Catena 86 (2011) 1–13

Contents lists available at ScienceDirect

Catena

journal homepage: www.elsevier.com/locate/catena

A participatory soil quality assessment in Northern Ethiopia's Mai-Negus catchment

Gebreyesus Brhane Tesfahunegn a,b,⁎,1, Lulseged Tamene c, Paul L.G. Vlek a a Center for Development Research (ZEF), University of Bonn, Walter-Flex-Str. 3, 53113 Bonn, Germany b Aksum University, Faculty of Agriculture and Rural Development, P.O. Box 287, Aksum, Ethiopia c International Center for Tropical Agriculture (CIAT), Chitedze Agricultural Research, Station, P.O. Box 158, Lilongwe, Malawi article info abstract

Article history: Local communities often have substantial knowledge related to trends in soil quality and the associated Received 6 October 2010 limiting factors. Despite this, soil quality (SQ) degradation is a critical problem in Ethiopia and there is little or Received in revised form 19 January 2011 insufficient scientific information documenting local community experience in assessing SQ. This paper Accepted 20 January 2011 presents experiences of local communities in diagnosis of SQ and assesses the contribution of local knowledge as a strategy for sustainable development decision making within the Mai-Negus catchment of northern Keywords: Ethiopia. Participatory transect-walks, group discussions and field observation which complemented by Soil quality fi Soil quality indicators household interview were used to acquire data. Farmers identi ed SQ indicators e.g., crop yield, soil depth, Soil quality categories erosion and sedimentation as their basis of categorizing the soils into high, medium and low SQ. They were Participatory survey also able to identify severely degraded areas (hotspots) and underlying causes. Significant variations Mai-Negus catchment (P≤0.05) were shown between the proportions of farmers used certain SQ indicator and those who didn't Northern Ethiopia while categorizing SQ. Local farmers involved in this study demonstrated their capability to suggest appropriate land management solutions for specific problems. This study demonstrates the benefitof involving local farmers in both problem identification and solution development so that anti-degradation technologies can easily be implemented and adopted. © 2011 Elsevier B.V. All rights reserved.

1. Introduction demonstrated that investments in rehabilitating degraded landscapes in tropical regions do payoff in economic terms (Boyd and Turton, Agriculture is the mainstay of Ethiopia, providing the major source 2000; Holden et al., 2005), the overall productivity of many areas in of employment and income. About 85% of the population living in the the country is often perceived to be so dramatically damaged by country are primarily engaged in agriculture or related activities human impact that recovery is deemed impossible (Dai Trung et al., (Federal Democratic Republic of Ethiopia Population Census Com- 2007). Regardless of this, there has been a great deal of effort to mission, FDREPCC, 2008). Thus, agriculture directly or indirectly forms address soil degradation problems in Ethiopia, though success in an important component of the livelihoods for more than 70 million reversing land degradation is minimal. Part of the reason for a lack of people in the country. However, changing environmental factors have success is that the introduced practices and technologies were not led to soil quality degradation which poses a critical risk for well suited to the conditions local farmers face and local communities agricultural productivity and food security (Bekele and Holden, were often not involved in the technology selection processes 1999; Krowntree and Fox, 2008). Soil quality is most often defined (Kebrom, 1999; Badege, 2001). as “the capacity of the soil to function” (Karlen et al., 1997, 2001), An active involvement of communities under consideration is vital although a variety of definitions exist in the current literature. for successful implementation of introduced land management Soil quality degradation is often associated with interactions practices. Participation of local communities in evaluating SQ, its among land use, soil management and local knowledge regarding determining factors and possible management options are crucial, not agricultural production with inherent soil forming and erosion factors only for the measures to be accepted and implemented, but also to (Karlen et al., 2001). Deforestation and accelerated soil erosion sustain those practices. Local knowledge also benefits our scientific causing soil quality degradation are serious problems in Ethiopia understanding of the entire land management and decision-making (Badege, 2001). Even though several impact assessment studies have processes (Sillitoe, 2000; Barrera-Bassols et al., 2008). Worldwide, traditional rural societies still encompass the majority of small farmers, and the use of conventional soil survey information ⁎ Corresponding author at: Center for Development Research (ZEF), University of frequently fails because it does not take into account or under- Bonn, Walter-Flex-Str. 3, D-53113 Bonn, Germany. Tel.: +49 15 225 262 887; fax: +49 estimates the importance of local knowledge (Osunade, 1994; 228 73 5097. E-mail address: [email protected] (G.B. Tesfahunegn). WinklerPrins, 1999; Sillitoe, 2000; WinklerPrins and Sandor, 2003; 1 Tel.: +251 34 775 3549; fax: +251 34 775 1931. Barrera-Bassols et al., 2008). Local cultures and people have

0341-8162/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.catena.2011.01.013 2 G.B. Tesfahunegn et al. / Catena 86 (2011) 1–13 substantial knowledge about soils and environments, attained by regarding SQ in the country, this study was designed to explore the experience through many generations of living close to the land. The experiences of local communities in diagnosis of SQ and to assess the environmental knowledge embedded in local communities provides a contribution of local knowledge as potential indicators for sustainable long-term perspective on land use and management (Romig et al., development decision making. The goal was to enhance our 1995; WinklerPrins, 1999; Barrios et al., 2000). The long-term understanding of both the determining factors causing land degrada- experience of local communities with natural resource use and tion at the local level and benefits of local participation in problem management, including successes and failures, can help in evaluating identification and solution prescription. If successful, the experience land use in relation to soil quality and sustainable agriculture through could also help re-design strategies, investment programs and participatory approach. projects that enhance SQ and thereby food security not only in the Participatory processes are useful for providing persons with study area but also throughout the Ethiopian highlands. different backgrounds opportunities to engage in an interactive dialogue, communicate their perspectives, and develop shared 2. Methodology abilities for discourse and reflection (Röling, 2002; Barrios and Trejo, 2003). There is also an increasing awareness and acceptance 2.1. Study area that information obtained from people at the local or “grassroots” level can both provide feedback on and enrich decisions made at even This study was conducted in Mai-Negus catchment in the Tigray the national or international level (Kasemir et al., 2003). Persons at region (12°00′–15°00′ N latitude and 36°30′–41°30′ E longitude) of the local level are usually those most affected by the issue at stake and northern Ethiopia (Fig. 1). The catchment has an area of 1240 ha with are often the greatest experts on many aspects affecting their own a rugged terrain and altitude ranging from 2060 to 2650 m above sea situation. Participation is thus for the most part, valued as a means to level. Land use is predominantly arable with teff (Eragrostis tef) being enable and enhance democracy (Mostert, 2003a,b), create empower- the major crop along with different proportions of pasture and ment, and a practical means for putting decisions into effect (Stave, scattered patches of trees, bush and shrubs. The major rock types are 2002). lava pyroclastic and meta-volcanic. Soils are dominantly leptosols on Despite the aforementioned importance, previous SQ studies using the very steep positions, cambisols on middle to steep slopes and participatory local communities are lacking in Ethiopia. Taking into vertisols on the flat areas. Soils are highly eroded in most landscape account such benefits of participatory research and information gaps positions and the overall terrain erosivity potential is high because the

260000 340000 420000 500000 580000

N 461000 462000 463000 464000 465000 (C) Mai-Negus catchment

N. western zone Eastern Western zone zone Central zone 1500000 1550000 1600000 1650000

0120.5 Southern 1450000 1450000 zone Kilometers

Mekelle 1400000 1400000(B) 1500000 1550000 1600000 1650000 Tigray N 15 TigrayTigray N 13 1560700 1561600 1562500 1563400 1564300 1560700 1561600 1562500461000 1563400 1564300 462000 463000 464000 465000 11 11 13 15 (A)Ethiopia 00.4 0.8 1.6 9 9 Kilometers

