Livestock Research for Rural Development 34 (5) 2022 LRRD Search LRRD Misssion Guide for preparation of papers LRRD Newsletter

Citation of this paper

In vitro methane production and ruminal fermentation parameters of tropical grasses and grass-legume associations commonly used for cattle feeding in the tropics

Xiomara Gaviria Uribe1,2, Diana M Bolívar Vergara1, Ngonidzashe Chirinda2,4, Isabel Cristina Molina-Botero3, Johanna Mazabel2, Rolando Barahona Rosales1 and Jacobo Arango2

1 Animal Production Sciences Department, Universidad Nacional de Colombia. Street 59 A # 63-20, Medellin, Antioquia, Colombia. C.P. 050034. Research Group BIOGEM
xgaviri0@unal.edu.co
2 Tropical Forages and Soil Programs of the International Center for Tropical Agriculture (CIAT). Km 17 Recta Cali-Palmira, Valle del Cauca, Colombia. C.P. 763537
3 Department of Nutrition, Faculty of Animal Science, Universidad Nacional Agraria La Molina, Av. La Molina s/n, La Molina, Lima, Perú, C.P. 12056
4 Mohammed VI Polytechnic University (UM6P), AgroBioSciences (AgBS), Agricultural Innovations and Technology Transfer Centre (AITTC), Benguerir, Morocco

Abstract

in vitro study was carried out to measure methane (CH4) production and ruminal fermentation parameters of tropical forages either commonly used and with a potential for inclusion in Colombian livestock systems. The forages evaluated wereUrochloa hybrid cv. Cayman, Leucaena leucocephala, Leucaena diversifolia, Megathyrsus maximus cv. Mombaza, Urochloa brizantha cv. Toledo, Canavalia brasiliensis,Urochloa decumbens, Tithonia diversifolia and Dichantium aristatum. which were incubated using the in vitro gas technique for 96 h. Treatments with higher neutral detergent fiber (NDF) and crude protein (PC) contents had higher gas production, dry matter (DM) degradability and the highest CH4 production (ml/g DMd) at 24 hours. Methane at 24 hours of incubation varied between 4.69 and 8.10 ml and increased by 43% on average at 48 hours. In all treatments, the highest proportion of volatile fatty acid (VFAs) corresponded to acetate, which was more than 50% of the total VFAs produced. Treatments with Urochloa hybrid cv. Cayman and their associations with Leucaena diversifolia had the lowest CH4 production values at 24 h. Similarly, treatments with Cayman grass, its associations with Leucaena and Toledo grass alone showed the highest DM degradability values. In conclusion, the inclusion of Leucaena and Tithonia diversifolia on a diet based on forage grass (Cayman or Toledo) had a positive effect on nutrient content and degradability and the group of treatments that included Cayman grass and its associations had lower CH4 production values and higher degradability than the rest of the treatments.

Keywords: enteric fermentation, greenhouse gasses, in vitro gas production technique, ruminants


Resumen

Se realizó un estudio in vitro para medir la producción de metano (CH4) y algunos parámetros de fermentación ruminal de forrajes tropicales utilizados comúnmente o con potencial de uso en los sistemas ganaderos colombianos. Los forrajes evaluados fueron Urochloa hybrid cv. CaymanLeucaena leucocephala, Leucaena diversifolia, Megathyrsus maximus cv. Mombaza, Urochloa brizantha cv. Toledo, Canavalia brasiliensis,Urochloa decumbens, Tithonia diversifolia and Dichantium aristatum, los cuales fueron incubados usando la técnica de gas in vitro durante 96 h. Se encontró que los tratamientos con mayor contenido de fibra en detergente neutro (FDN) y proteína cruda (PC) tuvieron mayor producción de gas, degradabilidad de la materia seca (MS) y la mayor producción de CH4 (ml/g MSd) a las 24 horas. La producción de CH4 después de 24 horas de incubación varió entre 4.69 y 8.10 ml (ml/g MS) y aumentó un 43% en promedio a las 48 horas. En todos los tratamientos, la mayor proporción de ácidos grasos volátiles (AGVs) correspondió al acetato, que fue más del 50% del total de AGVs producidos. Los tratamientos con el híbrido de Urochloa cv. Caimán y sus asociaciones con Leucaena diversifolia presentaron los valores más bajos de producción de CH4 a las 24 h. Asimismo, los tratamientos con pasto Caimán, sus asociaciones con Leucaena y el pasto Toledo mostraron los valores más altos de degradabilidad de la MS. En conclusión, la inclusión de Leucaena y Tithonia diversifolia en una dieta basada en pasto (Caimán o Toledo) tuvo un efecto positivo en el contenido de nutrientes y en la degradabilidad.


Introduction

Livestock production in the low-land tropics of Colombia (below 1.500 m.a.s.l) is based on the grazing of natural and introduced pastures without supplementation of concentrates or other types of forage (Ibrahim et al 2007). As a result, there is generally a low intake of nutrients and energy (Barahona and Sánchez 2005), which leads to low production efficiency and high negative impacts on the environment (Rivera et al 2017).

Global population growth and climate change have generated intense pressure and bad publicity for livestock production in recent years, primarily due to increased demand for livestock products and increased methane (CH4) emissions from cattle, mainly of enteric origin. However, while it is currently impossible to de-couple livestock production with CH4 emissions, several feed-based mitigation options have been suggested (de Souza et al 2021; Arango et al 2020). Although CH4 production may be higher under grazing and animals receiving no concentrate supplementation, this feeding system is attractive to farms due to low production costs. However, the use of high nutritional value fodder to reduce greenhouse gas (GHG) emissions per unit product (i.e., meat or milk), is attractive for production systems in the low-land tropics (Ku-Vera et al 2020).

Quantification of the benefits of mitigation actions requires reliable estimates of GHG emissions from enteric fermentation. However, most of the default Tier 1 emission factors (IPCC 2019) used to estimate emissions were developed with breeds and feeds in temperate countries, and there are huge uncertainties associated with their use in the tropics. Despite recent revisions of such factors (Gavrilova et al 2019), it remains unclear whether these emission factors reflect CH 4 emissions from cattle across different tropical locations. Thus, several developing countries are now aiming to develop local emission factors that will reduce uncertainties in GHG emission inventories and will also bring clarity on Nationally Determined Contributions (NDC). Yet, few facilities are available, and the costs are often beyond available analytical infrastructure, financial and personnel resources despite efforts to develop low-cost analytical methodologies (Gaviria-Uribe et al 2020a).

