r“ ‘ a
LIBRARY ’

“Mun-I State

1 University

x“; ,1

fl

 

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

5 [1T :3 r1200": ll

 

 

 

 

 

 

 

 

' 'l

w#_ _J

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

=‘1W

MSU Is An Affirmative Action/Equal Opportunity Institution
czgmmaa-pn

 

g

 

LOCATIONAL AND ECOLOGICAL ASPECTS OF PERI-URBAN FUELWOOD
PLANT ATIONS IN THE WEST AFRICAN SAHEL

By
E. Mark Pires

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF ARTS

Department of Geography

1990

ABSTRACT

LOCATION AL AND ECOLOGICAL ASPECTS OF PERI-URBAN FUELWOOD
PLANTATIONS IN THE WEST AFRICAN SAHEL

By
E. Mark Pires

Countries of the West African Sahel are dependent on fuelwood for meeting
their energy requirements. Traditional practices of collecting wood from areas
surrounding cities and towns have led to large—scale deforestation around many of the
region’s urban centers. An often-cited approach to assure urban energy needs, reduce
deforestation, and arrest environmental degradation in rural areas is the establishment of

fuelwood plantations.

The main objective of this study is to examine the conditions under which
fuelwood could be produced as a crop in a pen-urban plantation system. Concepts
related to 'I‘hunen’s model of agricultural land use are employed to determine the best
location for the establishment of a fuelwood plantation. Data are analyzed for
agricultural crops and a hypothetical plantation representative of the conditions in this
semi-arid area. Results indicate that for plantations to be economically viable fuelwood
market prices must be substantially higher than current levels. Economic and political
difficulties posed by higher fuelwood prices in urban areas render the establishment of

peri-urban plantations problematic.

ACKNOWLEDGMENTS

I wish to express my gratitude to the members of my thesis committee,
Professor David Campbell and Professor Assefa Mehretu, for their encouragement and
support in my undertaking this research. Professor Campbell, my advisor and thesis
director, has been a constant source of ideas, energy, and enthusiasm throughout.
Professor Mehretu has been instrumental in helping me to gain a better understanding of
location theory, particularly with regards to von Thunen’s model. Both have
contributed significantly to my increased understanding of the problems and challenges
now facing the African continent.

I would also like to thank Professors Bruce Pigozzi and Jack Williams for their
efforts in reading the thesis and participating in the final oral examination. I am
grateful for their questions, comments, and constructive criticism.

I am most thankful for the financial support for my graduate studies received
from the Department of Geography at Michigan State University, in addition to the
award of a U.S. Department of Education Title VI Foreign Language and Area Studies
Fellowship, administered by the African Studies Center at MSU.

My sincere thanks go to good friend Charlie Rader for his expert assistance in
creating Figure 2.2 and the graphics in Chapter IV, and for his valuable advice and help
with bringing me one step closer to computer literacy. Finally, I owe a special debt of
appreciation to Linda Beck, my companion and confidante. I am grateful for the long
hours she spent diligently editing earlier drafts of the thesis, and for her love and
support that always see me through.

iii

List of Tables

List of Figures

CHAPTER

I. INTRODUCTION

A.
B.
C.

IV. ANALYSIS OF DATA

TABLE OF CONTENTS

 

 

mm on e?

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Page
vi
vii
1
Statement of the Problem 2
II. REVIEW OF LITERATURE 4
Overview of the Study Area 4
A Theoretical Framework for Natural
Resource Management 9
Fuelwood Production and Consumption 13
Fuelwood Situation in Sahelian Urban
Areas 19
Strategies to Address Fuelwood Deficits 27
Need for Present Study 32
III. METHODS AND DATA COLLECTTON 35
The Thunen Model as a Method of Analysis 36
Data Collection 40
Calculations Used in the Analysis of Data 45
47
Agricultural Crop Data 47
Fuelwood Data 48
Calculation of Rent Gradients 53
Locating the Firewood/Charcoal Threshold 62
Preliminary Interpretation of Results 65

mpnw?

 

iv

 

 

 

 

 

DISCUSSION

A. Evaluation of the Thunen Model Application
B. Economic Impacts of Fuelwood Plantations
C. Ecological Considerations

D. Social Constraints and Opportunities

E. Policy Problems

 

SUMMARY AND CONCLUSIONS

 

LIST OF REFERENCES

 

Table
2.1
2.2
2.3

4. l
4.3
4.4
4.5
4.6

4.7

LIST OF TABLES

World Fuelwood Production 1985

 

Fuelwood Consumption Compared with the
Mean Annual Increment of Tree Stocks in the
Sahel

 

Population Growth in the Sahel

 

Agricultural Crop Data

 

Estimated Fuelwood Plantation Costs

 

Shadow Price for Fuelwood Based on
Kerosene Substitute

 

Maximum Rent and Distance for Crops

 

Conditions for Profitable Plantation Production
Maximum Rent and Distance for Fuelwood
Plantations

 

 

Production Cost of Firewood and Charcoal
Per Ton CE Energy

 

vi

13

18
21

48
50
52
56
57

LIST OF FIGURES

Principal Countries of the West African Sahel

 

Ecological Zones of the West African Sahel

 

Fuelwood Situation in Developing Countries

 

Fuelwood Supply Model for Dryland Population
Centers

 

Location Rent Gradients for Competing Crops

 

Land Use Zones in Thunen’s Isolated State

 

Administrative Map of Mali

 

 

Rent Gradients for Agricultural Crops
Rent Gradients for Fuelwood Plantations

 

Combined Rent Gradients

 

Optimal Fuelwood Plantation Rent Gradients

 

vii

CHAPTER I
INTRODUCTION

Wood has served humankind’s energy needs over the millennia. Its importance
as a source of energy has waned in societies that have gone through the stages of
economic development which have led to the use of fossil fuels such as coal, oil, and
natural gas. However, for the overwhelming majority of people in the developing
world, dependence on wood for satisfying energy needs remains prevalent and a decline

in such dependence is generally assumed to be unlikely in the near future.

In Africa, most nations depend heavily on wood-based fuels to meet their energy
requirements. This is especially true in the domestic sector, where firewood and
charcoal account for virtually all of the energy consumed for cooking and heating. In
the countries of the West African Sahel, an ecological zone located between the Sahara
desert to the north and the more humid subtropical savannahs to the south, the
environmental consequences of fuelwood dependence have reached severe proportions.
Large-scale deforestation has taken place around most of the urban centers in this
region, resulting in expanding zones of degraded land surrounding large cities and
towns. Although the exploitation of fuelwood resources is only one of many factors of
land degradation in the environs of Sahelian urban areas, the continued dependence on

this energy source promises to exacerbate the ecological deterioration now in progress.

The current practice of fuelwood production mainly involves the collection,
processing, and transportation of what has hitherto been considered a ‘free good’.1
With respect to urban areas, this process begins first with the collection of wood from

the most proximate areas outside the city or town, followed by a progressive outward

 

1 The idea of fuelwood as a ‘free good’ is reflected in the fact that no consideration is
given to the replacement costs of trees felled to provide wood energy. In essence,
forested lands are being ‘mined’ in order to produce fuelwood energy.

1

2

expansion of the collection zone as nearby resources are depleted. In the most extreme
cases, fuelwood in the West African Sahel is now transported to urban markets over
distances of several hundred kilometers. Due to the fact that the supply of fuelwood to
urban markets must now overcome increasingly greater distances, this energy resource
is no longer a ‘free good’ for urban inhabitants but instead a monetized good that entails

a significant cash expenditure.
Statement of the Problem

This study entails an analysis of a fuelwood production system for meeting
domestic energy needs in the urban centers of the West African Sahelian region.
Emphasis will be placed on the locational and ecological aspects of a possible

alternative spatial pattern of fuelwood production in peri-urban environments.

The objective of this study is to demonstrate how one could determine an
economically and ecologically viable location for peri-urban fuelwood plantations
designed to supply the energy needs of urban centers in the countries of the West
African Sahel. Although rural populations are undeniably faced with equally pressing
problems regarding fuelwood provision, this analysis is intended to deal exclusively
with fuelwood supply systems for urban centers. The nature of rural fuelwood supply
differs from that of urban areas primarily due to the absence of a medium of exchange
between buyer and seller. However, in the urban context, the evolution of a complex
market apparatus for what has now become a monetized good requires an analytical
approach specific to the question of fuelwood consumption in cities and towns.
Nevertheless, there remains an important connection between the supply of energy from
forested rural areas and the attendant demand for fuelwood on the part of urban

populations. Thus, the present study will address relevant rural-urban components of

the fuelwood problem.

Questions examined in this study are related to the possible spatial
reorganization of urban fuelwood supply systems.2 For example, where, in terms of the
agricultural space economy, could fuelwood be profitably produced as a ‘crop’ in
competition with other types of land use were it no longer considered to be a ‘free’
good? In addition, what would be the economic impact, ecological ramifications, social
constraints and opportunities, and policy problems associated with a different spatial

pattern of fuelwood production?

For the purposes of this study, the term ‘fuelwood’ will be used to denote
firewood (cut logs, trunks, branches, etc.) and charcoal. In the literature, the terms

‘fuelwood’ and ‘woodfuel’ are for the most part used interchangeably.

