Feeding concentrate pellets to early-lactation cows consuming spring ryegrass pasture did not reduce methane emissions
K. Garrett
A * , E. Stubbs A , E. Minnee A , K. Verhoek A and J. Kay A
A
Abstract
Feed additives containing methane-reducing compounds are effective where dairy cows are fed a total mixed ration. In pasture-based systems, there are challenges incorporating feed additives into the diet, although one potential delivery mechanism is through inclusion into concentrate supplementary feed. However, the use of supplementary feed encounters seasonal complexities, particularly in spring when high-quality pasture supply often equals feed demand. If feed additives were to become a methane mitigation option for pasture-based systems, it is important to determine the impact of concentrates on dairy cow methane emissions (g/day, g/kg DM, and g/kg milk solids), and system profitability.
This study aimed to evaluate the performance (dry-matter intake (DMI), milk production, and methane emissions) of early lactation dairy cows offered spring perennial ryegrass-based pasture (PAS) or PAS plus 4 kg DM/cow.day of a supplementary concentrate feed (SUP) fed as 2 kg DM/cow during AM and PM milkings.
Forty lactating (89 ± 2 days in milk (DIM), mean ± s.e.m.) Holstein-Friesian dairy cows were randomly assigned to one of two treatments (PAS or SUP). Following a 3-week dietary adaptation period outdoors, cows were housed indoors for 25 days in a Calan Gate System and continued in their treatment groups (PAS or SUP) with fresh pasture fed twice daily. Milk production was recorded daily, milk composition determined weekly, and gas emissions measured daily by using GreenFeed technology.
Cows in the SUP treatment had 7.6% greater total DMI (19.2 vs 17.9 ± 0.3 kg DM/day, P < 0.01) and 6.9% greater milk solids (MS) production (2.02 vs 1.89 ± 0.04 kg MS/day, P = 0.02) than did PAS cows. Cows in the SUP treatment tended to produce more methane (395 vs 376 ± 7 g CH4 g/day, P = 0.08); however, there was no difference in methane yield (g/kg DMI) or intensity (g/kg MS).
When cows are grazing high-quality spring pastures, there is no reduction in methane emission intensity when a concentrate pellet is fed.
The trend for an increase in methane production and the economic cost, must be considered when evaluating supplementary feeds/or feed additives as potential mitigation tools in pasture-based systems.
Keywords: concentrate, dairy cattle, methane, pasture, supplementary feed.
Introduction
In New Zealand, enteric methane (CH4) emissions from dairy cattle account for ~19% of national greenhouse gases (GHG) (Ministry for the Environment 2023), making it an important GHG to reduce to meet emission targets. Methane is a by-product of feed digestion released primarily through eructation (O’Brien et al. 2010), therefore, an animal’s dry-matter intake (DMI) is a key driver of CH4 emissions. While CH4 emissions are closely related to DMI (Grainger et al. 2007), they also vary with feed chemistry, both across different feed types (e.g. fodder beet or supplementary feeds) (Jonker et al. 2016; Haque 2018; Beauchemin et al. 2022), and within the same feed type (e.g. seasonal changes) (Robertson and Waghorn 2002; Koning et al. 2022; Lahart et al. 2024).
Supplementary feeds such as pellets, concentrates, or mixed rations have been proposed as a potential GHG mitigation option in pasture-based systems (Beauchemin et al. 2022; Bosher et al. 2024). These feeds may reduce CH4 emissions through several mechanisms: increasing dietary energy intake and diluting maintenance energy requirements, lowering rumen pH, shortening rumen retention time, and promoting propionate production via higher starch content (Arndt et al. 2022; Beauchemin et al. 2022). Despite these proposed benefits, many supplementary feeds have not been evaluated for their CH4 mitigation potential during periods of high pasture quality and supply (e.g. spring perennial rye-grass with metabolisable energy (ME) of >11.5 MJ/kg DM). Supplementary feeds can also serve as a delivery mechanism for CH4-reducing feed additives (e.g. 3-NOP or Bromoform), especially when fed at milking times to align with existing on-farm practices. Therefore, it is important to understand the effect of using supplementary feed in a pasture-based system during a period that is typically pasture-only, on CH4 emissions and farm economics.
There is limited research on the use of supplementary feed to mitigate CH4 emissions during spring in pasture-based systems. Understanding how feeding concentrates to cows grazing high-quality spring pasture affects CH4 production (g/day), yield (g/kg DMI) and intensity (g/kg milksolids) is essential, both for assessing the CH4 mitigation potential and the feasibility of using supplements as a delivery mechanism for CH4-reducing additives. It is also important to consider the economic cost/benefit of using supplementary feed during the spring period, because this will be a key driver of adoption of mitigation strategies. Marginal milk response (milk increase per kilogram of incremental increase in supplement DMI) is crucial because it directly affects profitability through the price of supplement versus the return on milk (Kay et al. 2023). Evaluating the milk solids (MS) response helps determine the economic viability of supplementary feeds in pasture-based systems, ensuring that the benefits outweigh the costs, and the mitigation will be viable and scalable across the industry.
