Forage yield and quality parameters of eight oat (Avena sativa L.) genotypes at multilocation trials in Eastern Oromia, Ethiopia

Abstract

Eight genotypes of oat (Avena sativa L.) were evaluated across four locations with two standard checks under rainfall conditions from 2017 to 2018. Both agronomic and chemical composition data were collected. The analysis of the results indicated a statistical significance difference (P < 0.01) in the days of 50% flowering stage, plant height, biomass yield, leaf-to-stem ratio, dry matter yield, date of maturity, and seed yield. For the collecting parameters, the overall mean was measured from days of 50% flowering stage taking (54.83–86.58) days, plant height (86.04 to 108.56 cm), biomass yield (36.19 to 47.47 t/ha), leaf-to-stem ratio (0.25 to 1.32), dry matter yield (6.96 to 9.79 t/ha) the maturity date (119.0 to 130.98 days) and seed yield (33.38 to 43.49 qu/ha). There were no statistically significant differences (P > 0.05) in ash, crude protein (CP), neutral detergent fiber (NDF), acid detergent fiber (ADF), acid detergent lignin (ADL), and in-vitro dry matter digestibility (IVDMD) among genotypes. The additive main effects and multiplicative interaction (AMMI) analysis for dry matter yield and biomass yield were influenced by environment and genotype. Among the study locations, Tulo was more of an ideal environment for oat crops followed by Chiro. Among the eight oat genotypes evaluated in this trial, the genotypes ILRI#5450 and ILRI#5442 revealed higher agronomic performance than the standard check and remaining studied genotypes. Therefore, oat genotypes ILRI#5450 and ILRI#5442 should be recommended for wider cultivation and need further breeding improvement investigation as animal feed in Eastern Oromia and similar agro-ecology to Ethiopia.

Keywords:

• agronomic parameters
• genotypes
• Hararghe
• oat and quality

PUBLIC INTEREST STATEMENT

Ethiopia has the largest livestock population in Africa. However, the production and productivity of the sector is low. The major factor is the inadequate quality and quantity of animal feed. The majority of livestock feed resources in Ethiopia are obtained from natural pasture, crop residues, and crop thinning, but such feed resources are available only in particular seasons and contain limited nutrients (low crude protein, minerals, and vitamins), resulting in lower intake and digestibility. Hence, forage development is one of the strategies to address feed scarcity and low livestock productivity in Ethiopia. The feed sub-sector is a pillar for all livestock growth and transformation from various perspectives.

1. Introduction

Ethiopian livestock is an integral component of agriculture. The subsector makes a significant contribution to the national income (Alemayehu et al., 2012) and livelihoods of rural and urban communities. However, the productivity of animals remained at a low level due to an unbalanced supply of quantity and quality feeds (Wada et al., 2019). The majority of livestock feed resources in Ethiopia are obtained from natural pasture, crop residues, and crop thinning (Alemayehu et al., 2012), but such feed resources are available only in particular seasons and contain limited nutrients (low crude protein, minerals, and vitamins), resulting in lower intake and digestibility (Talore, 2015). Therefore, this case is reported as a major problem in livestock production in developing countries like Ethiopia, particularly during the long dry season, when there is insufficient plant biomass carried over from the wet season to support domestic livestock species (Kebede et al., 2021).

The landholding per farmer in the Hararghe area is less than 1 ha (Alemu, 2016; Etafa et al., 2013). Therefore, producing enough crop residues as animal feed is challenging (Gemechu et al., 2016), and in contrast, ruminant fattening is unique (Abebe, 2019; Tefera et al., 2019). Using industrial by-product feed highly increases the cost and is not affordable for farmers. The system is largely based on cut-and-carry green feeds from maize and sorghum origin including thinning, leaf strip part of maize, and sorghum plants (Alemayehu et al., 2016; Anteneh et al., 2010; Ayalew et al., 2013; Goshu et al., 2017). In the livestock industry, efficiency and productivity are limited due to quality fodder production because different animals require feeds according to their growth stage and production stage (Rehman et al., 2017).

Oat (Avena sativa L.) is excellent fodder for animals and is cultivated in different regions of the country due to its diverse adaptability; it can grow in a wide range of soil types, rainfall situations, and altitudes. However, moderate and cool climate conditions are ideal for its development (Beyene et al., 2015). Compared to other cereals such as barley and wheat, oats are resistant to both drought and moisture stress. It is characterized by the growth habit of erect and bunch to basal. It can be a good source of animal feed in the dry season if harvested at the right stage of growth, cured, and stored as hay (Tulu et al., 2020). It is also a quick-growing, palatable, succulent, and nutritious fodder crop (Wada et al., 2019). According to Tulu et al. (2020) report, oat genotypes in Western Oromia can produce hay yields ranging from 5.07 to 13.57 t/ha, crude protein from 36.2 to 63.5 g/kg DM, neutral detergent fiber 685–728.7 g/kg DM, and acid detergent fiber 470–610.4 g/kg DM. However, although Eastern Oromia was well known for indigenous bull fattening and milk production practices, this good-quality fodder-like oat is not evaluated in the area. Therefore, to understand the quality of fodder, the current study was undertaken to identify the best adaptable, good nutritive value and digestibility characteristics of oat genotypes grown under different locations of Hararghe Western zone, Eastern Oromia.

2. Materials and methods

Eight oats (Avena sativa L.) genotypes were evaluated, including two standard checks (Bonsa and SRCPX80Ab2806). The experiment was conducted under rain-fed conditions at four locations for two consecutive years (2017–2018), main cropping season (June–December) at Daro Lebu district (Mechara Agricultural Research Center on station), Habro district (Bareda farmer training center), Chiro district (Arbarakate Farmer Training Center), and Tulo district (Gara Kufa Farmer Training Center).

Daro Lebu (08°27–08°69N; 040°31–040°65E) district is located about 434 km Southeast of Addis Ababa and 115 km from Chiro town, the capital of West Hararghe Zone. The district is characterized mostly by flat and undulating land features, with altitudes ranging from 1350 to 2450 m above sea level. The minimum and maximum temperature of the district ranges from 14°C to 26°C and the annual mean temperature is 20°C, while the annual rainfall ranges from 700 to 1494 mm/year with an average of 963 mm/year. The soil type of Daro Lebu experimental site was Nitisols with 2.35% organic carbon, 9.8 ppm available P, 0.03% of total N, and pH (H₂O) 7.5 (Geremu et al., 2021).

Habro (08°56–08°89N; 040°34–040°89E) district is located 391 km from Addis Ababa and 65 km from the zonal town Chiro in the West Hararghe zone, the Eastern part of Ethiopia. The district is characterized by plateaus, mountains, hills, plains, and valleys and is generally classified into three agro-ecologies, the lowland, the midland, and the high land constituting 5%, 80%, and 15% of the total area of the district, respectively. The mean annual rainfall in the district was 966.7 mm/year. The mean annual temperature was 19.97°C, with the hottest months being May and June and the coldest months being November and December. The major soil types found in the Habro district are vertic luvisols, rendzic leptosols, haplic luvisols, eutric vertisols, and eutric leptosols. Of these soil types, Habro experimental site vertic luvisols with 2.06% organic carbon, 10.67 ppm available P, 0.05% of total N, and pH (H₂O) 6.65 (Gizaw et al., 2021).

