Advertisement

Progressive inclusion of pearl millet herbage as a supplement for dairy cows fed mixed rations: Effects on methane emissions, dry matter intake, and milk production

Open ArchivePublished:December 23, 2020DOI:https://doi.org/10.3168/jds.2020-18894

      ABSTRACT

      The inclusion of grazing in dairy feeding systems can improve animal welfare and reduce feed costs and labor for animal care and manure management. This work aimed to evaluate the effects of including pearl millet herbage (Pennisetum glaucum ‘Campeiro') as a supplement for dairy cows fed total mixed rations (TMR). The treatments included 100% TMR offered ad libitum (control, TMR100), 75% TMR ad libitum intake + access to grazing of a pearl millet pasture between the morning and afternoon milkings (7 h/d; pTMR75), and 50% TMR ad libitum intake + access to grazing of a pearl millet pasture between the morning and afternoon milkings (7 h/d; pTMR50). Nine multiparous Holstein and F1 Jersey × Holstein cows were distributed in a replicated 3 × 3 Latin square design with 3 periods of 21 d (a 16-d adaptation period and a 5-d measurement period). Cows in the TMR75 and TMR50 groups strip-grazed a pearl millet pasture with pre- and postgrazing sward height targets of 60 and 30 cm, respectively. The herbage dry matter intake (DMI) increased with decreasing mixed ration supplies, and the total DMI decreased linearly from 19.0 kg/d in the TMR100 group to 18.0 kg/d in the pTMR50 group. Milk production decreased linearly from 24.0 kg/d in the TMR100 group to 22.4 kg/d in the pTMR50 group, and energy-corrected milk (ECM) production decreased linearly from 26.0 kg/d to 23.6 kg/d. Enteric methane (CH4) emissions decreased linearly from 540 g/d in the TMR100 group to 436 g/d in the pTMR50 group, and CH4 yields (g/kg of DMI) tended to decrease linearly. The CH4 intensity was similar between treatments, averaging 20 g of CH4/kg of ECM. The inclusion of pearl millet herbage in the dairy cow diets decreased the total DMI and milk production to a small extent without affecting CH4 intensity (g/kg of ECM).

      Key words

      INTRODUCTION

      The use of grazing in dairy production systems can improve animal welfare and reduce health problems (
      • Arnott G.
      • Ferris C.P.
      • O'Connell N.E.
      Review: Welfare of dairy cows in continuously housed and pasture-based production systems.
      ) as well as reduce labor for animal care and manure management (
      • White S.L.
      • Benson G.A.
      • Washburn S.P.
      • Green Jr., J.T.
      Milk production and economic measures in confinement or pasture systems using seasonally calved Holstein and jersey cows.
      ;
      • Schingoethe D.J.
      A 100-year review: Total mixed ration feeding of dairy cows.
      ). Grazing can also decrease feeding costs and improve income-over-feed costs (
      • Soriano F.D.
      • Polan C.E.
      • Miller C.N.
      Supplementing pasture to lactating Holsteins fed a total mixed ration diet.
      ;
      • White S.L.
      • Benson G.A.
      • Washburn S.P.
      • Green Jr., J.T.
      Milk production and economic measures in confinement or pasture systems using seasonally calved Holstein and jersey cows.
      ). In contrast, pastures alone are rarely able to meet the energy requirements of lactating dairy cows (
      • Kolver E.S.
      • Muller L.D.
      Performance and nutrient intake of high producing Holstein cows consuming pasture or a total mixed ration.
      ;
      • Delaby L.
      • Peyraud J.L.
      • Delagarde R.
      Effect of the level of concentrate supplementation, herbage allowance and milk yield at turn-out on the performance of dairy cows in mid lactation at grazing.
      ;
      • O'Neill B.F.
      • Deighton M.H.H.
      • O'Loughlin B.M.
      • Mulligan F.J.J.
      • Boland T.M.M.
      • O'Donovan M.
      • Lewis E.
      Effects of a perennial ryegrass diet or total mixed ration diet offered to spring-calving Holstein-Friesian dairy cows on methane emissions, dry matter intake, and milk production.
      ) and do not provide a constant herbage supply throughout the year (
      • Wilkinson J.M.
      • Lee M.R.F.
      • Rivero M.J.
      • Chamberlain A.T.
      Some challenges and opportunities for grazing dairy cows on temperate pastures.
      ). Thus, mixed feeding systems involving grazing pastures and TMR have been proposed and used worldwide (
      • Wales W.J.
      • Marett L.C.
      • Greenwood J.S.
      • Wright M.M.
      • Thornhill J.B.
      • Jacobs J.L.
      • Ho C.K.M.
      • Auldist M.J.
      Use of partial mixed rations in pasture-based dairying in temperate regions of Australia.
      ).
      Investigations of dairy cow systems where cows are both grazing on temperate pastures and receiving a mixed ration have, in some studies, not affected total DMI and milk production (
      • Vibart R.E.
      • Fellner V.
      • Burns J.C.
      • Huntington G.B.
      • Green Jr., J.T.
      Performance of lactating dairy cows fed varying levels of total mixed ration and pasture.
      ;
      • Mendoza A.
      • Cajarville C.
      • Repetto J.L.
      Short communication: Intake, milk production, and milk fatty acid profile of dairy cows fed diets combining fresh forage with a total mixed ration.
      ); in other studies, total DMI and milk production had been reduced up to 16% (
      • Soriano F.D.
      • Polan C.E.
      • Miller C.N.
      Supplementing pasture to lactating Holsteins fed a total mixed ration diet.
      ;
      • Bargo F.
      • Muller L.D.
      • Delahoy J.E.
      • Cassidy T.W.
      Performance of high producing dairy cows with three different feeding systems combining pasture and total mixed rations.
      ;
      • White S.L.
      • Benson G.A.
      • Washburn S.P.
      • Green Jr., J.T.
      Milk production and economic measures in confinement or pasture systems using seasonally calved Holstein and jersey cows.
      ). Reductions in milk production have been observed in high-production cows (
      • Bargo F.
      • Muller L.D.
      • Delahoy J.E.
      • Cassidy T.W.
      Performance of high producing dairy cows with three different feeding systems combining pasture and total mixed rations.
      ) and when the proportion of TMR is relatively low (
      • Vibart R.E.
      • Fellner V.
      • Burns J.C.
      • Huntington G.B.
      • Green Jr., J.T.
      Performance of lactating dairy cows fed varying levels of total mixed ration and pasture.
      ;
      • Mendoza A.
      • Cajarville C.
      • Repetto J.L.
      Short communication: Intake, milk production, and milk fatty acid profile of dairy cows fed diets combining fresh forage with a total mixed ration.
      ). Relatively large reductions in DMI and milk production have also been observed when the quality of herbage decreases according to the season (
      • White S.L.
      • Benson G.A.
      • Washburn S.P.
      • Green Jr., J.T.
      Milk production and economic measures in confinement or pasture systems using seasonally calved Holstein and jersey cows.
      ;
      • Vibart R.E.
      • Fellner V.
      • Burns J.C.
      • Huntington G.B.
      • Green Jr., J.T.
      Performance of lactating dairy cows fed varying levels of total mixed ration and pasture.
      ). However, compared with investigations done on cows grazing temperate pastures, studies assessing the effect of including tropical herbage on diets of dairy cows receiving mixed rations are rare.
      The effects of combining temperate pastures with mixed rations on enteric methane (CH4) emissions have shown that CH4 yields (g/kg of DMI) may be lower in dairy cows exclusively grazing a high-quality perennial grass in the spring than in dairy cows receiving a TMR diet (
      • O'Neill B.F.
      • Deighton M.H.H.
      • O'Loughlin B.M.
      • Mulligan F.J.J.
      • Boland T.M.M.
      • O'Donovan M.
      • Lewis E.
      Effects of a perennial ryegrass diet or total mixed ration diet offered to spring-calving Holstein-Friesian dairy cows on methane emissions, dry matter intake, and milk production.
      ). Therefore, CH4 intensity (g/kg of ECM) did not change in cows exclusively grazing on a temperate pasture or receiving TMR supplementation, which was explained by the similarity between herbage and TMR quality (
      • O'Neill B.F.
      • Deighton M.H.
      • O'Loughlin B.M.
      • Galvin N.
      • O'Donovan M.
      • Lewis E.
      The effects of supplementing grazing dairy cows with partial mixed ration on enteric methane emissions and milk production during mid to late lactation.
      ). Additionally, CH4 intensity decreased in cows receiving a mixture of corn silage and soybean meal with the inclusion of an annual temperate pasture because of the improved quality of herbage compared with that of corn silage (
      • Dall-Orsoletta A.C.
      • Almeida J.G.R.
      • Carvalho P.C.F.
      • Savian J.V.
      • Ribeiro-Filho H.M.N.
      Ryegrass pasture combined with partial total mixed ration reduces enteric methane emissions and maintains the performance of dairy cows during mid to late lactation.
      ). However, the quality of herbage may be relatively low in tropical pastures such as those of pearl millet (Pennisetum glaucum), which has shown great potential for use in dairy cows. Compared with other tropical pastures, pearl millet pastures have been shown to increase DM yields from 10 to 20 t/ha and have both a high protein content and high digestibility (
      • de Assis R.L.
      • De Freitas R.S.
      • Mason S.C.
      Pearl millet production pratices in Brazil: A Review.
      ). Thus, the effects of including pearl millet pasture on CH4 emissions from dairy cows receiving TMR warrant further study.
      We hypothesized that, due to greater herbage and diet NDF content, a progressive inclusion of pear millet in dairy cow diets would decrease total DMI, milk production, and CH4 emissions (g/d), but CH4 yields (g/kg of DMI) and CH4 intensity (g/kg of ECM) would increase. The aim of this work was to quantify the effects of partial TMR replacement by pearl millet on CH4 emissions and production responses in dairy cows.

