On-farm diversity of faba bean (Vicia faba L.) farmers’ varieties in Eastern Hararghe Zone, Ethiopia

Ethiopia is considered as the secondary centre of faba bean diversity. However, the extent of its diversity at present time is not well known in the eastern Hararghe Zone. Therefore, survey study was conducted from December 2018 to March 2019 to identify patterns of on-farm diversity of faba bean farmers’ varieties and their use; and to assess production attributes, constraints, and role of gender in the production and management of faba bean farmers’ varieties in eastern Hararghe Zone. Two stratified agro-ecological zones (Tepid moist mid-highland, M3 and Tepid sub-humid mid-highland, SH3) were selected from the zone. Three Kebeles from each agroecologies were randomly selected. From each Kebele, 12 general informants and two key informants based on their gender and wealth status were selected, making a total of 72 general and 12 key informants, a total of 84 informants. Structured and semi-structured questionnaires were used for the general and key informants, respectively. Data were collected on use of faba bean, cropping practices, diversity, attributes, and gender roles in production and postharvest handling. The data were analyzed using descriptive and inferential statistics using R (version 3.5.2) software. Eight farmers’ varieties of faba bean were identified. Highest varietal diversity (H′ = 1.35) was recorded at Gara Abdula kebele of M3 while the lowest diversity value (H′ = 0.81) was at Obi Kutir 1 kebele of SH3. Variety Safisa was reported for its highest market price (32.8 ETB kg–1) and variety Dabale was the highest yielder (1900 kg ha–1). Most farmers (94%) use traditional seeds and 72% of farmers grow faba bean on < 0.125 ha land area. Baqela Faranji was the widely (33%) cultivated variety. Diseases (100%), weeds (89%) and land shortage (85%) were the main faba bean production constraints. The male adult and female adult family members play the greatest roles in the production and post-harvest management practices of the crop. It is concluded that farmers grow eight faba bean varieties with varied yield potential and market preference by consumers and the majority of them use informal seeds acquired from family members. The results also revealed that the vast majority of the farmers cultivate the crop on small average landholding of less than 0.2 hectare. The results imply that faba bean production and diversity is constrained by small landholding and lack of quality seeds of high yielding improved varieties.


Introduction
Faba bean (Vicia faba L.) is one of the earliest domesticated food legumes in the world (Singh et al. 2013). It is the fourth most important pulse crop in cultivated area and production in the world after common bean (Phaseolus vulgaris L.), dry peas (Pisum sativum ssp. arvense) and chickpeas (Cicer arietinum L.) (Kumari and van Leur 2011). It is believed that Central Asia is the origin for faba bean, whereas secondary centres of diversity are postulated in Ethiopia and Afghanistan (Hajjar and Hodgkin 2007). Faba bean is among the most important pulse crop grown in Ethiopia at the elevations of 1800-3000 m above sea level and receiving average annual rainfall of 700-1100 mm. It was the first crop among the pulses grown in the country both in terms of area coverage and volume of annual production (CSA 2020). The production obtained from faba beans was 3.10% of the grain production (CSA 2015). Faba bean grows in several ecogeographical regions of the country including Arsi and Bale highlands, Central highlands of Ethiopia (South-West, West and North Showa), Tigray, North and South Wollo, North and South Gondar, East and West part of Gojam, Wollega, Guji highlands, Hadiya, Sidama and Gamogofa (Tekalign, 2014). It has an ability to grow over a wide range of climatic situations and broad adaptability to a range of soil conditions (Yahia et al. 2012). It has high nutritional value and it is used almost daily in human diet of many Ethiopians (Abebe et al. 2015).
Ethiopia considered as one of the richest genetic resource centres and crop genetic diversity conservation area in the world. However, a report by IBC (2005) showed changes in the development of agriculture or land use, displacement of indigenous landraces by new and improved varieties, genetically uniform crop cultivars, drought, and production constraints such as diseases and pests are the main causes of crop biodiversity loss in Ethiopia. More specifically, Tamiru and Abdela (2021) identified that loss of the diversity of farmer' crops varieties including faba bean is the main problem of crop production in Eastern Hararghe Zone and loss of diversity was occurred as a result of khat expansion (45%) and due to replacement of modern improved seed varieties (40%). In similar way, faba bean is among the different crops grown in this zone and in recent years, its cultivation is declining from time to time and become among the endangered farmers' varieties that the farming communities unable to meet their livelihood needs from the crop and more exposed to food insecurity and malnutrition (NCA 2014;WFP 2014). Andualem et al. (2022) also identified a total of seventeen faba bean farmers' varieties in northeastern and southwestern Ethiopia in which the tepid submoist mid-highlands of South Wollo had the highest number of varieties (five) of any area sampled. This indicates the decline in the diversity of faba bean farmers' varieties that need urgent monitoring and studying its on-farm diversity, their uses, management, and to assess production attributes, production constraints, and the role of gender in the production and management of faba bean farmers' varieties in eastern Hararghe zone.
In the recent developed intensive cropping systems, crop diversity is reduced to one or very few species that are genetically homogeneous. Sustainable management of crop genetic resources assure crops diversity, both in trust collections or gene banks (ex-situ) and on farms (in-situ) (Smale 2006;Bezabih 2008). For farmers, crop diversity is important to combat production risks found in the changing environments. Understanding farmers' preferences for crop attributes and their incentives to grow diverse varieties are critical to the success of on-farm conservation (di Falco et al. 2010). Many farmers still depend on a few varieties and mostly prefer to grow their own varieties due to its better selling price locally, environmental adaptability (resistance to drought, poor soil and frost occurrences), cooking quality, better suit with their production system, and yield stability of the variety despite occurrences of disease and pest problems (Tafere et al. 2012).
Effective on-farm management and conservation of genetic resources takes place where the genetic resources are valued and used to meet the needs of local communities. The probability that farmers choose to cultivate a certain crop variety, on the other hand, is dependent on the key attributes the farmers associate with the variety. Faba bean research in Ethiopia focused mostly on developing varieties released from research centres (Tekalign 2014) and has given less emphasis on studying its diversity and improving quality traits of preferred farmers' varieties. Therefore, it is imperative to investigate the diversity of faba bean farmers' varieties grown in the study area for further research work. Hence, the objectives of this study were to identify patterns of on-farm diversity of faba bean farmers' varieties grown, their uses, and management practices; and to assess production attributes, production constraints, and the role of gender in the production and management of faba bean farmers varieties grown in Tepid moist mid-highland and Tepid sub-humid mid-highland of eastern Hararghe Zone, Ethiopia.