Chiro (08°97–09°21N; 040°64–041°09E) district is located 326 km from Addis Ababa. Chat, coffee, sorghum, and maize are the major crops grown in the district. The district is founded at an average altitude of 1800 above sea level. It has a maximum and minimum temperature of 23°C and 12°C, respectively, and a maximum and minimum rainfall of 1800 mm and 900 mm, respectively. The district is mainly dominated by sandy soil, clay soil (black soil), and loamy soil types covering 25.5%, 32%, and 42.5%, respectively, with 2.03% organic carbon, 13.8 ppm available P, 0.18% of total N, and pH (H₂O) 7.5 (Dinkale et al., 2021).

Tulo (09°25–09°303N; 041°008–041°26E) is one of the districts in the West Hararghe Zone and is located 370 km distance from Addis Ababa and 47 km from Chiro (Zone). Sorghum, maize, chat, and coffee are the major crops grown in the area. The average altitude of the district is 1750 m above sea level, the annual rainfall range from 427 to 2302 mm/year, and the mean annual temperature of 23°C (Dinkale et al., 2021). The soil type of Tulo experimental site is sandy soil with 2.35% organic carbon, 9.8 ppm available P, 0.03% of total N, and pH (H₂O) 7.5 (See Figure 1).

Figure 1. Geographical location map of selected districts of West Hararghe Zone.

2.1. Planting materials

Planting materials used for this study were initially obtained from the International Livestock Research Institute (ILRI). Then, the eight genotypes used for the current study, namely, ILRI#5442, ILRI#5445, ILRI#5447, ILRI#5450, ILRI#5457, ILRI#5467, ILRI#5478, and ILRI#7251 were selected based on their adaptability to the climatic conditions of Eastern Oromia, Ethiopia, from the previous preliminary variety trial work carried (Birmaduma et al., 2016 unpublished).

2.2. Experimental land preparation and planting

Appropriate experimental sites were selected in four locations. Based on the nature of the soil, the experimental land was plowed twice using a bull. During the sowing at the end of June 2017 (first year) and 2018 (second year), the plot was prepared using hand. Planting/sowing was conducted using seeds in both experimental years. The planting and weed were done manually. During the experimental period, the plant gets only rain fed.

2.3. Experimental design and layout

A randomized complete block design with three replications was used at all locations. The plot size was 3.6 m²; with six rows of 2 m in length at 20 cm interspacing. Recommended fertilizer rate of 100 kg/ha NPS at planting and a seed rate of 70 kg/ha was used. All agronomic practices were done uniformly as recommended for oats (Tulu et al., 2020).

2.4. Data collection

2.4.1. Days 50% flowering stage (days)

The mean value for the days of the 50% flowering stage was recorded, while 50% of plants from each plot were blooming. The length of the days of the 50% flowering stage was calculated as the day of the 50% flowering stage minus the emergency date.

2.4.2. Plant height (cm)

The average plant height was measured from the ground to the tip of the main stem. The measurement was done by taking 10 random plants on the days of the 50% flowering stage from each plot.

2.4.3. Biomass yield (t/ha)

The vegetation from each plot was sampled using a quadrant of 0.25 m² (0.5 m × 0.5 m) size during a predetermined sampling period (10% flowering stage). All biomass yield was the weight of the spot from the area and converted the value to hectare.

2.4.4. Leaf-to-stem ratio

The mean value of leaf to stem was recorded, while 10% of plants from each plot were bloomed. The sample was harvested to measure the fresh biomass, and then the stem and leaf were separated and weighted individually. After weighing, the weight value of the leaf was divided by the weight value of the stem then equal to the leaf-to-stem ratio.

2.4.5. Dry matter yield (t/ha)

The vegetation from each plot was a sample for determining biomass yield production; then about 250 g was taken from the paper as sampled to determine dry matter yield at Mechara Agricultural Research Center and dried in an oven at 105°C overnight. After the samples were removed from the oven-dry; reweighed and dry matter yield was calculated as weight before dried minus weight after dry divided by weight before dried multiplied by 100.

2.4.6. Date of maturity (days)

The date of maturity was recorded when about 90% of the plant reach the physiological maturity stage.

2.4.7. Seed yield (qu/ha)

After the oat crop was matured, the whole plot was harvested and threshed individually in each plot. The total seed yield weight from the plot was converted to hectares and presented in quintals.

2.5. Chemical analysis and in-vitro dry matter digestibility

The oven-dried samples at a temperature of 65°C for 72 hours were used for laboratory analysis to determine the chemical composition and in-vitro organic matter digestibility of oat genotypes. The dried samples were then ground to pass a one millimeter (mm) sieve, and the ground samples were used for laboratory analysis. The samples were analyzed on a DM (%) basis for ash, crude protein (CP), neutral detergent fiber (NDF), acid detergent fiber (ADF), acid detergent lignin (ADL), and in-vitro dry matter digestibility (IVDMD). Total ash content was determined by oven drying the samples at 105°C overnight and by combusting the samples in a muffle furnace at 550°C for 6 hours (AOAC, 1990). Nitrogen (N) content was determined following the micro-Kjeldahl digestion, distillation, and titration procedures (AOAC, 1990), and the CP content was estimated by multiplying the N content by 6.25. The structural plant constituents (NDF, ADF, and ADL) were determined according to the Van Soest et al. (1991) procedure. The in vitro dry matter digestibility (IVDMD) was determined according to Tilley and Terry (1963).

2.6. Statistical analysis

Data analysis before computing the combined analysis, the error variance homogeneity test was verified using Hartley‘s test (F-max test) (Gomez & Wiley, 1984). In the combined analysis of variance, the location was considered a random variable, and genotypes were considered a fixed variable. Data analysis was performed using the statistical software version 9.0. Genotype by Environment analysis with gen stat Version 18th. Additive main effect and multiplicative interaction AMMI (Zobel et al., 1988) models were used to compute stability. In the AMMI model, the magnitude obtained in the first principal component (IPCA1) of each genotype was used as an indicator stability. The lower absolute value of IPCA1 indicated a stable genotype. The genotypes were tested using the analysis of variance (ANOVA) procedure of the SAS general linear model (GLM) to compare treatment means (SAS, 2002). The least significant difference (LSD) at a 5% significance level was used for the comparison of means. The data were analyzed with the model:

Yijk = µ + Gi + Ej + (GE) ij + Bk (j) + eijk;

where, Yijk = measured response of genotype i in block k of environment j; µ = grand mean; Gi = effect of genotype i; Ej = effect of environment j; GEij = genotype and environment interaction; Bk (j) = effect of block k in environment j; eijk = random error effect of genotype i in block k of environment j.