 

2 Spatial reorganization refers here to a pattern of fuelwood production other than the
current practice of ‘mining’, without subsequent reforestation, of rural forested lands
located at increasing distances from urban market centers. '

CHAPTER II
REVIEW OF LITERATURE

This chapter serves as an introduction to the study area and as a review of recent
studies relevant to the exploitation of forest resources for the production of fuelwood
energy. The first section gives an overview of the study area, outlining its political
composition and ecological characteristics. In the second section, linkages between
politico-economic and ecological factors related to present environmental conditions in
the Sahel are discussed in the context of a conceptual framework for analyzing
problems of economic development and environmental protection. The third section
discusses the levels of production and consumption of fuelwood energy resources in the
countries of the study area. Following this, the fourth section examines the specifics of
fuelwood use in the urban context. As the problem of supplying fuelwood energy to
urban areas becomes evident, the fifth section discusses several strategies commonly

cited to address this problem.
Overview of the StudyLArea

The term ‘Sahel’ comes from the Arabic word for coast or border, which, in the
geographic context, refers to this region’s location along the southern extent of the
Sahara desert. Based on its political composition, the West African Sahel is comprised
of the following seven states: Burkina Faso, Chad, The Gambia, Mali, Mauritania,
Niger, and Senegal (see Figure 2.1). In 1985, the total population of these seven
countries was approximately 35 million inhabitants. Of this total, about 22 percent is

considered to live in urban areas (World Bank 1988).

 

 

 

 

 

 

 

 

 

 

 

 

Figure 2.1 Principal Countries of the West African Sahel
Source: Jayne, Day and Dregne 1989

6

The Sahel extends across West Africa defined primarily by its semi-arid
climatic characteristics and is subdivided into different zones based on precipitation
(World Bank 1985, p. 2). The Sahel is generally considered to extend from the
southern limit of the Sahara desert to the northern limit of the tropical savannah of West
Africa’s coastal zone. More specifically, in ecological terms, the Sahel proper is
bounded by the northern limit of rain-fed cultivation at approximately the 250 My
isohyet, and by the northern limit of the Sahelo-Sudanian zone at approximately the 400

mm/yr isohyet (see Figure 2.2).

Sinclair and Fryzell (1985) identify important physical characteristics of the
Sahel that are related to the region’s relatively poor biomass productivity. In the
Sahelian region, soils are generally of low fertility due to low mineral and organic
content, as well as poor water retention capacity. Vegetation consists mostly of open
‘ Acacia scrub and short grasses, which become increasingly scarce in the northern

reaches of the region.

Annual cycles of wet and dry seasons in the Sahel are the result of the seasonal
migration of the Intertropical Convergence Zone. This discontinuity between dry
continental and moist tropical air masses is the primary driving force controlling
precipitation in the Sahelian region. Since the beginning of the meteorological record
in the Sahel recurrent drought has been a predominant feature of the region, presumably
common in earlier times as well (Nicholson 1982). Oscillating periods of dry and
humid conditions spanning several thousands of years have been discerned through
archaeological research (National Academy of Science 1983a). Research on
contemporary conditions in the Sahel indicates that cycles of severe drought occur at
intervals of approximately thirty years; however, some suggest that human agency is

currently contributing to more frequent drought conditions (Rasmusson 1987).

 

 

    
     

 
   
   

ALGERIA

AFRCAN REPUBLE

 

 

Cllmetlc Zones
Saharan
Sahelo-Saharan
Sahelian
“”;5,._ Sahelo—Sudanian
1N Sudanian

Sudano-Guinean

 

-— eoo Isohyets in mm

 

w” m ANGOLA

 

 

 

Figure 2.2 Ecological Zones of the West African Sahel.

After: World Bank 1985

Agricultural activity in the Sahel is restricted mostly to the cultivation of
drought tolerant crops such as millet and sorghum, which are main staples in the drier
parts of the region. Other important crops grown in the Sahel include groundnuts,
irrigated rice, and, to a lesser degree, maize (Jayne et a1. 1989). Sugar cane - in
irrigated areas such as the Senegal River valley - and cotton are two significant cash
crops produced in the region. Perhaps more important than cultivation, in terms of
extensive land use, is the wide-spread nomadic pastoralism practiced in the Sahel (Swift
1977). Whereas crop production is localized in areas of highest agricultural potential,
cattle and other livestock are herded over a much larger spatial range. Competition
between farmers and herders for land and water resources is intense, particularly in
areas where population pressures begin to exceed local carrying capacities, resulting in

the exploitation of increasingly marginal areas.

As the work of Gritzner (1988) has detailed, these three factors — recurrent
drought, crop cultivation on marginal lands, and the demands of extensive pastoralism -
have contributed significantly to the environmental degradation apparent in the Sahel
‘ today. Due to the fact that the natural resource base of the Sahel is inherently limited,
and that population pressures continue to rise while levels of agricultural technology
remain relatively static, the increasing demands for food, forage, and water contribute
-to the deterioration of local ecological equilibrium. Under these conditions, it is argued
that many areas within the Sahel are losing their productive capacity to the insidious

process of desertification (Gorse and Steeds 1985; Grainger 1982).

In light of the scenario described above, one must also consider the contribution
of fuelwood demand toward greater environmental degradation. It is important to
realize that in addition to the demands for food, forage, and water, the limited Sahelian

resource base must also provide the overwhelming majority of the population’s energy

9

supply. Whereas most rural inhabitants satisfy their fuelwood requirements by
selectively culling wood from locally available resources, urban fuelwood demand is
usually met by the intensive cutting of remaining forested stands in remote rural areas.
Hence, energy requirements, and particularly those of rapidly expanding urban

populations, contribute significantly to overall environmental degradation.

The discussion in the next section indicates how the demand for fuelwood goes
beyond the specific issue of energy supply to touch upon broader concerns of overall
economic development. It is argued that fuelwood production, environmental
degradation, and the productive capacity of the natural resource base (hence the
economic development potential) are interconnected in an environment-economy-

energy nexus.

A Theoretical Framework for Natural Resource Management

Threatened by the specter of continued environmental degradation, the countries
of the Sahel must begin to devise improved strategies and implement effective plans of
action designed to stabilize and reverse the deterioration of their productive natural
resources. This remains difficult, however, for as Redclift (1987) has pointed out, most
theories of regional growth, decline, or development fail to incorporate environmental
considerations. He insists that those concerned with planning development must adapt
existing theories to take into account the variability of natural resources over space and

through time.

In attempting to provide a conceptual framework for incorporating natural
resource management issues into overarching theories of regional development, Blailde

and Brookfield (1987) seek to combine economic and ecological considerations. Their

10

main ideas are put forward within the context of what they term ‘regional political .

ecology’.

Regional political ecology seeks to link specific natural resource management
issues with broader issues of political economy. The ‘regional’ in this concept refers to
the spatial variations in capability, resilience, and sensitivity of land resources. The
term ‘political .ecology’ seeks to encompass concerns of ecology and political economy.
It attempts to examine the dialectic between society and land-based resources (Blaikie
and Brookfield 1987, p. 17). Regional political ecology also recognizes the importance
of the role played by the state in natural resource management. It explores the
dynamics of the allocation of public resources to land users, as well as the effects such

allocations have on natural resource use.

Kostrowicki (1986) complements the ideas put forth by Blaikie and Brookfield
when he discusses the notion of moving from a "technico-economic" approach, to an
"ecologico-economic" approach in reconciling environmental concerns with issues of
economic development. Kostrowicki acknowledges, as do Blaikie and Brookfield, that
problems of environmental degradation are, in effect, societal problems as much as they
are problems of physical circumstances of the biosphere. He sees the need to bring
together two diametrically opposed viewpoints on ways of seeking solutions to
problems of environment and development; he suggests that a solution may be more
effectively sought through the "ecologization of econorrrics and other social sciences,
[together with] the economization and humanization of the natural sciences"

(Kostrowicki 1986, pp. 5-6).

In addition to regional political ecology, Blaikie and Brookfield offer another
concept - that of ‘landesque capital’ - which attempts to bring together the two concerns

of environment and development. The idea of landesque capital basically represents the

11

production and/or safeguarding of capital in the fonrr of natural resources for future.
productive use. It is essentially an investment in land with an anticipated life beyond
that of the present crop cycle, or rotation period when dealing with forest resources
(Blaikie and Brookfield 1987, p.9). Landesque capital investments are considered as
any purposive land management intervention which has the specific intent to create
capital for future maintenance of land capability. Examples of such investment would
include soil stabilizing wind breaks or live fences, terracing on steep slopes, or the

design of proper field drainage systems.

These interventions necessitate a long-term perspective on the use of productive
natural resources. Unfortunately, many natural resource management decisions in
developing countries seek to maximize short-term gains. This is understandable in
environments where risks are high and production is plagued by many uncertainties.
However, long-term productive capacity is frequently placed in jeopardy due to the

short time horizon of such resource management decisions.

In the case of fuelwood resource exploitation in the Sahel, the regional political
ecology framework offers a useful context for examining how political decision making
affects the status of the natural resource base, the population living in rural collection
areas, and the urban consumer of fuelwood energy. For example, policies that allow
fuelwood mining in rural areas to continue unabated contribute not only to
environmental degradation, but also to the economic marginalization of farmers whose
land-based capital is adversely affected by increased soil erosion and the loss of other
service functions provided by trees (e.g., nitrogen fixation by leguminous species). It
follows that the marginalization of rural-based agricultural systems affects the status of

the overall national economy.