The objective of the current experiment was to evaluate performance (DMI, milk production, and CH4 emissions) of dairy cows in early lactation eating spring perennial ryegrass-based pasture (PAS) or PAS plus 4 kg DM/cow.day of a supplementary (SUP) concentrate feed (Denver Spring Pellet, Denver Stock Feeds, Palmerston North, New Zealand) fed as 2 kg DM during the AM and PM milking.
Materials and methods
Ethical approval
This study was conducted from September to November 2022, at the DairyNZ Lye Farm (Hamilton, New Zealand; 37°76′S, 175°36′E). All animal manipulations were conducted with approval from the AgResearch Animal Ethics Committee (Application #0574).
Experimental design and treatments
In mid-September, 50 multiparous lactating Holstein-Friesian cows were randomly allocated to one of two treatments for dietary adaption (see section Adaptation), balanced for age (5.4 ± 0.2 year; mean ± s.e.m.), DIM (42.4 ± 2.0), milk production (190 ± 3 L/cow.week), and liveweight (500 ± 6 kg). Dietary treatments were (1) spring perennial ryegrass-based pasture (PAS), and (2) spring perennial ryegrass-based pasture plus 4 kg DM/cow.day of supplementary concentrate (Spring Pellet, Dever Stock Feeds, Palmerston North) fed as 2 kg DM after each of the AM (06:30 h) and PM (14:00 h) milking (SUP; see Grazing Management and Supplementary Feed section for further details). Following the adaptation period, 40 cows (n = 20 per treatment) were selected (on the basis of meeting criteria of 90% target supplement intake) to continue to the measurement period for which cows were housed in the indoor Calan Gate Facility for 25 days. The 40 cows had an average age of 5.3 ± 0.2 years; DIM of 42.2 ± 2, milk production of 189 ± 3 L/cow.week, and liveweight of 496 ± 6 kg.
Grazing management and supplementary feed
During the 3-week adaptation period, animals were managed in their treatment groups (n = 25) grazing a perennial ryegrass and white clover-based pasture with cows in the SUP treatment transitioned at the group level onto the final allocation of 4 kg DM/cow.day of supplementary feed over the first 4 days, by using increments of 1 kg DM per day. Rotational grazing management was used with decisions based on Macdonald and Penno (1998). Briefly, a fresh break of pasture was allocated after each milking to target an intake of approximately 18 kg DM per cow and post-grazing residuals of 1600–1700 kg DM/ha as determined at the group level by pasture disappearance, by using the rising plate meter (Farm works, Palmerston North, New Zealand).
During the first 2 weeks of the adaptation period, the supplementary feed was group fed during a 15-min period on a concrete pad near the milking shed. In the final week of dietary adaptation, the supplementary feed was fed to individual cows during a 15-min period in an adjacent tie-stall facility, with refusals measured twice daily to determine individual supplementary feed DMI. During this 15 min period, cows in the PAS treatment were kept on a standoff pad so all cows (PAS + SUP) returned to their allocated paddocks at the same time. To prevent oestrus occurring while cows were in the Calan Facility, all animals were enrolled in a synchrony program including insertion of a CIDR, established by a veterinarian.
The supplementary feed (Denver Spring Pellet: Denver Stock Feeds, Palmerston North, New Zealand) was selected on the basis of its advertised nutritional specifications, specifically a ME of 11.5 MJ/kg and a protein content of 13%, which were deemed suitable for feeding to lactating cows grazing spring pastures However, the actual ME and protein content of the supplementary feed (following wet chemistry laboratory analysis conducted post-experiment) were lower than specified (see Results section). The daily amount of 4 kg DM/c.day of supplementary feed (a mid-level supplementation amount) was chosen because this is reflective of typical amounts fed via in-shed feeding systems in New Zealand pasture-based systems (2 kg DM per milking) and is deemed an appropriate amount to incorporate feed additives into a pasture-based diet.
The 40 cows selected (n = 20 per treatment) for the measurement period were housed indoors for 25 days in a well-ventilated Calan Gate (American Calan, Northwood, NH) feeding facility. The facility was divided into four quarters, each housing 10 cows throughout the trial period. Animals from both treatment groups were randomly assigned to each quarter and stalls to ensure balanced distribution and minimise location bias. This 25-day period included: a 4-day re-adaptation (to the facility) period (all animals had received prior training to use the facility (June–July 2022) and the GreenFeed units (C-lock, Rapid City, USA; July–August 2022)) and a 21-day measurement period. During the indoor period, cows in the SUP treatment were offered their supplementary feed twice daily following the AM and PM milking in an adjacent tie stall facility on an individual basis with refusals collected twice daily to determine supplementary-feed DMI. Pasture was cut twice daily (immediately prior to feeding), at 7 cm above ground height with a Kuhn PZ 220 mower (Kuhn, Saverne, France) and picked up with a wagon (Bergmann, Ludwig Bergmann, Germany), and all cows had unlimited access to their individual allocation of fresh pasture, fed in equal portions after each of the AM and PM milkings. Each morning, individual pasture refusals were recorded and removed before the new AM allocation of pasture was offered. All cows had access to a GreenFeed unit (C-Lock Inc., Rapid City, USA) and approximately 1.2 kg DM of a bait pellet daily to facilitate CH4 measurements Total DMI was calculated as the daily sum of pasture plus GreenFeed bait, plus supplementary concentrate pellet (SUP treatment) intake. The milking and supplementary feed allocation times remained the same as in the adaptation period. However, after the 15 min of consuming the supplementary feed (SUP) or standoff (PAS), all cows were held on a concrete holding area for a further 30 min to represent the time cows spent walking back to a paddock after milking, as they would on-farm.