3. Results and discussions

3.1. Agronomic performance of eight oat genotypes

The combined analysis of variance for 50% flowering stage, biomass yield, leaf-to-stem ratio, plant height, maturity date, seed yield, and dry matter yield of oat genotypes tested over four locations and for 2 years is presented in Table 1. The result of the analysis variance showed that the genotype, environment, and years were highly significant differences (P < 0.01) in the 50% flowering stage, plant height, biomass yield, leaf-to-stem ratio, dry matter yield, maturity date, and seed yield. A similar result was reported by Kebede et al. (2021); Lodhi et al. (2009) who investigated the yield and nutritional quality of oat (Avena sativa L.) genotypes yield and under vertisols conditions in the central highlands of Ethiopia. The genotype interaction with the environment (G*E) was shown to have a significant difference (P < 0.01) only on the maturity date. The genotype interaction with years (G*Y) showed a statistically significant difference (P < 0.05) in biomass yield, dry matter yield, and maturity date. This was reflected by no change in the ranking order of genotypes over the years due to relatively uniform growing conditions during the experimental years (Kebede et al., 2017). The three ways of interaction genotype, environment, and years showed a non-significant (P > 0.05) difference for only seed yield. This indicated that the genotypes exhibit a highly specific response to a particular environment (soil, rainfall, and temperature), while others are uniform in performance over a range of environments (Kebede et al., 2017).

Table 1. Combined analysis of variance for the measured morphological trait of oat genotypes

Parameters	G	E	Y	G*E	G*Y	G*E*Y	Mean	CV (%)	R^2
50% (days)	***	***	***	NS	NS	***	70.23	8.64	0.92
PH (cm)	***	***	***	NS	NS	***	101.72	7.01	0.93
Bm (t/ha)	*	***	***	NS	**	***	41.02	31.19	0.73
LSR	**	*	***	NS	NS	***	0.67	30.94	0.65
DMY (t/ha)	***	***	***	NS	*	***	7.62	27.82	0.71
DM (days)	***	***	***	***	***	***	126.04	2.28	0.99
Sy (qu/ha)	***	***	**	NS	NS	NS	38.1	25.51	0.78

G = Genotype, E = Environment, Y = Year, CV (%) = Percentage of coefficient of variation, 50% (days) = 50% flowering stage, LSR = Leaf-to-stem ratio, Bm (t/ha) = Biomass yield in ton per hectare, PH (cm) = Plant height in centimeter, MD = Date of maturity, Sy (qu/ha) = Seed yield in quintal per hectare, DMY (t/ha) = Dry Matter Yield in ton per hectare.

3.1.1. Days of 50% flowering stage

The combined analysis for days of 50% flowering stage of the eight oat genotypes tested at four locations over the two experimental years is shown in Table 2. The combined analysis of oat genotypes at four locations for years revealed highly significant variation (P < 0.01) on days of 50% flowering stage. All genotypes flowered early than the standard check from 10 to 32 days. The mean value of genotype ILRI#5445 (54.83 days) was reached early at the days of 50% flowering stage, followed by ILRI#5457 (60.04 days) and ILRI#7251 (61.17 days), with a mean value of 70.23 days. Both standard checks Bonsa (86.58 days) and SRCPX80Ab2806 (85.33 days) reached 50% flowering dates. Of the total genotypes tested, only four genotypes earlier reached the days of 50% flowering stage than the overall mean value. This result indicated that the genotypes are distinctly different for days of 50% flowering stage attributes. Concerning environmental grouping, the mean value of flowering dates ranged from an earlier mean value of 23 days from ILRI#5445 genotype in 2018 at the Tulo location to a late flowering of 106 days for genotype ILRI#5447 in 2017 at the Habro location. The lowest overall mean value of 50% flowering stage was recorded at 54.83 days from ILRI#5445 genotype and the late dates of 50% flowering were recorded from Bonsa (86.58 days) variety (check). The current result is taken an early day to attain 50% flowering than the report by (Tessema & Getinet, 2020) who reported 147 days to 92 days. The variation among oat genotypes in days taken to flower may be due to their genetic makeup (Nazakat et al., 2004).

Table 2. Location by year analysis of days of 50% flowering stage of oat genotypes during 2017 and 2018, at Daro Lebu, Habro, Chiro, and Tulo

Genotypes	Daro Lebu		Habro		Chiro		Tulo		Mean Average
	2017	2018	2017	2018	2017	2018	2017	2018
ILRI#5442	59.00	53.33	85.33	52.67	80.00	65.00	75.00	48.33	64.83
ILRI#5445	55.67	50.33	75.33	49.67	66.00	50.67	68.00	23.00	54.83
ILRI#5447	74.00	60.00	106.00	58.67	90.00	65.00	89.00	65.00	75.96
ILRI#5450	70.00	58.67	90.67	56.00	90.00	65.00	77.00	65.00	71.54
ILRI#5457	56.67	50.33	76.33	52.00	71.00	51.33	77.00	45.67	60.04
ILRI#5467	61.33	58.67	96.67	54.33	90.00	65.00	84.00	65.00	71.88
ILRI#5478	65.33	56.00	89.67	56.00	87.00	65.00	77.00	65.00	70.13
ILRI#7251	58.00	51.67	81.00	52.00	76.67	51.33	73.00	45.67	61.17
Bonsa	87.67	74.00	110.00	72.00	107.00	75.00	89.00	78.00	86.58
SRCPX80Ab2806	89.33	68.67	109.67	72.00	102.00	75.00	89.00	77.00	85.33
Mean	67.7	58.17	92.01	57.53	85.97	62.83	79.8	57.77	70.23
CV (%)	5.27	9.82	4.87	3.61	8.14	0.93	8.24	20.56	10.01
LSD (5%)	6.12	9.80	7.67	3.57	12.00	1.00	11.28	20.37	4.009
PV	<0.001	0.0009	<0.0001	<0.0001	<0.0001	<0.0001	0.007	0.0006	<0.0001

CV (%) = percentage of coefficient of variance, LSD = least significant difference.