      MATERIALS AND METHODS

      The Ethics Committee of the University of Santa Catarina State (Brazil) approved all the procedures in this study under protocol number 4373090816.

      Treatments, Experimental Design, and Animals

      The experiment was conducted according to a 3 × 3 Latin square design replicated 3 times. Nine multiparous cows were divided into 3 homogeneous groups (squares) of 3 animals, having 1 Holstein and 2 Holstein × Jersey cows as similar as possible in terms of milk production. Each square was then assigned a different treatment sequence. The following variables were determined during the week before the beginning of the experiment (means ± SD): milk production (25.1 ± 3.93 kg/d), DIM (136 ± 40 d), BW (533 ± 41.7 kg), and number of lactations (2.6 ± 0.69). Each experimental period lasted 21 d, with a 16-d adaptation period and a 5-d measurement period.
      The treatments were as follows: 100% TMR (TMR100) offered ad libitum (control), 75% TMR ad libitum intake + access to grazing of a pearl millet pasture (pTMR75), and 50% TMR ad libitum intake + access to grazing of a pearl millet pasture (pTMR50). All treatments received the same mixed ration, which was balanced after chemical analysis of the ingredients to meet the net energy and metabolizable protein (PDI) requirements of the control treatment, according to equations developed by the
      • INRA
      Alimentation Des Bovins, Ovins et Caprins.
      . The TMR was composed of corn silage and a concentrate (60:40 ratio on a DM basis). The ingredients, chemical composition, and nutritive value of the TMR are presented in Table 1.
      Table 1Chemical composition and nutritive value of mixed rations offered to dairy cows
      Mineral supplement composition available in the feeding area and paddocks (on natural basis): 150 g/kg of calcium, 78 g/kg of phosphorus; 26 g/kg of sulfur, 20 g/kg of magnesium, 114 g/kg of sodium, 100 mg/kg of cobalt, 1,500 mg/kg of copper, 30 mg/kg of chromium, 2,000 mg/kg of iron, 80 mg/kg of iodine, 2,300 mg/kg of manganese, 30 mg/kg of selenium, 5,000 mg/kg of zinc, and 780 mg/kg of fluorine.
      ItemContent
      Ingredient, g/kg of DM
       Corn silage600
       Ground corn260
       Soybean meal140
      Chemical composition, g/kg of DM
       DM, g/kg fresh380
       OM962
       CP149
       NDF340
       ADF173
      Nutritive value
      Estimated from chemical analysis and via equations proposed by INRA (2007); PDIN = metabolizable protein when N is limiting for microbial synthesis in the rumen; PDIE = metabolizable protein when energy is limiting for microbial synthesis in the rumen.
       Gross energy, MJ/kg of DM18.7
       OM digestibility0.77
       NEL, MJ/kg of DM7.02
       PDIN, g/kg of DM97.8
       PDIE, g/kg of DM99.0
      1 Mineral supplement composition available in the feeding area and paddocks (on natural basis): 150 g/kg of calcium, 78 g/kg of phosphorus; 26 g/kg of sulfur, 20 g/kg of magnesium, 114 g/kg of sodium, 100 mg/kg of cobalt, 1,500 mg/kg of copper, 30 mg/kg of chromium, 2,000 mg/kg of iron, 80 mg/kg of iodine, 2,300 mg/kg of manganese, 30 mg/kg of selenium, 5,000 mg/kg of zinc, and 780 mg/kg of fluorine.
      2 Estimated from chemical analysis and via equations proposed by
      • INRA
      Alimentation Des Bovins, Ovins et Caprins.
      ; PDIN = metabolizable protein when N is limiting for microbial synthesis in the rumen; PDIE = metabolizable protein when energy is limiting for microbial synthesis in the rumen.
      The individual voluntary DM TMR intake was quantified before the experiment started in a 14-d preexperimental period where cows been housed and fed individually. The average DMI of the last 5 d was considered for calculating the amount of mixed ration to be offered in the pTMR75 and pTMR50 treatments for each cow throughout the experiment. During the experiment, the cows were housed individually, where either TMR or pTMR were offered in covered outdoor feeders. The cows in the TMR100 group were fed twice a day after the morning and afternoon milkings; these cows received a daily quantity that was 20% greater than the voluntary DMI measured the prior day. In pTMR75 and pTMR50 treatments, cows received 75 and 50% of their TMR ad libitum intake measured during the preexperimental period, respectively. They also had access to a pasture between the morning and afternoon milkings (7 h/d of access to the pasture, from 0800 h to 1500 h) and received pTMR after the afternoon milking (13 h/d of access to pTMR, from 1700 h to 0600 h). The TMR and pTMR refusals were individually collected and weighed once per day during the morning milking. Water and mineral supplements (Bovigold, DSM Tortuga, São Paulo, Brazil) were continuously available in the feeding area and paddocks.

      Pasture and Grazing Management

      The experiment was performed in Lages, SC, Brazil (50.18° W, 27.47° S; 920 m above sea level), from January 25 to March 29, 2019. An area of 2 ha of pearl millet sown in 2017 was used. During the experimental period, the average temperature was 19.6°C, and the cumulative rainfall was 215 mm. The 10-yr climatic average temperature and rainfall during the months of the experiment were 14.5°C and 161 mm, respectively. Before the first grazing cycle (after pearl millet had developed the third-leaf stage) and after each experimental period, the experimental area was fertilized with 50 kg of N/ha, which was supplied as urea.
      The area was divided into 2 paddocks, with 1-third and 2-thirds of the surface assigned to the pTMR75 and pTMR50 groups, respectively. This size ratio of the surface was chosen because the expected herbage DMI for the pTMR75 and pTMR50 groups were 25% and 50% of the ad libitum DMI, respectively, and thus the TMR75 group was expected to require half the area of the pTMR50 group. The paddocks were strip-grazed with pre- and postgrazing sward height targets of 60 and 30 cm, respectively. To achieve these targets, the areas allocated daily to the pTMR75 and pTMR50 groups were 41 and 82 m2/cow, respectively, which was defined on the basis of a 1-wk preexperimental period. As the actual pre- and postgrazing sward heights throughout the experiment were close to the pre- and postgrazing target heights, no other adjustments for area allocation were necessary. To minimize variations in herbage quality between periods, different areas were used during the last 14 d of each period. These areas were mowed 18 d before starting the measurement periods for controlling of pregrazing sward height and chemical composition. Grazing management processes aimed to ensure that animals in the pTMR75 and pTMR50 groups removed the same proportion of forage in relation to the pregrazing height and that this removal did not exceed 50% of the initial height. The aim of 50% was chosen because it is the threshold at which the grazing management and structural characteristics of the herbage at the end of the occupation period can impose restrictions on herbage intake (
      • Zanini G.D.
      • Santos G.T.
      • Schmitt D.
      • Padilha D.A.
      • Sbrissia A.F.
      Distribuição de colmo na estrutura vertical de pastos de capim Aruana e azevém anual submetidos a pastejo intermitente por ovinos.
      ;
      • Mezzalira J.C.
      • Carvalho P.C.F.
      • Amaral M.F.
      • Bremm C.
      • Trindade J.K.
      • Gonçalves E.N.
      • Genro T.C.M.
      • Silva R.W.S.M.
      Rotational grazing management in a tropical pasture to maximize the dairy cow's herbage intake rate.
      ).