12

One could also argue that the lack of an effective resource management policy
regarding fuelwood production basically amounts to a government subsidy to urban
consumers on the price of fuelwood at the expense of rural farmers. This is especially
the case when there is little or no subsequent reforestation following fuelwood
collection. Hence, in the absence of effective management policies, the allocation of
natural resources for fuelwood production in many developing countries leads to an
inequitable distribution of the benefits and costs attributable to exploiting such

I'CSODI'CCS.

Without effective measures in place to conserve fuelwood energy, more realistic
prices for wood-based fuels, and better management of remaining tree stocks, Sahelian
countries will find it difficult to reverse the stress placed on natural productive
resources by current fuelwood exploitation practices. The lack of investment in future
land productivity presents a serious threat to the prospects of economic growth in

resource-poor Sahelian nations.

Within the regional political ecology framework, the interdependent variables of
economics, politics, and ecology can be brought together in order to seek more effective
solutions to problems facing the future prospects of development in the Sahel. The
current situation regarding fuelwood exploitation in this region is an example of the
type of issue that can be addressed by this holistic concept. This approach is consistent
with the suggestion that the societal goal of economic development, which includes
meeting energy requirements for domestic and industrial purposes, be met in an
environmentally sound and sustainable manner (Matthews and Siddiqi 1981). Thus, the
complexity of the fuelwood issue in developing countries must be seen in terms of its

interconnected environment, economy, and energy components.

13
Fuelwood Production and Consumption

The developing world’s dependence on wood-based fuels to meet domestic
energy requirements was not a major concern of environmental researchers or
development assistance organizations until fairly recently. In the wake of the
tumultuous repercussions of the 1973 oil crisis in the industrial nations, there was an
increase in research attention given to the "other energy crisis," namely that of

fuelwood (Eckholm 1975; Arnold and Jongma 1977).

Whereas industrial nations rely almost exclusively on conventional fuels such as
petroleum, coal, natural gas, hydroelectricity, and nuclear power generation,
approximately 90 percent of the peeple in most developing countries depend on

fuelwood as their chief source of energy (Eckholm 1984).

To illustrate the dependence of developing nations on fuelwood energy, the
following statistics contrast world production of fuelwood with that of the developing
world. In 1985, total world roundwood production was estimated at 3.2 billion cubic
meters, with fuelwood accounting for 53 percent of this figure (FAO 1986). Eighty-five
percent of the fuelwood production for the same year took place in developing countries

(see Table 2.1).

Table 2.1 World Fuelwood Production - 1985

 

(Thousand m3)
Developed countries 254,587
Developing countries 1,408,384
Total 1,662,971

Source: UNFAO, Yearbook of Forest Products, 1986.

 

l4

Fuelwood use represents approximately 5.4 percent of total world energy
consumption. Although this may appear to be a relatively small amount on a global
scale, estimates for total energy requirements met by fuelwood in developing countries
range from 50 to 80 percent (dc Montalembert and Clement 1983, p. 16). Some
countries, such as Burkina Faso and Nepal, are even more dependent on wood-based
fuels. In these two countries figures for wood as a percentage of total energy
consumption reach 94 and 98 percent respectively (Eckholm 1984, p. 11). This
dependence is most pronounced in the rural areas of developing countries, particularly
those of sub-Saharan Africa, where in many cases 100 percent of the domestic energy

requirement is met by fuelwood (de Montalembert and Clement 1983, p. 28).

The figures mentioned in the preceding paragraphs are to be interpreted with
caution. Quantifying the amount of fuelwood consumed in thousands of small, remote
villages around the globe is a difficult, if not impossible task. However, the point to be
stressed is that without viable alternative sources of energy to satisfy domestic needs,

most people in the developing world are highly dependent on fuelwood.

What is probably one of the most comprehensive attempts made to assess the
fuelwood supply situation in developing countries was reported by the United Nations
Food and Agriculture Organization (FAO) in a 1983 study (de Montalembert and
Clement 1983). The report gives a systematic overview of fuelwood production vis-a-
vis consumption requirements on a region by region basis for all developing countries.
Data were gathered on a country by country basis, and then analyzed and aggregated in
order to develop a global picture of the fuelwood situation. The results culminated in
an important update to an earlier FAO study which produced a map of the fuelwood

situation for developing countries (see Figure 2.3).

 
   

- 0mm on near: mu
m cum ecu-mutt

M mm: um"

cussrncanou or All“:
- armrest mutton

 

   

~.\\"‘\i\‘\a\)l‘ \
M 5 / . \‘ ‘
. ‘ ', v/ /

. k" V J
S N) ‘3
. «a

n 32
>2-

“ti-.9
% '
I . I ' I.

 

Countries

ing

in Develop

rtuation

Figure 2.3 Fuelwood S

UNFAO 1981

Source

16

In the study by de Montalembert and Clement, categories of varying severity
were developed in order to indicate the status of the fuelwood situation in a given
region. The categories and their definitions are given as follows:

Acute scarcity situations: zones or countries with a negative balance where the
fuelwood supply level is so notoriously inadequate that even overcuttin g of the resource
does not provide a sufficient supply and fuelwood consumption is, therefore, clearly
below minimum requirements.

Deficit situations: present level of supplies is insufficient to ensure provisioning
on a sustained basis, and overcutting leads to degradation or progressive destruction of
the resources.

Prospective deficit situations: zones or countries in which supplies still
exceeded demand in 1980, but which in 2000 will be in a deficit situation if present
trends continue.

Satisfactory situations: zones which on the whole will still have sufficient
supplies by the year 2000.

As indicated on the map in Figure 2.3, most of the territory in the West Afiican
Sahel is classified as zones of acute scarcity/deficit (western Senegal, southwestern
Niger, and central Burkina Faso), or prospective deficit (southwestern Mali).

According to the map, a small area classified as satisfactory is located in southeastern

Senegal, a region in which the population is relatively sparse and rainfall relatively

good.

In their consideration of why many areas are experiencing expanding fuelwood
deficits, de Montalembert and Clement cite the combined effects of 1) rapid population
growth, 2) deterioration of natural forest resources, and 3) the absence of viable
alternative energy sources. These points will be addressed in greater detail later in this

chapter.

Reports by Floor and Gorse (1988), and Bouttoud (1988), concentrate more

specifically on the fuelwood problem in countries of the West African Sahel. In

l7

addition, Keita (1982) employs a similar approach to that of de Montalembert and '
Clement in classifying the fuelwood situation for the seven countries of the Sahel
region. However, rather than reporting his findings on a country by country basis, Keita
divides the region into four ecological sub-divisions according to rainfall parameters.
These divisions are given as follows: 1) Saharan zone - less than 100 mm/yr, 2)
Sahelian zone - 100—400 mm/yr, 3) Soudanian zone - 400-1200 mm/yr, and 4) Soudano-
Guinean zone - 1200-1600 mm/yr.

Keita notes that of the sub-divisions defined above, the Soudanian zone (with 17
percent of the total land area and 60 percent of the total population of the region) is
experiencing the greatest degree of deforestation and land degradation. Unfortunately,
no attempt is made to estimate the proportion of this degradation due to fuelwood

consumption.

In his study, Keita describes the different spatial patterns of fuelwood supply for
various sized settlements in the Sahel. For rural villages, he notes a ‘fuelshed’
proscribed by a circle with an average radius of 10 km, within which firewood is
collected mostly by women and children. For small-town settlements, the average
fuelshed radius is within a range of 10-50 km. In these areas, fuelwood is usually
brought to the settlement by animal carts, or, in some instances, small vehicles. Major
urban centers are shown to have a minimum fuelshed radius of 50-60 km and a
maximum radius ranging from 200-400 km. Most of the fuelwood consumed in these
centers is in the form of charcoal and is transported from the collection site by large-

capacity trucks.

From the general conclusions of Keita’s study, the future fuelwood situation for
the Sahel is regarded as a crisis looming on the horizon. In light of projections

comparing future energy supply from natural tree growth with expanded fuelwood

l8

consumption (see Table 2.2), it appears that Keita’s conclusions warrant serious
attention. The author notes that due to problems of substituting conventional fuel
sources such as oil, coal, or natural gas, "no other source of energy will be able, in the
near future, to replace fuelwood products under appropriate technical and economic
conditions on a large scale" (Keita 1982, p. 51, my translation). In light of projected
supply and consumption, however, it is evident that dependence on fuelwood will

require innovative interventions in order to satisfy future energy demand.

Table 2.2 - Fuelwood Consumption Compared with the Mean Annual Increment of Tree

 

 

Stocks in the Sahel
Accessible Supply Fuelwood Excess of Consumption
from Tree Growth Consumption over Accessible Supply
Year 000m3/yr 000m3/yr 000m3/yr %
1980 17,000 22,000 5,000 30
2000 15,000 34,000 19,000 127

Source: Anderson and Fishwick 1984, p. 15.

 

Finally, in the 19808, researchers and development assistance agencies began to
pay greater attention to the problem of meeting increasing demands for fuelwood, and
to mitigating the environmental degradation ensuing from the ‘mining’ of forested
lands. Several writers examined the relationship between fuelwood consumption,
clearing of land for agriculture, and increasing rates of deforestation in Africa and other
developing regions (Allen and Barnes 1985; Anderson 1986; Anderson and Fishwick
1984; Bowonder 1985/86; French 1986; Gritzner 1982; Hosier 1988; and Pimentel et a1.