Description of the study area
The study was conducted in eastern Ethiopia of Oromiya regional state, east Hararghe Zone, in Deder, Kersa, Meta, and Jarso districts. East Hararghe Zone is characterized by plateaus, rugged mountains, deep gorges and flat plains. It is located at latitude: 7°30′-9°45′ N; longitude: 41°10′-42°50′ E; altitude: 500-3040 m above sea level. It receives a rainfall of 400-2000 mm and an average minimum and maximum temperature of 10-25 °C (Tolossa and Tafesse 2008). The major crops grown in eastern Hararghe Zone include sorghum and maize are cultivated as sole and also intercropped with other crops like chat, common bean, faba bean, field pea, sweet potato, potato, etc. Vegetable crops like cabbage, onion, tomato, and fruit crops like mango, banana, and papaya are some of crops grown in the zone (CSA 2014).

Informant selection, sampling sizes and procedures
Two stratified agro-ecological zones (Tepid moist mid-highland, M3 and Tepid sub-humid mid-highland, SH3) were selected from the zone based on an agro-ecological classification developed by the Ethiopian Institute for Agricultural Research (EIAR 2011). These agro-ecological classifications were defined by temperature as a function of elevation and length of the growing season as a function of rainfall. They were described with an alphanumeric code consisting of two parts: the letters refer to the length of the growing period (based on rainfall) and the number refers to the thermal zone (based on elevation). SH3 refers to tepid sub-humid mid-highlands. 'SH' refers to 'sub-humid', or a growing period of 181-240 days. M3 refers to tepid moist mid highlands. 'M' refers to 'moist' or a growing period of 121-180 days. The '3' refers to 'tepid mid-highlands' or an area with an average annual temperature of 16-21 °C.
Potential faba bean producing districts were purposively selected from each agro-ecological zone to represent the two production systems and three sub-districts (kebeles) were randomly selected from each agroecology. The survey included 12 randomly selected farmers from each Kebele. The main criteria for selection of interviewed farmers were based on wealth status and gender of the farmers (6 lowincome and 6 middle to high-income households among which 3 women from high-income households, 3 men from high-income households, 3 women from low-income households and 3 men from lowincome households), making a total of 72 general and 12 key informants, a total of 84 informants. Structured and semi-structured questionnaires were used for the general and key informants, respectively. The purpose of the study was explained to the informants and prior informed oral consent was obtained before the study was conducted.

Major data collected
Quantitative and qualitative ethnobotanical primary data were collected on faba bean farmers' varieties from the informants. The structured surveys applied with general informants were documented using Open Data Kit (ODK), an application on the smart phone. The semi-structured interview questions employed and interviewed the key informants and transcribed into a Microsoft Excel spreadsheet. Data were collected on general information about informants, use of faba bean, cropping practices, inter-and intra-specific diversity, attributes, and gender roles in production and postharvest handling. A sample of seeds for each variety that were named by the farmers in that Kebele was collected. A list of names given by farmers for each management unit was established from this survey. The consistency of the names among farmers was verified in further investigation in a participatory way using seed characteristics and plant morphological traits. Hence samples of seed from each type were prepared and shown to different farmers across the kebele. The key informants were also asked if two or more local names refer to the same variety to avoid problems related to synonymy. During the study, market values of faba bean farmers' varieties in the local markets were done and recorded the prices of faba bean varieties in Ethiopian Birr per kg.