3.1.2. Plant height (cm)

A combined analysis of plant height of the eight oat genotypes tested at four locations over the two experimental years is shown in Table 3. Plant height is a major factor contributing to the forage yield of different crops. The result of the analysis of variance indicated a highly significant difference (P < 0.01) in plant height among genotypes. Of the total genotypes tested in this trial, all genotypes revealed a higher mean value of plant height than both the standard checks. These indicated that the tested genotypes have a good potential for fodder production and alleviate feed scarcity in the area, which is a good indication for the further breeding experiment. Concerning environmental grouping, the measured mean value of plant height ranged lowest 58 cm from ILRI #5442 at Daro Lebu in 2018 to 145.89 cm from SRCPX80Ab2806 at Tulo in 2017 with an overall mean value of 101.72 cm. These great different mean values might be due to agro-ecology of oat being well performed in the highland than in the midland. The current finding agrees with Mesgana et al. (2020) who reported 89.2 cm to 153.1 cm plant height for different oat genotypes in the Amhara region, Ethiopia. The highest mean value of the present study is in agreement with Beyene et al. (2015) who reported 145 cm from 8235-CI genotype and 137 cm from Jasari. Wada et al. (2019) reported 123 cm from the Lampton variety Abate and Wegi (2011) reported 128.4 cm from the Bonsa variety and 156.2 cm from the Bona-bas variety. The minimum mean value recorded in the current result was in agreement with Bibi et al. (2021) who reported ranging from 79 to 119 cm and 63 cm overall mean for Ukrainian varieties in Pakistan.

Table 3. Location by year analysis of plant height (cm) of oat genotypes during 2017 and 2018, at Daro Lebu, Habro, Chiro, and Tulo

Genotype	Daro Lebu		Habro		Chiro		Tulo		Mean Average
	2017	2018	2017	2018	2017	2018	2017	2018
ILRI#5442	72.33	58.00	94.33	73.53	88.33	90.93	119.56	91.27	86.04
ILRI#5445	98.73	71.33	108.78	87.73	129.33	112.07	139.89	116.33	108.03
ILRI#5447	102.67	69.53	111.44	87.13	113.78	108.27	133.00	113.00	104.85
ILRI#5450	105.20	70.00	109.11	92.40	128.22	109.13	137.78	116.67	108.56
ILRI#5457	100.80	73.47	107.11	80.20	112.67	104.00	137.44	111.00	103.34b
ILRI#5467	101.40	68.80	92.78	81.87	111.00	100.53	139.22	110.33	100.74
ILRI#5478	105.07	72.07	103.22	90.13	127.56	105.20	138.22	111.93	106.68
ILRI#7251	90.13	65.60	106.44	80.33	120.33	103.47	129.56	105.27	100.14
Bonsa	90.07	59.00	101.56	81.53	113.56	98.73	136.00	103.07	97.94
SRCPX80Ab2806	97.87	59.60	100.00	86.33	115.44	97.87	145.89	103.73	100.84
Mean	96.43	66.74	403.48	84.12	116.02	103.02	135.66	108.26	101.72
CV (%)	7.93	7.18	6.70	5.11	6.03	5.54	7.60	4.59	6.96
LSD (5%)	13.12	8.22	11.89	7.38	12.00	9.8	17.69	8.53	4.0371
P-Value	0.0016	0.0035	0.047	0.0016	<0.0001	0.0103	0.404	0.0002	<0.0001

CV (%) = Percentage of coefficient of variance, LSD (5%) = Least significant difference at 5%.

3.1.3. Biomass yield (t/ha)

The analysis of the results for the biomass yield of the eight oat genotypes tested at four locations over the two experimental years is shown in Table 4. The analysis of the variance (ANOVA) of the tested genotypes showed significant variation (P < 0.01) in biomass yield (t/ha). The mean value of biomass yield ranged from the lowest in Bonsa (36.19 t/ha) to the highest in ILRI#5450 (47.47 t/ha) genotype, with an overall mean value of 41.01 t/ha. Regarding environmental grouping, the highest mean value of biomass yield was recorded for ILRI#5442 (98.2 t/ha) followed by ILRI#5450 (86.13 t/ha) at Tulo in 2017, in contrast, the lowest mean value was produced by ILRI#5445 (19.42 t/ha) followed by ILRI#5478 (19.89 t/ha) at Daro Lebu in 2018. The higher biomass yield recorded at the Tulo location is due to the district‘s high rainfall distribution (1850 mm/yr) in the West Hararghe zone, which created a conducive environment for oat crops. In nature, the oat crop is well adapted to high to midland, mainly under rain-fed conditions (Wada et al., 2019). As the area is downward to the lowland, the productivity decreases (Mesgana et al., 2020), which was confirmed in the current trial at Tulo to Daro Lebu. Annual rainfall might be the major factor in the recorded different mean values of biomass yield between Tulo (1850 mm/year) and Daro Lebu (963 mm/year), which is more than double the difference. However, the current trial recorded a higher mean value of biomass yield than Wada et al. (2019) who reported higher biomass for CV-SRCP X 80 Ab 2291 (42.4 t/ha) following by CV-SRCP X 80 Ab (36.2t/ha). Beyene et al. (2015) reported that oats can produce biomass yield in the range of 60–70 t/ha.

Table 4. Location by year analysis of biomass (t/ha) of oat genotypes during 2017 and 2018, at Daro Lebu, Habro, Chiro, and Tulo

Genotypes	Daro Lebu		Habro		Chiro		Tulo		Mean Average
	2017	2018	2017	2018	2017	2018	2017	2018
ILRI#5442	26.60	23.98	63.60	23.20	41.53	52.19	98.20	31.75	45.13
ILRI#5445	29.93	19.42	54.47	23.27	40.07	26.07	70.67	28.23	36.52
ILRI#5447	31.80	20.76	55.12	33.61	42.60	38.39	66.47	34.78	40.44
ILRI#5450	36.80	20.38	70.33	28.37	45.73	44.97	86.13	47.07	47.47
ILRI#5457	33.20	24.98	51.87	22.28	40.60	35.47	63.93	29.55	37.73
ILRI#5467	33.93	20.06	63.80	29.11	40.67	48.83	61.80	38.88	42.13
ILRI#5478	31.40	19.89	61.07	28.12	45.40	48.59	79.60	33.74	43.48
ILRI#7251	33.93	21.79	74.27	25.80	41.60	33.60	86.00	29.89	43.36
Bonsa	25.33	21.69	42.06	36.28	32.33	43.60	42.80	45.40	36.19
SRCPX80Ab2806	29.20	21.16	53.93	32.19	42.40	40.80	39.73	41.74	37.65
Mean	31.21	21.41	59.05	28.22	41.29	41.25	69.53	36.11	41.01
CV (%)	25.68	22.80	27.48	31.16	25.71	20.57	35.76	24.65	32.40
LSD (5%)	13.75	8.37	27.8	15.01	18.2	14.56	42.66	15.27	7.5745
P-Value	0.7963	0.9142	0.4581	0.5682	0.9389	0.0353	0.1587	0.149	<0.0001

CV (%) = Percentage of coefficient of variance, LSD (5%) = Least significant difference at 5%.