1986). In-depth studies and analyses of the situation began to examine the varied and

l9

interrelated effects of unsustainable fuelwood exploitation on both physical and human
environments at the local level in urban and rural areas, as well as on an overall national
level (Moss and Morgan 1981; Eckholm et a1. 1984; Bogach 1985; Timberlake 1985;
Agarwal 1986; Leach and Meams 1988; and Munslow et a1. 1988). These effects range
from soil erosion and watershed deterioration, to poorer human health and declining

rural incomes, to reduced agricultural productivity and gross domestic product.

The broad range of interrelationships between fuelwood resource exploitation
and the condition of natural and human environments spans several contextual and
scalar levels of analysis. The problems of meeting fuelwood energy requirements can
thus be examined from ecological, political, social, and economic perspectives. In
addition, analyses can be performed at scales ranging from the local to the national

level.

Fuelwood Situation Around Sahelian Urban Areas

As in most of sub-Saharan Africa, wood-based energy makes up the majority of
primary energy supply in the Sahelian region. On average, the amount of energy
produced from fuelwood accounts for two-thirds of the national energy budget in most
sub-Saharan countries (World Bank 1989, pp. 128-130). This section examines more
closely the important characteristics of fuelwood use at the urban level within the

Sahelian region.

Major distinctions exist between energy consumption in urban and rural areas.
First, the majority of fossil fuel energy consumed on a national basis takes place in the
urban areas. It is here that the majority of industry is located, where fuel needs for

transportation are most concentrated, and where the vast majority of electricity is

20

consumed. For example, in Dakar, Senegal, one-sixth of the country’s population '
comprises 90 percent of the nation’s electricity consumers, who purchase 55 percent of
the total national electricity output (Di-Meo et a1. 1985, pp. 1-2). In addition, 75
percent of imported petroleum products are destined to meet the local demand of the
greater Dakar area.

Second, fuelwood meets almost all rural energy requirements for domestic
purposes, e.g., cooking, heating, and myriad other uses. In urban areas, fuelwood meets
approximately 90 percent of domestic energy requirements. However, there is a
significant difference between rural and urban areas in terms of the form in which
wood-based energy is consumed. In rural areas, fuelwood is almost always consumed
in the form of firewood - mostly dead, lapped-off branches of trees collected from
around villagers’ homesteads.1 On the other hand, almost all of the fuelwood consumed
in major urban areas (e.g., 98 percent in Dakar) is in the form of charcoal (Floor and

Gorse 1988, p. 72).

This is significant due to the nature of the carbonization process. Although
charcoal is more efficient than firewood in terms of calorific energy per unit weight,
approximately 55 percent of the heat energy in firewood is lost in the conversion
process to charcoal (Earl 1975, pp. 26-28). The net effect of this energy loss is that in
order to obtain an equal amount of useful energy, an urban inhabitant utilizing charcoal
consumes twice as much wood as a villager who obtains the energy in the form of
firewood (Goodman 1987, p. 114). Thus, it is apparent that urban domestic energy
consumption (in the form of charcoal) has a more deleterious effect on the natural

resource base than firewood consumed in rural areas.

 

1 This situation is beginning to change, however, in countries suffering from almost
complete deforestation. It appears that rural charcoal markets begin to appear in some
areas when local forest resources are depleted (see Whitney 1987).

21

The differential in fuelwood consumption between rural and urban areas is even
more important in light of the rapid population growth rate of Sahelian cities. Table 2.3
indicates that urban population growth rates in the Sahel are approximately twice those

of overall national growth rates.

Table 2.3" - Population Growth in the Sahel

 

 

 

Urban Population National Populap'pn

Total Pop % Avg An Growth % Avg An Growth %

Country 1965 1985 1965-80 80-85 1965-80 80-86
Burkina Faso 6 8 3.4 5.3 2.0 2.5
Chad 9 27 9.2 3.9 2.0 2.3
Mali 13 20 4.9 4.5 2.1 2.3
Mauritania 7 31 12.4 3.4 2.3 2.6
Niger 7 15 6.9 7.0 2.7 3.0
Senegal 27 36 4.1 4. 2.5 2.9

Source: World Bank 1988, pp. 274 & 284.
*Data for the Gambia not available.

 

The p0pulation growth rate of urban areas is an important indicator of demand
for fuelwood in cities and towns. It is argued that this will remain the case as long as
the substitution of more conventional fuels, i.e., fossil fuels, remains an unlikely option
for urban dwellers due to prohibitive costs, irregular supply, and poor distribution

networks.

Few studies have examined the specific problems of fuelwood supply and
consumption in and around urban centers of the Sahel. However, some attempts have

been made to document the situation pertaining to a small number of cities and towns.

22

One of the earliest studies undertaken regarding fuelwood production for a '
Sahelian urban center was reported on Niamey, the capital of the Republic of Niger
(Delwaulle and Roederer 1973). Although this information is now dated, what was
reported at the time were the first signals of a deteriorating fuelwood situation. The
authors observed a 15-20 km radius of deforested land surrounding Niamey. No natural
regeneration was taking place, as most of the land was kept in cultivation to supply food
for a growing urban population. Because of the growing need for agricultural land,
drastically reduced fallow periods were contributing further to the process of
environmental degradation. Thus, the authors predicted a radius of complete
‘desertification’ of 40-50 km around Niamey. In addition, due to the poor rate of
natural forest regeneration, it was noted that fuelwood was already being collected from

remaining savannah-forested zones more than 60 km from the city.

Another note of interest from this study relates to the transportation
infrastructure available to bring fuelwood to the Niamey market. Delwaulle and
Roederer discuss the geographic distribution of collection areas as being a function of
the quality of the transportation infrastructure. To the north of Niamey, the road was
unpaved; hence, the distance transporters were willing to travel ‘off-road’ was less than
that which they would travel to the east or west on paved surfaces. Situated on the
north bank of the Niger River, Niamey had been poorly served by deliveries coming
from the south until the construction of a bridge facilitated the transportation of
fuelwood. The authors envisioned greater traffic via the southern route due to this
improvement. It should also be noted that fuelwood was being transported to the
Niamey market by boat along the Niger River. Unfortunately, the authors did not

provide information regarding the significance of this mode of transportation.

23

In another study related to the problem of urban fuelwood supply in the Sahel,
Chauvin (1981) examined the potential of bringing fuelwood from surplus areas in the
Ivory Coast to the critical fuelwood deficit city of Ouagadougou, Burkina Faso
(formerly Upper Volta).

At the time of his study, Chauvin determined the average distance from
Ouagadougou to existing wooded savannah areas to be 70-100 km along the main
roads. With natural regeneration rates of forested lands, i.e., the mean annual increment
of growth, lagging behind urban fuelwood demand (due to a reported 8 percent per
annum urban population growth rate), the author suggested a three-pronged approach to
relieve the pressure on natural forests to supply Ouagadougou. This entailed 1) a
program to conserve fuelwood, 2) the creation of plantation forests, and 3) the import of

charcoal from distant resources, i.e., the abundant forests of the Ivory Coast.

Overall, Chauvin concluded that the cost of producing charcoal in the Ivory
Coast and transporting it to Ouagadougou would, considering the local retail price and
marketing costs, be wonomically viable. He examined alternative transportation modes
(road and road/rail combination), and determined that the margin between the delivered
cost and retail price would be greatest for charcoal produced in the central region of the
Ivory Coast and shipped via a road/rail combination to Ouagadougou. To date, the
degree to which fuelwood may be entering the Ouagadougou market via Chauvin’s

proposed channel has not been documented.

In a study of a region with similar ecological and socio—economic
characteristics, Digemes (1977, 1979) examined the contribution of fuelwood use to the

overall degradative process of desertification around the town of Bara, Sudan.2 The

 

2 Although this case study comes from outside the present study area, the conditions
surrounding Bara are quite similar to those found in many West African Sahelian

24

study noted changes in wood/charcoal consumption which indicated a trend from
individual household collection of a ‘free’ firewood resource, to the commercialization
of the domestic energy supply. This was due primarily to the expanding distance
between the site of collection and the town. Fuel that was formerly collected and
returned to the town by the end-users themselves was now brought to the market for
urban consumers by rural villagers. The villagers converted firewood into charcoal

since the latter is less bulky and easier to transport over longer distances.

For the purposes of the present study, it is interesting to note the spatial aspects
of the fuel supply system for towns such as Bara (see Figure 2.4). Although schematic
in nature, Digemes’ graphic model illustrates the process of fuelwood ‘mining’ and the

resultant deforestation around urban centers.

Stage A in the model relates to the collection of fuelwood by endusers
themselves, requiring no monetary or other exchange in order to obtain the necessary
supply. This is the ‘free good’ stage, not accounting for the opportunity cost of time and
labor. In stage B, the expanding zone of resource depletion becomes evident. Stage C
is perhaps the most representative of existing conditions in the Sahel, where fuelwood
resources are being collected at great distances from cities and towns. One note of
caution is necessary regarding stage D in this model: the amount of alternative energy
originating from external (5) or internal (6) sources is usually highly dependent on
access to these alternatives. In the case of conventional energy sources, i.e., imported
petroleum products, such access is restricted in many developing countries due to lack
of foreign exchange, efficient distribution networks, appropriate cooking devices and/or

sufficient income to purchase alternative forms of energy. Most likely, greater amounts

 

countries. Therefore, the conclusions drawn by Di gemes are relevant to the present
discussion.

