Assessing the dry matter intake and enteric methane emissions of pre-partum dairy cows offered grass clover or grass-only silage from two different silage systems

Over the winter period, the low grass growth and availability in pasture-based dairy systems results in animals being housed and predominantly fed a diet of grass silage. There is limited availability of methane (CH₄) data evaluating the impact of forage type on dairy cows over the pre-partum period. The objective of the current experiment was to evaluate the impact of feeding grass clover (GC) silage and grass-only (GO) silage on the dry matter intake (DMI) and enteric CH₄ emissions of dairy cows pre-partum. A complete randomised block design was utilised for the 6-week experiment over two winter periods: from December 2020 to January 2021 and from December 2021 to January 2022. In each year, 30 non-lactating pregnant dairy cows were randomly allocated to two treatments (n = 15). In both years, cows in the GC treatment were offered grass clover bale silage, while cows in the GO treatment were offered grass-only pit silage. The DMI and gaseous emissions of individual animals were monitored daily using Hokofarm RIC (roughage intake control) feed stations and the GreenFeed technology. GC silage consistently had greater (p< 0.05) organic matter (OM) digestibility and lower (p< 0.05) neutral detergent fibre (NDF) and acid detergent fibre (ADF) contents when compared with GO silage. Cows in the GC treatment had significantly greater (p< 0.05) total DMI (TDMI) compared with cows in the GO treatment. The daily CH₄ emissions (in grams per day) were not affected by treatment; however, cows in the GC treatment had reduced (p< 0.05) CH₄ yield (in grams per kilogram TDMI). Offering dairy cows GC silage over the pre-partum period resulted in greater DMI with reduced CH₄ yield when compared with cows offered GO silage.

Introduction

Globally, agriculture is a significant source of greenhouse gas (GHG) emissions, accounting for 9.3 million tonnes of carbon dioxide (CO₂) equivalent, with 39% related to enteric methane (CH₄) from ruminant livestock (FAO, 2020). Ruminant livestock systems serve as a vital source of human edible protein and play a crucial role in food security for growing global populations, particularly in developing nations (Arndt et al., 2022). The agricultural sectors in Ireland (36.8%) and New Zealand (49%) are the largest contributors to the total GHG emissions (EPA, 2023; MOE, 2023). Within the EU, there has now been major sectoral targets implemented in an effort to sharing regulations to reduce total GHG emissions (EEA et al., 2021). The Irish agricultural sector has been tasked with lowering the total GHG by 25% relative to the 2018 levels (EEA et al., 2021; Teagasc, 2023). Effective CH₄ mitigation strategies are crucial to achieve this target as CH₄ contributes 62% to the agricultural GHG emissions in Ireland (Teagasc et al., 2019; EPA, 2023). The majority of CH₄ research in grazing dairy systems has been predominantly derived during the lactating period, with limited data available during the pre-partum period (Lahart et al., 2023).

Pasture-based dairy systems, globally, are characterised by the diet of the animal, which mainly comprises grazed grass, although variability in the grass supply due to climatic conditions is inevitable across the year (Delaby et al., 2021). During the winter period, cows are dried off, housed or kept outdoors, and then enter the non-lactating period. The non-lactating dry period is an integral part of the lactation cycle for pasture-based dairy cows due to the replenishment of the mammary glands, the body reserves, foetal maturation, and the preparation for the onset of lactation (Butler et al., 2011; Mann et al., 2015). Research over the pre-partum period has been primarily focused on periparturient health, nutritional performance, and lactation performance (Richards et al., 2020), with less focus on enteric CH₄ emissions. The quality of the diet fed during the pre-partum period is crucial for meeting the nutritional requirements and maintaining the metabolic status of dairy cows over the pre-partum and post-partum periods (Butler et al., 2011; Janovick et al., 2022). Pasture-based dairy systems such as those in Ireland and New Zealand traditionally offer cows a high-fibre diet pre-partum, predominantly grass silage (O’Brien et al., 2018) or forage crops (Pinares-Patiño et al., 2007; Jonker et al., 2017; Waghorn, 2018). The nutrient demand of cows for energy pre-partum will be dependent on the dairy system and the cow type, although it has been reported that offering dairy cows a low-energy, high-fibre diet is adequate for meeting the nutrient requirements during this time (Butler et al., 2011; Richards et al., 2020).

Grass silage is the second largest component of the diet of Irish dairy cows, accounting for 18% of the annual feed budget on a fresh weight (FW) basis (O’Brien et al., 2018). It is an inexpensive source of feed compared with concentrates (Finneran et al., 2012; Doyle et al., 2022) and aids feed security by reducing reliance on external sources of imported feed on dairy farms. Grass silage is generally harvested as precision chop or baled silage (Mceniry et al., 2011) during periods of high grass growth over the year. Traditionally, grassland swards in Ireland contain perennial ryegrass (PRG) (Lolium perenne L.) due to its high yield potential, nutritive value, persistency (Hearn et al., 2021), and low production cost (Doyle et al., 2022). In order to achieve adequate silage yields and forage digestibility, grass-only silage swards require a certain level of chemical nitrogen (N) (Keady et al., 2000). The inclusion of legumes such as white clover (Trifolium repens L.) within grassland swards has increased in recent years not only due to improvements in animal performance when offered but also to restrictions in the application of chemical N fertiliser under the EU Nitrates Directive (Council Directive 19/676/EEC) (Egan et al., 2017; Herron et al., 2021). Grass clover swards are seen as an effective mitigation strategy for environmental emissions, such as ammonia and nitrous oxide, thereby increasing their use on Irish dairy farms (Herron et al., 2021). For this reason, the current study focused on comparing the impact of feeding grass-only (GO) silage and grass clover (GC) silage over the pre-partum period to non-lactating dairy cows.

Feeding GC silage has been reported to improve the dry matter intake (DMI) and milk production over the lactation due to clover having greater organic matter (OM) digestibility compared with PRG (Johansen et al., 2017, Johansen et al, 2018). Previous research has reported that the mitigation of CH₄ is possible by altering the composition of the diet (Loza et al., 2021), forage processing (Beauchemin et al., 2008), lipid supplementation (Boland et al., 2021; Muñoz et al., 2021), and by the inclusion of feed additives (Haisan et al., 2017; Van Gastelen et al., 2024). Among these strategies, the diet composition is of primary importance as it shows a substantial difference in CH₄ production (Soder and Brito, 2023). Using high-quality forages that are more digestible can decrease CH₄ emissions as they are fermented more efficiently, leading to less CH₄ production per unit of feed intake (Van Gastelen et al., 2024). In this context, the beneficial effect of clover inclusion in silage swards on mitigating enteric CH₄ is still under investigation (Bica et al., 2022). The pre-partum period has been reported to have lower levels of CH₄ emissions when compared with the lactation period as a result of the lower energy demand for the production and suppression of DMI due to foetal maturation (Jonker et al., 2017; Lyons et al., 2018). There is a clear seasonal variation in the CH₄ emissions of dairy cows, which is related to the level of DMI achieved (Ulyatt et al., 2002; Lahart et al., 2023), although both the DMI and enteric CH₄ emissions are dependent on the digestibility of the diet that is offered to dairy cows (Jonker et al., 2017).

Research reporting on the levels of CH₄ emitted when GC silage is offered to dairy cows has been predominantly during the lactating period (Baldinger et al., 2011; Brask et al., 2013; Johansen et al., 2017), with its impact over the pre-partum period still under investigation.

The objectives of the current study were to evaluate the digestibility of GC and GO silage and to assess their impact on the animal DMI, the enteric CH₄ emission, and the post-partum milk production when offered to dairy cows during the pre-partum period. The hypothesis is that GC silage will have a greater digestibility, resulting in differences in both the animal DMI and enteric CH₄ emissions compared with GO silage during the pre-partum period.

