Assessment of farmers’ perceptions of production constraints, and their trait preferences of sorghum in western Ethiopia: implications for anthracnose resistance breeding

ABSTRACT

Sorghum, Sorghum bicolor (L.) Moench is an important food security crop widely grown by smallholder farmers in sub-Saharan Africa (SSA), including Ethiopia. In SSA, the potential of sorghum production and productivity has not been realised due to an array of constraints. Colletotrichum sublineolum disease is one of the main biotic constraints causing significant yield losses. The objectives of this study were to assess farmers’ perception, preferences and constraints to sorghum production and productivity in western Ethiopia, and to identify key drivers for anthracnose resistance breeding. A participatory rural appraisal study was undertaken in six selected districts in the East Wellega and West Shewa Administrative Zones in Ethiopia. Data were collected through structured questionnaires involving 165 respondent farmers. Further focus group discussions were held with 180 farmers and development agents. Sorghum is the third most preferred cereal crop after Eragrostis tef and Zea mays in western Ethiopia. About 79% of interviewed farmers cultivated sorghum during the study season. Farmers perceived that sorghum production is constrained by anthracnose disease, bird attack, loose smut and covered smut diseases. The most important farmer-preferred traits in sorghum varieties in the study areas were anthracnose resistance and tolerance to bird attack. Breeding sorghum varieties with the farmer-preferred traits is an important consideration to enhance productivity and adoption of improved sorghum cultivars in western Ethiopia.

KEYWORDS:

• Farmers’ preferences
• PRA
• production constraints
• Sorghum bicolor
• western Ethiopia

Introduction

Sorghum bicolor (L.) Moench is the fifth most important cereal crops next to maize (Zea mays L.), rice (Oryza sativa L.), bread wheat (Triticum aestivum L.) and barley (Hordeum vulgare L.) in the world (FAO 2017). It is the third most important cereal crop after maize and rice in sub-Saharan Africa. Ethiopia is the third largest sorghum producer in Africa after Nigeria and Sudan (FAO 2017). In Ethiopia, sorghum is the third most important crop in area of production after tef and maize, and third in volume of production following maize and tef (CSA 2017).

Potential sorghum production is affected by abiotic and biotic factors, and socio-economic constraints. Diseases, insect pests and weeds are key biotic impediments to yield and quality losses in sorghum production.

In Ethiopia, the sorghum varieties that have been officially promoted were sourced from other countries, or locally bred using externally sourced genetic resources and landraces. However, their adoption by sorghum growers has been low because of their lack of farmer-preferred traits such as resistance to bird attack and plant diseases, poor cooking and food quality, and other undesirable plant attributes such as being short-stemmed (McGuire 2008). Therefore, there is a need to develop sorghum varieties that accurately reflect the needs and preferences of sorghum growers and the marketplace.

During sorghum variety and production technology development, mostly researchers conceive the research agenda, and then develop and release the new technology to farmers with limited participatory contributions by the farmers. For this reason, most modern sorghum varieties were less adopted by the farmers for whom the varieties were bred. Several authors have reported that the majority of sorghum farmers in Ethiopia continue to grow their own low yielding and unimproved local landraces because there are no improved varieties that have been bred with a full complement of farmer-preferred traits, a lack of awareness or a lack of appropriate technologies for their growing environment (Mulatu and Zelleke 2002; Lacy et al. 2006; Gebretsadik et al. 2014). In particular, the farmers have made it clear that they prefer their local sorghum landraces because they are multipurpose varieties that meet the criteria for grain quality, animal feed and high biomass production for fuel, thatching and construction material (Mekbib 2006).

Farmers’ involvement in problem identification, setting the priority and technology evaluation processes enables plant breeders to develop client-oriented varieties and new production technologies that will achieve high adoption levels. The views and preferences of farmers during variety development and technology evaluation are necessary preconditions for researchers to design, develop and prioritise their research agenda. Participatory rural appraisal is one of the multidisciplinary tools that helps to capture farmers’ perceptions and preferences. Understanding farmers’ needs and preferences enhances the adoption of new technologies and fosters diffusion through farmer-to-farmer technology exchange, among others (Mulatu and Belete 2001).

