Effects of replacing natural pasture hay with Napier grass on lactation performance and estimated methane emissions in dairy cows

Abstract

This experiment was conducted to evaluate the effects of replacing natural pasture hay (NPH) with Napier grass hay (NGH) on feed intake, milk yield and composition, feed efficiency, and estimated enteric methane (CH₄) emission intensity of lactating crossbred dairy cows supplemented with a concentrate mixture. Five cows were assigned to treatments in a 5 × 5 Latin square design. The treatments were 100% NPH, 75% NPH + 25% NGH, 50% NPH + 50% NGH, 25% NPH + 75% NGH, and 100% NGH, with concentrate supplementation at 0.5 kg dry matter (DM) per liter of milk. Increasing the level of NGH in the diet increased dietary crude protein (CP) content and improved dry matter (DM) and nutrient intakes (P < 0.001). Milk yield increased significantly (P < 0.001) from 11.4 to 16.1 L/day between cows fed 100% NPH and 100% NGH, respectively, accompanied by higher milk protein yield and improved feed efficiency. Estimated enteric CH₄ emission and intensity decreased (P < 0.001) from 26.9 to 20.6 g/kg milk yield with increasing NGH inclusion. These results demonstrate that replacing NPH with NGH enhances lactation performance, while reducing estimated enteric CH₄ emission intensity, providing a practical climate-smart feeding strategy for smallholder dairy systems in Ethiopia.

Introduction

Ethiopia possesses the largest livestock population in Africa¹, and the dairy sector has been identified as a key government priority to increase milk production and quality². In the Ethiopian farming system, feeding dairy cattle often depends on crop residues and natural pasture¹, with little or no concentrate supplementation. These types of feed resources are characterized by high fiber content, low digestibility, and low CP content³, and marked seasonal availability⁴, resulting in low feed intake, digestibility, and reduced milk yield, particularly during the dry season, when feed scarcity is most severe⁵.

Beyond productivity constraints, low-quality diets also have important environmental implications. Poorly digestible feeds reduce rumen efficiency and increase enteric CH₄ emissions per unit of product, thereby lowering overall system efficiency⁶. Many smallholder farmers in Ethiopia feed dairy cows without considering nutrient requirements, which limits milk yield, increases CH₄ emission intensity, and results in greater nitrogen losses to the environment⁷. Addressing these issues requires appropriate feeding strategies that improve nutrient supply while simultaneously mitigating enteric CH₄ emissions and environmental pollution⁸.

One promising strategy during periods of critical feed shortage is the use of improved forage species that provide better biomass yield and superior nutritive value compared to natural pasture hay (NPH) and crop residues⁵. Supplementing a low-quality roughage-based diet with improved forage grass species has been shown to enhance nutrient intake, digestibility, milk yield, and composition, and feed efficiency in lactating dairy cows in Ethiopia^5,9. Consequently, integrating cultivated forage with traditional feed resources represents a sustainable feeding strategy for smallholder dairy systems⁸. This situation has encouraged the cultivation and utilization of well-adapted, improved forage such as Napier (Pennisetum purpureum) grass.

Napier grass is widely cultivated in tropical and subtropical regions due to its high biomass yield, better nutritional value compared to natural pasture, and relatively simple agronomic management¹⁰. Its capacity to produce large quantities of biomass under tropical conditions makes it a promising alternative to natural pasture, particularly during the dry season¹¹. However, despite these advantages, most previous studies have focused primarily on production responses, with limited attention to the combined evaluation of productivity, feed efficiency, and methane emission intensity under controlled experimental conditions. In addition, quantitative evidence on the progressive replacement of NPH with NGH across multiple inclusion levels remains scarce, particularly under Ethiopian smallholder dairy systems where concentrate supplementation is constrained.

Therefore, this study aimed to evaluate the effects of progressively replacing natural pasture hay with Napier grass hay on feed intake, milk yield, milk composition, feed efficiency, and estimated enteric CH₄ emission intensity in lactating crossbred dairy cows.

Materials and methods

Experimental animal, design, and treatment

The experiment was conducted at the Injibara University dairy farm, Injibara, Ethiopia. Five mid-lactation crossbred dairy cows (13.5 ± 0.61 L/day; mean ± SD) were selected at approximately 60 days in milk. The study used a 5*5 Latin square design in which each cow received all five dietary treatments across five experimental periods. Each period lasted 21 days, consisting of 14 days for dietary adaptation and 7 days for data collection. Treatments were arranged in a balanced sequence to minimize carry-over effects. The five roughage-based dietary treatments were 100% NPH (T1), 75% NPH + 25% NGH (T2), 50% NPH + 50% NGH (T3), 25% NPH + 75% NGH (T4), and 100% NGH (T5). All cows received a concentrate supplement at a rate of 0.5 kg DM per liter of milk produced. Cows were milked twice daily at 08:00 and 16:00 h.

