Participatory evaluation of malt barley (Hordium disticum L.) varieties in barley-growing highland areas of Northwestern Ethiopia

Abstract

Although in Ethiopia, improved varieties of malt barley are released year after year, on-time promotion and distribution of released varieties to smallholder farmers are major research limitations. The study was initiated to explore the magnitude and extent of the performance of recently released malt barley varieties through farmers’ participation. The study was conducted at Guagusa Shikudad, Debaytilatgn, Farta and Lay Gaint, which represent barley-growing highland areas of Northwestern Ethiopia. Twelve malt barley varieties, Bahati, Bekoji-1, EH1847, Fanaka, Grace, HB1963, HB1964, Holker, IBON174/03, Sabini, Sington and Traveller, were used as experimental treatments. Treatments were laid out in a randomized complete block design with three replications. Malt barley varieties EH1847, HB1963 and IBON174/03, showed wider adaptabable as well as relative stable across tested climatic zones. The farmers’ preference traits in malt barley varieties were relatively similar across tested highland areas. Moreover, disease resistant was ranked first across all climatic zones. The rank correlation analysis between varieties’ rank by farmers and varieties’ grain yield rank was positive across the tested climatic zones. Hereby, farmers could select the higher grain yield malt barley varieties based on agronomic preference traits to their agro-ecologies. Therefore, by considering both grain yield performance of the varieties and varieties’ rank by farmers, the recently released malt barley variety HB1963 should be pre-scale in barley-growing highland areas; also relatively older varieties EH1847 and IBON174/03 are used as genetic materials for the seed source in Northwestern Ethiopia.

Keywords:

• farmers preference traits
• malt barley
• matrix ranking
• pairwise ranking
• preference ranking
• rank correlation
• stable variety

PUBLIC INTEREST STATEMENT

In Ethiopia, there have been released and recommended technologies that boost the production and productivity of malt barley. Moreover, it has huge areas suitable for barley production. Particularly, it has released malt barley varieties that are higher in grain yield, that are tolerant to barley diseases and that meet the quality standards of brewery factories. On the other hand, farmers grow older varieties which are lower in grain yield performance. Hence, due to the emergence of brewery factories, the local row malt barley production and two malt factory supply do not meet the demand of 12 brewery factories. Therefore, it is vital to consider the production and productivity of malt barley through farmers’ participation for the purpose of increasing farmers’ income and to slow down the foreign currency in case of high amount of malt import.

Competing Interests

The authors declare no competing interests.

1. Introduction

Malt barley is becoming a major income source to smallholder farmers in the highland areas of Ethiopia, particularly where the agro-ecologies are not more productive to other cereal crops (Ministry of Agriculture [MoA], 2018). However, in Ethiopia, barley productivity (2.66 t/ha) is lower compared to that of other barley-producing countries such as United Arab Emirates, Belgium and Netherlands (8, 7.59 and 7.0 t/ha, respectively) (FAOSTAT, 2018). This is due to the combination of genetic, socioeconomic constraints and inappropriate use of integrated technologies (Bayeh & Stefania, 2011). In addition, malt barley requires optimum environmental factors (altitude, rainfall and soil pH are between 2,300 and 3,000 masl, 500 and 1,000 mm and 5.5 and 6.5, respectively and soil types are well-drained light brown and red soil) due to quality standards of beer factories (Bayeh & Stefania, 2011).

Ethiopia has large suitable cultivated land for barley production, which covers about 970,053 ha per year. It produces 347,497 tons barley per year (Central Statistics Agency, Ethiopia [CSA], 2018). Similarly, Ethiopia has a high demand for raw malt barley products due to the older established and new emerging malt and brewery factories. It has a total of four malt (two on process) and 12 brewery factories (Asokoinsight, 2019). In consideration of suitable agro-ecology and high demand for malt barley products by malt and brewery factories, Ethiopia has established a malt barley market value chain from the farmers to malt and brewery factories (Addisu, 2018; NBE, 2017). As Berhanu (2013) reported, farmers in the highland areas of Ethiopia have simply produced barley for food security and local markets at lower prices due to lack of market value chains. Now due to the emergence of malt and brewery factories, contract farming is evolved in malt barley commercialization among farmers, cooperatives, unions, seed enterprises and malt and brewery factories which have a 10% price advantage over noncontracted farming (Addisu, 2018). Even if the malt barley production and productivity are increased year after year, the supply does not meet the demand of emergence of malt and brewery factories. The brewery factories demanded about 118,000 tons malt per year, while the local malt source is 52,000 tons which covers only about 50% of it (Addisu, 2018; Business innovation facility BIF, 2018; NBE, 2017).

