ArticlePDF Available

Cost of Land Degradation in Ethiopia: A Critical Review of Past Studies

Cost of Land Degradation in Ethiopia:
A Critical Review of Past Studies
Mahmud Yesuf
, Alemu Mekonnen
, Menale Kassie
, and John Pender
December 2005
Environmental Economics Policy Forum in Ethiopia
International Food Policy Research Institute
The authors are grateful to the World Bank for its financial support of this research; to
Gunnar Kohlin, Jan Bojö, Ernst Lutz, Sergio Margulis, Josef Loening, Stefano Pagiola, Ben
Sonneveld, and others who provided advice on the research and access to valuable documents
and sources of data. The authors are solely responsible for any errors or omissions.
1. Introduction
Over the last few decades, there have been several efforts to evaluate the cost of land
degradation in Ethiopia. These studies have reached widely varying conclusions concerning
the nature, extent and economic cost of the land degradation problem. The diversity in
methodological approaches and underlying assumptions used by these studies, contributes to
this diversity of conclusions. This, together with the fact that the costs reported by different
studies are not always reported in a manner that allows comparisons across them to be made,
limits the ability to draw clear implications for policy and program decisions related to
sustainable land management.
To address this problem, this study will review the past studies of the cost of land degradation
in Ethiopia, assess the major methodological and conceptual issues and problems existing in
the different approaches, compare the findings across these studies considering the relative
merits of the different approaches, and draw implications for policies and programs, as well
as for future research related to land management in Ethiopia.
The major objectives of this review are, therefore, the following:
To review the major methodological and conceptual issues involved in estimating the
costs of land degradation;
To critically review the methods, assumptions and drawbacks of past estimates of the
cost of land degradation in Ethiopia;
To compare the estimates across the available studies and compare those with other
estimates for sub-Saharan Africa and other developing countries.
The rest of the paper is organized as follows. In section 2, we present major theoretical and
conceptual issues related to estimating the cost of land degradation. In section 3, past studies
estimating the cost of land degradation in Ethiopia are critically reviewed. In section 4 we
compare the past estimates of land degradation in Ethiopia, considering the reliability of the
findings and how these compare to other estimates of costs of land degradation in other
countries of sub-Saharan Africa. In section 5, we draw conclusions and implications.
2. Methodological and conceptual issues in cost estimation of land degradation
Before entering into the assessment of previous studies of cost of land degradation in
Ethiopia, there is a need to clarify the major conceptual and methodological issues that are
raised by attempts to estimate the cost of land degradation.
2.1. Definitional problems/confusions
It is important to define and distinguish related concepts such as land degradation, soil
degradation, and soil erosion, as these are sometimes (incorrectly) used interchangeably.
Land degradation is a broad term, reflecting the fact that land itself is a broad term, including
more than just the soil. The U.N. Convention to Combat Desertification defines land as “the
terrestrial bio-productive system that comprises soil, vegetation, other biota, and the
ecological and hydrological processes that operate within the system”, and land degradation
as “reduction or loss . . . of the biological or economic productivity and complexity of rainfed
cropland, irrigated cropland, or range, pasture, forest and woodlands resulting from . . .
processes . . . such as (i) soil erosion caused by wind and/or water; (ii) deterioration of the
physical, biological or economic properties of the soil; and (iii) long-term loss of natural
vegetation” (Pagiola 1999).
Soil degradation is a narrower term for declining soil quality, encompassing the deterioration
in physical, chemical and biological attributes of the soil (Enters 1998). Soil erosion is a
particular physical process that causes land and soil degradation, and refers to the wearing
away of the land surface by water and/or wind as well as to the reduction in soil productivity
due to physical loss of topsoil, reduction in rooting depth, removal of plant nutrients, and loss
of water (Ibid.). Other forms of soil degradation include other types of physical degradation,
such as compaction, surface sealing and crusting, waterlogging and aridification; chemical
degradation, including depletion of soil nutrients, acidification, salinization, and pollution;
and biological degradation, including loss of soil organic matter (which also affects physical
and chemical properties of the soil), flora and fauna populations or species in the soil (e.g.,
earthworms, termites and microorganisms) (Scherr 1999).
Land degradation is a complex phenomenon influenced by natural and socio-economic
factors. Most cost estimates of land degradation do not distinguish between soil erosion, soil
degradation and land degradation in their analyses. In many economic analyses, there is a
tendency to attribute soil fertility decline only to soil erosion. Erosion is treated as the sole
contributing factor to soil/land degradation and yield declines, as the impacts of nutrient
depletion on crop yields are underestimated or completely neglected. In fact, soil erosion
could be a major component of on-site costs, particularly on steeper slopes (Kerr and Pender,
2005). However, in many developing countries, especially in Africa, soil nutrient mining is
also often identified as a very serious problem (Stoorvogel, et al. 1993; Bojö and Cassells,
1995). Research on soil nutrient balances in Africa highlights that besides erosion, soil
fertility is also reduced by the removal of harvested crop parts and residues, leaching and
volatilization/denitrification losses (Stoorvogel et al, 1993; Smaling et al, 1996). The fact that
different studies estimate different components of land degradation makes comparison of cost
estimates across studies a difficult task.
Understanding the contribution of each component of land degradation to the total cost of
land degradation is vital, not only for appreciating the type and degree of the problem but
also for pinpointing appropriate policy interventions.
2.2. Estimating soil erosion rates
Soil erosion is a complex phenomenon influenced by many natural and socio-economic
factors. Soil erosion is a flow concept and is measured as a rate; e.g., x tonnes/ha/year. Soil
erosion includes a range of phenomena such as sheet erosion (the removal of thin layers of
topsoil from an eroded site), rill and gully erosion (the formation of small incised rills or
large gullies on a slope), mass wasting (the structural failure of part of a hillside as in the case
of a landslide); wind erosion, tunnel erosion (caused by percolation of groundwater through a
porous mantle) and fluvial (stream bank) erosion (Wright and Adamseged 1984). Often gully
erosion and mass-wasting contribute more to erosion than surface erosion phenomena such as
sheet erosion in uplands and highlands, even without human influences (El-Swaify, 1997).
The Universal Soil Loss Equation (USLE) is a widely applied methodology used to estimate
sheet and rill erosion rates. The USLE, which was first developed by Wischmeier and Smith
(1978), is given by the following equation:
),,,( LTSCfE =
E is the average annual erosion
C is the climatic factor (rainfall, wind, etc.)
S is the soil factor (soil erodibility of a specific soil type)
T is the topography factor (slope and length of the field)
L is the land utilisation factor (cropping management and erosion control practices).
Due to its relatively modest data demands and transparent model structure, the USLE remains
the most popular tool for estimating water erosion (sheet and rill erosion) rates. However,
although it is designed to predict long term average annual soil erosion for alternative
combinations of crop systems and management practices in association with specified soil
types, rainfall patterns, and topography; it does not predict soil deposition or sediment yield
from gully or channel erosion. Furthermore, most of the USLE data are from the croplands.
Extension of the USLE to range lands, forest lands or other sites without an extensive data
base is highly controversial (Stocking, 1995). Furthermore, since it was developed for soils
and conditions in the United States, we should expect errors if applied in the tropics (Ibid).
To estimate the economic significance of land degradation, it is necessary to develop models
that link the soil erosion rates and other aspects of land degradation to the changes in crop
production and farming systems. The most common approaches in the literature are discussed
in the next section.
2.3. Approaches to measuring the on-site costs of soil degradation
Two methods are most commonly used to estimate the financial on-site costs of soil erosion:
Replacement cost approach
Productivity change approach
Other methods of valuation include contingent valuation and hedonic pricing (Enters 1998).
Contingent valuation involves asking individuals about their willingness to accept
compensation for the negative effects of erosion or their willingness to pay for the benefits of
conservation. This approach can also be used to assess off-site costs of land degradation. We
are not aware of any studies using this approach to assessing the costs of land degradation in
Ethiopia, although such an approach could be used.
Hedonic pricing uses land prices to assess the economic value of land degradation. This
approach depends upon well functioning land markets, which is a doubtful proposition in
Ethiopia since land sales are prohibited. It is possible that impacts of land degradation are
also reflected to some extent in land rental markets, though probably only to a very limited
extent, since most rental contracts in Ethiopia are short-term in nature. For this review we
therefore consider only the replacement cost and productivity change approaches in more
2.3.1 The replacement cost approach
The replacement cost approach calculates the costs that would have to be incurred in order to
replace a damaged asset (soil); e.g., the annual marginal costs of fertilizer applications to
compensate for the loss of soil nutrients due to erosion. The replacement cost approach is
appealing but can be misleading for two basic reasons: (i), the replacement good (in many
cases fertilizer) is not a perfect substitute for the original good (in this case soil depth and
fertility), and (ii) the replacement costs does not necessarily reflect the opportunity costs of
the good being degraded (i.e., lost benefits of soil depth and fertility due to land degradation).
Errors caused by the first reason will tend to lead to underestimation of the cost of land
degradation, while errors due to the second reason will often tend to overestimate the cost.
More specifically:
Soil erosion does not only affect nutrient status of the soil but also its organic
matter content and its physical structure.
Soil nutrients may not be the most limiting factor in crop production (Bojö, 1996)
Fertilizer applications are not necessarily the most cost effective option available
for farmers for maintaining yields; in extreme cases, e.g., on deep and fertile soils,
farmers may not even experience any yield decline (Stocking, 1995).
The replacement cost is a measure of the cost of avoiding or mitigating land
degradation and not a measure of the cost of allowing land degradation (which is
the same as the benefit of avoiding land degradation). The replacement cost may
be much larger than the benefit of avoiding land degradation, which is one reason
why farmers may allow their land to degrade. For example, the costs of replacing
the lost nutrients associated with soil erosion may be much higher than farmers
would be willing to invest to mitigate this loss.
In summary, the replacement cost approach is simple to apply when nutrient loss data are
available (Bojö, 1996). However, given the above weaknesses it is an imperfect proxy.
2.3.2. The change of productivity approach
Under this approach, the value of land degradation damage equals the value of the lost crop
production valued at market prices (with future losses discounted by market interest rates).
As a physical measurement, it relies on projected crop yields with and without soil erosion.
Crop yields are then multiplied by the unit price of the crops.
The change in productivity approach is logical and is straightforward to apply, but requires
substantial information regarding production and degradation processes, and how the specific
nature of land degradation affects production. However the methodology has some inherent
problems too. Enters (1998) noted the following weaknesses of this approach.
Crop production is highly variable and depends, especially in rainfed
production, on the reliability of the onset of the rain. Thus yield decline is not
always ascribable to land degradation. The impacts of land degradation would
therefore need to be applied either on an assumed average situation or,
preferably, on distributions of climatic and other key factors affecting
Under this approach, yield declines are compared with hypothetical
benchmarks of undegraded soils. The question of what production would be in
the “without land degradation” case is not obvious in most studies.
The possible existence of irreversibility means that a higher cost should be
charged against irreversible land degradation than just the estimated relation
between yield decline and degradation. This is because the natural capital
stock may have an option value that is lost when irreversible damages occur in
the presence of uncertainty about the future value of that stock (Arrow and
Fisher 1974). That option value is not taken into account by conventional
cost-benefit approaches (Ibid.; Dixit and Pindyck 1994).
To the extent that the loss in natural capital results in reduced future productivity, this loss will be reflected in
the gross discounted future loss measure (see Appendix 1). However, there may be a loss in option value
resulting from land degradation even if the expected impact on future productivity is zero. For example,
suppose that soil erosion reduces soil depth in a site from 2 to 1 meter, and that this loss is expected to have no
impact on future productivity under normal conditions. However, the buffer offered by additional soil depth
may prevent future losses under unforeseen adverse future conditions (e.g., future droughts or landslides). This
loss in buffering capacity against unforeseen changes in future conditions represents the loss in option value due
In summary, the change of productivity approach is more appealing than the replacement cost
approach for valuing on-site costs of land degradation, although it also has drawbacks as
noted above. In addition, a number of analysts have noted that the simulation models based
on the replacement and change of productivity approaches do not represent fundamental
hydrological and erosion processes and their results are therefore inaccurate (Enters, 1998).
Physical soil loss is only a rough proxy for soil fertility decline (Bishop, 1995) and it is
difficult to attribute crop yields to differences in past erosion (Olson et al, 1994). In fact as
Pagiola (1992) notes, even high erosion rates may affect crop yields only marginally on deep
soils with favourable subsoil characteristics. Soil erosion can take many years to affect crop
productivity. There is rarely a one-to-one relationship between the amount of soil lost and the
effects on yields. While the displacement of soil is irreversible, the effects of erosion on
productive capacity depend on the depth and quality of soil remaining and not on soil lost
(Scherr and Yadav, 1996). In addition, in semi-arid regions, reductions in moisture-retention
capacity due to erosion are often a more significant contributor to yield decline than the loss
of soil per se (Pagiola, 1992; Pender and Kerr; 1998, Kerr and Pender, 2005). Furthermore,
most analysts consider “moved soil” as “lost soil” even though much of it may be deposited
on other agricultural land, trapped by indigenous technologies such as soil harvesting
structures or trapped and redistributed to fields as an inexpensive source of fertilizer (Lutz et
al, 1994; Stocking, 1995; Kerr and Pender, 2005). Finally, it should be remembered that the
short term impacts of land degradation on agricultural productivity can often be partly or
fully compensated by use of other productive inputs such as labour, capital, and fertilizers.
When there is such adaptive behaviour, a narrow view of the impacts of land degradation
to irreversible land degradation. If land degradation is reversible (with no sunk cost), this option value does not
exist, because the less degraded soil condition could be readily restored when its value is recognized.
may underestimate the impacts of land degradation. The focus should therefore be on the
long-term impacts on the profitability of agriculture (Dejene, 2003).
Cost-Benefit Analysis (CBA)
CBA of land degradation and of sustainable land management interventions relies on
estimates obtained using the above approaches to impute costs to land degradation or benefits
to soil conservation. Deducting the stream of net income earned from land degrading
practices from the stream of net income from a soil conserving practice determines the
“incremental” value which in this case is the cost of erosion (Eaton, 1996). This is the very
essence of a CBA. The time aspect is crucial when soil conserving practices are compared
with exploitative practices. In terms of crop productivity, practices usually become more
distinguishable only in the future. In other words, yields are expected to differ more
significantly in the future than at present or after a period of only a few years. Absolute costs
and benefits are expected to differ between these two scenarios and they may also occur at
different times.
The full CBA framework is a comprehensive approach to estimate the cost of land
degradation. It should take into account all of the current and future costs/benefits,
technological progress and associated costs. Furthermore, off-site costs could be explicitly
incorporated into the analysis.
Since CBA depends on methods to estimate such as those discussed above to estimate costs
of land degradation or benefits or conservation, it faces the same weaknesses already
discussed. CBA is also sensitive to the discount rate and the time horizon used for analysis.
The social rate of discount and time horizon is a much debated issue (Ekbom, 1992). A
“weak spot” in many cost-benefit studies is the rationale for the choice of discount rate and
the time horizon (Bojö, 1986a; Ekbom, 1992). As the selected interest rate and time horizon
obviously influences the results of the CBA, they need careful consideration. In many studies,
the empirical basis for the rates used is at best weak; in most cases no explanations are
provided (Bojö, 1986b). Studies also often fail to provide any rationale for their choice of
time horizon, which ranges from six to 100 years (Enters, 1998).
Apart from the issue of choosing an appropriate discount rate and time horizon, many
attempts to use CBA are also frustrated by imperfect information as this method demands
huge and detailed datasets for its precise application.
Another issue often not addressed in a standard CBA (especially financial CBA) is the use of
shadow prices that take risk and imperfect markets into consideration. This is particularly
important when one computes economic viz-a-viz financial costs of land degradation.
2.4. The problem of aggregating (scaling up) the cost of land degradation
On-site costs of land degradation can be estimated at various hierarchical scales, such as the
cropping/farming system, village/catchment, regional or national scales. Theoretically, the
valuation methods discussed above can be used at any level. But practically it is not possible
to estimate the cost of erosion for each individual field or farm and then aggregate to national
level. Therefore, authors usually identify landscape units for which they expect the same or
similar impacts. The most straightforward approach is then to add up the annual costs of soil
erosion for each unit and thereby derive the total cost due to erosion nationwide (Bishop,
1995). The straightforward approach described above has a number of problems that are not
immediately apparent when only aggregate figures are provided. Enters (1998) noted the
following problems.
It ignores possible price effects of differences in total agricultural production resulting
from land degradation. Such price effects may be limited in many circumstances,
however, because land degradation usually affects only a portion of cultivated land
and may be offset by farmers’ use of inputs, so that effects on total production are
small. Furthermore, price impacts for tradable commodities will be limited by world
market prices.
It does not allow for a change in crop choice over time. In response to declining
productivity, farmers may change to less nutrient demanding or less erosive crops.