Animal management, feed, and feeding

Before the experiment, cows were treated against internal and external parasites using tetraclozan and ivermectin, respectively. Animals were housed in individual pens with free access to clean drinking water. NPH was harvested from the Injibara University grazing area. Napier grass (Pennisetum purpureum, acc. 15743) was established following recommended agronomic practices using a combined application of vermicompost and urea. The forage was harvested at appropriate maturity, field-dried with frequent turning, chopped into 10–15 cm lengths, and offered ad libitum. The concentrate mixture was obtained from Mulle Animal Feed Factory (Bahir Dar, Ethiopia). Feed offered, and refusals were recorded daily, and DMI was calculated as the difference between feed offered and refused on a DM basis. Nutrient intake was calculated on a feed-component basis. First, DMI of each feed ingredient was determined. Then, nutrient intake was calculated by multiplying the DMI of each feed ingredient by its respective chemical composition. Finally, total nutrient intake was obtained by summing the contributions of all feed components.

Feed sample analysis

Feed samples were dried at 60 °C for 48 h in a forced-air oven and ground to pass through a 1 mm sieve using a Wiley mill. The grounded samples were kept in sealed plastic bags at room temperature pending analysis. DM and Ash were analysed according to the procedures of AOAC¹². NDF, ADF, and ADL were determined according to Vansoest and Robertson (1985)¹³. The nitrogen content was measured by the Kjeldahl method¹², and CP was calculated as N*6.25.

Milk yield and composition

Milk yield was recorded at each milking by weighing the milk and summing morning and afternoon yields to obtain daily milk production. Average daily milk yield was calculated for each cow during the data collection period. Milk samples were collected during morning and afternoon milkings in each period. Composite milk samples were prepared, stored in sterile bottles, kept in an icebox, and analyzed at the Injibara University laboratory. Milk fat, protein, and lactose contents were determined using a Lactoscan milk analyzer.

Fat- and protein-corrected milk production (FPCM) (kg/d) was calculated as - FPCM = (0.337 + 0.116 × milk fat [%] + 0.06 × milk protein [%]) × kg of milk yield¹⁴. Fat-corrected milk yield (kg/d) = 0.4*milk yield + 15*(milk fat/100)*milk yield¹⁵. Feed efficiency (also referred to as dairy efficiency) was calculated as the ratio of 3.5% FCM yield to DMI.

Estimation of enteric methane emission

The enteric CH₄ emissions (g/day) of lactating dairy cows were estimated according to a recommended intercontinental equation according to Niu et al.¹⁶.

CH₄ production (g/day per cow) = 124 (± 10.44) + 13.3 (± 0.32) × DM intake (DMI, kg/day).

This estimation equation was chosen because of its low root mean square prediction error (RMSPE = 17.5) compared with those of other equations in the same publication. Correspondingly, Appuhamy et al.¹⁷ also evaluated the performance of more than 40 empirical models for predicting enteric CH₄ emissions and suggested that using DM intake alone can be sufficient for satisfactory prediction of enteric CH₄ emissions.

Statistical analysis

The data obtained from feed intake, nutrient intake, milk yield, and composition were subjected to the analysis of variance (ANOVA) using R Studio 4.2.3. Software. Differences among treatment means were tested by using the least significant difference (LSD). The statistical model was:- Yijk = µ + Ti + Pj + C_K + εijk.

Where:-.

Yijk = observation on cow k given treatment i at period j.

µ = overall mean.

Ti = fixed effect of the ith treatment (i = 1 to 5).

Pj = random effect of jth period (j = 1 to 5).

Ck = random effect of the kth cow.

εijk = random residual error.

Ethical approval

All experimental procedures involving animals were conducted in accordance with the guidelines of Injibara University for the care and use of agricultural animals in research. The study protocol was reviewed and approved by the Department of Animal Science Research and Ethics Committee of Injibara University. Animal handling and management were carried out to ensure animal welfare and to minimize stress throughout the experimental period.

Results

Chemical composition of experimental feeds

Results of proximate and detergent analysis is presented in Table 1. The NPH had a low CP concentration (5.37%), whereas the NGH hay contained a higher CP concentration (11.3%). The CP content of the roughage increased progressively with increasing proportions of NGH in the diet. Conversely, fiber fractions decreased as the proportion of NGH increased. The highest neutral detergent fiber (NDF), acid detergent fiber (ADF), and acid detergent lignin (ADL) concentrations were observed in 100% NPH, while the lowest values were recorded in 100% NGH. The concentrate mixture used as a supplement contained 17% CP, with NDF, ADF, and ADL concentrations of 36.6%, 31.3%, and 6.1%, respectively.

