastos y Forrajes 97.

Pedreira M Primavesi O Lima M A Frighetto R Oliveira S G and Berchielli T T 2009 Ruminal methane emission by dairy cattle in Southeast Brazil Scientia Agricola 66 (6): 742-750

Pinares-Patiño C S and Clark H 2008 Reliability of the sulfur hexafluoride tracer technique for methane emission measurement from individual animals: an overview Australian Journal of Experimental Agriculture 48(2): 223-229.

Pinares-Patino C S Ulyatt M J Waghorn G C Lassey K R Barry T N Holmes C W Johnson D E 2003 Methane emission by alpaca and sheep fed on lucerne hay or grazed on pastures of perennial ryegrass/white clover or birdsfoot trefoil. The Journal of Agricultural Science, 140(2): 215-226.

Preston T R and Leng R A 1990 Producción pecuaria tropical. Círculo de impresores Ltda., Cali, Colombia.

Rivera-Herrera J E Molina-Botero I Chará-Orozco J Murgueitio-Restrepo E and Barahona-Rosales R 2017 Sistemas silvopastoriles intensivos con Leucaena leucocephala (Lam.) de Wit: alternativa productiva en el trópico ante el cambio climático Pastos y Forrajes 40(3): 171-183.

Rivera J E Chará J Gómez-Leyva J F Ruíz T and Barahona R 2018 Variabilidad fenotípica y composición fitoquímica deTithonia diversifolia A. Gray para la producción animal sostenible Livestock Research for Rural Development 30:12.

Rosales M 1996 In vitro assessment of the nutritive value of mixtures of leaves from tropical fodder trees. PhD Thesis Departament of Plant Sciences Oxford University Oxford, UK. 214 pp

SAS 1989 Statistical Analysis System Institute Inc. SAS/STAT user’s guide version 6, vol. 2, 4th ed. Cary, NC: SAS Institute Inc.; 1989.

Sepúlveda Jiménez G Porta Ducoing H and Rocha Sosa M 2003 La participación de los metabolitos secundarios en la defensa de las plantas Revista mexicana de fitopatología 21(3).

Sun X Z Hoskin S O Muetzel S Molano G and Clark H 2011 Effects of forage chicory (Cichorium intybus) and perennial ryegrass (Lolium perenne) on methane emissions in vitro and from sheep Animal Feed Science and Technology 166: 391-397.

Thom E C 1959 The discomfort index Weatherwise 12: 57-59.

Van Soest P J Robertson J B and Lewis B A 1991 Methods for dietary fiber neutral detergent fiber and nonstarch polysaccharides in relation to animal nutrition Journal Dairy Science 74(10): 3583-3597

Vargas J Cárdenas E Pabón M and Carulla J 2012 Emisión de metano entérico en rumiantes en pastoreo Archivos de Zootecnia 61(237): 51-66.

Villalobos L and Arce J 2016 Efecto del picado sobre las características nutricionales y fermentativas de ensilajes de pastos kikuyo, ryegrass perenne y alpiste forrajero Agronomía costarricense 40(1): 65-74.