The study was conducted in Geshy sub-catchment of Gojeb river catchment, Omo-Gibe basin, Ethiopia which covers a total area of 9628.5 ha. Geographically, it lies between 07˚22' - 7˚26'N latitude and 36˚12' - 36˚24'E longitude with altitude ranges from 1600 to 1800 m.a.s.l. (Figure 1).

Agro-ecologically, it falls in the wet/moist (woina dega) regime and is found in warm sub-moist lowlands and tepid sub-moist mid highlands and tepid humid to sub-humid mid highlands, and warm sub-humid lowlands [27] .

The rainfall is uni-modal with low rainfall from November to February and

the wettest months between May and September. The mean annual and monthly rainfall of 1762 mm and 136.3 mm and monthly mean maximum and minimum temperature of 25.3˚C and 15˚C (Figure 2), respectively, was recorded at Dirri meteorological station. In the study area, five major soil types, namely: Fluvisols, Cambisols, Vertisols, Regosols and Leptosols have been identified [27] . The land use pattern is characterized by extensive cropland and mainly dominated by cropland, forestland, shrublands, woodland, and swamp area [20] . The total population in Geshy sub-catchment is 14518 from which 7261 are men with total number of households 3060 from which 2793 men [27] .

¹The lowest administrative level within a district or Wereda.

Table 1 and Table 2 present basic household and socio-demographic characteristics of the sample households in Geshy sub-catchment. The overall mean family size, 5.88 persons per household was above the average family size of 4.9

persons per household [30] . Similar to all rural area of Ethiopia, land is the most important livelihood asset for the surveyed households. The landholding size of the respondents ranged from 0.25 to 25 ha with an average of about 3.37 ha, which is by far larger than the 2007 national average of 0.85 ha [30] . The majority of the respondents (44.2%) however possessed 1.0 to 3.0 ha followed by less than 1 ha (24.7%) and 3.1 to 7 ha (24.7% and only 1.3% possessed more than 10 ha of land). Despite the landholding size is by far above the national average, most of them (about 67.5%), acknowledged that their holdings were increasingly inadequate to support their households because of rural population growth, soil erosion, and declining land productivity. The respondents have used to live for an average of about 35 years, of which the majority of them (50.6%) were lived more than 30 years which made them to have basic knowledge about the environmental problems in the area.

In the study area, the majority of respondents (39% of total) were illiterate with only 6.5% of them having secondary school education (9 - 12 grades). Both age and literacy levels have implications for the respondents’ perception and adoption of soil erosion and conservation. Furthermore, unlike the majority (67.5%) of farmers, some 32.5% of them were involved in local administrations. This is critically important because the involvement affords farmers a measure of social status as well as access to critical land management and farming information. Moreover, 70.1% of respondents’ livelihood entirely depends on on-farm activities and their major farming system is crop-livestock mixed (80.5%), while 18.2% of the households used to practice crop only and very few of them (1.3%) used to herd livestock only. This might have an implication on the dependency of farmers on natural resources for their agricultural expansion.

In Geshy sub-catchment, farmers’ perception of soil erosion was presented in Table 3. A farmer’s awareness of soil erosion is consequently a critical parameter for their willingness to adopt. The majority of the respondents (79.2%) perceived soil erosion as a problem constraining their crop production and lead serious land degradation, while the rest 20.8% didn’t perceived it as a problem, this might be due to its gradual and unseen nature of the progress. The results obtained from the individual farmers and focus group discussion revealed that the main causes of soil erosion were deforestation due to agricultural expansion, heavy and long rainy season, steepness of their cropland and runoff coming from the upslope area. Similarly, soil erosion still remains a significant problem in large parts of Ethiopia due to the forecasted increase in population and extreme precipitation events [16] . The major consequences of soil erosion understood by the respondents are loss of soil fertility (67.5%), the decline in land productivity (22.5%), followed by creation of rill and gullies (7.5%) on their farm land. They also reported that the most important topsoil for crop production

was deteriorating and reducing the depth of the topsoil over time due to erosion. Other studies indicated that soil erosion problem has been widely recognized by farmers [9] [14] [31] .

In the study area, the respondents were able to rate the extent of soil erosion problem as severe (29.9%), minor (27.3%), moderate (23.4%); while, 19.5% of them were not perceived its severity. The majority of them (79.2%) hoped that the trends of soil erosion rate will become less severe following the implementation of sustainable land management practices. Accordingly, 77.9% of them expected that the severity of soil erosion will get better, while 15.6% perceived the severity will get worse followed by will remain the same (2.6%) but only 3.9% don’t know. Generally, almost all of them (97.4%) believed that soil erosion can be controlled and they could control erosion on their farm plots. In line with our finding, various studies [14] [20] [24] [32] [33] reported that the majority of farmers confirmed that erosion can be controlled.

