1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Assessment of Upland Erosion Processes and Farmer Perception of Land
Conservation in Debre-Mewi Watershed, near Lake Tana, Ethiopia
Assefa D.Zegeye1,2, Tammo S. Steenhuis2,3, Robert W. Blake2,4 , Selemyihun Kidnau5 ,
Amy S. Collick2,3, Farzad Dadgari5
1
Adet Agricultural Research Center, ARARI, PO.box 1326 Bahir Dar, Ethiopia
(adz6@cornell.edu),
2
Master’s Program in Integrated Watershed Management and Hydrology, Cornell University,
at Bahir Dar University, Bahir Dar, Ethiopia.
3
Department of Biological and Environmental Engineering, Cornell University, Ithaca, NY
14853, USA (tss1@cornell.edu, asc38@cornell.edu )
4Center for Latin American and Caribbean Studies, 300 International Center, Michigan State
University, East Lansing, MI 48824 (rwblake@msu.edu)
5
SWHISA, Bahir Dar, Ethiopia (selamyhun@yahoo.com, fdadgari@sympatico.ca)
For presentation at the International Symposium: “Ecohydrology for water ecosystems and
society in Ethiopia” Addis Ababa, Ethiopia: 18-20 November 2009 and submitted to
Ecohydrologica
Abstract
22
23
24
25
26
27
28
29
30
Erosion is of great concern in the Ethiopian Highlands. The objective of this study was to
determine the soil erosion rates under actual farming conditions by measuring the dimensions
and number of rills in 15 agricultural fields in the Debre-Mewi Watershed near Lake Tana
and to understand farmer attitudes towards land conservation through personal interviews
held with one-third of the watershed households. The annual rill erosion rate was 8 to 32 t/ha.
Greatest rates of erosion occurred at planting early in the season but became negligible in
August. Major factors influencing land conservation decisions-were the demand of labor and
lack of technical support for implementing new conservation measures from experts.
31
Key words: Rill erosion, soil loss, farmer perception, SWC,
32
Running headline: Upland Erosion and Perception of Land Conservation
33
1
Introduction
34
Soil erosion creates severe limitations to sustainable agricultural land use, as it
35
reduces on-farm soil productivity and causes food insecurity (Tadesse 2001, Sonneveld 2002,
36
Beshah 2003, Moges and Holden 2006, Bewket 2007). In most developing countries,
37
including Ethiopia, human activity triggers these losses (Mohammad et al., 2001, Belyaev et
38
al., 2004, Bewket and Sterk, 2005, Hurni et al., 2005). This is associated with rapid
39
population growth, inadequate attention to the basic natural resources, and the need to
1
40
maximize production to meet the needs of the growing population (Shiferaw and Holden,
41
1999, 2000, Bewket, 2002, Feoli et al., 2002). This situation is more serious in poor
42
developing countries like Ethiopia (Feoli et al., 2002), where subsistence production
43
predominates. The Ethiopian farmer, who on average cultivates one hectare of food crops and
44
keeps some livestock, is dependent on natural conditions and cannot tolerate further
45
deterioration of soil productivity (Sonneveld and Keyzer, 2003). Farms in the Debre-Mewi
46
watershed suffer from severe soil erosion, with gullies and their deleterious effects increasing
47
at alarming rate. Basic natural resources like soil, water and vegetative cover in the watershed
48
are deteriorating. Farmers are not satisfied with the status of their current land holding.
49
Visible erosion features, such as rills, gullies and concentrated accumulations, are
50
features that often indicate erosion hot spots, those parts of an area that are most seriously
51
affected by soil erosion (Mitiku et al., 2006). Rill erosion is the very visible mechanism of
52
soil loss from sloping cultivated land. Rills are very shallow channels formed by the
53
concentration of surface runoff along depressions or low points in sloping lands. Rill erosion
54
is a result of surface runoff and associated sheet wash, which selectively removes fine
55
material and organic matter that are very important determinants of land productivity
56
(Bewket and Sterk, 2003). The shearing power of water can detach, pick up and remove soil
57
particles making these channels the preferred routes for sediment transport.
58
The evaluation of soil erosion was undertaken through measurement of rill erosion
59
features (Bewket and Sterk, 2003, Belyaev et al., 2004, Casalí et al., 2006). This method is
60
used to estimate the amount of material removed from the field by concentrated runoff. As
61
suggested by Bewket and Sterk (2003), rill survey is a key for practical conservation planning
62
purposes. Most quantitative research on soil erosion has dealt with soil loss rates due to rill
63
and inter-rill erosion, mostly from runoff plots. The magnitude of soil erosion and its effect
64
has been extrapolated based on plot level and catchment level to larger spatial scales (Bewket
2
65
and Sterk, 2003, Nyssen et al, 2004, Belyaev et al., 2004). These methods have drawbacks
66
when the results are extrapolated and conclusions are made about the wider landscape
67
(Brazier, 2004).
68
Field surveys of rills have an integral role to play in the sustainable management of
69
agricultural lands in particular. Without involving expensive instrumentation and
70
sophisticated modeling of soil loss, this strategy may yield more economical (and efficient)
71
solutions in local areas than the application of the existing generation of erosion models
72
(Herweg, 1996; Bewket and Sterk, 2003). Surveys must also be treated as a means to aid soil
73
conservation (Herweg 1996) and to inform catchment managers and decision/policy makers.