All cows were allocated pasture to target total daily intakes of ~18–20 kg DM, similar to that which they had while grazing in the paddock. Individual cow intakes were adjusted on a daily basis on the basis of a rolling average from the previous 3 days to target pasture refusals of approximately 5% (e.g. ~1 kg DM refusal). This allocation ensured that cows were not underfed but reflected management in rotational grazing scenarios. During the indoor period, all animals had access to GreenFeed Technology (one GreenFeed unit per 10 cows with GreenFeed units rotated each week). GreenFeed parameters were set to allow a maximum of six visits per 24-h period, with a minimum of 3.5 h between visits. A visit was set to dispense eight drops of 30 ± 2 g of bait (Country Harvest Alpaca Pellets, Dunstan Nutrition, Cambridge, New Zealand), with 25 s between each drop. The drop size of 30 g was determined using a 10-drop calibration test, performed prior to the measurement period, in which 10 drops of Greenfeed pellets were dispensed, collected, and weighed to calculate a standard drop size. All cows had sufficient GreenFeed visitation for reliable CH4, hydrogen (H2), and carbon dioxide (CO2) measurements, 31 ± 5 (mean ± s.d.) visits per cow per week, meeting the minimum inclusion criteria of 20 visits averaged over 7 days (Manafiazar et al. 2017). Further, both treatments had similar visitation patterns throughout the day, and good representative data of the temporal emission pattern (data not shown).
Feed measurements and sampling
Fresh cut pasture was weighed twice daily for individual cows at ~07:00 hours 14:30 h. As each individual cow’s fresh pasture allocation was weighed, a subsample was taken. These samples were bulked for each feeding event to acquire a representative sample of the pasture fed to all cows twice daily. Bulked pasture samples were then subsampled into three 150 ± 5 g samples, oven dried at 95°C for 72 h and reweighed to determine DM content. Additional subsamples of the bulked feed (~200 and 50 g) were taken. The 200 g sample was stored at −20°C, freeze dried and ground to pass through a 1 mm sieve (Christy and Norris Mill, United Kingdom) for laboratory analysis (Hills Laboratories, Hamilton, New Zealand). The 50 g subsample was bulked at the daily level (AM + PM) and used to determine botanical composition of ryegrass, white clover, other grass species, weeds, and dead material. Sorted botanical samples were dried in a force-draught oven at 95°C for 48 h to determine dry-weight composition. Daily refusals were weighed, and then subsampled by bulking according to size of refusal (0–5, 5–10, 10+ kg wet weight), and DM content was determined following the same procedure as for allocated feed.
Representative samples (800 g) of the allocated concentrate pellets were taken daily. Each daily sample was subsampled into four 200 g samples, three of which were oven dried at 95°C for DM analysis and the remaining sample was stored at −20°C, freeze-dried and ground in a mill (Christy and Norris, Ipswich, UK, 1 mm screen). Samples were bulked by measurement week and analysed for chemical composition by wet chemistry (Hill Laboratories, Hamilton, New Zealand). In addition, any concentrate pellets that remained after the allocated 15 min consumption time were weighed and recorded for the individual animal. If the concentrate pellet refusal was ≥200 g, a 200 ± 5 g sample was taken and DM was determined, otherwise DM was assumed to be the same as for the allocated feed.
Cow measurements
Individual milk yields were recorded at each milking (Westfalia Surge Metatron Milk Meter; GEA Farm Technologies, Bönen, Germany). On two occasions each week (Monday PM and Tuesday AM), representative milk samples (1.25% of total volume) were collected from individual animals using in-line milk meters. Milk samples were refrigerated at 4°C until analysis to determine milk fat, total protein, and lactose concentrations by using a calibrated infra-red milk analyser (Fourier Transform Infrared Spectroscopy (FT120); Foss Electric, Hillerød, Denmark). To calculate the MS response to the additional supplementary pellet, the Greenfeed bait was included in the base diet.
Statistical analysis
Analyses were conducted using R (ver. 4.3.0; 2023; R Core Team 2023). Statistical significance was considered at P ≤ 0.05 and tendencies were considered at 0.05 < P ≤ 0.10. Data were analysed using a mixed model ANOVA using the ‘lme4’ package (Bates et al. 2015). Treatment groups, week, and their interaction were included as fixed effects and animal was included as a random effect. Model diagnostics were evaluated for all models for goodness of fit. Normality of residuals was assessed, and where residuals deviated from normality (e.g. H2), data were square-root transformed to better approximate a normal distribution.