Projection: UTM 8 8 Zone: 37N Datum: Adindan Spheroid: Clarke 1880 6 6 4 4

33 36 38 41 44 47

Fig. 1. Location of Mai-Negus catchments in Tigray, Northern Ethiopia. The blue color shaded area is the reservoir in the study catchment. G.B. Tesfahunegn et al. / Catena 86 (2011) 1–13 3 slope gradient often reaches 80% or more. Surface cover is also poor across the landscape) were explained to the farmers. Once awareness with high human disturbance often facilitating soil degradation had been created, the participants in consultation with facilitators processes throughout the catchment. planned the route considering diversity of topography, soil types, land uses and catchment related degradation problems. In short, the 2.2. Research approach and sampling strategy transect was designed to cross as many land uses and soil types as possible, with at least part of it aligned perpendicular to the direction A participatory field survey which complemented by household of the primary drainage course. The route was neither a single straight head interview was carried out to collect relevant information related line nor confined to the most accessible roads or paths, as such a to SQ within the catchment. The first field visit was executed strategy could give a false impression of the area. Once the route to be throughout the catchment in June 2009 to provide a clear, overall followed was agreed upon, the walk began at a high point near the impression about the area. Local farmers were then selected randomly hydrological divide of the catchment and continued downhill toward from different wealth categories to be involved in a transect-walk. the drainage line. Finally, in addition to the participatory transect-walk, arrangements A specific checklist of issues that guided the discussions during the were made to supplement the data from field observations with group transect-walk and field observation was developed. It included: (1) and informal discussions with farmers, Bureau of Agriculture experts identifying observable SQ indicators of erosion such as rills, sheet and development agents (DAs). wash, gullies, root exposure, pedestals, rock exposure, sedimentation Categorizing farmers into different socio-economic groups was or deposition and relative severity of the erosion indicators, and done entirely by farmers themselves using local criteria such as (1) reasons for continued soil erosion processes. The frequency of rills, food security status and (2) draught oxen ownership and the number gullies and other soil erosion indicators on the surface was counted of other livestock held by their households. The goal for differentiating per 100 m2 quadrants at several points along the transect to estimate farmers into different socioeconomic groups was to include farmer the relative severity of erosion indicators on SQ; (2) observing soil knowledge associated with different levels of resources, as they may color, texture, thickness of topsoil, workability, drainage, dominant also have different views on SQ degradation. Three types of farmer landforms, soil fertility and management requirements and practices wealth group were identified in the study catchment. These were: across different soil quality soils; (3) observing crop and weed species poor, medium and rich. SQ indicators in the landscape; (4) observing general land husbandry Rich farmers were defined as those who are able to feed household practices, and their relationship to SQ degradation and (5) categoriz- members throughout the year; medium farmers were those who ing SQ conditions and identifying hotspot degraded areas using the sometimes have problems with their daily food supply; and poor indicators and finally (6) determining the spatial variability based on farmers were those with no means of getting daily food and are thus field observation by the farmers participating in the transect-walks. dependent on the sale of fuel-wood, grass and wage labor for most Geographical Positioning System (GPS) readings according to farmer months of the year. Physical assets such as the condition of the house, understanding of soil quality and hotspot degraded area were farm size and ownership of permanent trees and other crops were observed in the field transect-walk. also considered as additional criteria for categorizing farmers into During the transect-walks, the first author and development different wealth groups. Resource rich farmers account only for 13% of agents took note on indicators of SQ degradation associated with the total households in the study area. Medium farmers constituted water erosion, soil fertility, weed and crop, and management practices about 47% of the total household heads and the rest are poor. Based on based on information given by the participant farmers. Occasionally, the list of farmers assigned to each wealth category, their names were open-ended interviews were also carried out with farmers within the randomly drawn to participate in the catchment's SQ assessment catchment along and after the transect-walk. Group discussions were survey. conducted after the transect-walk focusing on the existing status of SQ and the diagnosis indicators employed in the catchment. 2.3. Participatory transect-walk and group discussion Observations during the transect-walk were presented to the household heads to discuss, review, and reach consensus about A field transect-walk was held with a group consisting of five (5) (sometimes to vote) by a designated presenter. The transect-walks farmers from each of the three wealth categories (total of 15 farmers). were carried out in the morning, whereas group discussions were The individuals were randomly selected to participate because it was held in the afternoon of the next day. impractical due to group size for all the household heads to In order to have a common understanding about the SQ indicators participate. Doing so would have been problematic not only for the (e.g., soil erosion, soil fertility, soil thickness, yield), assessment walk but also for discussion and consensus building needed to extract categories, severity of the degradation and main causes, the group accurate and representative information from all economic groups. discussion meeting was held among the 15 farmers involved in the The transect-walks and discussions were carried out by the first transect-walk and the other 52 household head farmers in the author with two development agents serving as facilitators. catchment. During this group discussion, the 15 farmers presented the The participatory transect-walks and field observations were common SQ indicators, SQ categories and general resource variability conducted in two different months. The first was in June 2009 and across the study catchment. They also described the appearance of each the second in September 2009, before planting and at vegetative of the indicators in the fields and their causes. The farmers who stages, respectively. In June it was easy to identify and differentiate SQ participated in the transect-walk analyzed all the soil quality indicators indicators such as erosion, texture, color, hard surfaces, and terrain during the group discussions to establish a final consensus list of factors. While the land was being prepared for planting, it was also observable SQ indicators and the criteria used to categorize SQ condition very easy to visually identify management practices, soil conservation into categories of high, medium or low for areas throughout the efforts and tillage effects resulting in both good and poor SQ catchment. Farmers used their experiences to decide which of the listed conditions. Similarly, in September it was easy to observe differences indicators relatively implied a more severe SQ damage than the others. in SQ across the landscape using biological indicators e.g., weeds, grasses and crop performance at the vegetative and flowering stages 2.4. Household interview of growth. Prior to conducting the transect-walk, the research goals and type Forty-two household heads were chosen at random from the of preliminary information that was to be obtained (i.e. the dominant farmers that participated in the transect-walk and group discussion soils and land use practices in relation to SQ degradation indicators meeting for the structured questionnaire interview. Household 4 G.B. Tesfahunegn et al. / Catena 86 (2011) 1–13

Table 1 Diagnostic criteria of soil quality (SQ) indicators for categorization into high, medium and low SQ categories agreed on by farmers in Mai-Negus catchment, Northern Ethiopia.