If originally soil loss was measured or estimated on the smallest spatial units, then
aggregation will also only deal with soil loss. Adding up all the soil losses basically
means that soil deposition in downstream areas is ignored. Stocking (1995) points out
that net soil losses from major catchments and river systems are typically only a few
percent of the gross soil losses from land in the catchment, as a result of deposition of
eroded soil within the same catchment. As a result, the total cost of erosion to the
national economy may be greatly overestimated if based on gross erosion rates as
predicted by USLE or similar plot scale erosion estimates.
The cost of erosion depends on economic as well as biophysical processes. Variables
such as crop prices, off-farm employment opportunities, discount rates, credit
availability, and farmers’ perspectives affect the cost of agricultural production and
erosion as much as slope, soil quality, and rainfall intensity. These variables are
usually not considered in aggregation.
2.5. Off-site costs of land degradation: neglected impacts
Once soil and excessive runoff leave the boundary of individual farms they cause off-site or
off-farm impacts and result in costs or benefits that are external to the farm household. Clark
(1996) lists 16 off-site costs of soil erosion, of which the most important are likely in-stream
problems of water quality and quantity, sedimentation effects on reservoirs, the degradation
of potable water, a decrease in the availability of irrigation water, increased dredging or
siltation, accelerated runoff leading to localized flooding , reduced hydrological cycling and
recharge of ground water. Such costs are usually not internalized by upstream land users and
affect downstream resource users. In addition to these downstream impacts, which are
usually felt mostly within catchments and national borders, there can also be international and
global impacts of land degradation. Pagiola (1999) identifies potential impacts of land
degradation on climate change (by affecting soil organic carbon sinks as well as above
ground carbon stocks in the form of vegetation, and effects on albedo (reflectivity) of the
earth’s surface); biodiversity (both above and below ground); and international waters as
potentially important international impacts of land degradation (though little research has
established the magnitude of such impacts).
It is important to recognize that not all external impacts of land degradation are necessarily
negative. As mentioned above, soil erosion may lead to beneficial inflows of soil and plant
nutrients to farmers downstream (e.g., farmers in the Nile delta have long benefited from
erosion occurring in Ethiopia and elsewhere in the Nile Basin). Increased runoff of water
caused by land degradation in uplands may also be captured and used by farmers or other
water users downstream. We know of no studies that have attempted to quantify such
positive externalities due to land degradation, however.
So far, most studies estimating the costs of land degradation have concentrated on the on-site
impacts; the analysis of off-site effects, despite often being recommended, has been rare, and
is usually conducted only in qualitative terms, due to the difficulties of measuring such
3. Review of past estimates on on-site costs of land degradation in Ethiopia: Basic
assumptions, major findings and drawbacks
3.1. Scope of the review
There are a number of studies that estimate the on-site costs of land degradation in Ethiopia.
The following are the main works that we have reviewed:
1) The Ethiopian Highlands Reclamation Study (EHRS) (FAO 1986a, 1986b);
2) The Soil Conservation Research Project (SCRP) (Hurni, 1988);
3) The National Conservation Strategy Secretariat (NCSS) (Sutcliffe 1993);
4) The World Bank’s reassessment (Bojö and Cassells 1995); and
5) Estimates by Sonneveld, 2002.
6) The Woody Biomass Inventory and Strategic Planning Project (WBISPP) (WBISPP
2001a, 2001b, 2001c, 2002, 2003a, 2003b, 2004a, 2004b)
The selection of the studies reviewed was mainly based on national or regional level coverage
from available sources.
Most of the studies reviewed focus on the Ethiopian highlands (EHRS, SCRP, NCSS, Bojö
and Cassells), which include areas over 1500 meters above sea level but including associated
valleys. This appears to be restrictive as the Ethiopian highlands cover only about 44 percent
of the country’s total area. However, it may not be so restrictive when we consider the facts
that about 90 percent of the human population, about two-thirds of the livestock population
and well over 90 percent of the regularly cropped land are found in the Ethiopian highlands.
The studies reviewed mainly use data from before 1993, and national level data used in these
studies include Eritrea, which was part of Ethiopia until 1993.
We may also note that the review focuses on on-site costs/impacts although we make
reference to off-site (downstream) impacts depending on the nature of the studies reviewed.
For each study reviewed in this section we present the basic assumptions and methods used to
estimate the extent and costs of land degradation followed by a discussion of the key findings
and drawbacks.
3.2. Review of previous studies
3.2.1. The Ethiopian Highlands Reclamation Study (EHRS)
This is perhaps the most comprehensive and detailed study addressing the issue of land
degradation in Ethiopia. This review is mainly based on two volumes of reports by the EHRS
(FAO 1986a, 1986b), which are based on over 20 working papers prepared as part of the
EHRS. The EHRS addresses a wide range of topics but in this review we focus on the extent
and costs of land degradation. Discussion of these issues in the EHRS is fairly detailed,
including analysis done by (three) zones (high potential cereal (HPC), low potential cereal
(LPC) and high potential perennial (HPP) zones) and also by altitudinal belt (class). In this
review we focus on the country level estimates for the highlands.
A. Basic assumptions and methods
A.1 Extent of land degradation
Broad categories of soil degradation were presented, including water erosion (which includes
sheet (or inter-rill) erosion, rill erosion, gully erosion, bank erosion, tunnel erosion and land
slides), wind erosion, chemical degradation, physical degradation and biological degradation.
It is noted that the severity of soil degradation depends largely on the remaining soil depth
(which is affected by soil erosion) and the fertility of that soil (which is affected by all forms
of biological, physical and chemical degradation) (FAO 1986a: 162).
The extent of erosion was estimated using soil depth and geomorphological maps and data
compiled by the Land Use Planning and Regulatory Department (LUPRD) of the Ministry of
Agriculture, together with extensive field trips. FAO (1986a) notes that the estimates of the
existing erosion according to severity were done on an admittedly subjective basis.
After discussing the presence of wide variation in estimates of soil formation in situ
(1986a: 181) reported that soil formation rates were calculated for the highlands from data on
temperature, rainfall, length of growing period, soil units, soil depth, slope gradient, land
cover and use. While estimates arrived at using this method were considered to provide a
reasonable indicator of relative rates of soil formation in the highlands, the estimates were
tentative and the method used for estimation was still being developed in light of field
verifications, although no further refinements were completed as a part of the EHRS.
Three processes were identified for soil formation at a given site: 1) the re-deposition of sediment by runoff
erosion from a higher site; 2) the natural weathering of rock beneath the soil; and 3) the formation of soil at the
surface by decaying organic and inorganic materials caused by both natural processes and by cropping (FAO
1986a: 181).
Both geological erosion and accelerated (human-induced) erosion contribute to current rates
of erosion. After quoting various estimates of soil loss for Ethiopia, it is noted that the
variation is indicative of conceptual, methodological and statistical difficulties inherent in
making such estimates and the variability of such estimates over time and space. Though
details are not provided,
estimates of gross soil erosion rates are based on an FAO method
derived from the Universal Soil Loss Equation (USLE). FAO (1986a: 182) notes that while
the application of the method provides reasonable estimates of relative rates of erosion in the
highlands, it probably overestimates rates of erosion in absolute terms.
It was also assumed that 80 percent of the total gross soil loss was from croplands while most
of the remaining 20 percent was from overgrazed grasslands and a little from wastelands and
other lands. This assumption was based on records of SCRP which show much lower soil
losses from grass plots, allowing for overgrazing of some grasslands and taking into account
the grasslands area of the highlands (FAO 1986a: 182). Ninety percent of the soil eroded was
assumed to be redeposited evenly across land use types.
Projections of soil depth in 2010 were made based on the assumption that estimated rates of
erosion would continue from 1985 to 2010, although as noted in FAO (1986a), this is
unlikely since population growth and the processes of erosion are likely to accelerate rates of
Sheet and rill erosion were estimated to be by far the most important types of accelerated
erosion in all zones and altitudinal belts. In terms of areas affected and soil irretrievably lost,
It is indicated in the FAO (1986a) report that the method is described in LUPRD/UNDP/FAO (1984) which we
have not been able to locate.
wind erosion and all other forms of erosion combined were considered to be less significant
than the sheet and rill erosion caused by water (FAO 1986a: 166-170).
FAO (1986a) also noted that biological degradation adversely interacts with and aggravates
accelerated soil erosion, as it was considered to be both widespread and severe in the
highlands. It was also noted that obvious indicators of biological degradation include removal
of crop residues and dung from fields, overgrazing and deforestation (FAO 1986a: 170).
Physical degradation of soils was also considered as being commonly associated with
biological degradation and water erosion in the highlands, while chemical degradation was
said to be rarely found in actively eroding soils and more common in stable soils under more
favorable conditions of high and long rainfall (FAO 1986a: 170-171). As a conclusion of the
discussion of biological, physical and chemical degradation, FAO (1986) concluded that the
consensus from the EHRS and other informed opinion “is that sheet and rill erosion and inter-
related biological degradation are by far the most important forms of soil degradation in the
Highlands, both in their area coverage and in their impact on reducing soil productivity”
(FAO 1986a: 171) Estimated rates of soil loss were thus based only on the impact of water
A.2 Costs of land degradation
FAO (1986a) tried to assess the magnitude of the degradation problem in social and
economic terms. Referring to the conclusion in 1981 of a USA National Soil Erosion/Soil
Productivity Research Planning Committee that “erosion reduces productivity first and
foremost through loss of plant-available soil water capacity”, FAO noted the need to consider
the relation between erosion induced yield reductions and remaining soil depth. However,
the difficulty and even impossibility of precise quantification of many of the effects of
degradation due to conceptual problems and lack of data was noted. FAO (1986a) noted the
difficulty of estimating these in the Ethiopian highlands as almost all of the experimental
studies of soil erosion – production relationships were conducted in the USA and Australia,
these studies were field pilot trials and hence the results are site-specific. The analysis of
costs of land degradation was therefore based on guesstimates and sensitivity analysis was
also conducted to examine the sensitivity of results to alternative assumptions.
After a review of international evidence on estimates of yield losses due to soil erosion that
mainly focused on temperate regions with soils on slopes of 10 percent or less, it was noted
that yield declines tend to be higher in the tropics. Slopes in much of the cropland in the
Ethiopian highlands are much steeper than those reviewed and small seed cereals require
finer seedbeds and hence more ploughings than maize. These arguments and estimates of soil
loss discussed above were used to estimate annual yield reductions by FAO (1986a: 196). No
formal model was presented for yield reduction estimates; rather, the estimates were based on
“guesstimates” using the international evidence reviewed as a starting point.
Downstream effects were also discussed but the conclusion was that it is difficult to quantify
many of these effects due to lack of data on sedimentation and the limited time and resources
available. The importance of the effects was, however, noted such that there are some areas
of the highlands where downstream costs of degradation may even be larger than on-farm
Tentative on-site monetary costs of degradation at the national and farm levels were also
estimated. Methodological weaknesses were noted which arise from weak data and limited
understanding of relations between soil losses and yield reductions and the difficulties in
projecting soil losses. The effects of different assumptions and guesstimates were considered
in a sensitivity analysis. Estimates of reductions in crop and grass production were derived
for the 25-year period (1985-2010) classified by three zones and three altitudinal classes.
Four types of costs were specified, viz., lost cropland, lower crop yield, lost grazing land, and
lower grass yield. These costs were compared to a situation where there is no soil erosion.
We should also note that a productivity loss of 2 percent per annum was applied to the entire
crop production area irrespective of soil depth.
A discount rate of 9 percent was used in the computation of present value of costs of
degradation (FAO 1986a: 201).
The sensitivity of the cost estimates was also analyzed by
assuming lower rates of soil erosion.
FAO (1986a: 184) notes that while soil erosion rates are likely to accelerate in the absence of
conservation measures,
the analysis uses the conservative assumption that rates of soil
erosion would be constant over time. No change in agricultural productivity is also assumed
which may be a conservative assumption if baseline agricultural productivity is expected to
rise over time so that yield reductions due to degradation become more than assumed. Prices
paid by the Agricultural Marketing Corporation (AMC) were used, which would
underestimate costs as these were between 20 and 30 percent lower than market prices at the
time of the study.
This rate was the interest rate for agricultural loans in Ethiopia in the mid-1980s and is considered to
approximate the opportunity cost of capital (FAO 1986a: 201). We may also note that the FAO report (FAO
1986a) discusses the issue of whether one should discount future costs of soil erosion as it would have negative
implications for the future generation of Ethiopians.
Arguments to support this include reductions in soil productivity result in thinner vegetative cover, giving rise
to greater erosion; progressive selective removal of the lighter particles and organic materials from the soil by
erosion would make soil increasingly more erodible as it leaves behind the coarser silt and sand; most sub-soils
are already heavier and inherently less permeable than top soils because of their structure and texture so that
run-off increases. In addition to these technical reasons FAO (1986) also mentioned population growth as a
contributor to increased erosion. However, it is also noted that only in some of the already most severely
degraded areas of the highlands are erosion rates likely to decline as the amount of soil left to erode is less (FAO
1986a: 184).
While downstream costs and most social costs were not included, estimates of the number of
persons affected and some social costs were provided (FAO 1986a: 203). FAO (1986a: 198)
argues that within Ethiopia any farming benefits accrued from deposition of sediment are
insignificant and short-lived.
B. Major findings
B.1 Extent of land degradation
Based on the estimates of the severity and extent of erosion in the mid-1980s, FAO (1986a:
166-169) concludes that about half (about 27 million hectares) of the highlands’ land area
was “significantly eroded”, and over one-fourth (14 million hectares) was “seriously eroded”.
It is also concluded that over 2 million hectares of farm lands have reached the “point of no
return” in the sense that they are unlikely to sustain economic crop production in the future.
While the other half of the highlands was estimated to be without significant accelerated
erosion, over half of this was considered to be at future risk due to the likelihood that
cropping will spread in this area and the inherent erodible nature of the soils. Only about 10
million hectares of land (which is about 20 percent of the highland area) was considered to be
“fairly free from the risks of serious accelerated erosion”.
Annual re-deposition of soil in the lower lying highland areas was estimated to be about 90
percent of the total soil erosion (or about 1700 million tons of soil). Assuming this is
distributed evenly in the highlands this is equivalent to about 35 tons of soil per hectare. Soil
FAO (1986a: 198) indicates that in general the deposit of sediments may have both positive and negative
effects. While much land in Egypt might never have been developed if it were not for the deposits coming from
the Ethiopian highlands, the effect on farm benefits in Ethiopia is insignificant and short-lived. Top soil
sediment is sooner or later covered by less fertile sub-soil sediment, and almost always the deposition of
sediment leads to a reduction in the permeability of soil whereon it is deposited, as the fine particles settle into
the water-conducting pores of the soil profile. Although only as low as ten percent of the soil may be totally lost,
these fine particles are probably the most fertile and valuable part. Sedimentation can also damage vegetation.
formation in higher areas is, however, confined to in situ formation. The soil formation rate in
situ was estimated to be from 2 to 22 tons/ha/year for the highlands, which was considerably
higher than estimates reviewed for other countries (FAO 1986a: 181).
After noting data and methodological limitations in the estimation of soil erosion, estimates
from LUPRD data show that total annual erosion in the highlands is about 1900 million tons
of soil. As indicated above the loss from croplands is assumed to be 80 percent of the total
estimated gross soil loss. This would imply that average annual gross loss from cropland was
130 tons per hectare while the average annual loss for all land in the highlands was estimated
at 35 tons per hectare.
On the assumption that 90 percent of the eroded soil was re-deposited evenly per hectare, net
soil losses from cropland were estimated to be about 100 tons per hectare leading to a net
cropland soil loss of about 1100 million tons per annum from the highlands.
While most of
this is deposited on grassland and forestland, the part that is carried into rivers is lost forever
from highland cropping. FAO (1986a) noted that these estimates of erosion rates appear high
compared with estimates for other countries. A comparison was also made with other
estimates such as those by SCRP for Ethiopia, and FAO concluded that the estimated erosion
rates in the EHRS could be overestimated, but probably not by more than 30 percent.
Assuming no change in rates of erosion, estimates of soil depth for 2010 indicate that about
100,000 km
(or about 18 percent of the total area of the highlands) would be bare rock or
While the rate of soil erosion is very high on cropland compared to other land uses, the assumption that soil is
re-deposited evenly in the highlands and concentration of cropland at higher elevations appear to be the main
reasons for the relatively smaller reduction in soil loss due to re-deposition.
FAO (1986a: 183) concluded that “such comparisons suggest that the overestimation implicit in the method
used is probably not more than 30 percent. Thus, even if the calculated erosion rates are reduced by 30 percent
they would still be high … in comparison to rates calculated for other countries. This reflects the fact that most
cultivation in the Highlands is on slopes which are very steep …in comparison to slopes … on croplands in
countries for which such calculations have been made … and that cultivation in the Highlands may involve up
to six ploughings, some at angles to contours, and leave the ground with very little cover during periods of very
erosive rain.”
have less than 10 cm soil depth (11 percent) by 2010 compared to 4 percent under these
categories in 1985. FAO (1986a: 190) concludes that on top of the 20,000 km
of land which
is not capable of sustaining arable cropping in the mid-1980s, an additional 76,000 km
likely to be similarly degraded by 2010. This claim seems to contrast with actual land use
trends, which shows not much change in arable or permanent crop lands in Ethiopia since the
early 1990’s (Figure 1). Of course, the validity of these aggregate land use figures may be
, and they may mask important changes between or within land use categories, such
as expansion of arable land into marginal areas while formerly cropped land is abandoned
due to land degradation.