Materials and methods

Animals and experimental design

The current experiment was conducted at the Teagasc, Animal and Grassland Innovation Centre, Moorepark, Fermoy, Co., Cork, Ireland, over two 6-week periods from December 20, 2020, to January 29, 2021 (year 1), and from December 20, 2021, to January 31, 2022 (year 2). Within each year, 30 non-lactating pregnant spring calving dairy cows were selected from the Moorepark dairy herd. In year 1, all 30 cows were multiparous, while in year 2, the group consisted of four primiparous and 26 multiparous cows. In years 1 and 2, all cows were dried off on December 7, 2020, and on December 2, 2021. The cows were then blocked for parity and breed (Holstein Friesian and Holstein × Jersey) and balanced for expected calving date, bodyweight, body condition score (BCS), and economic breeding index (EBI) (Table 1). The cows were then randomly assigned to one of two treatments: the grass and clover silage (GC) treatment or the grass-only silage (GO) treatment. Within the GC treatment, cows were offered grass and clover bale silage for the duration of the experiment in both years, whereas cows within the GO treatment were offered grass-only pit silage for the duration of the experiment. All cows were housed in a purpose-built cubicle shed for the duration of the experiment, with the cubicles fitted with a 15-mm rubber mat (Huber Technik, Erding, Germany), which were cleaned and limed daily, with cows having free access to water. All forages were fed once daily at 9 a.m. using a Keenan Mechfibre 350 mixer wagon (Alltech Farming Solutions Ltd., Borris, Co., Carlow, Ireland). Both treatments were offered silage ad libitum by ensuring a 5%–10% refusal rate. The cubicle shed was equipped with 16 individual feed stations used to monitor the daily feed intake of cows over the course of the experiment. In both years, a 7-day acclimation period was implemented prior to commencing the experiment to ensure adequate animal use of each feed station. All silage was placed directly into 16 individual automatic feed stations (eight for the GC treatment and eight for the GO treatment) (RIC Feed-Weigh Trough, Hokofarm Group BV, Marknesse, Netherlands), similar to those described by Kelly et al. (2019).

Table 1

Table 1. Pre-experimental data (mean ± SD) on breed, lactation number, days in calf, expected calving date, economic breeding index, pre-experimental bodyweight, and pre-experimental body condition score of dairy cows used in year 1 (2020) and year 2 (2021) of the experiment.

Cows were removed from the experiment up to 7 days prior to calving and were placed in a calving pen on a grass silage diet, with the average calving dates being February 17, 2021, and February 6, 2022, in years 1 and 2, respectively. Silage treatments were stopped post-partum, with all the cows managed as one group. Post-partum, all of the cows were kept indoors and offered ad libitum grass silage plus 3 kg of concentrate for an average of 1 week, depending on the weather and ground conditions. All animals were turned out to grass and managed in a rotational grazing system and offered a diet of, on average, 13 kg dry matter (DM) grazed grass plus 3 kg DM of concentrate for the first 12 weeks of lactation. During periods of inclement weather, 4 days in year 1 and 1 day in year 2, all cows were housed and offered a diet of ad libitum grass silage and 3 kg of concentrate.

Silage preparation

The areas harvested for the GC silage were part of a systems grazing study (Fitzpatrick, 2023). All areas cut for the GC silage were grazed previously during the first rotation (March) and then closed for a period of 8 weeks before the harvesting of silage. The GC swards received 50 kg of chemical N over two splits, in February and March, and then received a further 60 kg N/ha 8 weeks prior to cutting. The GO silage was harvested from an area used for two cut silage systems with no animal grazing. The GO swards received 80 kg chemical N/ha 8 weeks prior to cutting. Prior to harvesting, the pre-cutting herbage mass of the GC and GO swards was determined using a method described by O’donoVan and Dillon (1999), i.e., by placing a quadrat (0.25 m²) at three random locations per hectare, cutting all grass within the quadrat to 4 cm from the ground level using a Gardena hand shears (Gardena GmbH, Ulm, Germany). Fresh cut grass from within each quadrat was placed in a bag and weighed. Once weighed, a 100-g subsample was taken and placed into an oven at 90°C for 16 h to determine the DM content. The sward clover content from the GC swards was determined using a method described by Egan et al. (2017). All of the bale GC silage for the GC treatment was harvested from swards that have been established for 5 years, with the swards containing PRG with a sown mixture of PRG (c.v Astonenergy and Tyrella in a 50:50 mixture sown at 27.2 kg/ha) and white clover (Trifolium repens cv. Chieftain and Crusader in a 50:50 mixture sown at 5 kg/ha). The GC was harvested in year 1 on June 8, 2020, and in year 2 on May 30, 2021. In year 1, the pre-cutting herbage mass was 3,838 kg DM/ha and the average clover content was 25% ± 13.4%. In year 2, these values were 4,107 kg DM/ha and 23% ± 11.9%, respectively. The GC silage was cut and left to wilt for 24 h to ensure that an adequate DM content (~30% DM) was achieved. Once wilted, the cut grass was raked into a 1.2-m swath and then baled and wrapped at harvesting using a McHale fusion baler (McHale, Ballinrobe, Co., Mayo, Ireland). All bale grass clover silage for the GC treatment was harvested from swards that were 5 years established, swards contained a mixture of PRG (c.v Astonenergy, Tyrella in a 50:50 mixture sown at 27.2 kg/ha) and white clover (Trifolium repens cv. Chieftain and Crusader in a 50:50 mix sown at 5 kg/ha) The GO silage was harvested in year 1 on May 20, 2020, and in year 2 on May 28, 2021. The pre-cutting herbage mass in year 1 was 5,350 kg DM/ha and that in year 2 was 5,050 kg DM/ha. The GO silage was cut and allowed to wilt for 24 h to ensure that an adequate DM content (~30% DM) was achieved at harvesting. Once wilted, the grass silage was raked into a 1.2-m swath and then harvested using a self-propelled precision chop harvester. All GO silage was ensiled in a silage clamp, rolled, and covered with a polythene sheet to ensure adequate conditions for fermentation (Teagasc, 2016).

Chemical composition

Over the course of the experiment, fresh silage samples were collected twice weekly from the RIC feed stations. From each sample, 100 g was weighed and dried at 40°C for 48 h to determine the silage DM. Fresh concentrate samples were collected twice weekly from the GreenFeed emissions monitoring (GEM) unit (C-Lock Inc., Rapid City, SD, USA). From each sample, 100 g was dried at 60°C for 48 h to determine the DM content. In both years, 12 silage and six concentrate samples were collected. Once dried, the silage and concentrate samples were milled through a 1-mm screen (Cyclotec 1093 Sample Mill, Foss Electric, Hillerød, Denmark) prior to chemical analysis. The ash content of each milled sample was determined by burning the sample at 550°C for 16 h using a muffle furnace (Nabertherm, GmbH, Lilienthal, Germany). The crude protein concentration was determined using a Leco FP-428 nitrogen analyser (Leco Australia Pty Ltd., Castle Hill, NSW, Australia) as described by Sweeney (1989). The neutral detergent fibre (NDF) and acid detergent fibre (ADF) contents were determined using an Ankom 2000 Fiber Analyzer as outlined by ANKOM Technology Corporation (Macedon, NY, USA) (Van Soest et al., 1991). The organic matter digestibility of the silage samples was determined using an in vitro NDF cellulase procedure, as described by Morgan et al. (1989). The gross energy content of the milled silage and concentrate samples was determined using a Parr 6050 bomb calorimetry system (Parr Instrument Company, Moline, IL, USA).

Animal measurements

Bodyweight and body condition score

Bodyweight and BCS were recorded at three time points in year 1 (i.e., December 6, January 8, and February 11) and four time points in year 2 (i.e., December 15, December 20, January 6, and February 5) of the experiment. The cows were individually weighed using an electronic portable weighing scale with the Animal Performance software package (Gallagher Europe, Groningen, the Netherlands). BCS was assessed by a trained professional using a scale from 1 to 5 (where 1 = emaciated and 5 = extremely fat), increasing in 0.25 increments, as described by Lowman et al. (1976).

Dry matter intake

The individual animal FW intake was recorded daily using the RIC feed-weight trough. Each feed station was equipped with an antenna above the access gate, which read a cow’s radio frequency identification (RFID) ear tag (HDX EID Tag, Allflex Livestock Intelligence, Dallas, TX, USA) prior to allowing access to the silage. This ensured that GO treatment cows could not access the GC treatment silage, and vice versa. Once a cow visited a feed station, a feed event is recorded, with the duration of the visit, the starting weight (in kilograms FW), the ending weight (in kilograms FW), and the total roughage (in kilograms FW) removed from each feed station recorded. Prior to feeding fresh silage, all refused silage was removed from each feed bin. The daily silage dry matter intake (SDMI) for each individual cow was calculated by summing the total FW of silage consumed at each feed event per day, followed by converting the daily FW silage intake to SDMI using the corresponding DM% of the silage, which was then averaged across each week of the experiment. The fresh concentrate intake of individual cows was monitored using the GEM unit, and this was converted to dry weight using the DM% of the corresponding weekly concentrate sample to calculate the concentrate DM intake (CDMI). The individual cow SDMI was combined with the CDMI to derive the TDMI.