The link between soil erosion and decline in soil fertility levels appeared to be obvious to the respondents because the declining soil fertility was attributed to “over cultivation” caused by overuse and occurrence of soil erosion. Some of the interviewed farmers (14.3%) and those who participated in the focus group discussion confirmed that there have been a decline in soil fertility and land productivity in their farm plots over time. Unlike the previous time, the majority of farmers (77.9%) confirmed that they observed an increasing trend in the soil fertility status following the introduction of soil conservation practices. Field observation and group discussion also indicated that farmers undertook a range of practices for soil fertility maintenance: crop rotation; application of organic matter (animal manure, compost, household wastes, and crop residues); use of chemical fertilizers; and erosion control practices.

The reason for the increasing trend of soil fertility might be attributed to the integrated use of both organic and inorganic fertilizers. Accordingly, the majority of the respondents (61%) used both organic and inorganic fertilizer simultaneously; while 19.5% and 16.9% of them used only organic and inorganic fertilizers respectively, but only 2.6% of them were used none of these. Moreover, the majority of them (58.8%) confirmed that the fertility of their land has been improved and rated the soil fertility status of their land as medium (77.9%), followed by high (15.6%) and low (6.5%) which might indicated their awareness on the positive effects of organic fertilizers in improving soil quality and reduced runoff. However, farmers in the study area used to collect cow dung and crop residues from their land for fuel and hatching which limited the use of organic matter and resulted in the deterioration of biological processes in the soil. In the study area, among household characteristics, the most significant factors influencing farmers’ perception of soil erosion were farming system (χ² = 13.845, P = 0.001), and income sources (χ² = 5.379, P = 0.020); but sex (χ² = 3.375, P = 0.066), educational level, relative position (χ² = 5.398, P = 0.067) and their involvement in local administration (χ² = 3.673, P = 0.055) didn’t influence them. While, perceived trend of soil erosion rate was significantly affected by farmers’ relative positions in the catchment (χ² =11.648, P = 0.020), and land holding adequacy (χ² = 25.213, P = 0.000) of respondents. Moreover, the severity of soil erosion was significantly influenced by their farming system (χ² = 21.642, P = 0.001), income source (χ² = 12.901, P = 0.005), relative position (χ² = 15.763, P = 0.015). Similarly, their expectations in soil erosion severity changes in the future were significantly influenced by farming system (χ² = 23.895, P = 0.001), income sources (χ² = 8.656, P = 0.034), landholding size adequacy (χ² = 12.582, P = 0.050) and relative position (χ² = 19.674, P = 0.003). This finding concurs with similar studies across different parts of Ethiopia [9] [24] [34] findings that farmers have a holistic understanding of the attendant soil erosion and their management methods as integral to soil degradation.

In Geshy sub-catchment, farmers have been used to practice both indigenous and introduced SWC measures to conserve and maintain their farm land. The physical measures, mainly soil bund and Fanya juu, have been practiced by integrating them with multipurpose biological measures such as vetiver grass (Vetiverial zizanioides), Desho grass (Pennisetum pedicelluatum), and elephant grass (Pennisetum purpureum). Besides the stabilization of soil bund, they also used as thatching house, mulch material and forage for animals. Constructions of the physical SWC measures are costly and highly labour-intensive task to accomplish only by the effort of subsistence farmers. During field work and focus group discussion, we observed that farmers were involved in physical works in mass, while development agents were working as “facilitators” and wereda (district) agricultural officers were working as superintendent. It is realized that farmers were not allowed to be absent; otherwise they would be subjected to serious punishment. The conservation effort is truly meritorious, but only if it is really effective and can be sustainable. To be effective, it has to be carefully engineered; and to be sustainable, committed participation of the stakeholders―the farmers are crucial. Community participation is one of the factors assumed in this study as influencing the degree of adoption of soil conservation measures by farmers. If one needs to achieve success in conservation activities, farmers’ active participation should be ensured from the beginning [25] [35] . Hence, farmers’ willingness to participate in SWC activities should be examined in their communities.