74
Hence, assessment of soil loss by surveying rill erosion plays a great role for soil and water
75
conservation planning.
76
Farmer perception of land degradation by erosion is a key social factor that is also
77
important in deciding options for controlling soil losses (Graaff, 1993). Some authors who
78
studied in different parts of the Ethiopian highlands reported that farmers are more likely to
79
adopt conservation measures in plots that are highly prone to soil erosion, such as plots where
80
slopes are steep and erosion features are visible (Shiferaw and Holden 1998; Bekele and
81
Drake 2003; Gebremedhin and Swinton 2003). Amsalu and Graaff (2006) concluded that
82
under current conditions in the Ethiopian central highlands, soil and water conservation
83
interventions should consider farmer conservation knowledge and practices to improve the
84
possibility of acceptance and adoption of the recommendations.
85
The overall goal of this research was thus to determine the severity and rate of soil
86
erosion and to identify factors affecting the decision-making process of farmers regarding
87
their land conservation practices. The specific objectives were: (1) to compare the rate of soil
88
loss within different slope positions and crop covers, (2) to assess some causes of soil
89
erosion, (3) to understand farmer perception and awareness of erosion processes and SWC
3
90
measures as a land management practice, and (4) to examine the local farmer acceptance and
91
adoption of land conservation technologies.
92
2
Materials and methods
93
2.1
Description of the study sites
94
This paper is based on a survey conducted during the months of July to October 2008
95
in the Debre Mewi watershed located between 11o20’13’’ N and 37o25’’55’’ E. The
96
watershed is located about 30 km south of Lake Tana at Bahir Dar Town, the capital of
97
Amhara Regional State in the northern Ethiopian highlands (Fig. 1). The elevation ranges
98
between 1950 and 2309 m a.s.l. The total area of the watershed is about 523 ha. The total
99
number of households living in this watershed is estimated to be 276, of which 35 (13%) are
100
women-headed households. It is also estimated that about 1092 people are living in the
101
watershed.
102
Based on 10 years (1996-2005) of metrological data for the Adet station (7 km from
103
the site), the maximum annual temperatures range from 22 to 29.4 oC, whereas minimum
104
annual temperatures range from 5.4 to 12.1oC. The average annual rainfall is 1238 mm.
105
2.2
106
2.2.1 Measurement of rill erosion
Methodology
107
For determining erosion by the rill measurements, 15 representative fields were
108
selected from the Debre Mewi watershed, representing a cumulative field area of 3.56 ha.
109
These fields were classified into three slope positions: upslope, mid-slope and down-slope. A
110
series of cross-slope transects was established (marked using sticks and stones) with an
111
average distance of 10m between two transects, positioned one above another to minimize rill
112
measurement errors (Hudson, 1993). During the months of July and August when the greatest
113
rainfall amounts causing significant soil loss were recorded, each field was repeatedly visited
114
immediately after rainfall storms had occurred.
4
115
Though the channel size and shape of rills had a great influence on measurement
116
accuracy (Casalí et al., 2006), the length, width and depth of the rills were carefully measured
117
along two successive transects. The length of a rill was measured from its starting point up to
118
the place where the eroded soil was deposited. The width and depth of rill were measured as
119
shown in Figure 2. Widths were measured at several points along a rill to give a better
120
approximation of the mean width because the width varied along the rill (Herweg, 1996).
121
Fewer depth measurements were required because the depths of most rills were homogenous.
122
From these measurements, different magnitudes of rill erosion were determined, including rill
123
volumes, rates of erosion, density of rills, area of actual damage by the rills and the
124
percentage of area covered by the rills to the total area of surveyed fields (Herweg, 1996,
125
Hagmann, 1996, Bewket and Sterk, 2003). The percentage crop canopy coverage was
126
estimated whenever each rill measurements were undertaken.
127
2.2.2 Farmer perceptions of SWC measures
128
In order to capture the perception of soil erosion process and factors affecting farmer
129
land conservation decision-making processes and related reasons, both a formal and informal
130
interview were conducted. In addition, the researcher periodically visited the entire research
131
area and the wider farming system so that he was able gain additional insights by observation.
132
Data and information about farmer perceptions of soil erosion processes and SWC
133
technologies were collected using formal interviews with the sampled households. To obtain
134
information about the same fact from multiple methods and to increase validity and reliability
135
of data, focus group discussions (composed of elders, male and female farmers and
136
community leaders) and informal interviews with development agents and Woreda
137
agricultural experts were carried out.
138
2.3
Methods of data analysis
139
In each field, maximum development of rills, both in number and dimensions, was
140
attained by August 1, 2008, after which the rill dimensions did not show significant change
5
141
although there was still soil loss when rainfall occurred. Therefore, the maximum value was
142
analyzed in this paper to estimate the total soil loss due to rills. The eroded soil volumes, rill
143
densities, areas of actual damage and other quantities were calculated from the measured rill
144
length, width and depth (Herweg, 1996). The soil volume from a given field was calculated
145
using the following formula.
L W D146
N

X 
i
i
i
………………………………………….