25

of energy are being supplied from the far-removed forested areas as indicated in the
diagram.

According to the major findings from the review of studies related to Sahelian
urban centers, it is evident that urban fuelwood consumption is an important factor in
land degradation in peri-urban environments and in rural hinterland areas. The
progression of fuelwood collection sites into increasingly distant rural areas has the
combined effect of raising the price of fuelwood paid by urban consumers, and
accelerating the rate at which deforestation and environmental degradation take place.
This is particularly worrisome in light of the effect that the change from firewood to
charcoal use has on the rate of forest clearance, as distance increases between collection
and consumption points. Thus, as related to the present study, the findings from the
work examined so far indicate that exploring the establishment of peri-urban fuelwood
plantations could possibly lead to a more environmentally sustainable and less costly

urban energy supply system.

26

 

 

 

Legend: I. lnhabited area. 2. Forest: and fields. 3. 1

Deaenifled area. 4a. Charcoal. 4b. Wood. 3. Alternative
4a 0.... 5 m 6
413

external energy. 6. Alternative internal energy.
Figure 2.4 Fuelwood Supply Model for Dryland Population Centers
Source: Digemes 1979

 

27
Strategies to Address Fuelwood Deficits

In light of the growing difficulties of supplying affordable fuelwood energy to
urban areas, and doing so without desu'oying the ecological balance in rural areas,
research attention has focused on a variety of strategies to ameliorate current deficit
situations. Generally recommended policy options to redress fuelwood deficits fall into
three main categories: 1) the use of substitute fuels (sometimes referred to as
‘fuelswitching’), 2) conservation in the use of existing fuel resources, and 3) enhancing
the supply of fuelwood energy (Anderson and Fishwick 1984, p. 3). Each strategy is
addressed in turn.

Replacing fuelwood energy with conventional fuels (kerosene, bottled gas
(LPG), coal, electricity, etc.) has been the focus of several recent studies in the
developing world (Elkan 1988; Hosier et al 1982; Foley 1986; Whitney et a1. 1987).
However, in many instances the implementation of the substitution option has faced

numerous obstacles.

The quantity of fossil fuel energy required to meet the domestic needs of the
entire population of the developing world currently using fuelwood is relatively small,
equivalent to approximately 2 percent of current global oil consumption (Goodman
1987, p. 116). However, foreign exchange expenditures for petroleum imports in many
developing countries already exceed 50 percent of export eamin gs. In a recent study in
the Sudan, Whitney and colleagues (1987, p. 320) state that given the high cost of
importing fossil fuels, forest ‘mining’ appears, in the short term, to be an economically
rational policy decision. Reducing fuelwood consumption to prevent deforestation
through the use of fossil fuels would require an increase of over 100 percent in Sudan’s

annual petroleum imports. The added foreign exchange expenditure that fossil fuel

28

substitution would require is a fiscal burden that Sahelian countries, among the poorest

in the world, can hardly afford.

In addition to concerns about foreign exchange availability, there are several
other issues that make switching to more conventional fuels problematic. One of these
is the problem of assuring delivery of imported fuel resources (Leach and Meams 1988,
p. 244). Given that most Sahelian countries are landlocked, the delivery of imported
fuel resources to interior urban centers is particularly difficult. Long distances from
port locations, along with the poor state of transportation infrastructure, pose serious

obstacles to making fossil fuel products accessible to these resource poor countries.

Assuming that a viable supply system existed to deliver alternative fuels to
Sahelian consumers, another problem to be overcome is that of the additional costs
associated with the use of fossil fuels. For example, the use of such fuels would require
modern cooking and heating devices. Since incomes in Sahelian countries are generally
low, most households cannot afford the large lump-sum investment in an expensive
kerosene or gas cooker. In addition, due to the day-to-day functioning of most family
budgets, higher occasional payments for utility bills or replacement gas bottles are more
difficult to make than smaller, more frequent payments made to cover daily fuelwood

needs.3

Fuelswitchin g strategies involving the use of renewable sources of energy have
also been exarrrined for potential application in the Sahelian region. A significant
amount of research into solar, aeolian, and bio-gas digestion techniques has been
canied out (Harrison 1987b). Although feasible in some cases (particularly small-scale

applications), these renewable energy strategies have proven difficult to apply on a

 

3 These budget problems, along with cultural preferences for fuewood or charcoal to
121';ng local dishes, have been examined in Dakar, Senegal, by Tibesar and White
(1 5).

29

large-scale basis. In some instances (windmills for example), the initial costs of
applying this technology have been prohibitive. Other difficulties have been
encountered in terms of adapting renewable energy systems for use under local
circumstances. Experiments with solar cooking devices in the Sahel have demonstrated
that they are not particularly well-suited for meeting the requirements of preparing local

foods (nor can they be used in sheltered areas or at night).

The success of fuelswitching interventions will be determined by the level of
economic development in a given country as measured by national foreign currency
reserves and consumer purchasing power. Unfortunately, the slow pace of economic
development in the Sahelian region does not augur well for reducing fuelwood

dependence through the substitution option.

The second strategy for addressing fuelwood deficiencies is that of energy
conservation. In the Sahel, conservation of existing fuelwood resources is a somewhat
foreign concept. Although most Sahelian fuel use is fairly judicious, no extraordinary
attempts are currently being carried out to conserve what is considered by many
outsiders to be a dwindling resource base. For the most part, this situation is due to
indigenous perceptions of fuelwood scarcity. In contrast to estimates made by outside
observers, the indigenous population does not necessarily view this resource scarcity to

be as severe or critical (author’s personal observations).

Nevertheless, programs have been initiated in Sahelian countries with the
objective of conserving, or making more efficient use of fuelwood resources. These
programs may be categorized under two basic types: the first deals with the
dissemination of improved cooking devices (i.e., more efficient wood-burning stoves),
and the second is concerned with the design of more efficient charcoal kilns (Gill 1987;

Harrison 1987a; Krugmann 1988; and Meier and Munasinghe 1987).

30

The dissemination of improved cookstoves has been carried out through two
types of programs. In the first case, stoves have been designed and constructed for the
use of firewood in rural villages. These are made from local materials (usually a
mixture of sand and clay) and built by the villagers themselves. The stoves are
designed so that more heat energy is applied to cooking the meal, thus reducing the
amount that is dissipated into the surrounding air. The construction of these cookstoves
requires little or no cost to the builder/user, except for the opportunity cost of labor and
time required to make the stove. The benefit of building an improved cookstove is

expected to result in less time spent collecting firewood, and thus less wood consumed.

The second design is a metal cookstove intended for use by urban (mostly
charcoal) consumers. The urban version of the improved cookstove is produced by
local blacksmiths and sold in neighborhood markets. Through the use of the metal
cookstove, charcoal consumers in urban areas are expected to realize a monetary
savings per unit time in the amount of fuel required to meet their needs. The production
of these stoves also provides a small degree of employment opportunity to local

artisans, most of whom work in the so-called informal sector of the economy.

The specific design of improved charcoal kilns, unlike cookstoves, is as varied
as the locations in which they are used (see Harrison 1987b). Improved kilns are
expected to reduce the energy loss inherent in the carbonization process, thereby
contributing to a reduction in deforestation rates. The use of these devices is marginal
in most Sahelian countries, especially where materials for their construction must be
imported. Currently, more research is required into the development of effective

improved ldlns made from less costly materials such as recycled metal products.

31

Evaluating the performance of the conservation intervention has proven to be
quite difficult. No one knows for certain how many stoves have been built or sold, nor
how much firewood or charcoal (or money) may have been saved by their use.
Generally, the urban metal cookstoves have experienced greater popularity and success
because they are relatively fuel efficient and, more importantly, show a tangible benefit
to their user through reduced cash expenditure on fuel. On the other hand, the improved
stove intended for use in villages has experienced several problems. Among the
difficulties are 1) poor acceptance/usage rates, 2) poor fuel efficiency performance, 3)
short life spans, and 4) less widely perceived benefit in terms of reducing amount of

labor and time required to collect fuelwood.

Due to the difficulties in assessing the contribution that conservation measures
have made to reducing fuelwood consumption, tentative conclusions indicate only that
these measures are at best rather marginal. However, conservation measures should
continue to be promoted as a corollary activity to other fuelwood-deficit reduction

efforts.

The third strategy for reducing fuelwood deficits, and the one most germane to
the present study, is the enhancement of the fuelwood supply. This approach addresses
the problem of meeting urban fuelwood requirements through the establishment of
plantations in the periphery of urban centers. There exists a wide range of studies on
this topic (Anderson and Fishwick 1984; Floor and Gorse 1988; Anderson 1986; El-
Hinnawi 1982; Tanticharoen and Gururaja 1982; Leach and Mearns 1988; Munslow et
al., 1988; and Goor and Barney 1976).