Methods of data analysis
Ethnobotanical data were analyzed through descriptive and inferential statistical analysis using R software of version 3.5.2. Microsoft Excel was used for its graphic visualization. A two-factor analysis of variance (ANOVA) was conducted to determine whether there were significant differences in the area planted to agro-ecological zone followed by a post hoc Tukey test to identify pair-wise significance. The diversity of faba bean varieties across the study kebeles and agroecological zones were subjected to Shannon Diversity Index (SDI) analysis as: where: pi is the proportion of the total number of varieties in the ith class and n is the number of households producing faba bean.

Faba bean farmers' variety identification
Thirty six respondents (18 male and 18 female) of faba bean growing farmers from each agroecology (M3 and SH3) participated in the interview having different age ranges. The age of respondents ranged from 18 to over 60, and the greatest number of informants were between the ages of 31 to 45 years. Farmers have developed traditional naming systems for faba bean varieties they grow in both SH3 and M3 production areas. Based on the names and descriptions given by farmers, eight faba bean farmers' varieties were identified (Table 1). These farmers' varieties varied in seed size, colour, maturity, yield potential, and other agronomic traits. The names given by the farmers are also related to those characters. For example, of the eight local names given by the farmers the names Baqela Faranji and Baqela Habasha were referred to their seed colour as they were cream and brown, respectively. Batte refers to multiple traits describing the seed shape and size as flat and very big, respectively. Dabale name was related with its high yielding capacity. Dhera refers to the crops morphology related to its growth habit as it grows tall.
Hadho is the name given to the faba bean farmers' variety that was cultivated for long period of time and transferred from generation to generation and experienced a long cultivation history in the farming society. Very small size is the characteristic feature of the name Bukuri that one key informant described as this farmers' variety has as similar taste as that of field pea. The name Safisa refers to relatively early maturing faba bean farmers' variety and two key informants said that it matures in 90 days. The specific names usually related to morphology, maturity, origin, and yield components.

Distribution of faba bean farmers' varieties in both agroecologies
Distribution of farmers' varieties across agroecology and in the whole eastern Hararghe Zone explored in terms of type of farmers' varieties cultivated and the proportion of farmers cultivating the varieties ( Table 2). The types of faba bean farmers' varieties grown by the farmers of the two agroecologies were different (Table 3). The results of the study revealed that tepid moist mid-highland (M3) agroecology cultivated one more number of farmers' faba bean variety than the farmers dwelling in tepid sub-humid mid-highland (SH3) agroecology. Thus, in SH3 five varieties were cultivated and in the M3 agroecology, six varieties were cultivated. This difference in the number of faba bean varieties cultivated in the two agroecological zone is 20%. Baqela Faranji, Baqela Habasha and Batte varieties were cultivated in both agroecologies. However, Dabale and Bukuri varieties were cultivated only in the SH3 whereas Hadho, Safisa and Dhera varieties were grown only in the M3 agroecology. The differences of the varieties in distribution across the agroecological zones were attributed to the differences in climate, adaptability, topography, soil factors, and the farmers' preferences (Chen et al. 2014). In SH3 agroecology, about 53% of the farmers grow Baqela Faranji followed by Baqela Habasha. However, a few farmers (3%) grow Bukuri variety. Dabale was grown at par with Batte (8%) by farmers of the SH3 agroecology. The majority of the farmers (44%) cultivated faba bean variety Hadho in M3 agroecology, followed by Baqela Faranji and Safisa (14%) each. The lowest numbers of farmers grow Baqela Habasha and Batte (8%). The varieties that were popularly grown by the farmers could be those giving preferred traits in addition to reasonable yield under stress conditions and provide the necessary values for the smallholder households. Asnakech (2014) reported that farmers have some specially preferred traits which may not be considered by breeders NB: AEZ = Agro-ecological zone, SH3 = tepid sub-humid mid highlands, M3 = tepid moist mid highlands   stating that 13.4% of the respondents preferred the local landraces for their good biomass, and 19.4% for their resistance to disease and very few respondents (2.4%) preferred the landraces due to the unavailability of improved faba bean varieties (Fig. 1). The proportions of farmers cultivating the documented farmers' faba bean varieties in both agroecologies in eastern Hararghe Zone cultivating different faba bean varieties were analysed (Fig. 2). The results of the analysis indicated that farmers in the Zone cultivated a total of eight farmers' faba bean varieties. Among these farmers' varieties, Baqela faranji was the most widely cultivated variety (33%), followed by Hadho (22%) by the interviewed farmers in the study area. Bukuri was the farmers' variety cultivated by only a few (1%) of the interviewed farmers. The family members of the farmers determine the variety type to be cultivated depending on their preferences. The preference for a variety and land allocation decision is dependent on farm household characteristics, their attitudes and concerns (Wale 2004).