Sorghum is an important cereal crop in western Ethiopia for home consumption and income generation. The region is characterised by warmer temperatures and higher relative humidity associated with high annual rainfall which varies from 1200 to 1600 mm. Consequently, sorghum diseases including anthracnose, grain mold, leaf blight diseases cause reduced yields and quality of grain. No recent studies have been conducted in this region to document farmers’ perceptions on the production constraints affecting sorghum production including sorghum anthracnose disease, and the preferred traits that farmers demand of a new sorghum variety. This information can be useful for designing client-oriented sorghum improvement and variety design. Therefore, the objective of this study was to assess farmers’ perception, preferences and constraints affecting sorghum production and productivity in western Ethiopia.

Materials and methods

Description of the study areas

The study was conducted during November to December 2016 in two zones (East Wellega and West Shewa) in the Oromia Regional State of Ethiopia (Table 1) in western Ethiopia where sorghum is widely grown and sorghum diseases are prevalent owing to the high humidity and rainfall conditions of the region. Diga, Sibu Sire and Wayu Tuqa districts were selected from the East Wellega Zone, while Bako Tibe, Dano and Illu Gelan districts were included from the West Shewa Zone (Figure 1).

Figure 1. Map of Ethiopia showing the study sites.

Table 1. Number of producers, cultivated areas, total production and yield of sorghum in some selected zones in Oromia Regional State of Ethiopia (CSA 2017).

Zones	HHLDs	Area (hectare)	Production (tonnes)	Yield (tonnes ha^-1)
East Wellega	96,874	36,605	111,254	3.0
Ilu Aba Bora	176,052	41,595	120,150	2.9
Jimma	210,037	50,155	124,397	2.5
West Shewa	162,643	72,176	220,888	3.1
West Wellega	235,942	63,816	191,358	3.0

Note: HHLDs = Households; ha = hectare.

The study districts are known for their high total rainfall, which varies from 1200 to 1600 mm per year. The main rain season starts in May and ends in October. The areas receive maximum rainfall in the months of June to August. The main crops cultivated include cereals (finger millet, maize, sorghum and tef), pulses (faba bean, field pea, haricot bean and soybean), oil crops (groundnuts, noug [Guzotia abyssinica Cass.] and sesame) and horticultural crops (hot pepper, potato, sweet potato, pumpkin and tomato). These crops are mainly grown for cash and home consumption.

Sampling procedure

Multi-stage sampling was applied, based on the dominance of sorghum production and prevalence of diseases (anthracnose, grain mold and leaf blight), weeds and insect pests. From each district, two peasant associations were selected, except in the Sibu Sire district where only one peasant association was included. In total, 11 peasant associations were included in the study (Table 2). In each peasant association 15–18 farmers were sampled, providing a total of 165 and 180 farmers that participated during the interviews and focus group discussions, respectively. Among the interviewed farmers 14 were females and 151 were males, while 14 females and 166 males participated in the focus group discussions (Table 2).

Table 2. Number of sampled farmers for individual interview and focus group discussion in selected districts and peasant associations of the East Wellega and West Shewa Zones.

Zones	Districts	Peasant associations	No. of interviewees			No. of focus group discussants
			Female	Male	Total	Female	Male	Total
East Wellega	Diga	Arjo	3	12	15	3	13	16
		Bech Bech	–	15	15	–	16	16
	Sibu Sire	Jarso Wama	2	13	15	2	14	16
	Wayu Tuqa	Boneya Molo	3	12	15	3	13	16
		Gara Hudha	–	15	15	–	16	16
West Shewa	Bako Tibe	Amerti Gibe	–	15	15	–	18	18
		Gudina Welqite	2	13	15	2	14	16
	Dano	Kersa Dorane	2	13	15	2	14	16
		Seyu Gudetu	1	14	15	1	15	16
	Illu Gelan	Obera Benesu	–	15	15	–	16	16
		Refso Qameno	1	14	15	1	17	18
Total			14	151	165	14	166	180