To reduce the uncertainty of emission estimates it might be necessary to focus research efforts on studying the nutritional value of feeds and in evaluating possible associations between increased quality of cattle diets, increasing productivity and GHG emissions. Therefore, in this study, we aimed to quantify in vitro CH4 production and ruminal fermentation parameters of several commonly used pastures and forages with potential for widespread use in livestock production systems in Colombia.


Materials and methods

Treatments

The following forages were evaluated in this study: Urochloa hybrid cv. Cayman-CIAT BR02/1752, Leucaena leucocephala, Leucaena diversifolia, Megathyrsus maximus cv. Mombaza, Urochloa brizantha cv. Toledo,Canavalia brasiliensis, Urochloa decumbens, Tithonia diversifolia, Dichantium aristatum. The composition of the treatments and the percentage of inclusion (base dry matter- DM) of each forage are shown in Table 1. The inclusion of each forage corresponds to the proportion of intake as fresh weight basis calculated in previous in vivo experiments (Gaviria-Uribe et al 2020b).

Table 1. Forages and percentage of inclusion of the treatments evaluated. For each forage, the percentage of inclusion is followed by the days of regrowth before harvest

Abbreviation

Treatment

Ca1

Urochloa hybrid cv. Cayman (100%,65 days)

Ca2Ll

Urochloa hybrid cv. Cayman (77.5% 45 days) +Leucaena leucocephala(22.5%, 58 days)

Ca2

Urochloa hybrid cv. Cayman (100%, 45 days)

Ca2Ld

Urochloa hybrid cv. Cayman (81.8%, 45 days) +Leucaena diversifolia(18.2%, 58 days)

Mo

Megathyrsus maximus cv. Mombaza (100%,46 days)

MoLl

Megathyrsus maximus cv. Mombaza (74%, 40 days) +Leucaena leucocephala(26%, 58 days)

To

Urochloa brizantha cv. Toledo (100%, 42 days)

ToLdCn

Urochloa brizantha cv. Toledo (74.2%, 39 days) +Leucaena diversifolia(18%, 59 days) +Canavalia brasiliensis(7.8%,59 days)

De

Urochloa decumbens (100%, 47 days)

DeTd

Urochloa decumbens (78%, 47 days) + Tithonia diversifolia (22%, 66 days)

HA

Dichantium aristatum hay (100%, 52 days)

Chemical composition determination of forages

Forage samples were collected from experimental plots at the Tropical Forages Program of the Alliance of Bioversity International and the International Center for Tropical Agriculture (CIAT) located in Palmira, Valle del Cauca, Colombia (3°30’7’N; 76°21’22’W) at 1.050 m.a.s.l. This site has an annual mean temperature of 27°C and rainfall of 1008 mm. The chemical analyses and in vitro trials were conducted in the Forage Quality and Animal Nutrition laboratory of CIAT. Dry matter content was determined utilizing the thermogravimetric method by drying at 105°C for 48 h (ISO method 6496) (ISO 1999); Concentrations of neutral detergent fiber (NDF) and acid detergent fiber (ADF) were measured sequentially, according to operating instructions, using an ANKOM 2000 fiber analyzer (Ankom Technology 2011) and according to the methods of Van Soest and Robertson (1985). Crude protein (CP) was determined using the Kjeldahl method (AOAC 984.13; 1990) and gross energy (GE) content by the bomb calorimeter method ISO 9831(1998). The following equation was used to calculate the metabolizable energy (ME). , where: is organic matter digested at 96 hours (Lindgren 1983).

In vitro gas production

An in vitro gas production procedure was carried out using the technique described by Theodorou et al (1994). The forage samples were dried and ground to pass a 1mm sieve and stored for later analysis. Samples were accurately weighed (1 g) into 110ml glass flasks. Four repetitions for each sample were incubated, and additionally, twelve flasks containing only the rumen fluid/buffer solution (blanks) were included for corrections of gas production.

The culture medium used in trials consisted of buffer solution, macro-mineral solution, micro-mineral solution, reductive solution and resazurin (Goering and Van Soest 1970). The solutions of the culture medium were prepared the day before the experiment and were mixed in the following order: distilled water, buffer solution, macro-mineral solution, micromineral solution and indicator solution (Menke and Steingass 1988). Rumen fluid collected from three rumen-cannulated brahman steers was filtered, mixed and transferred into pre-warmed thermos flasks for transfer to the laboratory; 10 ml of this fluid was mixed with 89 ml of culture medium at 39°C under constant stirring and continuous CO2 flow.

Gas production was recorded at 3, 6, 9, 12, 24, 36, 48, 60, 72 and 96 hours of incubation using a pressure transducer (Lutron Electronic Enterprise Co. Ltd., Taipei, Taiwan) connected to a digital wide-range manometer (Sper Scientific, Arizona, USA). To convert the pressure values obtained in psi into units of volume (ml), the equation from the Laboratory of Forage Quality and Animal Nutrition of the International Center for Tropical Agriculture: Y = 0.49x - 1.51; where: Y= Volume in ml and x = Pressure in psi was used.

Dry matter degradation

The dry matter in vitro degradability (DMd) was determined at different time points by weighing the residue recovered from the fermentation. First, bottles were removed after 24, 48 and 96 hours of incubation and cooled 4ºC to stop the fermentation process. The contents of each bottle were then filtered and dried for 48 h at 65ºC in an oven with forced air circulation (Memmert® UF 750, Schwabach, Germany). Finally, the final DM content of each bottle was weighed on an analytical balance (Mettler Toledo®, USA). Degradability was calculated by the difference between the initial and final DM content (undegraded DM) and expressed as a percentage of the initial DM.

Determination of volatile fatty acids (VFAs)

The concentration of acetic, propionic, butyric and isobutyric acid was determined for each treatment; samples of the effluent of each bottle were taken from the in vitro gas production at 24 and 48 hours after incubation and were deposited in 2 ml eppendorf® vials. A deproteinization and acidifying solution was added (10% metaphosphoric acid). The obtained solution was centrifuged at 13.000 rpm for 12 minutes at 4°C, and the supernatant was placed on a vial and stored at 4°C for later analysis by Shimadzu® ultrafast liquid chromatograph (Prominence UFLC, 20 series), equipped with an UV/Vis detector (SPD-20AV) and a 300mm x 7.8mm BIO-RAD Aminex HPX-87H column.

Methane quantification

Methane concentration was determined from the gas samples accumulated on each incubated bottle at 24 and 48 h post-incubation. The gas samples were stored in 5-ml glass vials under vacuum, until their subsequent concentration analysis at the Greenhouse Gas Laboratory of CIAT using a Shimadzu GC-2014 gas chromatograph (Shimadzu GC-2014, Shimadzu®, Japan), equipped with a flame ionization detector and an electron capture detector.