Methane emissions

Over the course of the experiment, the daily CH₄, CO₂, and hydrogen (H₂) emissions were monitored using the GEM unit. Cows had free access to one GEM unit for the duration of the experiment in both years. One GEM unit was allocated to all cows 7 days prior to commencing the experiment in both years. The entrance chute for cows to enter the GEM unit was modified to ensure that each individual cow had access to the unit at a given time. The GEM unit measures the spot breath samples of individual cows as they feed on a known amount of concentrate at a given time point over the day. In both years, a standard 14% crude protein dairy concentrate was used (Dairygold Co-Operative Society Limited, Cork, Ireland). The concentrate pellet consisted of barley (16.5 g/kg DM), maize (10.3 g/kg DM), wheat feed (5 g/kg DM), rapeseed extract (16.4 g/kg DM), maize gluten feed (14.1 g/kg DM), maize distillers (2.8 g/kg DM), soya hulls (12.3 g/kg DM), palm kernel extract (10.2 g/kg DM), molasses (5.1 g/kg DM), delactosed permeate (3.1 g/kg DM), and minerals and vitamins (4.2 g/kg DM). Once a cow visited the GEM unit, 34 g FW of the concentrate was dispensed at 25-s intervals for a total of eight drops per visit. The duration between each visit was set at 6- to 8-h intervals (Figure 1), with the cows allowed to visit the GEM unit for a maximum of four visits per day to minimise the level of concentrate fed (average = 0.7 ± 0.12 kg DM/cow per day). As described by Della et al, 2021, every 3 days, automatic gas calibrations were performed, where known amounts of zero calibration gas containing nitrogen (N₂) followed by span calibration gasses containing CH₄ and CO₂ are released to ensure accuracy of the non-dispersive infrared spectrometry sensor. Prior to beginning the experiment, then once monthly and at the end of the experiment, manual CO₂ recoveries were performed to calibrate the airflow. Here, a known amount of CO₂ is released directly into the feed face of the GEM unit over three timed intervals (300 s). The change in the weight of the CO₂ canister was then compared with the amount measured by the GEM unit calculating the percentage recovered (average CO₂ recovery = 99.5% ± 0.95%.)

Figure 1

Figure 1. (A) Total visits (n) at each 4 hour interval to the Greenfedd emissions monitoring unit and (B) average daily CH₄ emissions (g/d) at each 4 hour interval over the day for pre-partum dairy cows within the grass only silage treatment in both year 1 and year 2 of the experiment.

Milk production

The post-partum milk production of individual cows was monitored over the first 12 weeks of lactation. Daily individual milk yields (in kilograms) were measured at each milking at 0700 and 1500 hours (Dairymaster, Causeway, Ireland). The milk composition was determined weekly using one milk sample taken from a weighted consecutive a.m. and p.m. milking. The milk samples were analysed for protein, fat, and lactose concentrations using mid-infrared spectroscopy analysis (Milkoscan 203, FOSS Electric, Hillerød, Denmark).

Data handling and calculations

All data cleaning and the handling of data were performed using R software version 4.2.2 (R Core Team). A total of 120,933 feed events from the feed bins were recorded over the two consecutive years. All raw intake data used were cleaned for outliers by performing regression equations of the visit duration (in seconds) versus the roughage intake (in kilograms) or the roughage intake (in kilograms) versus the visit duration (in seconds) for each feed event. A linear model was then utilized to calculate the predicted value and residuals from each of the measured values. The residuals were then divided by the standard deviation of the visit duration (in seconds) or the roughage intake (in kilograms). Feed events with a residuals/SD greater than ±3 were replaced with a predicted value from the linear model.

All raw emissions data were checked for irregularities before being finalised. Once finalised, raw emission data were checked for outliers using a regression equation described by Coppa et al. (2021). This process involved performing regression equations of CO₂ versus CH₄ and CH₄ versus CO₂. These equations gave the predicted and residual values of each measurement. The standard deviation of each gas was determined and then used to divide the residuals of the linear model. Any measurement with residuals/SD greater than ±3 was removed from the dataset.

The methane yield (in grams per kilogram TDMI) for individual cows was derived by dividing the average daily CH₄ emissions for each cow with their corresponding TDMI. The silage and concentrate DMI measurements were multiplied by the corresponding gross energy content of the feed and summed together to calculate the gross energy intake (GEI) of individual cows. The GEI of individual cows was used to calculate the percentage of GEI converted to CH₄ energy (Eugène et al., 2019). This was calculated by multiplying each cow’s daily CH₄ (in kilograms per day) by the energy (55.65 MJ) density found in a kilogram of CH₄. The total megajoules (MJ) of methane energy emitted by a cow was then divided by the cow’s GEI, calculating the percentage of GEI as CH₄ energy (Liu et al., 2017).

The predicted DMI and CH₄ emissions were calculated following country-specific IPCC tier 2 methodology (IPCC, 2019), as outlined in Ireland’s national inventory for the calculation of national CH₄ emissions (Herron et al., 2022; EPA, 2022). Predicted DMI (in kilograms) was calculated using the net energy (NE) system derived by Jarrige et al. (1986) and modified by O’Mara, 2006 following the INRA methodology for the calculation of dairy cow NE requirements for specific feeding systems. In Ireland, when dairy cows are housed and on a grass silage-based diet, a country-specific equation developed by Yan et al. (2000) is used to calculate the enteric CH₄ emissions, as outlined in Ireland’s national inventory (O’Mara, 1996; EPA, 2022).

$$\mathrm{C}{\mathrm{H}}_4= \mathrm{D}\mathrm{E}{\mathrm{I}}^{*}[0.096+ (0.035\times S_{DMI}/T_{DMI} )] - {2.298}^{*}(\mathrm{F}\mathrm{L} - 1)$$

where DEI is the digestible energy intake (in megajoules per day), S_DMI is the silage DM intake (in kilograms per day), T_DMI is the total DM intake (in kilograms per day), and FL is the feeding level (multiples of maintenance energy requirement). The digestible energy intake was derived using the OM digestibility of the corresponding offered silage during the experimental period. The individual predicted DMI and CH₄ emissions were then compared with the individual measured DMI and CH₄ emissions for each individual animal.

Statistical analysis

All data analysis was conducted using SAS 9.4 (SAS Institute Inc., Cary, NC, USA). Data were checked for normality using PROC UNIVARIATE and analysed with a linear mixed model, which allowed for repeated measurements using PROC MIXED. Forage data were analysed using the following statistical model:

$$Y_{ijk}=\mathit{\mu }+T_i+W_j+Y_k+T_i\times W_j+T_i\times Y_k+e_{ijk}$$

where Y_ijk is the dependent variable, μ is the overall mean, T_i is the effect of treatment (i = GC or GO), W_j is the effect of week (j = 1–6), Y_k is the effect of year (k = 1 or 2), T_i × W_j is the interaction between treatment and week, T_i × Y_k is the interaction between treatment and year, and e_ijk is the residual error.

The bodyweight, average daily gain, and BCS were analysed using the following model:

$$Y_{wjimnk}=\mathit{\mu }+B_w+P_j+T_i+C_m+P_n+Y_k+T_i\times P_n+T_i\times Y_k+X_{wjimnk}+e_{wjimnk}$$

where Y_wjink is the dependent variable, μ is the overall mean, B_w is the breed (w = 1 or 2), P_j is parity (j = 1 or 2 or 3+), T_i is the effect of treatment (i = GC or GO), C_m denotes cow within treatment (m = 1–15), P_n is the effect of week (n = 1–6), Y_k is the effect of year (k = 1–2), T_i × W_n is the interaction between treatment and week, T_i × Y_k is the interaction between treatment and year, X_wjimnk denotes the pre-experimental bodyweight variables, and e_wjimnk is the residual error.

The enteric CH₄ emissions, SDMI, CDMI, TDMI, and the predicted CH₄ emissions and predicted DMI were analysed using the following model:

$$\begin{array}{l} \hfill \\ Y_{wjimnk}=\mathit{\mu }+B_w+P_j+T_i+C_m+W_n+Y_k+T_i\times W_n+T_i\times Y_k+DT{C}_{wjimnk}+e_{wjimnk}\hfill \end{array}$$

where Y_wjink is the dependent variable, μ denotes the overall mean, B_w is the breed (w = 1 or 2), P_j is parity (j = 1 or 2 or 3+), T_i is the effect of treatment (i = GC or GO), C_m denotes the cow within treatment (m = 1–15), W_n is the effect of week (n = 1–6), Y_k is the effect of year (k = 1–2), T_i × W_n is the interaction between treatment and week, T_i × Y_k is the interaction between treatment and year, DTC_wjimnk denotes the days until calving, and e_wjimnk is the residual error.