Farmers were randomly selected for the key informant interviews based on their locally recognised experience of sorghum production, management and utilisation. Participants were selected randomly from all wealth groups and generations (seniors, adults and youth) to ensure representativeness of the farmers. Development agents were involved in focus group discussions as key informants and being known by the farmers in each peasant association. Semi-structured questionnaires were designed on topics related to sorghum production constraints, focusing on anthracnose, grain mold, leaf blight diseases; sorghum trait preference, and coping mechanisms towards anthracnose disease. Fresh leaf samples showing sorghum leaf blight and anthracnose were collected from sorghum fields and shown to the farmers during the individual interview and focus group discussions to differentiate the two diseases.

Data collection

Both primary and secondary data were collected. Primary data were collected through interviews and focus group discussions. Demographic features, crop production and uses, sorghum production constraints, knowledge and importance of anthracnose and disease management strategies were recorded. Researchers from different disciplines of the Bako Agricultural Research Centre (BARC) (Socio-Economist, Breeders, Pathologist and Weed Scientist) were involved in facilitating and collecting data. The local language, ‘Afaan Oromo’, was used throughout the interview and focus group discussions to communicate easily and to discuss freely. Secondary data were collected from zonal and district agricultural offices of the respective zones and districts used in the study.

Data analysis

Both quantitative and qualitative data were subjected to analyses using SPSS computer software (SPSS 2005). The quantitative data were coded before analysis. Descriptive statistics such as frequencies and percentages were calculated. In addition, chi-square and t-tests were performed using the cross-tabulation procedure of SPSS.

Results and discussion

Demographic features of households in the study area

Among the interviewed farmers, 79% of them cultivated sorghum during the 2015 main cropping season. Ninety-two per cent of the interviewed farmers were males and 8.5% were females. Relatively few female farmers participated in this survey except in the Sibu Sire district, which had 13.3% female respondent farmers (Table 3). Most of the household heads were aged between 30-50 years (53.9%) (Table 3). There were significant differences (P < .05) in family size among districts. A majority of interviewed farmers (70.9%) had more than 5 children, while 28.5% had 3–5 children per household, 0.6% had less than 3 children per family (Table 3). Large families provide labour for farming activities (Gebretsadik et al. 2014), but they also lead to an increased demand for food (Abraha et al. 2017). Among the surveyed districts, Sibu Sire followed by Dano had the highest proportion of illiterate farmers with 46.7% and 23.3%, respectively. Most of the respondent farmers in Bako Tibe, Diga, Illu Gelan, Dano and Wayu Tuqa had education levels ranging from Grades 1 to 6. It is believed that a better level of education is a precondition for accepting improved agricultural technologies and farm practices in northern Ethiopia (Abraha et al. 2017). Most of the respondent farmers (40.6%) attended elementary school (Grade 1 to 6 education) and 27.3% attended secondary education (above Grade 6). 15.8% could read and write but 16.4% respondents could not read and write (Table 3). Households who could read and write attended formal school or adult education programmes.

Table 3. Proportion (%) of respondents’ sex, age, family size and level of education in the study districts.