There has been much debate concerning the advantages and disadvantages of
promoting plantations for the express purpose of growing fuelwood. Among the

benefits, the idea of reclairrring deforested lands in the urban periphery in order to meet

32

an important need for urban energy supply would seem eminently sensible. Anderson
(1986, p. 857) argues that due to rising scarcity prices of fuelwood, peri-urban
plantations show potential for good financial and economic returns. Additional benefits
that he cites include: 1) lower transport costs due to shorter distances between points of
production and consumption, 2) regeneration and stabilization of biomass productivity

in degraded areas, and 3) the potential for employment and income generation.

The disadvantages of fuelwood plantation development arise primarily from
competition for other uses of land close to urban areas, large initial establishment costs,
and competition from other sources of ‘free’ fuelwood, i.e., fuelwood for which the
market price reflects little more than collection and transport costs (Anderson and
Fishwick 1984; Elkan 1988). Other studies have noted problems related to improper
species selection and climatic constraints such as rainfall variability in dryland

environments (Catterson et a1. 1984; Freeman and Resch 1985/86).

Despite these disadvantages, Anderson and Fishwick (1984) state that

plantations for fuelwood, if appropriately located, have the potential to be economically

 

viable. The task of the present study is to examine a method for determining where,
and under what conditions, peri-urban fuelwood plantations could compete effectively
with other agricultural land uses in order to satisfy the demand for energy in Sahelian

urban centers.
Need for Present Study

It is evident from the literature that dependence on fuelwood energy in regions
such as the West African Sahel is unlikely to diminish in the near future. However,

research has also indicated that, under current conditions of resource exploitation,

33

fuelwood dependence will exacerbate the problems of environmental degradation in
rural areas while contributing to economic impoverishment in both rural and urban

population sectors.

The review of strategies to redress the fuelwood deficit situation has indicated
that: l) the ‘fuelswitching’ option is unlikely to prove successful due to financial
difficulties, and 2) conservation measures can be expected to have only a marginal
effect on reducing fuelwood consumption. In contrast, the establishment of peri-urban
plantations for the purpose of growing trees to meet urban energy needs may prove to
be an economically viable solution. Although much work has been undertaken to detail
the advantages and disadvantages of establishing peri-urban fuelwood plantations, little
attention has been paid to the important question of where, in terms of the space

economy, such plantations should be located.

There are two general issues involved in determining the best site for locating
peri-urban fuelwood plantations. The first deals with the problem of site specificity
related to the physical requirements of a selected tree species (soil type, rainfall, etc.),
and the second relates to the competitiveness of a specific site for tree production with
other potential land uses. This study will demonstrate one way in which the second

issue may be addressed.

The literature has also indicated that the increasing distances between fuelwood
collection and consumption sites is a major factor contributing to higher fuelwood
prices and accelerated rates of deforestation and environmental degradation. It is
suggested that the peri-urban plantation strategy can help to resolve this problem by
investing in the rehabilitation of degraded lands on the periphery of urban areas. Such

an intervention would follow in the spirit of Blaikie and Brookfield’s idea of ‘landesque

34

capital’, and represent an attempt at solving natural resource management issues within

the framework of the regional political ecology concept.

The investment in a renewable resource, energy-producing forestry system
would be consistent with the maintenance of land capability for future production. At
the same time, this type of energy production system would present an opportunity to

meet urban demand without sacrificing rural (and national) productive resources.

A spatial reorganization of the present system of fuelwood exploitation in the
West African Sahel will have inevitable repercussions on the region’s economic and
ecological systems. It will also generate a set of social opportunities and constraints
based on the dynamics of the land use decisions that will have to be implemented. This
last issue will present new challenges to those responsible for determining natural
resource management policy. This study will address these issues as well, attempting to
describe some of the changes that can be expected in association with a shift in the

location of fuelwood production.

CHAPTER III
METHODS AND DATA COLLECTION

While set within the theoretical framework of regional political ecology, this
study approaches the more specific locational question of where peri-urban fuelwood
plantations could be established. This is accomplished, through the application of
Johann H. von T'hunen’s model of agricultural land use as a method of analysis. Thunen
is credited by many geographers as being the first to put forward a comprehensive
theoretical model of agricultural land use. His original work, The Isolated State, was
published in 1826. Thunen’s work laid the foundation for numerous subsequent studies
by economic geographers.1 The first section of this chapter is devoted to a discussion
of the Thunen model of agricultural land use and the rationale behind its application to

this study as a method of analysis.

The second section of this chapter discusses the selection of data used in the
analysis. Limitations inherent in the data due to secondary collection methods and
applied assumptions are also mentioned. These assumptions have a critical bearing on
the outcome of the data analysis, and therefore must be well understood in advance.
Finally, the last section of this chapter explains the calculations that will be performed

in the data analysis.

 

1 For a description of the Thunen model see: de Souza 1990; Conkling and Yeates
1976; Chisholm 1979; Dunn 1954; and Barlowe 1978. For a translation of Thunen’s
original work, The Isolated State, published in 1826, see Hall 1966. Examples of works
applying Thunen’s model include: Horvath 1969; Griffin 1973; Moran 1979; Feet
1969; Sinclair 1967; and Stevens and Lee 1979.

35

36
The Thunen Model asfia Method of Analysis

By maintaining meticulously detailed records of his farnring operations in the
northern German state of Mecklenburg, Thunen formulated a model of the patterns of
agricultural activity around a single market center. His idealized agricultural region is
represented as a flat, featureless plain upon which a single mode of transport (horse-
drawn cart) is used to bring crops to a central market from any and all directions. Land
is deemed to be unifomrly fertile, and all farmers are assumed to act in such as manner

as to maximize their profits.

Departing from the Ricardian concept of ‘econorrric rent’, i.e., the difference in
the net profits attributable to parcels of land inherently different in terms of fertility,
Thunen developed an alternative view of this concept known as ‘location rent’. Thunen
considered location rent to represent the net revenue produced at a given site compared
with the return generated by cultivation at the margin of production, i.e., where land is
capable of yielding a return just sufficient to cover production expenses. The equation

for determining location rent is given as follows:

R = E(P-C) - (Efk)

where R = location rent per unit of land
E = crop yield per unit of land
p = market price per unit of output
c = production cost per unit of output
f = transport rate per unit of distance
per unit of output
k = distance to market.

This concept of location rent demonstrates how ‘rent’ is a function of distance
from a market center. Essentially, rents decline with increasing distance from the
market center until there is ‘negative rent’, i.e., beyond the spatial margin of

profitability. In cases where more than one agricultural activity competes for the same

37

parcel of land, the one that commands the highest rent will be deemed the best suited to

occupy that location (see Figure 3.1).

SOP

 

Net Income l Dollars)

   
 

 

 

 

 

 

Figure 3.1 Location Rent Gradients for Competing Crops. Source: de Souza 1990

In Thunen’s isolated state, the idealized pattern of land use is represented by a
series of concentric circles around the market center (see Figure 3.2). The innermost
zones include products that are perishable (e.g., fresh vegetables) or heavy and costly to
transport in relation to their value (e.g., fuelwood and building lumber). Immediately
beyond these inner zones, field crops are produced which employ less intensive
methods of farming as distance from the market increases. Finally, forms of
agricultural production that require extensive land, use such as livestock grazing, are

located in the zone most distant from the market center.

This idealized pattern of agricultural land use is of course modified by real

world situations. Thunen recognized that the concentric circles zonation would be

38

 

Field Crass with Emphasis
on Dairy Products

  
 
 

Three Course System

   

Extensive Cattle Raising

 

Kilometers
trom Market

Central City

 

 

 

Figure 3.2 Land Use Zones in Thunen’s Isolated State
Source: de Souza 1990

altered by many different factors. Among these he discussed the effects of competing
secondary market centers, as well as improved transportation systems, e.g., navigable

rivers, which reduce shipment costs compared to overland carts.

Given present-day industrial economics, the land use zones of Thunen’s Isolated
State are not necessarily reflected in modern agricultural land use patterns. In
particular, T'hunen’s location of a forest products zone might seem somewhat anomalous
as modern societies have acquired new sources of energy. However, for the towns of
early nineteenth-century Germany, nearby forests were an important source of building
materials and fuel. Today, in industrial economies with improved transportation
systems, higher transportation costs to move forest products over longer distances have

been compensated for through economies of scale and regional comparative advantages

39

in modern production processes. Hence, a forested zone surrounding a major urban.
settlement in today’s industrial nation would more likely be attributed to aesthetic
reforestation than to a source area for fuel and building supplies. It is argued, however,
that the current status of economic development in most developing countries may
explain certain resemblances to nineteenth century Europe in terms of the patterns of

agricultural activity surrounding urban centers (Sinclair 1967, p. 73).

In a study by Horvath (1969), the author describes agricultural land use patterns
found around the capital city of Addis Ababa, Ethiopia. The results of his study suggest
an interesting parallel between the pattern of agricultural zonation in the Addis Ababa
area and that of the idealized Thunen model. Among his findings, Horvath noted a
forested zone of Eucalyptus surrounding the city which served to meet the needs for
fuelwood and building materials. In addition, the study described the elongated shape
of this forested land as it extends along the areas surrounding an improved
transportation artery. Although Horvath had to adapt some of the basic assumptions of
the Thunen model to fit the situation found in the Addis Ababa study area (e.g., lack of
the flat featureless plain around the city), his work is a good example of how this model

can be applied in the context of an urban area in the developing world.