The post-partum milk production, bodyweight, and BCS were analysed using the following model:

$$Y_{wjimnk}=\mathit{\mu }+B_w+P_j+T_i+C_m+W_n+Y_k+T_i\times Y_k+e_{wjimnk}$$

where Y_wjink is the dependent variable, μ is the overall mean, B_w represents breed (w = 1 or 2), P_j denotes parity (j = 1 or 2 or 3+), T_i is the effect of pre-partum treatment (i = GC or GO), C_m denotes cow within the pre-partum treatment (m = 1–15), W_n is the effect of week of lactation (n = 1–12), Y_k is the effect of year (k = 1–2), T_i × Y_k is the interaction between pre-partum treatment and year, and e_wjimnk is the residual error.

Differences between the predicted and the measured DMI and enteric CH₄ emissions were analysed using a PROC TTEST procedure for individual measurements within each treatment.

All results reported are least square means values, with significant effect declared when p ≤ 0.05 and tendencies declared from 0.05 ≤ p ≥ 0.10.

Results

Chemical composition

There were significant (p< 0.001) effects of silage treatment on all chemical composition variables, except for the crude protein content (Table 2). The chemical composition of silage was not consistent across experimental years, and the DM content of GC silage was lower (p< 0.05) in year 2 when compared with year 1 of the experiment (199.1 and 374.1 g/kg, respectively). GC silage had greater (p< 0.001) OM digestibility when compared with GO silage across both years of the experiment (716.6 and 676.8 g/kg DM, respectively). The ash content was not significantly (p > 0.05) different for each silage treatment; however, the ash contents of both silage types increased (p< 0.001) in year 2 when compared with year 1. The crude protein content of the GO silage decreased by 35.2 g/kg DM from year 1 to year 2, resulting in a significant (p< 0.05) treatment-by-year interaction in year 2. The GC silage had a lower (p< 0.001) NDF content when compared with the GO treatment silage (484.3 and 558.6 g/kg DM, respectively). The acid detergent fibre content was greater (p< 0.001) in the GO silage when compared with that in the GC silage (335.1 and 292.3 g/kg DM, respectively). The gross energy content of the silage was not affected (p > 0.05) by treatment; however, the GE content was lower (p< 0.001) in year 2 compared with that in year 1 (17.7 and 18.6 MJ/kg DM, respectively).

Table 2

Table 2. Chemical composition of grass clover and grass-only silages offered to dairy cows over the pre-partum period in years 1 and 2 analysed for treatment, year, and their relative interaction.

Bodyweight, body condition score, and dry matter intake

The bodyweight, BCS, and the DMI parameters are reported in Table 3. Bodyweight tended to be greater (p = 0.071) for cows within the GC treatment when compared with cows in the GO treatment. Over the course of the experiment, the average daily gain was not affected by treatment. The BCS were not affected by treatment (p > 0.05) across both years of the experiment.

Table 3

Table 3. Bodyweight, average daily gain, body condition score, silage dry matter intake (DMI), total DMI, gross energy intake (GEI), and predicted DMI of dairy cows within the grass clover and grass-only treatments during the pre-partum period analysed for treatment, year, and their relative interaction.

There was no effect of treatment on CDMI. Silage DMI was affected by treatment, with the GC treatment having greater (p< 0.05) SDMI compared with the GO treatment (10.9 and 8.1 kg/cow per day, respectively). There were differences in SDMI across treatments and years: in year 1, the GC treatment consumed 3.7 kg of additional (p< 0.001) SDMI when compared with the GO treatment. In year 2 of the experiment, the SDMIs were similar between treatments, highlighting a significant (p< 0.001) treatment-by-year interaction. The GO treatment cows had greater (p< 0.05) SDMI (+1.4 kg) in year 2 compared with year 1 (11.0 and 8.7 kg/cow per day, respectively). The total DMI was affected by treatment, with the key differences across treatments and years similar to those of SDMI (Table 3). The GC treatment had greater (p< 0.05) GEI compared with the GO treatment (214.6 and 191.6 MJ/cow per day, respectively). There were differences in the GEI across treatments and years, with the GO treatment cows having greater (p< 0.001) GEI in year 2 compared with year 1 (229.9 and 183.8 MJ/cow per day, respectively), while the GC treatment cows showed a decline in GEI (p< 0.001) from year 1 to year 2 (252.8 and 196.1 MJ/cow per day, respectively).

The predicted DMIs for both treatments are reported in Table 3. There was no difference for predicted DMI between treatments (9.5 and 9.9 kg, respectively). The predicted DMI was greater (p< 0.05) for all animals in year 2 when compared with the animals in year 1 of the experiment (9.1 and 10.3 kg/cow per day, respectively). For both treatments, the predicted DMIs were significantly lower (p< 0.05) than the measured DMIs. For the GC treatment, the predicted DMI was reduced (p< 0.001) by 2.4 kg, while the GO treatment exhibited a predicted DMI reduction (p< 0.05) of 0.5 kg.

Enteric methane emissions

There was no effect (p > 0.05) of treatment on the daily CH₄ emissions. The methane yield (in grams per kilogram TDMI) was affected (p< 0.05) by silage treatment, with the cows offered GC silage having lower CH₄ yields (in grams per kilogram TDMI) compared with the cows in the GO treatment (20.9 and 22.0 g/kg TDMI, respectively). There was a significant (p< 0.001) effect of year for both treatments due to the greater CH₄ yield (in grams per kilogram TDMI) in year 2 when compared with that in year 1 of the experiment (20.0 and 22.9 g/kg TDMI, respectively). Across the 2-year experiment, the greatest difference in CH₄ yield was evident in year 1 of the experiment, with the GC treatment having a 17% lower CH₄ yield, with no difference reported in year 2, highlighting a significant (p< 0.001) treatment-by-year interaction. There was a significant (p< 0.001) treatment-by-year interaction for the percentage of GEI as CH₄ energy, with the GC treatment having a lower (p< 0.05) percentage of GEI as CH₄ energy when compared with the GO treatment in year 1. For cows on both treatments, the percentage of GEI as CH₄ energy increased from year 1 to year 2 (5.7% and 6.6%, respectively).

The predicted CH₄ emissions were not significantly (p > 0.05) affected by silage treatment (Table 4). The predicted CH₄ emissions were greater (p< 0.05) than the measured CH₄ emissions, with 10.9 and 26.6 g/cow greater predicted CH₄ emissions in the GC and GO treatments, respectively, compared with the measured CH₄ emissions.

Table 4

Table 4. Methane emissions, carbon dioxide emissions, methane yield, percentage gross energy intake (GEI) as methane energy, and predicted methane emissions of dairy cows within the grass clover and grass-only treatments during the pre-partum period analysed for the effect of treatment, year, and their relative interactions.

Post-partum bodyweight, body condition score, and milk production

The post-partum milk production was not affected by the silage treatment imposed in the pre-partum period for all measured variables (Table 5). In addition, the silage treatment during the pre-partum period did not have a significant (p > 0.05) effect on the bodyweight and BCS during the first 12 weeks of lactation post-partum (Table 5). However, there was a significant (p< 0.05) effect of year on post-partum BCS.

Table 5

Table 5. Bodyweight, body condition score, and milk production of dairy cows over the first 12 weeks of the lactation within the pre-partum grass clover and grass-only treatments analysed for treatment, year, and their relative interaction.

Discussion

The findings of the current experiment support the hypothesis that the silage system impacts the forage digestibility, which in turn impacts both the animal DMI and enteric CH₄ emissions in pre-partum dairy cows. Grass silage is well acknowledged to be a cost-effective source of feed in the Irish pasture-based dairy system and is subsequently fed over the housed winter period (Butler, 2014; Doyle et al., 2022). During the pre-partum period, the digestibility of the diet offered to dairy cows impacts the level of DMI and the metabolic status (Richards et al., 2020); however, the level of CH₄ emissions during this time has received less attention. It is widely acknowledged that the digestibility of the offered diet dictates the level of DMI achieved, which is a key driver of the CH₄ emissions in dairy cows (Parnian-Khajehdizaj et al., 2023).