Individual-level data, including daily milk meter recordings and GreenFeed visit data, were used in the analysis. Methane data were averaged sequentially: by tag, week, and hour, and finally by tag across the week to generate weekly means. Liveweight and milk composition were linearly interpolated between measurement points and used to calculate daily averages for derived variables such as CH4 per unit of liveweight (CH4/LW), prior to weekly averaging. Fat and protein-corrected milk (FPCM) was calculated according to the equation FPCM (kg/day) = (0.337 + 0.116 × fat % + 0.06 × protein %) × milk yield (kg/day) (Centraal Veevoederbureau 2022). Feed composition data are summarised using the mean and standard deviation (SD) and are presented as descriptive statistics.
Results
Diet component chemical analysis
The nutritive composition of each of the diet component are displayed in Table 1.
| Nutritive value | Diet component | ||||||
|---|---|---|---|---|---|---|---|
| Pasture | s.d. | Supplement | s.d. | GreenFeed bait | s.d. | ||
| DM (% as-fed) | 16.4 | 2.9 | 88.8 | 0.8 | 88.1 | 0.5 | |
| CP (% DM) | 15.3 | 1.8 | 9.5 | 0 | 13.2 | 0.1 | |
| NDF (% DM) | 41.2 | 3.25 | 34.6 | 0.3 | 46.5 | 0.6 | |
| ADF (% DM) | 23.7 | 1.7 | 21.2 | 0.3 | 26.5 | 0.7 | |
| OM (% DM) | 91.8 | 0.4 | 87.0 | 0.3 | 91.2 | 0.2 | |
| ME (MJ/kg DM) | 12.1 | 0.4 | 9.9 | 0.1 | 10.1 | 0.1 | |
| Nutritive value | Average diet composition by treatment | ||||
|---|---|---|---|---|---|
| PAS | s.d. | SUP | s.d. | ||
| DM (% as-fed) | 20. 8 | 1.5 | 34.7 | 1.5 | |
| CP (% DM) | 15.2 | 1.4 | 14.0 | 1.1 | |
| NDF (% DM) | 41.5 | 2.5 | 40.1 | 2.0 | |
| ADF (% DM) | 23.9 | 1.3 | 23.3 | 1.0 | |
| OM (% DM) | 91.8 | 0.3 | 90.8 | 0.2 | |
| ME (MJ/kg DM) | 12.0 | 0.3 | 11.6 | 0.2 | |
Values presented as mean ± standard deviation (s.d.).
DM, dry matter; CP, crude protein; NDF, neutral detergent fibre; ADF, acid detergent fibre; OM, organic matter; ME, metabolisable energy.
Dry-matter intake and milk production
There was a treatment effect on DMI (Table 2), where cows fed the SUP treatment had a 7.3% greater total DMI (19.2 vs 17.9 kg DM/day; P < 0.01) than did those in the PAS treatment. Cows in the SUP treatment consumed, on average, 3.9 kg DM/day of supplementary concentrate pellet. Conversely, cows in the SUP treatment ate 14.3% less pasture (14.4 vs 16.8 kg DM/day; P < 0.01) than did cows in the PAS treatment (Table 2). All cows consumed the GreenFeed bait; however, cows in the SUP treatment ate 0.13 kg DM/day less GreenFeed bait (0.93 vs 1.06 kg DM/day; P < 0.01) than did cows in the PAS treatment.
| Item | Treatment (TRT) | P-value | |||||
|---|---|---|---|---|---|---|---|
| PAS | SUP | s.e.m. A | TRT | Week | TRT × Week | ||
| Total DMI, mean (kg DM/day) | 17.8 | 19.2 | 0.3 | <0.01 | <0.01 | 0.38 | |
| Pasture intake (kg DM) | 16.8 | 14.4 | 0.3 | <0.01 | <0.01 | 0.19 | |
| Supplement intake (kg DM) | – | 3.9 | 0.1 | – | – | – | |
| GreenFeed bait intake (kg DM) | 1.1 | 0.9 | 0.1 | <0.01 | <0.01 | 0.31 | |
| DMI/LWT (kg DM/kg) | 0.037 | 0.040 | 0.001 | <0.01 | <0.01 | 0.92 | |
| Milk (kg/day) | 24.0 | 26.0 | 0.6 | 0.02 | <0.01 | 0.13 | |
| FPCM (kg/day) | 25.4 | 27.3 | 0.5 | 0.01 | <0.01 | 0.52 | |
| Milk solids (kg/day) | 1.89 | 2.02 | 0.04 | 0.02 | <0.01 | 0.90 | |
| Fat yield (kg/day) | 1.06 | 1.14 | 0.02 | 0.01 | <0.01 | 0.43 | |
| Protein yield (kg/day) | 0.83 | 0.88 | 0.02 | 0.06 | <0.01 | 0.08 | |
Although the effect of week was included in all models to account for potential temporal variation, it did not explain any biologically meaningful patterns in the data. As such, it is not discussed further in the results presented.
Milk production parameters are presented in Table 2. Cows in the SUP treatment produced more (P < 0.05) milk (kg/day; 8.3%), MS (kg/day; 6.9%), and FPCM (kg/day; 7.5%) than did cows in the PAS treatment. This equated to an average MS response of 34.5 g MS/kg DM supplementary feed. Milk composition was similar between treatments, with no differences in fat or protein percentage (P > 0.05). There was an effect of week on milk production but no treatment by week interactions. Milk production (week milk) declined as the week increased (P < 0.01).