SQ indicator Diagnostic criteria

High soil quality Medium soil quality Low soil quality

Crop yielda Soils give teff (Eragrostis tef) yield more than Soils give teff (Eragrostis tef) yield 1.0 to Soils give teff (Eragrostis tef) yield less than 1.5 t ha− 1 1.5 t ha− 1 1tha− 1 Crop appearance Overall crop is dark green, large, tall, in a dense stand, even Overall crop is light green, small, thin stand, Overall crop is poor, stunted, discolored, and vigour growth, matures on time uneven growth and late to mature uneven stand, never seems to mature Weed infestation/ Low weed biomass and incidence but high in diversity but Somehow high biomass and diversity, Higher biomass due to high weed infestation incidenceb demanding least labor demanding less labor but low diversity, demand high labor Soil fertility Soil is high potential nutrient with little or no fertilizer need Soil needs some help of inputs Potential is very low, poor and needs more fertilizer inputs Soil erosion Soil is little or no erosion evident and topsoil resists to Signs of sheet and rill erosion and some Considerable top soil moved and rills, gullies erosion topsoil blows, moderate erosion formed, severe erosion Soil compaction Soil stays loose, does not pack Thin hardpan or plow layer Soil is tight and compacted, can't get into it, thick hardpan Moisture in Soils hold moisture well, and give and take water easily Soil is drought prone in dry weather Soils dries out too fast dry season Topsoil thickness Soils is deep to a root or water-restricting layer Topsoil is shallow Subsoil exposed or near surface Earthworm Soil has numerous worm holes and castings, bird behind Few worm holes and castings No casts or holes of worm activity population tillage Fertilizer Soils are responsive to some fertilizer Demanding somehow high fertilizer Need higher fertilizer rate response of soil Soil tilth/ Soil is easy to work or soil flows and falls apart Soils difficult to work or needs extra passes Plow hard or soil never works down workability Soil color Surface soil color is black, dark brown or dark gray It is brown, gray or reddish It is light, light yellow, orange white or light gray color Soil texture Texture is clay loam, loamy, loam clay Texture is too heavy or too light but presents Texture is a problem, has extremely sandy, no or little problem clayey rocky Drainage Water goes, no ponding, soils drains at good rate, water Soil drains slowly, slow to dry out, water lays Poor drainage, soil is often waterlogged or moves through soil steadily, soil not too wet, not too dry for short period of time, eventually drains oversaturated for long time, very wet ground

a Teff is the most commonly grown crop in the catchment regardless of soil quality categories. That is why farmers selected and used its grain yield to categorize soil quality. b Farmers can identify weed species that grow on productive soils, heavily eroded surfaces and heavily degraded gullies, or that indicate extreme shortage of nutrients and moisture, and the trend in declining soil quality.

interview was carried out to complement the information collected and determinants of soil degradation. Local communities identified during the transect-walk and group discussion meeting. This was many SQ indicators in the transect-walk that attempted to describe done to collect data on individual farmer experiences on a range of the SQ status based on their own diagnostic criteria (Tables 1 and 2). issues of SQ indicators that used to identify the SQ categories. The Table 1 shows that the local farmers' SQ indicators ranged from interview thus addressed information not covered during the physical (soil related indicators) to biological (yield and yield transect-walks and group discussion meeting. This was because components) while Table 2 indicates how they categorized their soil each SQ indicator indicated in the group discussion might not be in local terms without yield and yield component information. represented for every farmer when categorized the levels of SQ. Indicators related to crop yield, and erosion (e.g. soil depth color) Having such information was helpful to identify the most frequently were often used by the farmers to classify their soils into three soil used indicators by the local farmers as diagnostic criteria for the SQ quality conditions: high, medium and low. Their classification was not categories into high, medium and low. The interview was also given limited to the soils' nutrient status but also considered soil erosion, the chance to explore the status of the fields farmers possessed with fertility, color, thickness, water-holding capacity, yield and crop respect to the SQ categories described in the group discussion. performance indicators. Soil quality was seen as dynamic by the farmers, since a particular unit of land can have high or low SQ based 2.5. Data management and analysis on the type of management imposed or natural processes, including erosion, that were observed. Data management was handled using a MicroSoft Excel spreadsheet. The farmers used popular local terms for good, medium and low All spatial data (point location using GPS) for each SQ category, and SQ as shown in Table 2. They witnessed that dark soils are fertile with source of runoff or sediment were identified based on the consensus of high water-holding capacity and that they generally produce good the participant farmers and entered into the spreadsheet. The data was crop yields. The local term ‘Diqua’ meaning fertile soil was commonly subsequently accessed by Geographical Information System (GIS) used by farmers to describe good SQ. According to the farmers' software and used to develop a GIS SQ map that helped identify critical medium soil depth, mixed red and dark colors, and the presence of sources of runoff and sediment delivery. In addition, data analysis was some stone out-crop on the soil characterized medium SQ. Red, white carried out using SPSS release 18.0 software. Descriptive statistics such and yellow colored soils were commonly used by local farmers to as frequencies and percentages were used. Chi-square (χ2) was applied describe poor SQ. The farmers thought that poor soils had low fertility, to test whether a particular SQ indicator was significantly used by the a tendency to dry up quickly and generally produced lower crop interviewed farmers or not while categorized the SQ. yields, particularly in low rainfall seasons. The farmers added that poor soil can be described by shallow depth, high weed infestations, a 3. Results and discussion sandy texture, and a very loose surface that is easily detached by raindrops and runoff (Table 2). 3.1. Participatory soil quality diagnosis In addition, according to the farmers' point of view, reguid (deep soil) has better water holding capacity, is more fertile and therefore more The participatory SQ survey indicated that farmers have the productive. This is consistent with research findings reported by Haile experience and knowledge to assess SQ status as well as the severity (1995) and Corbeels et al. (2000) in reports that are focused on soil G.B. Tesfahunegn et al. / Catena 86 (2011) 1–13 5

Table 2 broad range of criteria was used by farmers in describing their field Consensus-based description of local soil quality indicator terms used for classifying SQ. It was noted that farmers are not only simply concerned with high, medium and low soil quality by local farmers in Mai-Negus catchment, Northern factors, such as soil properties, fertility, and suitability for a specific Ethiopia. crop but also take into account the whole range of other properties Soil descriptor (local terms) Translation related to social and cultural issues such as history of previous Reguid High soil quality management and how it relates to other nearby fields to describe soil Tselimo hamed Darkish soil quality status. Similar findings have been documented in Southern Aeman zeibilu Not stony out-crop Ethiopia by Elias (2000), where farmers based their management Diqua' Highly fertile soil fi Reguid hamed Deep soil decisions on a range of parameters. Such ndings contrast with the Maekelay Medium soil quality classic approaches in evaluating soil, which only use the physical Hawsi Walka/tselimo Mix of red and dark soil aspects to assess its inherent or dynamic soil qualities in determining Maekelay Medium soil depth agricultural or environmental values. Despite such knowledge of the Kirub Aeman zelebo Some stone out-crop local communities, the problem of soil quality degradation still Rekik Low soil quality Keih, hamekushtay hamed Red soil, light, yellow soil continues in many areas of Ethiopia. Therefore, to tackle this common Aeman zelebo Stone-out crop dominated problem, approaches that fully involve the indigenous community Enda-Tsihayay Highly weeds infestation should be designed in such a way that they should address the Hashewama Sandiness concern of resource degradation in Ethiopia and other similar areas. Teferkashay Loose soil fi Rekik hamed Shallow soil In general, this SQ study is rooted in eld experiences of local farmers which translate the descriptive indicators based on look, feel, smell, workability, productivity and presence of biota. That could be fertility, which is just one indicator of SQ. Systematic studies using local part of the reason why Pawluk et al. (1992) and Harris and Bezdicek farmers' knowledge of soil and their soil classification have been carried (1994) conclude that farmer derived descriptive properties are out in a few developing countries, such as Nigeria (Osunade, 1988), valuable for describing in meaningful terms the SQ assessment. It is Indonesia (Grobben, 1992), Zambia (Sikana, 1993), Rwanda (Habarur- also providing a foundation for developing and validating an ema and Steiner, 1997) and Kenya (Macharia and Ng'ang'a, 2005). This analytical component of SQ based on quantifiable soil properties study adds to the scientific knowledge by incorporating the farmers' that can be used as a basis for management and policy decisions. We experience or understanding of soil quality in northern Ethiopia. also conclude that farmer knowledge regarding management of SQ The SQ indicators identified in this participatory survey revealed throughout Ethiopia should be utilized and well documented. that the valley bottom of the catchment had medium to high SQ whereas low soil quality is widespread on the rolling-hills, central- 3.2. Severity of degradation in soil quality indicators ridge, and mountainous landforms of the catchment. This is illustrated by the transect-walk description map (Fig. 2) and general spatial For evaluation of SQ, it is desirable to select indicators that are distribution of SQ categories (Fig. 3). It is interesting to note that a directly related to the intended soil function (Karlen et al., 1997). In