(1)
i
10000 A
147
148
where X is the volume of rills in m3 ha-1, L is the rill length (m), W is the rill width (cm), D is
149
the rill depth (cm), A is the field area (ha), N is the number of rills, and i is the number of
150
homogeneous dimensions. The calculated volume is equivalent to the volume of soil lost due
151
to rill formation. The total volume of soil loss was obtained simply by summing the volumes
152
of all homogenous rill segments as shown in Equation 1. The eroded soil volume was also
153
expressed in terms of weight of eroded soil by multiplying the calculated volume by the
154
measured bulk density of the soils at each of the 15 fields in the site (Hagmann, 1996). The
155
total soil loss was converted into per unit hectare of land to express the annual rate of soil
156
loss. The area of actual damage per unit hectare (surface area covered by rills themselves)
157
was obtained from the product of length and width dimensions of each homogenous rill
158
segment by using Equation 2. The rill density was calculated by dividing the total rill lengths
159
(obtained by summing up the length measurements of all rills) by the total area of the
160
surveyed fields.
161
AAD 
 L W N
i
i
162
100 A
i
………………………………
(2)
163
where AAD is the area of actual damage by rills in m2 ha-1, L is the rill length (m), W is the
164
rill width (cm) , D is the rill density (m ha-1), A is the field area (ha), N is the number of rills,
165
and i is the number of homogeneous dimensions. The number of rills originating from upper
6
166
fields and rills initiated within the field and their contributions to soil loss were identified in
167
the study. The significant difference of rill magnitudes between each slope position and crop
168
cover types were analyzed statistically using SPSS software.
169
For social survey analysis, the data generated by the structured questionnaires was
170
analyzed using Microsoft Excel and SPSS software, version 12. The relevant qualitative
171
information generated by the informal discussions with farmers and other concernedgroups
172
were integrated with the quantitative data for a better understanding of the issues covered in
173
the study.
174
3
Results
175
3.1
The nature of rill erosion in differing slope positions
176
Rill erosion was the most visible mechanism of soil loss from sloping cultivated land
177
(Herweg 1996). The rate of soil loss due to rills from the downslope fields was 1.5 times
178
greater than that of midslope fields and 3.3 times greater than that of upslope fields (Table I).
179
ANOVA analysis indicated that
180
significantly differed among the three slope positions (Table I). In the downslope fields the
181
rills covered 884 m2/ha or almost 9% of the area. The average rill density before and after
182
sowing crops from the total rill length of 12,854m was 3,611 m/ha.
183
3.2
soil loss, area of actual damage and rill density all
Nature of rill erosion in crop types
184
Tef is the dominant cereal crop in the watershed, followed by maize, finger millet and
185
wheat. The amount of soil eroded from the tef plots after land preparation and prior to sowing
186
(July 11, 2008) was 8 t/ha, with an additional 26 t/ha lost after sowing (between July 12 and
187
August 27, 2008), yielding a total soil loss in tef fields wof 34 t/ha (Table II). The soil loss in
188
tef fields was three times greater than that of finger millet and wheat, and twice that of maize.
189
Erosion was low in wheat fields because the surface was rough both before and after sowing,
190
which increased infiltration and in turn decreased runoff. Moreover, wheat growth was more
7
191
rapid than other crops and a canopy cover was established before the soil roughness had been
192
dinminished due to high rainfall. ANOVA analysis in Table II idicated that the soil loss
193
differences among crop types was significant.
194
3.3
195
3.3.1 Perceptions of erosion as a problem
Farmer perception of land conservation
196
All the interviewed farmers perceived soil erosion as a problem constraining crop
197
production (Table III). They reported that the most important top soil for crop production
198
activity was deteriorating over time due to erosion. Hence, they frequently observed how the
199
loss of soil from cultivated fields has been reducing the depth of the topsoil and increasing
200
the number of stones in their farmlands over time. Moreover, the occurrence of unproductive
201
soil (locally called zinza or sometimes gel) in their farmland is used as an indicator of
202
decreasing soil depths.
203
The majority of the farmers reported that the occurrence of rill erosion was the
204
dominant erosion feature (89%) on their farmlands (Table III). This percentage of
205
respondents also compared rill erosion problem as the highest with other erosion features.
206
Gully erosion was also reported by 11% of farmers along their farm boundaries and
207
waterways. Surprisingly, no respondents perceived sheet erosion as a problem. From all
208
respondents, 64% of the farmers rated the extent of the problem as severe, and all
209
respondents mentioned that the rate of soil erosion has been increasing over time. All of them
210
also answered that erosion can be controlled.
211
3.3.2 Causes of soil erosion, soil fertility and productivity declines
212
As indicated in Table IV, the major cause of soil erosion mentioned by 98 % of
213
farmers was the lack of conservation structures. Indeed, the transect surveys in the entire
214
watershed confirmed that SWC technologies were poorly constructed and most of them are
215
damaged. The majority of the farmers (61%) believed that damaged structures contributed to
216
soil loss. Almost all of the interviewed farmers and those who participated in the group
8
217
discussion reported that there is a declining in soil fertility in their farm plots over time. All
218
the respondents suggested that soil erosion was the major cause of soil fertility decline.
219
Repeated cultivation (continuous cultivation without fallow) was also the other major causes
220
mentioned by 79% of the respondents. Almost all of the interviewed farmers reported
221
productivity decline. The respondents (83%) mentioned that soil erosion was the main cause
222
of productivity decline followed by loss of soil fertility (76%).