      Animal Measurements

      Individual milk production values were recorded twice daily (at 0700 h and 1600 h), and milk samples were collected at each milking via an electronic milk meter (Waikato Milking Systems, Hamilton, New Zealand) approved by the International Committee for Animal Recording (ICAR). The milk composition (fat, milk CP, and MUN concentrations) was individually measured for samples collected during each milking of the last 5 d of each period via infrared spectrophotometry (
      • IDF
      Milk and liquid milk products—Guidelines for the application of mid-infrared spectrometry. ISO standard number 9622, IDF 141.
      ). The ECM production standardized to 4.0% fat and 3.3% protein was calculated according to the equation proposed by
      • Tyrrell H.F.
      • Reid J.T.
      Prediction of the energy value of cow's milk.
      as follows: ECM (kg/d) = milk production kg × [37.6 × fat (g/kg) + 20.9 × protein (g/kg) + 948]/3,138.
      The TMR and pTMR intake, as well OM, CP, NDF and ADF intakes, and diet concentration, were measured as the average difference between the supplied quantity and the remaining quantity from each of the last 5 d of each period, when the DM, OM, CP, NDF and ADF content of offered TMR, pTMR, and refusals were measured separately. The individual herbage intake was measured according to the n-alkane technique (
      • Mayes R.W.
      • Lamb C.S.
      • Colgrove P.M.
      The use of dosed and herbage n-alkanes as markers for the determination of herbage intake.
      ) via the C31 (naturally present in the forage):C32 (supplied to the animals) ratio. Animals received cellulose pellets (Carl Roth, GmbH, Karlsruhe, Germany) containing 186 mg of C32 twice per day after each milking from d 8 to 21 of each experimental period. During the last 5 d of each experimental period, fecal grab samples were collected from each cow after each milking. The fecal samples were oven-dried at 60°C for at least 72 h, composited by period and cow, and then ground (Solab SL-31, Piracicaba, Brazil) to pass through a 1-mm screen for subsequent chemical analyses.
      The daily grazing time in the pTMR75 and pTMR50 groups was measured individually via visual observations every 5 min between 0800 h and 1500 h during the last 5 d of each period. The cows were previously accustomed to humans, and the time spent watching each individual animal was no more than 10 s, during which grazing behavior or no grazing was recorded (
      • Penning P.D.
      • Rutter S.M.
      ). No behavior was recorded indoors when the cows were milked or received the TMR. The herbage intake rate (g of DM/min) was estimated per cow and period by dividing the average daily herbage intake by the average daily grazing time.
      The gross energy (GE) of herbage was estimated as proposed by
      • INRA
      as GE (kcal/kg of OM) = 4,531 + 1.731 × CP (g/kg of OM) − 71 (n = 166, R2 = 0.89), whereas the GE for the mixed ration was calculated from tabulated values for the concentrates and corn silage (
      • INRA
      ). The OM digestibility was estimated from the chemical composition of forages according to specific equations for corn silage, herbage, and concentrates (
      • INRA
      ). The NEL and PDI balances were estimated per cow and period according to the difference between the NEL and PDI supply and requirements, according to the methods of the
      • INRA
      Alimentation Des Bovins, Ovins et Caprins.
      . The NEL requirements were estimated considering BW (kg) and FCM (kg/d) as follows: NEL requirements (Mcal/d) = 0.080 × BW0.75 + FCM × 0.7476. The PDI requirements were estimated with consideration of the BW (kg), actual milk production (kg/d), and CP content (g/kg of milk) as follows: PDI requirements (g/d) = 3.25 × BW0.75 + milk production × (CP × 0.93) × 10. The NEL and PDI supplies were estimated considering herbage and TMR intakes and their NEL and PDI contents, respectively.
      Daily CH4 emissions were measured individually according to the sulfur hexafluoride (SF6) tracer gas technique described by
      • Johnson K.
      • Huyler M.
      • Westberg H.
      • Lamb B.
      • Zimmerman P.
      Measurement of methane emissions from ruminant livestock using a sulfur hexafluoride tracer technique.
      . Each cow received 1 SF6 capsule 21 d before beginning the experiment, with an average SF6 release rate of 3.68 ± 0.10 mg/d. This average release rate was quantified by immersing the capsules in a 39°C water bath and then measuring weight loss during a period of 6 wk. The gas samples were collected on the last 5 d of each period, from the afternoon milking of d 16 to the afternoon milking of d 21, which was possible due to the calibration of flow regulators and storage capacity of the air-sampling devices (
      • Pinares-Patiño C.
      • Gere J.
      • Williams K.
      • Gratton R.
      • Juliarena P.
      • Molano G.
      • MacLean S.
      • Sandoval E.
      • Taylor G.
      • Koolaard J.
      Extending the collection duration of breath samples for enteric methane emission estimation using the SF6 tracer technique.
      ). Thus, from 5 d of gas sampling, only a half-day was not concomitant with intake measurements; however, both variables were measured for 120 h consecutively.
      Cows with or without access to pastures received the same kind of air-sampling devices concomitantly; each device was put on the head halters such that the sampling point was positioned above the nostrils. The air-sampling devices consisted of stainless-steel cylinders (0.5-L volume) with the sample flow regulated by a brass ball bearing (
      • Gere J.I.
      • Gratton R.
      Simple, low-cost flow controllers for time averaged atmospheric sampling and other applications.
      ). The cylinders were cleaned with high-purity N gas and preevacuated before each sample collection. The flow regulators were calibrated to allow for an expected remaining vacuum of approximately 500 mbar (which represents half of the total cylinder volume) in the cylinder at the end of the sample collection period (5 consecutive days). In addition to breath samples, 2 identical apparatuses were placed 1.5 m above the soil in the paddocks, and 2 others were placed where the TMR was offered to measure the background concentrations of CH4 and SF6 in the environment.
      To ensure the most successful individual gas samples, 2 gas-sampling cylinders were used simultaneously per animal. When 2 concomitant air samples per cow were collected successfully, the average was used. Operation of the gas-sampling apparatus was considered successful if the residual vacuum was between 350 and 650 mbar, which correspond to 37.6 and 69.8% of the initial vacuum, respectively (
      • Pinares-Patiño C.
      • Gere J.
      • Williams K.
      • Gratton R.
      • Juliarena P.
      • Molano G.
      • MacLean S.
      • Sandoval E.
      • Taylor G.
      • Koolaard J.
      Extending the collection duration of breath samples for enteric methane emission estimation using the SF6 tracer technique.
      ). The average residual vacuum in the gas-sampling apparatus was similar between treatments, and overall collection of the 54 gas samples (2 gas-sampling cylinders × 9 cows × 3 periods) was 69% successful. In 4 situations, samples from both gas-sampling cylinders of the same cow within the same period were considered lost.
      The CH4 emissions (g/d) were calculated in relation to the known release rate of SF6 by subtracting the background concentrations of CH4 and SF6 (
      • Berndt A.
      • Boland T.M.
      • Deighton M.H.
      • Gere J.I.
      • Grainger C.
      • Hegarty R.S.
      • Iwaasa A.D.
      • Koolaard J.P.
      • Lassey K.R.
      • Luo D.
      • Martin R.J.
      • Martin C.
      • Moate P.J.
      • Molano G.
      • Pinares-Patiño C.
      • Ribaux B.E.
      • Swainson N.M.
      • Waghorn G.C.
      • Williams S.R.O.
      ) as follows:
      RCH4=RSF6[CH4]M-[CH4]BG[SF6]M-[SF6]BG×MWCH4MWSF6×1,000,


      where RCH4 is the enteric CH4 (g/cow/d), RSF6 is the release rate of SF6 (mg/d), MWCH4 is the molecular mass of CH4 (16 g), and MWSF6 is the molecular mass of SF6 (146 g). [CH4]BG and [SF6]BG are the background concentrations of CH4 (ppm) and SF6 (ppt), respectively. The background CH4 and SF6 concentrations in the treatments with access to pearl millet pastures were calculated according to the weighted average of indoor and outdoor background concentrations, according to the length of time the animals spent in the pastures (7/24 h) or in confinement (17/24 h).