Faba bean farmers' varieties richness and diversity
The proportion of faba bean grown was compared to the proportions of other legume crops grown in the study area for highlighting the interspecific diversity of legume crops grown (Fig. 3). Accordingly, all faba bean growing farmers in both agroecologies were also found to grow mainly fenugreek crop (75% in M3 and 63.89% in SH3) and field pea (66.67% in SH3 and 41.67% in M3) agroecologies. A few farmers (2.78%) were found to grow chick pea in SH3 and 5.56% of the farmers in M3 agroecology were found to grow common bean. The probability that farmers choose to cultivate a certain crop variety is dependent on the key attributes the farmers associate with the variety. Some key informants indicated that fenugreek seed was among the essential seed to prepare sauce or stew called Hulbeta which is commonly consumed with injera in Hararghe Zones. They also mentioned that Field pea and lentil was preferred to make very good taste of shiro wot (sauce made of its seed flour) faba bean, chick pea, and common bean also used to prepare sauce and consumed as shummoo (boiled seed) along with maize grain. The result of the study indicated that the richness and diversity of farmers' faba bean varieties for each kebele were different (Table 4). Richness is one concept of diversity that refers to the number of different kinds of individuals regardless of their frequencies (Brown and Hodgkin2015). The number of varieties (varietal richness) per household is the same (1) in all kebeles. The total number of farmers' faba bean varieties (gamma diversity) was higher (5) in Gara Abdula and Afgug, followed by Efa Jalela (4) in which all kebeles were located in M3 agroecology. However, the varietal richness is slightly lower in SH3 agroecology in which case only three varieties were recorded in each kebele. The reason for the diversity decline in SH3 agroecology was the substitution of faba bean crop production fields to chat crop areas forced the loss of diversity of faba bean in the area. According to Hailu and Aune (2003)   chat production in Habro district, is rapidly replacing cereal and coffee production fields. Species richness and evenness as well as their relative abundance (pi) were also determined using the Shannon-Wiener Diversity Index (SDI). The highest value of varietal diversity (H′ = 1.35) was recorded for Gara Abdula kebele. Similarly, the varietal diversity was high (H′ = 1.3) in Efa Jalela kebele, followed by the varietal diversity in Afgug kebele (H′ = 1.12). The lowest SDI (H′ = 0.81) was recorded from Obi Kutir 1 kebele of SH3 agroecology. Generally, the index values are higher in tepid moist mid highland (M3) and slightly lower in tepid sub-humid mid highland (SH3) areas. Such low Shannon diversity values in kebeles were found under SH3, indicating that most of the area planted to faba bean is dominated by only three similar faba bean varieties. Beside to replacing the faba bean to khat farming land which causes crops decline in diversity, the farmers also continue to grow the faba bean variety which best suit to their agroecology putting other varieties aside. Similar result was reported for field pea by Mulugeta (2017) where the lowest SDI (0.9) was recorded for SH3 agroecology compared H3 and M3 agroecologies.

Use values of faba bean farmers' varieties
The respondent farmers in eastern Hararghe zone use faba bean for different purposes (Fig. 4). They use the crop for food (100%) and the crop residue is used as fodder (100%) for their livestock feeding purposes in both selected agroecological zones of the study area. This implies that the primary purposes of faba bean production by the farmers were majorly for family consumption and feeding of their livestock. The high protein content of faba bean seed makes the crop one of the most preferred sources of food by the farming communities (Crepon et al., 2010). It can be observed that 6% of the respondents used faba bean for bee forage as the additional advantage obtained from faba bean production.

Human consumption (food) value
The respondents revealed the importance of faba bean farmers' varieties ( Fig. 5a) and ranked variety Bukuri (100%) as the extremely important variety for consumption. The key respondents explained that Bukuri variety was preferred for its taste and used to prepare the best quality wot like field pea. All of them (100%) also ranked Dabale and Dhera varieties as a very important variety. Key informants indicated that different types of foods are prepared from faba bean such as sauces (Kiki and Shiro wot), Shumo (boiled beans), Ashuqi (roasted and boiled), Fuli (split and boiled sauce mainly consumed with bread), and Asheti (raw green) are the major food types that farmers prepared from seed of faba bean. The result from key informants indicated that Baqela Habasha and Bukuri are mostly preferred to prepare Kiki and Shiro wot due to its good taste related to its inherent genetic difference compared to other faba bean farmers' varieties. The genetic characteristics of faba bean varieties has sufficient variability in carbohydrate, fiber, and protein contents and quality that can determine the taste and cooking qualities for consumption (Gasim et al., 2015). However, all varieties are equally preferred for Shumo, Ashuqi and Asheti. Asnakech et al. (2016) reported that faba bean local landraces were preferred

Fodder value
The respondent farmers highlighted the use of faba bean crop residue as fodder for livestock (Fig. 5b). About 67% of the respondents reported that Dabale variety was very valuable for use as livestock fodder, followed by Batte (50%) and Safisa (40%). In contrast, Hadho (19%) and Baqela Habasha (15%) were reported to be undesirable for using their residues as fodder. It is after the farmers separated the seeds from the straw that the faba bean plant residues were used for livestock feeding. The key informant farmers related the low use of these varieties for fodder to low biomass and low palatability of the faba bean residues to livestock due to its stem hardiness. The farmers' faba bean varieties that had larger biomass were more preferred for fodder purposes. A similar result was reported by Teklu et al. (2017) that there were apparent differences in palatability of faba bean varieties in that the straw obtained from Walki variety was consumed more than straws from Mosisa, Shallo, and Local faba bean varieties due to its difference in protein and fibre contents.