223
3.3.3 Farmers’ conservation practices in the watershed
224
Almost all of the farmers reported that SWC measures were very helpful for erosion
225
control and better crop production. Farmers commonly used to protect their farm lands from
226
soil erosion were stone terraces/bunds, cultural drainage ditches, soil bunds, waterways and
227
contour plowing.
228
Traditional ditches (locally called feses or shina) were indigenous practices widely
229
used by 75% of the surveyed farmers for erosion control (Table V). These are micro-channels
230
constructed on cultivated fields to direct excess water from cultivated fields. Most fields have
231
traditional ditches and farmers believed that these ditches are used to conserve soils, seeds
232
and fertilizers by decreasing the dimensions of rills.
233
Soil bunds were reported as the other means to conserve soil practiced by 61% of the
234
surveyed farmers. However, from field observation, soil bunds were rarely constructed and
235
poorly maintained. Permanent waterway structures constructed alongside cultivated fields
236
were used by 13% of the respondents. Farmers connect the ditches to waterways (which are
237
wider and deeper than the ditches and normally require maintenance) for the safe disposal of
238
excess water. Contour plowing was widely used in the area mentioned by 48% of the
239
respondents. This measure was commonly used not only to conserve soil from erosion but
240
also to decrease required traction power of animas during plowing. As SWC measure, it is an
241
efficient technique for reducing runoff mainly in moderately and gently sloping areas. On
9
242
steep slopes, as farmers noted, contour plowing alone may not be effective, requiring other
243
techniques such as bunds to effectively control erosion.
244
3.3.4 Farmer perception, acceptance and adoption of SWC measures
245
As it can be seen in Table VI, almost all respondents reported that the technologies
246
were effective in arresting soil erosion. Similarly, all respondents believed that the new SWC
247
technologies had the potential to improve land productivity. Farmers who tried to implement
248
some conservation measures in their plots were interviewed to determine how they measure
249
the effectiveness of SWC technologies. They had already observed better growth and
250
development of crops, particularly along the structures where fertile sediments were trapped.
251
They also evaluated that the amount of sediment trapped by the structure was very high and
252
would have been lost from the field had that conservation structure not been built. During
253
group discussion, participants who treated their lands by some conservation structures gave
254
witness to the group that the technology they have been using improved their land
255
productivity.
256
The farmers were also asked about their future intentions regarding using introduced
257
SWC technologies (Table VI). Almost all respondents expressed their willingness to continue
258
maintaining the established structures. In addition, of the farmers interviewed whether they
259
would like to implement the SWC technologies in the rest of their farm fields, all planned to
260
implement SWC measures on plots that were not treated at the time of the survey. To assess
261
their attitudes towards government support needs, farmers were asked whether they should be
262
paid for constructing and maintaining the SWC technology in their farm. The majority (85%)
263
responded “No” while the rest answered, “Yes”.
264
3.4.5 Factors Affecting Adoption of Introduced SWC Technologies
265
In the watershed, introduced soil and water conservation measures have rarely been
266
constructed. Though farmers indicated willingness to adopt the newly introduced SWC
267
structures, they did not practice these measures to their farmlands. From the interviewed
10
268
farmers, 87% reported that some conservation measures like terraces, bunds and fanyajuus
269
were time-consuming and labor-demanding for construction (Table VII). The other issue that
270
affected their conservation practices was lack of technical support (locally known as kiyessa)
271
from agricultural experts for constructing bunds and terraces (23%). In the group discussion
272
and informal interviews, lack of kiyessa was the a primary issue raised by the participants.
273
4
Discussion
274
The total amount of soil loss due to rills (before and after sowing) from all of the
275
surveyed fields was 78.2m3 (94.6 tones) with rate of 22m3 ha-1, which is equivalent to 27t ha-1
276
(Table I) of soil loss given the average soil bulk density of 1.21g cm-3. During the survey, the
277
contributions of other erosion features were not considered. However, rills are not the only
278
mechanisms for soil erosion and are always accompanied by impacts of raindrops such as
279
sheet or inter-rill erosion. According to Zachar (1982), rill erosion assessments underestimate
280
the actual soil loss by 10 to 30 %. Govers (1991) also reported that the contribution of inter-
281
rill erosion can be more than 30 % of the total soil loss in fields where rills are present.
282
Bewket and Sterk (2003) also assumed 30% of the actual soil loss to calculate the
283
contribution of inter-rill erosion to soil loss. If we assume for this study that the measured rill
284
erosion underestimates soil loss by 25%, then actual annual soil loss rates would be estimated
285
as around 36 t ha-1. This is equivalent to 3.6 mm yr-1, based on 1t ha-1 being equivalent to 0.1
286
mm yr-1 (Morgan, 1996; Tadesse, 2001). According to Basic et al. (2004), the erosion risk in
287
the watershed can be estimated by the formula:
288
Erosion risk = [Erosion rate (t/ha/yr)] ÷ [Soil loss tolerance]
(3)
289
Assuming the mean soil loss tolerance be 10 t ha-1, which was accepted as appropriate for
290
moderate thickness of soil (Morgan, 1996; Mwendera et al., 1997; Tadesse, 2001), then the
291
soil loss obtained from this study is nearly 70% greater (approximately fourfold of tolerable
11
292
soil loss). According to this assumption, the Debre Mewi watershed can be characterized as a
293
high erosion risk area.