Although genetic advances in malt barley varieties have increased year after year through agricultural research institutes and nongovernmental agricultural research organizations, timely adaptation, promotion and diffusion of improved varieties of malt barley by smallholder farmers in the Western Amhara Region are limited, which widely and commonly grows older malt barley variety like Holker released in 1979. However, in Ethiopia, there have been recently released malt barley varieties that have attainable grain yield and malt quality in barley-growing areas (MoA, 2018).

Farmers' participation in variety evaluation and development in Amhara Region also is not emphasized as much as demanded in the research system. However, improved varieties may not be adaptable to different agro-ecologies without the smallholder farmers’ participation and involvement of their indigenous knowledge in variety development due to climate variability (Chiara et al., 2017) and may also affect the speed and duration of variety development and the acceptance of released varieties by farmers without their involvements (Assefa et al. 2006). Moreover, farmers’ participation in variety evaluation is a pertinent technique to alleviate the farmers doubt’ on the technologies, to select demanded varieties, to create skills and knowledge on genetic diversities in the breeding program (Ceccarelli, 2012) and to reduce time and resource wastage for technology diffusion as well (Bellon & Reeves, 2002). Farmers can be evaluating the number of varieties for performance-based criteria adapting to their agro-ecosystems (Smolders, 2006). Hereby, it is important to assess the genetic potential and farmers' preference traits in malt barley varieties to their agro-ecologies. The study was initiated with the objectives of evaluating recently released malt barley varieties through farmers’ participatory variety evaluation approach to select and promote the high-grain-yielding malt barley varieties and to create awareness of the variety development process and technology distribution to the end-users.

2. Materials and methods

2.1. Description of study areas

The study was conducted in Guagusa Shikudad, Debaytilatgn, Farta and Lay Gaint, which represent the barley growing areas of the Western Amhara Region. The experimental sites of agro-ecological data are listed in Table 1.

Table 1. Agro-ecological data of the experimental areas

Districts	Code	Altitude (masl)	Soil type	Rainfall (mm)	Temperature (°C)		Humidity
					Maximum	Minimum
Lay Gaint	E1	3,002	Brown	1,106	18.5	9.4	75.8
Lay Gaint	E2	2,950	Light brown	1,106	18.5	9.4	75.8
Farta	E3	2,883	Light brown	779	21.8	11	69.7
Farta	E4	2,650	Light brown	779	21.8	11	69.7
Debaytilatgn	E5	2,862	Brown	1,082	21.1	6.2	-
Debaytilatgn	E6	2,660	Light brown	1,082	21.1	6.2	-
Guagusa Shikudad	E7	2,413	Red	-	-	-	-

Source: Ethiopia Meteorology Agency, Bahirdar Branch.

2.2. Experimental materials and design

Twelve malt barley varieties, namely, HB1963, HB1964, Singitan, Traveler, Fanaka, Grace, EH1847, IBON174/03, Sabini, Bahati, Bekoji-1 and Holker, were used as experimental treatments (Table 2). The treatments were laid out in a randomized complete block design with three replications. The gross and net harvestable plot area were 2.5 m by 2 m and 2.5 m by 1.6 m, respectively. Spacing between rows, plots, and replications was 0.2 m, 0.4 m and 1.5 m, respectively.