The descriptive statistics on farmers’ participation in SWC activities in Geshy sub-catchment was shown in Table 4. The majority of respondents (92.2% of the total) were participated in the SWC activities and only 7.8% of them were not participated. Accordingly, most of the respondents (87%) were participated voluntarily, while the remainder, 6.5 percent of them participated simply because they were forced to do so by the kebele administration and development

agents. But, the rest 6.5 percent were not participated in any of the activities. The implementation of SLM program might play a vital role in motivating the community in participation of SWC activities. Since the majority of respondents were participated voluntarily, it becomes clear that SWC works are taking into account participatory approach as it is one of the basic principles of community based participatory watershed management. Similar study in the Northern part of Ethiopia [35] indicated that the newly introduced SWC measures have got good perception, acceptance and widely practiced by farmers. In contrary to our findings, studies [14] [33] revealed that only 35 - 40 percent of the interviewed farmers participated in the SWC works voluntarily. On the contrary to our finding, they concluded that where the majority felt coerced to participate which did not take into account participatory principles. In the study area, our findings revealed that among basic household socio-economic characteristics, only sex, educational level and period of residence have significantly influenced both farmers’ level of participation and how they were participating in SWC activities. Since, the numbers of females headed respondents were very much less than the male respondents, the chi square test indicated a significant association between respondents’ sex and their participation as they are dependent to each other. The majority of male headed household are willing to participate in SWC activities voluntarily than female headed households. The Chi-square test result showed a significant relationship between farmers’ educational level and level of participation (χ² = 9.267, P = 0.050) as well as how they were participating in SWC activities (χ² = 19.001, P = 0.015). Those farmers who have been used to live for long time were found to be significantly higher level of participation (χ² = 7.991, P = 0.046) and more willing to participate voluntarily in SWC activities (χ² = 12.270, P = 0.056) that might be attributed to continued awareness. Similar findings also reported by [36] [37] . In the contrary, other scholars revealed that age has a profound effect on farmers’ participation in SWC activities [31] [38] .

The descriptive statistics of farmers’ knowledge and adoption of SWC technologies was shown in Table 5. In the study sub-catchment, most of the respondents

(94.8%) had been getting extension service by developments agents, though the frequency of visiting varied with individual farmers. The acceptance of adoption is the decision to implement SWC measures or not, first of all farmers should know and aware of the relevance of introduced SWC technologies. Farmers’ decisions to conserve soil and water resources are largely determined by their knowledge of the erosion problems and perceived benefits of conservation. The success of any SWC intervention depends on the extent to which the introduced conservation measures are accepted and adopted by the farming community [24] [25] . In other words, acceptance and farm-level adoption of the newly introduced conservation measures by the farmers is the decisive element for the success of a watershed management intervention [25] [39] . In this study context, acceptance refers to the farmers’ evaluation of the introduced SWC technologies in terms of their effectiveness in arresting soil erosion and their potential to improve land productivity; while adoption refers to the farmers’ expression of commitment for a sustained utilization of the technologies as part of the local agricultural system after the external assistance is withdrawn.

In the study area, almost all of the respondents (97.4%) knew about the presence of introduced SWC measures such as soil/stone bund, Fanya juu (Table 5). Surprisingly, unlike some (31.2%) of the respondents, the majority of them (68.8%) were aware of the introduced SWC measures before the implementation of SLM program. They also confirmed that introduced SWC technologies (soil bund (46.8%), Fanya juu (23.4%) and afforestation (22.1%)) are the most efficient technologies in arresting soil erosion. Moreover, farmers were also asked to compare the problem of soil erosion in their farm plots after conservation structures were built; accordingly, the majority of them (93.5%) confirmed that soil erosion rate had decreased after the implementation of SLM program. Besides, almost all of them (94.8%) believed that the new SWC technologies had the potential to improve land productivity and lead to increased crop yields. Similar findings elsewhere [24] [25] [32] [33] reported that the introduced SWC technologies had the potential to improve land productivity and control erosion.

Farmers who had been tried to implement conservation measures in their plots were interviewed to determine how they had measured the effectiveness of SWC technologies. Accordingly, they had already observed better growth and development of crops, particularly along the structures where fertile sediments were trapped. They also evaluated that the amount of sediment trapped by the structure was very high and would have been lost from the field had that conservation structure not been built. Furthermore, during group discussion, participants who treated their lands by conservation structures gave witness to the group that the technology they have been using improved their land productivity and obtained more yield than before. Therefore, farmers were able to differentiate indicators of “acceptance” such as knowledge about introduced SWC technologies, its effectiveness in arresting soil erosion as well as its potential to improve land productivity. On the other hand, acceptance of the technologies as effective measures against soil erosion and as having potential to improve land productivity cannot warrant its adoption at the farm level. While acceptance depends more on the design characteristics of technologies as related specifically to its effectiveness in arresting soil loss, farm level adoption of the technologies depends also on several socioeconomic and institutional factors [25] .