Throughout the experiment, key differences in the silage quality became evident between treatments and years. The greater DM content in the GC silage resulted in the GC treatment cows having greater DMI, which resulted in greater CH₄ emissions in year 1. The inclusion of clover in swards has been reported to increase the OM digestibility due to the lower NDF content (Egan et al., 2018), which was clear in the current experiment findings. Offering forages with a lower fibre content can impact the physical effective NDF, which is a key driver of DMI and rumen fermentation (Grant, 2023). This is attributed to the differing lignin contents and compositions, resulting in variations in the cell wall concentrations between PRG and clover (Jung, 1989). GC silage has been reported to increase DMI, particularly due to the lower fibre content increasing the rate of digestion (Lind et al., 2020). In year 2 of the experiment, no differences were evident in the SDMI, as both silage treatments had similar DM, NDF, and OM digestibility, highlighting that forage digestibility had the greatest impact on individual animal SDMI, over sward species or silage system (Bica et al., 2022). To maintain a high level of white clover in grazing swards, it is recommend to keep a lower herbage mass (Barthram and Grant, 1994; Murray et al., 2022), resulting in a lower pre-cutting herbage mass in the current study for the GC silage sward as the GC sward was grazed prior to being closed for silage harvesting (Egan et al., 2018). In contrast, as the GO silage sward was not grazed before silage harvesting, it had a greater pre-cutting herbage mass. This resulted in the GO silage having lower OM digestibility and greater fibre content (Humphreys and O’Kiely, 2006; Pang et al., 2021; Álvarez et al., 2022). Offering forages with greater fibre content has been reported to have lower rumen degradability within the rumen, which in turn increases the rumen retention times, thus reducing the animal DMI (Gregorini et al., 2013; McCarthy et al., 2023). This response was clear in cows on the GO treatment, as the GO silage had greater fibre content and lower OM digestibility, which reduced the DMI (−3.7 kg DM/cow), whereas the GC treatment silage had a greater level of OM digestibility, which is associated with greater rumen passage rates, resulting in greater propionate concentrations within the rumen (Van Gastelen et al., 2019). Legumes such as white clover, due to their different plant morphology, have been reported to have a greater rumen passage rate due to the smaller particle size in the rumen when compared to PRG with long fibres that reduce the passage rate and the rumen effective degradability (Dewhurst et al., 2009).

It is well acknowledged that, during the pre-partum period, dairy cows have lower levels of DMI and enteric CH₄ emissions when compared with cows in the lactating period (Ulyatt et al., 2002; Lyons et al., 2018; Meese et al., 2020). The levels of CH₄ emissions reported over the dry period in the current experiment, 200–254 g/cow per day, are similar to those reported by Ferris et al. (2017) and Jonker et al. (2017), who reported CH₄ emissions of 173–215 g/cow per day when cows were offered grazed grass or grass silage and beet. In the current experiment, similar to those in Pinares-Patiño et al. (2007), the enteric CH₄ emissions were correlated (R² = 0.46) with the DMI over the dry period (Figure 2). This was evident in year 1 of the experiment, where the GC treatment emitted 18% greater daily CH₄ emissions compared with the GO treatment, due to a 26% increase in the TDMI. In year 2, differences in the CH₄ emissions and the CH₄ yields were not as apparent due to reductions in the DM content, the OM digestibility, and the NDF and ADF contents in both silage treatments. Greater intake of digested OM is positively correlated with daily CH₄ due to the greater rumen fill and rumen passage rate when feeding forage-based diets (Parnian-Khajehdizaj et al., 2023). The overall contribution of legumes to CH₄ emissions in ruminant livestock is still under investigation, with contrasting responses dependent on the animal species, the production system, and their proportions in the diet (Bica et al., 2022). Numerous studies have reported elevations in the performance of dairy cows with the inclusion of clover in the diet, mainly due to the enhanced forage digestibility, which increases ruminal passage (Bica et al., 2022), thus increasing the intake capacity of the ruminant animal (Brask et al., 2013; Van Dorland et al., 2007; Dineen et al., 2018). Similar to the current experiment, Van Dorland et al. (2007) reported a 64% lower fibre content in white clover silage, elevating the DMI of dairy cows, which resulted in greater CH₄ emissions, although no difference in the CH₄ yield was evident (in litres per kilogram OM digested) when compared with red clover and PRG silage.

Figure 2

Figure 2. Relationship between (A) methane emissions (g/cow/d) and total dry matter intake (kg/cow/d) and (B) methane yield (g/kg TDMI) and organic matter intake (kg/cow/d) of dairy cows within both the grass clover and grass only treatments with each ● and ▲ represents one measurement point for each animal across the pre-partum experimental period in both years.

It has been reported that the digestibility and energy content of the diet offered during the pre-partum period affect the subsequent post-partum milk production and the metabolic status of dairy cows (Butler et al., 2011; Mann et al., 2015). Janovick et al. (2022) reported that dairy cows fed above their energy requirement pre-partum showed a reduced milk production post-partum, with greater incidences of metabolic disorders due to the greater lipid deposition on the liver on day 14 post-partum as these cows had greater BCS at calving. In the current experiment, the GC treatment had greater levels of TDMI over the experimental period, particularly in year 1; however, no effects on post-partum milk production were evident mainly due to there being no differences in the BCS during the experimental period and over the first 12 weeks of lactation. In pasture-based dairy systems, the diet offered during the pre-partum period does not have the main goal of improving milk production (Butler, 2014) when compared with confined production systems (Mann et al., 2015; Richards et al., 2020). The benefit in the DMI reported in the present study would suggest that offering high-quality GC silage to dairy cows is more beneficial during the lactation to help improve milk production, particularly in the spring or autumn, when grass growth is reduced, in pasture-based dairy systems (O’Brien et al., 2018; Walsh et al., 2023). Post-partum dairy cows experience negative energy balance due to the onset of milk production, therefore supporting the nutrient demands with highly digestible forages being key to maximising production in pasture-based dairy cows (Butler, 2014; Delaby et al., 2021). According to reports, the supplementation of grass silage over highly digestible grazed grass during the lactation has been shown to reduce animal performance, particularly with regard to the milk protein concentration (Claffey et al., 2019; Walsh et al., 2023). Therefore, improving the quality of the offered silage could potentially improve the DMI and, in turn, improve or maintain animal performance when low grass availability necessitates silage supplementation.

The substantial positive association between DMI and enteric CH₄ emissions renders the estimation of DMI indispensable for accurately predicting enteric CH₄ emissions (Appuhamy et al., 2016). Ferris et al. (2017) reported a stronger relationship between DMI and CH₄ emissions compared with that between bodyweight and CH₄ emissions when predicting CH₄ from non-lactating dairy cows offered grazed grass. The NE system used to predict DMI (Jarrige et al., 1986) does not take into account the rumen fill or the intake capacity of the cow with the corresponding forage digestibility, which can result in an overestimation of the prediction of CH₄ emissions, as reported in the current study. The use of predicted DMI to calculate daily CH₄ emissions resulted in 11% greater predicted CH₄ emissions in the GO treatment when compared with the measured CH₄ emissions (Table 4). The methodology used in the current experiment is highly dependent on the fill value and the digestible energy content of the consumed forages (Binggeli et al., 2022). Moraes et al, 2014 highlighted that models that account for dietary NDF and ether extract are effective in accurately predicting enteric CH₄ over the non-lactating period in dairy cows. Greater levels of DMI result in increased rumen fermentation, which can lead to higher CH₄ emissions (Hristov et al., 2013). The current study demonstrated a positive correlation between TDMI and daily CH₄ (R² = 0.47) (Figure 2). However, both silage types had contrasting OM digestibility content, which would result in different rumen passage rates (McCarthy et al., 2023). The rumen fermentation kinetics and rumen fill are highly dictated by both the animal bodyweight and the digestibility of the offered forages (Moraes et al., 2014), explaining the contrasting results between the predicted and the calculated DMIs and enteric CH₄ emissions in the current experiment.

Data on forage digestibility and herd-averaged DMI have the potential to be commercially available in ruminant livestock systems in Ireland (O’Mara, 2006). The estimation of CH₄ emissions using IPCC methods is dependent on the data availability in a specific country. The use of tier 2 calculations requires information on animal categories, feeding systems, production systems, and manure management (IPCC, 2019). When all elements connected to CH₄ are understood, a country-specific tier 3 methodology can be utilised to create in-depth descriptive data on feeding systems that link the conversion factors to the calculated energy requirements of the specific animal (Eugène et al., 2019). The current experiment offers valuable insights into the impact on the animal DMI and CH₄ emissions when GC bale silage or GO pit silage is offered to dairy cows during the pre-partum period in pasture-based dairy systems. The use of GC bale silage during the pre-partum period is not advantageous as the substantial increase in DMI due to greater digestibility has no benefit on milk production, but results in greater CH₄ during the pre-partum period. The digestibility of the offered silage plays a pivotal role in the levels of DMI and CH₄ emissions in dairy cows; therefore, considerations should be taken when feeding silage types with superior digestibility.

Conclusions

This study evaluated the digestibility of GC silage and GO silage, focusing on their effects on the DMI, the enteric CH₄ emissions, and the post-partum milk production of dairy cows during the pre-partum period. GC silage had greater OM digestibility and lower fibre content, which resulted in greater DMI when offered to GC treatment cows. This resulted in the GC treatment cows having higher CH₄ emissions, but a lower CH₄ yield, than the GO treatment cows. No benefit in post-partum milk production was evident from offering GC silage during the pre-partum period. The study also highlighted discrepancies in the current national inventory calculations, which overestimated CH₄ emissions by up to 10% for both silage types. Overall, the study provides valuable insights into the impact of silage systems and forage quality on enteric CH₄ emissions and offers essential data on DMI and CH₄ emissions over the pre-partum period, which are crucial for the development of effective CH₄ mitigation strategies in dairy systems.