Table 2. Description of malt barley varieties used for the study

Variety name	Code	Pedigree	Breeder center	Year of release	Altitude (masl)	Rainfall (mm)	DM	Grain yield (qt/ha)		Protein (%)	Extraction %)
								Research field	On farm
HB1963	V6	PFC9215/3/ZHEDAR#SHYRI//OLMO	HARC	2016	>2,300	500–700	146	35-60	-	10.6	81
HB1964	V7	RECLA78//SHYRI/3/ATAH92/GOB	HARC	2016	>2,300	500–700	138	3,356	-	11.5	80
Singitan	V11	IBON-MRA	SARC	2016	2,200–2,600	750–1,000	119	3,141	21–35	11.5	78
Fanaka	V4	HKBL1512-5	HARC	2015	2,000–2,600	500–1,000	125	2,638	23–31	10.5	78
Traveler	V12	-	HARC	2013	2,000–2,600	500–1,000	145	2,545	20–40	10.5	-
Grace	V5	-	HARC	2013	2,000–2,400	500–1,000	125	2,040	18–35	10.4	-
IBON17/03	V9	-	HARC	2012	2,000–2,800	500	120	3,057	-	10	-
Sabini	V10	-	KARC	2011	2,300–2,500	>700	135	49	25–40	8.5	-
EH1847	V3	EH1847/F4.2p.5.2(BEA/IBON64/91)	HARC	2011	2,200–2,800	>500	141	44	35	11	-
Bahati	V1	-	KARC	2011	2,300–2,800	>700	142	48	25–40	8.7	-
Bekoji-1	V2	EH1293/F2-18B-11-1-14-18	KARC	2010	2,300–2,800	>700	142	50	23–28	11.7	-
Holker	V8		HARC	1979	2,300–3,000	500–800	-	2.4–3.1	-	-	-

Source: MoANR (1979–2016).

2.3. Management practices

The trial was planted between the end of May up to the second week of June, with a seed rate of 100 kg/ha. Fertilizer rates were NPS 100 kg/ha for all environments and urea for debaytilatgn was 100 kg/ha, while that for Guagusa Shikudad, Farta and Lay Gaint were 150 kg/ha. All NPS and one-third of urea were applied at planting, whereas the remaining two-third of urea applied at the tillering stage. Weeding was done two times at tillering and booting growth stage across environments.

2.4. Data collection and statistical analysis

2.4.1. Biological data collection and analysis

The collected agronomic traits such as days to physiological maturity, plant height, spike length, number of seeds per spike, thousand seed weight and test weight were analyzed using GenStat software (17^th edn) for the analysis of variances of varieties, environments and their interactions. Fisher’s protected least significant difference method (P ≤ 0.05) was used for mean separation among varieties.

2.4.2. Social data collection and analysis

Participant farmers were selected by Kebele Agricultural Experts regardless of their age, sex, religion and education level per district. Farmers were drawn up the preference traits of malt barley varieties based on their interest through focus group discussion. A total of 40 farmers, with 10 farmers per environment, participated in between dough to physiological maturity growth stages of malt barley varieties. Among seven environments, four environments were selected which show vigor in performance for variety evaluation by farmers. Each farmer per environment ranked each variety per preference traits in three groups as very good (1), good/medium (3) and poor (5). Farmers’ preference traits were analyzed using pairwise ranking. According to the collected social data, varieties’ rank by farmers was done using preference ranking for Debaytilatgn and Guagusa Shikudad nd using matrix ranking for Farta and Lay Gaint. Rank correlation analysis was done between varieties’ grain yield rank and varieties’ rank by farmers using Spearman’s rank correlation analysis as per the following equation:

Rs = 1-(6∑d²/(n³-n)) experesed in percentasge,

where d= difference in the ranks assigned to the same individual or phenomenon and n= number of individuals or phenomena ranked.

3. Results and discussion

3.1. Performance of yield-related traits in malt barley varieties across environments

The varieties, environments and their interaction were showed significant (P ≤ 0.05) differences, except the interactions of varieties by environments for trait number of seeds per spike in the tested malt barley varieties. The variety Singitan was relatively early maturing, while Traveler, Grace, Fanaka, Bahati, HB1963 and HB1964 were late-maturing varieties. In height performance, the varieties HB1964, HB1963, Bekoji-1 and Holker were taller, whereas Grace and Traveler were shorter varieties. In spike length performance, HB1964 was higher, while Holker was of a lower variety. The varieties HB1963 and Fanaka were higher for thousand seed weight (TSW) and hectoliter weight (HLW), respectively, whereas Grace was lower for both TSW and HLW response (Table 3). The grain yield performance was significantly influenced by environmental factors and their interactions (Misganaw, 2016; Fentaw et al., 2015; Arega et al., 2013; Tesfaye et al., 2013).