294
The main reason for the high soil loss recorded in tef fields is that the period of land
295
preparation occurs during the middle of the rainy season with high intensity rains, while
296
preparation for the other crops tillage occurred earlier when rains were less intense. Another
297
contributing factor may be compaction (or trampling) by animal feet just before sowing tef.
298
In the upslope fields, very shallow rills were observed throughout the rainy season.
299
This was because (1) there was no any run-on coming from uplands as all fields in this
300
position are found in the edge of the watershed, and (2) the slope in these fields is very low
301
and infiltration is more rapid, hence the critical distances before concentration sufficient to
302
develop rills is longer than downslope fields. As described later in section 4.2, runoff from
303
upslope fields is an important contributor to rill formation in the more vulnerable downslope
304
fields. During the rainy season, these fields were highly saturated and infiltration rate was
305
relatively lower which resulted high concentration of runoff. Therefore, runoff in these fields
306
may be generated due to saturation excess. Due to these and other factors the downslope
307
position fields had the greatest rate of soil loss.
308
4.1
Effects of rainfall to soil erosion
309
The peak erosion losses occurred in the month of July which accounted for more than
310
90% soil loss, of which about 42% occurred at the beginning of rill measurements in early
311
July. This 42% soil loss corresponded to 18% of the rainfall recorded in July. This result was
312
similar to the findings of Herweg, et al. (2002As described in the Figure 3, after the first
313
measurement of rill volumes, the rate of soil loss was increasing but at a decreasing rate.
314
Only a few more rills were formed in the month of August. Thus, most erosion occurred
315
before the time of highest rainfall. Though the rainfall increased, the change in rill erosion
316
magnitude between each monitoring date was decreasing. Finally, redistribution of sediment
12
317
affected the rill dimensions and hence no significant change in magnitude of rill erosion was
318
observed starting from mid-August. This finding is in agreement with findings by Bewket
319
and Sterk (2003) and Herweg et al (2002).
320
Therefore, a large proportion of annual soil loss occurred during a few rainstorm
321
periods (Herweg and Ludi 1999), and thus the runoff generated from these periods was the
322
most important direct driver of severe soil erosion. Rills were initiated mostly by these few
323
destructive storms, but their continued growth throughout much of the wet season was the
324
result of the cumulative rainfall (Bewket and Sterk 2003). Therefore, the effectiveness of
325
SWC technology depends on the extent to which it can resist such ‘extreme’ rainstorm
326
periods.
327
4.2
Effects of run-on to soil erosion
328
Run-on (surface runoff coming from the upper fields), was the cause of rill formation
329
in the midslope and downslope fields. Surface runoff coming as a run-on had considerable
330
contributions to rill formation and then soil loss. The surveyed upslope fields did not receive
331
any surface runoff from the uplands due to the main road that crossed the slope and acted as a
332
divide. Therefore, the rills in the upslope positions were initiated by surface runoff generated
333
inside the fields, which created very small dimensions of rills.
334
In general, from the upper boundaries of 13 fields found in the mid-slope and down
335
slope positions of the study site, about 39.3% of the rills that contributed more than half
336
amount of soil loss (52.6%) were initiated due to run-on. This indicates a few number of rills
337
generated from run-on can induce a large amount of soil loss, which would be minimized if
338
there was appropriate conservation structures in the area. This result has important
339
implications in designing SWC measures and used as a starting point for the watershed
340
practitioners to plan what type and location of SWC measures are appropriate to reduce soil
341
erosion risk and to increase productivity.
13
342
4.3
Effects of crop cover on soil erosion
343
The role of crop cover is extensively studied in the literature. Cover reduces the
344
direct impact of raindrops on the soil, it increases the flow depth, infiltration and surface
345
roughness and it reduces the speed of runoff. Thus, cover reduces the amount of soil detached
346
by flowing water and the capacity of water flow to transport sediment (Mwendera et al. 1997,
347
Tamene and Vlek 2007). Therefore, the percentage cover of each crop was estimated
348
349
whenever each rill measurements were undertaken.
During field observation, the peak soil loss was measured when the canopy cover was
350
approximately zero. Though there were other relevant factors such as root density (effects not
351
known since they were not measured), plant cover was expected to play a primary role. The
352
effect of the temporal variation of rainfall and canopy coverage of the four crop types (tef,
353
wheat, millet and maize) on soil loss for individual storms is clear in Figure 4, where the soil
354
loss rate in all fields decreased slowly as the crop canopy cover increased. The correlation
355
coefficient between soil loss and cover percentage of tef, millet, and maize was -0.7; with
356
wheat the coefficient was -0.5. At the beginning of rill measurement, the coverage of tef and
357
wheat plots was zero whereas coverages on the other two crop plots were greater. As clearly
358
shown in the Figure 4, the soil loss in all crops decreased as coverage increased.
359
A unit increase of canopy coverage can bring about a much greater reduction in soil
360
loss only where the soil is properly managed, so the effectiveness of canopy coverage can be
361
greatly increased if it is combined with good soil management practices.