West East

- Upland (steep)areas with shallow soil depth - Protected dense - Scattered rock exposure - shallow soil depth vegetation - most part subsoil exposure - Loose soil and red soil - stone cover - bush, shrub and tree on color dominated high some parts of closed area - Big to small active -Reddish soil -Scattered acacia spp. - Big and active gullies gullies, rills, root color - Gully erosion active - Some settlement/inhabitants exposure & pedestals - rills & small - Infertile cultivated and at the shoulder slope up-to the - Settlement with gullies active marginal land dominated - Church surrounded foot slope scattered tress & shrubs - Escarpment with - Stone terraces breached by small patches of - Cultivated land dominated around, some scattered - Rills, sheet wash and trees by mixed dark & red soil at - broken terraces, less settlers pedestals dominant - Very shallow soil the valley fertile (low soil quality) - Valley landform -medium to high - Light color soil depth around the - Most soil at the valley in most distance farms between the central soil quality dominated by loose soil church and increasing bottom are fertile with deep but medium to high soil ridge and plateau dominated of low soil quality the soil depth down soil of high to medium soil quality near homestead landforms cultivated land - Rock out crops at the the slope quality but some pocket areas - Rock exposure covered - Big to medium -truncation of A border of grazing land - low soil quality of low soil quality large part of the gullies active to less horizon on many - Grazing land connected dominated - Open grazing land at the toe landscape active (some part) fields connected to micro-dam - Very light soil color of the valley bottom - Open grazing with - high soil quality to the valley - On the grazing land - Rills & sheet wash

Decreasing slope steepness/elevation connected to the micro-dam. alluvial soils at the foot mixed black and red bottom. erosion indicators of soil erosion indicators of this landform color -Grazing land quality such as rills, common connected to the micro- - Some irrigation with alluvial soil pedestals, root exposure, - Less common SWC dam. vegetables & fruits is connected with loose soils, and sediment measures such as - Protected pasture the reservoir. deposits are common. terracing observed - Bordered by grazing land some part land connected to the - Grazing land around micro-dam. Transect-III micro-dam Transect-I Transect-II Transect-IV Transect-V Transect-VI Valley bottom of the catchment

Reservoir (Micro-dam)

Fig. 2. Transect-walk diagram that shows soil quality indicators and other resource variability across the landscape according to farmers' view at Mai-Negus catchment in Tigray, Northern Ethiopia. Each column designated by I–VI was subjected to each transect-route in the catchment. 6 G.B. Tesfahunegn et al. / Catena 86 (2011) 1–13

460300 461200 462100 463000 463900 464800

N

Soil quality categories

Low soil quality Medium soil quality High soil quality

catchment boundary Reservoir

Projection: UTM Zone: 37N 1561600 1562400 1563200 1564000 1564800 Datum: Adindan Spheroid: Clarke 1880

00.4 0.8 1.6

1560800 1561600Kilometers 1562400 1563200 1564000 1564800 460300 461200 462100 463000 463900 464800

Fig. 3. General spatial distribution of soil quality categories identified during participatory farmer transect-walks in a Northern Ethiopia catchment. Pocket small fields were generalized during classification to each of the soil quality categories because tracing them manually would have been too time and labor consuming.

doing so, key indicators of SQ were identified and assessed during the were perceived not to be continued for sustainable crop production, transect-walk by a group of farmers (Figs. 2 and 3). During the unless appropriate remedies are taken. Where soil surfaces showed transect-walk, the group of farmers having different economic status evidence of erosion indicators such as splash pedestal, sheet wash and identified SQ indicators that describe the severity of SQ degradation soil structure starting to become loose, it is understood by the farmers and then brought to discussion with other household-head farmers in that SQ is being deteriorated. Therefore, the order of ranking reflected the study catchment. After the discussion, a consensus was reached on the degree of soil damage caused by increasing erosion features as the list of SQ indicators and ranked according to their relative shown by the high frequencies of soil quality indicators (Table 3). This importance in the study area. The frequency of rills, gullies and other also helps in identifying possible causes and solutions from the SQ indicators that were counted at several points along the transect- farmers' point of view. walk was summed to estimate the abundance of such indicators and When farmers met, they actively participated in describing the then ranked according to the severity of degradation with respect to nature of fields they possessed and production constraints and the SQ indicators (Table 3). SQ indicators identified as the most potentials in detail. In such condition, they were able to distinguish important in the study catchment by farmers were thus rills, followed between the SQ indicators such as erosion that had evolved because of by root and subsoil exposure (Table 3). Indicators of SQ in the form of current or past soil erosion events and other soil management erosion and sedimentation processes were easy to identify by the practices (Table 3). This helped for the local farmers to understand the group of farmers during the transect-walk. history of soil erosion in a segment of field or landscape profile, and The process of achieving community consensus (involving above judge whether the soil erosion situation is high, moderate or low. 60% of the land owners) on ranking SQ indicator in the form of Many farmers were able to evaluate the conditions of their own fields erosion, soil depth and color raised a huge debate among farmers in using changes in topsoil characteristics due to the effect of erosion. the meeting. In some instances, farmers had to visit the actual eroding They are also able to link the changes in soil conditions due to erosion sites so that they can verify the differences in the severity status of an to the crop productivity. indicator. In order to verify further whether the communities in the The evidence of the existing soil erosion as one SQ indicator was catchment have agreed with the farmer group who participated in the demonstrated by identification of several on-site erosion and soil transect-walk, discussions on the overview of the SQ descriptors from fertility and off-site reservoir sedimentation indicator measures the transect-walk were carried out, which resulted into important during the participatory survey (Table 3). According to this survey reactions among the participants. Final resolutions on relative severity result, the most oftenly observed erosion indicators were rills, root of the widespread SQ indicators showed that the presence of gullies, exposure, and subsoil exposure (Table 3; Fig. 4). Even though the subsoil exposure and rills in the catchment soil surface (Table 3) numbers of gullies are few as compared to the other indicators, the indicated the erosion situation was severe, and such soil conditions contribution for sedimentation and soil loss might be significantly G.B. Tesfahunegn et al. / Catena 86 (2011) 1–13 7

Table 3 Consensus-based list of soil quality indicators, total frequency and relative severity as ranked by local farmers in Mai-Negus catchment, Northern Ethiopia.

Soil quality descriptors Total frequency count Severity ranking a Indicator due to b, c Measure

Presence of gullies 11 9 Past and ongoing soil losses Gully expansion or develop Rills 89 1 Ongoing soil losses Presence of rills Root exposure 82 2 Past soil losses Difference soil depth Soil color change 43 4 Ongoing soil losses Direct observation Rock exposure 7 10 Past soil losses Difference soil depth Sedimentation 23 7 Ongoing soil losses Sediment thickness Sheet wash 38 5 Ongoing soil losses Direct observation Splash pedestals 18 8 Ongoing soil losses Soil depth difference Build-up of soil against barriers 27 6 Past soil losses Soil height difference Subsoil exposure 51 3 Past soil losses Soil depth difference

a Where severity ranking order as 1 = severe degradation and 10 = low degradation. This ranking is based on the count made for each soil quality indicator using erosion features along the transect-walk in a 10 m×10 m area. Farmers noticed the presence of few gullies as compared to the other indicators. But gullies may contribute to high soil loss as they are active in the study area, despite the fact that this demands further investigation. b Farmers defined ongoing erosion indicators as those indicators that developed within a single or couple of rainfall events but where evidence of these indications were easily obliterated during tillage operations; they are thus also referred to as reversible indicators. c Whereas the past erosion indicators were those erosion indicators that had developed progressively to more severe erosion conditions mainly due to negligence of the recurring current indicators. These can either be or not be obliterated through tillage operations or restorative management; indicators that cannot be obliterated in a short-term period are termed as irreversible indicators.