      Feed and Pasture Measurements

      Offered TMR and pTMR were sampled twice daily from d 15 to 20 of each period, and the samples were composited per period. Samples of the orts left by each cow were collected during the last 5 d of each period and were used to create a composite sample for each cow and period. All samples were dried in an oven for 72 h at 60°C and then ground (Solab SL-31, Piracicaba, Brazil) to pass through a 1-mm screen for subsequent chemical analyses.
      The pregrazing herbage mass was measured at ground level by cutting four 1-m2 squares of pearl millet with scissors per treatment every day during the last 5 d of each period. The herbage DM concentration was determined for each square from an 800-g subsample. The pre- and postgrazing sward heights were measured daily via a 1.0-m sward stick (
      • Barthram G.
      ) by averaging the first contact of 60 readings taken randomly throughout the area allocated for grazing by each group. Selected herbage samples were collected by the hand-plucked method daily during the last 5 d of each experimental period. The samples were dried in a forced-ventilation oven for 72 h at 60°C and then stored for chemical analyses. The morphological composition of the canopy was determined on d 18 and 20. In each treatment, 20 handfuls of randomly selected herbage were cut at ground level. These samples were separated into leaf (lamina + sheath), stem, and senescent material. Each component was oven-dried at 60°C for 72 h and subsequently weighed.

      Chemical Analyses

      After the samples were ground, their DM content was determined by drying at 105°C for 24 h. The ash content was quantified by combustion in a muffle furnace at 550°C for 4 h, and the OM was quantified on the basis of the mass difference. The total N content was measured according to the Dumas combustion method 968.06 (
      • AOAC International
      Official Methods of Analysis of AOAC International. CD-ROM.
      ) via a Leco FP 528 instrument (LC, Leco Corporation, Saint Joseph, MI). The CP content was calculated as N content multiplied by 6.25. The NDF concentration was assessed according to the methods of
      • Mertens D.R.
      Gravimetric determination of amylase-treated neutral detergent fiber in feeds with refluxing in beakers or crucibles: Collaborative study.
      , with the exception that the samples were weighed in filter bags and treated with a neutral detergent in an Ankom A220 system (Ankom Technology, Macedon, NY). This analysis included α-amylase and residual ash, but did not include sodium sulfite. The concentration of ADF was analyzed according to method 973.18 of the AOAC (
      • AOAC International
      Official Methods of Analysis of AOAC International. CD-ROM.
      ).
      The n-alkane content was quantified on the basis of the protocol described by
      • Dove H.
      • Mayes R.W.
      Protocol for the analysis of n-alkanes and other plant-wax compounds and for their use as markers for quantifying the nutrient supply of large mammalian herbivores.
      , which was adapted for the use of the columns, as proposed by
      • Oliveira D.E.
      • Tedeschi L.O.
      Comparing plastic syringes and disposable columns to extract n-alkanes in forage and concentrate feeds.
      . The n-alkane content was analyzed via gas chromatography by a Clarus 580 instrument (PerkinElmer, Inc., Waltham, MA) equipped with a flame ionization detector and capillary column (PerkinElmer Elite-1, 100% dimethyl polysiloxane; 30 m × 0.25 mm and a 0.25-µm film thickness).
      The concentrations of CH4 (ppm) and SF6 (ppt) were determined via a GC-2014 gas chromatograph (Shimadzu, Kyoto, Japan). The chromatograph was equipped with a flame ionization detector at 250°C and a 1/8” Shimalite Q packed column (0.7 m, 80/100 mesh; Shinwa Chemical Industries Ltd., Kyoto, Japan) for the detection of CH4, and equipped with an electron capture detector at 325°C and a 1/8” Porapak N packed column (1.5 m, 100/180 mesh) for the detection of SF6. A mixture comprising 5% CH4 and argon was used as the compositional gas in the SF6 analysis (electron capture detector). The gas chromatograph column was maintained at 80°C during the analysis, and N gas was used as a carrier with a flow of 25 cm3/min. Calibration curves were established by the use of certified standards (White Martins Development Laboratory, Osasco, Brazil), with CH4 concentrations of 2.5, 5.0, 10, and 20 ppm, and SF6 concentrations of 11, 30, and 100 ppt. The minimum detection limit, which is usually critical because of the low concentration of background CH4 and SF6, were 0.15 ppm and 5.2 ppt, respectively.

      Statistical Analyses

      The dependent variables were subjected to ANOVA via the PROC MIXED function of SAS software (version 9.4, SAS Institute, Cary, NC). The animal variables, which were averaged per cow and per period (n = 27), were analyzed using the following model:
      Yijk = μ + squarei + periodj + treatmentk + squarei × treatmentk + cowl(i) + eijk,


      where Yijk, μ, squarei, periodj, treatmentk, squarei × treatmentk, cowl(i), and eijkl represent the analyzed variable, the overall mean, the fixed effects of the square, the fixed effects of period, the fixed effects of treatment, the fixed effects of square × treatment interaction, the random effect of cow nested in square, and the residual error, respectively. The fixed effect of treatment × period interaction was not significant for DMI, milk production, and methane variables, and thus was removed from the model.
      The variables were tested via orthogonal polynomial contrasts to determine the linear and quadratic effects of the proportion of herbage inclusion in the diet. The least squares means were considered as significantly different if P < 0.05; P-values between 0.05 and 0.10 were considered trends, and standard errors of the mean were reported to describe variations.