Figure 1: Land use (1993-2002)
1,000 ha
Arable Land
Non-arable &
Source: FAOSTAT, 2005
For example, the FAOSTAT (2005) figures for permanent pasture area (part of non-arable and non-permanent
crop area) are implausibly given as exactly 20 million ha in every year, while forest area and other land uses
were not estimated after 1994. Estimates of cropped area may be more reliable, since they may be based on
agricultural surveys or crop area reports from local Bureau of Agriculture officials, although the source of such
estimates are not specified in the FAOSTAT database. Unfortunately, other national level estimates of land use
changes for Ethiopia are not readily available.
Several studies using remote sensing information at a micro scale have found large changes
in land use in different parts of Ethiopia during the past several decades (Kebrom Tekle and
Hedlund 2000; Reid, et al. 2000; Gete Zeleke and Hurni 2001; Belay Tegene 2002; Muluneh
Woldetsadik 2003; Tesfaye Demissie, et al. 2004; Dwivedi, et al. 2005). Most of these
studies find that cropland has expanded, usually at the expense of forests or grazing area,
although there are exceptions in different places and times. For example, Reid, et al. (2000)
found that cropland area declined between 1973 and 1987 in the Ghibe Valley region in
southwestern Ethiopia due to drought and other factors, though it expanded between 1957
and 1973 and even more rapidly between 1987 and 1993. In the Dembecha area of Gojjam in
northwestern Ethiopia, Gete Zeleke and Hurni (2001) found that cropland area expanded
from 39% to 70% of the study area between 1957 and 1982, but only increased to 77%
between 1982 and 1995, because of limited remaining area available suitable for expansion.
Belay Tegene (2002) found a similar slowing of cropland expansion after 1986 in a
catchment in South Wello, northern Ethiopia. In west Gurageland, Muluneh Woldetsadik
(2003) found that tree density increased while the area of cropland increased. In the Awash
River basin in eastern Ethiopia, Tesfaye Demissie, et al. (2004) found major cropland
expansion and deforestation between 1965 and 1984, but the deforestation trend reversed
between 1984 and 1996 due to a reforestation program. In a site in Gamo Gofa, Dwivedi, et
al. (2005) found a rapid increase in cropland (22%) and shrubland/grassland (31%) between
1994 and 1997, both replacing barren land as a result of soil and water conservation
measures. These studies highlight the diversity of situations concerning land use change and
land degradation in Ethiopia, and the risks of relying on available macro level studies to draw
B.2 Costs of land degradation:
Based on assumptions and methods discussed above, FAO (1986a) presents estimates of
annual soil losses and the corresponding yield reductions in the highlands as a share of 1985
levels for crops and grass. The results indicate that the predicted average annual yield decline
for crops and grass in the highlands compared to the 1985 level were 2.2 percent and 0.6
percent respectively. Sensitivity analysis was also conducted for “low”and “high” scenarios
and the results suggest that average annual yield decline for crops and grass were 0.6 and 0.3
percent in the “low” scenario respectively. The corresponding figures for the “high” scenario
were 3.4 and 1.1 percent respectively (FAO 1986a: 196, Table 7.1).
Reduction in national grain production due to soil erosion was estimated at about 2 percent
per annum. The equivalent in terms of cereal production lost in the early 1980s was over
120,000 tons per annum, losses which are expected to increase substantially over time (FAO
1986a: 201).
The present value of costs of degradation over the 25-year period after the mid-1980s was
estimated at EB 4,200 million which is equivalent to an average annual reduction of over 2
percent of agricultural GDP in 1982/83.
In 2010 estimates show that about 100,000 km
of the highlands would have a soil depth of
10 cm or lower. Assuming an even distribution of the rural population over the whole land
area, which implied that in 2010 about 10 million persons would have to derive their food
and income from sources other than cropping of their own lands. FAO (1986a: 204) notes
that “if present trends continue, it is likely that nature will impose its own socio-ecological
checks by increasingly frequent and severe famine.”
C. Drawbacks
C.1. Extent of land degradation
It is admitted in FAO (1986a: 182) that the data used for estimation of erosion rates in the
highlands is very limited and necessarily site-specific and that limited use is made of remote
sensing techniques. Data used by the LUPRD with the assistance of a GOE/FAO/UNDP
project for calculation of erosion rates for the highland areas is at the scale of 1:2,000,000 and
a USLE-type of method is used (FAO 1986a: 182). FAO also admitted that the erosion rates
may well be overestimated, as noted above due to poor quality data. Estimates of soil loss and
crop land lost may have been overestimates for the following additional reasons: 1) It appears
that EHRS extended the definition of “cropland” to an additional 21.1 million hectares of
“open and wooded grassland” which increased the figure for cropland soil losses; and 2)
increase in non-cultivable areas was applied not just to cropland but to all land in the
highlands which contributes to overestimation of the cropland lost. On the other hand, the
costs of forms of land degradation other than water erosion were not explicitly estimated,
which underestimates the estimates of the extent of land degradation. Subsequent studies
(discussed below) address both of these issues.
C.2. Costs of land degradation:
Estimates of effects of erosion on yield are only guesstimates without any formal model.
Exclusion of forms of land degradation other than sheet and rill erosion may contribute to the
underestimation of the costs of land degradation, as would some other assumptions such as
use of crop prices that were lower than market prices. On the other hand, the extent of
reduction in yield due to erosion depends on soil depth. It appears that this was not taken into
account in the estimation of costs, which may overstate the costs of erosion. Factors that
contribute to overestimation of soil loss estimates discussed about would also have
implications for crop loss estimates.
3.2.2. The Soil Conservation Research Project (SCRP)
was a project with the mandate to conduct research on land degradation in Ethiopia.
It was part of the Ministry of Agriculture and was implemented in collaboration with Berne
University, Switzerland (Hurni 1988). Hurni (1988) noted that it is only since 1981 with the
inception of SCRP that scientific approaches to soil conservation problems in Ethiopia were
systematically and continuously implemented. The research program was conducted at three
levels (Bojö and Cassells 1995):
1. a national level exploratory survey on rainfall erosivity;
2. regional- and district-level data collection on land use;
3. catchment and plot-level collection of soil conservation data.
We review the work by Hurni (1988) which estimated soil loss in Ethiopia as part of SCRP’s
A. Basic assumptions and methods
As did the EHRS, Hurni (1988) assumed that soil erosion by water must be considered the
most important of all land degradation processes. He also assumed that soils in Ethiopia were
originally more than 50 cm deep and probably had rooting depths of more than 100 cm.
Soil loss rates for different land unit classes (land cover types) were estimated using the
Universal Soil Loss Equation (USLE) adapted for Ethiopian conditions. Hurni (1988) noted
SCRP was funded by the Swiss Directorate for Development Cooperation and the Ethiopian Government.
There were a network of eight stations (out of which two were in now independent Eritrea) measuring losses,
runoff, and crop yield on experimental plots using different conservation techniques (Bojö and Cassells 1995).
that these results correlated well with test-plot measurements made by the SCRP, although, as
noted previously, such estimates can greatly overstate the net impact of erosion on soil loss,
because of re-deposition of soils elsewhere in the landscape.
Hurni (1988) reports gross soil losses and hence the loss figures are not net of soil re-
deposition and soil formation. Soil formation was separately reported so that gross losses
could be compared to soil formation, as an indicator of sustainability. Estimation of soil
formation was “based on assumptions of soil regeneration as a function of geology and soils
(unit and depth), slope gradient, climate (temperature, rainfall, and length of growing period),
and land use” (Hurni 1988: 126). Hurni also noted that the model was not validated
thoroughly with field data.
B. Main findings
Hurni (1988) reported (approximate) gross soil loss rates using land cover types classified by
the Ethiopian Highlands Reclamation Study (EHRS). These gross soil loss rates are reported
in Table 1. The figures in Table 1 indicate considerable variation in soil loss rates by land use
Table 1. Estimated gross soil loss rates by land cover type
Soil loss Land cover Area(%)
t/ha/year t/year (in millions)
Cropland 13.1 42 672
Perennial crops 1.7 8 17
Grazing and browsing land 51.0 5 312
Currently unproductive (Former
3.8 70 325
Currently uncultivable 18.7 5 114
Forests 3.6 1 4
Wood and bushland 8.1 5 49
Total country 100 12 1493
Source: Hurni (1988: 127) and Bojö and Cassells (1995: 7).
We note from Table 1 that the average gross soil loss in the country was estimated to be 12
t/ha/year and 1,493 million t/year. Note that in spite of the lower gross soil loss rates
compared with those for EHRS, Hurni’s estimates of total soil loss per year are not very far
from those of EHRS due to the fact that the former presents estimates for the entire country.
We also note that the soil loss rate is much higher for cropland than for grazing, perennial
crops, forests and bushland. Thus, in spite of the fact that cropland takes only about 13
percent of the total area of the country it contributes to about 45 percent of the total estimated
soil loss. This has its own implications for agricultural expansion leading to increased share
Note that some of the land cover types reported in Bojö and Cassells (1995) quoting the same work by Hurni
(1988) are different and they note that these are revised based on personal communication with Hurni on June
15 and 16, 1994. A major difference is that what is called ‘Grazing and browsing land’ in Hurni seems to have
been divided into ‘Grazing land’ and ‘Swampy land’ with area shares of 47% and 4% respectively in Bojö and
Cassells. Bojö and Cassells also use the term ‘Former cropland’ which seems to be instead of ‘Currently
of cropland and hence increased land degradation. Hurni (1988) also reported considerable
variation in soil loss rates by agro-ecological zones (dry and hot lowlands, moist middle
slopes, cold and wet highlands). Estimated loss rates were the highest in cultivated slopes in
the Dega highland zone (2,300-3,200 m a.s.l), where there was the highest rainfall and the
most intensive cultivation while soil loss from grassland was much lower (3-10 times less
than from cultivated land) (Hurni 1988: 127). While Hurni’s estimates of gross soil loss are
lower than those of the EHRS, it appears that Hurni’s estimates are more reliable for at least
two reasons: 1) they were calibrated on field measurements and a more detailed subdivision
of soil texture factor is used in deriving erodibility; 2) improvements have been made in the
estimates of rainfall erosivity (Krauer 1988; Bojö and Cassells 1995)..
Based on soil formation rates estimated by Hurni (1988), and using a rule of thumb on soil
, Bojö and Cassells report soil formation rates of 3-7 tons per hectare per year, well
below the estimated loss rates.
Bojö and Cassells (1995) also report Hurni’s estimates of the productivity implications of his
erosion results, which postulate a 2 percent decrease in agricultural production per year due
to land degradation.
Hurni (1988: 127) indicates that 1 millimeter of soil depth corresponds to about 10 metric tons per hectare.
Note, however, that the bulk density of Ethiopian soils varies (Bojö and Cassells: 6). It seems likely that
Ethiopian soils are more dense than water, which has a density of 1.0 g/cm
, so this estimate seems low.
Stocking (1996) estimates that 1 mm of soil typically weighs about 13 tons per ha. (i.e., a bulk density of 1.3
Bojö and Cassells (1995: 7) note that this appears to be based on empirical measurements from Hurni’s work
in 1985 in which data from 52 plot samples of barley showed an annual decline in the yield of barley on the
order of 2 percent per year, based on an erosion rate of 66 tons per hectare per year. Bojö and Cassells (1995: 7)
indicate that Hurni confirmed through personal communication that the yield equation is: Y=0.38+0.032X
(where Y=barley yield in tons and X=depth of reworkable soil in centimetres), which implies a yield reduction
of 320 kilograms per 10 centimeters of soil lost and not 250 kilograms as given in Hurni (1985). With an
average yield of 1,820 kilograms per hectare (X=45 centimeters), this implies a relative loss of about 1.5 percent
per year, given the loss rate of 66 tons/ha.
C. Drawbacks
Only soil erosion by water was considered, as with the EHRS. Though it is the major
contributor to soil erosion it would lead to an underestimation of the cost of land degradation
as the latter include other forms of land degradation such as biological degradation.
The effects and implications of soil re-deposition are also not considered. Moreover, referring
to Hurni’s work, Bojö and Cassells (1995: 6) note that the “soil formation” concept utilized
by Hurni involves three separate processes; “(1) accumulation on the surface of the soil
(decomposition of organic material, deposition of material transported by water and wind),
(2) soil formation within the A horizon (humus formation, root development, and
decomposition), and (3) weathering of parent material (development of the B horizon)”.
Given the three processes above, there will be double counting if the sum of re-deposition
and soil formation is deducted from gross soil loss to calculate net soil loss. Information on
the three processes mentioned above is required to determine the level of this double
counting. This information is, however, not available as only the total is given. Bojö and
Cassells (1995) note that conceptually, this would not be a problem if all re-deposition is
actually included, but the figures for “soil formation” are much lower than the figures for “re-
deposition” reported for example in the EHRS. This appears counterintuitive given the first
process in the definition of the soil formation concept above which should include re-
deposition as part of soil formation.
3.2.3. The National Conservation Strategy Secretariat (NCSS)
This review is based on Bojö and Cassells’ (1995) discussion of work by Sutcliffe (1993).
Using additional data, Sutcliffe (1993) presented new estimates of the extent and impacts of
on-site soil erosion and the loss of nutrients through the use of dung and residues for fuel.
A. Basic assumptions and methods
Sutcliffe looks at the impact of water erosion followed by an analysis of the impact of
nutrient losses due to removal of dung and crop residues. Sutcliffe followed the following
steps to derive the impact of water erosion (Bojö and Cassells 1995: 8):
1. Sutcliffe used the erosion rate estimates by Hurni (1988), which he judged to be more
realistic than the higher rates estimated by EHRS.
2. Sutcliffe established the minimum required depth for cultivation of different crops and the
maximum required (90 cm) beyond which erosion does not immediately affect crop
cultivation. He used a soil life model developed by Stocking and Pain (1983) as the analytical
framework. The future user cost of decrease in soil depth was also incorporated in his
3. Using a model developed by FAO, Sutcliffe related monthly values of rainfall, water-
holding capacity, evapotranspiration, and crop water requirements to derive a Water
Requirements Satisfaction Index (WRSI). The relationship between the WRSI and relative
reductions in crop yield was then derived.
4. Estimates from the EHRS of crop mix, absolute yields, and areas for three altitudinal
ranges and three major agro-ecological zones (HPP, HPC, and LPC), were used to project
losses in crop production. Losses in crop residue production were converted to tropical
livestock unit (TLU) equivalents. Estimates were made for the period 1985 to 2010 to be
comparable to the EHRS.
5. Financial implications of the results were estimated using market prices in 1985/86, while
world market prices were used to determine the economic impact for society as a whole.
Sutcliffe then builds on a methodology developed by Newcombe (1984) to analyze the
impacts of breaches in the nutrient cycle due to use of dung and residue as fuel. Using
estimates of actual use of dung and residues as fuel from the Ethiopian National Energy
Commission, the estimates were derived in the following three steps:
1. The amount of nitrogen (N) and phosphorus (P) lost and their impacts in terms of lost crop
production were calculated;
2. The burnt crop residues were evaluated in terms of their potential value as livestock feed
and converted to potential increases in TLUs, which in turn are priced.
3. The replacement value of nutrients lost, using the price of chemical fertilizers, was
B. Main findings
Estimates of soil erosion damage and nutrient breaches are given separately below.
Sutcliffe’s soil erosion damage estimates are presented in Table 2. Sutcliffe’s estimates of
crop production losses and cropland losses for 1985 were only about 7 percent and 6 percent
of the respective EHRS estimates.
Table 2. Soil erosion damage estimates
Year Estimates
Crop production losses (rounded to thousands of tons)
1985 9
2010 332
Cropland lost (thousands of hectares)
2010 489
Pastures lost (thousands of hectares)
2010 5,747
Source: Adapted from Sutcliffe (1993) by Bojö and Cassells (1995: 9)
As noted above, Sutcliffe also estimated the cost of breaches in the nutrient cycle. The results
show losses that are much larger than those for erosion. The 1990 losses in physical terms are
estimated as (1) almost half a million tons of grain; and (2) more than 1 million TLUs. These
losses are much higher than those for soil erosion in 1985 (see Table 2). The corresponding
differences in monetary costs between soil erosion and nutrient losses would also be large.
Bojö and Cassells (1995: 9) report that Sutcliffe estimates total crop losses due to losses from
both erosion and nutrient loss to be at about 0.5 million tons in 1985 (the base year for his
calculations). When livestock losses are added, he derives a total cost of EB 581 million in
1985 which corresponds to about EB934 million in 1994.
These results imply that the cost of land degradation due to erosion is much less than
suggested by previous studies. On the other hand, the cost of nutrient losses was estimated to
be very large.