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Ethics statement

The animal study was approved by Teagasc Animal Ethics Committee (TAEC) (TAEC247/2019). The study was conducted in accordance with the local legislation and institutional requirements.

Author contributions

MK: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Resources, Software, Visualization, Writing – original draft, Writing – review & editing. BL: Data curation, Formal analysis, Writing – review & editing. JH: Data curation, Formal analysis, Writing – review & editing. TB: Conceptualization, Funding acquisition, Supervision, Writing – review & editing. CF: Data curation, Investigation, Writing – review & editing. ME: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – review & editing.

Funding

The author(s) declare financial support was received for the research, authorship, and/or publication of this article. The authors would like to thank the Irish Dairy Levy Funding administered by Dairy Research Ireland, the Teagasc Walsh Scholarship scheme,VistaMilk Science Foundation Ireland research centre and the University College Dublin for providing the financial support, which facilitated the current study.

Acknowledgments

The authors would like to thank D. Hennessey, E. Fitzpatrick, J.P. Murphy and M. Liddane for their technical assistance and the staff of Moorepark Research Farm (Animal & Grassland Research and Innovation Centre, Teagasc Moorepark, Ireland) for their care, assistance and support of the experimental cows and procedures.

Conflict of interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Álvarez C., Nielsen N., Weisbjerg M., Volden H., Eknæs M., Prestløkken E. (2022). High-digestible silages allow low concentrate supply without affecting milk production or methane emissions. J. Of Dairy Sci. 105, 3633–3647. doi: 10.3168/jds.2021-21479

Crossref Full Text | Google Scholar

Appuhamy J., France J., Kebreab E. (2016). Models for predicting enteric methane emissions from dairy cows in north america, europe, and Australia and New Zealand. Global Change Biol. 22, 3039–3056. doi: 10.1111/Gcb.13339

Crossref Full Text | Google Scholar

Arndt C., Hristov A. N., Price W. J., Mcclelland S. C., Pelaez A. M., Cueva S. F., et al. (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. Proc. Of Natl. Acad. Of Sci. 119, e2111294119. doi: 10.1073/Pnas.2111294119

Crossref Full Text | Google Scholar

Baldinger L., Baumung R., Zollitsch W., Knaus W. F. (2011). Italian ryegrass silage in winter feeding of organic dairy cows: forage intake, milk yield and composition. J. Of Sci. Of Food And Agric. 91, 435–442. doi: 10.1002/jsfa.v91.3

Crossref Full Text | Google Scholar

Barthram G. T., Grant S. A. (1994). Seasonal variation in growth characteristics of lolium perenne and trifolium repens in swards under different managements. Grass And Forage Sci. 49, 487–495. doi: 10.1111/J.1365-2494.1994.Tb02026.X

Crossref Full Text | Google Scholar

Beauchemin K., Mo K., O’mara F., Mcallister T. (2008). Nutritional management for enteric methane abatement: A review. Aust. J. Of Exp. Agric. - Aust. J. Exp. Agr 48, 21–27. doi: 10.1071/Ea07199

Crossref Full Text | Google Scholar

Bica R., Palarea-Albaladejo J., Lima J., Uhrin D., Miller G. A., Bowen J. M., et al. (2022). Methane emissions and rumen metabolite concentrations in cattle fed two different silages. Sci. Rep. 12, 5441. doi: 10.1038/S41598-022-09108-W

PubMed Abstract | Crossref Full Text | Google Scholar

Binggeli S., Lapierre H., Lemosquet S., Ouellet D. R., Pellerin D. (2022). Comparison of feed evaluation models on predictions of milk protein yield on québec commercial dairy farms. J. Of Dairy Sci. 105, 3997–4015. doi: 10.3168/Jds.2021-21182

Crossref Full Text | Google Scholar

Boland T. M., Pierce K. M., Kelly A. K., Kenny D. A., Lynch M. B., Waters S. M., et al. (2021). Feed intake, methane emissions, milk production and rumen methanogen populations of grazing dairy cows supplemented with various C 18 fatty acid sources. Animals 10, 2380. doi: 10.3390/Ani11020280. Correction: Boland, T.M.; Et AlAnimals: An Open Access Journal From Mdpi, 11, 280.

Crossref Full Text | Google Scholar

Brask M., Lund P., Hellwing A., Poulsen M., Weisbjerg M. (2013). Enteric methane production, digestibility and rumen fermentation in dairy cows fed different forages with and without rapeseed fat supplementation. Anim. Feed Sci. And Technol. 184, 67–79. doi: 10.1016/j.anifeedsci.2013.06.006

Crossref Full Text | Google Scholar

Butler S. (2014). Nutritional management to optimize fertility of dairy cows in pasture-based systems. Animal 8, 15–26. doi: 10.1017/S1751731114000834

PubMed Abstract | Crossref Full Text | Google Scholar

Butler M., Patton J., Murphy J. J., Mulligan F. J. (2011). Evaluation of A high-fibre total mixed ration as A dry cow feeding strategy for spring-calving holstein friesian dairy cows. Livestock Sci. 136, 85–92. doi: 10.1016/J.Livsci.2010.08.003

Crossref Full Text | Google Scholar

Claffey A., Delaby L., Galvin N., Boland T. M., Egan M. (2019). The effect of spring grass availability and grazing rotation length on the production and quality of herbage and milk in early spring. J. Of Agric. Sci. 157, 434–448. doi: 10.1017/S0021859619000613

Crossref Full Text | Google Scholar

Coppa M., Jurquet J., Eugène M., Dechaux T., Rochette Y., Lamy J.-M., et al. (2021). Repeatability and ranking of long-term enteric methane emissions measurement on dairy cows across diets and time using greenfeed system in farm-conditions. Methods (San Diego Calif.) 186, 59–67. doi: 10.1016/J.Ymeth.2020.11.004

PubMed Abstract | Crossref Full Text | Google Scholar

Delaby L., Buckley F., Mchugh N., Blanc F. (2021). Characteristics of robust animals for grass-based production systems. Irish J. Of Agric. And Food Res. 59, 246-257. doi: 10.15212/Ijafr-2020-0147

Crossref Full Text | Google Scholar

Della M. R., Jonker A., Waghorn G. (2021). A review of technical variations and protocols used to measure methane emissions from ruminants using respiration chambers, sf6 marker dilution and greenfeed, to facilitate global integration of published data. Anim. Feed Sci. And Technol. 279, 115018. Available at: https://hdl.handle.net/11440/8519.

Google Scholar

Dewhurst R. J., Delaby L., Moloney A., Boland T. M., Lewis E. (2009). Nutritive value of forage legumes used for grazing and silage. Irish J. Of Agric. And Food Res. 48, 167–187. Available at: www.jstor.org/stable/20720367.

Google Scholar

Dineen M., Delaby L., Gilliland T., McCarthy B. (2018). Meta-analysis of the effect of white clover inclusion in perennial ryegrass swards on milk production. J. Of Dairy Sci. 101, 1804–1816. doi: 10.3168/jds.2017-12586

Crossref Full Text | Google Scholar

Doyle P., Tubritt T., O’donoVan M., Crosson P. (2022). “Economic value of grass and grass white clover,” in Proceedings of the swards for the future conference & Workshop teagasc Moorepark. Ed. Hennessy D. (Co. Cork, Ireland: Teagasc).

Google Scholar

EEA E. E. A., Graichen J., Gores S., Schmid C., Nicole M. (2021). Trends and projections under the effort sharing legislationin Ed. Agency E. E. Overview on developments and drivers. European topic centre on climate change and energy Boeretang 200, B-2400 Mol, Belgium: European Topic Centre on Climate change mitigation and energy.

Google Scholar

Egan M., Galvin N., Hennessy D. (2018). Incorporating white clover (Trifolium repens L.) into perennial ryegrass (Lolium perenne L.) swards receiving varying levels of nitrogen fertilizer: effects on milk and herbage production. J. Of Dairy Sci. 101, 3412–3427. doi: 10.3168/Jds.2017-13233

Crossref Full Text | Google Scholar

Egan M., Lynch M. B., Hennessy D. (2017). Including white clover in nitrogen fertilized perennial ryegrass swards: effects on dry matter intake and milk production of spring calving dairy cows. J. Of Agric. Sci. 155, 657–668. doi: 10.1017/S0021859616000952

Crossref Full Text | Google Scholar

EPA E. P. A. (2022). “Irelands’s national inventory report 2022: greenhouse gas emissions 1990-2020,” in Agency E. E. (ed.). Reported to the united nations framework convention on climate chnage. PO Box 3000, Johnstown Castle, Co. Wexford, Ireland, Y35 W821: Environmental Protection Agency.