large since most of the gullies were active. The severity of sheet wash empowerment of local farmers on soil quality improvement should and pedestals can be masked as these are easily destroyed by human be re-designed to halt such problems of soil quality degradation. and animal activities. To do so, farmers who participated in the transect-walk identified Stone terraces were observed dominantly in the catchment as that the steepest landscape parts are the source of large amount of compared to other structures designed to reduce soil loss. However, runoff and sediment (hotspots of degradation) (Fig. 6). This the failure or breaching of soil and water conservation (SWC) confirmed the finding in Hurni et al. (2005) which reported structure due to runoff force from upper slopes and human and degradation is not uniform, even in the same landscape. Farmers livestock interferences has resulted in subsequent erosion damages in noticed that high runoff and sediment source areas of the landscape the downhill fields by creating new gullies or changing the direction need first priority while introducing management practices than the of the flow and breaking other conservation structures. Generally, the others. This is because as what farmers underlined during the participatory survey confirmed that farmers have adequate knowl- transect-walk the sediment sources such as the gullies at the lower edge related to factors determining SQ. However, they are not being part of the catchment are formed as a result of the runoff coming from able to tackle the problem of SQ deterioration mainly because of lack the steeper source areas. But cultivation close to the margin of the of capital, labor and technical options in addition to their reluctance. gullies and over grazing increases the chance of collapsing gully sides They suggested that the food insecurity problem reduces farmer and developing wider gullies due to high runoff and thereby to be the willingness to take action against degradation as they give priority to source of large sediment in the catchment. This was supported by activities related to their immediate daily food requirements. farmers during the group meeting after the argument of small areas of The SQ indicators identified by the farmers were also ranked as land tending to be the source of disproportionately large amounts of more problem,2 less problem,3 no problem4 and don't know5 (Fig. 5) runoff and sedimentation within the catchment. Therefore, this as compared to the status in the past 5–10 years. Soil erosion, indicated that confining mitigation to erosion source areas costs less followed by soil fertility was considered by the majority of farmer than targeting wider areas in a resource poor country such as Ethiopia respondents (about 80% in the group discussion) as ‘more problem’. using the local knowledge as an input for decision making. This study This was followed by soil dryness and compaction in decreasing order. suggested that a strategy on environmental programs should be On the other hand, soil workability followed by soil compaction was focused on critical problem source areas within a hydrological unit pointed-out as ‘less problem’ in the study catchment. The number of instead of introducing universal controls. farmers who voted as ‘no problem’ for soil workability was larger than Furthermore, when farmers asked to suggest actions and solutions the other indicators, but the number of farmers who voted as ‘no to the problem of SQ degradation, the farmers in the group discussion problem’ for soil fertility and erosion was very small. Most farmers in suggested more than one action. The most important practices the catchment also understood that the problem of SQ was declining suggested were constructing intensive terraces throughout the as evidenced by the high fertilizer demand of soils and they thought catchment integrated with planting economic trees and shrubs, that the problem might be getting worse with time. Therefore, a close enclosed low SQ areas, use of fertilizer and appropriate cropping follow-up and monitoring systems with the involvement and system related management practices. The assumption is that integration of such practices considering the land use and terrain factor differences can rehabilitate degraded areas rapidly. Zero- grazing using cut and carry system of grasses introduced recently in 2 When clearly visible erosion, soil fertility, water holding capacity, etc., are the study catchment was also appreciated by farmers as part of the observed compared to productive soils. Soil is virtually lost and is not suitable for important approaches to improve SQ degradation as compared to agricultural systems; the original resources are largely destroyed and need major stocking the livestock for the whole year on the grazing land (Fig. 7). investments and work to restore to its full productivity. 3 Fields that formerly showed good but productivity is strongly declining as shown But the area used protected land for cut and carry grass system was by a particular SQ indicator, but is still suitable for local farming system. Inherent very small in proportion. Farmers also pointed-out that in order to quality and biotic resources are partially destroyed and as a result demand major successfully rehabilitate degraded areas by enclosing, active involve- improvement efforts from the land users. ment of the farmers within the catchments is needed and potential 4 Soil fulfills intended function and is still suitable for local farming systems; full conflicts between land users should be resolved. Farmers whose land productivity with some additional inputs. 5 Not enough observations or knowledge whether the trend of a particular soil is to be enclosed should get a compensation land or other equivalent quality indicator is changed or not. incentives from the government or any supporting agents. Strict local 8 G.B. Tesfahunegn et al. / Catena 86 (2011) 1–13

Rills

Root exposure

sediment

Subsoil exposure

Stone out-crop

Fig. 4. Most frequently observed soil quality indicators due to water erosion during transect-walk with farmers in June 2009: rills, sediment deposition, root exposure, subsoil exposure and stone out-crop in Mai-Negus catchment, Northern Ethiopia. regulations should be set out by the farmers themselves to manage exploitation from the enclosures can provide support for the farmers effectively any destruction or interferences by human or livestock to and the environment sustainably. enclosed areas. Strategies should also be designed to grow “trees” that increase income for the local farmers while improving soil quality in 3.3. Farmers' understanding of causes for declining soil quality the enclosures. Using such an approach in a long-term, resource Farmers' group discussion based on the transect-walk information more problem less problem no problem don´t know as a brain storming pointed out that erosion negatively impacted the 90 soil quality (crop production) and the overall environmental 80 condition such as sedimentation of reservoir and field boarders. It also revealed that rainfall intensity is high resulting in severe soil 70 losses, when the soils are bare. Soil erosion levels in the study 60 catchment are still high due to the fact that farmers are not making 50 much effort on SWC measures and land use re-design. This might be because many farmers are engaged on off-farm activities to maximize 40 income regardless of the seriousness of the on-going soil erosion on 30 soil quality. Besides, a lack of full involvement of local community on 20 problem identification and suggesting of remedies to problems before farmer respondents (%) 10 the implementation of new recommendation might contribute for farmers to be reluctant in taking soil conservation measures. By 0 fi soil erosion soil compaction soil fertility soil dryness soil workability involving farmers from the beginning to the nal stage of introduced technology, the constraints of the recommended techniques from soil soil quality indicators productivity and environment perspective can be understood better Fig. 5. Severity of soil quality indicators from different problem perspectives of local and they also feel ownership. Generally, the observation from the farmers in Mai-Negus catchment, Northern Ethiopia. transect-walk implied that steep slopes tend to be relatively G.B. Tesfahunegn et al. / Catena 86 (2011) 1–13 9

460300 461200 462100 463000 463900 464800

Erosion sources N Runoff and sediment source (steep areas) Runoff source (middle slope) Sediment source (cultivated and grazing disected by gullies) Stable area (flat to gentle slopes) Reservoir

00.4 0.8 1.6 Kilometers 1561600 1562400 1563200 1564000 1564800

Projection: UTM Zone: 37N Datum: Adindan Spheroid: Clarke 1880 1560800 1561600 1562400 1563200 1564000 1564800