Data analysis

The Gompertz non-linear model was used to describe the dynamic of cumulative gas production over time (Casas et al 2010). The curves were adjusted with CurveExpert professional 2.0.0. program.

Y=a*exp(-exp(b-(c*x))

Where:

Y= accumulated gas production over time x

a > 0: maximum gas production

b > 0: difference between the initial gas and the final gas in a time x

c > 0: specific gas accumulation rate

Using a, b and c, the following parameters were calculated to give practical explanation and biological meaning to the data:

TIP - Time at inflection point (h) =b/c

GPI - Gas to inflection point (ml) =a/e*

MGPR - Maximum gas production rate (ml/h) = (a*c)/e*

PL – Lag phase (h) = (b/c) - (1/c)

*The value of e or exp corresponds to Euler number » 2.7183

Variables generated (accumulated gas production, DMd, TIP, GPI, MGPR, PL, VFAs and CH4) were analyzed with a completely randomized design with 11 treatments and four repetitions. The comparison of means was made by testing Tukey using SAS® software, version 9.1 (SAS Institute Inc., Cary, NC, USA, 2008). The model used was the following:

Yij = μ + Mi + ҽij

Where:

Yij: jth variable response of the ith treatment

μ: General average of treatments

Mi: Difference between the average of the ith treatment and the general average

ҽij: experimental error


Results

Chemical composition of treatments

The nutrient content for each evaluated treatment is shown in Table 2. The observed nutrient content of the treatments with only grass is typical for low-tropical grasses in Colombia, which are generally low in protein and high in NDF contents. Protein content in grass-only treatments varied between 39.4 and 93.5 g/kg DM, while CP contents in treatments with other forages were in average 1.7 times greater. While NDF decreased in the evaluated forage mixes, differences were not as striking as in CP content. Gross energy values ranged from (14-18 Mj/kg DM), and no significant variations were found between treatments. Metabolizable energy values ranged from 5.8 to 8.99 (MJ/kg DM), with the HA treatment showing the lowest content.

Table 2. Chemical composition of tropical grasses and grass-legume associations commonly used for cattle feeding in low-land tropics of Colombia

DM

CP

NDF

ADF

GE

EM

(g/kg fresh matter)

(g/kg DM)

(g/kg DM)

(g/kg DM)

(Mj/kg DM)

(Mj/kg DM)

Ca1

391

44

710

414

15.9

8.18

Ca1Ll

211

96

683

359

17.0

8.35

Ca2

223

94

606

298

17.0

7.59

Ca2Ld

238

128

581

299

18.0

7.06

Mo

335

41

760

451

17.9

8.23

MoLl

325

123

661

345

18.0

7.75

To

319

64

692

329

16.7

7.70

ToLdCn

306

103

682

380

17.4

7.49

De

315

39

749

381

17.7

7.68

DeTd

282

53

717

401

17.0

8.99

HA

632

62

613

389

14.0

5.80

Abbreviatures= DM: Dry matter; CP: Crude protein; NDF: Neutral detergent fiber; ADF: Acid detergent fiber; GE: Gross energy; ME: Metabolizable energy. Ca1=Urochloa hybrid cv. Cayman(100%, 65 days); Ca1Ll= Urochloa hybrid cv. Cayman (77.5% 65 days) +Leucaena leucocephala (22.5%, 58 days); Ca2= Urochloa hybrid cv. Cayman (100%, 45 days); Ca2Ld=Urochloa hybrid cv. Cayman(81.8%, 45 days) +Leucaena diversifolia(18.2%, 58 days); Mo=Megathyrsus maximus cv. Mombaza(100%,46 days); MoL1=Megathyrsus maximus cv. Mombaza(74%, 40 days) +Leucaena leucocephala(26%, 58 days); To= Urochloa brizantha cv. Toledo(100%, 42 days); ToLdCn=Urochloa brizantha cv. Toledo(74.2%, 39 days) +Leucaena diversifolia(18%, 59 days) +Canavalia brasiliensis(7.8%,59 days); De= Urochloa decumbens(100%, 47 days); DeTd=Urochloa decumbens (78%, 47 days) +Tithonia diversifolia(22%, 66 days); HA= Dichantium aristatum hay (100%, 52 days)
In vitro gas production

Gas production per gram of incubated organic matter (iOM) accumulated at 96 hours, varied between 215 and 297 ml (Figure 1) with TIP being 16 h on average, going from 12.5 to 20.8 h (Table 3). Colonization time of the feed particles by bacteria or phase Lag varied between 1.16 and 2.00 hours. The Maximum gas production rate had its highest values in the To (P<0001) and Ca2Ld (P<0001) treatments and the lowest values in theCa2, Mo, DeTd and HA treatments.

Figure 1. Accumulated gas production during fermentation time of tropical grasses and
grass-legume associations commonly used for cattle feeding in the tropics

 Abbreviatures= iOM=incubated organic matter; Ca1=Urochloa hybrid cv. Cayman (100%, 65 days); Ca1Ll= Urochloa hybrid cv. Cayman (77.5% 65 days) +Leucaena leucocephala (22.5%, 58 days); Ca2= Urochloa hybrid cv. Cayman (100%, 45 days); Ca2Ld=Urochloa hybrid cv. Cayman (81.8%, 45 days) +Leucaena diversifolia (18.2%, 58 days); Mo= Megathyrsus maximus cv. Mombaza (100%,46 days); MoL1=Megathyrsus maximus cv. Mombaza (74%, 40 days) + Leucaena leucocephala (26%, 58 days); To= Urochloa brizantha cv. Toledo (100%, 42 days); ToLdCn= Urochloa brizantha cv. Toledo (74.2%, 39 days) + Leucaena diversifolia(18%, 59 days) + Canavalia brasiliensis(7.8%,59 days); De= Urochloa decumbens(100%, 47 days); DeTd= Urochloa decumbens(78%, 47 days) + Tithonia diversifolia (22%, 66 days); HA= Dichantium aristatum hay (100%, 52 days)

Table 3. Gompertz model parameters for gas accumulation of tropical grasses and grass-legume associations commonly used for cattle feeding in the tropics

TIP (h)

GIP (h)

MGPR (ml)

PL (h)

