Previous PageTable Of ContentsNext Page

5    Land degradation and its impact in Amhara Region


5.1 Soil erosion

5.2 Nutrient depletion

5.3 Deforestation

5.4 Impact of land degradation


Land degradation in the Ethiopian highlands (i.e. areas above 1500 m a.s.l.) has been a concern for many years. Soil erosion, nutrient depletion and deforestation are common, but little has been done to determine their impact on productivity.

5.1 Soil erosion

Loss of arable land due to soil erosion is a widespread phenomenon in the highlands, which account for about 45% of Ethiopia's total land area and about 66% of the total land area of Amhara Region. On steep hillsides, soil losses of and exceeding 200 t/ha per year have been recorded (Kappel 1996). The potential threat of land degradation to the country's fragile economy and food security has been emphasised by several publications (e.g. Wright and Adamseged 1986; Hurni 1988; MNREP 1994: Abegaz Gizachew 1995; Kappel 1996; BoA 1997; NEC 1997; FEC 1998). The threat is credible as about 90% of the population of the Amhara Region lives in the highlands and 90% of the regularly cropped land is found there.

Soil erosion by water is the dominant form of erosion. The areas that are severely affected can be found in Wag Hemra and North Wello followed by North and South Gonder, eastern parts of South Wello and northern parts of North Shewa zones. The soil depth in these places is very shallow (Leptosols), soil fertility is poor and farmers squeeze a living from pockets of shallow soils. Gullies are a frequent and permanent phenomenon everywhere in the region. According to CoSAERAR (1997), Kobo, Gubalafto and Habru weredas (all of North Wello) have lost 3700 ha of their 284,950 ha total land area to gullies. In addition to reducing cultivable area, soil erosion and gully formation and expansion reduce the water holding capacity of the soil and, consequently, result in poor crop yields.

The Ethiopian Highland Reclamation Study (EHRS) has developed a 1:1,000,000 scale soil loss rate map, which shows the types of soil degradation processes, causes, severity and extent . The map is based on the universal soil loss equation (USLE) and soil-erodibility and land use maps. EHRS assesses the national soil loss rate as 'moderate to high', which is estimated at 30–100 t/ha per year (Wright and Yeshinegus Adamseged 1986). Depending on land use practices, however, the real rate is claimed to be <2 to >300 t/ha per year (Wright and Yeshinegus Adamseged 1986). Based on the EHRS calculations, the region's soil loss rate is estimated at about 58% of the national rate (BoA 1997; CEDEP 1999). Thus, given that the spatial coverage of the region is only about one-sixth of the nation, the soil loss rate per unit area is very high in the region, compared with other regions. The USLE over estimates net soil losses, since it accounts for erosion off plots above but not sedimentation on plots below. Therefore, the net soil loss may be significantly lower than estimated and, consequently, needs to be interpreted with caution. For example, the Soil Conservation Research Project (SCRP) estimates the national soil loss rate at about 21% less than the EHRS estimate (Dawit Kebede 1996). The study, therefore, concludes that the soil loss from croplands in the highlands of Amhara is much less than that estimated by the EHRS. Table 4 shows that about 29% of the total area of the region experiences a high erosion hazard (between 51 and 200 t/ha per year) and 31% exhibits a moderate erosion hazard (16–50 t/ha per year). Although, the highest soil loss rate (>200 t/ha per year) occurs in only 10% of the region, it is estimated that this contributes almost 50% of the total soil mass that is moved (CEDEP 1999).

Table 4. Estimated erosion hazard classes in Amhara Region.

Erosion classes

Range of soil loss rate
(t/ha per year)

Area coverage

ha (× 103)

Percentage
 (%)

Very high

>200

1660

10

High

51–200

4796

29

Moderate

16–50

5284

31

Slight

0–15

5020

30

Total

9–300

16,760

100

Source: Abegaz Gizachew (1995).

The FEC (1998) study indicates that the plain areas of the highlands in Gojam, Awi and the south-eastern parts of the region have a moderate soil erosion hazard. However, as these are currently the most productive and surplus-producing areas of the region, due to the ample rainfall and gentler slopes, these estimates should not be interpreted as though land degradation is not problematic but, rather, as an indication for remedy. Another relevant study conducted in the region is NEC (1997). This study was a soil survey on erosion hazard assessment in the northern parts of the region covering about 56% of the total land area. In the study, 4 zones and 23 weredas were identified as soil and water conservation implementation priority areas.

Site-specific test plots and experiments in 1987 and 1988 at SCRP stations in the region show soil loss rates between 0.04 and 212 t/ha per year, which depend on the soil type, slope, vegetation and type of conservation structure (see Tables 5 and 6). Grass cover tremendously reduces soil loss, as runoff is significantly diminished, implying that more water finds its way into the soil. Land under cereal cultivation also had higher soil loss rates than land under legume cultivation (Table 5) and conservation structures significantly reduced soil loss, with fanya juu7 terraces performing better than graded bunds, but not better than grass strips (Table 6).