460300 461200 462100 463000 463900 464800

Fig. 6. Overview of high runoff and sediment source (hotspot) areas based on information from the farmers who had participated in a transect-walk in Mai-Negus catchment, Northern Ethiopia. susceptible to water erosion as signified by a wide spread exposure of interrelated and call for comprehensive approach that takes into subsoil, rills, root exposure and shallow active gullies; and sediment account environmental variables such as slope, soil, crop cover and deposition and big gully formation on flat to gentle slopes of the study rainfall condition, besides the management practices to tackle SQ catchment (Table 4). degradation. Regular bund maintenance with emphasis on height Farmers were also asked to list and rank the main causes for increment is crucial to maintain the effectiveness of bunds to allow declining soil quality indicators (Table 4). Qualitatively, farmers in the the continued reduction of slope gradients and overland flows, in study catchment reached a consensus that the main causes for addition to the introduction of appropriate techniques. increasing the observed soil erosion features as a SQ indicator were The soil thickness and vegetation observed in the field and identified as poor soil cover, steep slopes/terrain, high intensive informal discussion with farmers and experts indicated that SQ has rainfall, inappropriate spacing and not timely maintenance of SWC been declining due to erosion, and nutrients are also mined because of measures and also the presence of loose soil fields (Table 4). both erosion effect and continuous cropping with minimal crop The participant farmers in the group meeting agreed that the rotation and fertilizer inputs. Gullies have long been established and numbers of observation for rills were highest as compared to the still continued the expansion (Fig. 8A); which in turn enhanced the other indicators. The main observable cause for this indicator was reduction in farm and grazing land size and therefore aggravated land identified as poor soil cover followed by the steep slopes. The fragmentation and land pressure. Rills and sheet erosion are also presence of poor soil cover and steep landscape in many parts of the frequently visible on cultivated land mainly on teff (Eragrostis tef) catchment caused the formation of more rills by runoff. In support of field and other crop lands located on the hillside slopes of the study farmers' observations, Poesen (1984) and Herweg and Ludi (1999) catchment as a challenge for soil quality maintenance. observed a tendency of rill formation as slopes became steeper and This study also suggested that to improve the SQ degradation in poor soil cover as a result of concentrated overland flow that Ethiopia, limitations of the existing conservation measures and land use increased the depth and number of rills. systems should be assessed from the context of local community in each Farmers also agreed that for the other indicators such as gullies, catchment, besides the empowerment of the contribution of local sheet-wash, red soil color, subsoil and root exposure and sedimen- communities to sustain the technologies. The basis for such remark is tation, the main cause was identified as excess runoffs which were that field observation indicated that past efforts on SWC measures did aggravated by high rainfall intensity, terrain, and poor soil cover not improve much the situation of soil degradation in the large part of (Table 4). The interference of human activities on steep terrain can the study catchment, though some areas are getting substantially better. aggravate more the effect of runoff on SQ. Broken SWC structures, There were evidences that supported this such as many fields were not coupled with the wide spacing and inappropriate structure may also conserved, terraces were destroyed and there was no regular increase the runoff amount and its effect on SQ. The reasons for maintenance, continuation of gully expansion and development, continued soil erosion processes in the study catchment are thus presence of highly degraded shallow marginal soils and poor soil 10 G.B. Tesfahunegn et al. / Catena 86 (2011) 1–13

(A) Table 4 Observed soil quality indicators and their causes based on local farmers' consensus in meetings following transect-walks in Mai-Negus catchment, Northern Ethiopia.

Soil quality No. of Proportion of observable causes (%) indicators observed Poor soil Steep Excess Poor Loose Othersb indicator cover slopes rainfall/ terracing soilsa runoff

Rills c 89 62 14 11 5 6 2 Gullies d 11 9 4 75 7 4 1 Root 82 4 6 56 3 7 24 exposure e Red soil color f 43 3 12 28 1 4 52 Stoniness g 8 7 16 37 6 1 23 Rock out-crops h 745656911 Sedimentation i 23 4 8 59 8 2 19 Sheet-wash j 38 5 3 62 4 16 10 Splash 18 3 2 45 0 16 34 pedestals k Broken SWC 53 4 9 12 19 7 49 structures l Subsoil 51 2 5 65 3 20 5 exposure m Soil fertility 45 14 11 25 3 4 43 loss n

a Implies soils are prone to erosion and easily scoured by runoff water. They are neither dark nor red but have poor water-holding capacity, and are interspersed (B) between red and darker soils. b This includes management practices such as tillage, fertilizer, removal of crop residues, grazing pressure, deforestation and human and live stock interferences. c Continuous or discontinuous channels. They developed after an intensive rainfall event commencing from a short distance of ridge to crest that concentrate into channel but it can be easily destroyed by tillage. d Larger than rills and locally distinguished from rills when a 7 year-old child cannot jump across it. e Exposure of aerial roots after topsoil is stripped off by runoff and splash effect of raindrop. Indicates that topsoil had been removed thus weakening the crop stability. f Implies top-dark soils have been removed by runoff, also used as a strong indicator of severe erosion leaving unproductive soils. g Small loose stones lying on soil surface signified that overlaying topsoil and subsoil layers have been removed by water erosion. h Partly exposed rocks indicate soils are shallow and have been washed off by runoff flow, exposing tips of underlying parent rock. i Identified by the burying of crops/grass or deposition of “new soil”. Such is marked by fertile or infertile zone in a field depending on its effect on soil functions. Soil material could be dark nutrient-rich or coarse sandy/stony deposit. j Marked by runoff flow path leaving a smoothened surface that shows direction of the flow. k Describes the created craters by raindrop and protected soil column by stone, root or crop residues. Found under and outside tree canopies. l Marked by gaps or breaching in formally continuous bunds of conservation measures. m Implies top-dark soils have been removed by runoff, also used as a strong indicator of severely eroded leaving unproductive subsoils. n Fields marked by shallow soil depth and poor crop vegetative performance. Fig. 7. (A) Land rehabilitation after two years of enclosed pasture under cut and carry grass system; and (B) livestock stocked throughout the year in Mai-Negus catchment, Northern Ethiopia, July 2009.

cover on many parts of the catchment (e.g., Fig. 8). Therefore, 3.4. Proportion of farmers used the indicators in diagnosis soil quality approaches on designing solutions to the processes of soil quality degradation in the study catchment and other similar areas should The chi-square test revealed that the proportion of the interviewed consider the landforms, potential erosion source areas, appropriateness farmers that used most SQ indicators identified during the participa- of the technology and full involvement of local farmers in all processes in tory group discussions significantly differed from those who did not such a way as to ensure sustainable natural resource management. use (Tables 5 and 6). Statistically significant chi-squares indicate a Moreover, to clearly understand farmers' knowledge of SQ degradation high proportion of differences in rating between the farmers on the and the impact of the technologies employed, different approaches need independent use and not use of each of the indicators. The proportion to be tried out. As a general remark, participatory assessment of soil of respondents who used crop yield (95%), top soil thickness (90%), quality using the experience of local communities is crucial to rapidly crop vigour (86%), soil fertility (78%) and soil erosion (83%) indicators monitor the sustainability of land management systems related to to categorize SQ significantly differed from those who did not use such agricultural production and environmental practices. This can assist indicators (P=0.000). farmers, decision makers and scientists in formulating and evaluating In addition, the chi-square probability levels showed significant agricultural land use and soil management systems against SQ differences between the proportions of respondents who used soil degradation from the end users' perspectives. color, fertilizer response of soils, moisture in dry season, weed G.B. Tesfahunegn et al. / Catena 86 (2011) 1–13 11