Table 3. Combined mean of yield and yield-related trait performance in malt barley varieties across environments

Varieties	Traits
	GY (qt/ha)	DM	PH (cm)	SL (cm)	NSPS	TSW (g)	HLW (kg/hl)
Bahati	40.33^abcd	135.7^ab	87^e	7.02^ef	27.25	47.76^bc	61.22^def
Bekoji 1	38.9^bcde	134 ^cd	98.1^a	6.64^fg	27.96	46.65 ^c	64.89^b
EH1847	42.13^ab	134.8 ^cd	91.61 ^cd	7.58^d	25.22	44.83^d	63.49^bc
Fanaka	36.71^e	139.8^ab	88.35^de	6.88^ef	28.41	54.68^a	64.37^b
Grace	33.14 ^f	137.1^ab	65.16 ^h	8.63^b	28.94	36.4^e	60.53 ^f
HB1963	41.04^abc	139.9^ab	93.53^bc	6.73^efg	28.14	48.76^b	66.9^a
HB1964	36.37^ef	139.1^ab	94.36^bc	9.47^a	27.27	55.04^a	61.14^def
Holker	37.24^de	135.5 ^cd	95.4^ab	6.43 ^g	27.78	44.84^d	64.81^b
IBON174/3	42.24^a	132.4^de	83.2 ^f	7.08^e	25.9	46.8 ^c	62.27^cde
Sabini	37.38^de	129.9^e	80.54 ^f	7.73 ^cd	27.43	43.23^d	64.66^b
Sington	37.82^cde	126 ^f	76.52 ^g	8.6^b	26.33	48.15^bc	62.46 ^cd
Traveler	39.99^abcd	140.8^a	63.86 ^h	8.08 ^c	27.29	44.14^d	60.92^ef
Mean	38.61	135.41	84.8	7.58	27.33	46.77	63.14
CV %	13.8	4.1	9	8.4	8	6.1	3.6
LSD (5%)	8.56	8.88	6.6	1	3.5	4.56	3.72
Var	**	**	**	**	**	**	**
Env	**	**	**	**	**	**	**
Var*Env	*	**	*	**	NS	**	**

** ighly significant a;t P ≤ 0.01; * significant at P ≤ 0.05; NS, nonsignificant; GY, grain yield; DM, days to maturity; PH, plant height; SL, spike length; NSPS, number of seeds per spike; TSW, thousand seed weight; HLW, hectoliter weight; Var, variety, Env, environment; Var ×Env = interaction of varieties by environments.

3.2. Grain yield performance of malt barley varieties across environments

The additive main effects and multiplicative interaction(AMMI) analysis of variances for grain yield in the varieties, environments and their interaction showed significant (P ≤ 0.05) differences. Grain yield performance variation of the tested malt barley varieties expressed by environments, variety by environment interactions and varieties was 73.58%, 16.62% and 9.79%, respectively (Table 4). The malt barley varieties EH1847 (V3) followed by HB1963 (V6) and IBON174/03 (V9) were higher grain yielders and stable across tested environments. While the malt barley varieties Grace (V5), Fanaka (V4) and HB1964 (V7) were inadaptable and unstable across the tested environments (Table 5 and Figure 1). In the plot, the best-performing malt barley varieties are closer to the average environmental coordinate (AEC) circle with the ranking lines (Yan & Kang, 2003). The varieties in the biplot with PC1 scores>0 and PC2 scores near to zero are adaptable and stable, respectively, whereas PC1 scores<0 and higher PC2 scores both + and—sings from the biplot lines inadaptable and unstable, respectively, across the environments (Zerihun, 2011). As depicted in Figure 2, malt barley varieties EH1847 (V3) followed by IBON174 (V9) and HB1963 (V6) were highest grain yielders which are enclosed by concentric circles closer to the ideal circle and/or AEC than the tested varieties across the tested environments. The varieties closest to the ideal genotype drawn on the center of concentric and/or AEC are the highest yielders (Zerihun, 2011; Yan & Kang, 2003).