3.1.4. Leaf-to-stem ratio

The analysis of the results for the leaf-to-stem ratio of the eight oat genotypes tested at four locations over the two experimental years is shown in Table 5. The result of analysis of variance (ANOVA) indicated significant (P < 0.05) variation in leaf-to-steam ratio. The average mean value of leaf-to-stem ratio ranged from the less leafy variety Bonsa (0.49) to the more leafy genotype ILRI#5467 (0.8) with a mean value of 0.67. Out of the total genotypes, about seven genotypes were recorded with a greater percentage of leaves than the overall mean value, and all genotypes were recorded with more leaf-to-steam ratio mean values than both standard checks. Relating to the environmental grouping, the highest leaf-to-stem ratio was produced from ILRI#5467 (1.32) by ILRI#5467 in 2018 at Daro Lebu location, followed by SRCPX80Ab2806 (1.21) in 2018 at the Habro location. The lowest leaf-to-stem ratio was recorded equally from Bonsa and SRCPX80Ab2806 (0.25) at Chiro in 2017 and Habro in 2017, respectively. The average mean value of leaf-to-stem ratio in the current finding was comparable with (Abate & Wegi, 2011) who reported in the range of 0.64 to 0.78 with an overall mean value of 0.72. Similarly, Sharma et al. (2019) reported oat leaves to stem 0.73 to 0.88, whereas Befekadu and Yunus (2015) reported a lower leaf-to-stem ratio (0.43) in Arsi highland, Ethiopia. The significant variations reported by different authors on the leaf-to-stem ratio at different locations are due to environmental differences, rainfall conditions, soil nutrient status, and genetic variation.

Table 5. Location by year analysis of leaf-to-stem ratio of oat genotypes during 2017 and 2018, at Daro Lebu, Habro, Chiro, and Tulo

Genotypes	Daro Lebu		Habro		Chiro		Tulo		Mean Average
	2017	2018	2017	2018	2017	2018	2017	2018
ILRI#5442	0.41	1.07	0.63	0.76	0.28	0.99	0.84	0.73	0.71
ILRI#5445	0.40	0.89	0.67	0.70	0.61	0.88	0.61	0.61	0.67
ILRI#5447	0.62	0.79	0.51	1.09	0.56	0.84	0.53	0.69	0.71
ILRI#5450	0.61	0.77	0.37	0.72	0.50	0.90	0.39	0.94	0.65
ILRI#5457	0.53	0.84	1.01	0.60	0.53	0.88	0.54	0.58	0.69
ILRI#5467	0.61	1.32	0.67	0.73	0.54	0.95	0.67	0.87	0.80
ILRI#5478	0.53	0.95	0.58	0.64	0.47	0.85	0.68	0.78	0.69
ILRI#7251	0.47	0.95	0.67	0.78	0.48	0.72	0.62	1.01	0.71
Bonsa	0.37	1.05	0.29	0.88	0.25	0.34	0.42	0.37	0.49
SRCPX80Ab2806	0.45	0.93	0.25	1.21	0.51	0.37	0.61	0.43	0.60
Mean	0.5	0.96	0.57	0.81	0.47	0.77	0.59	0.70	0.67
CV (%)	35.93	34.81	30.59	27.71	32.59	23.69	26.68	29.89	36.60
LSD (5%)	0.31	0.57	0.30	0.39	0.26	0.31	0.27	0.36	0.14
P-Value	0.5957	0.7057	0.0017	0.0611	0.1604	0.0022	0.097	0.0219	0.002

CV (%) = Percentage of coefficient of variance, LSD (5%) = Least significant difference at 5%.

3.1.5. Dry matter yield (t/ha)

The analysis result of the dry matter yield of the eight oat genotypes tested at four locations over the two experimental years is shown in Table 6. The result of the analysis of variance (ANOVA) indicated a statistically significant difference (P < 0.01) in dry matter yield (DMY) among genotypes. Out of the total average mean value, the highest DMY was recorded from genotype ILRI#5450 (9 t/ha) followed by ILRI#5442 (8.7t/ha), whereas the lowest DMY was obtained from genotype ILRI#5478 (6.2t/ha) followed by ILRI#5451 (6.59t/ha), with overall mean value 7.61t/ha. Out of the eight genotypes considered in the trial, only four (ILRI#5442, ILRI#5447, ILRI#5450, and ILRI#5457) genotypes produced more DMY than the overall mean value. Concerning the environmental grouping for the trail, the highest DMY was recorded from ILRI#5457 (15.53 t/ha) at the Chiro location in 2017, followed by ILRI#5442 (13.3t/ha) at the Chiro location in 2017. The lowest mean value of DMY was recorded from ILRI#5457 (2.57t/ha) at Daro Lebu in 2017 followed by ILRI#5451 (2.96t/ha) at Daro Lebu in 2017. The variation might be attributed to the variability in the amount and distribution of rainfall, locations, and genetic difference.

Table 6. Location by year analysis of dry matter (t/ha) of oat genotypes during 2017 and 2018, at Daro Lebu, Habro, Chiro, and Tulo

Genotype	Daro Lebu		Habro		Chiro		Tulo		Mean Average
	2017	2018	2017	2018	2017	2018	2017	2018
ILRI#5442	4.82	7.47	8.91	5.52	13.3	11.64	10.41	7.47	8.73
ILRI#5445	3.91	5.29	8.59	4.43	12.76	5.08	9.34	5.29	6.96
ILRI#5447	4.86	6.15	9.82	5.73	15.53	6.97	11.19	6.15	8.34
ILRI#5450	5.29	10.53	7.00	6.06	13.62	9.06	9.94	10.53	9.00
ILRI#5457	2.57	6.93	9.41	5.38	15.47	6.67	8.78	6.93	7.77
ILRI#5467	4.57	6.81	6.69	4.78	11.23	8.23	8.16	6.81	7.16
ILRI#5478	4.53	5.68	5.98	4.71	7.93	7.94	7.25	5.65	6.2
ILRI#7251	2.96	6.17	7.82	5.5	10.42	6.44	7.43	6.07	6.59
Bonsa	5.31	6.52	11.66	5.29	10.91	7.72	9.56	6.94	8.04
SRCPX80Ab2806	5.27	6.94	11.63	5.03	7.64	7.94	8.52	6.62	7.38
Mean	4.74	6.83	8.75	5.34	11.88	7.83	9.09	6.83	7.61
CV (%)	29.18	30.53	26.5	32.74	30.48	22.29	16.32	30.53	27.80
LSD (5%)	2.19	3.56	3.98	3	6.21	3	2.54	5.58	1.21
P-Value	0.1031	0.2363	0.087	0.995	0.1478	0.0252	0.059	0.2367	<0.0001

CV (%) = Percentage of coefficient of variance, LSD (5%) = Least significant difference at 5%.