(A) infestation, texture, drainage condition and earthworm population and those who did not use such indicators to categorize the SQ (Table 5). Even though the test was significantly different, the proportion of farmers who used (14%) earthworm as a soil quality indicator to classify the SQ was few compared to the farmers that did not use it (86%). The proportion of farmers used the SQ indicators such as soil compaction or soil tilth and workability with those who did not use to categorize SQ levels was not significantly different by the chi- square test at P≥0.05 (Table 5). This indicates that the proportion of respondents who used such indicators was similar with those who did not use. Those farmers who did not use soil compaction, and tilth and workability as criteria to categorize their field SQ might be associated to the confusion with soil dryness. Because some farmers assumed that dry soils are compact and difficult to work. For all the physical soil quality (SQ) indicator local terms that were identified in each SQ category by the participatory approach, the chi- square test showed a significant difference between the proportion of farmers used as criteria and those who did not use during categorization of the SQ (Table 6). For instance, the proportion of respondents who used the local term Diqua' (fertile soils) (98%) to indicate high SQ was significantly different (P=0.000) from those who did not use it. The same holds true for all the other local indicator terms that describe high, (B) medium and low SQ (Table 6). The soil indicator termed as Diqua' (fertile soil) showed the highest proportion of respondents who used it as criteria to identify the high SQ category as compared to the others. Similarly, the terms Maekelay hamed (medium soil depth) for medium SQ, and Rekik hamed (shallow soil depth) for low SQ category, showed the highest proportion of respondents as compared to the remaining indicators in each category (Table 6). In this study, the chi-square test indicates that the proportion of farmer who owned high, medium and low SQ in the study catchment significantly differed at P=0.011. The proportion of the respondents who classified their soil into high, medium and low SQ categories was, 12, 40 and 48%, respectively. According to farmers (88%), most SQ category in the area was in the range of low to medium SQ categories. This reveals that much work has to be done in improving the existing situation of SQ degradation. Generally, this study indicates that crop yield, top soil thickness, crop vigour, soil erosion and soil fertility were the most frequently cited indicators by farmers to categorize SQ levels, besides the local indicator terms described. The reason for rating these as a frequent local SQ indicator was due to their simplicity for visual measurement or (C) judgment than the others. Romig et al. (1995) reported that crop appearance and erosion indicators were ranked first by farmers in northern U.S Corn and Dairy Belt as the top properties for describing SQ, which is consistent with the present result. A similar result with the present finding has been also reported in Saito et al. (2006) and Mairura et al. (2008) which describe soil color as an important SQ indicator by farmers. The composition and abundance of weed species growing on agricultural soils are also useful indicators of SQ frequently used by farmers, despite the fact that the local knowledge of plant species has not been fully and well documented. In addition, some weeds that grew in one season may not do so in the other (next) season. In general, this study indicates that such experience of visual approach of soil quality classification is rapid, less expensive and participatory in its nature, with reasonable accuracy for practical decision making as compared to scientific evidences taken in the catchment (data not shown). In line with our study, case studies elsewhere have shown that there is a consistent rational basis to the use of local indicators of SQ in decision making processes (e.g., WinklerPrins, 1999; Ramisch, 2004). Fig. 8. Continued gully development (A); rock out-crop exposure after top soil has beenremovedbyerosion(B),andbreachedstonebundswithoutmaintenanceon 4. Conclusion shallow soil depth in marginal area (C), Mai-Negus catchment, Northern Ethiopia, July 2009. The results of this study show that the assessment of soil quality using participatory survey has become an important approach to sustain soil functions as it is quick, less costly and has high reproducibility. Such 12 G.B. Tesfahunegn et al. / Catena 86 (2011) 1–13

Table 5 Table 6 Proportion of the interviewed farmers that used the soil quality (SQ) indicators to Proportion of the interviewed farmers that used the physical soil quality (SQ) indicator categorize the SQ in Mai-Negus catchment, Northern Ethiopia (na =42). local terms in classifying into high, medium and low SQ in Mai-Negus catchment, Northern Ethiopia (n=42) . SQ indicators Farmers who mentionedb χ2 probability Soil quality descriptor Translation Respondents (%) χ2 Yes (%)c No (%)d (local terms) a,b probability Crop yield 40 (95) 2 (5) 0.000 Reguid High soil quality Top soil thickness 38 (90) 5 (10) 0.000 Tselim hamed Dark soil 86 (36) 0.000 Crop performance/vigour 36 (86) 6 (14) 0.000 Aeman zeibilu No stone-out crop, pure 76 (32) 0.000 Soil fertility 33 (78) 9 (22) 0.000 soil Soil erosion 35 (83) 10 (17) 0.000 Diqua' Highly fertile soil 98 (41) 0.000 Soil color 31 (74) 11 (26) 0.002 Reguid hamed Deep soil 95 (40) 0.000 Fertilizer response of soil 30 (71) 12 (29) 0.005 Moisture holding in dry season 28 (67) 14 (33) 0.031 Maekelay Medium soil quality Weed infestation/abundance 27 (64) 28 (36) 0.031 Hawsi Walka/ Mix of red and dark soil 88 (37) 0.000 Soil compaction 16 (38) 30 (62) 0.123ns tselimo Soil tilth and workability 22 (52) 20 (48) 0.758ns Maekelay hamed Medium soil depth 95 (40) 0.000 Earthworm population 6 (14) 36 (86) 0.009 Kirub Aeman zelebo Some stone-out crop 86 (36) 0.000 Texture 29 (69) 15 (31) 0.014 Drainage condition 28 (67) 14 (33) 0.031 Rekik Low soil quality χ2 is chi-squared and ns is non significant at probability level N0.05. Keih, hamekushtay Red soil, white, yellow 88 (37) 0.000 a Number of household heads interviewed. hamed soil b Values in parentheses are percentages of respondents and without are counts. The Aeman zibeziho Stone-out crop 81 (34) 0.000 percentage total is more than 100% because each respondent was chosen more than one Enda-Tsihayay Highly weeds infestation 71 (30) 0.005 SQ indicator. Hashewama Sandy dominant 88 (37) 0.000 c Indicates farmers that used the SQ indicator to categorize their soils in the field into Teferkashay Loose soil 90 (38) 0.000 high, medium or low levels. Rekik hamed S hallow soil 98 (41) 0.000 d Shows those farmers who did not use the indicator for such purposes. a The percentage total is more than 100% because each respondent used more than one indicator on each SQ category. b Values in parentheses are percentage of respondents who used such local terms while classifying their soils and without are the corresponding counts. an approach supports the quality of technology introduction and dissemination targeting the soil quality problem areas. Generally, a References well-structured soil quality local knowledge-base is detected at the study area. Knowledge was not homogeneous among farmers; but Badege, B., 2001. Deforestation and land degradation in the Ethiopian highlands: a – many of them exhibited a refined and robust local knowledge and strategy for physical recovery. Northeast African Studies 8 (1), 7 25. Barrera-Bassols, Zinck, J.A., Ranst, E.V., 2008. Participatory soil survey: experience in understanding of soil quality indicators that can support for decision working with a Mesoamerican indigenous community. Soil Use and Management making. The local communities used soil erosion, soil fertility and 25, 43–56. biological (crop and weed) indicators together to describe the soil Barrios, E., Bekunda, M., Delver, R., Esilaba, A., Mowo, J., 2000. Identifying and Classifying Local Indicators of Soil Quality. International Center of Tropical quality categories as high, medium and poor levels but there was a Agriculture, Cali, Columbia. significant variation between the proportion of farmers who used a Barrios, E., Trejo, M.T., 2003. Implications of local soil knowledge for integrated soil certain soil quality indicator and those who did not use. Since soil quality management in Latin America. Geoderma 111, 217–231. fi Bekele, S., Holden, S., 1999. Soil erosion and smallholders' conservation decisions in the measurement using scienti c techniques is so expensive, time consum- highlands of Ethiopia. World Development 27, 739–752. ing and limited in representation to large areas and complex Boyd, C., Turton, C., 2000. The contribution of soil and water conservation to sustainable catchments, the participatory survey approach of assessing soil quality livelihoods in semi-arid areas of Sub-Saharan Africa. Network Paper, 102. The can be useful in developing countries where resources are scarce. It can Agricultural Research and Extension Network, London. Corbeels, M., Shiferaw, A., Haile, M., 2000. Farmers' knowledge of soil fertility and local be thus noted that farmer derived SQ indicators are crucial in providing management strategies in Tigray, Ethiopia. Managing Africa's Soils, 10. IIED, London. the basis for sustainable management and policy decision making. Dai Trung, N., Verdoodt, A., Dusar, M., Tan Van, T., Van Ranst, E., 2007. Evaluating However, for effectively putting anti-degradation technologies, farmers ethnopedological knowledge systems for classifying soil quality. A case study in BoHamlet with Muong people of Northern Vietnam. Geographical Research 46 (1), should understand the issue of the technologies and be fully aware of 27–38. soil quality degradation especially its nature, scope, and responsible Elias, E., 2000. Soil enrichment and depletion in southern Ethiopia. In: Hilhorst, T., — factors and suggest possible solutions from local perspectives so that Muchena, F.M. (Eds.), Nutrient on the Move Soil Fertility Dynamics in African Farming Systems. llED, London, UK, pp. 65–82. technologies can be implemented easily and adopted sustainably. Federal Democratic Republic of Ethiopia Population Census Commission, FDREPCC, Participatory survey also promoted cooperation between local and 2008. Summary of Statistical Report of the 2007 Population and Housing Census. external participants and forms the basis for an agreed land manage- Addis Ababa, Ethiopia. Grobben, P., 1992. Using farmers' knowledge about soil. A Case Study of a Farming ment planning, implementation and evaluation that can be part of the Systems Research (FSR) Project in Java. AT-source, No. 3/92, Wageningen. robust approaches for sustainable management of natural resources. Habarurema, E., Steiner, K.G., 1997. Soil suitability classification by farmers in southern However, further research that verifies the soil quality categories Rwanda. Geoderma 75, 75–87. fi fi Haile, M., 1995. Indigenous knowledge and agricultural practices in central Tigray. identi ed by local communities using scienti c soil measurement Proceedings of the Workshop on Agro-Ecological and Health Studies in the Central should be done so as to know discrepancies and similarities between Zone of Tigray, (Unpublished), March 1995, Mekelle, Ethiopia. local and scientific knowledge. Harris, R.F., Bezdicek, D.F., 1994. Descriptive aspects of soil quality/health. In: Doran, J.W., Coleman, D.C., Bezdicek, D.F., Stewart, B.A. (Eds.), Defining Soil Quality for a Sustainable Environment: SSSA Spec. Pub. No. 35, Soil Sci. Soc. Am., Am. Soc. Agron., Acknowledgements Madison, WI., USA, pp. 23–35. Herweg, K., Ludi, E., 1999. The performance of selected soil and water conservation — – The authors sincerely acknowledge the financial support by DAAD/ measures case studies from Ethiopia and Eritrea. Catena 36, 99 114. Holden,S.,Shiferaw,B.,Pender,J.,2005.Policyanalysisforsustainableland GTZ (Germany) through the Center for Development Research (ZEF), management and food security in Ethiopia—a bioeconomic model with market University of Bonn (Germany), and field work supported by Aksum imperfections. Research Report, 140. International Food Policy Research Institute, University (Ethiopia). The authors highly appreciate the cooperation Washington DC. Hurni, H., Tato, K., Zeleke, G., 2005. The implications of changes in population, land use, of the participant farmers and assistance offered by the local and land management for surface runoff in the upper Nile Basin Area of Ethiopia. administration and experts during the field study. Mountain Research and Development 25, 147–154. G.B. Tesfahunegn et al. / Catena 86 (2011) 1–13 13