Google Scholar

EPA E. P. A., Environmental Protection Agency, E (2023). Ireland provisional greenhouse gas emissions In: ENVIRONMENTAL PROTECTION AGENCY, E. (ed.). (Johnstown Castle Estate, Co. Wexford Enviromental Protection Agency).

Google Scholar

Eugène M., SauVant D., Nozière P., Viallard D., Oueslati K., Lherm M., et al. (2019). A new tier 3 method to calculate methane emission inventory for ruminants. J. Of Environ. Manage. 231, 982–988. doi: 10.1016/J.Jenvman.2018.10.086

Crossref Full Text | Google Scholar

FAO (2020). “Emissions due to agriculture. Global, regional and country trends 2000–2018,” in Faostat analytical brief series (Rome: FAOSTAT Analytical Brief Series).

Google Scholar

Ferris C. P., Jiao H., Murray S., Gordon A. W., Carson A. F. (2017). Methane emissions from non-lactating pregnant dairy cows while grazing. Livestock Sci. 206, 121–124. doi: 10.1016/J.Livsci.2017.10.003

Crossref Full Text | Google Scholar

Finneran E., Crosson P., O’kiely P., Shalloo L., Forristal D., Wallace M. (2012). Stochastic simulation of the cost of home-produced feeds for ruminant livestock systems. J. Of Agric. Sci. 150, 123–139. doi: 10.1017/S002185961100061x

Crossref Full Text | Google Scholar

Fitzpatrick E. (2023). A comparative study into dietary strategies for optimising the efficiency of dairy cow performance (Queen’s University Belfast).

Google Scholar

Grant R. J. (2023). Symposium review: physical characterization of feeds and development of the physically effective fiber system. J. Of Dairy Sci. 106, 4454–4463. doi: 10.3168/Jds.2022-22419

Crossref Full Text | Google Scholar

Gregorini P., Minnee E. M. K., Griffiths W., Lee J. M. (2013). Dairy cows increase ingestive mastication and reduce ruminative chewing when grazing chicory and plantain. J. Of Dairy Sci. 96, 7798–7805. doi: 10.3168/Jds.2013-6953

Crossref Full Text | Google Scholar

Haisan J., Sun Y., Guan L., Beauchemin K. A., Iwaasa A., Duval S., et al. (2017). The effects of feeding 3-nitrooxypropanol at two doses on milk production, rumen fermentation, plasma metabolites, nutrient digestibility, and methane emissions in lactating holstein cows. Anim. Production Sci. 57, 282–289. doi: 10.1071/An15219

Crossref Full Text | Google Scholar

Hearn C., Egan M., Berry D., Geoghegan A., O’leary M., Lynch M., et al. (2021). Agronomic performance of ten perennial ryegrass varieties on commercial grassland farms. J. Of Agric. Sci. 159, 604–614. doi: 10.1017/S0021859621000927

Crossref Full Text | Google Scholar

Herron J., Hennessy D., Curran T. P., Moloney A., O’Brien D. (2021). The simulated environmental impact of incorporating white clover into pasture-based dairy production systems. J. Of Dairy Sci. 104, 7902–7918. doi: 10.3168/Jds.2020-19077

Crossref Full Text | Google Scholar

Herron J., O’Brien D., Shalloo L. (2022). Life cycle assessment of pasture-based dairy production systems: current and future performance. J. Of Dairy Sci. 105, 5849–5869. doi: 10.3168/Jds.2021-21499

Crossref Full Text | Google Scholar

Hristov A. N., Oh J., Firkins J. L., Dijkstra J., Kebreab E., Waghorn G., et al. (2013). Special topics-mitigation of methane and nitrous oxide emissions from animal operations: I. A review of enteric methane mitigation options1. J. Of Anim. Sci. 91, 5045–5069. doi: 10.2527/Jas.2013-6583

Crossref Full Text | Google Scholar

Humphreys J., O’Kiely P. (2006). Amount and quality of grass harvested for first-cut silage for differing spring-grazing frequencies of two mixtures of perennial ryegrass cultivars with contrasting heading date. Grass And Forage Sci. 61, 77–88. doi: 10.1111/j.1365-2494.2006.00511.x

Crossref Full Text | Google Scholar

Ipcc, I. P. O. C. C (2019). Refinement to the 2006 ipcc guidelines for national grennhouse gas inventories. In: Agriculture, F. A. O. L. U (ed.). Switzerland: Intergovernmental Panel on Climate Change.

Google Scholar

Janovick N., Dann H., Loor J., Drackley J. (2022). Prepartum dietary energy intake alters hepatic expression of genes related to peroxisome proliferator-activated receptor and inflammation in peripartal dairy cows. J. Of Dairy Sci. 105, 8069–8086. doi: 10.3168/jds.2021-21669

Crossref Full Text | Google Scholar

Jarrige R., Demarquilly C., Dulphy J. P., Hoden A., Robelin J., Beranger C., et al. (1986). The inra “Fill unit” System for predicting the voluntary intake of forage-based diets in ruminants: A review. J. Of Anim. Sci. 63, 1737–1758. doi: 10.2527/Jas1986.6361737x

Crossref Full Text | Google Scholar

Johansen M., Lund P., Weisbjerg M. R. (2018). Feed intake and milk production in dairy cows fed different grass and legume species: A meta-analysis. Animal: Int. J. Of Anim. Bioscience 12, 66–75. doi: 10.1017/S1751731117001215

Crossref Full Text | Google Scholar

Johansen M., Søegaard K., Lund P., Weisbjerg M. R. (2017). Digestibility and clover proportion determine milk production when silages of different grass and clover species are fed to dairy cows. J. Of Dairy Sci. 100, 8861–8880. doi: 10.3168/Jds.2017-13401

Crossref Full Text | Google Scholar

Jonker A., Scobie D., Dynes R., Edwards G., De Klein C., Hague H., et al. (2017). Feeding diets with fodder beet decreased methane emissions from dry and lactating dairy cows in grazing systems. Anim. Production Sci. 57, 1445–1450. doi: 10.1071/An16441

Crossref Full Text | Google Scholar

Jung H. G. (1989). Forage lignins and their effects on fiber digestibility. Agron. J. 81, 33–38. doi: 10.2134/Agronj1989.00021962008100010006x

Crossref Full Text | Google Scholar

Keady T., Mayne C., Fitzpatrick D. (2000). Prediction of silage feeding value from the analysis of the herbage at ensiling and effects of nitrogen fertilizer, date of harvest and additive treatment on grass silage composition. J. Of Agric. Sci. 134, 353–368. doi: 10.1017/S0021859699007674

Crossref Full Text | Google Scholar

Kelly D. N., Murphy C., Sleator R. D., Judge M. M., Conroy S. B., Berry D. P. (2019). Feed efficiency and carcass metrics in growing cattle1. J. Of Anim. Sci. 97, 4405–4417. doi: 10.1093/Jas/Skz316

Crossref Full Text | Google Scholar

Lahart B., Buckley F., Herron J., Fitzgerald R., Fitzpatrick E., Galvin N., et al. (2023). Evaluating enteric methane emissions within A herd of genetically divergent grazing dairy cows. J. Of Dairy Sci. 107, 383-397. doi: 10.3168/Jds.2022-22646

Crossref Full Text | Google Scholar

Lind V., Weisbjerg M. R., Jørgensen G. M., Fernandez-Yepes J. E., Arbesú L., Molina-Alcaide E. (2020). Ruminal fermentation, growth rate and methane production in sheep fed diets including white clover, soybean meal or porphyra sp. Animals 10, 79. doi: 10.3390/ani10010079

PubMed Abstract | Crossref Full Text | Google Scholar

Liu Z., Liu Y., Shi X., Wang J., Murphy J. P., Maghirang R. (2017). Enteric methane conversion factor for dairy and beef cattle: effects of feed digestibility and intake level. Trans. Of Asabe 60, 459–464. doi: 10.13031/issn.2151-0032

Crossref Full Text | Google Scholar

Lowman B., Scott N., Somerville S. (1976). Condition scoring of cattle. The East of Scotland College of Agriculture, Edinburgh, Scotland, Edinburgh School of Agriculture.