Ca1

16.6bc

90.9c

5.4bc

1.6cd

Ca2Ll

17.6b

92.8bc

5.44bc

1.66c

Ca2

17.3b

81.d

4.70def

1.46d

Ca2Ld

12.5d

80.6d

5.9ab

1.68c

Mo

17.8b

79.4d

4.15f

2.00a

MoLl

14.8cd

82.3d

4.86cde

1.84b

To

16.4bc

106a

6.23a

1.56d

ToLdCn

16cb

93.5bc

5.39bc

1.27e

De

20.8a

99b

5.03cd

1.16e

DeTd

18.3ab

90.1c

4.83cdef

1.54d

HA

12.4d

76.1d

4.27ef

0.99f

P-value

<.0001

<.0001

<.0001

<.0001

RMSE

1.08

0.71

0.53

1.12

Abbreviatures= TIP: Time to inflection point; GIP: Gas at the inflection point; MGPR: Maximum gas production rate; PL: Phase lag. a, b, c, d, e, f= mean values among the same column with different superscript are significantly different (P<0.05); RMSE: Root Mean Square Error; Ca1=Urochloa hybrid cv. Cayman (100%, 65 days); Ca1Ll= Urochloa hybrid cv. Cayman (77.5% 65 days) + Leucaena leucocephala (22.5%, 58 days); Ca2= Urochloa hybrid cv. Cayman (100%, 45 days); Ca2Ld= Urochloa hybrid cv. Cayman (81.8%, 45 days) + Leucaena diversifolia (18.2%, 58 days); Mo= Megathyrsus maximus cv. Mombaza (100%,46 days); MoL1=Megathyrsus maximus cv. Mombaza (74%, 40 days) + Leucaena leucocephala (26%, 58 days); To= Urochloa brizantha cv. Toledo (100%, 42 days); ToLdCn= Urochloa brizantha cv. Toledo (74.2%, 39 days) + Leucaena diversifolia (18%, 59 days) + Canavalia brasiliensis (7.8%,59 days); De= Urochloa decumbens (100%, 47 days); DeTd= Urochloa decumbens (78%, 47 days) + Tithonia diversifolia (22%, 66 days); HA= Dichantium aristatum hay (100%, 52 days)

Dry matter degradation

As expected, degradability values showed great variability and differences (P<0001) between treatments due to differences in nutritional value. Thus, the treatment with the lowest degradability was HA, which was different (P<0001) from all other treatments, having a maximum degradation of 48.3% after 96 hours. Treatment (To) had the highest degradation values (P<0001), reaching a degradability of 62% at 48 h and 68% at 96 hours, despite having 11.4% more FDN content than HA (Table 2 and 4).

The treatments with mixtures of U. cayman grass and Leucaena ( Ca1, Ca2Ll, Ca2 and Ca2Ld) did not show differences between them in degradability values. However, these treatments had higher degradability values (significant statistical difference, P<0001) than treatments of mixtures with Mombasa and B. decumbens grasses (Mo, MoLl, De, DeTd) at 24 and 48 hours (Table 4).


Methane quantification

Table 4. In vitro degradability of dry matter and fiber at different times of tropical grasses and grass-legume associations commonly used for cattle feeding in the tropics

Dry matter degradability (g/kg DM)

24 h

48 h

96 h

Ca1

46.1b

57.2b

64.1b

Ca2Ll

45.7b

57.3b

63.0b

Ca2

45.3b

55.8bc

60.1cd

Ca2Ld

46.6b

55.1cd

58.8d

Mo

38.1d

48f

56.1e

MoLl

41.3c

52.1e

59.4cd

To

49.9a

62.3a

68.4a

ToLdCn

42.6c

53.2e

59.9cd

De

42.1c

55.2cd

63.4b

DeTd

41.1c

53.8ed

60.4c

HA

37.6d

43.7g

48.3f

P-value

<.0001

<.0001

<.0001

RMSE

1.02

0.705

0.58

a, b, c, d, e, f, g = mean values among the same column with different superscript are significantly different (P<0.05); A bbreviatures=RMSE: Root Mean Square Error. TIP: Time to inflection point; GIP: Gas at the inflection point; MGPR: Maximum gas production rate; PL: Phase lag. a, b, c, d, e, f= mean values among the same column with different superscript are significantly different (P<0.05); RMSE: Root Mean Square Error; Ca1=Urochloa hybrid cv. Cayman (100%, 65 days); Ca1Ll= Urochloa hybrid cv. Cayman (77.5% 65 days) + Leucaena leucocephala (22.5%, 58 days); Ca2= Urochloa hybrid cv. Cayman (100%, 45 days); Ca2Ld= Urochloa hybrid cv. Cayman (81.8%, 45 days) + Leucaena diversifolia (18.2%, 58 days); Mo= Megathyrsus maximus cv. Mombaza (100%,46 days); MoL1=Megathyrsus maximus cv. Mombaza (74%, 40 days) + Leucaena leucocephala (26%, 58 days); To= Urochloa brizantha cv. Toledo (100%, 42 days); ToLdCn= Urochloa brizantha cv. Toledo (74.2%, 39 days) + Leucaena diversifolia (18%, 59 days) + Canavalia brasiliensis (7.8%,59 days); De= Urochloa decumbens (100%, 47 days); DeTd= Urochloa decumbens (78%, 47 days) + Tithonia diversifolia (22%, 66 days); HA= Dichantium aristatum hay (100%, 52 days)

Treatments with lower CH4 (ml/g DMd) production were Ca1, Ca2, Ca2Ll and Ca2Ld at 24 h (P<0001). However, at 48 h, there were changes, with treatments showing the lowest emissions of CH4 being Mo and MoLl that at 24 h were not classified as such. On the other hand, at 24 h, the higher values of CH4 (ml/g DMd) were for HA,Mo, MoLl, To,ToLdCn (P<0001) and at 48 h, treatment HA showed the highest value of CH4 (P<0001) (Table 5).

Treatments such as To had the highest DMd values and one of the highest CH4 production values in ml and ml/kg DM at 24 h. In contrast, Ca1, Ca2 and Ca2Ld had degradability values within the lowest range and a CH4 production value within the low production range. However, other treatments such as HA and Mo had lower degradability values than these treatments.

Volatile fatty acids (VFAs)

Content of VFAs total at 24 h varied between 35.27 and 45.64 mmol/L. These values increased on average by 34.6% at 48 h (41.73 and 56.27 mmol/L). Propionate concentrations were more variable than those of acetate. At 24 h, propionate proportions varied between 14.28 and 27.62% and at 48 h between 19.58 and 30.04%. In all treatments, the highest proportion of VFAs corresponded to acetate, which was more than 56% of total VFAs, and the acetate to propionate ratio was higher than 1.98 in all treatments. The highest acetate to propionate ratio at 24 h were for Mo and MoLl (P<0001) and the lowest for Ca2 (P<0001). At 48 h, the highest values were for Mo, MoLl and HA (significant statistical difference with the other treatments, P<0001) and the lowest for Ca2, Ca2Ld, To and De (P<0001) (Table 5).