Gas emissions
Gas emissions (CH4, H2, and CO2) are presented in Table 3. Cows in the SUP treatment tended (P = 0.08) to produce 5.1% more CH4 per day (395 vs 376 g CH4 per day) than did cows in the PAS treatment (Table 3). There was no treatment effect on CH4 yield (g/kg DMI; P = 0.14), or CH4 intensity (g/kg MS; P = 0.51). In addition, there was no difference (P = 0.16) between treatments in CH4 per kilogram liveweight. There were no treatment by week interactions.
| Item | Treatment (TRT) | P-value | |||||
|---|---|---|---|---|---|---|---|
| PAS | SUP | s.e.m. A | TRT | Week | TRT:week | ||
| CH4 (g/day) | 376 | 395 | 7.3 | 0.08 | <0.01 | 0.78 | |
| CH4 yield (g/kg DMI) | 21.2 | 20.6 | 0.27 | 0.14 | <0.01 | 0.94 | |
| CH4 intensity (g/kg MS) | 202 | 199 | 4.2 | 0.51 | <0.01 | 0.49 | |
| CH4 liveweight (g/kg liveweight) | 0.79 | 0.82 | 0.02 | 0.16 | <0.01 | 0.96 | |
| CO2 (g/day) | 12,176 | 12,364 | 169 | 0.44 | 0.14 | 0.39 | |
| H2 (g/day) | 1.90 | 1.51 | 0.11 | 0.02 | <0.01 | 0.02 | |
There was a treatment effect on H2 emissions, such that the SUP cows produced 20.5% less H2 emissions (1.51 vs 1.90 g/day; P < 0.05) than did the PAS cows. There was a treatment by week interaction for H2 emissions, with cows in the SUP treatment emitting less (P < 0.05) H2 than did cows in the PAS treatment during Weeks 1 and 2; however, there was no difference (P = 0.24) in H2 emissions between treatments for Week 3. There was no treatment or week effect on CO2 emissions.
Discussion
In the current experiment, supplementing pasture-fed cows with 4 kg DM pelleted concentrate during spring did not reduce CH4 intensity (g CH4 per kg MS), and tended towards greater total CH4 production (g per day) than for unsupplemented cows. The lack of a reduction in CH4 intensity is in contrast to Van Wyngaard et al. (2018), and Bosher et al. (2024; see also Bosher et al. (2024) thesis, for further details on this trial), who reported a linear decrease in CH4 intensity, as the amount of supplementary feed in the diet increased (0, 4 and 8 kg DM, and 0, 2, 4 and 6 kg DM/cow.day respectively). Potential reasons for the different effects of supplementary feeding on CH4 intensity are differences in the base pasture diet and the quality of supplementary feeds. Van Wyngaard et al. (2018), and Bosher et al. (2024) fed base pasture diets of lower nutritional quality; characterised by greater NDF content and less ME than those used in the present experiment. Specifically, Van Wyngaard et al. (2018) utilised a summer kikuyu (C4) dominant pasture, and Bosher et al. (2024) fed a combination of very high-mass ryegrass pasture and low-quality pasture silage. In comparison, cows in the current experiment consumed good-quality ryegrass-dominant pasture with approximately 12.1 MJ ME/kg DM. Previous reviews have established that the type of forage significantly influences CH4 emissions from dairy cattle (Deramus et al. 2003; McCourt et al. 2007; Beauchemin et al. 2020, 2022), and seasonal variations also alter these emissions, with spring ryegrass-dominant pastures typically resulting in lower total CH4 intensity and yield than do summer and autumn pastures (Robertson and Waghorn 2002; Lahart et al. 2024). Moreover, the characteristics of the base diet influence the efficacy of potential mitigation strategies, including supplementary feeds and feed additives (Hristov et al. 2013;van Gastelen et al. 2022). Thus, when assessing supplementary feed as a mitigation strategy in pasture-based systems, it is important to consider the base diet and ensure that this reflects realistic on-farm conditions.
Spring perennial ryegrass-based pastures in temperate climates such as New Zealand and Ireland, are typically rich in digestible carbohydrates and lower in fibre content than in other times of the year (O’Neill et al. 2011; Hristov et al. 2013; Box et al. 2017). Although the pasture used in the current study had slightly lower protein content than expected (DairyNZ Limited 2021), potentially owing to reduced nitrogen fertiliser application following a very wet winter/early spring (Box et al. 2017; Martin et al. 2017), the DM%, NDF, and ME values aligned with typical good-quality New Zealand North Island dairy pastures in October/November. Moreover, cow performance indicators, including DMI (3.7% LWT) and milk production (25.4 kg FPCM), supported the nutritional adequacy and energy content of the pastures.