362
4.4
Farmer perceptions
363
The farmers in the study area have understanding about the problem of soil erosion
364
and fertility loss and the soil-stabilizing effects of conservation measures. They are able to
365
identify erosion and fertility loss indicators that are relevant, and sufficiently express the
366
degree and the nature of the problems in their area. Though detailed socio-economic analysis
367
was not undertaken, the result indicates that labor requirements and the lack of technical
14
368
support from concerned bodies influenced their conservation practices. We did not assess the
369
efforts of agricultural experts to motivate farmers to implement the conservation
370
technologies. Farmers suggested that they would be very motivated to adopt the technologies
371
if there were a cooperative work as was done in the past.
372
5
Conclusions and recommendations
373
The measurement and description of rill erosion was an important aspect of erosion
374
research in the Debre Mewi watershed. At the study site, highest erosion rates were observed
375
in early July, which can be attributed to the higher erosivity of rains, higher erodibility of the
376
soil surface after a warm and dry season, and the poor soil cover due to land preparation for
377
sowing. The results obtained from the surveyed fields indicated that about 3.6mm soil depth
378
is lost annually, which characterizes the watershed as a high erosion risk area.
379
Based on interviews and field observations, the intensity of the rainfall coupled with
380
poor vegetation cover has aggravated the soil erosion in the watershed. Hence, crop
381
production and soil productivity have been decreasing over time. Thus far, farmers undertake
382
little action to reduce erosion. Only few soil conservation structures accompanied with poor
383
management practices at farm hold level were observed frequently during the survey. If
384
nothing is done to correct the existing situation, more land will become unsuitable for crop
385
production, putting even more strain on existing resources. Therefore, sustainable soil
386
management systems must be developed to reduce further degradation and restore the
387
productivity of the eroded land.
388
The management options to reduce the soil loss are to implement conservation
389
measures and/or to change the crop to be planted. Changing to crops that need less tillage and
390
improve the soil structure can reduce the problem of erosion. Soil conservation planners
391
should find a mechanism in such a way that farmer can plant tef to earn higher profits while
392
minimizing its contribution to erosion, which would need further investigation. Cultural
15
393
ditches currently implemented by farmers are generally effective and should be an integral
394
part of any soil and water management practices proposed by soil conservation designers.
395
Though farmers have an awareness of soil degradation and knowledge of possible
396
remedial measures, they should be motivated to adapt the introduced SWC technologies
397
based on their indigenous knowledge with relevant scientific advice. During discussions with
398
local farmers, they showed great interest in practicing terraces/bunds and grass strips and
399
plantation in their croplands. However, accessibility of grasses and trees seedlings impairs
400
their capacity to alleviate the problem. The authors hope that Adet Agricultural Research
401
Center together with the local office of Agriculture can help farmers in this regard. Generally,
402
the soil degradation problem in the area is complex and encompasses environmental,
403
economic, social and political issues. Therefore, detail investigations should be undertaken
404
through the collaboration of stakeholders to identify better management options that decrease
405
soil erosion and improve food productivity.
406
Acknowledgements
407
This research was made possible by funds provided by Cornell University and logical
408
and technical support given by personnel of SWISHA, ARARI and IWMI’s CP19 project.
409
We are especially grateful to the landholders for permission to carry out the rill
410
measurements in their cropped fields and their time in answering the questionnaire. Finally,
411
Adet Agricultural Research Center is gratefully acknowledged for their overall support of the
412
project.
413
16
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
References
Amsalu, A., Graaff, J.D. 2006. Determinants of adoption and continued use of stone terraces
for soil and water conservation in an Ethiopian highlands. Ecological economics 61,
294-302.
Basic, F., Kisic, I., Mesic, M., Nestroy, O., Butorac, A. 2004. Tillage and crop management
effects on soil erosion in central Croatia. Soil & Tillage Research 78, 197–206.
Bekele W., Drake, L. 2003. Soil and water conservation decision behavior of subsistence
farmers in the eastern highlands of Ethiopia: a case study of the Hunde-Lafto area.
Ecological Economics 46, 437–451.
Belyaev, V., Wallbrink, P.J., Golosov, V., Murray A.S., Sidorchuk, A .2004. A comparison
of methods for evaluating soil redistribution in the severely eroded Stavropol region.
Southern European Russia, Faculty of Geography, Moscow State University, Russia.
Beshah, T. 2003. Understanding farmers: explaining soil and water conservation in
Konso,Wolaita and Wello, Ethiopia. Tropical Resource Management Papers 41, 245.
Wageningen University.
Bewket, W. 2002. Land Cover Dynamics since the 1950s in Chemoga watershed, Blue Nile
Basin, Ethiopia. Mountain Research and Development 22, 3, 263-269.
Bewket, W., Sterk, G. 2003. Assessment of soil erosion in cultivated fields using a survey
methodology for rills in the Chemoga watershed, Ethiopia. Agriculture, Ecosystems
and Environment 97, 81–93.
Bewket, W., Sterk, G. 2005. Dynamics in land cover and its effect on stream flow in the
Chemoga watershed, Blue Nile basin, Ethiopia. Hydrological Process 19, 445–458.
Bewket, W., 2007. Soil and water conservation intervention with conventional technologies
in North western highlands of Ethiopia: Acceptance and adoption by farmers. Land
use policy papers 24, 404–416, Department of Geography and Environmental Studies,
Addis Ababa University.
Brazier, R. 2004. Quantifying soil erosion by water in the UK: a review of monitoring and
modeling approaches. Progress in Physical Geography 28, 3, 340–365.