Table 5. Average values of in vitro methane (CH4) and volatile fatty acid production parameters at 24 and 48 hours of fermentation of tropical grasses and grass-legume associations commonly used for cattle feeding in the low-land tropics of Colombia

Ca1

Ca2Ll

Ca2

Ca2Ld

Mo

MoLl

To

ToLdCn

De

DeTd

HA

P-value

RMSE

24 h

CH4 (ml)

5.69de

6.43bcd

5.29de

4.69e

5.98cd

7.28ab

8.10a

6.91abc

5.85cde

6.53bc

7.06abc

<.0001

0.499

CH4 (ml/ g DM)

12.4de

14.1cd

12.0de

10.0e

15.7bc

17.6ab

16.3abc

16.2abc

13.9cd

15.9bcd

18.8a

<.0001

1.170

VFA total (mmol/L)

37.0ef

38.8bcd

42.3ab

41.8bc

35.3f

37.7ef

45.6a

41.2bcd

37.8ef

38.9de

35.3f

<.0001

1.383

VFA Proportions

    Acetate (%)

61.9d

63.2bc

56.2e

59.3e

64.5a

64.5a

59.3e

63.6ab

60.5e

62.3cd

62.3cd

<.0001

0.395

    Propionate (%)

18.4cd

18.2d

27.6a

23.9b

14.3e

15.6e

24.9b

18.4cd

20.2c

18.4cd

16.8de

<.0001

0.839

    Butyrate (%)

14.8de

15.3cd

10.4g

12.0fg

20.1a

20.9a

15.5cd

18.0b

13.1ef

14.8de

17.1bc

<.0001

0.722

Ratio of acetate to propionate

3.37c

3.39c

2.04f

2.48d

4.35a

4.11a

2.38e

3.46bc

2.99d

3.39c

3.71b

<.0001

0.123


48 h

CH4 (ml)

6.8cd

7.1bc

9.3b

8.3bcd

7.00bc

8.5bc

11.6a

13.5a

7.73c

6.74c

13.5a

<.0001

0.801

CH4 (ml/ g DM)

11.9e

12.4e

16.7d

15.1cde

14.6de

16.3cd

18.7c

25.4b

14.0de

12.5ce

30.9a

<.0001

1.645

VFA total (mmol/L)

51.9abc

50.6bc

47.5c

49.7bc

47.8c

50.0bc

56.3a

51.0bc

53.8ab

51.0bc

41.7d

<.0001

1.963

VFA Proportions

    Acetate (%)

60.9d

62.4bc

57.6f

59.2e

63.5b

64.2a

59.6e

63.4b

59.6e

61.2d

62.4c

<.0001

0.247

    Propionate (%)

26.3bc

23.5cd

29.2a

28.3ab

20.8de

20.5f

30.0a

23.0de

29.3a

25.9bc

19.6f

<.0001

1.166

    Butyrate (%)

16.2ef

16.9de

11.3i

12.3h

20.5b

23.1a

16.1fg

18.0c

14.7g

15.5de

17.8cd

<.0001

0.389

Ratio of acetate to propionate

2.31cd

2.67b

1.99e

2.09de

3.05a

3.13a

1.98e

2.76b

2.03e

2.36c

3.18a

<.0001

0.098

a, b, c, d, e, f, g = mean values among the same row with different superscript are significantly different (P<0.05); A bbreviatures= DMd: Dry matter degradability. RMSE: Root Mean Square Error. TIP: Time to inflection point; GIP: Gas at the inflection point; MGPR: Maximum gas production rate; PL: Phase lag. a, b, c, d, e, f= mean values among the same column with different superscript are significantly different (P<0.05); RMSE: Root Mean Square Error; Ca1=Urochloa hybrid cv. Cayman (100%, 65 days); Ca1Ll= Urochloa hybrid cv. Cayman (77.5% 65 days) + Leucaena leucocephala (22.5%, 58 days); Ca2= Urochloa hybrid cv. Cayman (100%, 45 days); Ca2Ld= Urochloa hybrid cv. Cayman (81.8%, 45 days) + Leucaena diversifolia (18.2%, 58 days); Mo= Megathyrsus maximus cv. Mombaza (100%,46 days); MoL1=Megathyrsus maximus cv. Mombaza (74%, 40 days) + Leucaena leucocephala (26%, 58 days); To= Urochloa brizantha cv. Toledo (100%, 42 days); ToLdCn= Urochloa brizantha cv. Toledo (74.2%, 39 days) + Leucaena diversifolia (18%, 59 days) + Canavalia brasiliensis (7.8%,59 days); De= Urochloa decumbens (100%, 47 days); DeTd= Urochloa decumbens (78%, 47 days) + Tithonia diversifolia (22%, 66 days); HA= Dichantium aristatum hay (100%, 52 days)


Discusion

In tropical livestock systems, fodder resources provide more than 90% of the energy consumed by the animals. However, compared to any other animal feed, forages have a highly variable nutritional value affected by factors such as NDF content and degradability, among others (Barahona and Sanchez 2005). As mentioned above, in the present study, a large variability was found between the values of the variables studied. Thus, it was found that for the same pasture with different harvesting times (Ca1=45 and Ca2=65 days), there was a significant difference in fiber and protein contents as harvesting time increased, fiber values increased, and protein values decreased (Table 2).

Treatments with U. cayman grass alone and mixtures of U. cayman grass and Leucaena had high gas production values per iOM and high degradability at 24 and 48 h. Also, in these treatments CH4 (ml/ g DMd) production did not show significant differences during the first 24 h of fermentation between them. However, after 48 h, there was a remarkable change, with greater production of CH4 in treatments Ca2 and Ca2Ld, probably due to slower degradation rate. This trend suggests that this relationship in values responds to forage digestibility, mainly due to NDF contents (Durmic et al 2010). Jayanegara et al (2011) obtained values between 0.43 and 0.56 when correlating cell wall compounds and in vitro dry matter digestibility. Thus, an increase in cell wall components (NDF and FDA) suppresses microbial activity by reducing the availability of fast fermenting carbohydrates and is negatively related to gas production (Doane et al 1997; Torres-Salado et al 2018). Soluble carbohydrates such as free sugars and starch result in lower CH4 production than structural carbohydrates because the latter are fermented more slowly. Thus, the inclusion of feeds with high structural carbohydrates increases methanogenesis (McAllister et al 1996; Yan et al 2006). In addition, fermentation of structural carbohydrates produces a high acetate:propionate ratio. However, depending on the rate of fiber degradation, CH4 production levels may change (Relling and Mattioli 2003).