Data from the current study indicated that adding supplementary feeds to high-quality spring pasture diets did not change dietary NDF and ME content, nor did it change ruminal pH, or volatile fatty acid production, and thus did not reduce CH4 intensity (Jonker et al. 2016; Beauchemin et al. 2020). This contrasts with studies where supplementary feed was offered to cows fed a lower-quality forage diet, resulting in changes to dietary composition, energy content, and rumen kinetics, which reduced CH4 intensity (Van Wyngaard et al. (2018) and Bosher et al. (2024)).
The lack of reduction in CH4 intensity, and a tendency for greater total daily CH4 production, may also be attributed to the lower quality of the supplementary feed offered in the current study. Although the supplementary pellet was specified at 11.5 MJ ME/kg DM, wet chemistry analyses indicated a lower energy content of ~10 MJ ME per kg DM. Consequently, it is likely that the tendency for a greater total CH4 production (grams per day) resulted from a greater total DMI, (1.4 kg DM/cow.day) without a concomitant improvement in dietary composition (e.g. ME content) and reduction in CH4 yield (per kg DMI) (Jonker et al. 2016; Beauchemin et al. 2020).
In some experiments, greater total DMI per cow from the addition of supplementary feeds has been associated with reduced CH4 yield, partly owing to faster rumen passage rates, reduced rumen retention time, and digestive fermentation (Arndt et al. 2022; Beauchemin et al. 2022). In addition, when total DMI and energy intake are increased, the reduction in CH4 intensity may be due to a reduction in the proportion of total dietary energy required for maintenance (Capper et al. 2009), such that a greater proportion of energy is partitioned to MS production, reducing CH4 production per kilogram MS (Arndt et al. 2022; Beauchemin et al. 2022). However, in the current experiment, even though the supplemented cows consumed 3.9 kg DM concentrated pellet, there was a small increase in the total DM and energy intake (7% and 4% respectively). This was largely due to pasture substitution, such that cows in the SUP treatment substituted 0.65 kg DM pasture for every 1 kg DM of supplement eaten (65% substitution).
The high substitution rate (65%) of the current study was not unexpected because substitution of pasture for supplementary feed is greatest during spring (Stockdale 2000; Macdonald and Roche 2024). Consequently, the incremental response of MS production is at its seasonal lowest (Stockdale 2000; Poole 2018). Macdonald and Roche (2024) reported that the low MS response during spring is due to the small net increase in DM and ME intake, with substitution rate being greatest alongside high pasture quality. It must be noted that in on-farm scenarios, there are many feed, animal and system factors that influence substitution rates and MS responses to supplementary feeds (Bargo et al. 2003). For example, lower rates of substitution and greater MS responses are associated with lower pre-supplement post-grazing residuals. In the current study, cows were housed indoors and fed via individual feeding cubicles, thus, although cows were offered ~20 kg DM pasture to target a 5–10% refusal in an attempt to mimic grazing situations targeting best practice post-grazing residuals (e.g. 1500–1600 kg DM), it is difficult to mimic on-farm grazing behaviour. When cows are housed indoors, they tend to consume more feed (Dohme-Meier et al. 2014), which would be reflective of cows being offered a higher pasture allowance and thus leaving behind higher residuals if they were grazing in situ, and this may have contributed to the high rates of substitution. If grazing cows were offered supplementary feeds during other seasons, and target post-grazing residuals were met, this may result in a reduced substitution rate and, consequently, greater energy intake and marginal MS response, further resulting in reduced CH4 intensity. The MS and CH4 response to supplementary feed inclusion in different seasons (which different forage quality and composition) requires further research to determine the suitability of using supplementary feeds as a mitigation strategy (Bargo et al. 2003).
The trend toward increased daily CH4 production in the current experiment when supplementary feed was offered has important implications for using these supplementary feeds as carriers for CH4-reducing additives (e.g. 3-NOP, Asparagopsis taxiformis). In spring, pasture quality and quantity often meet herd nutrient and energy demands, and thus many pasture-based farm systems do not routinely offer supplementary feeds during this period. Therefore, if feed additives are to be used successfully in pasture-based systems, the feed additive and the supplementary feed must significantly reduce daily CH4 production, below baseline pasture-only diets, to justify the use of the additive from an environmental perspective.
Additionally, although CH4 is a key GHG target in dairy production, it is important to consider the total daily GHG emissions (i.e. CH4 + nitrous oxide + carbon dioxide). Market-driven GHG targets (e.g. set by Nestle and Fonterra) include embedded CO2 emissions associated with supplementary feed manufacturing, land use change, transport and storage (Reisinger et al. 2017). Thus, the additional CO2 emissions incurred by using supplementary feeds must be evaluated against any potential CH4 reduction benefits to assess the sustainability and market outcomes of different strategies.
Another consideration is the economic viability of the use of supplementary feed as a mitigation per se, or as a carrier for CH4-reducing additives. In the current study, the high substitution rate resulted in a low marginal MS response (34 g MS per kg DMI). Assuming a milk price of NZ$8.00/kg MS, this equates to a return of NZ$0.28 per kg DM supplementary feed offered. The supplementary feed used in the current study retailed for $0.60 per kg DM, and it is well established that there are additional costs to the system when supplementary feeds are purchased (Neal and Roche 2019; DairyNZ 2025). Data from the last 15 years of DairyNZs DairyBase farm performance database indicated that for every NZ$1 spent on imported feed, total system costs increased by NZ$1.66 for the Waikato region (Neal and Roche 2019; DairyNZ 2025). Therefore, from an economic perspective, the use of supplementary feed in the current study was unprofitable.