Despite the many studies undertaken which have built upon Thunen’s work,
there is still a certain degree of debate as to whether his model is basically descriptive
or normative in nature (Harvey 1966, p. 363). Chisholm (1979, p. 14) has discussed
how Thunen’s model can be seen as an analytical methodof general applicability, one
that is essentially a normative statement on the form of the agricultural landscape,
according to certain assumptions. Sinclair (1967) would concur, arguing especially for
the validity of applying the Thunen model to agricultural zones around urban centers in

less developed countries. It is in such cases, Sinclair states, that T'hunen’s assumptions

40

and driving forces, i.e., transport costs as a function of distance, continue to operate as

primary factors in shaping the patterns of agricultural land use.

From the preceding discussion it is argued that Thunen’s model can be adapted
to help determine viable locations for peri-urban plantations in present-day Sahelian
countries. Where the cost of fuelwood in urban markets is directly influenced by
distance-related transport costs, Thunen’s model should be able to provide some useful

insight into answering the question of where fuelwood plantations could be located.
D‘ata Collection

The determination of the best location for a peri-urban fuelwood plantation,
according to the Thunen location rent equation, requires data for agricultural land uses
surrounding a selected market center. In this study, data are examined for a set of
agricultural creps from an agro-ecological zone typical of the West African Sahel. In
addition, hypothetical data based on estimates for plantation forestry under similar

conditions are presented.

Data on agricultural products are given for the Fifth Administrative Region of
Mali, which has as its main market and administrative center the town of Mopti, located
on the Niger River (see Figure 3.3). The selection of this agricultural market center for
the current study is based on the availability of recently collected (secondary) farm-
level data, and the fact that ecological conditions and fuelwood dependence in the

Mopti region are representative of the Sahelian agro-ecological zone.

The Mopti region covers a total land area of approximately 77,800 km2, and is
situated between 14 and 16 degrees north latitude, and 4 and 6 degrees west longitude

(Henry de Frahan et a1. 1989). The climate is representative of the Sahelian and the

41

Khm..\\\

  

$2 .3 8 55am on Eco: 623m
me: no 9:2 o>uab££Ee< m.m Semi

tee. a - 8.0

i . mocSU

it I
‘\ . .

. .
BnQKiOSVKSQ Its-talerrtiz It, 922me \\
I a as

on \Yeoi ~33. \
Q
a ml film: EN

~ Omnotm
. r , . m.
N

 

u.Eu~.tau\\

 

158828

   
      

. oi.
. may.» law
sx
\
\ :3qu cu :uZiCuzu I
_
" 33.96 a
_ w L336 33. 32.0: gauge“. tit/i
_ e.g.,—so... a.
. 5:: 2.030.. new 31E: it..-
I
It 3.33:9...— I
I Irllllll
/ ucsugucoeoz

42

Sudano-Sahelian ecological zones, with annual rainfall ranging fi'om 200 to 600 mm.
Clay soils predominate in the inland delta area of the Niger, with mostly sandy soils
found throughout the remainder of the region. The economy of the region is based

essentially on the primary activities of crop production, livestock raising, and fishing.

Crop data used in this study are from the Malian Ministry of Agriculture
(Republique du Mali 1989), and two other sources (Henry de Frahan et a1. 1989;
Stryker et al. 1987). In order to calculate rent gradients according to the Thunen
equation, figures are given for the following variables: 1) crop yield in tons/hectare, 2)
market price per ton of crop, 3) total production costs per ton of crop, and 4) an
estimated freight rate for transporting crops to market. The craps for which data are
available in the aforementioned sources include rice, millet, maize, groundnuts, and
cotton. Production of most crops involves techniques such as the use of improved seed
varieties, some use of chemical fertilizers, and the rotation of crops from one growing
season to the next. In some cases, particularly flooded rice cultivation, animal traction

is used.

Data for yields and production costs of plantation-produced fuelwood are highly
variable depending on the source of data. In most instances, data are generalized and
reported in terms of cost per hectare of production, or unit weight or volume of product,
without reference to the yield realized or species grown - factors which vary widely
from one location to another. Wardle and Palrrrieri (1981), and Foley (1986), have
indicated that there is a dearth of reliable data regarding the true value of fuelwood, its
hidden costs (related to the free good notion), and the numerous factors that influence

eventual market prices.

Many factors must be taken into account in order to arrive at a reliable

estimation of the expected costs of production and potential yields. These include

43

physical conditions such as soil properties, irrigation versus rain-fed production, and
fertilizer use. Other important factors relate to species selection, stocking rate, the
degree of mechanization used versus manual labor employed, and whether or not
multiple products (such as building materials, fruits, gums, resins, etc.) are expected to

accrue from the production system.

In light of the lack of reliable data for plantation costs of fuelwood in the
Sahelian region, hypothetical data are developed for this study based on a range of
figures generally cited in the literature. Extreme caution must be used in any attempt to
interpret the results generated by the analysis of such data. The data for the
hypothetical fuelwood plantation do not necessarily represent site-specific conditions of
the agro-ecological zone of the M0pti region; they are at best general approximations.
In addition, the range of production cost figures is rather wide (from US$ 500 to US$
1,300 per hectare). Although the reliability of the data is questionable, the analysis of
such a wide range of values may generate some important information regarding the

conditions under which fuelwood plantations would be feasible.

For the purposes of this study, the data for the cost of plantation-produced
fuelwood reflect a range of figures cited by Floor and Gorse (1988), and Anderson and
Fishwick (1984). This range varies from US$ 500 to US$ 1,300 per hectare for
plantation-produced Eucalyptus species that are widely promoted in dryland
environments such as the Sahel. Over this range, three production cost scenarios are
devised based on arbitrarily selected values of US$ 500, US$ 1,000, and US$ 1,300 per
hectare, representing ‘low’, ‘medium’ and ‘high’ cost situations respectively. The dollar
amounts are converted into local currency (the Franc CFA) in order to allow for

comparison with the data on agricultural crops.

44

Furthermore, it is assumed in this study that fuelwood production takes place in
a taungya2 agroforestry system. Earl (1975, pp. 52 & 71) states that the
implementation of the taungya method is an effective, indeed almost essential, means
by which the high establishment costs of fuelwood plantations can be compensated for.
The production of agricultural crops interspersed between trees during the early stages
of plantation development results in a substantially reduced final cost per unit of
fuelwood. For example, Weinstabel (1984) describes how a similar intervention in
Burkina Faso turned a net loss fuelwood-only plantation scheme into a net gain
agroforesty system. Earl has estimated that by using the taungya method, plantation
fuelwood costs per ton of wood can be as low as 50 percent of the costs of producing
fuelwood alone. Hence, the production costs given for the three plantation scenarios in
this study are recalculated to adjust for fuelwood produced in an agroforestry system
affording maximum financial benefit from associated crop production according to
Earl. It must be pointed out that this assumption may not necessarily hold for present
conditions in the Sahelian region, despite Weinstabel’s supporting evidence. Much
would depend on the specific crop components, prevailing local market prices, and site-

specific physical conditions of the particular agroforestry system.

Yield estimates for Eucalyptus plantations in the Sahelian ecological zone used
in this study reflect the median to lower values of the expected range for this
environment, according to data from the National Academy of Sciences (1980), and
Anderson and Fishwick (1984). The use of moderate yield estimates in the analysis
should more closely represent actual conditions. The best estimate available for a

fuelwood freight rate is given by Shaikh (1985), and assumes the use of large-capacity

 

2 Taungya, a forrrr of intercropping, refers to a system of establishing forest plantations
whereby agricultural crops, e.g., groundnuts, cowpeas, etc., are grown between the
newly planted trees until the canOpy is closed.

45

vehicles on paved surfaces. Again, caution is required in the use of this information.
Transportation in the Sahel is very difficult and variable according to seasonal

conditions. Shaikh’s estimate may very well underestimate actual transport costs.

The determination of an estimated market price for plantation-produced
fuelwood presents another type of problem. Under the current fuelwood exploitation
and marketing structure in most Sahelian countries, market prices for fuelwood are,
relatively speaking, so low that peri-urban plantations would be infeasible. However, if
it is assumed that wood energy is perfectly substitutable for kerosene, then a shadow
price for fuelwood can be derived by equalizing the costs of these energy sources in
terms of delivered energy content (see Christophersen and Karch 1988, pp. 31-32).
Shadow pricing is frequently used as a means of determining the intrinsic value of
wood energy. The derived increase in the fuelwood price has the effect of rendering

plantation production systems economically feasible.

Given the assumptions made in the preceding paragraph, a shadow market price
for fuelwood is used in the present analysis. The calculation of this price, detailed in
the next chapter, is based on an assumed retail price for the kerosene energy substitute,
calorific energy conversions between kerosene and wood, and estimated thermal

efficiencies of both kerosene and wood cooking devices.

Calculations Used in the Analysis of Datp

There are two calculations performed in the next chapter in order to determine
locational characteristics of the hypothetical fuelwood plantation. The first is based on
Thunen’s location rent equation (p. 42). Subject to this equation, rent gradients for the

various agricultural and tree products are derived in order to determine the resultant

46

patterns of land uses generated by the model. This procedure will help to ascertain the
best location for fuelwood production relative to the location of the market and other

agricultural land uses in the urban periphery.