Treatments that included other fodders and grass had reduced fiber content (Table 2). However, this reduction was not associated with increases in DMd for all treatments (Table 4) and probably, for this reason, there was no decrease in CH4 production in treatments likeTo vs. ToLdCn and De vs. DeTd (Table 5). In this study, MoLl and HA were associated with higher CH4 production values, lower degradability, and high acetate:propionate ratio values. In general, in the present study, no differences were found in CH4 production associated with the inclusion of forages such as Leucaena leucocephala, Leucaena diversifolia, Tithonia diversifolia or Canavalia brasiliensis. However, there is evidence of a positive effect on the nutritional value of the diets that could influence the productive behavior of the animals consuming these diets and that are shown when emissions per unit of product are reported (i.e., CH 4 production per dry matter intake or unit of meat and/or milk produced). These are parameters that, although cannot be evaluated with in vitro assays, have already been reported in vivo studies (Molina et al 2015; Gaviria et al 2020).

On the other hand, it is known that VFA production is directly related to the degradability of the forage. Thus, a forage of higher quality and higher degradability will have higher VFA production. (Judd and Kohn 2018). The acetate:propionate ratio is one of the essential factors affecting CH 4 production as it regulates the availability of H2 and, therefore, the production of CH4. Acetate and butyrate generate CH4 due to greater availability of CO2 and H 2 for methanogenic archaea, while for propionate formation in the rumen, it is considered a competitive form in H2 uptake that causes less CH4 synthesis (Gidlund et al 2015). Values of this ratio can vary from 1.99 to 4.35, and it is known that the utilization of available energy in the feed is more efficient if the ratio is close to 1.0 (Danielsson et al 2017). On the contrary, a high value of this ratio is related to low efficiency in energy use, which generates higher CH 4 production. However, for treatment To, this ratio was lower, and CH4 production was within the range of highest emissions and coincided with high digestibility values. Possibly, these findings could be explained by the fact that CH4 production does not depend on a single variable. That is, CH4 emissions are related to the type of carbohydrate fermented, the type of carbohydrate fractions in the forage and their digestibility, among other factors (Yan et al 2006; Singh et al 2012).

With the inclusion of Leucaena in a Cayman grass-based diet, there was no significant change in CH4 production or DMd, However, these treatments performed better compared to the other treatments such as Mombasa-grass and B. decumbens mixtures. Other authors have reported differences in CH4 production associated with legume inclusion and changes in nutritional quality and degradability (Chaves et al 2006; Donney's et al 2015; Rivera et al 2015; Jiménez-Santiago et al 2019).

Reported results are variable, and no clear-cut conclusions can be derived, because in some cases, the addition of the legume increases CH4 production (Carulla et al 2005), in other cases, no differences are found (Donney's et al 2015), and in other cases, CH4 production decreases (Ley de Coss et al 2018). The differences in in vitro CH 4 production among forage species may be explained by differences in the proportion of digestible carbohydrates in forages and cellulose (Hess et al 2003).

In most cases, and the current study, the addition of legumes or other forages of high nutritional quality such as Tithonia diversifolia to grasses increases the concentration of CP and decreases the concentration of total carbohydrates in the mixture (Hess et al 2003; Durmic et al 2017). In practice, the impact of including legumes in the diet has been promoted due to the positive effect they have on feed intake, animal production and the positive environmental impacts on the system (Castro et al 2008; Tiemman et al 2008; Gaviria et al 2015; Molina et al 2016).


Conclusions

Differences in nutritional quality generated changes in diet degradability, gas production, methane emissions, and VFAs production. In the present study, the inclusion of forages such as Leucaena and Tithonia diversifolia had a positive effect on nutrient (fiber and protein) content and degradability of the diets. However, these changes were not reflected in decreases in CH4 production, as significant differences were observed between treatments of grasses alone and associations with other forages. However, it is noteworthy that the group of treatments that included Cayman grass and its associations had lower CH4 production values and higher degradability than the rest of the treatments. Due to the good performance observed in this study with Cayman grass, future evaluations are recommended as a possible option in mitigating methane emissions and increasing the nutritional quality of animal diets. The CH4 production and fermentation parameters depend on many factors, some of which were not evaluated in this study, such as type of carbohydrates and ruminal microorganisms; therefore, it is suggested to extend the analysis to these variables. Although the gas technique is a valuable tool for evaluating the kinetics of fodder fermentation and methane production and also allows the simultaneous evaluation of many forages it is recommended to evaluate emissions in vivo and to estimate the intensity of emissions per unit product to better visualize the benefits of legume inclusion in the diet of grazing ruminants.


Acknowledgements

This research is part of a doctoral thesis funded by a doctoral grant from Colciencias scholarship Program No. 647 of Colombia and was implemented as part of the new OneCGIAR initiative Livestock, Climate and System Resilience (LCSR).We also acknowledge the financial assistance of BBSRC grants: UK—CIAT, RCUK-CIAT Newton Fund—Toward climate-smart forage-based diets for Colombian livestock (BB/R021856/1) and advancing sustainable forage-based livestock production systems in Colombia (CoForLife) (BB/S01893X/1).


References

Arango J, Ruden A, Martinez-Baron D, Loboguerrero A M, Berndt A, Chacón M, Torres C F, Oyhantcabal W, Gomez C A, Ricci P, Ku-Vera J, Burkart S, Moorby J M and Chirinda N 2020 Ambition Meets Reality: Achieving GHG Emission Reduction Targets in the Livestock Sector of Latin America. Frontiers in Sustainable Food Systems. 4:65. https://doi.org/10.3389/fsufs.2020.00065

Association of Official Analytical Chemists (AOAC) 1990 Protein (Crude) Determination in Animal Feed: Copper Catalyst Kjeldahl Method. (984.13). In: Official Methods of Analysis. 15th ed. Gaithersburg, USA.