Figure 1. Genotype and genotype by environment (GGE) ranking biplot of grain yield in malt barley varieties across environments

V1 = Bahati, V2 = Bekoji-1, V3 = EH1847, V4 = Fanaka, V5 = Grace, V6 = HB1963, V7 = HB1964, V8 = Holker, V9 = IBON174/03, V10 = Sabini, V11 = Sington, V12 = Traveller, E1 = Lay Gaint, E2 = Lay Gaint, E3 = Farta, E4 = Farta, E5 = Debaytilatgn, E6 = Debaytilatgn, E7 = GGuagusa Shikudad, AEC = average environmental coordinate,PC = principal component.

Figure 2. GGE comparison biplot of grain yield in malt barley varieties across environments

V1 = Bahati, V2 = Bekoji-1, V3 = EH1847, V4 = Fanaka, V5 = Grace, V6 = HB1963, V7 = HB1964, V8 = Holker, V9 = IBON174/03, V10 = Sabini, V11 = Sington, V12 = Traveller, E1 = Lay Gaint, E2 = Lay Gaint, E3 = Farta, E4 = Farta, E5 = Debaytilatgn, E6 = Debaytilatgn, E7 = GGuagusa Shikudad, AEC = average environmental coordinate; PC = principal component.

Table 4. AMMI analysis of variances for grain yield in 12 malt barley varieties across seven environments in 2017 cropping season

Source	DF	SS	MS	F pr	% SS (Var+Env+Var*Env)
Total	251	19,279	76.8
Treatments	83	15,794	190.3	<0.001
Genotypes	11	1,547	140.7	<0.001	9.79
Environments	6	11,622	1,937.1	<0.001	73.58
Block	14	1,107	79.1	<0.001
Interactions	66	2,625	39.8	<0.001	16.62
IPCA 1	16	1,226	76.6	<0.001
IPCA 2	14	786	56.2	<0.001
Residuals	36	613	17	0.334
Error	154	2,377	15.4

3.3. Farmers’ preference traits in malt barley varieties across environments

Farmers were ranked by their preference traits in malt barley varieties using pairwise ranking methods. The farmers’ preference traits were relatively similar across environments. Among the preference traits, disease resistant was ranked first across all environments as well as other trait ranks were relatively similar across tested environments (Table 6). The study was in line with Semagn et al. (2017) reports; farmers’ selection criteria were very diverse and different in potato varieties across agro-ecologies and growing seasons. On the other hand, Chiara et al. (2017) reported that farmers’ selected desirable trait rank was varied in durum wheat genotypes across environments. In this study, disease reaction was more emphasized by farmers, but in Chiara et al. (2017) study, early maturity was prioritized by farmers. It might be because of the fact that the studies were conducted in different agro-ecologies.

Table 5. Grain yield (qt/ha) performance in 12 malt barley varieties across seven environments

Code	Varieties	Environments							Mean
		Laigaint-E1	Lay Gaint (E2)	Farta (E3)	Farta (E4)	Debaytilatgn (E5)	Debaytilatgn (E6)	Guagusa Shikudad (E7)
V1	Bahati	53.34	34.96	42.51	27.61	50.76	39.06	34.7	40.42
V2	Bekoji 1	50.84	35.56	38.93	27.01	38.9	41.36	39.71	38.9
V3	EH1847	57.12	38.57	44.57	30.61	44.92	39.19	39.91	42.13
V4	Fanaka	52.38	38.22	36.08	28.16	35.28	31.95	34.91	36.71
V5	Grace	46.46	31.34	31.76	20.66	46.16	33.03	23.91	33.33
V6	HB1963	50.85	37.7	44.21	30.95	47.6	40.17	35.11	40.94
V7	HB1964	48.66	35.67	34.12	33.3	35.67	32.33	35.82	36.51
V8	Holker	43.40	38.33	36.28	20.41	43.3	43.69	35.92	37.33
V9	IBON174/3	57.35	41.6	42.54	36.1	41	38.25	37.16	42
V10	Sabini	52.17	34.83	41.64	28	45.4	33.95	29.02	37.86
V11	Singitan	52.11	35.79	34.94	26.95	40.55	38.67	34.07	37.58
V12	Traveler	58.46	44.47	37.95	29.69	42.58	35.52	31.25	39.99
	Mean	51.93	37.25	38.79	28.29	42.68	37.26	34.29	38.64
	CV %	9.4	9	10.9	11.6	10.6	10	9	11.6
	LSD (5%)	8.26	5.7	7.16	5.57	7.69	6.3	5.25	7.32
	P-level	0.03*	0.01**	0.01**	<0.01**	0.01**	0.01**	<0.01**	<0.01**

** Highly; sign;ificant at P ≤ 0.01; * significant at P ≤ 0.05; NS, nonsignificant GY, grain yield; Var, variety Env, environment; Var ×Env, interaction of varieties by environments.