Variables	Districts						Mean
	Bako Tibe	Illu Gelan	Dano	Sibu Sire	Wayu Tuqa	Diga
Sex
Male	93.3	96.7	90.0	86.7	90.0	90.0	91.5
Female	6.7	3.3	10.0	13.3	10.0	10.0	8.5
Chi-Square test	df = 5		X^2 = 1.87		P-value = .87
Age of household head (year)
≤29	33.3	36.7	26.7	40.0	36.7	16.7	30.9
30–50	46.7	46.7	70.0	53.3	46.7	60.0	53.9
≥ 50	20.0	16.7	3.3	6.7	16.7	23.3	15.2
Chi-Square test	df = 10		X^2 = 11.13		P-value = .35
Number of children
≤ 2	–	–	–	–	–	3.3	0.6
3–5	23.3	6.7	26.7	26.7	50	36.7	28.5
≥ 5	76.7	93.3	73.3	73.3	50	60.0	70.9
Chi-Square test	df = 10		X^2 = 20.08		P-value = .03
Education status
Illiterate	3.3	16.7	23.3	46.7	13.3	10.0	16.4
Read and write	20.0	10.0	10.0	20.0	13.3	23.3	15.8
Grade 1–5	46.7	40.0	40.0	26.7	36.7	46.7	40.6
≥Grade 6	30.0	33.3	26.7	6.7	36.7	20.0	27.3
Chi-Square test	df = 15		X^2 = 22.07		P-value = .11

Note: df = Degrees of freedom; X²= Chi-Square; P = Probability.

Major crops grown and production area, management practices and uses

Maize covered larger production area in the study districts, except in Illu Gelan where tef was the most widely grown crop (Figure 2). Sorghum was the third most important crop in mean area coverage after maize and tef in most surveyed districts. In the Dano and Diga districts sorghum was the second most important crop after maize (Figure 2).

Figure 2. Mean area of farm land (ha) allocated for different crops grown in 2015/2016 cropping season in the study districts.

Table 4 summarises cereal crops grown, management practices and yields reported by respondent farmers in the study areas. Maize, sorghum and tef were the most widely grown cereal crops. Almost all respondent farmers used approximately the same seeding rate (25 kg ha^-1) recommended for maize, while a mean seeding rate of 12.1 and 12.4 kg ha^-1 was used for sorghum production in East Wellega and West Shewa, in that order. A higher mean seeding rate of 22.2 kg ha^-1 was used for tef production in the East Wellega zone, with a slightly higher rate used in the West Shewa Zone (28.2 kg ha^-1).

Table 4. Comparison of management practices and yields of major cereal crops grown in the study districts of East Wellega and West Shewa Zones in western Ethiopia.

		East Wellega					West Shewa
Crop	Management practice and yield	Mean	SD	Std.E	Df	t-value	Mean	SD	Std.E	Df	t-value
Maize	DAP fertilizer (kg ha^-1)	97.3	24.6	2.9	71	33.5**	101.7	23.3	2.5	87	40.9**
	Urea fertilizer (kg ha^-1)	147.6	59.1	7.0	71	21.2**	159.7	57.2	6.1	87	26.2**
	Ploughing frequency	3.6	0.7	0.1	71	45.6**	3.8	0.6	0.1	87	55.7**
	Yield (ton ha^-1)	4.0	1.1	0.1	71	30.3**	4.5	1.2	0.1	87	35.9**
Sorghum
	Seeding rate (kg ha^-1)	12.1	4.0	0.5	69	25.1**	12.4	3.7	0.4	70	27.9**
	DAP fertilizer (kg ha^-1)	4.6	18.5	2.2	69	2.1*	12.3	31.0	3.7	70	3.3**
	Urea fertilizer (kg ha^-1)	4.1	18.0	2.1	69	1.9ns	1.6	8.5	1.0	70	1.6ns
	Ploughing frequency	2.3	0.7	0.1	69	28.8**	2.3	0.7	0.1	70	27.5**
	Yield (ton ha^-1)	2.0	0.9	0.1	69	20.0**	2.2	0.7	0.1	70	21.3**
Tef
	Seed rate (kg ha^-1)	22.2	11.1	1.8	36	29.9**	28.2	11.9	1.4	73	20.4**
	DAP fertilizer (kg ha^-1)	35.6	32.0	5.3	36	6.8**	60.3	37.2	4.3	73	13.9**
	Urea fertilizer (kg ha^-1)	26.4	28.2	4.6	36	5.7**	35.5	31.8	3.7	72	9.5**
	Ploughing frequency	3.9	0.8	0.1	36	29.9**	4.1	0.8	0.1	73	45.3**
	Yield (ton ha^-1)	0.9	0.3	0.04	36	21.8**	1.1	0.4	0.04	73	22.7**

Notes: *P < .05; **P < .01; ns = non-significant; SD = Standard deviation; Std.E = Standard error; Df = Degrees of freedom and DAP = Diammonium phosphate.