Barahona R y Sánchez M 2005 Limitaciones físicas y químicas de la digestibilidad de pastos tropicales y estrategias para aumentarla. CORPOICA Ciencia y Tecnología Agropecuaria 6 (1): 69–82. doi: 10.21930/rcta.vol6_num1_art:39

Carulla J E, Kreuzer M, Machmüller A and Hess H D 2005 Supplementation of Acacia mearnsii tannin decreases methanogenesis and urinary nitrogen in forage-fed sheep. Australian Journal of Agricultural Research, 56: 961-970. https://doi.org/10.1071/AR05022

Casas G A, Rodriguez D and Afanador G 2010 Propiedades matemáticas del modelo Gompertz y su aplicación al crecimiento de los cerdos. Revista Colombiana de Ciencias Pecuarias. 23 (3): 349-358. https://dialnet.unirioja.es/servlet/articulo?codigo=3635370

Castro E, Mojica J E, León J M, Pabón M L, Carulla J E y Cárdenas E A 2008 Productividad de pasturas y producción de leche bovina bajo pastoreo de gramínea y gramínea + Lotus uliginosus en Mosquera, Colombia. Revista de la Facultad de Medicina Veterinaria y de Zootecnia, vol. 55. núm. I. pp. 9 -21. Universidad Nacional de Colombia Sede Bogotá. https://www.redalyc.org/pdf/4076/407639217002.pdf

Chaves AV, Thompson LC, Iwaasa A D, Scott S L, Olson M E, Benchaar C, Veira D M, McAllister T A 2006 Effect of pasture type (alfalfa vs. grass) on methane and carbon dioxide production by yearling beef heifers. Canadian Journal Animal Science . 86:409–418. https://doi.org/10.4141/A05-081

Danielsson R, Dicksved J, Sun L, Gonda H, Müller B, Schnürer A, Bertilsson J 2017 Methane production in dairy cows correlates with rumen methanogenic and bacterial community structure. Frontiers in Microbiology.8:A-226. https://doi.org/10.3389/fmicb.2017.00226

de Souza Congio G F, Bannink A, Mogollón O L, Hristov A N, Jaurena G, Gonda H, ... & Monteiro A L G 2021 Enteric methane mitigation strategies for ruminant livestock systems in the Latin America and Caribbean region: A meta-analysis. Journal of Cleaner Production, 127693. https://doi.org/10.1016/j.jclepro.2021.127693

Doane P, Schofield H P, Pell A N 1997 Neutral detergent fibre disappearance and gas and volatile fatty production during the in vitro fermentation of six forages. Journal of Animal Science. 75:3342–3352. https://doi.org/10.2527/1997.75123342x

Donney’s G, Molina I C, Rivera J E, Villegas G, Chará J & Barahona R 2015 Producción in vitro de metano de dietas ofrecidas en sistemas silvopastoriles intensivos con Tithonia diversifolia y sistemas tradicionales. In: 3° Congreso Nacional de Sistemas Silvopastoriles y VIII Congreso Internacional de Sistemas Agroforestales INTA. Puerto Iguazú, Argentina, 7–9 mayo, pp 672–677.

Durmic Z, Hutton P, Revell D K, Emms J, Hughes S, Vercoe P E 2010 In vitro fermentative traits of Australian woody perennial plant species that may be considered as potential sources of feed for grazing ruminants. Animal Feed Science and Technology. 160:98–109. https://doi.org/10.1016/j.anifeedsci.2010.07.006

Durmic Z, Ramírez‐Restrepo CA, Gardiner C, O'Neill CJ, Hussein E, & Vercoe P E 2017 Differences in the nutrient concentrations, in vitro methanogenic potential and other fermentative traits of tropical grasses and legumes for beef production systems in northern Australia. Journal of the Science of Food and Agriculture, 97(12), 4075-4086. https://doi.org/10.1002/jsfa.8274

Gaviria-Uribe X, Chirinda N, Barahona-Rosales R, Bolívar-Vergara D M, Rosenstock T, Arango J 2020a Measurement of methane emissions in cattle with infrared gas analyzer and gas chromatography. Revista U.D.C.A Actualidad & Divulgación Científica. 23(2):e1365. http://doi.org/10.31910/rudca.v23.n2.2020.1365

Gaviria-Uribe X, Bolivar D M, Rosenstock T S, Molina-Botero, I C, Chirinda N, Barahona R and Arango J 2020b Nutritional quality, voluntary intake and enteric methane emissions of diets based on novel cayman grass and its associations with two Leucaena shrub legumes. Front. Vet. Sci. 7:579189. doi: 10.3389/fvets.2020.579189

Gaviria-Uribe X, Naranjo-Ramírez J F, Bolívar-Vergara D M, Barahona-Rosales R 2015 Consumo y digestibilidad en novillos cebuínos en un sistema silvopastoril intensivo. Archivos de Zootecnia. 64, 21–27. doi: 10.21071/az.v64i245.370

Gavrilova O, Leip A, Dong H, MacDonald J D, Gomez Bravo C A, Amon B, Barahona Rosales R, Del Prado A, Oyhantçabal W, Van Der Weerden T j, Widiawati Y 2019 Emissions from livestock and manure management. En: CALVO BUENDIA, E.; TANABE, K.; KRANJC, A.; BAASANSUREN, J.; FUKUDA, M.; NGARIZE, S.; OSAKO, A.; PYROSHENKO, Y. SHERMANAU, P.; FEDERICI, S.(Ed.). 2019 Refinement to the 2006 guidelines for National Greenhouse Gas Inventories. Agriculture, forestry and other land use. Geneve: IPCC, 2019. v. 4. cap. 10.

Gidlund H, Hetta M, Krizsan S J, Lemosquet S, Huhtanen P 2015 Effects of soybean meal or canola meal on milk production and methane emissions in lactating dairy cows fed grass silage-based diets. Journal Animal Science.98(11):8093-8106. https://doi.org/10.3168/jds.2015-9757

Goering H K and Van Soest P J 1970 Forage fibre analysis (apparatus, reagents, procedures and some applications). Agricultural Handbook, no 379. ARS-USDE. Washington, DC.

Hess H D, Monsalve L M, Lascano C E, Carulla J E, Díaz T E and Kreuzer M 2003 Supplementation of a tropical grass diet with forage legumes and Sapindus saponaria fruits: effects on in vitro ruminal nitrogen turnover and methanogenesis. Australian Journal Agricultural Research, 54: 703-713. https://doi.org/10.1071/AR02241

Ibrahim M, Chacón M, Cuartas C, Naranjo J, Ponce G, Vega P, Casasola F, Rojas J 2007 Almacenamiento de carbono en el suelo y la biomasa arbórea en sistemas de usos de la tierra en paisajes ganaderos de Colombia, Costa Rica y Nicaragua. Agroforestería las Américas (45):27-36. http://hdl.handle.net/11554/7934

Intergovernmental Panel on Climate Change (IPCC) 2019 Summary for policymakers. In: Climate Change and Land: An IPCC Special Report on Climate Change, Desertification, Land Degradation, Sustainable Land Management, Food Security, and Greenhouse Gas Fluxes in Terrestrial Ecosystems. Geneva: IPCC (2019).