Soil fertility management value
In terms of soil fertility management, all of the respondents preferred variety Batte followed by variety Baqela Faranji (83%) as they perceived that these varieties have very positive effect on soil fertility management (Fig. 5c above). About 100% respondents responded as the variety Baqela Habasha and Bukuri have only positive effect in soil fertility management role. Faba bean is normally a legume crop capable to fix free nitrogen that increase soil fertility as the expense of the varieties ability to symbiosis with the rhizobia for nodulation. According to the report of Anteneh and Abere (2017), the effectiveness of faba bean nodule in nitrogen fixation depends on the variety of the crop and its ability to symbiosis with the rhizobia found in soil.

Importance for income
The result of the study showed (Fig. 5d) variety Batte (33%) the extremely important faba bean farmers' variety fetching high income. All the respondents (100%) equally responded Dabale, Bukuri and Safisa were the very important varieties for their income value. However, a few respondents (8%) and (6%) reported Baqela Habasha and Hadho, respectively, as not so important for their market value. The reasons for the variability of faba bean varieties in market value were the consumers' preference towards the varieties for its seed size, taste, ease of cooking, tolerant to natural calamities, straw, palatability, and seed storability (Asnakech et al., 2016).
The analysis of the relationship between the average seed yield and average market price values revealed that the market price that one kg seed of Safisa variety fetches was 32.8 ETB, which was higher than the market prices fetched by the other farmers' faba bean varieties. However, seed of one kg of Dabale variety fetched 17.33 ETB and that of Bukuri fetched 18 ETB, which was lower than the market price of the seed of the Safisa variety and amounted to only half of the former variety ( Table 5). The low price values in some varieties were related with the distance of the producers from consumers in need as the big markets where many buyers available were too far from the potential farming communities producing ample amount of faba bean yield in their local area. Important Attributes of faba bean farmers' varieties Important characteristic features of the eight farmers' faba bean varieties collected from the study area were ranked by the farmers for their selected properties (Fig. 6). Farmers were asked to rank the varieties by yield levels especially during time of drought. The ranking result revealed that not all varieties produce high yield at times of drought (Fig. 6a). However, the variety Bukuri produced a medium average yield than the yields of other varieties at the time of drought. The result of the study showed that Safisa and Baqela Habasha are the varieties that are least yielders during drought time. Faba bean is a cool season crop that requires sufficient moisture to grow and yield high. It has relatively shallow roots as a result of which the crop may suffer from moisture stress in soils that dry quickly (Asnakech et al., 2016).
Farmers identified Bukuri as a variety that was sometimes affected by frost followed by Batte (Fig. 6b). However, Safisa is rarely affected and relatively resistance to frost. The ranking result made by farmers for resistance against common diseases showed that Batte, Bukuri and Dabale varieties were sometimes affected by common diseases in the study area (Fig. 6c). Two key informants explained as Safisa and Hadho varieties were the more tolerant varieties to common diseases. This may be due to genetic variation between varieties. Comparison based on taste during consumption of the seeds by humans showed that Bukuri and Baqela Habasha were ranked the varieties having very good taste (Fig. 6d). Three key informants explained that the taste of Bukuri and Baqela Habasha was similar to the taste of field pea and favourite for preparing wot or sauce. Both varieties also resemble in possessing small seed size and the inherent genetic difference in seed nutrient content of the varieties made them to be very good in taste. According to the report of Gasim et al. (2015), the genetic characteristics of faba bean varieties determine the nutrient constituents in the seed and the taste qualities of the varieties.

Seed source of the varieties
The respondents grow faba beans of traditional which the farmers handled for long period of time and new released varieties from research centre and universities on their farm (Fig. 7a). The survey results indicated that the majority of respondents (94%) cultivate farmers' faba bean varieties that stayed on their hand for long years and only 6% of the respondents cultivate improved faba bean varieties. The farmers acknowledge for the low yield of the local faba bean varieties and still use a source of seed. A similar result was reported by Kibret et al.
(2020) that only a few farmers used improved seeds of cereals, oil crops, pulse vegetable and fruit crops, the majority of the farmers used local varieties in all farming systems in Eastern Hararghe Zone. Respondent farmers reported that they obtain the seeds of faba bean varieties from different sources such as family, market, neighbours and formal through universities and research centres (Fig. 7b). About 47% of the farmers obtained the seed of the traditional varieties from their family members like their fathers, uncles, brothers for the first time and 35% of the respondents obtained from market and 13% obtained seeds from their neighbours. Only few (3%) respondents reported as they got new improved variety from the market for the first time and 3% of the respondents mentioned that they got new variety from formal way. This was due to the reason that the farmers' access to get improved faba bean varieties from formal institutions was very limited. Similar result was reported by Asnakech (2014) that the main source of faba bean seed was farmer to farmer seed exchange as indicated by 42.9% of the respondents, while 27.2% obtained the seed from their parents, 11.6% purchased from the market, and 7.5% of the farmers from Ministry of Agriculture (MoA).