During the focus group discussions, farmers explained the use of variable seeding rates, which depended on soil fertility. They used lower seeding rates under fertile soil condition because these soils favoured more tillering by sorghum, whereas higher seeding rates were used for less fertile soils in which the tillering ability of sorghum was reduced.

Farmers applied the recommended rates of inorganic fertilizers (DAP and Urea) for maize and tef production. However, most of the respondent farmers did not apply inorganic fertilizers for sorghum. A limited number of farmers used manure to improve the soil fertility of their sorghum fields. In rare cases, farmers practiced fallowing. In addition, farmers occasionally practiced crop rotation (sorghum followed by finger millet, tef and pulses) to improve soil fertility and increase sorghum productivity. Respondent farmers were aware that appropriate crop management practices (proper seeding rates, fertilizer doses and optimal land preparation) resulted in higher mean grain yield of maize of 4.0 ton ha^-1 in the East Wellega zones, and 4.5 ton ha^-1 in the West Shewa Zones. In contrast, sorghum achieved mean grain yields of 2.0 and 2.2 ton ha^-1 in the East Wellega and West Shewa Zones, respectively, which are below the mean national yield of 2.5 ton ha^-1 (FAO 2017).

Table 5 presents major crops grown and their role in western Oromia. The farmers identified several roles of sorghum in their farming systems, including grain for home consumption and sale of grain as a source of cash income. It was recognised as requiring less labour than the other crops, having a greater tolerance to moisture stress and for its strong weed suppression ability (Table 5).

Table 5. Proportion of farmers (%) who grow diverse crops for their various roles in western Oromia Regional State of Ethiopia.

Crops	Home consumption		Income source		Not labour intensive		Tolerant to moisture stress		Suppress weeds
	ES	WS	ES	WS	ES	WS	ES	WS	ES	WS
Sorghum	29.7	28.9	5.4	2.2	47.3	42.2	67.6	65.6	70.7	53.9
Maize	23.0	25.6	28.4	25.6	10.8	6.7	6.8	20	4	7.9
Tef	13.5	34.4	8.1	35.6	2.7	8.9	4.1	10	0	10.1
Finger millet	29.7	3.3	–	–	1.4	2.2	12.2	–	1.3	–
Wheat	–	1.1	–	–	–	1.1	–	–	–	2.2
Haricot bean	–	1.1	1.4	–	2.7	1.1	–	1.1	–	1.1
Groundnut	1.4	–	17.6	–	1.4	–	4.1	-	4	4.5
Hot pepper	2.7	5.6	18.9	13.3	–	–	–	1.1	–	1.1
Sesame	–	–	9.5	–	2.7	–	–	–	–	–
Faba bean	–	–	–	1.1	–	4.4	–	–	–	1.1
Chickpea	–	–	2.7	–	1.4	–	2.7	–	–	–
Noug	–	–	8.1	22.2	29.7	33.3	2.7	2.2	20	18
Total	100	100	100	100	100	100	100	100	100	100
X^2	29.43		49.15		13.47		25.78		15.94
df	7		9		10		8		9
P-value	.00		.00		.20		.00		.07

Note: EW = East Wellega; WS = West Shewa; df = Degrees of freedom ; X²= Chi Square and P = Probability.

(–) denote no role described in respective zone/s.

Many of the respondent farmers in East Wellega (29.7%) and West Shewa (28.9%) primarily consumed sorghum products. 34.4% of the respondents in West Shewa Zone mentioned that tef followed by sorghum is the major food items included in their daily diet. In East Wellega sorghum and finger millet were consumed by approximately similar proportion of respondents (29.7%) (Table 5).