Casalí, J., Loizu, J. Campo, M.A., De Santisteban L.M., Álvarez-Mozos J. 2006. Accuracy
of methods for field assessment of rill and ephemeral gully erosion. Public University
of Navarre, Department of Projects and Rural Engineering, Campus de Arrosadia s/n,
31006 Pamplona, Navarre, Spain.
Feoli, E., Laura, G. V., Woldu, Z. 2002. Evaluation of environmental degradation in northern
Ethiopia using GIS to integrate vegetation, geomorphological, erosion and socioeconomic factors. Agriculture, Ecosystems and Environment 91, 313–325.
17
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
Gebremedhin, B., Swinton, S.M. 2003. Investment in soil conservation in northern Ethiopia:
the role of land tenure security and public programs. Agricultural Economics 29, 69–
84.
Govers, G. 1991. Rill erosion on arable land in central Belgium: Rates, controls, and
predictability. Catena 18, 133-155.
Graaff, J.D. 1993. Soil conservation and sustainable land use: An economic approach Royal
Tropical Institute, Amsterdam. Ethiopia: The agricultural sector: an overview. Vols. II
and I.Rome: FAO.
Hagmann, J. 1996. Mechanical soil conservation with contour ridges: cure for, or cause of,
rill erosion? Land degradation & development 7,145-160.
Herweg, K., 1996. Field manual for assessment of current erosion damage. Soil conservation
research programme (SCRP), Ethiopia and centre for development and environment
(CDE), University of Berne, Switzerland.
Herweg, K., Ludi E. 1999. The performance of selected soil and water conservation
measures, case studies from Ethiopia and Eritrea. Catena 36, 99–114.
Herweg, K., Stillhardt, B., Krauer, J., Frey, L.,
http://www.sfiar.infoagrar.ch/documents/posters/frey.pdf.
Hurni,
H.
2002.
Hudson, N.W. 1993. Field measurement of soil erosion and runoff. Food and Agricultural
Organization (FAO) of the united nations Ampthil, United Kingdom.
Hurni, H., Kebede, T., Zeleke, G. 2005. Implications of Changes in Population, Land Use and
Land Management for surface runoff in the upper basin area of Ethiopia. Mountain
Research and Development 25, 2, 147-154
Mitiku, H., Herweg, K., Stillhardt, B. 2006. Sustainable land Management- a New Approach
to Soil and Water Conservation in Ethiopia.pp.79-89, Mekele University, Ethiopia.
Moges, A., Holden, N.M. 2006. Farmers’ perceptions of soil erosion and soil fertility loss in
southern Ethiopia. Land Degrad. Develop. (in press). Published online in Wiley Inter
Science (www.interscience.wiley.com) DOI: 10.1002/ldr.795.
Mohammad, A., Jabbar, M.A., Mohamed, S., Hugo, L. 2001. Towards transdisciplinarity in
technology and resource management research: a project in Ethiopia. Outlook on
Agriculture 30, 4, 257–260.
Morgan, R.P.C. 1996. Soil Erosion and Conservation, second ed. Longman Silsoe College,
Cranfield University.
Mwendera, E., Jabbar, M.A., Mohamed, S. 1997. Infiltration rates, surface runoff, and soil
loss as influenced by grazing pressure in the Ethiopian highlands. Soil Use and
Management 13, 29-35.
18
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
Nyssen, J., Poesena, J., Moeyersonsc, J., Deckersd, J., Mitiku, H., Andreas, L. 2004. Human
impact on the environment in the Ethiopian and Eritrean highlands. A state of the art
Earth-Science Reviews 64, 273–320.
Shiferaw, B., Holden, S. 1998. Resource degradation and adoption of land conservation
technologies in the Ethiopian highlands: a case study in Andit Tid, North Shewa.
Agric. Econ. 18, 233–247.
Shiferaw, B., Holden, S. 1999. Soil Erosion and Smallholders' Conservation Decisions in the
Highlands of Ethiopia. World Development 27, 4, 739 – 752.
Shiferaw B., Stein T. Holden, 2000. Policy instruments for sustainable land management: the
case of highland smallholders in Ethiopia. Agricultural Economics 22, 217–232.
Sonneveld, B.G.J.S. 2002. Land Under Pressure: The Impact of Water Erosion on Food
Production in Ethiopia. Shaker Publishing, Maastricht, the Netherlands.
Sonneveld, B. G. J. S., Keyzer, M. A. 2003. Land under pressure: soil conservation concerns
and opportunities for Ethiopia. Land Degrad. Develop. 14, 5–23
Tadesse, G. 2001. Land degradation: a challenge to Ethiopia. Environmental Management 27,
815–824.
Tamene, L., Vlek, P. L. G. 2007. Assessing the potential of changing land use for reducing
soil erosion and sediment yield of catchments: a case study in the highlands of
northern Ethiopia, Soil Use and Management 23, 82–91.
Zancher, D. 1982. Soil Erosion, Developments in Soil Sciences, Elsevier, Amsterdam.
541
542
543
544
545
546
19
547
List of Figures and Tables
548
Figure 1:
Location of the area of Debre Mewi watershed, Ethiopia
549
Figure 2:
Photographs of rill dimension measurements: depth on the left and width on
550
551
the right side
Figure 3:
552
553
Average soil loss due to rills for the 15 upland agricultural fields in the DebreMewi watershed
Figure 4:
Graph of crop cover versus soil loss rate
Table I:
Rill magnitudes in three different slope positions. Values followed by
554
555
556
differing letters in a given column were significantly different (p=0.05).