      RESULTS

      The pre- and postgrazing sward heights and the pregrazing herbage mass of pearl millet averaged 62 cm, 32 cm, and 3,500 kg of DM/ha, respectively (Table 2). The CP, NDF, and ADF contents of the ingested pearl millet averaged 201, 625, and 306 g/kg of DM, respectively. The OM digestibility, energetic value, and PDI content of selected herbage averaged 69.4%, 5.81 MJ of NEL/kg of DM and 97 g/kg of DM, respectively. Throughout the experiment, CP content of ingested pearl millet herbage were 198, 215, and 188 g/kg of DM in periods 1, 2, and 3, respectively. The pearl millet NDF content was 611, 631, and 632 g/kg of DM, and OM digestibility was 0.70, 0.69, and 0.69 in periods 1, 2, and 3, respectively.
      Table 2Herbage characteristics and grazing management of a pearl millet pasture (Pennisetum glaucum ‘Campeiro') grazed by dairy cows receiving mixed rations
      ItemTreatment
      pTMR75 = 75% ad libitum TMR intake + grazing herbage after the morning milking (7 h/d); pTMR50 = 50% ad libitum TMR intake + grazing herbage after the morning milking (7 h/d).
      pTMR75pTMR50
      Herbage mass, kg of DM/ha3,4243,592
      Pregrazing sward height, cm62.961.0
      Postgrazing sward height, cm31.831.7
      Daily offered area, m2/cow4182
      Herbage allowance, kg of DM/d
       Aboveground level13.929.4
       Living leaves6.411.9
      Pregrazing herbage morphological composition, g/kg of DM
       Leaves (lamina + sheath)470403
       Stems516576
       Dead material1317
      Herbage chemical composition, g/kg of DM
       DM, g/kg158156
       OM915917
       CP193208
       NDF625625
       ADF302310
      Herbage nutritive value
      PDIN = metabolizable protein when N is limiting for microbial synthesis in the rumen (INRA. 2007); PDIE = metabolizable protein when energy is limiting for microbial synthesis in the rumen (INRA. 2007).
       Gross energy, MJ/kg of DM18.518.6
       OM digestibility0.690.70
       NEL, MJ/kg of DM5.825.80
       PDIN, g/kg of DM128134
       PDIE, g/kg of DM9698
      1 pTMR75 = 75% ad libitum TMR intake + grazing herbage after the morning milking (7 h/d); pTMR50 = 50% ad libitum TMR intake + grazing herbage after the morning milking (7 h/d).
      2 PDIN = metabolizable protein when N is limiting for microbial synthesis in the rumen (
      • INRA
      Alimentation Des Bovins, Ovins et Caprins.
      ); PDIE = metabolizable protein when energy is limiting for microbial synthesis in the rumen (
      • INRA
      Alimentation Des Bovins, Ovins et Caprins.
      ).
      The CH4 emissions (g/d) decreased linearly with the progressive inclusion of grazed herbage in the diet (linear effect: P < 0.01), and the CH4 yield (g/kg of DMI) tended to decrease linearly (P < 0.07; Table 3). For each kilogram of pearl millet herbage inclusion, there was a reduction of 13.3 g/d of CH4 production and a reduction of 0.2 g/kg of CH4 yield. The CH4 intensity was similar between treatments, averaging 20 g of CH4/kg of ECM. The herbage DMI increased with decreasing mixed ration supply, and the total DMI decreased linearly (Table 4). The mixed ration DMI decreased quadratically with decreasing mixed ration supply, with a greater reduction occurring between TMR100 and pTMR75 than between pTMR75 and pTMR50. The concentrate DMI decreased from 7.6 kg/d in the TMR100 group to 5.5 and 4.1 kg/d in the pTMR75 and pTMR50 groups, respectively. The grazing time (+48 min/d) and herbage DMI rate (+9.6 g of DM/min) were 22 and 42% greater (P < 0.001), respectively, in the pTMR50 group than in the pTMR75 group.
      Table 3Enteric methane emissions by dairy cows receiving mixed rations and with or without grazing access to a pearl millet pasture
      Data from 23 observations.
      MethaneTreatment
      TMR100 = total mixed ration ad libitum; pTMR75 = 75% ad libitum TMR intake + grazing herbage after the morning milking (7 h/d); pTMR50 = 50% ad libitum intake TMR + grazing herbage after the morning milking (7 h/d).
      SEMP-value
      TMR100pTMR75pTMR50ANOVALinearQuadratic
      g/d54048143616.30.0050.0010.59
      g/kg of DMI26.926.925.61.110.090.070.45
      g/kg of ECM
      ECM calculated as follows: kg of milk production × [37.6 × fat (g/kg) + 20.9 × protein (g/kg) + 948]/3,138 (Tyrrell and Reid, 1965).
      20.419.919.60.770.480.250.78
      % gross energy intake8.038.017.510.4230.390.280.49
      1 Data from 23 observations.
      2 TMR100 = total mixed ration ad libitum; pTMR75 = 75% ad libitum TMR intake + grazing herbage after the morning milking (7 h/d); pTMR50 = 50% ad libitum intake TMR + grazing herbage after the morning milking (7 h/d).
      3 ECM calculated as follows: kg of milk production × [37.6 × fat (g/kg) + 20.9 × protein (g/kg) + 948]/3,138 (
      • Tyrrell H.F.
      • Reid J.T.
      Prediction of the energy value of cow's milk.
      ).
      Table 4Dry matter intake, behavior, and chemical composition of the diet of dairy cows receiving mixed rations with or without grazing access to a pearl millet pasture
      ItemTreatment
      TMR100 = total mixed ration ad libitum; pTMR75 = 75% ad libitum TMR intake + grazing herbage after the morning milking (7 h/d); pTMR50 = 50% ad libitum TMR intake + grazing herbage after the morning milking (7 h/d).
      SEMP-value
      TMR100pTMR75pTMR50ANOVALinearQuadratic
      DMI, kg/d
       Total19.018.418.00.270.040.020.51
       Herbage4.67.80.10<0.001
       TMR19.013.810.20.19<0.001<0.0010.007
      Grazing time, min/d2162646.60.03
      Herbage DMI rate, g/min23.032.62.340.001
      Chemical composition, g/kg of DM
       OM962951942
       CP155169175
       NDF348429463
       ADF177215232
      NEL supply,
      Net energy for lactation supply.
      MJ/d
      136124118
      NEL balance,
      Net energy for lactation balance (NEL supply − NEL requirements).
      MJ/d
      201310
      1 TMR100 = total mixed ration ad libitum; pTMR75 = 75% ad libitum TMR intake + grazing herbage after the morning milking (7 h/d); pTMR50 = 50% ad libitum TMR intake + grazing herbage after the morning milking (7 h/d).
      2 Net energy for lactation supply.
      3 Net energy for lactation balance (NEL supply − NEL requirements).
      The milk production and ECM production decreased in cows with access to grazed herbage compared with that of cows in the TMR100 group (Table 5). For each kilogram of pearl millet herbage inclusion, there was a 0.2 kg of milk yield reduction. The milk fat and MUN concentrations were similar between treatments, and the milk protein content decreased linearly as the mixed ration supply was reduced.
      Table 5Milk production and milk composition of dairy cows receiving mixed rations and with or without grazing access to a pearl millet pasture
      ItemTreatment
      TMR100 = total mixed ration ad libitum; pTMR75 = 75% ad libitum TMR intake + grazing herbage after the morning milking (7 h/d); pTMR50 = 50% ad libitum TMR intake + grazing herbage after the morning milking (7 h/d).
      SEMP-value
      TMR100TMR75TMR50ANOVALinearQuadratic
      Milk production, kg/d24.022.722.40.83<0.001<0.0010.07
      4% FCM production, kg/d
      4% fat-corrected milk production.
      26.324.824.00.73<0.001<0.0010.12
      ECM, kg/d
      ECM calculated as follows: kg of milk production × [37.6 × fat (g/kg) + 20.9 × protein (g/kg) + 948]/3,138 (Tyrrell and Reid, 1965).
      26.024.523.60.68<0.001<0.0010.16
      Milk fat, g/kg46.946.845.42.530.190.110.37
      Milk protein, g/kg33.833.432.30.630.0090.0030.33
      Milk fat production, g/d1,1121,0491,00812.7<0.001<0.0010.34
      Milk protein production, g/d80875471612.90.003<0.0010.50
      MUN, mg/L18.418.018.00.820.640.420.64
      1 TMR100 = total mixed ration ad libitum; pTMR75 = 75% ad libitum TMR intake + grazing herbage after the morning milking (7 h/d); pTMR50 = 50% ad libitum TMR intake + grazing herbage after the morning milking (7 h/d).
      2 4% fat-corrected milk production.
      3 ECM calculated as follows: kg of milk production × [37.6 × fat (g/kg) + 20.9 × protein (g/kg) + 948]/3,138 (
      • Tyrrell H.F.
      • Reid J.T.
      Prediction of the energy value of cow's milk.
      ).