The following are noted by Bojö and Cassells (1995: 9) as important reasons for the
differences in methodologies used and hence the results between EHRS and Sutcliffe’s work:
1) Sutcliffe used Hurni’s lower estimates of gross soil loss than those used by EHRS for
reasons mentioned above; 2) EHRS used a loss in productivity of about 2.2 percent per
annum to the entire crop production area, regardless of soil depth. As noted above Sutcliffe
uses the assumption that crop losses set in only after reaching a depth of not more than 90
centimeters while still accounting for the future user cost; 3) about 21.1 million hectares of
“open and wooded grassland” appears to have been included by EHRS in the definition of
“cropland”; this would overstate the total soil loss estimates as there are even actual gains in
topsoil cover on grazing land; and 4) the EHRS applied the percentage increase in non-
cultivable areas to all land in the highlands, not only cropland which increased the estimated
area of wasteland to 7.6 million hectares of “cropland” loss in the year 2010.
C. Drawbacks
Bojö and Cassells (1995: 11-12) point out a number of drawbacks of the study by Sutcliffe
(1993). First, there may be an element of double counting involved in relation to the use of
residues or dung as fuel and their opportunity cost in terms of lost crop production (including
loss of future residue production). Since there is also the additional option of feeding residues
to livestock, the opportunity cost is the higher of the two use values. But they cannot logically
be added. Adjusting for this would reduce Sutcliffe’s estimate downward by more than 20
Second, Bojö and Cassells (1995) note that there is probably double counting in estimation of
the cost of livestock losses from burning residues instead of using them as livestock feed.
The sum of the value of flow products and services from livestock and the value of livestock
as a “store of wealth” is taken as a measure of the cost. Since the market value of livestock
should reflect the net value of the total flow of goods and services that emanates from the
livestock over its expected (remaining) lifetime plus the store of wealth value and other
intangibles, double counting will occur since the flow values are separately accounted for.
Moreover, the investment before the sale of livestock is not deducted when calculating the
net value of the entire (remaining) life cycle of a livestock unit. While adjustment for these
would revise the damage estimates downwards considerably, such adjustment of Sutcliffe’s
figures was not possible as the necessary details were not provided.
Third, Sutcliffe followed Hurni not only in using his estimates of gross soil losses but also in
not adjusting for re-deposition of soil to arrive at net losses. Bojö and Cassells (1995) note
also that in subsequent communications, Sutcliffe indicated that more detailed USLE
calculations taking account of the likely distribution of slope and soil depth classes reduced
average losses from cropland to about 20 tons per hectare per year (without any allowance for
re-deposition processes). Thus water erosion losses may still be overstated in Sutcliffe’s
study, even though these are estimated to be much lower than in either the EHRS or Hurni’s
Fourth, the availability to plants of removed nutrients contained in burned dung and residues
may be limited. The implicit assumption appears to be that such nutrients are highly available
for plant growth, in the sense that the nutrient content is a perfect substitute for commercial
fertilizer. This is more likely for dung than for residues, but may not hold even for dung,
depending on how it is managed.
Thus, the costs associated with burning dung and crop
residues are probably overstated by Sutcliffe (1993).
3.2. 4. World Bank’s reassessment
After a review of previous major studies on land degradation in Ethiopia, Bojö and Cassells
(1995) reassessed the problem using concepts of economic costs defined in terms of (1) the
counterfactual scenario assumed and (2) the time horizon used. They used three concepts of
economic loss in their work, viz. gross annual immediate loss (GAIL), gross discounted
future loss (GDFL) and gross discounted cumulative loss (GDCL) (see Appendix 1 for
definitions of these concepts). Their reassessment is reviewed below.
A. Basic assumptions and methods
A.1 Extent of land degradation
Bojö and Cassells’ estimates of land degradation are mainly based on the approach followed
by Sutcliffe, where both soil erosion and nutrient losses are taken into account. In cases
where data is available, the drawbacks of Sutcliffe’s work reviewed above were addressed by
Bojö and Cassells to arrive at revised estimates. Moreover, Bojö and Cassells (1995) pay
particular attention to the issue of re-deposition. They noted the importance of attempting to
take into account re-deposition in spite of problems of lack of information. Based on previous
work they argue that “the net rate at which the soil is lost to agriculture is lower than the
gross rate of soil erosion as estimated by even the most rigorous geomorphic applications of
USLE-type calculations. … Hence, there is reason to adjust previous calculations of the rate
of soil loss down even further than suggested by the National Conservation Strategy
Secretariat” (Bojö and Cassells 1995: 40). Thus, they argued, due to re-deposition “net soil
When manure is not covered, much of the nitrogen can be lost to the atmosphere. If the nitrogen content of
manure is too low relative to the carbon content, it can actually reduce crop yields in the near term by reducing
the availability of soluble nitrogen to plants (Giller et al. 1997).
loss to cropland is probably on the order of half or less of the gross figures used in previous
analyses” (p. 12).
A.2. Costs of land degradation
Costs of soil erosion:
Using a matrix derived by Sutcliffe (1993), Bojö and Cassells (1995) related various rates of
soil loss per annum with relative decline in yield. An average decline of 0.4 percent per year
was obtained for all cereals as a function of soil loss. Information on distribution of soil depth
was used to calculate the decline in crop yield.
The figure for yield impacts of erosion used by Bojö and Cassells is similar in order of
magnitude to most estimates of yield impacts of soil erosion in several studies for other parts
of the world (Wiebe 2003). For example, Crosson (1995a) estimated an annual average
agricultural productivity loss of 0.4 percent globally on degraded lands between 1945 and
1990, based upon estimates from the Global Assessment of Soil Degradation (GLASOD) by
Oldeman, et al. (1991). Oldeman (1998) estimated annual productivity losses averaging 0.1
to 0.2 percent, though considerably higher considering less favorable loss rates and in
particular regions (e.g., 0.5 percent per year in Africa and 0.7 percent per year in Central
America). Pagiola and Bendaoud (1994) estimated productivity loss rates of 0.4 to 0.7
percent annually on steeper slopes in Morocco, while Pagiola (1996) estimated substantially
higher impacts of erosion on maize and beans yields in the Machakos district of Kenya (2.2
percent annual losses). Based on case studies in Africa, Lal (1995) estimated annual yield
losses for major food crops (cereals, pulses, roots and tubers) due to erosion in Africa to be
between 1970 and 1990 to average 0.5 percent, and 0.3 percent in sub-Saharan Africa. For
China, Huang and Rozelle (1995) estimated that environmental degradation, mainly soil
erosion and salinization, reduced grain yields by about 0.4 percent per year on average
between 1976 and 1989, although the impacts varied substantially by crop and region. For
the United States, Pimentel, et al. (1995) estimated annual losses due to soil erosion of 8
percent, but this figure has been criticized by Crosson (1995b) and others. Crosson’s own
estimates (1986) imply annual losses in the United States of only about 0.1 percent for maize
and less for other crops; similar to findings of other studies (Pierce, et al. 1983; Alt, et al.
1989). Using data from an exhaustive review of published experimental studies, Weibe
(2003) found that mean yield losses due to soil erosion are usually in the range of 0.01 to 0.04
percent per ton of soil lost, and generally lower in temperate regions. Based on estimates of
potential erosion rates for different crops in different parts of the world, Weibe (2003)
estimated average annual yield losses for major crops to be 0.3 percent globally, ranging from
as low as 0.04% in Europe to 0.94% in Latin America, and estimated at 0.49 percent for
Africa. These figures show that the estimated yield loss due to erosion assumed by Bojö and
Cassells (1995) is well within the range of published estimates of erosion induced yield
losses, and quite close to other estimates for Africa.
As noted earlier, while the EHRS discussed the issue of re-deposition and estimated its
extent, it appears that the effect on yields was considered negligible. On the other hand, based
on estimates of re-deposition discussed above and assuming that the productivity impacts are
approximately linear within the minimum and maximum soil depth thresholds assumed by
Sutcliffe, Bojö and Cassells divide the erosion damage estimates presented by Sutcliffe
roughly by two.
Based on Sutcliffe’s analysis, a critical depth of 100 centimeters was assumed for all crops
concerned. Thus for cases where soil depth exceeds 100 centimeters (which is the case for 69
percent of the total cropland) Bojö and Cassells (1995: 35) assumed that the effect of soil
erosion can be disregarded as being negligible in the near term. Even taking a relatively long
time horizon they note that for most of these cases the user cost is negligible. The effect of
soil loss on crop yield is therefore calculated for only about 30 percent of the cropland.
Bojö and Cassells (1995) modified Sutcliffe’s estimates of decreased livestock feed
availability due to lower crop yields caused by soil erosion by taking into account the effect
of re-deposition.
Cost of nutrient loss (due to burning of dung and crop residues):
Bojö and Cassells (1995: 36) referred to a letter from Sutcliffe in 1994 who reported revised
calculations that show the opportunity cost for burning of crop residues to be livestock
production forgone, not grain loss. They then noted that given that the focus of their work
was on land degradation, we should not concern ourselves with use of crop residues as energy
rather than livestock feed and that the removal of residues from land is rational as its value is
higher as livestock feed than as fertilizer. They noted that “removal of crop [residues] does
not constitute ‘land degradation’ in a meaningful sense of the word, as it is taken from its best
alternative use as livestock feed to be used as fuel.” They therefore concluded that crop
residues should be taken out of the calculation of estimates of the total nutrient loss.
Bojö and Cassells (1995: 36) noted that the same basic model can be used as the one used by
Newcombe and Sutcliffe to reconstruct a scenario using functions that relate dung quantities
removed to nutrient content and nutrient loss to crop loss (6 kilograms of crop loss per
kilogram N or P).
Price adjustment:
Since Sutcliffe’s estimates use 1985/86 as the base year, Bojö and Cassells (1995) converted
the effect on crop yields into monetary values in 1994 using average rate of inflation and the
retail price index for Addis Ababa.
Economic vs. financial costs:
Bojö and Cassells (1995) did not make a distinction between financial (market) and economic
(social or shadow price-based) costs, arguing that such adjustments are of minor importance
due to the liberalization of major markets in Ethiopia after 1991.
Discount rate, time horizon and future use of dung and crop residues:
Bojö and Cassells (1995) used a discount rate of 10 percent in their calculations of GDFL and
GDCL, noting that it is the World Bank standard for public investment projects. An infinite
time horizon was assumed for computation of GDFL while a period of about one century was
considered for GDCL calculations. For computation of GDCL, due to expectation of
increased demand in the future partly because of population growth, an annual growth rate of
2.5 percent was assumed for use of dung and residue. Assuming limits to the livestock herd
and dung production, and limits to what can be collected and returned to cropland, Bojö and
Cassells (1995) imposed a cap of 20 million tons of dung.
B. Main findings
Bojö and Cassells (1995) used the following three economic concepts of costs: gross annual
immediate loss (GAIL); gross discounted future loss (GDFL); and gross discounted
cumulative loss (GDCL). They used Sutcliffe’s (1993) estimates as the best available
Starting from a price per ton of EB 890 in 1985/86, they used an average inflation rate of 2.4 percent for the
period 1986-1990 and retail price index for Addis Ababa for January 1991 and May 1994. The figure they arrive
at is 890X(1.024)
X1.428=EB 1,431.
benchmark. They modified Sutcliffe’s (1993) estimates of soil loss rates to an average of
about 20 tons per hectare per year (or about 2 millimeters per year). After making some
rough adjustments to Sutcliffe’s calculations based on the assumptions presented above, their
analysis leads to the erosion and nutrient loss figures reported in Table 3.
Soil erosion, which is assumed to affect about 30 percent of cropland, leads to a loss of grain
of about 7,000 tons per year. Using EB 1,431 as the price per ton of grain in 1994, the value
of grain lost due to soil erosion comes to about EB 10 million. The cost of decreased
availability of livestock feed due to soil erosion leading to lower crop yields is estimated at
EB 0.8 million in 1994 prices.
This means a total current income loss of EB 10.8 million
due to soil erosion.
Table 3. Reassessment of land degradation costs (EB million)
Erosion Nutrient loss
Source: Bojö and Cassells (1995: 13).
Bojö and Cassells (1995) started from the estimated cost of EB 1 million arrived at by Sutcliffe as a measure
of cost of decreased availability of livestock feed due to soil erosion in 1985. This is reduced to 0.5 million as
the physical impact of soil erosion could be halved due to re-deposition. This figure is then multiplied by 1.608
to adjust the prices and arrive at the final figure of EB 0.8 million as the cost decreased availability of livestock
feed due to soil erosion in 1994 prices.
Based on assumptions presented above the loss due to nutrient breaches is estimated at EB
626 million in 1994 prices. Thus, the total GAIL due to erosion and nutrient loss is EB 637
which amounts to 3 percent of agricultural GDP
The GDFL due to soil erosion with infinite time horizon is estimated at about EB 108
million. This represents the present value of the loss of productive capacity due to soil
erosion over an infinite horizon. As mentioned earlier Bojö and Cassells (1995) noted that the
accounting for losses in the future due to nutrient breaches is not strictly parallel to that of
erosion losses and hence the nutrient losses should be counted only once.
The concept of GDCL is used as layers of costs are added on top of each other due to the fact
that land degradation is a cumulative process— another follows the cost of a year’s erosion
and nutrient removal. For soil erosion, using a decline in crop yield at the rate of 0.4 percent
per year due to erosion, presented above, Bojö and Cassells (1995) project the decline in crop
yield 100 years into the future, which they deduct from a hypothetical “without” land
degradation scenario. Using prices in 1994 and a 10 percent discount rate, the GDCL is
estimated at about EB 3.4 billion (in present value terms).
Computation of GDCL for nutrient breaches is based on assumptions of growth in dung and
crop residue use and imposing limits to the amount of dung that can be used. With a discount
This is the sum of EB 10 million (due to grain loss caused by erosion), EB 0.8 million (due to decreased
livestock feed availability caused by erosion) and EB 626 million (due to nutrient losses)
The agricultural GDP figure used is based on what is reported for 1992 adjusted for inflation of the dollar
until 1994 (3.9 percent per year) (Bojö and Cassells 1995: 36).
To justify this conclusion Bojö and Cassells (1995: 36) argue that “in circumstances in which the supplies to
the available nutrient store ….were higher than the total of both crop removals and the nutrient breach,
restoration of nutrient retention should allow full recovery of productivity; however, this does not mean that the
crop yield would not have been even higher in year two, had the nutrient loss in year one not taken place.
…hence, there could be some lasting impact even if the original level of nutrients is replenished. The impact is
not lasting over several years, however, as the nutrient store available in a particular year, if available to plants
and used would be absorbed by that year’s crop, and hence (largely) removed.”
rate of 10 percent, the GDCL for the period until 2100 was a present value of about EB 8.3
billion (Bojö and Cassells 1995: 37).
Bojö and Cassells (1995) noted that their results only illustrate the order of magnitude of the
problems due to the high level of uncertainty associated with the data used in the calculations
of. A comparison of the GAIL and GDCL measures illustrates both the immediate
significance of nutrient breaches and the longer-term cumulative effects of physical soil
C. Drawbacks
In addition to other problems with data and methodology that also apply to Sutcliffe’s work
on which Bojö and Cassells’ reassessment is based, the issue of re-deposition should be
examined more carefully. Assuming that re-deposition reduces the costs of erosion by 50% is
an arbitrary assumption. We should note however that Bojö and Cassells did a sensitivity
analysis and reasonable variation in the rate of re-deposition did not critically impact the end
result. Perhaps more data is needed to take into account the effects of deposition on yield.
3.2. 5. Sonneveld (2002)
Sonneveld (2002) estimated the impact of soil degradation (SD) on agricultural production
mainly using two approaches: empirical (non-parametric) and engineering approaches. In the
empirical approach Sonnveled estimated the impact of SD on crop yields for dominant cereal
crops. The soil degradation data (inventory) was based on the work of UNEP/GRID (1992).
In the second approach the soil losses (SD) estimated from combined spatial water erosion
models are related to all agricultural production (all crops and livestock). In the following
section we discuss each approach separately.
44 Empirical approach
Under this approach Sonneveld analyzed the effect of SD on crop yields for the dominant
cereal crops and studied this relationship with the population level and fertilizer use within a
map unit. Further, he evaluated the effects of soil conservation and application of fertilizer on
future crop yields.
A. Basic assumptions, methods and data
It was assumed that soil erosion by water is the most important form of land degradation.
The relationship of SD and production was estimated using a non-parametric (kernel density)
regression model for rainfed areas. The value of yield ratio (actual/potential yield) served as
dependent variable. The reason for adjustment of the actual yield by potential yield was to
correct for agro-ecological and crop-genetic differences. The modeling compared yields in
relation to indexes of soil degradation and soil fertility. The data set was cross-sectional and
comprised qualitative classifications of soil degradation and soil fertility, mean monthly
annual rainfall, slope gradients, altitude, temperature, length of growing period and crop
yields of dominant cereal crops. All data were geo-referenced according to 460 polygons
(map units) of the Crop production System Zones (CPSZ) (FAO, 1998). These map units
correspond to former administrative units in Ethiopia (Aurajas). The above data were
overlaid with a map on soil degradation derived from an African continental inventory
(UNEP/GRID, 1992). Experts on the region and physiographic zone qualitatively carried out
the soil degradation assessment.