7. Fanya juu is a type of terrace adopted from Kenya. In Swahili, fanya means 'throw' while juu means 'up'. It thus means, 'throwing up the slope' as opposed to 'throwing down the slope' in the conventional soil bund construction. With fanya juu, less cultivable land is taken up by the structure and benching is faster than the conventional soil bund; however, fanya juu requires a higher amount of labour.

Table 5. Soil loss estimates for Soil Conservation Research Project (SCRP) sites in Amhara Region, 1987 and 1988.

Year

SCRP site

Slope (%)

Soil type

Crop

Runoff (mm)

Soil loss rate (t/ha per year)

1987

Maybar

16

Haplic phaeozem

Maize

63.6

4.5

   

37

Stony haplic phaeozem

Barley/horse bean

50.4

11.2

   

43

Haplic phaeozem-lithosol

Grass

21.9

0.04

   

64

Haplic phaeozem-lithosol

Grass

13.9

0.1

 

Andit Tid

23

Eutric regosol

Lentil

273.3

121.8

   

39

Chromic cambisol

Lentil

321.2

160.1

   

48

Eutric regosol

Fallow

222.9

69.1

   

48

Orchic andosol

Lentil

115.3

42.1

 

Anjeni

12

Eutric regosol

Teff

836.0

175.6

   

16

Eutric regosol

Grass

627.5

14.4

   

22

Stony eutric regosol

Wheat

692.2

184.2

   

28

Vertic luvisol

Barley/gibto

989.2

210.9

1988

Maybar

16

Haplic phaeozem

Barley/beans

569.9

36.0

   

37

Stony haplic phaeozem

Wheat/beans

374.3

53.8

   

43

Haplic phaeozem-lithosol

Grass

28.2

0.5

   

64

Haplic phaeozem-lithosol

Grass

21.4

0.2

 

Andit Tid

23

Eutric regosol

Wheat

583.5

212.4

   

39

Chromic cambisol

Barley

703.6

199.5

   

48

Eutric regosol

Fallow

586.1

142.6

   

48

Orchic andosol

Barley

382.9

152.4

 

Anjeni

12

Eutric regosol

Beans

836.7

40.2

   

16

Eutric regosol

Grass

620.4

1.6

   

22

Stony eutric regosol

Grass

690.7

73.6

   

28

Vertic luvisol

Wheat

840.2

199.2

Source: Grunder and Herweg (1991a, 1991b).

Table 6. Soil loss estimates for Soil Conservation Research Project (SCRP) experiments in Amhara Region, 1987 and 1988.

Year

SCRP site and conditions

Conservation structure

Runoff 
(mm)

Soil loss rate 
(t/ha per year)

1987

Maybar (slope = 28%; soil = Haplic phaeozem; crop = Horse, bean/ barley/wheat)

Grass strip
Graded fanya juu
Graded bund
Control

7.7
10.2
12.1
15.6

0.4
1.2
1.7
1.1

 

Andit Tid (slope = 24%; soil = Eutric regosol; crop = Barley/lentil)

Grass strip
Graded fanya juu
Graded bund
Control

57.3
113.7
79.9
182.8

3.9
15.0
14.6
43.4

 

Anjeni (slope = 28%; soil = Vertic luvisol crop = Barley/gibto)

Grass strip
Graded fanya juu
Graded bund
Control


441.2
495.8
595.6


40.1
45.6
144.0

1988

Maybar (slope = 28%; soil = Haplic phaeozem; crop = Maize)

Grass strip 
Graded fanya juu
Graded bund
Control

19.7
15.9
15.3
22.0

0.5
0.2
0.5
0.7

 

Andit Tid (slope = 24%; soil = Eutric regosol; crop = Peas/beans)

Grass strip 
Graded fanya juu
Graded bund
Control

607.5
651.3
705.5
685.4

51.6
54.8
85.6
140.8

 

Anjeni (slope = 28%; soil = Vertic luvisol; crop = Wheat)

Grass strip 
Graded fanya juu 
Graded bund 
Control


282.0
345.5
481.8


36.1
42.0
104.0

Source: Grunder and Herweg (1991a, 1991b).

5.2 Nutrient depletion

Loss of fertility is manifested through using dung and crop residues as household fuels and animal feeds, low use of chemical fertilisers, declining fallow periods, soil and organic matter burning (guie), and soil erosion. Even though the farming system in the highlands of Amhara is mixed crop–livestock, nutrient flows between the two are predominantly one sided, with feeding of crop residues to livestock but little or no dung being returned to the soil. For example, in a household level socio-economic survey of the Amhara Region, even though almost all households (90%) fed crop residues to their livestock, only 40% used manure on their farmlands (UNECA 1996). This phenomenon is common in most of the highlands of Ethiopia, where the nutrient balance is highly negative. Deficits of more than 100 kg of nitrogen/ha per year have been reported for the highlands of Ethiopia, compared with only 15 kg of nitrogen/ha per year in Mali (de Wit et al. 1996, cited in Steinfeld et al. 1998). Furthermore, estimates of soil nutrient loss in Ethiopia show a net removal of 41 kg of nitrogen/ha of agricultural land between 1982 and 1984, and losses projected to reach 47 kg/ha by the year 2000 (Stoorvogel et al. 1993).