International Organization for Standardization (ISO) 1998 Animal feeding stuffs: Animal products, and faeces or urine - ISO 9831: Determination of gross calorific value - Bomb calorimeter method. Geneva, Switzerland.

International Organization for Standardization (ISO) 1999 Animal feeding stuffs - Determination of moisture and other volatile matter content. ISO 6496. Geneva, Switzerland.

Jayanegara A, Wina E, Soliva C R, Marquardt S, Kreuzer M, Leiber F 2011 Dependence of forage quality and methanogenic potential of tropical plants on their phenolic fractions as determined by principal component analysis. Animal Feed Science and Technology. 163:231–243. https://doi.org/10.1016/j.anifeedsci.2010.11.009

Jiménez-Santiago A, Jiménez-Ferrer G, Alayón-Gamboa A, Pérez-Luna E J, Piñeiro-Vázquez A T, Albores-Moreno S, Pérez-Escobar M G & Castro-Chan R 2019 Fermentación ruminal y producción de metano usando la técnica de gas in vitro en forrajes de un sistema silvopastoril de ovinos de Chiapas, México. Revista mexicana de ciencias pecuarias, 10(2), 298-314. https://dx.doi.org/10.22319/rmcp.v10i2.4529

Judd L M and Kohn R A 2018 Test of conditions that affect in vitro production of volatile fatty acids and gases, Journal Animal Science,96, 694—704.

Ku-Vera J C, Castelán-Ortega O A, Galindo-Maldonado F A, Arango J, Chirinda N, Jiménez-Ocampo R, Valencia-Salazar S, Flores-Santiago E J, Montoya-Flores M D, Molina-Botero I C, Pineiro-Vázquez AT, Arceo-Castillo J I, Aguilar-Perez C F, Ramirez-Avilés L, Solorio-Sanchez F J 2020 Strategies for enteric methane mitigation in cattle fed tropical forages. Animal, 1-11. doi:10.1017/S1751731120001780

Ley de Coss A, Guerra C, Montañez O, Guevara F, Pinto R, Reyes J 2018 In vitro production of gas methane by tropical grasses. Revista MVZ Córdoba. 23(3):6788-6798, 2018. http://dx.doi.org/10.21897/rmvz.1368

Lindgren E 1983 Recalibration of the VOS method for determination of energy content in forages (Nykalibrering av VOS-metoden för bestämning av energivärde hos vallfoder). Working Paper. Swedish University of Agricultural Sciences, Uppsala (In Swedish)

McAllister T A, Cheng K-J, Okine E K, Mathison G W 1996 Dietary, environmental and microbiological aspects of methane production in ruminants. Canadian Journal of Animal Science. 1996 76:231–43. doi: 10.4141/cjas96-035

Menke K and Steingass H 1988 Estimation of the energetic feed value obtained from chemical analysis and in vitro gas production using rumen fluid. Animal Research and Development (28):7-55.

Molina I C, Donneys G, Montoya S, Villegas G, Rivera E, Chará J, Lopera J J y Barahona R 2015 Emisiones in vivo de metano en sistemas de producción con y sin inclusión de Tithonia diversifolia. 3° Congreso Nacional de Sistemas Silvopastoriles y VIII Congreso Internacional de Sistemas Agroforestales. Puerto Iguazú, Misiones, Argentina, pp. 678-682. https://www.researchgate.net/publication/276207652_Emisiones_in_vivo_de_metano_en_sistemas_de_produccion_con_y_sin_inclusion_de_Tithonia_diversifolia

Molina I C, Angarita E A, Mayorga O L, Chará J, Barahona R 2016 Effect of Leucaena leucocephala on methane production of Lucerna heifers fed a diet based on Cynodon plectostachyus. Livestock Science 185:24-29. https://doi.org/10.1016/j.livsci.2016.01.009

Relling A E, Mattioli G A 2003 Fisiología digestiva y metabólica de los rumiantes. (en línea). La Plata, UNLP. Facultad de Ciencias Veterinarias. 72 p. https://ganaderiasos.com/wp-content/uploads/2014/08/fisiologia-digestiva-y-met-de-los-rumiantes.pdf

Rivera J E, Molina I C, Donney’s G, Villegas G, Chará & J, Barahona R 2015 Dinámica de fermentación y producción de metano en dietas de sistemas silvopastoriles intensivos con Leucaena leucocephala y sistemas convencionales orientados a la producción de leche. Livestock Research Rural Development. 27, Article #76. Available: http://www.lrrd.org/lrrd27/4/rive27076.html .

Rivera-Herrera J E, Molina-Botero I, Chará J, Murgueitio-Restrepo R & Barahona-Rosales R 2017 Intensive silvopastoral systems with Leucaena leucocephala (Lam.) de Wit: productive alternative in the tropic in view of the climate change. Pastos y Forrajes, 40(3), 171-183. http://scielo.sld.cu/scielo.php?script=sci_abstract&pid=S0864-03942017000300001&lng=es&nrm=iso&tlng=en

SAS 2008 SAS/STAT User's Guide, Versión 9.2, Cary, NC: SAS Institute Inc.

Singh S, Kushwaha B P, Nag S K, Mishra A K Singh A & Anele UY 2012 In vitro ruminal fermentation, protein and carbohydrate fractionation, methane production and prediction of twelve commonly used Indian green forages. Animal Feed Science and Technology. 178:2-11. https://doi.org/10.1016/j.anifeedsci.2012.08.019

Theodorou M K, Williams B A, Dhanoa M S, McAllan A B and France J 1994 A simple gas production method using a pressure transducer to determine the fermentation kinetics of ruminant feeds. Animal Feed Science and Technology. 48: 185-197. https://doi.org/10.1016/0377-8401(94)90171-6

Van Soest P J, Robertson J B, Lewis B A 1991 Methods for dietary fiber neutral detergent fiber and nonstarch polysaccharides in relation to animal nutrition. Journal of Dairy Science. 74, 3583–3597. https://doi.org/10.3168/jds.S0022-0302(91)78551-2

https://doi.org/10.3168/jds.S0022-0302(91)78551-2 Yan T, Mayne C S & Porter M G 2006 Effects of dietary and animal factors on methane production in dairy cows offered grass silage-based diets. International Congress Series, 1293, 123-126. https://doi.org/10.1016/j.ics.2006.02.024