Table 1.

Chemical composition of experimental feeds.

Treatments	Chemical composition of feed (%)
	DM	OM	CP	NDF	ADF	ADL
T1	90.1	86.1	5.37	69.1	41.7	13.4
T2	89.7	86.4	6.84	66.6	39.1	11.1
T3	89.2	86.7	8.32	64.1	36.5	10.6
T4	88.8	86.9	9.78	61.6	33.9	9.3
T5	88.3	87.2	11.3	59.1	34.3	7.9
Concentrate	90	89.9	17	36.6	31.3	6.1

T1 = 100% NPH, T2 = 75% NPH + 25% NGH, T3 = 50% NPH + 50% NGH, T4 = 25% NPH + 75% NGH (T4), T5 = 100% NGH, DM = Dry matter, OM = Organic matter, CP = crude protein, NDF = Neutral detergent fiber, ADF = Acid detergent fiber, ADL = Acid detergent lignin, NPH = Natural pasture hay, NGH = Napier grass hay.

Dry matter and nutrient intake

Daily DM and nutrient intakes of lactating crossbred dairy cows are presented in Table 2. Dry matter intake (DMI) increased with increasing proportions of NGH in the diet (P < 0.001), ranging from 13.8 kg/day in cows fed 100% NPH (T1) to 15.5 kg/day in cows fed 100% NGH (T5). A similar increasing trend was observed for DMI expressed as a percentage of body weight and per unit of metabolic body weight (P < 0.001). Crude protein intake increased progressively across dietary treatments (P < 0.001). In contrast, neutral detergent fibre intake was higher in diets containing greater proportions of NPH (P < 0.001).

Table 2.

DM and nutrient intake of lactating crossbred dairy cows fed NPH and NGH supplemented with concentrate mix.

Parameters	Treatments
	T1	T2	T3	T4	T5	SE	P-value
DMI (kg/day)	13.8d	14.2c	14.6b	14.9b	15.5a	0.09	< 0.001
DMI (%BW)	3.22	3.33	3.41	3.49	3.62	0.9	< 0.001
DMI, g/kgBW0.75	146.7d	151.3c	154.9bc	158.7b	164.8a	0.02	< 0.001
CPI (kg/day)	1.47e	1.61d	1.75c	1.91b	2.1a	0.00	< 0.001
NDFI (kg/day)	7.51b	7.61ab	7.63ab	7.64ab	7.76a	0.06	0.008

T1 = 100% NPH, T2 = 75% NPH + 25% NGH, T3 = 50% NPH + 50% NGH, T4 = 25% NPH + 75% NGH (T4), T5 = 100% NGH, NPH = Natural Pasture hay, NGH = Napier Grass hay, TDMI = Total Dry matter intake, DMI = Dry matter intake, CPI = Crude protein intake, NDFI = Neutral detergent fiber intake, ^abc Means within a row with no common superscripts differ (p < 0.05) significantly.

Milk yield and composition

Milk yield was significantly affected by dietary treatment (P < 0.001) and increased progressively with NGH inclusion. Cows fed 100% NGH produced 16.1 L/day compared with 11.4 L/day for cows fed 100% NPH, corresponding to a 41.2% increase (Table 3). Milk component yields also improved as the inclusion level of NGH increased. Protein yield increased from 0.42 to 0.46 kg/day (+ 14.3%; P < 0.001), while lactose yield increased from 0.46 to 0.68 kg/day (+ 47.8%; P < 0.001). Although milk yield was highest in T5, the lower milk protein concentration resulted in a slightly reduced protein yield compared to T4. Fat yield showed a modest but significant increase (P < 0.001). Energy-corrected milk, fat-corrected milk, and fat- and protein-corrected milk were higher in T4 and T5 than in T1 and T2 (P < 0.001). Feed efficiency (3.5% FCM/DMI) improved from 0.88 to 0.96. Milk fat and protein concentrations declined (P < 0.001) with increasing NGH inclusion, while lactose concentration increased, reaching the highest value in T4.

Table 3.

Milk yield and composition of lactating crossbred dairy cows fed NPH and NGH supplemented with concentrate mix.