      DISCUSSION

      Methane Emissions

      The hypothesis concerning CH4 production was confirmed in part because daily CH4 emissions decreased linearly, but the CH4 yield and CH4 intensity did not increase with decreasing mixed ration supply. The linear reduction in daily CH4 emissions with decreasing mixed ration intake and the inclusion of grazing herbage is consistent with the linear reduction in total DMI, which is well known as the main driver of enteric CH4 emissions (
      • Hristov A.N.
      • Oh J.
      • Firkins L.
      • Dijkstra J.
      • Kebreab E.
      • Waghorn G.
      • Makkar H.P.S.
      • Adesogan T.
      • Yang W.
      • Lee C.
      • Gerber P.J.
      • Henderson B.
      • Tricarico J.M.
      Special topics — Mitigation of methane and nitrous oxide emissions from animal operations: I. A review of enteric methane mitigation options.
      ). These results are also in agreement with those of other studies (
      • O'Neill B.F.
      • Deighton M.H.H.
      • O'Loughlin B.M.
      • Mulligan F.J.J.
      • Boland T.M.M.
      • O'Donovan M.
      • Lewis E.
      Effects of a perennial ryegrass diet or total mixed ration diet offered to spring-calving Holstein-Friesian dairy cows on methane emissions, dry matter intake, and milk production.
      ;
      • Cameron L.
      • Chagunda M.G.G.
      • Roberts D.J.
      • Lee M.A.
      A comparison of milk yields and methane production from three contrasting high-yielding dairy cattle feeding regimes: Cut-and-carry, partial grazing and total mixed ration.
      ) that show that reductions in daily CH4 emissions occur because of reductions in DMI in dairy cows receiving fresh herbage plus pTMR compared with that of cows receiving TMR exclusively. Moreover, the CH4 emission values reported in the present study are within the range of values observed when cows consumed TMR (
      • Dall-Orsoletta A.C.
      • Almeida J.G.R.
      • Carvalho P.C.F.
      • Savian J.V.
      • Ribeiro-Filho H.M.N.
      Ryegrass pasture combined with partial total mixed ration reduces enteric methane emissions and maintains the performance of dairy cows during mid to late lactation.
      ) or grazed on a pearl millet pasture (
      • Alves T.P.
      • Dall-Orsoletta A.C.
      • Ribeiro-Filho H.M.N.
      The effects of supplementing Acacia mearnsii tannin extract on dairy cow dry matter intake, milk production, and methane emission in a tropical pasture.
      ).
      The tendency for linearly decreasing CH4 yields with decreasing TMR supply was unexpected because decreasing the concentrate content may increase CH4 emissions per unit of DMI (
      • Hristov A.N.
      • Oh J.
      • Firkins L.
      • Dijkstra J.
      • Kebreab E.
      • Waghorn G.
      • Makkar H.P.S.
      • Adesogan T.
      • Yang W.
      • Lee C.
      • Gerber P.J.
      • Henderson B.
      • Tricarico J.M.
      Special topics — Mitigation of methane and nitrous oxide emissions from animal operations: I. A review of enteric methane mitigation options.
      ). However, both the NDF (
      • Niu M.
      • Kebreab E.
      • Hristov A.N.
      • Oh J.
      • Arndt C.
      • Bannink A.
      • Bayat A.R.
      • Brito A.F.
      • Boland T.
      • Casper D.
      • Crompton L.A.
      • Dijkstra J.
      • Eugène M.A.
      • Garnsworthy P.C.
      • Haque M.N.
      • Hellwing A.L.F.
      • Huhtanen P.
      • Kreuzer M.
      • Kuhla B.
      • Lund P.
      • Madsen J.
      • Martin C.
      • McClelland S.C.
      • McGee M.
      • Moate P.J.
      • Muetzel S.
      • Muñoz C.
      • O'Kiely P.
      • Peiren N.
      • Reynolds C.K.
      • Schwarm A.
      • Shingfield K.J.
      • Storlien T.M.
      • Weisbjerg M.R.
      • Yáñez-Ruiz D.R.
      • Yu Z.
      Prediction of enteric methane production, yield, and intensity in dairy cattle using an intercontinental database.
      ) and concentrate (
      • INRA
      ) contents of the diet have been shown to be strongly related to enteric CH4 yields. Though the dietary NDF content was linear and positively related to CH4 yields, the dietary concentrate content exhibited a curvilinear relationship, where the maximum methanogenesis per kilogram of OM intake occurred with the inclusion of 35% concentrate (
      • Sauvant D.
      • Giger-Reverdin S.
      Modélisation des interactions digestives et de la production de méthane chez les ruminants.
      ). In the present study, the diet concentrate content averaged 22, 30, and 40% in the TMR50, TMR75, and TMR100 groups, respectively. Therefore, it is logical to assume that reductions in CH4 yields due to a lower NDF content in the TMR100 group compared with the other groups were offset by an increase in amount of ruminal fermentable OM. This probably occurred because the concentrate content in the TMR100 diet was not high enough to affect the CH4 yield compared with that in the other treatments.
      The similarity in CH4 intensity (g/kg of ECM) between treatments may be explained because the ECM production and DMI decreased linearly in similar proportions when the mixed ration supply decreased. The average value of CH4 intensity (20 g/kg of ECM) was close to the values estimated by
      • Moate P.J.
      • Deighton M.H.
      • Williams S.R.O.
      • Pryce J.E.
      • Hayes B.J.
      • Jacobs J.L.
      • Eckard R.J.
      • Hannah M.C.
      • Wales W.J.
      Reducing the carbon footprint of Australian milk production by mitigation of enteric methane emissions.
      as emissions from the Australian dairy industry (19.9 g/kg of ECM) and values of studies assessing dairy cows receiving TMR or pTMR plus fresh temperate herbage grazing in Europe (16–24 g/kg of ECM;
      • O'Neill B.F.
      • Deighton M.H.H.
      • O'Loughlin B.M.
      • Mulligan F.J.J.
      • Boland T.M.M.
      • O'Donovan M.
      • Lewis E.
      Effects of a perennial ryegrass diet or total mixed ration diet offered to spring-calving Holstein-Friesian dairy cows on methane emissions, dry matter intake, and milk production.
      ,
      • O'Neill B.F.
      • Deighton M.H.
      • O'Loughlin B.M.
      • Galvin N.
      • O'Donovan M.
      • Lewis E.
      The effects of supplementing grazing dairy cows with partial mixed ration on enteric methane emissions and milk production during mid to late lactation.
      ). This is evidence that dairy cow diets including tropical herbage may have similar CH4 intensities as do those of confined or temperate herbage-based diets.

      Dry Matter Intake and Grazing Behavior

      The linear reduction in total DMI in cows grazing on the pearl millet pasture compared with that in cows in the TMR100 group (−0.1 kg for each kilogram of pearl millet herbage inclusion) could be explained primarily by the reduction in mixed ration intake, and thus the reduction in the concentrate intake. The substitution rate between forages and concentrates is typically in the range of 0.0 to 0.7 at grazing (
      • Delagarde R.
      • Valk H.
      • Mayne C.S.
      • Rook A.J.
      • González-Rodríguez A.
      • Baratte C.
      • Faverdin P.
      • Peyraud J.L.
      GrazeIn: A model of herbage intake and milk production for grazing dairy cows. 3. Simulations and external validation of the model.
      ), leading to a reduction in total intake when the concentrate supply is reduced within a large range of herbage quality and herbage allowance values (
      • Bargo F.
      • Muller L.D.
      • Kolver E.S.
      • Delahoy J.E.
      Invited review: Production and digestion of supplemented dairy cows on pasture.
      ;
      • Faverdin P.
      • Baratte C.
      • Delagarde R.
      • Peyraud J.L.
      GrazeIn: A model of herbage intake and milk production for grazing dairy cows. 1. Prediction of intake capacity, voluntary intake and milk production during lactation.
      ). In this study, this reduction was no larger than 6%, which could be explained by the forage-concentrate substitution increase with increasing concentrate intake (
      • Faverdin P.
      • Baratte C.
      • Delagarde R.
      • Peyraud J.L.
      GrazeIn: A model of herbage intake and milk production for grazing dairy cows. 1. Prediction of intake capacity, voluntary intake and milk production during lactation.
      ), because high substitution rate values (approximately 1.0) have little or no effect on total DMI. Additionally, the average percentage of reduction in total DMI observed when grazing partly replaced the mixed ration is also in agreement with the variation range (−4 to −7% of total DMI) observed for dairy cows when the concentrate proportion in the diet decreased in a similar range of roughage (corn silage or grass silage):concentrate (high starch) ratios (60:40 to 80:20;
      • Faverdin P.
      • Dulphy J.P.
      • Coulon J.B.
      • Vérité R.
      • Garel J.P.
      • Rouel J.
      • Marquis B.
      Substitution of roughage by concentrates for dairy cows.
      ).
      The similar reduction in total DMI in the pTMR75 and pTMR50 groups compared with the TMR100 groups may be explained by the reduction in the TMR supply from 75 to 50% of ad libitum intake being partly offset by an increase of 3.2 kg in herbage DMI. This increase was mediated through a greater grazing time (+48 min/d) and herbage intake rate (+9.6 g of DM/min) in the pTMR50 group than in the pTMR75 group. These results agree with those of other studies where dairy cows grazing on temperate herbage presented an increased herbage DMI as the feed supplement amount decreased (
      • Pérez-Ramírez E.
      • Delagarde R.
      • Delaby L.
      Herbage intake and behavioural adaptation of grazing dairy cows by restricting time at pasture under two feeding regimes.
      ;
      • Vibart R.E.
      • Fellner V.
      • Burns J.C.
      • Huntington G.B.
      • Green Jr., J.T.
      Performance of lactating dairy cows fed varying levels of total mixed ration and pasture.
      ). For instance,
      • Pérez-Ramírez E.
      • Delagarde R.
      • Delaby L.
      Herbage intake and behavioural adaptation of grazing dairy cows by restricting time at pasture under two feeding regimes.
      reported that cows increased their herbage DMI (+3.0 kg/d) by increasing their daily grazing time (+36 min/d) and herbage intake rate (+7 g of DM/min) when the feed supplement (corn silage + soybean meal) was reduced from 10 to 5 kg of DM/d.