Adoption of SWC technologies is a difficult concept to measure. The factors affecting adoption determine the sustainable utilization of the measures by the farmers [24] [40] . Hence, the newly introduced technologies can be considered as adopted if the land users (farmers) continue to utilize them as a part of their production system after the external assistance is withdrawn. In the study area, almost all the respondents (98.7% of the total) showed willingness to adopt SWC measures. However, despite their interest, farmers were asked about the status (degree) of use of improved SWC structures. Accordingly, they responded that partially removed (41.6%), while 33.8% of them replied that the structures were modified (adopted); whereas, only few (7.8%) of them were totally removed. Results obtained from observation and group discussion showed that construction of introduced SWC measures were time-consuming, high labour-demanding and need frequent technical support for design construction. In the study area, almost all of the respondents (97.4%) believe that construction and maintenance of SWC is their responsibility. They also added that family labour (66.2%), community participation (29.9%), labour exchange (2.6%) and financial incentives (1.3%) were the most important means to construct and maintain the SWC structures. Moreover, almost all respondents (94.8%) expressed their willingness to continue maintaining the established structure and most of their adoption judgment of SWC measures was very good (57.1%) and good (41.6%) (Table 5).

In the study area, the sex of respondents was the potential factors that significantly influence the farmers’ knowledge about the presence of SWC structures (χ² = 24.298, P = 0.000); provision of extension service (χ²= 10.461, P = 0.001) as well as their frequency of being visited by development agents (χ²= 12.713, P = 0.013). In addition, farmers’ period of residence in the area is significantly influence the provision of extension services (χ² = 10.619, P = 0.014). However, age, family size, educational level, landholding size, land size adequacy, period of residency, the involvement in local administration, farming system, sources of income, relative position in the sub-catchment of house head was not significantly influence about farmers’ knowledge about the presence of SWC technologies. Moreover, our findings revealed that the farmers’ knowledge on the introduced SWC technologies was significantly influenced by their age (χ² = 9.686, P = 0.046), farming systems (χ² = 8.017, P = 0.018), household income sources (χ² = 5.022, P = 0.025), and relative position in the sub-catchment (χ² = 18.856, P = 0.000).

In the study area, though not well organized and implemented improperly, farmers have been applied various types of SWC technologies to arrest soil erosion. The most common practices recognized by farmers are soil/stone bund, Fanya juu, cut-off drain, waterways, and afforestation. Soil bund is the most common and dominant SWC practices applied by farmers which is followed by Fanya juu. The potential of the newly introduced SWC technologies to improve land productivity and respondents’ intention to maintain the constructed structures was significantly affected by household income sources (χ² = 4.102, P = 0.043), relative position (χ² = 8.776, P = 0.012). The chi square test result also revealed that farmers’ perception on their responsibility of constructing and maintaining SWC technologies was found to be significantly influenced by sex of respondent (χ² = 5.091, P = 0.024), household income sources (χ² = 4.821, P = 0.028). Besides, our result also showed that the responsible body to construct and maintain the structures was significantly depends on sex (χ² = 17.882, P = 0.000), landholding size (χ² = 23.734, P = 0.022), farming system (χ² = 17.281, P = 0.008), income sources (χ² = 17.281, P = 0.001), relative position in the watershed (χ² = 33.160, P = 0.000). In the study sub-catchment, the status (degree) of use of improved SWC structures was assessed. Accordingly, the respondents’ educational level (χ² = 23.663, P = 0.023), involvement in local administration (χ² = 31.436, P = 0.000), farming systems (χ² = 21.584, P = 0.001), sources of income (χ² = 9.357, P = 0.025) and relative position (χ² = 64.406, P = 0.000) as the farmers located in the downstream were found to be the best adopters of the newly introduced SWC technologies than the upper and the midstream ones. When farmers compared the problem of soil erosion in their farm land after conservation structures were built, it was found to be significantly influenced by landholding size adequacy (χ² = 26.143, P = 0.000). Moreover, the chi-square test result also indicated that respondents’ adoption judgment of SWC measures was significantly influenced by sex (χ² = 12.590, P = 0.002). Similar findings elsewhere [24] [32] also confirmed that adoption of conservation practices alone may not lead to sustained land rehabilitation unless the technologies are utilized continuously.