3.1.6. Maturity dates (days)

The analysis of the results on the maturity date of the eight oat genotypes tested at four locations over the two experimental years is shown in Table 7. Statistical analysis of the data showed that the days to maturity were significantly affected (P < 0.01) by genotypes. The average days to maturity ranged from 119 days from early genotype ILRI#5442 to late maturity and were measured at 121.76 days from genotype ILRI#5457, with an overall mean value of 122.49 days. All genotypes tested in this experiment matured early than both the standard checks. Concerning the environmental grouping, the early maturity days were recorded from genotypes ILRI#5450, 5467, and 5478 (82 days) at the Habro location in 2018 equally; whereas the late maturity dates were recorded from four genotypes ILRI#5442, ILRI#5447, ILRI#5457, ILRI#5467 and both standard check Bonsa and SRCPX80Ab2806 (154) equally in 2018 at Chiro location. The present found mean value for days to maturity was shorter than Dinkale (2019) who reported 182.2 days and Abate and Wegi (2011) who reported oat maturity date with the range of 149–166 days with a mean value of 157 Nazakat et al. (2004) who reported 210 days to attain maturity dates. The variation in days of maturity might be due to genetic makeup, environment, and interaction of genotype with environmental effects.

Table 7. Location by year analysis for maturity dates of oat genotypes during 2017 and 2018, at Daro Lebu, Habro, Chiro, and Tulo

Genotype	Daro Lebu		Habro		Chiro		Tulo		Mean Average
	2017	2018	2017	2018	2017	2018	2017	2018
ILRI#5442	117	111	138	83.33	154	113	136.33	103.33	119
ILRI#5445	117	121	138	86	140	126	131.67	97	119.58
ILRI#5447	119.3	111	141	83.33	154	112.67	138.11	113	121.56
ILRI#5450	117	114.33	138	82	154	112.67	136.33	113	120.92
ILRI#5457	117	121	138	84.67	149.33	126	134.77	103.33	121.76
ILRI#5467	117	111	139	82	154	113	136.66	113	120.71
ILRI#5478	117	111	141	82	154	113	137.33	113	121.04
ILRI#7251	117	121	138	84.67	140	126	131.67	103.33	120.21
Bonsa	117	128.33	147	104	154	134	139.33	105	128.58
SRCPX80Ab2806	128.7	132	147	104	154.00	134	143.22	105	130.98
Mean	118.4	118.17	140.5	87.6	150.73	121.03	136.54	106.9	122.49
CV (%)	2.3	2.42	4.17	1.69	1.7	0.20	1.1	4.96	3.65
LSD (5%)	14.68	4.91	1.73	2.54	4.38	0.42	1.57	9.09	2.6228
P-Value	0.0011	<0.0001	0.0003	<0.0001	<0.0001	<0.0001	<0.0001	0.0118	<0.0001

CV (%) = Percentage of coefficient of variance, LSD (5%) = Least significant difference at 5%.

3.1.7. Seed yield (qu/ha)

The analysis of the results on seed yield of the eight oat genotypes tested at four locations over the two experimental years is shown in Table 8. The result of the analysis of variance revealed a statistical significance difference (P < 0.05) in seed yield among genotypes tested in the experiment. The average mean value seed yield ranged from the highest in ILRI#5447 (33.46 qu/ha) to the lowest in ILRI#7251 (43.79 qu/ha), with an overall average mean value of 38.1 qu/ha. Similar to other parameters, only four genotypes produced seed yield more than the overall mean value; however, about six genotypes produced more seed yield than both standard checks. Concerning environmental grouping, the highest seed yield was obtained from ILRI#5445 (69.66qu/ha) at the Chiro location in 2018, followed by ILRI#7251 (69.49 qu/ha) at the Chiro location in 2018, and the lowest seed yield was measured from Bonsa (12.43qu/ha) followed by SRCPX80Ab2806 (12.93 qu/ha) at Daro Lebu in 2018. The variation in environmental conditions is reflected by the large differences in the average seed yield observed across the environment, which might be attributed to the variability in the amount and distribution of rainfall genotypes, which are expected to vary greatly across locations and may influence the yield of oat genotypes. The current finding on seed yield was better than Mesgana et al. (2020) who reported a combined grain yield performance range of 39.04 to 30.45 qu/ha in the Amhara region and (Dawit & Mulusew, 2014) who reported an oat seed yield ranges from 21.7 to 29.8 qu/ha in Bale zone, Ethiopia. As a result, the performance of grain yield of oat genotypes was significantly affected by the main genetic, environmental, and interaction of genotype with environmental effects.

Table 8. Location by year analysis of seed yield (qt/ha) of oat genotypes during 2017 and 2018, at Daro Lebu, Habro, Chiro, and Tulo

Genotype	Daro Lebu		Habro		Chiro		Tulo		Mean Average
	2017	2018	2017	2018	2017	2018	2017	2018
ILRI#5442	19.71	25.40	37.96	36.63	52.04	60.12	48.20	43.27	40.42
ILRI#5445	16.34	22.10	47.16	36.12	55.21	69.66	40.59	49.61	42.10
ILRI#5447	19.81	19.09.	21.08	30.31	42.46	66.13	26.42	42.42	33.46
ILRI#5450	21.23	13.85	27.01	30.16	44.32	60.66	28.44	41.37	33.38
ILRI#5457	25.18	26.17	33.59	34.86	57.08	66.28	52.62	49.57	43.17
ILRI#5467	25.45	20.72	28.57	35.41	48.11	64.90	37.77	46.94	38.48
ILRI#5478	22.66	16.50	30.84	39.67	52.17	52.98	35.02	47.24	37.13
ILRI#7251	21.67	25.63	39.71	37.59	46.46	69.49	60.45	49.31	43.79
Bonsa	17.56	12.43	27.74	32.70	56.71	46.46	36.66	44.04	34.29
SRCPX80Ab2806	19.48	12.93	26.40	17.42	55.03	48.78	51.90	46.09	34.76
Mean	20.91	19.48	32.01	33.09	50.96	60.55	41.8	45.99	38.1
CV (%)	28.51	16.01	10.07	16.69	13.9	14.3	47.23	15.83	24.28
LSD (5%)	10.22	5.35	5.53	9.63	12.15	14.85	33.86	12.48	5.27
P-Value	0.6655	<0.0001	<0.0001	0.008	0.167	0.028	0.507	0.8318	0.0095

CV (%) = Percentage of coefficient of variance, LSD (5%) = Least significant difference at 5%.

3.2. Environment and interaction effect on the performance of oat genotypes

The biomass yield is one of the most important traits, and the ultimate goal of forage production for smallholders is to obtain high biomass with reasonable quality (Malosetti et al., 2013). The ANOVA table results for the environment showed significant (P < 0.01) variation in the biomass yield; however, their genotypes and interaction (GxE) did not show a statistically significant difference (P > 0.05) in the green forage yield Table 9. The higher percentage of contribution for variation was due to the environment (22.25%) than genotypes (3.41%). These results indicate that the genotypes tested in the trial have low genetic distances. The significance among environments demonstrated that genotypes responded differently to different environments confirming the need to assess the performance of oat genotypes across environments to identify genotypes with stable and superior yield across environments.