Table 6. Pairwise ranking of traits by farmers across environments in 2017 cropping season

Rank	Environments
	Farta	Lay Gaint	Debaytilatgn	Guagusa Shikudad
1	Disease resistant	Disease resistant	Disease resistant	Disease resistant
2	Spike morphology	Grain size	Spike morphology	Spike morphology
3	Tillering	Spike length	Tillering	Tillering
4	Plant height	Tillering	Plant height	Plant height
5		Plant height

3.4. Rank of malt barley varieties by farmers across environments

Although the farmers’ preference traits were relatively similar across the tested environments, the ranks of varieties by farmers were relatively different across the tested environments (Tables 7, 8 and 9). Hence, it is important to correlate varieties’ grain yield performance rank and farmers’ rank in malt barley varieties. As depicted in Table 10, the Spearman rank correlation between grain yield rank of malt barley varieties and farmers’ rank of malt barley varieties was at Farta (r = 0.5) and Guagusashikuda (r = 0.56). Therefore, at Farta and Guagusashikudad, Farmers were strong positive to select the higher grain yielding malt barley varieties,while at Lay Gaint (r = 0.05) and Debaytilatgn (r = 0.12) were weak positive to select the higher grain yielding malt barley varieties, it might be due to when the environments were not discriminating the varieties which shows relatively similar in phenotypic performance. The study was in line with Mohammadi et al. (2011) Farmers were efficient in identifying the best genotypes for their specific environment, Zerihun et al. (2012) Farmers were able to identify the higher yielding varieties as breeders, Mahmoud et al. (2014) reported that a significant positive correlation between the farmers’ score and the grain yield response (r = 0.6) in barley genotypes and Molla and Tsedalu (2012) indicated the statistically significant correlation (P < 0.01) among farmers and breeders with grain yield response of the varieties as well as between breeders and farmers.

Table 7. Matrix ranking of malt barley varieties by farmers in Farta and Lay Gaint in 2017 cropping season

Districts	Traits weighted score	Score of the varieties by farmers
		Sabini	EH1847	Grace	Holker	HB1963	HB1964	Bahati	Bekoji 1	Traveler	IBON174	Fanaka	Sington
Farta (E1)	Disease resistant (1)	19	12	30	18	10	11	13	11	21	20	12	30
	Spike morphology (2)	26	28	22	44	34	24	36	44	32	32	40	26
	Tillering (3)	45	39	39	63	33	54	60	60	48	39	60	90
	Plant height (4)	84	40	116	48	40	52	72	48	116	84	96	108
Total score		174	119	207	173	117	141	181	163	217	175	208	254
Rank		6	2	9	5	1	3	8	4	11	7	10	12
Lay Gaint-E1	Disease resistant (1)	18	16	30	21	18	17	17	18	20	14	18	21
	Grain filling status (2)	40	34	38	50	28	40	44	34	54	26	26	50
	Spike length (3)	48	54	51	78	42	30	54	63	51	36	45	36
	Tillering (4)	96	68	100	100	60	80	64	48	84	72	52	80
	Plant height (5)	105	80	140	95	60	70	65	55	150	100	55	100
Total score		307	252	359	344	208	237	244	218	359	248	196	287
Rank		9	7	12	10	2	4	5	3	12	6	1	8

Note: score of the variety = summation of rank of varieties (1–3) for each trait × number of farmers (1–10) and total score = summation of traits weighted score × score of the variety, the lower the score the variety was desirable by farmers.