      Milk Production and Milk Composition

      The linear reduction in ECM production in the pTMR75 and pTMR50 groups compared with that in the TMR100 group (−0.3 kg for each kilogram of pearl millet herbage inclusion) was a consequence of the concomitant linear reductions in total DMI and in concentrate DMI, both of which reduced the net energy intake. When the reduction in milk production was calculated as a function of only the decrease in concentrate intake, the milk production decreased by only 0.5 kg/d for each kilogram of reduced concentrate intake. This reduction is lower than that for classic milk production responses to concentrate supplementation (1 kg of milk/kg of concentrate intake) described in the literature (
      • Peyraud J.L.
      • Delaby L.
      Ideal concentrate feeds for grazing dairy cows.
      ;
      • Delagarde R.
      • Valk H.
      • Mayne C.S.
      • Rook A.J.
      • González-Rodríguez A.
      • Baratte C.
      • Faverdin P.
      • Peyraud J.L.
      GrazeIn: A model of herbage intake and milk production for grazing dairy cows. 3. Simulations and external validation of the model.
      ), which can be explained by the reduction in NEL supply from the TMR being partly offset by the NEL supply from the herbage intake. Finally, cows with access to the pearl millet pasture presented an important reduction in concentrate intake, but produced more than 90% of the amount of milk produced by the TMR100 cows. However, owing to the potential of shifting herbage nutritive values throughout the growing season, long-term continuous studies with larger number of cows grazing both pearl millet and other tropical forage species are highly recommended.
      The linear reduction in milk fat production with the progressive reduction in mixed ration was a consequence of milk production being lower than that of cows without access to grazing of the pearl millet pasture because the milk fat content was similar between the treatments. The high milk fat content observed in this study (46.2 g/kg) can be explained by the breed characteristics and agrees with the milk fat content reported in another study involving cows from the same herd (
      • Dall-Orsoletta A.C.
      • Oziemblowski M.M.
      • Berndt A.
      • Ribeiro-Filho H.M.N.
      Enteric methane emission from grazing dairy cows receiving corn silage or ground corn supplementation.
      ). The reduction in milk protein content in cows with access to the grazing herbage compared with that in cows in the TMR100 group is in good agreement with variations in the energy supply. The role of the energy supply, rather than the protein or AA supply, in improving milk protein content has already been demonstrated in 2 comprehensive literature reviews (
      • Coulon J.B.
      • Rémond B.
      Variations in milk output and milk protein content in response to the level of energy supply to the dairy cow: A review.
      ;
      • Beever D.E.
      • Sutton J.D.
      • Reynolds C.K.
      Increasing the protein content of cow's milk.
      ).

      CONCLUSIONS

      Including pearl millet in dairy cow diets decreased the total DMI and milk production, but even at the greatest level of herbage inclusion, cows were able to achieve more than 90% of the total DMI and milk production recorded for cows that were fed only TMR. As the relative reduction in milk production was similar to that of the DMI, CH4 emissions (g/d) decreased, but the CH4 intensity (g/kg of ECM) was unaffected by the progressive inclusion of herbage in the diet. Additional studies with dairy cows grazing tropical forages throughout the whole growing season are strongly encouraged.

      ACKNOWLEDGMENTS

      This work was supported in part by Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq; Brasília, Brazil, finance codes 403754/2016-0 and 308591/2019-4), Fundação de Amparo à Pesquisa e Inovação do Estado de Santa Catarina (FAPESC; Florianópolis, Brazil, finance code TR 584 2019), and Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES; Brasília, Brazil, finance code 001). The authors have not stated any conflicts of interest.