During focus group discussions, respondents indicated that 25% of maize harvest was allocated for home consumption while 75% was marketed for income generation. Conversely, tef, sorghum and finger millet were primarily grown for home consumption as sole or mixed with other crop products. The use of maize products for home consumption has decreased over the last 10 years due to its commercial significance in western Ethiopia. Whereas higher proportion of sorghum grain harvested (75%) was used for home consumption and the remaining was sold at local markets. The trend of using sorghum for home consumption mixed with other cereals or as sole product is increasing across all study districts. White seed types of sorghum is preferred by farmers for home consumption. However, white sorghum is susceptible to sorghum diseases including anthracnose (Duncan et al. 1990).

Respondent farmers cultivated various crops to diversify their cash income. Relatively higher proportion (35.6%) of respondents indicated deriving cash income from the sale of tef followed by maize (25.6%) in West Shewa. However, farmers in East Wellega derive higher cash income from the sale of maize (28.4%) followed by hot pepper (18.9%). Contribution of sorghum as cash income source is minimal in the study area because sorghum is mainly used for home consumption. Groundnut (17.6%) in East Wellega and noug (G. abyssinica Cass.) (22.2%) in West Shewa were regarded as good sources of cash income. Higher number of farmers in East Wellega and West Shewa considered sorghum to be non-labour intensive, tolerant to moisture stress and suppresses weeds (Table 5).

Socio-economic factors influencing sorghum production in western Ethiopia

Table 6 presents the reasons given by farmers for not applying inorganic fertilizers in sorghum production. Most of the farmers in East Wellega (76.5%) and West Shewa (50%) in were aware of the importance of inorganic fertilizers for sorghum production. However, most farmers indicated that a shortage of cash limited their ability to use inorganic fertilizer. This finding is in agreement with the report of Gebretsadik et al. (2014) who indicated that the majority of farmers in Metekel, North Wollo and North Shewa areas of Ethiopia did not use inorganic fertilizers for sorghum production. Further, 20.6% of East Wellega and 46.2% of West Shewa farmers were unsure how to use inorganic fertilizers for sorghum production, while a few farmers (5.5%) thought that inorganic fertilisers are toxic to the soil and are not important for sorghum production (Table 6). About 39.0% of respondent farmers in East Wellega and 22.2% in West Shewa were unable to access improved sorghum varieties released by plant breeders (Table 7).

Table 6. Reasons for not applying inorganic fertilizers for sorghum production by farmers in selected zones of western Ethiopia.

Reasons	East Wellega Zone		West Shewa Zone
	Frequency	Percent	Frequency	Percent
Lack of money	52	76.5	39	50.0
Lack of knowledge	14	20.6	36	46.2
Not useful for sorghum	2	2.9	2	2.6
Lack of labour	0	0	1	1.3
Total	68	100	78	100
X^2	11.91
df	3
P-value	.01

Note: df = Degrees of freedom and X²= Chi Square.

Table 7. Major socio-economic and environmental factors affecting sorghum production in western Ethiopia.

	East Wellega		West Shewa
Factors	Frequency	Percent	Frequency	Percent
Shortage of cash to purchase inputs	29	39.0	20	22.2
Limited availability of improved varieties	27	36.5	19	21.1
Late onset and early termination of rainfall	11	14.9	24	26.7
Shortage of labour	3	4.1	14	15.6
Shortage of fertilizer	2	2.7	6	6.7
Shortage of land	1	1.4	5	5.6
Declined soil fertility	0	0	2	2.2
Shortage of draft power	1	1.4	0	0
Total	74	100	90	100
X^2	21.30
df	7
P-value	.00

Note: df = Degrees of freedom and X²= Chi Square.

BARC had previously released a few novel sorghum varieties such as Lalo, Gemedi and Chemeda. However, these were not adopted by farmers due to limitations associated with each of the varieties, such as their poor grain quality, and the susceptibility to bird attack, except Lalo. Some respondent farmers reported that a late onset of rain hindered the planting of sorghum. The early termination of rainfall before sorghum had reached an appropriate growth stage is recognised as another challenge for sorghum production in the region. A shortage of labour, lack of cash for fertilizers, a lack of access to draft power, and declining soil fertility were identified as additional factors limiting sorghum production in the study areas (Table 7).