According to estimates cited by Sonneveld (and consistent with assertions of the EHRS and
Hurni (1988), water erosion is the dominant degradation process in those areas where cereals
are cultivated, followed by wind erosion (Table 4). The detachment of topsoil is the most
widespread soil degradation characteristic, followed by mass movements that mainly occur
on the steep soils of the highlands.
Table 4. Degradation type as a percentage of the total area of the rain fed cropping zone
Degradation type Percentage of total area
Water erosion top soil 57
Water erosion mass movements 29
Wind erosion 8
Source: Hakkeling, 1989 cited in Sonneveld, 2002
B. Main findings
The results were depicted in 3-D graphs (mollifier pictures). The relation found between yield
ratio and soil degradation was not very strong. However, the analysis identified the
SD has its major impact on soils of lower fertility and where population levels are
On fertile soils, SD is largely compensated by fertilizer application.
Many areas, populated by a large percentage of the people, are in a critical state,
where fertility loss needs urgently to be compensated by new external impacts and/or
soil conservation measures need to be implemented (see below). The most
vulnerable areas are in northern Ethiopia.
Further, Sonneveld simulated the impact of soil conservation and fertilizer use on yield by
changing SD and soil fertility over a period of 10 years and compared with actual yields. The
following three policy scenarios were developed.
1) No erosion controls and without fertilizer application (no control scenario)
2) With soil conservation and no fertilizer use (conservation scenario)
3) No erosion control and with fertilizer (fertilizer scenario)
In doing so the following assumptions were made:
1) SD increases by 2.5 per cent per year or is controlled at its current level by soil
erosion control.
2) Soil fertility declines by 2.5 per cent per year or is maintained by fertilizer use at its
current level.
In scenario 1 crop yields are projected to decrease by 22% over 10 years at national level
compared to the actual yields due to soil degradation and loss of soil fertility. In this scenario
the northern provinces of Wollo, Gondar and Tigray contain the real degradation ‘hot spots’
that mark them as priority areas for intervention. A similar spatial pattern of crop yield losses
(20% average nationally) was found in the conservation scenario, indicating that soil
conservation alone cannot compensate for the decreasing soil fertility levels. In the last
scenario yield decreased by 6% implying that fertilizer use may mask the effect of SD and
that fertilizer use alone also does not completely stop the decrease in productivity.
C. Drawbacks
Lack of quantified soil degradation (soil loss intensity) variable. Soil degradation
index was defined as qualitative variable (nil, low, moderate, severe, very severe) for
each CPSZ based on the qualitative judgment of experts on erosion hazard. Why
Sonneveld did not consider the intensity of soil losses was not indicated in the paper.
The basis for assuming that SD index would increase by 2.5 percent per year and that
soil fertility index would decline by 2.5 percent per year without controls is not clear.
Only general comments are discussed here. More technical comments are annexed in Appendix 2.
The estimated decline in yields (over 2% per year) due to land degradation is higher
than most other studies of erosion-yield relationships (see references cited earlier) and
appears inconsistent with assumptions in the subsequent engineering approach used
by Sonneveld (discussed below), which appear to imply a productivity decline of less
than 1 percent per year.
It was not clear whether the soil degradation index was developed specifically for
cereals or for all land use types of each CPSZ (map units)).
It was also not clear how the soil fertility variable was defined and hence difficult to
comment on it.
Different conservation types have different effect on soil erosion and production. In
the Sonneveld paper the type of conservation measures used in the simulation analysis
was not mentioned. Engineering Approach
The first approach was focused on annual cereal crops neglecting perennials and livestock
production and the impact of overgrazing on soil erosion. In the engineering approach the
impact of SD was evaluated for the entire agricultural sector by combining spatial water
erosion models with a spatial yield function that gives an estimate of the agricultural yield in
its geographical dependence on natural resources and population distribution.
A. Basic assumptions, method and data sources
It was assumed that soil erosion by water is the most important form of land degradation.
Specific assumption for each scenario include: free movement of people and easy access to
land within and outside ethnic origin of the farmers (migration scenario), access to
technology, infrastructures to sale surplus product generated due to improved technology use
and accelerated urbanization (non-agricultural sector activities) to absorb surplus rural labor
force (technology scenario) were assumed.
The soil losses were estimated using a combination of three spatial water erosion models:
USLE model, expert model and accessible data model. A descriptive comparison of model
characteristics and statistical comparison of the predictive power of each model showed that
the reliability (applicability) of the models varies in specific areas of the data domain. Given
this, USLE model was recommended in areas with low rainfall erosivity, defined as an R-
factor of less than 50 and slope gradients that are lower than 2%. The expert model applied
for land under rangeland or where silt percentages are lower or equal to 40 % and the
accessible data (accDat) model was applied for all other combination of land use or
biophysical variables.
The impact of soil loss on agricultural production was estimated as a function of soil fertility
and water holding capacity following Kassam et al. (1993) and Batjes (1996). Kassam et al
(1993) relate soil loss to the two most important soil productivity characteristics: soil fertility
and water holding capacity. Soil fertility depends on nutrient status and the regenerative
capacity of the topsoil (Smaling et al., 1997; Stoorvogel and Smaling, 1998). To estimate the
impact of soil loss on potential yield, Sonneveld approximated the susceptibility of soil
fertility to erosion by classifying soil types of each map units into three groups. These are
least, moderate and susceptible soil types, implying productivity reductions of 1, 2, and 7 per
cent per cm topsoil loss
. Based on Batjes (1996) water holding capacity was related to soil
depth, soil type, phase and textural class. Unlike the soil fertility case Sonneveld did not
Sonneveld did not mentioned how he derived these figures. However, these estimates are in the range of
productivity impacts cited by Weibe (2003). Assuming soil has a bulk density of 1.3 g/cm
, one cm of topsoil
weighs 130 metric tons. Productivity losses of 1 to 7 percent per cm. of topsoil lost translate to about 0.008 to
0.05 percent per metric ton of soil lost per ha.
report yield reduction figures due to reduced water holding capacity. The production impact
of soil loss was expressed as potential yield reduction and a reduction of the area under
cultivation. If the potential yield level dropped below 20 per cent of the maximum potential
yield, the land was considered unsuitable for production. This threshold value was based on
the land evaluation criteria of FAO (1993).
To assess the impact of water erosion on food production in Ethiopia; Sonneveld developed a
spatial optimization model that maximizes national agricultural revenues under alternative
scenarios (Table 5) subject to the possibilities for the rural population to migrate to other
rural areas with better prospects. The same data types and sources as in the empirical
approach were used.
B. Main findings
The results of the optimization model showed that in scenario 1 where there is no additional
land and water conservation measures are not adopted, total national agriculture remains
stagnant. The value of agricultural production per capita plunges from $372 U.S. in 2000 to
$220 in 2010. In this scenario the average annual production decline controlling for the
effects of population growth (which was not reported by Sonneveld) compared to the 2000
level was 2.93 per cent. (Column C of table 5). This is equivalent to losing U$D 5 billion
from the agricultural sector due to land degradation over the period 10-years (2000-2010)
with annual average of US$ 500 million. However, with introduction of soil conservation
measures production was projected to increase on average by 0.19 per cent per annum. This
contrasts with the simulation analysis carried out in the empirical approach (under the
conservation scenario) where yield are projected to decrease on average by 20 per cent over
10 years at national level.
The implications of other scenarios are indicated in Table 5. Other scenarios all assume
greater flexibility for migration and/or different technology impact. Soil conservation and
migration to other productive areas support a slow growth of agricultural production, but this
does not suffice to meet the expected food demand. Allowing trans-regional migration,
production in the highly degrades areas is exchanged for production in less degraded sites. A
shift to modern technology offers better prospects and moderates the migration, but soil
conservation, especially in the long term, remains important. In general, each scenario points
to the need for a sustained focus on rural production together with conservation of natural
resources. However, a vital agricultural sector will also depend on the accelerated growth of
non-agricultural activities as it further alleviates poverty in rural areas, contributing to higher
income levels for the total population and, simultaneously, relieving the pressure on the land
through rural-urban migration. This study emphasizes the need for a combined conservation
investment and infrastructural approach.
C. Drawbacks
The use of the engineering approach to generate soil losses for each CPSZ may be too
general given the diversity of the biophysical variables and land use systems within
each CPSZ.
The data or assumptions used to estimate the degradation – productivity relationship
are not clearly stated.
The type of soil conservation measures used in the simulation was not mentioned. In
addition to this, the cost of soil conservation not accounted for in the analysis of the
agricultural revenue (neither were these costs considered in the other studies
reviewed). This will overestimate agricultural revenue or GDP.
Prices of production will vary depending on the technology used, population size,
season, increasing quality of products, etc. Sonneveld used constant prices, ignoring
dynamism of price due to factors mentioned above and others (also not considered in
other studies).
It was not clearly spelled out how livestock production was affected by soil erosion;
i.e., whether production loss due to soil erosion (soil erosion may reduce grain yield
and grazing land) was included in the analysis.
The result under the control scenario (with soil conservation investment) where
production is projected to increase on average by 0.19% per annum was not in line
with the results of the conservation scenario in the empirical approach where yield are
projected to decrease 20 per cent.
Off site effects of soil erosion was not accounted for in the analysis (nor in the other
studies reviewed).
The model considered only soil losses by water erosion. Other forms of land
degradation were ignored (as in some of the other studies).
Table 5. Impact of soil degradation on food production: Summary of Scenario Results
Agricultural production
(in billion USD; PPP)
Per capita
agricultural production:
rural population
(in USD; PPP)
Average annual
production loss (gain)
due to soil erosion
(as % of AGDP)*
2010 2010 2010
1. Stationary
12.4 220 - 2.93
2. Control
17.8 324 + 0.19
3. Migration
15.9 283 -0.91
3.1. Restricted
23.2 412 +3.28
16.9 300 - 0.32
3.2. Free
24.2 430 + 3.8
4. Technology
43.5 773 + 14.89
4.1. Stationary (UN)
65.4 1163 + 27.41
43.5 773
Source: Sonneveld 2002 (pp 192) and own calculation.
+ 14.89
65.3 1661 + 27.36
* This is calculated as the yield difference between year 2000 and 2010, controlling for population growth.
Definition of Scenarios
Stationary: No soil conservation, current technology, continued in-situ population growth (no migration)
Control: Investment in soil conservation halting/reducing decline, continuing in-situ population-growth (no migration)
Migration: Restricted – indicates people movement within their ethnic administrative areas of origin
Free – indicates movement within and across ethnic boundaries (tarns-regional migration allowed)
Technology: Assume better quality produce and higher potential levels of production
UN and AccUrb: Scenario for rural labor force growth: population growth under UN prediction & the accelerated urbanization alternative
that presents a higher outflow from the agricultural sector to industrial and service activities.
3.2.6 Woody Biomass Inventory and Strategic Planning Project (WBISPP) (WBISPP
2001a, 2001b, 2001c, 2002, 2003a, 2003b, 2004a, 2004b)
The WBISPP was a project that, among other things, looked into issues of land degradation
in Ethiopia. This review is based on reports of the project for several regions of the country
prepared in collaboration with federal government and regional government offices of
Ethiopia (WBISPP 2001a, 2001b, 2001c, 2002, 2003a, 2003b, 2004a, 2004b). The WBISPP
reports dealt with various matters including biomass demand/consumption and
supply/production issues. However, given the objectives of this review, we will focus on
aspects of the reports dealing with the extent and costs of land degradation. We should note,
however, that while the reports cover most of the regions in Ethiopia, with different degrees
of completeness regarding land degradation, we have not been able to find a report for
Since this is a major omission we have not been able to make any conclusions at
the national level.
A. Basic assumptions and methods
The basic assumptions and methods applied in the WBISPP are similar to those used by
Sutcliffe (1993) with some changes introduced using information made available since 1993.
The reports indicate that the main types of land degradation in the regions are erosion of soil
by water and biological degradation involving the breaching of soil nutrient cycles (nitrogen
and phosphorous) by the burning of dung and residues.
A.1 Soil erosion
Using the Universal Soil Loss Equation (USLE) adapted for Ethiopia (Hurni 1988), soil loss
was calculated using the following five sets of factors: erosivity of rainfall, erodibility of soil,
We had access to reports on the following regions/administrative councils: Afar, Amhara, Benshangul-
Gumuz, Dire Dawa, Gambella, Harari, SNNP and Tigray. However, we present results only for those regions
for which comparable data is available.
slope, slope length, land cover, and land management. Values for these factors were obtained
from data sets within the GIS.
The extent of soil erosion was classified into as many as
eight groups.
It was noted that the greatest impact of soil loss on crop yields was through the reduction in
soil depth and thus the water holding capacity of the soil. Using a soil-life model (Stocking
and Pain 1983) and “Water Requirements Satisfaction Index” (WRSI) developed by FAO
(1983) which relates monthly rainfall, soil water holding capacity and evapotranspiration to
reductions in crop yield, critical soil depths at which reductions in crop yields start to occur
are calculated. As in Sutcliffe (1993), the concepts of maximum and minimum critical soil
depth were used. When soil depth is more than the maximum critical depth erosion does not
have an impact on yield. The minimum critical depth is that depth at which crop yields are
less than 20 percent of those obtained at the maximum critical depth, and are judged as
effectively useless for cropping.
These models were used to estimate the impacts of soil loss by water erosion on crop
production. It was noted, however, that since soil re-deposition on cultivated land has not
been accounted for, the results are over-estimates of crop production losses. Estimates of
annual and cumulative crop production and cultivated land lost are made for the period 2000-
The various reports of WBISPP indicate that the rainfall erosivity factor of the USLE was applied to the 1:2
million map of mean annual rainfall and the soil erodibility factors were applied to FAO soil units identified on
the MOA/UNDP/FAO 1:1 million soil-geomorphology map. Slope percentages were derived from a digital
terrain model created from the 1:250,000 topographic maps of Ethiopian Mapping Agency. Slope length was
assumed to be related to slope angle. The land cover factors were based on the Project’s land use and land cover
These groups are: 1) 0 - 3.125 ton/ ha/year (equivalent to 0 – 0.25mm of topsoil removed); 2) 3.125- 6.25
tons/ha/year (0.25 – 0.50mm); 3) 6.25 - 12.5 tons/ha/year (0.50 – 1.0mm); 4) 12.5 - 25 tons/ha/year (1 – 2 mm);
5) 25 - 50 tons/ha/year (2– 4 mm); 6) 50-100 tons/ha/year (4 – 8 mm); 7) 100 - 200 tons/ha/year (8 – 16 mm);
and 8) 200 - 400 tons/ha/year (16.0– 32 mm).
A.2 Biological degradation
Based on Sutcliffe (1993) which was updated using dung nutrient contents cited in Lupwayi
et al. (2000), one ton of dry dung was estimated to be equivalent to 11 kg. of DAP (P
equivalent) or 35 kg. of urea (N equivalent) while 1 ton of crop residues was estimated to be
equivalent to 1.1 kg. of DAP (P equivalent) or 11 kg. of urea (N equivalent). Dung for fuel
collected from the household’s own farmland was estimated for each region.
Estimates of the costs of nutrient breaches were made by converting the amount of nutrient
lost into the equivalent in terms of fertilizer which was then converted into the equivalent
crop loss. A nutrient-increment coefficient of 5 kg. of cereal grain per kg of nutrient (N) was
B. Major findings
As indicated above, we are not able to report results at the national level and make
comparison of results with other studies as the information available is incomplete. We
therefore present the general nature of the results for the regions for which data is available.
B.1 Soil erosion
Estimates of the maximum and minimum critical soil depths for reduction in crop
productivity due to soil loss are given in Table 6.
Table 6: Estimated maximum and minimum critical depths for reductions in crop
productivity due to soil loss.
Crop Maximum critical depth (cms) Minimum critical depth(cms)
Maize/wheat 93 28
Sorghum/pulses 77 22
Teff 91 10
Source: Various reports of WBISPP.
Table 6 shows that the lowest critical maximum soil depth was for sorghum and pulses (77
cm) indicating that these crops are more drought resistant than the rest included in the table.
However, the lowest critical minimum soil depth was for teff (10 cm).
For regions for which data is available, the percent of total area for each soil loss class is
shown in Table 3A of Appendix 3. The results show that soil loss rates are much higher for
cropland compared with the results for other land. The total weight of soil moved in gross
and net terms is presented by region in Table 3B of Appendix 3. The net soil loss figures
indicate the amount lost after accounting for re-deposition.