Table 9. ANOVA table for biomass yield in AMMI model at 2 years (2017 and 2018)

Source	DF	SS	MS	% Contribution to the variation	% G x E	F pro
Total	239	96720	404.7
Treatments	39	28070	719.8			<.001
Genotypes	9	3299	366.5	3.41		0.4032
Environments	3	21747	7249.1	22.48		<.001
Block	8	1523	190.4			0.8218
Interactions	27	3024	112	3.13		0.9995
IPCA 1	11	2018	183.5		66.73	0.8851
IPCA 2	9	634	70.5		20.97	0.9938
Residuals	7	372	53.1		12.3	0.9935
Error	192	67126	349.6

DF = Degree of freedom, SS = Sum square, MS = Mean square, G = Genotype, E = Environment.

3.3. Environment and interaction effect on the dry matter yield performance of oat genotypes

The ANOVA table results for the main effect (environment and genotypes) show significant (P < 0.01) variation in the dry matter yield (DMY); however, their interaction (GxE) did not show a statistically significant difference (P > 0.05) in DMY Table 10. The higher percentage of contribution for variation was due to environment (23.2%) than genotypes (7.45%). These results indicate that the genotypes tested in the trial have low genetic distances. The significance between genotypes and environment demonstrated that genotypes responded differently to different environments, confirming the need to assess the performance of oat genotypes across environments to identify genotypes with stable and superior yields across environments. Similarly, Tulu et al. (2020) reported that seven oat genotypes responded differently to biomass yield and quality traits across the test environments under different locations in Western Oromia.

Table 10. ANOVA table for dry matter yield in AMMI model at 2 years

Source	DF	SS	MS	% Contribution to the variation	% G x E	F pro
Total	239	2459.9	10.29
Treatments	39	900.9	23.1			<.001
Genotypes	9	183.3	20.36	7.45		0.0085
Environments	3	570.7	190.22	23.20		<.001
Block	8	29.2	3.65			0.8842
Interactions	27	146.9	5.44	5.97		0.8796
IPCA 1	11	93.5	8.5		63.65	0.3899
IPCA 2	9	50.7	5.64		34.51	0.7014
Residuals	7	2.7	0.38		1.85	0.9999
Error	192	1529.8	7.97

DF = Degree of freedom, SS = Sum square, MS = Mean square, G = Genotype, E = Environment.

3.4. Chemical composition of eight oat genotypes

The combined analysis variance for dry matter percentage, ash, crude protein (CP), neutral detergent fiber (NDF), acid detergent fiber (ADF), acid detergent lignin (ADL), in vitro dry matter digestibility (IVDMD), and organic matter (OM) is presented in Table 11. The result of the analysis variance indicated that there were no statistically significant differences (P > 0.05) in all chemical composition parameters among genotypes. The least-square mean value of DM % was ranging 92.63% from SRCPX80Ab2806 (check2) to ILRI 5457 (92.9%) with a mean value of 92.78%. This result is higher than that of Tulu et al. (2020) who reported 50.73% recorded DM% from ILRI 8237 to the higher 60.13% obtained from ILRI 6710. However, comparable to Wada et al. (2019) who reported 91.5% from CV-SRCP X 80Ab 2806 to 93.3% from CI-8237 of DM%. Abate and Wegi (2011) reported 90.6% from Bona-bas to 85.2% from CI-8235 at the high land of Bale, Ethiopia. This variation might be related to the difference in rainfall, genetic variability, soil fertility, forage harvesting stage, and other climatic conditions.

Table 11. Least square mean of eight oat genotypes with two check chemical compositions in Western Hararghe

Genotypes	DM%	Ash%	CP%	NDF%	ADF%	ADL%	IVDMD%	OM%
ILRI#5450	92.71	10.58	1.08	74.15	65.38	7.82	57.75	82.13
ILRI#5442	92.84	10.57	9.80	74.83	65.79	7.08	57.32	82.27
ILRI#5445	92.80	10.44	9.85	74.57	63.85	7.30	57.43	82.37
ILRI#5447	92.76	10.60	1.00	74.57	66.36	7.44	57.62	82.16
ILRI#5457	92.90	11.40	1.03	75.26	65.62	7.36	57.68	81.50
ILRI#5467	92.85	11.22	1.21	75.80	68.67	7.38	57.88	81.63
ILRI#5478	92.80	10.71	1.00	74.70	66.11	7.37	57.62	82.09
ILRI#7251	92.82	10.89	1.02	75.06	64.09	7.89	57.72	81.93
Bonsa	92.71	9.81	9.64	73.44	62.83	7.82	57.14	82.90
SRCPX80Ab2806	92.63	9.47	9.66	73.35	63.04	7.97	57.16	83.16
Mean	92.78	10.57	9.93	74.57	65.17	7.54	57.53	82.21
CV (%)	0.12	14.56	6.54	1.73	4.3	12.06	1.66	1.83
LSD(.05)	0.17	2.23	.94	1.88	4.07	1.32	1.39	2.18
P-V	0.0951	0.8081	.9599	0.2387	0.1705	0.8994	0.9730	0.8856

ADF (%) = Acid detergent fiber in percentage, ADL (%) = Acid detergent lignin in percentage, CV (%) = percentage of coefficient of variance, LSD = Least significant difference, DM (%) = Dry matter in percentage, CP (%) = Crude protein percentage, IVDMD (%) = In-vitro dry matter digestibility in percentage, NDF (%) = Neutral detergent fiber in percentage, OM (%) = Organic matter in percentage.

The ash content of oat genotypes ranged from 9.47 to 11.4 with a mean value of 9.93%. The amount of ash in forage is an indication of mineral concentrations in the feed. The mineral content is affected by the stage of maturity and the leaf-to-stem ratio of the forage plant. The current result is comparable with Kebede et al. (2021) who reported ash content ranging from 9.4 to 12.5 with a mean of 10.8%. However, Wada et al. (2019) reported lower ash content, ranging from 8.4 to 9.99 with a mean of 8.9%. The plant development stage, morphological fractions, climatic conditions, soil characteristics, and fertilization regime are some of the potential factors causing variation in mineral concentration in forage plants (Kebede et al., 2021; Tulu et al., 2020).

The analysis of the result does not show a statistical significance difference (P > 0.05) in CP among genotypes. Similarly, Kebede et al. (2021) reported a non-significant difference (P > 0.05) in CP and IVDMD among 15 oat genotypes under vertisols conditions in the Central Highlands of Ethiopia. The average mean value of the current CP content was ranging from 9.6 to 10.21 with a mean of 9.93%. This result was slightly comparable with Sterna et al. (2016), Kudake et al. (2017), and Mosissa et al. (2018) who reported 10–16.6 g/100 and higher than Kebede et al. (2021) who reported 6.9–8.1 with a mean of 7.7%. The minor difference could be related to the genetic differences among oat genotypes studied and agronomic practices applied during the production and harvesting stage.