Sorghum diseases, insect pests, and weeds in western Ethiopia

Important diseases, insect pests and weeds of sorghum in western Oromia reported by respondent farmers are summarised in Table 8. Sorghum anthracnose and head smut were identified to be prominent diseases affecting sorghum production. Most of the farmers (69%) mentioned anthracnose disease affecting sorghum crop, followed by head smut (21%) (Table 8). Anthracnose is the leading plant disease throughout the developmental stages of sorghum and causes reduced yields and quality of grain. Anthracnose disease has been observed in all sorghum growing areas of the world (Crouch and Beirn 2009; Prom et al. 2012; Tesso et al. 2012). The effects of anthracnose disease are more pronounced in warmer areas where temperature and relative humidity are high (Marley et al. 2005; Mehta et al. 2005; Crouch and Beirn 2009; Burrell et al. 2015). Sorghum anthracnose causes yield losses reaching 70% in susceptible varieties (Thomas et al. 1996). In Ethiopia and most sub-Saharan African countries, anthracnose is a major focus of sorghum breeding research (Chala et al. 2010a; Chala et al. 2010b). Different approaches have been recommended to control sorghum anthracnose such as the use of fungicides, management of the sources of inoculum, planting disease-free seed and crop rotation (Erpelding and Prom 2004; Chala et al. 2010a; Silva et al. 2015). However, the most economic, sustainable and environmentally friendly option to control anthracnose is the use of resistant sorghum varieties developed through breeding (Erpelding and Prom 2004; Singh et al. 2006; Tesso et al. 2012; Li et al. 2013).

Table 8. Importance of diseases, insects and weeds identified by respondents as constraints for sorghum production in western Ethiopia.

Diseases	Frequency	Percent	Bird and insects	Frequency	Percent	Weeds	Frequency	Percent
Anthracnose	113	69	Birds	123	76	Raphanus raphanistrum	96	59
Head smut	34	21	Stem borer	22	14	Sorghum halepense and Eleusine indica	67	41
Leaf blight	10	6	Termite	14	9
Covered smut	4	2	Shoot fly	2	1
Grain mold	3	2
Total	164	100.0		161	100.0		163	100

The other three sorghum diseases reported by farmers were covered smut, leaf blight and grain mold but these appear to have a minimal effect on sorghum production and productivity. Most farmers (76.0%) mentioned that bird attack was a major constraint hindering sorghum production and productivity in the area. Most respondents agreed that the area under sorghum production is decreasing mainly due to bird damage and anthracnose disease. In addition, insect pests such as stem borer, termites and shoot fly are additional problems in the area. A problem with broad-leaved weeds was mentioned by 59% of farmers, while 41% indicated grass weeds damaging sorghum fields (Table 8). Weed control measures in sorghum fields are mainly done manually by family labour (90.2%) and some farmers (8.5%) hired daily labourers, while 1.2% of the farmers used group labours (Table 9).

Table 9. Labour sources for weeding sorghum fields in western Ethiopia.

Source	Frequency	Percent
Family labour	148	90.2
Hired workers	14	8.5
Weeding by group of farmers	2	1.2
Total	164	100.0

Table 10 contained various options practiced by respondent farmers to control anthracnose disease in the area. About 40% of the respondents reported that they minimise the problem of anthracnose disease by growing resistant landraces or introduced varieties. Some interviewed farmers (17.1%) mentioned crop rotation to control anthracnose. Planting disease-free seeds, use of fungicides and improved soil fertility were indicated as anthracnose disease control strategies. Some of the farmers perceived that improved field sanitation, proper land preparation, early planting and roguing out diseased plant parts lowered the prevalence of anthracnose disease (Table 10).

Table 10. Various options used by farmers to control sorghum anthracnose disease in western Oromia.

