Subsistence agriculture is the main source of livelihood for more than 85% of Ethiopia’s population [1
]. However, the country faces challenging problems in its struggle to make agriculture sustainable and to achieve food security [2
]. In Ethiopia, land degradation in the form of soil erosion and nutrient depletion seriously threatens agricultural productivity and is a major cause of food insecurity [4
]. The problem is persistent in the highlands of Ethiopia, where the majority of the country’s population lives and depends on farming. Land degradation in the highlands of Ethiopia is primarily caused by intensive cultivation on steep and fragile farmlands with unsustainable land management practices [3
]. Furthermore, soil characteristics, topography and the cropping pattern (dominated by cereals) make the Ethiopian highlands vulnerable to soil erosion [7
]. In this regard, it has been estimated that 42 ton/ha [2
] to 179 ton/ha [9
] of soil is eroded from cultivated land every year. Hence, we can infer that the fate of Ethiopian smallholder agriculture relies on the quest for Sustainable Land Management (SLM).
In Ethiopia, SLM is an important issue and receives emphasis in the country’s development agenda, which aims to reverse land degradation, improve agricultural productivity and achieve food security through implementing soil and water conservation practices at a large scale [10
]. In this regard, many development projects and programs have been initiated and implemented by successive Ethiopian governments in collaboration with several consortia of donors since the 1970s [8
]. Between 1995 and 2009, the Ethiopian government incorporated SLM practices into agricultural extension packages/programs for individual farm-households [11
]. Recently, SLM practices have been promoted and implemented through community mass mobilization at a watershed level, as part of Ethiopia’s Growth and Transformation Plans (GTP I and II) [12
]. GTPs are a national development framework for five year periods: GTP I (2010/11 to 2014/15) was directed towards achieving the Millennium Development Goals by 2015 [12
], and GTP II (2015/16 to 2019/20) was directed towards achieving the country’s vision of becoming a middle income country by 2025 [13
]. The main SLM practices implemented through (community) mass mobilization include physical measures, such as stone/soil bunds, terraces and check-dams, as well as biological measures, such as tree planting and area enclosures [14
Despite considerable efforts made to promote SLM through different intervention strategies, limited adoption of SLM practices by local farmers is reported in many studies conducted, for instance, in the highlands of Ethiopia [2
]. At the same time, some farmers adopt SLM practices spontaneously, on their own initiative. Such farmers often adapt and implement these practices to make them fit to their farming system and limited available resources, and integrate them with other measures by using their own knowledge and family labor [18
]. However, there is limited research done to better understand which practices spread spontaneously and how these are adapted to fit the farming system. Having more insight into such spontaneous spreading would help to improve current SLM scaling-up strategies and better enable the inclusion of farmers’ knowledge and practices into a technology spreading strategy [21
This study provides insights into how stone bunds have spontaneously spread in the Girar Jarso woreda, the central highlands of Ethiopia, as such contributing to the understanding of the process of spontaneous spreading. Stone bunds are chosen because these are widely promoted and implemented on farmlands in the study area. The study further tests the hypothesis that spontaneously implemented stone bunds, as compared to stone bunds implemented by mass mobilization campaigns, are better integrated with other land management measures, and therefore lead to higher yields.
The aim of this study was to understand spontaneous spreading of stone bunds. Summarizing the results of this study, Table 2
shows that there are numerous significant differences between stone bunds implemented by SFs and NSFs. When looking at the farmland characteristics, the statistical analysis reveals that significant mean differences (p
< 0.01) exist concerning perceived soil fertility of the farmland, erosion levels and the locations of farmlands where stone bunds have been implemented. These three characteristics together are decisive in explaining where stone bunds are implemented by SFs. An explanation for this result is that farmers, who use farmlands in areas that are more vulnerable to soil erosion, perceive erosion problems and loss of soil more easily [11
]. Particularly when visible signs (rills and gullies) appear on the farmlands, farmers perceive soil erosion as severe [19
] and hence decide to use erosion control measures [39
]. Moreover, farmlands closer to the homesteads can be better monitored and taken care of, enabling more frequent supervision of the implemented conservation measures [2
]. Consistently, Pender and Gebremedhin [58
], Abebe and Sewnet [59
] and Cholo, et al. [60
] reported that farmers give more attention to farm plots closer to homestead areas than distant farm plots. This also enables farmers (farm owners) and/or other neighboring farmers to observe the effects of implemented stone bunds, suggesting that visible effects are crucial in terms of triggering spontaneous adoption and implementation of technologies.