Estimates of annual and cumulative crop loss due to soil erosion for the period 2000-2025 are
presented in Table 3C of Appendix 3. The reports also include cultivated land lost due to
erosion. We note from the results that the cumulative loss is particularly significant due to the
irreversible damage caused by soil erosion. For Tigray and Amhara regions crop and
Although net soil loss figures accounting for re-deposition are estimated in the reports, the impacts on crop
production are based on the gross soil losses, as noted earlier. It is noted in the reports that these estimates of
soil loss are generally much lower than those reported in the Conservation Strategies of the regional states.
cultivated land lost due to erosion in 2000 was estimated to be well below 1 percent of crop
production and cultivated land. However, predicted crop and cultivated land lost for 2025 is
about 3 percent of crop production and cultivated land in 2000.
B.2 Biological degradation
Using the methods described above, estimates of soil nutrient loss due to the removal of dung
and crop residues and their equivalent in terms of chemical fertilizer and grain production lost
are given in Table 3D of Appendix 3. Table 3E of Appendix 3 presents plant nutrient losses
and grain production foregone for an average family.
We may also note that a comparison of the cost estimates for soil erosion and nutrient
breaches shows that in terms of annual losses the costs of the later are much higher than those
of the former, similar to previous findings of Sutcliffe (1993) and Bojo and Cassells (1995).
In the long-run, however, one needs to take into account the fact that soil erosion leads to
irreversible damages.
C. Drawbacks
C.1 Soil erosion
Due to the similarity between the approach followed by WBISPP and Sutcliffe (1993), the
major drawbacks discussed there apply to this work as well. These include the use of the
USLE in the estimation of soil loss, and the absence of consideration of re-deposition in the
estimation of crop losses due to soil erosion.
C.2 Biological degradation
Here as well, the drawbacks of Sutcliffe (1993) generally apply to this study. Thus, whether
or not crop residues should be included in the calculation of crop loss is questionable
depending on the opportunity cost. The issue of limited availability to plants of removed
nutrients contained in burned dung and residues is also relevant.
4. Comparison of results across studies
In Tables 7 and 8, we summarize the basic assumptions, findings and weaknesses of past
studies of the cost of land degradation in Ethiopia. In order to compare the costs of land
degradation in Ethiopia with results from other countries, we provide additional information
on land degradation estimates of some Sub-Saharan African countries in Table 7, and report
the estimated costs as a percentage of agricultural GDP.
The estimated annual costs of land degradation in Ethiopia range from 2% to 6.75% of
agricultural GDP. The cost estimated by Sutcliffe (1993) is highest, because he accounted for
costs associated with burning dung and crop residues, as well as costs of soil erosion. As
mentioned previously, Bojö and Cassells (1995) revised downward Sutcliffe’s (1993)
estimates of the costs of burning dung and crop residues, as well as reducing his estimates of
the cost of soil erosion to account for re-deposition of soil, arriving at an annual income loss
estimate of about 3% of agricultural GDP. Bojö and Cassells’ estimate of the present value
of future cumulative losses due to land degradation was much larger than this, but as a
cumulative measure, should not be compared to annual losses or annual agricultural income.
The annual costs of land degradation estimated for Ethiopia are within the range of estimates
of the cost of land degradation in several other African countries. For example, Bishop and
Allen (1989) estimated costs of erosion in Mali to be 4% to 16% of agricultural GDP using
productivity and replacement cost approaches. Bishop (1995) estimated costs of erosion to
be 3% to 13% of agricultural GDP in Mali and 17% to 55% of agricultural GDP in Malawi.
By contrast, Grohs (1994) estimated the costs of erosion in Zimbabwe to be less than 0.4% of
agricultural GDP. It is beyond the scope of the present study to review the methods and
assumptions of these other studies. However, this small sample of other studies suggests that
the range of on-site costs of land degradation estimated by the various studies for Ethiopia is
not unreasonable as an order of magnitude estimate.
There are several problems that are common to many or all of the Ethiopian studies that may
have caused significant errors, including the following:
As discussed earlier, land degradation takes various forms including soil erosion
(caused by wind and/or water), deterioration of the physical, biological or economic
properties of the soil; and long-term loss of natural vegetation. With the exception of
Sutcliffe (1993) and Bojö and Cassells (1995), all studies impute value only to soil
erosion caused by water. Other forms of land degradation and downstream (off-site
effects) were completely ignored in the analyses. These omissions will tend to cause
underestimation of the costs of land degradation. On the other hand, Sutcliffe’s
estimates of costs due to burning of dung and crop residues likely overstate these
costs, since they assume that the plant nutrients available in these organic materials
would be readily available to crops and lead to immediate productivity impacts, which
is doubtful in many circumstances. Bojö and Cassells (1995) removed burning of
crop residues from their cost estimates, but still included the cost of burning dung,
although they note the problem of assuming that this would have immediate impacts
on productivity.
All studies admitted the poor quality of their data. Furthermore, in their estimation of
soil erosion rates, different studies applied different techniques to compute the soil
erosion rate. EHRS applied a USLE-type (though not specifically stated) approach,
while Sutcliffe and Bojö and Cassells’ reassessment was based on the USLE,
developed and adopted to Ethiopia by Hurni. As has been noted, USLE was
developed for application to croplands in the United States, and does not account for
soil re-deposition. This latter problem tends to overstate the costs of erosion, which
only a few of the studies addressed (albeit in an
ad hoc fashion). Sonneveld applied a
combination of approaches (such as map units, USLE and expert and accessible data
model), and found that different approaches to estimating erosion were better suited to
different conditions. Notwithstanding the limitations of the USLE itself and poor data
quality, differences in approaches of estimating soil erosion could possibly explain
differences in soil erosion and cost of soil erosion estimates.
In estimating erosion-yield relationships, different studies made different assumptions
concerning the minimum and maximum depth requirements for plant growth which,
we believe, will have considerable implications to their estimation of cost of soil
erosion. EHRS and SCRP implicitly assume no critical soil depth requirement for
plant growth, whereas Sutcliffe constructed a matrix of minimum and maximum soil
depth requirements for different crops, which attaches zero value for any level of soil
erosion unless the soil depth is below the critical level. Bojö and Cassells assumed a
critical depth of 100 cm for all crops, which excludes crop lands with soil depth above
100 cm in their estimation of costs of soil erosion. As a result, Bojö and Cassells
estimated the cost of soil erosion only for 30% of the cropland used. This ignores the
(possibly irreversible) loss in the capacity of soils that are deeper than 100 cm
resulting from erosion, assuming that this is no economic value; possibly tending to
understate the costs of soil erosion. As noted in earlier in this paper, in the context of
uncertainty, loss of deeper soils may entail a loss of option value associated with the
ability of the soil to buffer against unforeseen adverse events in the future.
With the exception of Bojö and Cassells, none of the studies distinguished soil loss as
an immediate loss, a future loss and cumulative losses to the national economy. EHRS
and Suttcliffe reported the discounted future loss, whereas SCRP and Sonneveld
reported only immediate losses. It is only in Bojö and Cassells that all the three types
of costs were clearly reported. Thus some of the estimates reported in different studies
are not comparable.
The soil erosion rates imputed in many of the studies, with the exception of Bojö and
Cassells, did not take re-deposition of soil into account. Even their inclusion of re-
deposition was based on a crude assumption (a 50% re-deposition rate and a reduction
of 50% of the productivity losses as a result) and an even distribution of re-deposited
soil across crop lands. The actual effects of re-deposition of soil on downstream
productivity are far from clear. Thus, it is difficult to say whether the reassessment
conducted by Bojö and Cassells leads to an overestimate, an underestimate, or a
reasonable estimate.
In all of the studies the benchmark levels of soil erosion and nutrient balances were
not clearly specified. This makes comparisons of estimates across studies difficult.
Furthermore, the discount rates and the time horizon used to impute value to cost of
soil erosion differ across the studies, also making comparisons difficult.
Price dynamics were not considered in any of the studies, with the exception of
Sonneveld who arbitrarily assumed a 2.5% annual growth in soil erosion and 2.5%
annual soil fertility decline. All the costs of soil erosion were computed with a
constant price and soil erosion rates over the years. The impacts of land degradation
on prices were not considered. Since land degradation tends to reduce agricultural
supply and hence increase prices of commodities (if not perfectly tradable), this
would tend to reduce the costs of land degradation to farmers but increase it for
consumers as a whole. Of course, the costs to particular farmers and consumers likely
vary greatly, depending on their net sales or net purchases of commodities, the extent
to which they are affected by land degradation, and other local factors.
Almost all studies did not indicate how they aggregated their estimates to national or
regional levels. Significant problems of aggregation could have resulted from the fact
that the relationships between land degradation and productivity are nonlinear, and
estimates basically represent averages of conditions at the lowest unit of aggregation
used in each study. To the extent that threshold effects are present, costs could be
substantially underestimated or overestimated as a result of averaging degradation
processes and productivity impacts over large spatial units.
As can be verified from FAO data on production and land use patterns in Ethiopia, we
don’t observe as much decline in cropland use, crop production, or yields in Ethiopia
over the last 20 years as predicted by some of the studies (see Figures 1, 2 and 3). For
example, the area of cultivated cropland has not changed to a large extent since 1993,
contradicting the dire predictions of the EHRS and Sutcliffe that large areas of
cropland would be abandoned by 2010 (see section 3.2.1. B.1 and table 2). Cereal
yields have also trended upward to some extent in the past decade, although with
large annual variations. Part of this upward trend is likely due to increased use of
fertilizer (Figure 4), which might be in part a response to soil nutrient depletion and
may be masking negative effects of soil erosion. This seems consistent with
Sutcliffe’s, Bojo and Cassells’ and Sonneveld’s assessments of soil fertility depletion
as a major form of land degradation in Ethiopian highlands.
It is impossible to say without a more detailed study whether the errors in the available
studies have caused them to over- or underestimate the costs of land degradation on balance,
since some of the errors tend to overestimate land degradation impacts and some tend to
underestimate them. Given the wide range of methods and assumptions used by the different
studies for Ethiopia, it is remarkable that their findings are of similar magnitude concerning
annual costs relative to agricultural GDP. Of course, there are large differences in the
estimated components of those costs, since several of the studies focused only on soil erosion,
while those that focused on soil nutrient depletion due to burning of organic materials found
that this has a larger impact in the near term than soil erosion. Nevertheless, based on the
available evidence, we believe that the true magnitude of annual on-site costs of land
degradation in Ethiopia (if they ever could be known with confidence) are likely to be of the
order of magnitude of the range of estimates shown in Table 8; i.e., a few percent of
agricultural GDP per year.
Source: FAOSTAT, 2005
Figure 3. Cereal Yield
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004
Source: FAOSTAT, 2005
Figure 4. Fertilizer Consumption
1993 1994 1995 1996 1997 1998 1999 2000 2001 2002
Source: FAOSTAT, 2005
5. Conclusions and implications
Although there has been considerable effort in the last few decades to estimate the on-site
costs of land degradation in Ethiopia, our review suggests that each of the previous studies
have produced only partial results subject to significant methodological and data weaknesses.
Nevertheless, there is general agreement across the studies that the overall cost of land
degradation in Ethiopia is substantial—probably on the order of a few percent of agricultural
GDP per year—though not as dire as some may believe.
This is not to say that the problem is not severe in many areas. Some of the studies highlight
particular land degradation “hotspots”, such as large areas of Tigray, Wello and Gonder, as
emphasized by Sonneveld’s findings. Furthermore, Sonneveld’s and other studies were
based on earlier assessments of land degradation, and the situation may have changed
markedly in the past few decades. It is not only low potential areas where severe land
degradation is occurring; for example, Gonder includes many high potential agricultural
areas. There is also evidence from more location specific studies of severe degradation
occurring in some very high potential areas, likely lead to high costs of land degradation per
hectare in such areas. For example, Gete and Hurni (2001) found that their study area
(Dembecha watershed in a high potential area of Gojjam) lost 3% of its total land area
between 1957 and 1995 due to land degradation, and 5% of the area reached critical soil
depth (i.e 25cm and below). They also predicted an additional 21% of the currently cultivated
land will reach critical soil depth over the coming 15 to 47 years. Similar problems high rates
of land degradation and negative productivity impacts have been noted in the Ginchi
watershed near Holetta in the Oromiya region (Okumu, et al. 2002); Andit Tid watershed in
north Shewa (Shiferaw and Holden 2001); and other high potential locations in Ethiopia (e.g.,
Lakew Desta, et al., 2001; and Bezuayehu Tefera, et al., 2002).
Some of the more recent studies emphasize the problem of soil fertility depletion as being at
least of equal importance to the problem of soil erosion, which was the main emphasis of the
earlier studies. Though soil erosion still contributes to a decline in agricultural productivity,
the reassessments by Sutcliffe (1993), Bojö and Cassells (1995) and WBISPP (various years)
are fairly persuasive that the costs of erosion are probably less than earlier thought while the
importance of soil nutrient depletion is more important. This finding is supported by
Sonneveld’s (2002) findings from his empirical study and actual production and yield data
since 1985. But we also think that Sutcliffe’s and Bojö and Cassells’ approaches to
estimating costs of burning organic materials overstates those costs, since the usefulness of
these organic materials as a fertilizer are probably less than assumed in many cases, due to
loss of nitrogen from manure, difficulties of applying manure on distant plots, etc. So our
sense is that the on-site costs of these types of land degradation may be still less than Bojö
and Cassells estimate. Nevertheless, there are still other aspects of land degradation that
haven’t been taken into account by any of the studies; particularly off-site costs. It is
important that such impacts be considered in designing policies and programs to address land
degradation and sustainable land management in Ethiopia.
For future research, we think the key gap is the need for a full cost-benefit analysis of feasible
options to reduce land degradation and improve productivity. Estimating the costs of land
degradation, no matter how well done, will only take us a little way towards deciding what to
do about it. Decision makers need to know what actions can be taken that are socially
profitable. That requires investigating off-site effects where those are likely to be important,
as well as on-site costs and benefits of land management options. Such off-site effects could
be quite important in particular watersheds, particularly where water resource development is
taking place. Further research is needed to address this topic. We recommend that such
research focus on areas where valuable downstream resources are most at risk due to land
degradation upstream.
To define what the government’s and other stakeholders’ roles should be, it is also important
to consider the difference between private and social costs and returns. If private net present
values are high for conservation investments, opportunities for adoption may be great, though
market imperfections and constraints such as land tenure insecurity or credit constraints may
need to be addressed for adoption of sustainable land management practices to increase. If
social returns are high but private returns are low, a more active approach involving
subsidies, taxes or regulations may be necessary to ensure socially optimal levels of
investment. By contrast, if both social and private returns are low, there is no efficiency
rationale to promote investments, unless these returns can be increased by improving the
technologies and/or the market environment. In such cases, alternative livelihood options
may be more effective to address land degradation problems where they are severe.
Alt, K., C.T. Osborn, and D. Colacicco. 1989.
Soil Erosion: What Effect on Agricultural
Productivity? Agriculture Information Bulletin No. 556, Washington, DC: Economic
Research Service, U.S. Department of Agriculture.
Arrow, K. J. and A.C. Fisher. 1974. Environmental preservation, uncertainty, and
Quarterly Journal of Economics 88, 312–319.
Barber, R. 1984. An assessment of the dominant soil degradation processes in the Ethiopian
highlands: their impacts and hazards. Working Paper 23, Part of the Ethiopian
Highlands Reclamation Study. Addis Ababa: Land Use Planning and Regulatory
Department, Ministry of Agriculture, Government of Ethiopia, and FAO.
Batjes, N. (1996) Development of a world data set of soil water retention properties using
pedo-transfer rules. Geoderma, Vol. 71, pp. 31-52.
Belay Tegene. 2002. Land-cover/land-use changes in the Derekolli Catchment of the South
Wello zone of Amhara Region, Ethiopia.
Eastern Africa Social Science Research
Review 18(1): 1-20. Michigan State University Press.
Bezuayehu Tefera, G. Ayele, Y. Atnafe, M.A. Jabbar and P. Dubale. 2002. Nature and
causes of land degradation in the Oromiya Region: a review. Socio-economics and
Policy Research Working Paper 36, International Livestock Research Institute,
Nairobi, Kenya.
Bishop, J. 1995. The economics of soil degradation: An illustration of the change in
productivity approach to valuation in Mali and Malawi. LEEC paper DP 95-02.
London: International Institute for Environment and Development.
Bishop, J., and J. Allen. 1989. The on-site costs of soil erosion in Mali. Environment
Working Paper #21, Environment Department. Washington: The World Bank.
Bojö, J., 1986a. A review of cost-benefit studies of soil and water conservation projects.
Report no. 3., Soil and Water Conservation and Land Utilization Programme,
Bojö, J, 1986b. An introduction to cost-benefit analysis of soil and water conservation
projects. Report no. 6., Soil and Water Conservation and Land Utilization
Programme, Maseru.
Bojö, J., 1996. The costs of land degradation in Sub-Saharan Africa.
Ecological Economics
16: 161-173.
Bojö, J., 1991. The economics of land degradation: theory and applications to Lesotho. The
Stockholm School of Economics, Stockholm.
Bojö, J. and D. Cassells 1995. Land degradation and rehabilitation in Ethiopia: a
reassessment. AFTES Working Paper No. 17, World Bank.
Clark, R., 1996. Methodologies for the economic analysis of soil erosion and conservation.