Growing experience and site of plantation
The respondents said they cultivated different faba bean varieties for varied number of years (Fig. 8). The result showed that Baqela Faranji was the faba bean variety cultivated by a greater number of respondents (46%), Baqela Habasha (38%), Bukuri (100%), Hadho (20%), and Safisa (56%) for greater than ten years in the study area. Although, there are several improved faba bean varieties released by research centres, many farmers still grow their own varieties. This may be due to its better selling price locally, lack of improved varieties, environmental adaptability (resistance to drought, poor soil and frost occurrences), cooking quality, better suit with their production system, and yield stability of the variety despite occurrences of disease and pest problems (Tafere et al., 2012).
Farmers of the study area that participated in this interview described the location of planting and the fertility status of the soil on which they planted the faba bean varieties (Fig. 9a). According to the respondents, all farmers in Efa Jalela kebele (100%) and most farmers in Burka Jalela kebele (92%) planted faba bean in their main crop fields. About 33% of the respondents planted faba bean in their home gardens in Obi kutir 1 of the SH3 agroecology and a small fraction of the interviewed farmers in Gara Abdula (6%) planted faba bean in their field margins. Farmers in Afgug kebele (83%) and Efa Jalela and Rameta (both 75%) planted faba bean on soils having low fertility status (Fig. 9b). On the other hands, in Obi Kutir 1 and Burka Jalela kebeles 58% of the respondents reported that they planted faba bean varieties on soil having medium fertility status. None of the farmers interviewed in the study area reported planting of faba bean in soils with high fertility status. That farmers often used land with lower soil fertility status to grow faba bean was attributed to their perception that faba bean itself has the capacity to improve soil fertility. This perception of the farmers may be attributed to the fact that the crop is a high fixer of atmospheric nitrogen through its association with Rhizobium bacteria. This postulation is consistent with the report of Kermah et al (2018) that grain legumes grown on poorly fertile fields add net N to the soil in poorly fertile fields.

Area of the farms covered by faba bean
The farmers in the study area grow faba bean on different-sized of land (Fig. 10). The result of the study showed that the average farm size of faba bean for the low income farmers was 0.091 ha and was 0.163 ha for the mid to high income farmers. The largest area the farmers possessed for faba bean production was 0.375 ha. The relatively wider areas of faba bean land were possessed by mid to high income farmers. The majority of the respondents (72%) possessed less than 0.125 ha of land for growing faba bean. A negligible number of farmers (1%) have greater than 0.375 ha of land ready for faba bean sowing in one cropping season. Two key informants mentioned that farmers give priority of allotting land for the cultivation of khat (Catha edulis Forsk.) compared to the cultivation of faba bean. The less area of land for faba bean planting is due to scarcity of land caused by the dense population and rugged topography of the Zone. The average land holding sizes of households were very small. Farmers in the East Hararghe Zone possessed an average farm size of 0.65 ha per household (Lemma and Wondimagegn, 2014).

Influence of agroecology and kebele on faba bean growing area
The area planted to faba bean by the respondent farmers varied between the agroecologies and kebeles ( Table 6). The faba bean growing area did not significantly (p > 0.05) different between the agroecologies in 2018 cropping season. However, the result of Tukey test indicated that the growing area was significantly (p < 0.05) varied by the across the two agroecologies during the 2017 cropping season and a higher value (0.13 ha) of growing area was recorded in SH3 agroecology. This indicates that there are good growing conditions like enough land, climate and availability of production inputs in SH3 of the 2018 cropping season. Consistent with this result, Menbere (2017) reported that there was a significant difference in cropped area of common bean due to agroecologies in which higher area was recorded in east Gojjam zone (0.45 ha) and M3 agroecology (0.54 ha). The growing area was also significantly (p < 0.001) varied among kebeles in both 2017 and 2018 cropping seasons ( Table 6). The highest faba bean area of production was recorded at Obi Kutir 1 kebele both in 2017 (0.26 ha) and 2018 (0.266 ha). Key informants in Obi Kutir 1 explained as there is relatively enough farm size in the kebele and the growers allocated higher area of land to grow faba bean. Betelhem et al. (2020) also reported significant differences in production areas in which on average, households in East Gojjam Zone planted more land to common bean (0.57 ha), followed by South Wollo (0.43 ha), Kefa (0.20 ha), and Bench Maji and Sheka (0.14 ha).