Neutral detergent fiber (NDF) content varied between 74.35 from SRCPX80Ab2806 (check2) to ILRI#5467 (75.8%) with a mean of 74.57%. A comparable result was reported by Kebede et al. (2021) who reported a mean value ranging from 70.1 to 74.8 with a mean value of 72.8%. Asmare et al. (2017) reported 73.58% for high land and 76.03% for midland in addition the authors also reported 72.28% to 77.68% at different harvesting stages for Desho grass in the Northern highland of Ethiopia. The higher NDF value is mostly affected by harvesting time (Molla et al., 2018). However, the mean NDF content of oat genotypes in the study is higher than that of (Negash et al., 2017) who reported a mean NDF content of 56.95%, and Wada et al. (2019) who reported NDF mean ranging between 41.6% and 51.4% for different oat varieties. Variations in genetic materials, harvesting stage, climatic conditions, and soil factors are the major causes of the difference in NDF content in oat genotypes.

Acid detergent fiber (ADF) content varied between 62.8 from Bonsa to ILRI#5467 (68.67%) with a mean of 65.17%. This result is higher than the result of Wada et al. (2019) who reported ADF mean values ranging from 24.2% to 28%. Abate and Wegi (2011) reported 32.7% to 38.3% for ADF mean value in the highland of the Bale zone, Ethiopia. Kebede et al. (2021) reported ADF ranging from 45.4 to 52 with a mean value of 48.5%. The variation might be recorded due to the harvesting stage, climatic conditions, and soil factors are the major causes of the difference in ADF content. Acid detergent lignin (ADL) content ranges from 7.08 to 7.89 with a mean value of 7.54%. The current result was higher than the report of Megersa (2021) who reported 4.56% to 6.52% for different oat genotypes in the highland of Ethiopia. In contrast, the current result was lower than the report of Kebede et al. (2021) who reported ADL content for different oat genotypes ranges 9.4 to 11.7 with a mean value of 10.5%. Generally, the presence of insoluble fiber, particularly lignin, lowers the overall digestibility of the feed by limiting nutrient availability (Van Soest, 1994). According to Tonamo et al. (2016); Negash et al. (2017) the NDF of more than 65% was categorized as low quality and 45–65% as medium quality and less than 45% categorized as high quality. Similarly, ADF greater than 40% is categorized as low quality, and less than 40% is categorized as high quality (Urbaityte et al., 2009). However, different authors in Ethiopia reported out of this interval needs further investigation on oat genotypes at different locations and growth stages. Tulu et al. (2020) reported for seven oat genotypes in western Oromia; NDF ranges between 68.5% and 72.87% and ADF ranges between 47.08% and 63.63%. Abate and Wegi (2011) in the highland of Bale, Ethiopia, reported NDF ranging from 51.7% to 62.5%. Wada et al. (2019) reported in South Ethiopia; NDF ranges from 41.6% to 51.4%, and Kebede et al. (2021) reported in the highland of central Oromia, Ethiopia; NDF ranges between 70.1% and 74% and ADF ranges between 45.4% and 52%.

The in vitro dry matter detestability values (IVDMD) ranged from Bonsa (51.14) to ILRI#5467 (57.88) with a mean of 57.53%. The current result is comparable with Kebede et al. (2021) who reported IVDMD for 15 oat genotypes ranging between 52.4 and 54.3 with a mean of 53.6%. However, the current result was lower than Wada et al. (2019) who reported IVDMD ranging between 68.55% and 73.78%, and Abate and Wegi (2011) reported 57.75–66.67% ranges. According to Negash et al. (2017) report the in vitro dry matter detestability values (IVDMD) greater than 65% indicated good nutritive value and below this level, results reduce intake due to lower digestibility. The result of the current trial revealed that the mean value of organic matter (OM) ranges between 81.16 from ILRI5467 to 82.9 from Bonsa. The current results were lower than the report of Abate and Wegi (2011) who reported OM ranges between 87.2% and 89.9% from different oats varieties. However, comparable to Dinkale et al. (2021) who reported OM percent mean value ranges 83.82–86.22 with a mean of 84.9% for different Elephant grass cultivars in western Hararghe, Ethiopia.

4. Conclusions and recommendations

The present study evaluated oat genotypes for agronomic performance and chemical composition at four locations for 2 years of the mean cropping season. The result was given in this direction as the oat crop can perform highly in Eastern Oromia, where traditional dairy and fattening practices are the models for a tropical country. All collected agronomic parameters were shown at significant differences (P < 0.01) at four locations over 2 years. AMMI analysis of DMY and biomass yield was influenced by environment and genotype. Among the study locations, Tulo was a more ideal environment for oat crops followed by Chiro. Among all the eight oat genotypes evaluated in the current study, genotypes ILRI#5450 and ILRI#5442 showed higher DMY and agronomic performance at all locations over the rest of the genotypes. The result of the analysis variance does not indicate any significant difference (P > 0.05) in all chemical composition parameters among genotypes. In general, out of eight study oat genotypes, ILRI#5450 and ILRI#5442 gave the higher plant height (cm), biomass (t/ha), dry matter yield (t/ha), seed yield, and chemical composition performed at all study locations.

Thus, both genotypes are recommended for general cultivation due to their better dry matter yield, agronomic performance, and chemical composition over both checks (Bonsa and SRCPX80Ab2806) and promising over the remaining genotypes studied. In general, the information generated in this study suggested that genotypes ILRI#5450 and ILRI#5442 have the potential to resolve the recurrent quality of feed scarcity and would serve as a guide to those who would like to adopt an improved fodder oat genotype for ruminant feeding.

Acknowledgments

The authors would like to thank the vital support and interest of the animal feed research team and technical staff of Mechara Agricultural Research Center for their assistance in data collection and in facilitating routine field management activities. The Oromia Agricultural Research Institute is also highly acknowledged for funding the research work. International Livestock Research Institute (ILRI) is also acknowledged for availing the genotypes.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data supporting the findings of this study are available from the corresponding authors upon request.

Additional information

Funding

The work was supported by the International Livestock Research Institute.

Notes on contributors

Birmaduma Gadisa

Birmaduma Gadisa was recruited in June 2013 by Oromia Agricultural Research Institute (ORAI) at Mechara Agriculture Research Center as a Forage Agronomy Researcher until June 2020. From June 2020 to date; Mr. Birmaduma has been working as a Researcher at Bako Agricultural Research Center.

Muleta Debela

Muleta Debela was recruited in June 2009 by the ORAI at Mechara Agricultural Research Center as a Forage Agronomy Researcher until October 2019. From October 2019 to date, Mr. Muleta has been working as Researcher at Bedele Agricultural Research Center.

Tamrat Dinkale

Tamrat Dinkale was recruited in April 2011 by ORAI at Mechara Agricultural Research Center as Forage Agronomy Researcher. Until now, Mr. Tamrat has been working as Researcher at Mechara Agricultural Research Center.

Abuye Tulu

Abuye Tulu was recruited in April 2012 by OARI at Bako Agricultural Research Center as Forage Agronomy Researcher. Until now, Mr. Abuye has been working as Researcher at Bako agricultural researcher center.