Concerning the characteristics of the stone bunds, Table 2
shows that stone bunds implemented by SFs were significantly better maintained than those implemented by NSFs. Observations during fieldwork confirmed this: stone bunds implemented by NSFs were poorly maintained and some were even destroyed and broken during ploughing. These poorly maintained and damaged stone bunds can even be the cause of additional erosion problems [19
]. However, the work of mass mobilization campaigns focuses more on constructing new structures rather than paying attention to maintaining the old or previously constructed structures [15
]. Modifying stone bunds to make them fit to local conditions is another important variable included to understand the process of spontaneous spreading. Our analysis showed that spontaneously implemented stone bunds were significantly more often modified during construction to fit to the local conditions of the farming system than those implemented by the mass mobilization campaigns. Several characteristics of these stone bunds were modified, including the recommended spacing between bunds as well as the height and length of the stone bunds to fit the needs of the farming system. This was also observed in a study conducted in Hunde-Lafto area [2
] and the Beressa watershed [19
], where farmers modified the original design of introduced technologies to fit the local conditions. Our findings support the diffusion of innovation literature [29
], which argues that technologies modified or changed by farmers during implementation spread better than other technologies, hence facilitating the spontaneous spreading of stone bunds.
Another important finding concerns the integration of stone bunds with other measures on the same field, as perceived by farmers. Integration means that soil management measures that are intended to conserve the soil, as well as to improve crop yields (such as fertilizer, compost and manure), are applied on the farmlands together with stone bunds. It was hypothesized that stone bunds implemented by SFs were better integrated with these soil fertility management measures and lead to higher yields, as compared to stone bunds implemented by NSFs. The analysis confirmed the stated hypothesis. With spontaneously implemented stone bunds, significantly more fertilizer, compost and manure are used in integration with the stone bunds. An explanation for this is the proximity of the stone bund plots implemented by SFs to their homesteads, because farmers are more likely to apply fertilizer, compost and manure on farmlands close to the homesteads [58
]. Another explanation is that SFs are more dedicated and aware about the importance of integrating stone bunds with these soil fertility practices.
Moreover, the better integration of the stone bunds with fertilizer, compost and manure has its beneficial effects on erosion control, soil moisture, soil productivity and crop yield. The statistical analysis shows that significant differences between the perceived effect of stone bunds implemented by SFs and NSFs exist for all these factors (p
< 0.01). Results showed that a large proportion of SFs perceive that erosion on the farmland has decreased, and that both soil productivity and crop yield have increased after stone bund implementation, and hence confirmed the hypothesis that stone bunds implemented by SFs lead to higher yields. These results may be explained by the fact that spontaneously implemented stone bunds are better maintained and integrated with more compost and manure. In line with this, Kessler [57
] suggests that manure use together with erosion control measures, such as stone bunds, is a promising alternative for more productive farmlands and agricultural production.
A study conducted by Hurni [8
] describes that erosion adversely affects crop productivity by reducing availability of water to crop growth and soil nutrients. As a result, when soil erosion control measures, such as stone bunds, are combined with the use of compost and manure, they enhance soil moisture and improve availability of soil nutrients [58
]. This is consistent with observations in this study, where a high number of SFs perceived that water availability on the farmland has increased after stone bunds were implemented. In addition, this will also result in higher crop yields, which, in this study, was significantly higher for SFs than for NSFs. This implies that the importance of ‘relative advantage’ [36
] holds true in our research, suggesting that farmers’ perceived relative advantage of stone bunds determines the process of spontaneous spreading.