557
Table II:
Rill magnitudes in different crop types
558
Table III:
Perception of respondents for soil erosion as a problem
559
Table IV:
Ranked farmer responses to the causes of soil erosion, fertility and
560
561
productivity decline (Totals may exceed 100% due to multiple responses)
Table V:
562
Farmer conservation practices in the Debre Mewi watershed, Ethiopia. (Totals
may exceed 100% due to multiple responses)
563
Table VI:
Indicators of acceptance and adoption of SWC technologies
564
Table VII:
Farmers’ reasons for not adopting the newly introduced SWC measures
565
(Totals may exceed 100% due to multiple responses)
566
567
568
20
569
570
571
572
573
574
575
576
577
578
Figure 2: Location of the area of Debre Mewi watershed, Ethiopia
21
579
580
581
582
583
584
Figure 2: Photographs of rill dimension measurements: depth on the left and width on the
right side
22
585
586
Soil loss in each obs.(t/ha)
25
1400
20
1200
1000
15
800
10
600
5
400
ep
t
S
ug
us
t
0
A
-5
200
Ju
ly
Ju
ne
0
587
588
589
590
RF (mm)
Rain fall
Soil loss (t/ha)
Com.soil loss (t/ha)
Figure 3: Average soil loss due to rills for the 15 upland agricultural fields in the DebreMewi watershed.
23
591
592
maize
wheat
millet
tef
soil loss (t/ha)
25.00
20.00
15.00
10.00
5.00
0.00
0
20
40
60
cover crop (%)
593
594
595
Figure 4: Graph of crop cover versus soil loss rate
24
80
100
596
597
598
599
Table VIII: Rill magnitudes in three different slope positions. Values followed by differing
letters in a given column were significantly different (p=0.05).
Slope
Soil erosion effects
Position
Rill
Soil loss
Area of actual damage
Rill density
count
(t/ha)
(m2/ha)
(m/ha)
Upslope
105
8a
256a
1029a
Midslope
453
23b
662b
2860b
Downslope
959
34c
884c
4946c
Average
1517
27
717
3611
600
601
25
602
603
604
Table IX: Rill magnitudes in different crop types.
AAD (m2-ha)
Crop type No of rills Soil loss (t-ha)
Rill density (m-ha)
Maize
48
14a
373a
2317a
Wheat
14
7.6b
169a
832b
Millet
131
7.7b
251a
977b
Tef
1324
34c
902b
4561c
Average
1517
27
717
3611
605
606
607
608
609
610
26
611
Table X: Perception of respondents for soil erosion as a problem
Farmers’ responses to:
Options
Percentage (n=80)
Occurrence of soil erosion Yes
100
No
0
Prevailing form of erosion
Extent of soil erosion
Rill erosion compared to
other erosion features
The rate of erosion over
time
Can soil erosion be
controlled?
Sheet erosion
Rill erosion
Gulley erosion
severe
Moderate
Minor
highest
medium
low
Increasing
Same
Decreasing
Yes
No
612
613
27
0
89
11
64
36
89
10
1
100
0
0
100
0
614
615
Table XI: Ranked farmer responses to the causes of soil erosion, fertility and productivity
decline (Totals may exceed 100% due to multiple responses)
Farmers’ response Options
Percentage
to:
n=80
Causes of soil
Lack of conservation structures
98
erosion
Damaged conservation structures
61
Steep land w/o conservation structures
10
Lack of diversion ditch
5
Other
Causes for soil
Soil erosion
100
fertility declining
Repeated cultivation
79
Lack of manure
9
Lack of fertilizer
0
Other
0
Causes of crop
Soil erosion
83
productivity
Fertility decline
76
decline
Continuous cultivation
31
Rainfall shortage
11
Other
3
616
617
618
619
620
621
622
623
624
625
626
28
627
628
629
Table XII: Farmer conservation practices in the Debre Mewi watershed, Ethiopia. (Totals
may exceed 100% due to multiple responses)
Management options
%, n=80
Cultural ditches
75
Soil bunds
61
Contour plowing
48
Waterways
13
Stone bunds
1
Grass strips
0
Others
0
630
631
29
632
Table XIII: Indicators of acceptance and adoption of SWC technologies
Farmers’ response
Options
Yes
No
to:
%
%
Indicators of
Their knowledge of SWC measures
99
1
acceptance
Effectiveness of SWC in arresting soil erosion
99
1
SWC have a potential to improve land
100
0
productivity
Indicators of
Plan to implement the new SWC tech.
99
1
adoption
Plan to maintain the constructed structures
100
0
Farmers should be paid for constructing and
15
85
maintaining SWC in their farm
633
634
635
636
637
638
639
640
641
642
643
644
30
645
646
647
Table XIV: Farmers’ reasons for not adopting the newly introduced SWC measures (Totals
may exceed 100% due to multiple responses)
Options
%, n = 80
Requires too much labor to implement
88
Decrease farm size and difficult to plow
19
Harbor mole-rats
11
Lack of knowledge
10
Land insecurity
5
Not considering erosion as a problem
0
Other (lack of technical support)
23
648
649
31