Although this is a component study, the data highlighted that even if supplementary feeds did reduce GHG metrics (i.e. different season, or different type/amount of supplementary feed), the economical feasibility of the mitigation needs to be carefully considered. The primary driver for profitable use of supplementary feeds is the MS response (Kay et al. 2023), which in turn is dependent on pasture management. This can be particularly challenging during spring, when pasture peak growth and quality can exceed peak energy demand in some regions (Clark et al. 2007; Macdonald et al. 2008; Macdonald and Roche 2024).
Conclusions
Offering supplementary concentrate feed at 4 kg DM/day to early-lactation cows consuming good-quality ryegrass-based pastures tended to increase total enteric CH4 production (5%) and did not reduce CH4 yield or intensity. The effectiveness of supplementary feed as a GHG mitigation strategy (on its own or as a delivery mechanism for CH4-reducing feed additives) needs to be determined across different base diets, and supplementary feed types and amounts, alongside evaluation of total farm-level GHG emissions, MS response, and economic viability.
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Declaration of funding
This work is part of the Less Methane research program, which is funded by New Zealand dairy farmers through DairyNZ Inc.
References
Arndt C, Hristov AN, Price WJ, McClelland SC, Pelaez AM, Cueva SF, Oh J, Dijkstra J, Bannink A, Bayat AR, Crompton LA, Eugène MA, Enahoro D, Kebreab E, Kreuzer M, McGee M, Martin C, Newbold CJ, Reynolds CK, Schwarm A, Shingfield KJ, Veneman JB, Yáñez-Ruiz DR, Yu Z (2022) Full adoption of the most effective strategies to mitigate methane emissions by ruminants can help meet the 1.5°C target by 2030 but not 2050. Proceedings of the National Academy of Sciences 119(20), e2111294119.
| Crossref | Google Scholar |
Bargo F, Muller LD, Kolver ES, Delahoy JE (2003) Invited review: production and digestion of supplemented dairy cows on pasture. Journal of Dairy Science 86(1), 1-42.
| Crossref | Google Scholar |
Bates D, Mächler M, Walker SC (2015) Fitting linear mixed-effects models using lme4. Journal of Statistical Software 67(1), 1-48.
| Crossref | Google Scholar |
Beauchemin KA, Ungerfeld EM, Eckard RJ, Wang M (2020) Review: Fifty years of research on rumen methanogenesis: lessons learned and future challenges for mitigation. Animal 14, S2-S16.
| Crossref | Google Scholar |
Beauchemin KA, Ungerfeld EM, Abdalla AL, Alvarez C, Arndt C, Becquet P, Benchaar C, Berndt A, Mauricio RM, McAllister TA, Oyhantçabal W, Salami SA, Shalloo L, Sun Y, Tricarico J, Uwizeye A, De Camillis C, Bernoux M, Robinson T, Kebreab E (2022) Invited review: current enteric methane mitigation options. Journal of Dairy Science 105(12), 9297-9326.
| Crossref | Google Scholar |
Bosher T, Della Rosa MM, Khan MA, Sneddon N, Donaghy D, Jonker A (2024) Methane emissions intensity in grazing dairy cows fed graded levels of concentrate pellets. New Zealand Journal of Agricultural Research 67(3), 296-302.
| Crossref | Google Scholar |
Box LA, Edwards GR, Bryant RH (2017) Diurnal changes in the nutritive composition of four forage species at high and low N fertiliser. Journal of New Zealand Grasslands 79, 111-118.
| Crossref | Google Scholar |
Capper JL, Cady RA, Bauman DE (2009) The environmental impact of dairy production: 1944 compared with 2007. Journal of Animal Science 87(6), 2160-2167.
| Crossref | Google Scholar |
Clark DA, Caradus JR, Monaghan RM, Sharp P, Thorrold BS (2007) Issues and options for future dairy farming in New Zealand. New Zealand Journal of Agricultural Research 50(2), 203-221.
| Crossref | Google Scholar |
DairyNZ (2025) DairyBase. Available at https://www.dairynz.co.nz/business/dairybase/ [accessed 23 May 2025]
Deramus HA, Clement TC, Giampola DD, Dickison PC (2003) Methane emissions of beef cattle on forages: Efficiency of grazing management systems. Journal of Environmental Quality 32(1), 269-277.
| Crossref | Google Scholar |
Dohme-Meier F, Kaufmann LD, Görs S, Junghans P, Metges CC, Van Dorland HA, Bruckmaier RM, Münger A (2014) Comparison of energy expenditure, eating pattern and physical activity of grazing and zero-grazing dairy cows at different time points during lactation. Livestock Science 162(1), 86-96.
| Crossref | Google Scholar |
Grainger C, Clarke T, McGinn SM, Auldist MJ, Beauchemin KA, Hannah MC, Waghorn GC, Clark H, Eckard RJ (2007) Methane emissions from dairy cows measured using the sulfur hexafluoride (SF6) tracer and chamber techniques. Journal of Dairy Science 90(6), 2755-2766.