      REFERENCES

        • Alves T.P.
        • Dall-Orsoletta A.C.
        • Ribeiro-Filho H.M.N.
        The effects of supplementing Acacia mearnsii tannin extract on dairy cow dry matter intake, milk production, and methane emission in a tropical pasture.
        Trop. Anim. Health Prod. 2017; 49 (28801856): 1663-1668
        • AOAC International
        Official Methods of Analysis of AOAC International. CD-ROM.
        AOAC International, Rockville, MD1998
        • Arnott G.
        • Ferris C.P.
        • O'Connell N.E.
        Review: Welfare of dairy cows in continuously housed and pasture-based production systems.
        Animal. 2017; 11 (27364762): 261-273
        • Bargo F.
        • Muller L.D.
        • Delahoy J.E.
        • Cassidy T.W.
        Performance of high producing dairy cows with three different feeding systems combining pasture and total mixed rations.
        J. Dairy Sci. 2002; 85 (12487461): 2948-2963
        • Bargo F.
        • Muller L.D.
        • Kolver E.S.
        • Delahoy J.E.
        Invited review: Production and digestion of supplemented dairy cows on pasture.
        J. Dairy Sci. 2003; 86 (12613846): 1-42
        • Barthram G.
        Experimental Techniques: The HFRO Sward Stick. Hill Farming Research Organization, Edinburgh, UK1986: 29-30
        • Beever D.E.
        • Sutton J.D.
        • Reynolds C.K.
        Increasing the protein content of cow's milk.
        Aust. J. Dairy Technol. 2001; 56: 138-149
        • Berndt A.
        • Boland T.M.
        • Deighton M.H.
        • Gere J.I.
        • Grainger C.
        • Hegarty R.S.
        • Iwaasa A.D.
        • Koolaard J.P.
        • Lassey K.R.
        • Luo D.
        • Martin R.J.
        • Martin C.
        • Moate P.J.
        • Molano G.
        • Pinares-Patiño C.
        • Ribaux B.E.
        • Swainson N.M.
        • Waghorn G.C.
        • Williams S.R.O.
        Lambert M.G. Guidelines for Use of Sulphur Hexafluoride (SF6) Tracer Technique to Measure Enteric Methane Emissions from Ruminants. New Zealand Agricultural Greenhouse Gas Research Centre, Wellington, New Zealand2014: 166
        • Cameron L.
        • Chagunda M.G.G.
        • Roberts D.J.
        • Lee M.A.
        A comparison of milk yields and methane production from three contrasting high-yielding dairy cattle feeding regimes: Cut-and-carry, partial grazing and total mixed ration.
        Grass Forage Sci. 2018; 73: 789-797
        • Coulon J.B.
        • Rémond B.
        Variations in milk output and milk protein content in response to the level of energy supply to the dairy cow: A review.
        Livest. Prod. Sci. 1991; 29: 31-47
        • Dall-Orsoletta A.C.
        • Almeida J.G.R.
        • Carvalho P.C.F.
        • Savian J.V.
        • Ribeiro-Filho H.M.N.
        Ryegrass pasture combined with partial total mixed ration reduces enteric methane emissions and maintains the performance of dairy cows during mid to late lactation.
        J. Dairy Sci. 2016; 99 (27016830): 4374-4383
        • Dall-Orsoletta A.C.
        • Oziemblowski M.M.
        • Berndt A.
        • Ribeiro-Filho H.M.N.
        Enteric methane emission from grazing dairy cows receiving corn silage or ground corn supplementation.
        Anim. Feed Sci. Technol. 2019; 253: 65-73
        • de Assis R.L.
        • De Freitas R.S.
        • Mason S.C.
        Pearl millet production pratices in Brazil: A Review.
        Exp. Agric. 2018; 54: 699-718
        • Delaby L.
        • Peyraud J.L.
        • Delagarde R.
        Effect of the level of concentrate supplementation, herbage allowance and milk yield at turn-out on the performance of dairy cows in mid lactation at grazing.
        Anim. Sci. 2001; 73: 171-181
        • Delagarde R.
        • Valk H.
        • Mayne C.S.
        • Rook A.J.
        • González-Rodríguez A.
        • Baratte C.
        • Faverdin P.
        • Peyraud J.L.
        GrazeIn: A model of herbage intake and milk production for grazing dairy cows. 3. Simulations and external validation of the model.
        Grass Forage Sci. 2011; 66: 61-77
        • Dove H.
        • Mayes R.W.
        Protocol for the analysis of n-alkanes and other plant-wax compounds and for their use as markers for quantifying the nutrient supply of large mammalian herbivores.
        Nat. Protoc. 2006; 1 (17487151): 1680-1697
        • Faverdin P.
        • Baratte C.
        • Delagarde R.
        • Peyraud J.L.
        GrazeIn: A model of herbage intake and milk production for grazing dairy cows. 1. Prediction of intake capacity, voluntary intake and milk production during lactation.
        Grass Forage Sci. 2011; 66: 29-44
        • Faverdin P.
        • Dulphy J.P.
        • Coulon J.B.
        • Vérité R.
        • Garel J.P.
        • Rouel J.
        • Marquis B.
        Substitution of roughage by concentrates for dairy cows.
        Livest. Prod. Sci. 1991; 27: 137-156
        • Gere J.I.
        • Gratton R.
        Simple, low-cost flow controllers for time averaged atmospheric sampling and other applications.
        Lat. Am. Appl. Res. 2010; 40: 377-381
        • Hristov A.N.
        • Oh J.
        • Firkins L.
        • Dijkstra J.
        • Kebreab E.
        • Waghorn G.
        • Makkar H.P.S.
        • Adesogan T.
        • Yang W.
        • Lee C.
        • Gerber P.J.
        • Henderson B.
        • Tricarico J.M.
        Special topics — Mitigation of methane and nitrous oxide emissions from animal operations: I. A review of enteric methane mitigation options.
        J. Anim. Sci. 2013; 91 (24045497): 5045-5069
        • IDF
        Milk and liquid milk products—Guidelines for the application of mid-infrared spectrometry. ISO standard number 9622, IDF 141.
        International Dairy Federation, Brussels, Belgium2013
        • INRA
        Alimentation Des Bovins, Ovins et Caprins.
        in: INRA Besoins Des Animaux - Valeurs Des Aliments. Tables INRA 2007. 4th ed. QUAE Editions, Versailles, France2007
        • INRA
        Noizère P. Sauvant D. Delaby L. INRA Feeding System for Ruminants. Wageningen Academic Publishers, Wageningen, the Netherlands2018
        • Johnson K.
        • Huyler M.
        • Westberg H.
        • Lamb B.
        • Zimmerman P.
        Measurement of methane emissions from ruminant livestock using a sulfur hexafluoride tracer technique.
        Environ. Sci. Technol. 1994; 28 (22176184): 359-362
        • Kolver E.S.
        • Muller L.D.
        Performance and nutrient intake of high producing Holstein cows consuming pasture or a total mixed ration.
        J. Dairy Sci. 1998; 81 (9621244): 1403-1411
        • Mayes R.W.
        • Lamb C.S.
        • Colgrove P.M.
        The use of dosed and herbage n-alkanes as markers for the determination of herbage intake.
        J. Agric. Sci. 1986; 107: 161-170
        • Mendoza A.
        • Cajarville C.
        • Repetto J.L.
        Short communication: Intake, milk production, and milk fatty acid profile of dairy cows fed diets combining fresh forage with a total mixed ration.
        J. Dairy Sci. 2016; 99 (26778319): 1938-1944
        • Mertens D.R.
        Gravimetric determination of amylase-treated neutral detergent fiber in feeds with refluxing in beakers or crucibles: Collaborative study.
        J. AOAC Int. 2002; 85 (12477183): 1217-1240
        • Mezzalira J.C.
        • Carvalho P.C.F.
        • Amaral M.F.
        • Bremm C.
        • Trindade J.K.
        • Gonçalves E.N.
        • Genro T.C.M.
        • Silva R.W.S.M.
        Rotational grazing management in a tropical pasture to maximize the dairy cow's herbage intake rate.
        Arq. Bras. Med. Vet. Zootec. 2013; 65: 833-840
        • Moate P.J.
        • Deighton M.H.
        • Williams S.R.O.
        • Pryce J.E.
        • Hayes B.J.
        • Jacobs J.L.
        • Eckard R.J.
        • Hannah M.C.
        • Wales W.J.
        Reducing the carbon footprint of Australian milk production by mitigation of enteric methane emissions.
        Anim. Prod. Sci. 2016; 56: 1017-1034
        • Niu M.
        • Kebreab E.
        • Hristov A.N.
        • Oh J.
        • Arndt C.
        • Bannink A.
        • Bayat A.R.
        • Brito A.F.
        • Boland T.
        • Casper D.
        • Crompton L.A.
        • Dijkstra J.
        • Eugène M.A.
        • Garnsworthy P.C.
        • Haque M.N.
        • Hellwing A.L.F.
        • Huhtanen P.
        • Kreuzer M.
        • Kuhla B.
        • Lund P.
        • Madsen J.
        • Martin C.
        • McClelland S.C.
        • McGee M.
        • Moate P.J.
        • Muetzel S.
        • Muñoz C.
        • O'Kiely P.
        • Peiren N.
        • Reynolds C.K.
        • Schwarm A.
        • Shingfield K.J.
        • Storlien T.M.
        • Weisbjerg M.R.
        • Yáñez-Ruiz D.R.
        • Yu Z.
        Prediction of enteric methane production, yield, and intensity in dairy cattle using an intercontinental database.
        Glob. Chang. Biol. 2018; 24 (29450980): 3368-3389
        • O'Neill B.F.
        • Deighton M.H.
        • O'Loughlin B.M.
        • Galvin N.
        • O'Donovan M.
        • Lewis E.
        The effects of supplementing grazing dairy cows with partial mixed ration on enteric methane emissions and milk production during mid to late lactation.
        J. Dairy Sci. 2012; 95 (22959938): 6582-6590
        • O'Neill B.F.
        • Deighton M.H.H.
        • O'Loughlin B.M.
        • Mulligan F.J.J.
        • Boland T.M.M.
        • O'Donovan M.
        • Lewis E.
        Effects of a perennial ryegrass diet or total mixed ration diet offered to spring-calving Holstein-Friesian dairy cows on methane emissions, dry matter intake, and milk production.
        J. Dairy Sci. 2011; 94 (21426985): 1941-1951
        • Oliveira D.E.
        • Tedeschi L.O.
        Comparing plastic syringes and disposable columns to extract n-alkanes in forage and concentrate feeds.
        Cienc. Rural. 2010; 40: 1844-1847
        • Penning P.D.
        • Rutter S.M.
        Penning P.D. Ingestive Behaviour. 2nd ed. The British Grassland Society, Maidenhead, UK2004
        • Pérez-Ramírez E.
        • Delagarde R.
        • Delaby L.
        Herbage intake and behavioural adaptation of grazing dairy cows by restricting time at pasture under two feeding regimes.
        Animal. 2008; 2 (22443829): 1384-1392
        • Peyraud J.L.
        • Delaby L.
        Ideal concentrate feeds for grazing dairy cows.
        in: Garnsworthy P. Wiseman J. Responses to supplementation in interaction with grazing management and grass quality. University Press, Nottingham, UK2001
        • Pinares-Patiño C.
        • Gere J.
        • Williams K.
        • Gratton R.
        • Juliarena P.
        • Molano G.
        • MacLean S.
        • Sandoval E.
        • Taylor G.
        • Koolaard J.
        Extending the collection duration of breath samples for enteric methane emission estimation using the SF6 tracer technique.
        Animals (Basel). 2012; 2 (26486921): 275-287
        • Sauvant D.
        • Giger-Reverdin S.
        Modélisation des interactions digestives et de la production de méthane chez les ruminants.
        INRA Prod. Anim. 2009; 22: 375-384
        • Schingoethe D.J.
        A 100-year review: Total mixed ration feeding of dairy cows.
        J. Dairy Sci. 2017; 100 (29153159): 10143-10150
        • Soriano F.D.
        • Polan C.E.
        • Miller C.N.
        Supplementing pasture to lactating Holsteins fed a total mixed ration diet.
        J. Dairy Sci. 2001; 84 (11768087): 2460-2468
        • Tyrrell H.F.
        • Reid J.T.
        Prediction of the energy value of cow's milk.
        J. Dairy Sci. 1965; 48 (5843077): 1215-1223
        • Vibart R.E.
        • Fellner V.
        • Burns J.C.
        • Huntington G.B.
        • Green Jr., J.T.
        Performance of lactating dairy cows fed varying levels of total mixed ration and pasture.
        J. Dairy Res. 2008; 75 (18701000): 471-480
        • Wales W.J.
        • Marett L.C.
        • Greenwood J.S.
        • Wright M.M.
        • Thornhill J.B.
        • Jacobs J.L.
        • Ho C.K.M.
        • Auldist M.J.
        Use of partial mixed rations in pasture-based dairying in temperate regions of Australia.
        Anim. Prod. Sci. 2013; 53: 1167-1178
        • White S.L.
        • Benson G.A.
        • Washburn S.P.
        • Green Jr., J.T.
        Milk production and economic measures in confinement or pasture systems using seasonally calved Holstein and jersey cows.
        J. Dairy Sci. 2002; 85 (11860126): 95-104
        • Wilkinson J.M.
        • Lee M.R.F.
        • Rivero M.J.
        • Chamberlain A.T.
        Some challenges and opportunities for grazing dairy cows on temperate pastures.
        Grass Forage Sci. 2020; 75 (32109974): 1-17
        • Zanini G.D.
        • Santos G.T.
        • Schmitt D.
        • Padilha D.A.
        • Sbrissia A.F.
        Distribuição de colmo na estrutura vertical de pastos de capim Aruana e azevém anual submetidos a pastejo intermitente por ovinos.
        Cienc. Rural. 2012; 42: 882-887