Google Scholar

Loza C., Verma S., Wolffram S., Susenbeth A., Blank R., Taube F., et al. (2021). Assessing the potential of diverse forage mixtures to reduce enteric methane emissions in vitro. Animals 11, 1126. doi: 10.3390/ani11041126

PubMed Abstract | Crossref Full Text | Google Scholar

Lyons T., Bielak A., Doyle E., Kuhla B. (2018). Variations in methane yield and microbial community profiles in the rumen of dairy cows as they pass through stages of first lactation. J. Of Dairy Sci. 101, 5102–5114. doi: 10.3168/Jds.2017-14200

Crossref Full Text | Google Scholar

Mann S., Yepes F. A. L., Overton T. R., Wakshlag J. J., Lock A. L., Ryan C. M., et al. (2015). Dry period plane of energy: effects on feed intake, energy balance, milk production, and composition in transition dairy cows. J. Of Dairy Sci. 98, 3366–3382. doi: 10.3168/Jds.2014-9024

Crossref Full Text | Google Scholar

McCarthy K., Lynch M., Pierce K., Fahey A., Gath V., Mcdonald M., et al. (2023). Rumen fermentation and forage degradability in dairy cows offered perennial ryegrass, perennial ryegrass and white clover, or A multispecies forage. Livestock Sci. 269, 105185. doi: 10.1016/j.livsci.2023.105185

Crossref Full Text | Google Scholar

Mceniry J., Forristal P. D., O’kiely P. (2011). Factors influencing the conservation characteristics of baled and precision-chop grass silages. Irish J. Of Agric. And Food Res. 50, 175–188.

Google Scholar

Meese S., Ulbrich S. E., Bollwein H., Bruckmaier R., Wellnitz O., Kreuzer M., et al. (2020). Methane emission, metabolism, and performance of holstein dairy cows with low, medium, and high lymphocyte proliferation during transition. J. Of Dairy Sci. 103, 4367–4377. doi: 10.3168/jds.2019-17584

Crossref Full Text | Google Scholar

MOE M. O. E., Environment, M. F. T (2023). New zealand’s greenhouse gas inventory: snapshot (Po Box 10362, Wellington 6143, New Zealand: New Zealand Government).

Google Scholar

Moraes L. E., Strathe A. B., Fadel J. G., Casper D. P., Kebreab E. (2014). Prediction of enteric methane emissions from cattle. Global Change Biol. 20, 2140–2148. doi: 10.1111/Gcb.12471

Crossref Full Text | Google Scholar

Morgan D. J., Stakelum G., Dwyer J. (1989). Modified neutral detergent cellulase digestibility procedure for use with the ‘Fibertec’ System. Irish J. Of Agric. Res. 28, 91–92. doi: 10.2307/25556234

Crossref Full Text | Google Scholar

Muñoz C., Villalobos R., Peralta A. M. T., Morales R., Urrutia N. L., Ungerfeld E. M. (2021). Long-term and carryover effects of supplementation with whole oilseeds on methane emission, milk production and milk fatty acid profile of grazing dairy cows. Anim. (Basel) 11, 2978. doi: 10.3390/Ani11102978

Crossref Full Text | Google Scholar

Murray Á., Delaby L., Gilliland T. J., Mcclearn B., Dineen M., Guy C., et al. (2022). A retrospective analysis of white clover (Trifolium repens L.) content fluctuation in perennial ryegrass (Lolium perenne L.) swards under 4 years of intensive rotational dairy grazing. Agriculture 12, 549. doi: 10.3390/agriculture12040549

Crossref Full Text | Google Scholar

O’Brien D., Moran B., Shalloo L. (2018). A national methodology to quantify the diet of grazing dairy cows. J. Of Dairy Sci. 101, 8595–8604. doi: 10.3168/Jds.2017-13604

Crossref Full Text | Google Scholar

O’donoVan M., Dillon P. (1999). Measurement of grassland management practice on commercial dairy farms (Teagasc, 19 Sandymount Avenue. Ballsbridge, Dublin 4: Teagasc).

Google Scholar

O’Mara F. (2006). Development of emission factors for the irish cattle herd (Po Box 3000, Johnstown Castle, Co. Wexford, Ireland: Environmental Protection Agency).

Google Scholar

O’Mara F. (1996). A net energy system for cattle and sheep (University College Dublin).

Google Scholar

Pang D., Yan T., Krizsan S. (2021). Effect of strategy for harvesting regrowth grass silage on performance in dairy cows. J. Of Dairy Sci. 104, 367–380. doi: 10.3168/Jds.2020-18888

Crossref Full Text | Google Scholar

Parnian-Khajehdizaj F., Noel S. J., Johansen M., Weisbjerg M. R., Hellwing A. L. F., Højberg O., et al. (2023). Methane emission, nutrient digestibility, and rumen microbiota in holstein heifers fed 14 different grass or clover silages as the sole feed. J. Of Dairy Sci. 106, 4072–4091. doi: 10.3168/Jds.2022-22638

Crossref Full Text | Google Scholar

Pinares-Patiño C. S., Waghorn G. C., Machmüller A., Vlaming B., Molano G., CaVanagh A., et al. (2007). Methane emissions and digestive physiology of non-lactating dairy cows fed pasture forage. Can. J. Of Anim. Sci. 87, 601–613. doi: 10.4141/Cjas06023

Crossref Full Text | Google Scholar

Richards B. F., Janovick N. A., Moyes K. M., Beever D. E., Drackley J. K. (2020). “Comparison of prepartum low-energy or high-energy diets with A 2-diet far-off and close-up strategy for multiparous and primiparous cows,” (Journal Of Dairy Science, 103, 9067-9080). doi: 10.3168/Jds.2020-18603

Crossref Full Text | Google Scholar

Soder K. J., Brito A. F. (2023). Enteric methane emissions in grazing dairy systems. GJournal of Dairy Science, Communications 4, 324-328.

Google Scholar

Sweeney R. A. (1989). Generic combustion method for determination of crude protein in feeds: collaborative study. J. Of Assoc. Of Off. Analytical Chem. 72, 770–774. doi: 10.1093/jaoac/72.5.770

Crossref Full Text | Google Scholar

Teagasc (2023). Marginal abatement cost curve 2023 (Teagasc: Agrculure And Food Development Authority).

Google Scholar

Teagasc L. ,. G., Donnellan T., Hanrahan K., Carsten P., Shalloo L., Krol D., et al. (2019). An analysis of abatement potential of greenhouse gas emissions in irish agriculture 2021-2030 (Teagasc).

Google Scholar

Teagasc (2016). Quality grass silage for dairy and beef production systems (Oak Park, Carlow R93 Xe12: Teagasc).

Google Scholar

Ulyatt M. J., Lassey K. R., Shelton I. D., Walker C. F. (2002). Seasonal variation in methane emission from dairy cows and breeding ewes grazing ryegrass/white clover pasture in New Zealand. New Z. J. Of Agric. Res. 45, 217–226. doi: 10.1080/00288233.2002.9513512

Crossref Full Text | Google Scholar

Van Dorland H. A., Wettstein H. R., Leuenberger H., Kreuzer M. (2007). Effect of supplementation of fresh and ensiled clovers to ryegrass on nitrogen loss and methane emission of dairy cows. Livestock Sci. 111, 57–69. doi: 10.1016/J.Livsci.2006.11.015

Crossref Full Text | Google Scholar

Van Gastelen S., Bannink A., Dijkstra J. (2019). “Effect of silage characteristics on enteric methane emission from ruminants. Cab reviews: perspectives,” in Agriculture, veterinary science, nutrition and natural resources, 14. doi: 10.1079/Pavsnnr201914051

Crossref Full Text | Google Scholar

Van Gastelen S., Burgers E. E. A., Dijkstra J., De Mol R., Muizelaar W., Walker N., et al. (2024). Long-term effects of 3-nitrooxypropanol on methane emission and milk production characteristics in holstein friesian dairy cows. J. Of Dairy Sci. 107, 5556–5573. doi: 10.3168/Jds.2023-24198

Crossref Full Text | Google Scholar

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

Crossref Full Text | Google Scholar

Waghorn G. C. (2018). Feeding fodder beet (Beta vulgaris L.) with either barley straw or pasture silage to non-lactating dairy cows. New Z. Veterinary J. 66, 178–185. doi: 10.1080/00480169.2018.1465484

Crossref Full Text | Google Scholar

Walsh S., Delaby L., Kennedy M., Mckay Z., O’donoVan M., Fleming C., et al. (2023). The effect of spring grass availability and silage supplementation on dairy cow performance and dry matter intake during early lactation. J. Of Agric. Sci. 161, 1–14. doi: 10.1017/S002185962300059x

Crossref Full Text | Google Scholar

Yan T., Agnew R. E., Gordon F. J., Porter M. G. (2000). Prediction of methane energy output in dairy and beef cattle offered grass silage-based diets. Livestock Production Sci. 64, 253–263. doi: 10.1016/S0301-6226(99)00145-1

Crossref Full Text | Google Scholar