Control options	Frequency	Percent
Use of resistant landraces or introduced varieties	59	40.4
Crop rotation	25	17.1
Planting disease free seed	10	6.8
Use of fungicides	8	5.5
Improved soil fertility	7	4.8
Field sanitation	6	4.1
Report to agricultural experts	5	3.4
Land preparation with sufficient tillage frequency	4	2.7
Early planting	4	2.7
Roguing out diseased plant	3	2.1
Crop rotation	2	1.4
Weed control	2	1.4
None	11	7.5
Total	146	100.0

Farmers preferred traits, and perceptions of major sorghum diseases

Table 11 presents the sorghum landraces and improved varieties that were grown in western Ethiopia, describing their traits. Participants of the focus group discussion (FGD) identified the desirable traits of widely grown local varieties. In the East Wellega zone, farmers grew some improved sorghum varieties, namely Lalo, Chemeda and Gemedi, which had been released by BARC. The variety Lalo is a brown seed type that is a high yielder and is resistant to bird attack. However, farmers do not prefer this variety due to its poor thresh-ability, poor food quality and low market demand. The recently released varieties Chemeda and Gemedi have high grain quality that is desirable for home consumption and has a greater market demand. However, these varieties are attacked by birds, and being sweet stem types, they are vulnerable to high levels of human theft. The varied needs and preferences of sorghum farmers in the region confirmed the critical need for farmers’ involvement in the development of novel sorghum varieties if they are to meet requirements of the farmers.

Table 11. Sorghum landraces and improved varieties grown by farmers in western Ethiopia with their distinguishing traits.

Landraces/variety	Preferred traits	Drawbacks
Bobe Dima	Resistant to anthracnose and other major diseases and not attacked by birds and tall stalk	Late maturing, brown/red grain color not preferable for home consumption
Bobe Adi	Early maturity and preferred for home consumption mixed well with tef flour	Susceptible to anthracnose and other major diseases, susceptible to bird and weevil and need fertile soil
Gababe	Early maturing and highly preferred for home consumption	Susceptible to anthracnose and other diseases
Lalo (released variety)	Resistant to bird attack	Difficult to thresh, unsuitable for home consumption and lower market demand
Chemeda and Gemedi (released varieties)	Resistant to anthracnose and other major diseases, creamy to light yellow grain color, good food quality and preferred for home consumption	Susceptible to bird attack

The current study established that sorghum was widely cultivated in western Ethiopia, as the third cereal crop, following maize and tef. However, sorghum production and productivity of this region are severely affected by anthracnose disease, which can cause yield losses of 70% in the presently grown varieties. Further, novel sorghum varieties released by BARC were not widely adopted in the study areas due to one or more critical limitations in their performance such as susceptibility to bird attack, lack of suitability for home consumption and difficulty to thresh. Therefore, breeding novel sorghum varieties with all the critical farmer-preferred traits is essential for the adoption of improved sorghum cultivars by farmers in western Ethiopia.

Acknowledgements

The Bako Agricultural Research Center is acknowledged for supporting this study, and the Oromia Agricultural Research Institute for providing study leave for the first author.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Girma Mengistu is a Ph.D. Student at the University of KwaZulu-Natal (Ph.D. – Plant Breeding) and Plant Breeder, Oromia Agricultural Research Institute, OARI, Addis Ababa, Ethiopia.

Hussein Shimelis is Professor of Plant Breeding at the University of KwaZulu-Natal, African Centre for Crop Improvement, and is the Deputy Director of African Centre for Crop Improvement.

Mark Laing is Professor of Plant Pathology at the University of KwaZulu-Natal, African Centre for Crop Improvement, and is the Director of African Centre for Crop Improvement.

Dagnachew Lule (Ph.D.) is a senior Plant Breeder, Oromia Agricultural Research Institute, OARI, Addis Ababa, Ethiopia.

ORCID

Girma Mengistu http://orcid.org/0000-0001-9569-4751

Additional information

Funding

The authors thank the National Research Foundation (NRF)/South Africa for financial support.