5.3 Deforestation

Excess removal of forests is contributing to land degradation. For example, based on population growth (demand) and forest increment (supply), the region recorded a deficit of about 16.6 million cubic metres of wood for fuel and construction in 1996 alone (BoA 1997). About 20 thousand hectares of forest are harvested annually in the Amhara Region for fuelwood, logging and construction purposes. Since harvested trees are not replaced adequately by tree planting, soils are exposed to high intensity of rainfall and about 1.9 to 3.5 billion tonnes of fertile topsoil are washed away annually into rivers and lakes due to deforestation alone (BoA 1997). While there is growing interest by farmers in planting eucalyptus trees, there is a strong debate as to whether eucalyptus should be planted on farmlands or not. Eucalyptus is believed to have a negative soil-moisture impact on neighbouring fields and renders the particular field unsuitable for future crop production. Consequently, current planting is limited mainly to homestead and field boundaries. Reliable data on area coverage, number of trees and annual growth rate of eucalyptus in the region are not available (BoA 1997; AFAP 1999). However, it is argued by many that, existing estimates are conservative. Thus, there is need to update most of the information for proper and effective policy intervention.8

8. The Woody Biomass Project is currently making an inventory of trees of different types in the Amhara Region.

5.4 Impact of land degradation

While a substantial number of research studies in the region (e.g. Yohannes Gebremichael 1989; Kassaye Goshu 1997; Ludi 1997; Azene Bekele 1998; Tamiru Sebsbe 1998; Zealbowesen Asfaw 1998) have examined the rate of land degradation, very few have looked at the impact on productivity. SCRP has monitored soil erosion and impact of many physical and biological soil and water conservation practices in the three research stations of the region (Maybar, Andit Tid and Anjeni). Thus, while ample biophysical data have been generated and are available, little quantitative analysis for policy action has been carried out.

The net soil loss from erosion, as reviewed by Kappel (1996), is estimated to range from 20 to 100 t/ha per year, with an annual productivity loss on cropland of 0.1% to 2% of total production for the country (see Table 7). Commenting on the wide variation of the estimates, Kappel (1996) argued that the economic implications of soil erosion in Ethiopia are neither as catastrophic as commonly believed nor negligible. Sutcliffe (1993, cited in Kappel 1996) and Bojo and Cassells (1994, cited in Kappel 1996) argued that soil fertility depletion may be a larger problem than erosion in the Ethiopian highlands.

Table 7. Soil erosion and productivity loss in Ethiopia.
 

Wright and Yeshinegus Adamseged (1986)

Hurni 
(1988)

Sutcliffe 
(1993, cited in Kappel 1996)

Bojo and Cassells 
(1994, cited in Kappel 1996)

Soil loss (t/ha per year)

Gross

130

n.e.

n.e.

40

Net

100

42

45

20

Productivity loss (% of total output per annum)

Potential

n.e.

n.e.

0.7

0.4

Effective

1.8

2.0

0.21

0.12

n.e. means not estimated.
Source: Kappel (1996).

Using 1994 prices, an annual yield loss of 1% of 1992 production was valued at about US$ 7.5 million (Kappel 1996). Belay Tegene (1992), at Gununo experimental station, an SCRP site, showed that there is a high positive correlation (r = 0.92) between topsoil depth and maize yield.

As vegetation cover and quality of soil resources deteriorate, the quantity and quality of water resources also deteriorates. A report on the water supply scheme of rural communities in 6 zones revealed that no more than 2.3% of the people had access to safe drinking water (BoA 1997). Many perennial springs and streams had become seasonal. Monitoring the impact of deforestation, soil erosion and conservation efforts on streams and rivers in particular and bodies of water in general is virtually non-existent. The negative impact of sedimentation on irrigation projects in the absence of catchment treatment is evident at the Borkena and Angereb water resources development projects in South Wello and North Gonder zones, respectively.

The Ethiopian Forestry Action Programme (MNREP 1994) and the AFAP (1999) also describe the land degradation problem in the country and region, respectively, its causes, impacts, and policy and institutional implications. However, these studies are based on land use in 1984 and, thus, the current situation is likely to be different, due to land-use dynamics. Nevertheless, a worsening scenario is envisaged as increasing amounts of forest, bush, grazing land and marshy areas are brought under intensive cultivation to meet the increase in food demand due to increasing population pressure. Already, degraded steep slopes are grazed continuously and are not allowed to regenerate. This phenomenon is observed especially in South Gonder, and North and South Wello zones, where wet and marshy (riparian) areas, which were used previously for grazing, are increasingly being drained for cropping.

Previous PageTop Of PageNext Page