Table 8. Preference ranking of malt barley varieties by farmers in Debaytilatgn in 2017 cropping season

Variety	Preference ranking by individual farmers							Rank index	Rank
HB1964	1 (1)	2 (6)	6 (1)	7 (1)	8 (1)			33	2
HB1963	1 (6)	2 (4)						14	1
Singitan	10 (2)	11 (1)	12(7)					115	12
Traveler	4 (1)	5 (1)	6 (1)	7 (5)	8 (1)	9(1)		67	7
Fanaka	3 (2)	4 (6)	6 (1)	8 (1)				36	3
Grace	2 (1)	3 (1)	6 (1)	7 (1)	8 (1)	9(1)	10(4)	75	9
EH1847	2 (1)	4 (1)	5 (1)	7 (1)	8 (2)	9(2)	11(2)	74	8
IBON174/3	4 (1)	6 (1)	7 (1)	8 (1)	9 (2)	10(2)	11(2)	85	10
Sabini	2 (1)	6 (1)	9 (2)	10 (1)	11 (2)	12(3)		94	11
Bahati	4 (2)	5 (3)	7 (1)	8 (1)	9 (2)	10(1)		66	6
Bekoji 1	4 (2)	5 (1)	6 (3)	7 (1)	8 (3)			62	5
Holker	3 (1)	5 (2)	6 (4)	7 (2)	9 (1)			60	4

Note: 1 (1) = first number was variety rank and numbers in brackets were number of farmers, rank index = summation of variety rank × number of farmers, the lower the rank index the variety was desirable by farmers.

Table 9. Preference ranking of malt barley varieties by farmers in Guagusa Shikudad in 2017 cropping season

Variety	Preference ranking by farmers						Rank index	Rank
HB1964	1 (8)	6 (1)	9 (1)				23	1
HB1963	1 (1)	6 (3)	7 (1)	8 (3)	10 (1)		60	8
Singiton	2 (6)	3 (1)	4 (1)	9 (1)	10 (1)		38	3
Traveler	11 (4)	12 (6)					116	12
Fanaka	5 (1)	6 (2)	7 (2)	8 (2)	9 (2)	10(1)	57	6
Grace	7 (1)	10 (1)	11 (5)	12 (3)			108	11
EH1847	3 (1)	4 (3)	5 (1)	6 (1)	7 (1)	8(1)	41	4
IBON174/3	2 (4)	3 (3)	4 (2)	5 (1)			30	2
Sabini	3 (4)	4 (1)	6 (1)	8 (1)	9 (1)	10(2)	59	7
Bahati	6 (2)	8 (1)	9 (2)	10 (5)			88	10
Bekoji 1	3 (2)	4 (2)	5 (6)				44	5
Holker	5 (1)	6 (1)	7 (5)	8 (1)	9 (1)	12(1)	75	9

Note: 1 (8) = first number is variety rank and numbers in brackets is number of farmers, rank index = summation of variety rank × number of varieties, the lower the rank index the variety was desirable by farmers.

Table 10. Spearman rank correlation analysis between variety rank by grain yield response and farmers’ rank across environments

Variety	Farta (E3)	GYR	FR	DNR	D^2	Lay Gaint (E1)	GYR	FR	DNR	D^2	Debaytilatgn (E6)	GYR	FR	DNR	D^2	Guagusa Shikudad (E7)	GYR	FR	DNR	D^2
Bahati	42.51	4	8	(4–8)	16	53.34	8	5	(8–5)	9	39.06	5	6	(5–1)	1	34.7	8	10	(8–10)	4
Bekoji 1	38.93	6	4	(6–4)	4	50.84	9	3	(9–4)	36	41.36	2	5	(2–5)	9	39.71	2	5	(2–5)	9
EH1847	44.57	1	2	(1–2)	1	57.12	3	7	(3–7)	16	39.19	4	8	(4–8)	16	39.91	1	4	(1–4)	9
Fanaka	36.08	9	10	(9–10)	1	52.38	5	1	(5–1)	16	31.95	12	3	(12–3)	81	34.91	4	6	(4–6)	4
Grace	31.76	12	9	(12–9)	9	46.46	11	12	(11–12)	1	33.03	10	9	(10–9)	1	23.91	12	11	(12–11)	1
HB1963	44.21	2	1	(2–1)	1	50.85	8	2	(8–2)	36	40.17	3	1	(3–1)	4	35.11	7	8	(7–8)	1
HB1964	34.12	11	3	(11–3)	64	48.66	10	4	(10–4)	36	32.33	11	2	(11–2)	81	35.82	6	1	(6–1)	25
Holker	36.28	8	5	(8–5)	9	43.40	12	10	(12–10)	4	43.69	1	4	(1–4)	9	35.92	5	9	(5–9)	16
IBON174	42.54	3	7	(3–7)	16	57.35	3	6	(3–6)	9	38.25	7	10	(7–10)	9	37.16	3	2	(3–2)	1
Sabini	41.64	5	6	(5–6)	1	52.17	6	9	(6–6)	9	33.95	9	11	(9–11)	4	29.02	11	7	(11–7)	16
Singitan	34.94	10	12	(10–12)	4	52.11	7	8	(7–8)	1	38.67	6	12	(6–12)	36	34.07	9	3	(9–3)	36
Traveler	37.95	7	11	(7–11)	16	58.46	1	11	(1–11)	100	35.52	8	7	(8–7)	1	31.25	10	12	(10–12)	4
∑d^2					142					273					252					126
SRC = 1-(6*∑d^2/(n^3-n))					0.5					0.05					0.12					0.56