| Crossref | Google Scholar |
Haque MN (2018) Dietary manipulation: a sustainable way to mitigate methane emissions from ruminants. Journal of Animal Science and Technology 60(1), 15.
| Crossref | Google Scholar |
Hristov AN, Oh J, Firkins JL, Dijkstra J, Kebreab E, Waghorn G, Makkar HPS, Adesogan AT, Yang W, Lee C, Gerber PJ, Henderson B, Tricarico JM (2013) SPECIAL TOPICS-Mitigation of methane and nitrous oxide emissions from animal operations: I. A review of enteric methane mitigation options 1. Journal of Animal Science 91, 5045-5069.
| Crossref | Google Scholar |
Kay J, Boom C, Robinson K, Chestnut K (2023) A quantitative case study assessment of the biophysical and economic effects of three different feed management systems in a Northland farmlet trial. Journal of New Zealand Grasslands 84, 147-154.
| Crossref | Google Scholar |
Koning L, Holshof G, Klop A, Klootwijk C (2022) Enteric methane emission of dairy cattle in relation to (fresh) grass quality annual report 2: 2021 – results of a multi-year grazing trial on methane emissions during grazing, summer stable feeding and grass silage feeding. Wageningen Livestock Research, Wageningen, Netherlands.
Lahart B, Buckley F, Herron J, Fitzgerald R, Fitzpatrick E, Galvin N, Shalloo L (2024) Evaluating enteric methane emissions within a herd of genetically divergent grazing dairy cows. Journal of Dairy Science 107(1), 383-397.
| Crossref | Google Scholar |
Macdonald KA, Penno JW (1998) Management decision rules to optimise milksolids production on dairy farms. Proceedings of the New Zealand Society of Animal Production 58, 132-135.
| Google Scholar |
Macdonald KA, Roche JR (2024) Farm systems research at Ruakura – a 60-year legacy underpinning profitable and sustainable pasture-based dairy systems. New Zealand Journal of Agricultural Research 67(2), 105-222.
| Crossref | Google Scholar |
Macdonald KA, Penno JW, Lancaster JAS, Roche JR (2008) Effect of stocking rate on pasture production, milk production, and reproduction of dairy cows in pasture-based systems. Journal of Dairy Science 91(5), 2151-2163.
| Crossref | Google Scholar |
Manafiazar G, Zimmerman S, Basarab JA (2017) Repeatability and variability of short-term spot measurement of methane and carbon dioxide emissions from beef cattle using GreenFeed emissions monitoring system. Canadian Journal of Animal Science 97(1), 118-126.
| Crossref | Google Scholar |
Martin KE, Bryant RH, Hodge S, Edwards GR (2017) Effect of autumn regrowth interval and nitrogen fertiliser on dry matter yield and plant characteristics of six forage species. Journal of New Zealand Grasslands 79, 67-72.
| Crossref | Google Scholar |
Neal M, Roche JR (2019) Profitable and resilient pasture-based dairy farm businesses in New Zealand. Animal Production Science 60(1), 169-174.
| Crossref | Google Scholar |
O’Brien D, Shalloo L, Grainger C, Buckley F, Horan B, Wallace M (2010) The influence of strain of Holstein-Friesian cow and feeding system on greenhouse gas emissions from pastoral dairy farms. Journal of Dairy Science 93(7), 3390-3402.
| Crossref | Google Scholar |
O’Neill BF, Deighton MH, O’Loughlin BM, Mulligan FJ, Boland TM, O’Donovan M, Lewis E (2011) 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. Journal of Dairy Science 94(4), 1941-1951.
| Crossref | Google Scholar |
Robertson LJ, Waghorn GC (2002) Dairy industry perspectives on methane emissions and production from cattle fed pasture or total mixed rations in New Zealand. Proceedings of the New Zealand Society of Animal Production, New Zealand Society of Animal Production, Palmerston North, New Zealand. pp. 213–218.
Stockdale CR (2000) Levels of pasture substitution when concentrates are fed to grazing dairy cows in northern Victoria. Australian Journal of Experimental Agriculture 40(7), 913-921.
| Crossref | Google Scholar |
R Core Team (2023) ‘R: a Language and environment for statistical computing.’ (R Foundation for Statistical Computing: Vienna, Austria) Available at https://wwwR-projectorg/ [accessed October 2023]
van Gastelen S, Dijkstra J, Heck JML, Kindermann M, Klop A, de Mol R, Rijnders D, Walker N, Bannink A (2022) Methane mitigation potential of 3-nitrooxypropanol in lactating cows is influenced by basal diet composition. Journal of Dairy Science 105(5), 4064-4082.
| Crossref | Google Scholar |
Van Wyngaard JDV, Meeske R, Erasmus LJ (2018) Effect of concentrate level on enteric methane emissions, production performance, and rumen fermentation of Jersey cows grazing kikuyu-dominant pasture during summer. Journal of Dairy Science 101(11), 9954-9966.
| Crossref | Google Scholar |