Influence of wealth status on area of faba bean production
The impact of wealth status on faba bean cropped area of farmers in eastern Hararghe Zone was analysed using t-test (Table 7). The result showed that area of faba bean the farmers possessed was not significantly (p > 0.05) affected by the wealth status of the farmers in 2017 cropping season. However, wealth group has a significant (p < 0.001) impact on farm size of faba bean in 2018 cropping season. The cropped area in 2018 by the mid to high income group was higher (0.17 ha) than the low income group (0.09 ha) which may be due to the large land area possessed by the mid to high income group and they proportionally allocate more land area to faba bean compared to low income group farmers. Similar result was reported by Betelhem et al. (2020) in that the mid-to-high income (0.39 ha) group planted significantly more common bean than low income (0.29 ha) group in 2015/16 cropping season in Bench Maji, Sheka, Kafa, east Gojjam and south Wollo Zones.

Cropping practices
In the study area of both M3 and SH3 agroecologies, faba bean was planted as an intercrop or in rotation with other crop species by all respondent farmers (Fig. 11a). Accordingly, most of the farmers (50%) in SH3 agroecology plant faba bean in intercropping with maize, and with chat and potato (33% each). In the M3 agroecology, faba bean is frequently intercropped with chat (47%), and less frequently with barley (11%). Three key informants explained that intercropping is usually practiced as crop intensification package in response to the shortage of land per household as a result of rise in population, to maximize profit and to minimize risk, moisture saving and to maintain soil fertility in the area. Similarly, Jima and Birhanu (2017) reported that all of the respondents participated in the interview practicing intercropping of different crops in west Hararghe Zone. Farmers in the SH3 agroecology rotate faba bean frequently with maize (47%) and barley (42%) and less frequently with potato (6%) as shown in Fig. 11b. In M3 agroecology also the respondents reported that they rotated faba bean mainly with wheat (43%) followed by maize (40%), garlic (36%), onions (29%), and less frequently with cabbage (10%) (Fig. 11b).
Five key informants explained that they plant faba bean in rotation in every year, every second or third year on the same field to increase yield of cereal crops involved in rotation. This is in agreement with the reports of Angus et al. (2015) that the yield of wheat after two successive legume break crops was 0.1-0.3 t ha −1 greater than after a single legume break crop.

Time of sowing
According to the response of faba bean growers that participated in the interview, planting of faba bean varieties differs in time (Fig. 12). Accordingly, the respondent farmers reported that the preferred time of sowing of faba bean farmers' varieties can be extended from early June for Baqela Faranji (4% of the respondents) up to mid July for Safisa variety (40% of the respondents) in the study area. Key informants mentioned that the time of planting of the varieties is highly dependent on type of varieties and the onset of rainfall. Similar reason was suggested by Berhanu (2017) for the planting time of field pea in Arsi and Kefa Zones.

Time of harvesting
The main harvesting times of the farmers' faba bean varieties were identified by the interviewed farmers (Fig. 13). The majority of the faba bean growing respondents (46%) reported that Baqela Faranji variety is harvested at the beginning of December and many of them (42%) also reported that it could be harvested in late November. The majority of respondents also reported that the harvesting time for the variety Baqela Habasha (54%) was in late December, for Dabale (67%) in early December, for Bukuri (100%) in mid-November, for Batte (50%) late November, for Hadho (50%) mid December, for Dhera (100%) in

Yield of faba bean farmers' varieties
The yield of faba bean farmers varieties produced in East Hararghe Zone was analyzed based on farmers yield obtained report. The farmers' faba bean varieties in the study area were different from each other in their average yields (Fig. 14). As it can be seen from the results of the two consecutive years (2018 and 2019 cropping years), the highest average yield was recorded for Dabale (1900 kg ha -1 ) followed by Baqela Faranji (1442.42 kg ha -1 ). However, according to the respondents, the lowest average yield was recorded for Safisa (890.67 kg ha -1 ). Four key informant farmers suggested that those faba bean farmers' varieties can perform better if accompanied by recommended cultural practices. The low yield of these varieties might be due to the inherent genetic variation (Sharifi, 2015) and shortage of moisture during 2018 growing seasons. In line with this result, Tafere et al. (2012) reported that grain yield of ten faba bean varieties ranged from 450 to 2490 kg ha -1 in which Selale (2500 kg ha -1 ), Wayu (2200 kg ha -1 ) and Dosha (1320 kg ha -1 ) were the top yielding while Moti (537 kg ha -1 ) and Gebelcho (450 kg ha -1 ) were the lowest yielding varieties.

Production constraints of faba bean farmers' varieties
The potential production constraints of faba bean in the study area were indicated by the informant farmers (Fig. 15). According to the result of the study, the major faba bean production constraint in the study area was the diseases (100%), followed by the weeds (89%). Four key informants reported that rust (Uromyces fabae) and chocolate spot (Botrytis fabea) were the dominant disease affecting faba bean production in the area. They indicated that cuqii or abbaa tabboo (Galinsoga parviflora L.), coqorsaa (Cynodon dactylon L.), parasitic weed (Orobanche sp) and migira saree (Digitaria sp) are among the most important weeds growing in faba bean fields. Shortage of land (85%) and insect pests (68%) are also very important constraints hindering faba bean production and the yields of the crop ultimately get less. The farmers totally consume the seeds obtained from small plot of land that was produced under the mentioned constraints and even sometimes lost the varieties and forced to buy the seeds from the market for the next plantation that leads to loss in faba bean diversity. Similar constraints were reported by Tekalign et al. (2016) in which case (87.9%) of the respondents indicated that reported faba bean chocolate spot disease was the major problem for faba bean production in Finfine Special, Arsi and North Showa Zones. Abebe and Nega et al. (2013) also reported that parasitic weeds like Orobanche crenate affect the yield of faba bean in Ethiopia.