4. Conclusion and recommendation

The AMMI analysis of variances for grain yield in the varieties, environments and their interaction showed significant (P ≤ 0.05) differences across environments. The variation for grain yield performance in malt barley varieties accounted by environments, variety by environment interactions and varieties was 73.58%, 16.62% and 9.79%, respectively. According to GGE ranking and comparison biplot analysis, among 12 malt barley varieties EH1847 (V3) followed by HB1963 (V6) and IBON174 (V9) showed higher grain yield and relative stability across tested environments. The preference traits and the rank of traits by farmers were relatively similar across tested environments because the study was conducted in relatively similar highland barley-growing areas. Disease resistant in malt barley was ranked first across all environments. The ranks of varieties by farmers were relatively different across the tested environments. Hence, it is important to correlate varieties’ grain yield performance rank and farmers’ rank in malt barley varieties. The rank correlation between varieties’ grain yield performance rank and farmers’ rank in Farta (rc = 0.5) and Guagusashikuda (rc = 0.56) was strongly positive to select the higher yielding potential malt barley varieties, while that in Lay Gaint (rc = 0.05) and Debaytilatgn (rc = 0.12) was weakly positive to select the higher grain yield malt barley varieties. This might be due to when the environments were not discriminating the varieties in performance. Hereby, farmers could select the higher grain yielder malt barley varieties only using few agronomic traits and disease reactions. In the study, therefore, by considering both statistical significant grain yield differences and varieties’ rank by farmers in malt barley varieties, recently released malt barley variety HB1963 and relatively older varieties EH1847 and IBON174/03 should be demonstrated and/or pre-scale up in Farta, Lay Gaint, Guagusa Shikudad and Debaytlatgn. Plant Breeders could be considered the farmers preference traits in malt barley breeding investigation to develop farmers demanded variety.

Cover Image

Source: Author

Acknowledgements

The authors acknowledge Adet Agricultural Research Center, Amhara Agricultural Research Institute, AGP-II project and ICARDA Malt Barley-Fababean Project for the financial and technical supports. The acknowledgment also forward to Farmers and Agricultural Experts thoroughly participated on the field works and selection of malt barley varieties..

Additional information

Funding

The authors received no direct funding for this research.

Notes on contributors

Misganaw Ferede

Misganaw Ferede was born on November 1988 in Ethiopia. He holds BSc degree in the field of Crop Science from Mekelle University in July 2010. Since December 2010, he was employed as Junior Researcher in Finoteselam Agricultural Research Sub-Center. He then received MSc degree in the field of Plant Breeding from Bahir Dar University in July 2015. Since July 2015 up to now, he has been working as a Plant Breeder on Cereal Crops Breeding program in Adet Agricultural Research Center. He served as a Director of Finoteselam Agricultural Research Sub-Center from 2012 to 2013 and a Coordinator of Cereal Crops Research Program in Adet Agricultural Research Center from 2017 to 2018. He has been developed improved varieties, published the journal articles in East Africa Science Journal, International Journal of Sustainable Agricultural Research,Cogent Food and Agriculture and Proceedings in Amhara Agricultural Research Institute.