Parameters	Dietary treatments
	T1	T2	T3	T4	T5	SE	P-value
Milk yield L/day	11.4b	12.1d	14.2c	15.1b	16.1a	0.13	< 0.001
Fat yield	0.50b	0.51b	0.53a	0.54a	0.56a	0.008	< 0.001
Protein yield (kg/day)	0.42c	0.41c	0.45b	0.46a	0.44b	0.008	< 0.001
Lactose yield	0.46d	0.49c	0.60b	0.66a	0.68a	0.007	< 0.001
ECM	12.5c	12.3c	13.7b	14.0ab	14.6a	0.16	< 0.001
FCM (kg/day)	12.4c	12.4c	13.9b	14.27b	14.86a	0.15	< 0.001
Feed efficiency	0.9b	0.88c	0.95a	0.95a	0.96a	0.009	< 0.001
Milk compositions
Fat	4.55a	4.18b	3.87c	3.65d	3.51e	0.05	< 0.001
Protein	3.68a	3.38b	3.19c	3.05c	2.74d	0.06	< 0.001
Lactose	3.98c	4.07c	4.22b	4.41a	4.21b	0.04	< 0.001
FPR	1.24a	1.24a	1.22a	1.20ab	1.17b	0.02	< 0.001

T1 = 100% NPH, T2 = 75% NPH + 25% NGH, T3 = 50% NPH + 50% NGH, T4 = 25% NPH + 75% NGH (T4), T5 = 100% NGH, ECM = Energy corrected milk, FCM = Fat corrected milk, FPCM = Fat protein corrected milk, FPR = Fat to protein ratio, SE = Standard error of mean, ^abc Means within a row with no common superscripts differ (p < 0.05) significantly.

Estimated enteric methane emission

Estimated daily CH₄ production increased with increasing NGH inclusion (P < 0.001), reflecting higher DMI (Table 4). However, CH₄ emission intensity decreased significantly. CH₄ per unit of DMI declined from 22.3 to 21.3 g/kg (–4.5%), and CH₄ per unit of milk yield decreased from 26.9 to 20.6 g/kg milk (–23.4%; P < 0.001). CH₄ emissions expressed per unit of FPCM differed among treatments, with higher values at intermediate NGH inclusion levels (P < 0.001). CH₄ scaled to metabolic body weight increased with NGH inclusion (P < 0.001).

Table 4.

Effect of replacing NPH with NGH on CH₄ emission of lactating crossbred dairy cows.

Parameters	Dietary treatments
	T1	T2	T3	T4	T5	SE	P-value
CH4 (g/day)	307.5e	313.4d	317.9c	322.4b	330.2a	1.14	< 0.001
CH4 (g/kg DMI)	22.3a	22.1b	21.8c	21.6d	21.3e	0.05	< 0.001
CH4 (g/kg milk)	26.9a	26.1b	22.5c	21.4d	20.6e	0.26	< 0.001
CH4 g/kg/FPCM	21.4c	22.8b	24.1a	24.3a	23.9a	0.21	< 0.001
CH4/BW0.75	3.27e	3.33d	3.38c	3.4b	3.51a	0.01	< 0.001

T1 = 100% NPH, T2 = 75% NPH + 25% NGH, T3 = 50% NPH + 50% NGH, T4 = 25% NPH + 75% NGH (T4), T5 = 100% NGH, CH4, Methane, BW^0.75 = Metabolic body weight, ^abc Means within a row with no common superscripts differ (p < 0.05) significantly.

Discussion

The CP content of NPH alone recorded in the current study was below the minimum threshold of 7–7.5% required to satisfy maintenance requirements of ruminants¹⁸, indicating that NPH alone cannot support milk production without additional protein supplementation. The CP content of NPH recorded in the current study was higher than the 3.5% reported by Asmare et al.¹⁹, but lower than the 6.7% reported by Adnew et al.²⁰, which shows NPH generally provides insufficient protein for optimal dairy production. The observed variation in CP content across research studies may be attributed to differences in agroecology, soil fertility, botanical composition, stage of maturity at harvest, and post-harvest handling.

In contrast, the CP concentration of NGH was sufficient to support rumen microbial activity and maintenance requirements, which was reflected in the increased milk yield as the proportion of NGH in the diet increased. Diets containing CP concentrations above 7% ensure sufficient nitrogen supply for optimal rumen activities, provided other factors such as lignification do not limit feed digestibility and nutrient utilization²⁰. Although the CP content of NGH did not reach the optimal range recommended to support moderate to high milk production (12–16%)¹⁵, it provided a nutritionally superior basal diet compared with NPH under smallholder dairy production systems. The higher NDF concentration observed in NPH classifies it as a low-quality feed, as roughages containing NDF levels greater than 65% are generally considered low in nutritive value²¹.