CSERGE Working Paper GEC 96-13. Norwich: University of East Anglia.
Convery, F. and Tutu, K., 1990. Evaluating the costs of environmental degradation in Ghana.
Applications of economics in the environmental action planning process in Africa.
University College Dublin, Dublin.
Crosson, P. 1995a. Soil erosion estimates and costs.
Science 269: 461-464. July.
Crosson, P. 1995b. Futures supplies of land and water for world agriculture. In:
and Food in the Early Twenty-First Century: Meeting Future Food Demand of an
Increasing Population
, edited by Nurul Islam. Washington, DC: International Food
Policy Research Institute, pp. 143-159.
Crosson, P. 1986. Soil erosion and policy issues. In
Agriculture and the Environment, edited
by T.T. Phipps, P.R. Crosson, and K.A. Price. Washington, DC: Resources for the
Future, pp. 35-73.
Dejene, A. 2003. “Integrated natural resources management to enhance food security. The
case for community-based approaches in Ethiopia.”
FAO Environmental and
Natural Resources Working Paper 16. Rome.
Dixit, A. K., and R. S. Pindyck. 1994.
Investment under Uncertainty. Princeton, NJ:
Princeton University Press.
Dwivedi, R.S., K. Sreenivas and K.V. Ramana. 2005. Land-use/land-cover change analysis
in part of Ethiopia using Landsat Thematic Mapper data.
International Journal of
Remote Sensing
Eaton, D., 1996. The economics of soil erosion: a model of farm decision making. Discussion
paper 96-01. Environmental Economics Programme. London: International Institute
for Environment and Development.
Ehui, S., T. Hertel, and P. Preckel. 1990. Forest resource depletion, soil dynamics, and
agricultural productivity in the tropics. Journal of Environmental Economics and
Management 18: 136-154.
Ekbom, A. 1992. Economic impact assessment of implementation strategies for soil
conservation. Environmental Economics Unit. Department of Economics,
Gothenburg University, Sweden.
El-Swaify, S. 1997. Factors affecting soil erosion hazards and conservation needs for tropical
steep lands. Soil Technology 11:3-16.
Enters, T. 1998. Method for economic assessment of the on-and off-site impacts of soil
Issues in Sustainable Land Management no.2.
FAOSTAT data 2005, last updated February 2005..
Food and Agriculture Organization (FAO) of the United Nations 1986a. Highlands
Reclamation Study: Ethiopia. Final Report. Volume I.
Food and Agriculture Organization (FAO) of the United Nations 1986b. Highlands
Reclamation Study: Ethiopia. Final Report. Volume II.
Food and Agriculture Organization (FAO) of the United Nations. 1998. Cropping production
system zones of the IGAD. Land and Water Digital Media Series 2. Rome.
Gete Zeleke and H. Hurni. 2001. Implications of land use and land cover dynamics for
mountain resource degradation in the northwestern Ethiopian highlands.
Research and Development
21(2): 184-191.
Giller, K.E., G. Cadisch, C. Ehaliotis, E. Adams, W.D. Sakala, and P.L. Mafongoya. 1997.
Building soil nitrogen capital in Africa. In: R. Buresh, P.A. Sanchez, and F. Calhoun
Replenishing Soil Fertility in Africa. Soil Science Society of America Special
Publication No. 51, American Society of Agronomy, Madison, WI. pp. 151-192.
Grohs, F. 1994. Economics of soil degradation, erosion, and conservation: A case study of
Zimbabwe. Arbiten zur Agrarwirtschaft in Entwicklungslandern. Kiel:
Wissenschaftsverlag Vauk Kiel KG.
Hakkeling, R.T.A. (1989). Global Assessment of Soil Degradation Eastern and Southern
Africa. Vol. 1. Main report. Vol.2 Matrix Netherlands Soil Survey Institute.
Wageningen, The Netherlands.
Huang, J. and S. Rozelle. 1995. Environmental stress and grain yields in China.
Journal of Agricultural Economics 77: 853-864.
Hurni, H. 1988. Degradation and conservation of the resources in the Ethiopian highlands.
Mountain Research and Development 8(2/3): 123-130.
Kassam, A.H., H.T., van Velthuizen, A.J.b. Mitchell, G.W. Fisher and M.m. Shah (1991)
Agroecological assessment for agricultural development planning. A case study of
Kenya. Resources data base and land productivity. Technical Annex 2. Soil erosion
and productivity. FAO/IIASA.
Kebrom Tekle and L. Hedlund. 2000.
Land cover changes between 1958 and 1986 in Kalu
District, Southern Wello, Ethiopia.
Mountain Research and Development: 20(1):
Kerr, J and J. Pender. 2005. Farmers’ perceptions of soil erosion and its consequences in
India’s semiarid tropics.
Land Degradation and Development (in press).
Krauer, J. 1988. Rainfall, erosivity, and isoerodent map of Ethiopia. Research Report 15.
Berne: Institute of Geography, University of Berne in Association with the United
Nations University.
Lakew Desta, Menale Kassie, S. Benin and J. Pender. 2001. Land degradation in the
highlands of Amhara region and strategies for sustainable land management.
International Livestock Research Institute, Livestock Policy Analysis Program,
Addis Ababa.
Working Paper No. 32.
Lal, R. 1995. Erosion-crop productivity relationships for soils in Africa. Soil Science Society
of America Journal
59(3): 661-667.
Lupwayi, N.Z., M. Girma and I. Haque 2000. “Plant nutrient contents of cattle manures from
smallholder farms and experimental stations in the Ethiopian Highlands”,
Agriculture, Ecosystems and Environment 78: 57-63.
Lutz, E., S. Pagiola, and C. Reiche. (eds.). 1994. Economic and institutional analyses of soil
conservation projects and policies in Central America and the Caribbean.
Environment Paper 8. Washington DC: The World Bank.
McKenzie, C. 1994. Degradation of arable land resources: policy options and considerations
within the context of rural restructuring in South Africa. Johannesburg.
Muluneh Woldetsadik, M. 2003.
Population growth and environmental recovery: more
people, more trees, lessons learned from West Gurageland.
Ethiopian Journal of the
Social Sciences and Humanities 1(1): 1-33.
Norse, D. and Saigal, R., 1992. National economic cost of soil erosion--the case of
Zimbabwe. Paper prepared for the CIDIE workshop on Environmental Economics
and Natural Resource Management in Developing Countries, World Bank,
Washington, DC, 22-24 January 1992. FAO, Rome.
Okumu, B., Jabbar, M., Colman, D., and Russell, N. 2002. A Bio-economic Model of
Integrated Crop-Livestock Farming Systems: The Case of the Ginchi Watershed in
Ethiopia. In: C.B. Barrett, F. Place and A. A. Aboud (eds.).
Natural Resources
Management in African Agriculture. CABI Publishing, New York.
Oldeman, L.R. 1998.
Soil Degradation: A Threat to Food Security? Report 98/01,
Wageningen, the Netherlands: International Soil Reference and Information Centre.
Oldeman, L.R., R.T.A. Hakkeling, and W.G. Sombroek. 1991.
World Map of the Status of
Human-Induced Soil Degradation: A Brief Explanatory Note. International Soil
Reference and Information Centre and United Nations Environment Programme.
Olson, K., R. Lal, and L., Vorton. 1994. Evaluation of methods to study soil erosion-
productivity relationships.
Journal of Soil and Water conservation 49(6):586-590.
Pagiola, S. 1992. Estimating the yield effect of soil degradation. Journal of Soil and Water
Conservation 47: 486-490.
Pagiola, Stefano. 1996. Price policy and returns to soil conservation in semi-arid Kenya.”
Environment and Resource Economics 8: 255-271.
Pagiola, S. 1999. The global environmental benefits of land degradation control on
agricultural land. World Bank Environment Paper No. 16. Washington, D.C.: The
World Bank.
Pagiola, S. 1993. Soil conservation and the sustainability of agricultural production. PhD
Dissertation. California: Stanford University.
Pagiola, Stefano, and Mohamed Bendaoud. 1994.
Long-Run Economic Effects of Erosion on
Wheat Production in a Semi-Arid Region of Morocco: A Simulation Analysis.
Agricultural Economics Staff Paper No. 95-12. Washington State University.
Pender J., and J. Kerr. 1998. Determinants of farmers’ indigenous soil and water conservation
investments in semi-arid India.
Agricultural Economics 19:113-125.
Pierce, F.J., W.E. Larson, R.H. Dowdy, and W.A.P. Graham. 1983. Productivity of soils:
assessing long-term changes due to erosion.
Journal of Soil and Water Conservation
38(1): 39-44.
Pimentel, D., C. Harvey, P. Resosudarmo, K. Sinclair, D. Kurz, M. McNair, S. Crist, L.
Shpritz, L. Fitton, R. Saffouri, and R. Blair (1995). Environmental and economic
costs of soil erosion and conservation benefits.
Science 267: 1117-1123.
Reid, R.S., R.L. Kruska, N. Muthui, A. Taye, S. Wotton, C.J. Wilson and W. Mulatu. 2000.
Land-use and land-cover dynamics in response to changes in climatic, biological and
socio-political forces: the case of southwestern Ethiopia.
Landscape Ecology 15(4):
339 – 355.
Scherr, S. 1999. Soil degradation: a threat to developing-country food security by 2020?
Food, Agriculture and the Environment Discussion Paper 27. Washington DC:
International Food Policy Research Institute.
Scherr, S., and S. Yadav. 1996. Land degradation in the developing world: Implications for
food, agriculture, and the environment to 2020. Food, Agriculture and the
Environment Discussion Paper 14. Washington DC: International Food Policy
Research Institute.
Shiferaw, B. and S. Holden 2001. “Farm-level benefits to investments for mitigating land
degradation: empirical evidence for Ethiopia.”
Environment and Development
Economics, Vol. 6, pp. 336-359.
Smaling, E., L. Fresco, and A. De Jager. 1996. Classifying, monitoring and improving soil
nutrient stocks and flows in African agriculture.
Ambio 25(8): 492-496.
Smaling, E.M.A., S.M. Nandwa and B.H. Jansen (1997) Soil fertility in Africa at stake. In:
R.G. Buresh, P.A. Sanchez and F Calhoun (eds) Replenishing Soil Fertility in
Africa. Soil Science Society of America; Special Publication, No 51, pp. 47-61.
Sonneveld, B. G. J. S. 2002. Land Under pressure: The Impact of Water Erosion on Food
Production in Ethiopia. Shaker Publishing (PhD dissertation). Netherlands.
Stocking, M. 1995. Soil erosion in developing countries: where geomorphology fears to
Catena 25: 253-267.
Stocking, M., 1986. The cost of soil erosion in Zimbabwe in terms of the loss of three major
nutrients. Consultant's Working Paper No. 3. FAO, Rome.
Stoorvogel J., E. Smaling, and B. Janssen. 1993. Calculating soil nutrient balances in Africa
at different scales.
Fertilizer Research 35: 227-235.
Stoorvogel, J.J. and E.M.A. Smaling (1998) Research on soil fertility decline in tropical
environments: integration of spatial scales. Nutrient Cycling in Agroecosystems,
vol. 50, pp. 153-160.
Sutcliffe, J. P. 1993. Economic assessment of land degradation in the Ethiopian highlands: a
case study. Addis Ababa: National Conservation Strategy Secretariat, Ministry of
Planning and Economic Development, Transitional Government of Ethiopia.
Demissie, D.E. Storm, M. White. 2004. A GIS-based watershed modeling and
water resources use optimization under dynamic changing land use/land cover
conditions. Paper number 042023, 2004 ASAE Annual Meeting. American
Society of Agricultural and Biological Engineers, St. Joseph, Michigan.
UNEP/GRID. 1992. World Atlas of Desertification. Edward Amold: A division of Holder
and Stoughton, London.
WBISPP, 2001a. Gambella Regional State: A strategic plan for the sustainable development,
conservation, and management of the woody biomass resources, Final Report, May.
WBISPP, 2001b. Harari People’s Regional State: A strategic plan for the sustainable
development, conservation, and management of the woody biomass resources, Final
Report, May.
WBISPP, 2001c. Southern Nations Nationalities and People’s (SNNP) Regional State: A
strategic plan for the sustainable development, conservation, and management of the
woody biomass resources, Final Report, May.
WBISPP, 2002. Amhara Regional State: A strategic plan for the sustainable development,
conservation, and management of the woody biomass resources, Final Report,
WBISPP, 2003a. Beneshangul-Gumuz Regional State: A strategic plan for the sustainable
development, conservation, and management of the woody biomass resources, Final
Report, December.
WBISPP, 2003b. Tigray Regional State: A strategic plan for the sustainable development,
conservation, and management of the woody biomass resources, Final Report,
WBISPP, 2004a. Afar Regional State: A strategic plan for the sustainable development,
conservation, and management of the woody biomass resources, Final Report,
WBISPP, 2004b. Dire Dawa Administrative Council: A strategic plan for the sustainable
development, conservation, and management of the woody biomass resources, Final
Report, February.
Weibe, K. 2003.
Linking Land Quality, Agricultural Productivity and Food Security.
Agricultural Economic Report No. 823, Economic Research Service, U.S.
Department of Agriculture.
Wischmeier, W., and D. Smith. 1978. Predicting rainfall erosion losses. Agricultural
Handbook 537. USDA, Washington, DC.
World Bank 1988. Madagascar--Environmental Action Plan. Vol I. In cooperation with
USAID, Coop. S., UNESCO, UNDP and WWF. The World Bank, Washington, DC.
World Bank, 1992. Malawi--Economic Report on Environmental Policy. Vols. I and I1.
Report No. 9888-MA1. The World Bank, Washington, DC.
Wright, C. and Y. Adamseged. 1984. Ethiopian highlands reclamation study: an assessment
of the causes, severity, extent and probably consequences of degradation in the
Ethiopian highlands, Working Paper 3. Land Use Planning and Regulatory
Department, Ministry of Agriculture, Government of Ethiopia, and Food and
Agriculture Organization of the United Nations.
Appendix 1. Definitions of measures of economic losses
Bojö and Cassells (1995: 12-13) use the following definitions of the three measures of
economic losses they report in their work.
1. Gross annual immediate loss (GAIL): This is the loss of gross agricultural output in a
single year due to land degradation in the same period of time. The comparative
format is generally if no land degradation took place. The concept is “gross” because
it does not account for (1) the cost of combating land degradation or (2) any cost
reductions related to lower production.
2. Gross discounted future loss (GDFL): Because the loss of soil particles is irreversible,
the loss in any one year will affect production in all future years. In other words, this
concept captures a loss of natural capital and is relevant for any discussion on
adjusting the System of National Accounts to better reflect environmental damage.
The value of future costs is translated to a current value using a discount rate (r). For
an infinite time horizon, the GDFL=GAIL/r. This concept does not apply to nutrient
3. Gross discounted cumulative loss (GDCL): This concept captures the fact that land
degradation is a cumulative process, in which each year’s erosion and nutrient
removal is followed by another, adding layers of cost on top of each other. This
concept also applies to nutrient breaches as there will be a stream of GAILs probably
increasing due to population growth and mounting scarcity of wood fuel.
Appendix 2: Technical comments on Sonneveld’s (2002) work
1. Empirical approach
The production function specification (particularly specification of the
explanatory variables) is incomplete, as they do not include at least the traditional
production inputs (labor, fertilizer, soil conservation, etc), likely leading to
omitted variable bias. Given the average number of observations (map units) of
8.8 per administrative area, it would have been possible to use fixed effects
(dummy variables for Aurajas), which would have helped to control for omitted
variables. The dependent variable is specified only as a function of the SD index
and soil fertility and was adjusted by the potential yield assuming that this
captures actual yield variations due to agro-ecological and crop-genetic
differences. However, this specification may only correct for inter-CPSZ
differences but will not capture intra (within) CPSZ variations due to variations in
agro-ecological, crop genetic traits, cropping pattern, infrastructure, population
density, landforms, land use and management that lead to yield and soil
degradation variation within each CPSZ, leading to aggregation bias.
Another problem related to the specification of the dependent variable was the
price used to convert physical yield into monetary values. The years where the
yield observations were taken (mid 1980s to 1995) and the prices (average of
1993 and 1994) used to convert physical yield into value do not coincide.
2. Engineering approach
The specification of the dependent variable was the same as the empirical
approach and thus the engineering approach encountered the same problem. The
weighting of livestock production only by grazing lands assuming livestock feed
source is only grazing lands overestimates livestock productivity and carry
capacity of grazing land since the author ignored livestock dependence for feed on
cultivated land (straw and stubble).
In the yield function only labor input (population in each CPSZ dived by potential
yield per ha times total area of cultivated land) was considered. First, this is a
crude representation of labor, as it does not differentiate rural labor into active and
non-active labor. Secondly, it is also not clear why the population in each CPSZ
after correcting for urban population is divided by total potential production in
order to derive the labor input, which is different from the traditional ways of
specifying labor (labor per ha). Thirdly, input variables such as conservation
inputs and agrochemical are not included in the yield function.