Most interestingly, the study found that many farmers (particularly SFs, but not significantly more than NSFs) believe that yield improvement was not due to the stone bunds only, but rather to the integration of these with soil fertility management measures. The analysis indicates that NSFs are also aware of the importance of integrating practices, but they perceive that they use less fertilizer, compost and manure. Because of this, NSFs do not experience higher crop yields. This finding is in line with a study conducted by Posthumus, et al. [62
] in Burkina Faso, reporting that conservation measures are more profitable when integrated with soil fertility measures. Therefore, technologies that enhance profitability are crucial in stimulating farmer’s adoption decision [4
], and hence are important to understand spontaneous spreading. In general, because the effect on yield is related to application of more fertilizer, compost and manure and they are constructed on erosion prone farmlands where the effect of stone bunds is more visible, spontaneously implemented stone bunds result in more short-term benefits than stone bunds implemented by the mass mobilization campaign. Therefore, in order to convince NSFs of becoming SFs, the mass mobilization campaign must focus on constructing stone bunds, where these are most needed, resulting in quick wins.
However, the study also found that the mass mobilization campaign is an important source of knowledge for farmers, and often motivates farmers to spontaneously implement stone bunds on their own farmlands [14
]. About 60% of the farmers acknowledge to have learned from the mass mobilization campaign and other projects (53%). Nevertheless, most knowledge about stone bunds comes from neighboring farmers (93%) and practical training (80%). This implies that spontaneous spreading of stone bunds is particularly enhanced by farmer-to-farmer exchanges in the community, by education and by training. This was also observed in a study conducted in Keita valley [42
], where farmers are inspired to adopt soil and water conservation practices by observing their neighboring practices and sharing their knowledge about the benefits of adoption. Our result is in line with the theory of diffusion of innovation discussed above, where Rogers [29
] argues that sources of information are important in learning and implementing an introduced technology. Therefore, this is an important requirement for the process of spontaneous spreading to take place.
6. Conclusions and Recommendation
The central aim of this study was to provide insights into how stone bunds have spontaneously spread in the central highlands of Ethiopia, thereby contributing to the understanding of the process of spontaneous spreading. A first conclusion drawn from this study is that stone bunds are spontaneously implemented where they are most needed: mainly on farmlands where farmers perceive severe erosion, poor soil fertility, steep slope gradient and located nearby the homestead area. This finding has important implications for the current mass mobilization campaigns, where farmers walk long distances to construct stone bunds on the selected farmlands. Essentially, long walking distances may discourage farmers to integrate stone bunds with soil fertility management measures, which is important to contribute to yield productivity and achieve household food security. Therefore, during the mass mobilization campaigns, stone bunds should be constructed based on farmers’ opinion of where they are most needed and where the short-term (visible) effects can be achieved.
The result of this study further indicated that spontaneously implemented stone bunds affect soil erosion, soil moisture and soil productivity, and that they lead to higher yields compared to stone bunds implemented through mass mobilization. These findings provide important insights to policy makers and extension workers on how to control erosion problems and improve soil fertility, simultaneously. Putting more stone bunds on farmland is hardly useful unless they are integrated with soil fertility management measures, such as compost, manure and improved tillage practices to contribute to improve yields. As a final conclusion, we recommend that the mass mobilization campaign should use a more participatory and integrated approach, in which there is ample space for awareness raising and learning concerning the benefits of integrated farm management, and in which farmers themselves have a more leading role in the decision on where to construct stone bunds. Such a strategy will lead to more sustainable impact on soil fertility and food security than the current top-down intervention approach.
In general, this article addressed the process of spontaneous spreading of SLM in the central highlands of Ethiopia. The study found that spontaneously implemented stone bunds had different characteristics than the conventional way of spreading technologies, through mass mobilization campaigns. However, there is still an important issue to be addressed with respect to spontaneous spreading, namely to what extent SFs and NSFs are different. The next research challenge is to compare the characteristics and motivations of farmers who spontaneously implement stone bunds with those of farmers who do not perform spontaneous conduct on stone bunds.