Increased DMI with increasing inclusion level of NGH can be explained by its higher CP content and lower fiber fractions compared with NPH, which favors rumen microbial growth and enhances digestion. In agreement with this finding, increasing the inclusion level of improved forage (Urochloa mutica) increased DMI in lactating crossbred dairy cows compared with NPH⁹. Similarly, Adnew et al.²⁰ demonstrated that the inclusion of Urochloa hybrid Mulato II in the diet of sheep increases DMI through their effects on nutrient digestibility. Moreover, Mutimura et al.²² and Mekuriaw et al.⁵ reported higher DMI of the diets based on U. brizantha cv. Piata and Urochloa hybrid cv. Mulato II in dairy cows, respectively, which was attributed to the higher CP content of the grasses. Intake expressed per metabolic body weight (BW⁰·⁷⁵) confirmed that differences in feed intake were not solely due to body size variation but reflected true dietary effects. In the present study, the lower NDF concentration of NGH relative to NPH likely enhanced ruminal degradation and passage rate, thereby improving feed intake, particularly in cows fed 100% NGH.

The increased milk yield with the increasing level of NGH can be attributed to greater intake of higher-quality feed (NGH) and increased nutrient supply for milk synthesis. The results of the present study are consistent with findings reported by Tadesse et al.⁹ and Tezera et al.²³, who observed higher milk yields in cows fed better-quality forages, whereas cows consuming low-quality roughages (NPH) exhibited lower milk production. Heavy reliance on low-quality feeds such as NPH results in lower milk yield due to inadequate CP content and high fiber concentration, which limit nutrient digestibility and metabolizable energy availability.

Feed conversion efficiency enhanced as the proportions of NGH increased, reflecting the combined effect of higher nutrient density, increased DMI, and increased milk yield. In contrast, the poor FCE observed in cows fed the sole NPH diet was likely due to insufficient CP intake and high fiber levels, which resulted in a greater proportion of nutrients being used for maintenance rather than milk production. These findings are consistent with those of Tadesse et al.⁹, who reported improved FCE with increasing inclusion of U. mutica in the diets of dairy cows. Moreover, Asmare et al.¹⁹ confirmed that increasing the inclusion level of Desho (Pennisetum pedicellatum) grass in the diet compared to NPH increases the FCE of sheep.

Methane emission intensity was also influenced by diet composition. The higher estimated CH₄ production observed in cows fed sole NGH was associated with higher DMI and higher fermentable substrate supply, which enhanced rumen fermentation. This result was consistent with previous reports^5,24. Higher feed intake accelerates digesta flow and fermentation extent, thereby increasing total CH₄ output. However, CH₄ emission expressed per unit of DMI and milk yield decreased significantly as the proportion of NGH increased, indicating improved efficiency of production per unit of estimated CH₄ emitted. This suggests that replacing natural pasture hay with Napier grass hay may improve emission efficiency in dairy production systems. This demonstrates that improved forage quality can reduce CH₄ emission intensity while enhancing productivity, supporting the development of climate-smart dairy production systems⁸.

Study limitations

This study has some limitations that should be considered when interpreting the results. First, enteric CH₄ emissions were not directly measured but estimated from dry matter intake, which may not fully capture the effects of dietary phytochemicals on rumen methanogenesis. Second, the study did not include an economic evaluation; therefore, the cost-effectiveness and practical feasibility of replacing natural pasture hay with Napier grass hay under smallholder conditions remain uncertain. Despite these limitations, the study provides important insights into the productive and environmental benefits of improved forage utilization.

Conclusion

Increasing the proportion of NGH as a basal diet for lactating crossbred dairy cows improved DMI, FCE, and milk yield compared with NPH. Higher inclusion levels of NGH also reduced estimated CH₄ emission intensity per unit of DMI, milk yield, and fat- and protein-corrected milk, indicating improved emission efficiency. Although total CH₄ production increased with higher intake, the reduction in CH₄ intensity demonstrates the advantage of feeding higher-quality forages. Overall, replacing NPH with NGH enhanced nutrient utilization, dairy productivity, and feed efficiency while lowering CH₄ emission intensity. This feeding strategy, therefore, represents a practical feeding strategy with potential environmental benefits in Ethiopian smallholder dairy systems, where access to concentrate feeds and precision feeding technologies is limited.

Author contributions

All authors contribute to designing and performing the experiment, analyzing and interpreting the data, writing the original draft, reviewing, editing, and validating the paper.

Funding

For completion, this work was supported by Injibara University.

Data availability

The datasets generated and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.