It was not clear which input costs were considered in the optimization model.
Sonneveld used a production function (Mitscherlich-baule function) different from
the familiar production functions. It’s not clear how well this model specification
fits the data; no specification tests reported or alternative specifications
Appendix 3. Estimates of soil loss, nutrient losses and crop losses due to land
degradation by WBISPP
Table 3A Soil erosion classes by region (mm or tons/ha/year removal of soil)
<0.25mm 0.25-0.5
<3.125 3.125-
12.5-25.0 25.0-50.0 50.0-
100-200 200-400
tons tons Tons tons Tons Tons tons
63% 10% 9% 8% 6% 3% 1% 0%
88% 5% 4% 1% 1% 0% 0% 0%
37% 4% 19% 26% 3%*
of total
area of
77% 7% 8% 6% 2% 1% 0% 0%
37% 8% 12% 19% 14% 7% 3% 0%
30% 13% 14% 24% 10% 3% 4% 0%
47% 11% 19% 15% 6% 1% 1% 0%
*This figure applies to 25-200 tons of soil loss class
Source: Various reports of WBISPP.
Table 3B Gross and net soil loss by region in millions of tons per year
Region Gross soil loss Net soil loss*
Amhara 147 25
5 0.8
Tigray 27 5
*Net soil loss is gross soil loss less soil re-deposited within the landscape.
Source: Various reports of WBISPP.
Table 3C
Estimated annual and cumulative reductions in crop production (tons of grain) 2000-2025 due to soil
loss on cultivated land within the critical soil depths by region
Region 2000 2005 2010 2015 2020 2025
9,726 27,985 49,995 71,567 92,699 113,273
509 1,413 2,451 3,400 4,270 5,087
1,324 3,906 7,025 10,030 12,930 15,803
Source: Various reports of WBISPP.
Table 3D Estimated loss of crop nutrients due to the burning of crop residues and dung in tons
Total Total
Urea .
Crop residues Dung
equiv. (P)
(N) (P) (N)
30,787 3,079 80,053 25,160 110,840 28,238 203,030
836 84 35 11 871 94 2,003
88 9 203
14580 1458 41621
1,277 128 9,263 2,911 10,541 3,039 24,112
Source: Various reports of WBISPP.
Table 3E
Estimated on-farm plant nutrient losses due to the burning of crop residues and dung
(kg per farm*)
Total Total
Crop residues Dung
(N) (P) (N)
10 1 26 8 32 8 73
3.5 0.4 0.32 0.1 8.04 0.87 18
3 0 20 6 23 7 52
*A farm here is defined as a farm family.
Source: Various reports of WBISPP.
... With a population of about 107.53 million (estimated as of December 2018) growing at an annual rate of 2.46%, Ethiopia is the most populous landlocked country in the continent of Africa, and the second-most populous nation in Africa [20]. Agriculture sector, which accounts for about 50% of the Gross Domestic Product (GDP), 85% of the total export revenue, and over 80% of the total employment, is the main source of the country's economy [21][22][23][24][25]. The great majority of the population is dependent on subsistence agriculture that is an overwhelming vulnerable to the recurrent droughts and land degradation [21][22][23][24][25][26][27][28]. ...
... Agriculture sector, which accounts for about 50% of the Gross Domestic Product (GDP), 85% of the total export revenue, and over 80% of the total employment, is the main source of the country's economy [21][22][23][24][25]. The great majority of the population is dependent on subsistence agriculture that is an overwhelming vulnerable to the recurrent droughts and land degradation [21][22][23][24][25][26][27][28]. Rapid population increase and growing demand posed a greater pressure on land resources, leading to severe soil erosion and land degradation in various parts of the country. ...
... Despite conservation measures taken over the past decades, land degradation is continued to threaten crop production and land productivity potential, and negatively affecting livelihood systems, food security, and the country's economy [29][30][31][32]. It was estimated that the land degradation cost to an annual agricultural GDP range from 2% to 6.75% [21]. The loss of topsoil by water erosion in Ethiopia was estimated at 1.5 billion tons per annum with a mean erosion rate of 42 t ha −1 y −1 [27,28]. ...
Full-text available
Land use and land cover change (LULCC) is a critical factor for enhancing the soil erosion risk and land degradation process in the Wabi Shebelle Basin. Up-to-date spatial and statistical data on basin-wide erosion rates can provide an important basis for planning and conservation of soil and water ecosystems. The objectives of this study were to examine the magnitude of LULCC and consequent changes in the spatial extent of soil erosion risk, and identify priority areas for Soil and Water Conservation (SWC) in the Erer Sub-Basin, Wabi Shebelle Basin, Ethiopia. The soil loss rates were estimated using an empirical prediction model of the Revised Universal Soil Loss Equation (RUSLE) outlined in the ArcGIS environment. The estimated total annual actual soil loss at the sub-basin level was 1.01 million tons in 2000 and 1.52 million tons in 2018 with a mean erosion rate of 75.85 t ha−1 y−1 and 107.07 t ha−1 y−1, respectively. The most extensive soil loss rates were estimated in croplands and bare land cover, with a mean soil loss rate of 37.60 t ha−1 y−1 and 15.78 t ha−1 y−1, respectively. The soil erosion risk has increased by 18.28% of the total area, and decreased by 15.93%, showing that the overall soil erosion situation is worsening in the study area. We determined SWC priority areas using a Multi Criteria Decision Rule (MCDR) approach, indicating that the top three levels identified for intense SWC account for about 2.50%, 2.38%, and 2.14%, respectively. These priority levels are typically situated along the steep slopes in Babile, Fedis, Fik, Gursum, Gola Oda, Haramaya, Jarso, and Kombolcha districts that need emergency SWC measures.
... Soil erosion is the one among major threats to the sustainability of environment and productive capacity of agriculture (Yang et al., 2003;Feng et al., 2010), which makes plant shallow root depth, removes plant nutrients and results in moisture loss (Mahmud et al., 2005). It results in serious food insecurity in many developing countries as it depletes productive soils (Blanco and Lal, 2008). ...
... In this case, the probable reason could be this study watershed made up of low proportion of forest cover than the watersheds in comparison. The soil loss rate is higher with respect to the soil formation rate for Ethiopia 2 t ha -1 yr-1 , which is very low compared to soil erosion rates (Mahmud et al., 2005). ...
Full-text available
Soil loss by runoff is a severe and continuous ecological problem in Neri watershed. It affects agricultural production, drinking water quality, ecosystem health, and landscape aesthetics. This study was conducted to assess spatial risk of soil by estimating the average annual soil loss, and creating a soil erosion hazard map for this watershed. Revised Universal Soil Loss Equation (RUSLE) model supported by a GIS framework was used. To this end, data for the model parameters were derived from, a digital elevation model (DEM) of (30*30 m), thirty years (1988-2017) rainfall data at rain gauge stations, soil erodibility data from field soil samples, Landsat-8 satellite image for cover management and conservation support practices. Once raster layer of the input parameters created, overlay analysis was carried to assess the spatial distribution of soil loss. The estimated annual soil loss varies from 0 to 465.16 t ha−1 yr−1 with a mean annual soil loss of 9.95 t ha−1 yr−1. The empirical analysis also confirmed that the watershed losses a total of about 463365.46 t of soil annually. Out of the total area; 54.9% of the watershed was categorized below moderate classes and the remaining 45.1% of land area was classified under high to very high classes, which is about several times the maximum tolerable soil loss rate. Based on the classification of soil loss hazard map, six out of eleven Neri sub-watersheds need prior intervention in terms of integrated cover-management and mechanical conservation measures. Furthermore, model results can be refined by analyzing along with sub-watersheds level real time monitoring for conservation practices.
... ship between environment and development in general, as well as low public participation and community-based organizations in environmental management efforts. Furthermore, bad farming methods, along with a lack of awareness and consciousness, contribute significantly to the degradation of natural resources, such as forest destruction, soil degradation, and water resource degradation (Girma, 2001;Pender et al., 2002;Mahmud et al., 2005;MoARD & WB, 2007). ...
Full-text available
Though there is plenty of theoretical evidence supporting the economic and social reasons of biodiversity loss, empirical evidence for the majority of these links is sparse, if not non-existent. The loss of habitats, the introduction of foreign species, over-harvesting of biodiversity resources, and species homogeneity in agriculture are all important biological drivers for the loss of biological diversity. All of these factors have one thing in common: they are all human-driven. More research is needed in this area. It is also debatable and questioned if existing biodiversity-conservation strategies provide adequate responses to these core causes of biodiversity loss and are capable of effectively counteracting the loss of biodiversity related cultural values, biological species, and ecosystems. This review study provides an overview of the economic and societal factors that contribute to biodiversity loss in Ethiopia, as well as prospective opportunities. It also indicates the obstacles and future directions that should be implemented. Only theoretical considerations and overviews of current estimations are used in the analysis. Better promotion of practical conservation methods, community-based management techniques, and sector-based conservation and integration should be applied throughout the entire resource region to scale up biodiversity conservation loss.
... Land degradation, deforestation and trees cutting for heating and cooking is very common practice in most of the rural areas of the African countries [1]. Increasing water yields and crop productivity are among the positive impacts of reducing soil erosion, deforestation and consequently sediment loads to streams [2]. ...
Full-text available
Recently, increasing concern in Africa has been registered on the potential impacts on base flow and stream flow due to the implementation of watershed management interventions. This research incorporates spatially monthly geographical hydrological data sets into a developed spreadsheet water balance model to estimate the changes in surface runoff, base flow and stream flow as a result of implementing watershed management interventions in the Abbay River Basin during the period (2010-2018). The model was implemented at sub-catchment level. Considering Year 2005 as a datum for watershed management intervention, results of the modeling and spatial analysis indicated that watershed management interventions relatively reduced surface runoff, increased deep infiltration to groundwater and accordingly increased base flow to the stream. Among the key results in the Abbay Basin that change from Year 2010 to Year 2018 was reduction of surface runoff from the Abbay Basin in the amount of 1,753 million m 3 /yr. As a result, an increase in annual base flow in the amount of 23 million m 3 /yr was estimated, leaving a reduction in annual mean flow of the Abbay River in the amount of 1,731 million m 3 /yr of the Abbay River at El-Diem site. It is expected that the flow of the Abbay River will continue to decrease due to the continuous implementation of the watershed management interventions and agriculture expansion in the Abbay Basin. Index Terms-Abbay river basin, base flow, hydrological modeling, GIS & RS, soil moisture water holding capacity, surface runoff, stream flow, watershed management interventions.
... In Ethiopia, land degradation is the most severe problem that affects agricultural productivity and negatively affects food security [31,35,120,121]. According to the World Bank statistical estimate, land degradation cost to annual agricultural GDP ranges from 2% to 6.75% [122]. To reduce the problem, the Ethiopian government adopted intensive SWC at the national level [35,42,43]. ...
Full-text available
Soil erosion by water has accelerated over recent decades due to non-sustainable land use practices resulting in substantial land degradation processes. Spatially explicit information on soil erosion is critical for the development and implementation of appropriate Soil and Water Conservation (SWC) measures. The objectives of this study were to estimate the magnitude of soil loss rate, assess the change of erosion risk, and elucidate their implication for SWC planning in the Gobele Watershed, East Hararghe Zone, Ethiopia. Applying remote sensing data, the study first derived the Revised Universal Soil Loss Equation (RUSLE) model parameters in an ArcGIS environment and estimated the soil loss rates. The estimated total soil loss in the watershed was 1,390,130.48 tons in 2000 and 1,022,445.09 tons in 2016 with a mean erosion rate of 51.04 t ha−1 yr−1 and 34.26 t ha−1 yr−1, respectively. The study area was divided into eight erosion risk classes ranging from very low to extremely high. We established a change detection matrix of the soil erosion risk classes between 2000 and 2016. The change analysis results have revealed that about 70.80% of the soil erosion risk areas remained unchanged, 19.67% increase in total area, and 9.53% decreased, showing an overall worsening of the situation. We identified and mapped areas with a higher and increasing erosion risk as SWC priority areas using a Multi-criteria Decision Rules (MCDR) method. The top three priority levels marked for the emergency SWC measures account for about 0.04%, 0.49%, and 0.83%, respectively. These priority levels are situated along the steep slope areas in the north, northwest, south, and southeast of the Gobele Watershed. It is, thus, very critical to undertake proper intervention measures in upslope areas based on the priority levels to establish sustainable watershed management in the study area. Keywords: ArcGIS; soil erosion risk; SWC; remote sensing; RUSLE; MCDR.
... Deforestation and forest degradation directly and indirectly affect livelihoods and food security of local communities who rely on the forest and its ecosystem services. These losses to deforestation were estimated at about USD 23 million (Yesuf et al., 2005), and the annual cost of land degradation associated with land use and cover change in Ethiopia is estimated at about USD 4.3 billion (Gebreselassie et al., 2013). ...
Full-text available
Introduction of the 7th Chapter Soils perform essentially five functions: biochemical and geochemical cycling, partitioning of water, storage and release, buffering, and energy production. These functions are essential for the support of ecosystems. The finite soil resource with its various functions is of direct and indirect benefit to humankind and the environment. In the context of this chapter, soil degradation is defined as a “process which lowers the current or the potential capability of soil to produce goods or services.” Adverse ecological and economic developments that are, at least in part, a result of soil degradation can be seen in Ethiopia and many other regions of the world. According to UNEP and World Soil Information (ISRIC), 24% of the world’s soils on settled land are already greatly impaired as a result of anthropogenic degradation, inappropriate soil use, the inputs of pollutants, and structural soil disturbance arising from infrastructure development and the building of housing. Soil degradation can have dire consequences for humans’ wellbeing such as income loss, food insecurity, and health risks. All forms of soil degradation, whether biological, chemical, or physical, occur across Ethiopia at different scales and levels of severity. Extensive experience in soil and water conservation (SWC) shows that land degradation can be tackled if SWC measures are designed and implemented with respect to local biophysical conditions.
... Deforestation and forest degradation directly and indirectly affect livelihoods and food security of local communities who rely on the forest and its ecosystem services. These losses to deforestation were estimated at about USD 23 million (Yesuf et al., 2005), and the annual cost of land degradation associated with land use and cover change in Ethiopia is estimated at about USD 4.3 billion (Gebreselassie et al., 2013). ...
... The population increase and the economy activities cause the increase of the over land function. Then the inefficency cost of the degraded land as the provider of the environmental services [7]. Taking for example, estimating the annual cost of the degraded land in the Mid Asia's villages, due to the land use and the change of the field in between 2001 and 2009 is about 6 billion USD, largely due to the desert degradation (4.6 billion USD), followed by the desertification (0.8 billion USD), deforestation (0.3 billion USD) and the abandon farm field (0.1 billion USD) [8]. ...
Technical Report
Full-text available
The Country Environmental Analysis (CEA) assesses the country’s key environmental challenges in this transformation process and the capacity to manage them. It takes a holistic perspective on environmental issues, covering the major sectors of the economy. The CEA focuses on four clusters of sectors that, together, are determining Ethiopia’s development path: (i) resilient rural landscapes, (ii) green industrialization, (iii) sustainable urbanization, transport and living conditions, and (iv) sustainable energy access. Achieving Ethiopia’s CRGE goals requires shifting the trajectory that each cluster is on toward more sustainable options. For each of these clusters, the CEA discusses their current and intended trajectories (environmental trends such as degradation and pollution, and the government plans and strategies to respond to these trends) as well as the policies, incentives, institutions, information (the upstream enabling environment) needed to …
Full-text available
Land use and land cover change (LULCC) is a critical factor for enhancing the soil erosion risk and land degradation process in the Wabi Shebelle Basin. Up-to-date spatial and statistical data on basin-wide erosion rates can provide an important basis for planning and conservation of soil and water ecosystems. The objectives of this study were to examine the magnitude of LULCC and consequent changes in the spatial extent of soil erosion risk, and identify priority areas for Soil and Water Conservation (SWC) in the Erer Sub-Basin, Wabi Shebelle Basin, Ethiopia. The soil loss rates were estimated using an empirical prediction model of the Revised Universal Soil Loss Equation (RUSLE) outlined in the ArcGIS environment. The estimated total annual actual soil loss at the sub-basin level was 1.01 million tons in 2000 and 1.52 million tons in 2018 with a mean erosion rate of 75.85 t ha–1 y–1 and 107.07 t ha–1 y–1, respectively. The most extensive soil loss rates were estimated in croplands and bare land cover, with a mean soil loss rate of 37.60 t ha–1 y–1 and 15.78 t ha−1 y−1, respectively. The soil erosion risk has increased by 18.28% of the total area, and decreased by 15.93%, showing that the overall soil erosion situation is worsening in the study area. We determined SWC priority areas using a Multi Criteria Decision Rule (MCDR) approach, indicates that the top three levels identified for intense SWC account for about 2.50%, 2.38%, and 2.14%, respectively. These priority levels are typically situated along the steep slopes in Babile, Fedis, Fik, Gursum, Gola Oda, Haramaya, Jarso, and Kombolcha districts that need emergency SWC measures.