Management practices of faba bean farmers' varieties
The respondents from both agroecologies practiced no insect pest and common diseases control measures. Five key informants addressed the reasons why they did not practice any pest and disease control measures. Some attributed the failure to practice pest and disease control measures due to lack of awareness towards the impact of pests and diseases on the yield of the crop.

Weed control
Seventy-eight percent of the respondents used hand weeding for controlling weeds. Some of the farmers used hoeing (14%) and a few respondents used slashing (8%) as weed controlling options in the study area. Some key informants indicated that some farmers used non-selective herbicides before sowing of faba bean to control weeds.

Soil fertility management
The soil fertility management practices of the farmers were analysed. The result of the study revealed that most of farmers applied compost to maintain soil fertility in all kebeles except in Gara Abdula kebele where the majority of the farmers (50%) did not apply any soil amendment practice (Fig. 16). However, all farmers (100%) in Burka Jalela applied compost and 25% of them said that they applied manure to maintain the fertility of the soil. The key informants reported that none of the farmers used mineral or chemical fertilizers such as DAP, Urea and NPS for producing faba bean. They believe that those mineral fertilizers increase vegetative growth of faba bean crops and lead to lodging. However, inorganic fertilizer application as the starter is essential to be applied for faba bean production as reported by MoARD (2008), that 46 kg ha -1 P 2 O 5 and 18 kg ha -1 N 2 is the amount of recommended inorganic fertilizer for faba bean variety Gachena.

Role of gender in production and management of faba bean
Regarding the gender roles, the respondents reported that the participation of female elders in faba bean production and management was nil in Meta district, 17% in Qersa, 25% in Deder, and 14% in Jarso districts (Fig. 17). The participation of male adults in all districts was the highest (100%) and the female adults were also highest at Qersa and Deder, but 97% at Jarso and 92% at Meta kebele. Less participation of male elders also reported in Meta (8%) district. Key informants mentioned that elders and children less participated in faba bean production as it needs more energy and endurance. Participation of all gender in the family of the respondents in faba bean production and management activities were analysed (Table 8). This study showed that family members of the farmers of both sex having different age category participate in different activities related to faba bean production and management. Two key informants mentioned that men are primarily responsible for ploughing and women are primarily responsible for food preparation. The key informants also reported that weeding is the dominant activity requiring large human power at critical time and can easily done by every age and gender groups of the family. Mashingaidze et al. (2009) reported that smallholder farmers in the tropics faces serious weed challenge than any part of the world and spend 75% of their time battling with weeds in their fields. Consistent with this report, three key informants mentioned that some farmers manage weeds with chemical herbicides which are applied before sowing the crop. Ploughing, hoeing, cleaning, cooking, marketing activities were not or with less degree practiced by male children and female children in the Zone. These activities require such high energy and stamina that the children cannot perform them.
Female adult farmers were involved in all activities in both agroecologies except in the activity of ploughing the land for faba bean production. Female elders were not involved in ploughing, hoeing and fodder collection in both agroecologies. In general, the key respondents underlined that there is no clearly visible sole responsibility of any age and gender group in majority of faba bean production and management activities. However, the most pronounced difference in gender roles of adult farmers was observed for ploughing (97% done by male adults compared to 3% done by female adults) and hoeing (97% done by male adults compared to 25% done by female adults) due to the heavy duty of the activities. Seed selection, storing and marketing, which bears important implications for the conservation of varietal diversity, was reported to be the responsibilities of both genders of adults and elders, excluding both ages of children in east Hararghe Zone.

Conclusion
Through field survey in eastern Hararghe Zone, farmers' knowledge related to faba bean on-farm diversity, production, use, and constraint management were collected and analyzed. Totally, eight faba bean farmers' varieties were identified. The diversity of faba bean farmers' varieties is higher in tepid, moist mid highlands (M3) agroecology as compared to tepid, subhumid mid highlands (SH3) agroecology. Respondents use the faba bean farmers' varieties for food (consumption), fodder, income, soil fertility management and as honeybee forage. The use value of faba bean varies depending on the faba bean variety type and agroecology. Farmers mostly grow faba bean in intercropping and rotation system that plays positive role in agroecological intensification and soil fertility maintenance. The wealth of farmers affected the area of land allocated for faba bean production where mid to high income farmers grown on a larger average land area. All gender and age groups of the famers' family participated in most of the faba bean production and postharvest activities. In general, the on-farm diversity of faba bean farmers' varieties is low at the study area. Thus, germplasm conservation, awareness rising of local farmers to properly manage the seeds, and further research to improve the yielding capacity of faba bean farmers' varieties is needed in order to maintain the diversity of faba bean.