Karlen, D.L., Mausbach, M.J., Doran, J.W., Cline, R.G., Harris, R.F., Schuman, G.E., 1997. Poesen, J., 1984. The influence of slope angle on infiltration rate and Hortonian overland Soil quality: a concept, definition, and framework for evaluation. Soil Science flow volume. Zeitschrift füf Geomorphology 49, 117–131. Society America Journal 61, 4–10. Pawluk, R.R., Sandor, J.A., Tabor, J.A., 1992. The role of indigenous knowledge in Karlen, D.L., Andrews, S.S., Doran, J.W., 2001. Soil quality: current concepts and agricultural development. Journal of Soil and Water Conservation 47, 298–302. applications. Advances in Agronomy 74, 1–40. Ramisch, J.J., 2004. Understanding soil fertility in its social context: integrating social Kasemir, B., Ja¨ger, J., Carlo, C., Jaeger, C.C., Gardner, M.T., 2003. Public Participation in and natural science research within AFNET. In: Bationo, A. (Ed.), Managing Nutrient Sustainability Science: A Handbook. Cambridge University Press, Cambridge. Cycles to Sustain Soil Fertility in Sub-Saharan Africa. ASP, Nairobi, pp. 501–522. Kebrom, T., 1999. Land degradation problems and their implications for food shortage Röling, N., 2002. Beyond the aggregation of individual preferences—moving from in south Welo, Ethiopia. Environmental Management 23 (4), 419–427. multiple to distributed cognition in resource dilemma. In: Leeuwis, C., Pyburn, R. Krowntree, K.M., Fox, R.C., 2008. Active learning for understanding land degradation: (Eds.), Wheelbarrows Full of Frogs. Royal Van Gorcum, Assen, pp. 25–47. African catchment game and risk map. Geographical Research 46 (1), 39–50. Romig, D.E., Garlynd, M.J., Harris, R.F., McSweeney, K., 1995. How farmers assess soil Macharia, P.N., Ng'ang'a, L.W., 2005. Integrating indigenous soil and land classification health and quality. Journal of Soil and Water Conservation 50 (3), 229–236. systems in the identification of soil management constraints in the tropics: a Kenya Saito, K., Linquist, B., Keobualapha, B., Shiraiwa, T., Horie, T., 2006. Farmers' knowledge case study. Tropical and Subtropical Agroecosystems 5, 67–73. of soils in relation to cropping practices: a case study of farmers in upland rice Mairura, F.S., Mugendi, D.N., Mwanje, J.I., Ramisch, J.J., Mbugua, P.K., Chianu, J.N., 2008. based slash-and-burn systems of northern Laos. Geoderma 136, 64–74. Scientific evaluation of smallholder land use knowledge in Central Kenya. Land Sillitoe, P., 2000. Let them eat cake. Indigenous knowledge, science and ‘the poorest of Degradation and Development 19, 77–90. the poor’. Anthropology Today 16, 3–7. Mostert, E., 2003a. The challenge of public participation. Water Policy 5, 179–197. Sikana, P., 1993. Mismatched models. How Farmers and Scientists See Soils. ILEIA Mostert, E., 2003b. Public participation and the European water framework directive: a Newsletter l/93, Leusden (Netherlands). framework for analysis. Inception Report of the HarmoniCOP Project, RBA-Centre. Stave, K.A., 2002. Using system dynamics to improve public participation in Delft University of Technology, Delft. environmental decisions. System Dynamics Review 18, 139–167. Osunade, A.M.A., 1988. Soil suitability classification by small farmers. Professional WinklerPrins, A.M.G.A., 1999. Local soil knowledge: a tool for sustainable land Geographer 40, 194–201. management. Society and Natural Resources 12, 151–161. Osunade, M.A.A., 1994. Community environmental knowledge and land resource WinklerPrins, A.M.G.A., Sandor, J.A., 2003. Local soil knowledge: insight, application, surveys in Swaziland. Singapore Journal of Tropical Geography 15, 157–170. and challenges. Geoderma 111, 165–170.