A second calculation is performed in order to determine the location of the
‘firewood/charcoal calorific threshold’. Poulsen (1978, p. 15) has noted that in a wood—
based urban energy production system a decisive point will be reached, in terms of
transportation costs, where an equal amount of energy (in calorific terms) can be
delivered to the market less expensively in the form of charcoal as opposed to firewood.
This is due mainly to the fact that per unit weight, charcoal provides approximately
twice as much useful energy (Earl 1975). This point can be considered as a
‘firewood/charcoal calorific threshold’. The location of this threshold is important in

that it determines the form in which fuelwood should ultimately be marketed based on

_ . transportation costs and distance to market. Furthermore, knowledge of where this

threshold is located can provide useful information for resource management strategies.
This is particularly important due to the larger area of forested land cleared to produce

charcoal versus firewood.

To summarize, the procedures followed in this study entail the analysis of data
on agricultural crops produced in the Fifth Region of Mali, along with estimated data
for a hypothetical peri-urban fuelwood plantation. Rent gradients are calculated for all
crops and fuelwood using the equation for location rent derived from the Thunen model
of agricultural land use. The analysis of the resultant crop patterns will help to answer
the question of where to locate fuelwood production in an economically viable fashion.
The determination of this location will provide some insight into answering other
questions, referred to in the previous chapters, related to the economic, ecological, and

social ramifications of an altered spatial pattern of fuelwood production.

CHAPTER IV
ANALYSIS OF DATA

This chapter first presents data for agricultural crops produced in the Fifth
Region of Mali, as well as data for fuelwood production in the hypothetical peri-urban
plantation. An explanation is given detailing how production costs and market price for
the fuelwood ‘crop’ were estimated. Second, rent gradients are calculated and presented
for each crop, including plantation-produced fuelwood. The rent gradients, calculated
according to the Thunen model, are used to display the theoretical zonation pattern of
these crops. Third, based on the calculated rent gradients, the best location for
plantation fuelwood production is then determined. Fourth, the location of the
firewood/charcoal calorific threshold is determined through an analysis of fuelwood
production and transportation costs. Locating this threshold provides further insight
into the ecological ramifications of the peri-urban fuelwood plantation system based on
the form in which fuelwood should be produced. Finally, a preliminary interpretation

of the results of the data analysis is offered.
Agricultural Crop Da_ta_

In Table 4.1, figures for yield, market price, production cost, and freight rate are
listed for five crops cultivated in the Mopti region. The data, collected during the 1986-
87 agricultural campaign, are based on estimates for production using improved manual
techniques and the sale of produce in the Mopti market. The freight rate estimate is
given for transportation costs via paved surfaces, and therefore should be considered an

underestimate since many farms are not located near modern transport arteries.l

 

1 The equal freight rates for all crops is not explained by the data source. In reality,
these rates can be expected to vary according to factors such as the bulkiness of
product, season, road conditions, and the demand for transportation.

47

48
Table 4.1 - Agricultural Crop Data

 

 

Yield Mkt.Pr-ice1 Prod. Costl Frt. Rate
Crop (ton/hat (thousand CFAZton) (CFAZtoM)
Cotton 1.2 330 318 25
Groundnuts 1.0 1 15 104 25
Maize 1.6 66 65 25
Millet 0.8 64 60 25
Rice 0.7 143 126 25

1 Market prices and production costs are given in local currency, Franc CFA
(Communaute Financiere Africaine), $1 U.S. = 270 F CFA approx. (1986).

Sources: Republique‘du Mali 1989, Henry de Frahan et al. 1989,
and Stryker et al. 1987.

 

Fuelwood Datp

As mentioned in chapter three, the determination of production costs for
fuelwood grown in a plantation setting depends on many factors, e.g., species planted,
soils, rainfall, and local labor costs. Plantation establishment in a semi-arid location
such as the Sahel is very tenuous, particularly in light of environmental uncertainties
such as highly variable rainfall. Thus, production costs may vary widely depending on
local conditions within the Sahel. The figures used here for fuelwood production costs
represent a range of possible scenarios for a plantation system in the Sahelian

environment.

Floor and Gorse (1988) have suggested that plantation costs in the semi-arid
environment of the Sahel can range between US$500-1,000 per hectare. Anderson and
Fishwick (1984) have stated that this cost may reach as high as US$1,300 per hectare.
These are the assumed total costs per hectare over the entire initial rotation period

(approximately 8-10 years) for a typical dryland tree species such as Eucalyptus

49

camuldalensis. Included in the total for the initial rotation are the costs of tree

seedlings, ground preparation, planting, maintenance, and interest on capital.

In Table 4.2, the total costs of fuelwood production (per ton of product) are
presented for several hypothetical Eucalyptus plantations based on three arbitrary levels
of assumed cost per hectare: US$ 1,300, 1,000, and 500 respectively. These figures
represent ‘high’, ‘medium’ and ‘low’ cost scenarios. The total cost per ton of fuelwood
is given for production both with and without using a Taungya system. The two yield
figures provided represent more modest harvest estimates expected from Eucalyptus
plantations in senri-arid regions, reflecting the lower end of a commonly-cited range of
between 2.0 and 11.0 tons/ha (National Academy of Sciences 1980; Anderson and
Fishwick 1984). The transportation cost (freight rate) given in the table is an estimate
from Shaikh (1985), and assumes the use of large-capacity vehicles on paved surfaces.
Again, estimates for yield and freight rate are used with caution as actual figures may

be quite different.

50

Table 4.2 - Estimated Fuelwood Plantation Costs

 

Scenario #1 - Hi h cost: assumin US 1 300 a (US$1 = 270 F CFA)

Expected yield ................................. 3.0 tons/ha
Total production costs without Taungya .............. . 117,000 F CFA/ton
Total production costs with T aungya .................. 58,500 P CFA/ton
Freight rate .................................... 4 0 CFA/ton/km
Expected yield ................................. 5.0 tons/ha
Total production costs without Taungya ................ 70,200 F CFA/ton
Total production costs with T aungya .................. 35,100 F CFA/ton
Freight rate .................................... 40 CPA/ton/km

Scenario #2 - Med. cost: assuming US$1,000[l_ra (U 881 = 270 F CFA)

Expected yield ................................. 3.0 tons/ha
Total production costs without Taungya ................ 90,000 F CPA/ton
Total production costs with Taungya .................. 45,000 F CPA/ton
Freight rate .................................... 4 0 CFA/ton/km
Expected yield ................................. 5.0 tons/ha
Total production costs without T aungya ................ 54,000 F CFA/ton
Total production costs with Taungya .................. 27,000 F CFA/ton
Freight rate .................................... 40 CFA/ton/km

 

Scenario #3 - Low cost: assuming US$500[l_ra (U 881 = 270 F CFA)

Expected yield ................................. 3.0 tons/ha
Total production costs without T aungya ................ 45,000 F CFA/ton
Total production costs with T aungya .................. 22,500 F CFA/ton
Freight rate .................................... 4 0 CPA/ton/km
Expected yield ................................. 5.0 tons/ha
Total production costs without, Taungya ................ 27,000 F CFA/ton
Total production costs with T aungya .................. 13,500 F CFA/ton
Freight rate .................................... 4 0 CPA/ton/km

 

Sources: According to estimates from Floor & Gorse 1988,
Anderson & Fishwick 1984, Earl 1975, and Shaikh 1985.

51

Determining the market price of fuelwood for this analysis poses a difficult '
problem. Since fuelwood sold in Sahelian urban markets does not reflect the
replacement cost of wood, i.e., the cost of growing a tree to replace one cut down, it is
not possible to empirically determine a market price for plantation-produced fuelwood.
Currently, fuelwood prices cited for Sahelian cities vary widely. For example, prices
range from approximately 4,600 F CFA/ton in Mopti (Christophersen & Karch 1988),
to 15,400 F CPA/ton in Dakar (Leach & Mearns 1988), to as much as 27,000 F
CPA/ton in Ouagadougou (Bertrand 1985/86). Based on these prices, and the costs of
producing fuelwood in plantations as shown in Table 4.2, it would appear that, in most

cases, growing wood for fuel in plantations would not be economically viable.

However, it is possible to determine a market price for fuelwood based on an
alternative method. Since energy used for domestic purposes is considered as a basic
human need, demand for fuelwood can be considered fairly price inelastic (Meier &
Munasinghe 1987, p. 127). Theoretically, plantation-produced fuelwood sales would be
viable as long as the price is lower than that of the next best energy substitute, e.g.,

kerosene.

The estimated market price for plantation-produced fuelwood may be deemed
more competitive if wood energy is assumed to be perfectly substitutable for imported
kerosene.2 Christophersen and Karch (1988) point out that a "shadow price" can be
derived for fuelwood, vis-a-vis kerosene, by equalizing the cost of the two energy
sources in terms of energy content delivered. It is argued that the shadow price of

fuelwood more accurately reflects its replacement cost and intrinsic value as a source of

 

2 In most cases, kerosene is considered to be the next fuel up the "energy ladder"
following fuelwood. Households with a regular, sufficient income are expected to
switch to using kerosene, assuming supply is regularly maintained.

52

energy in comparison to imported energy products that require foreign exchange

expenditures.

Table 4.3 indicates the shadow price of fuelwood based on two different
estimates for the price of imported kerosene. To perform the calculation, one must
determine the relative cost of producing an equal amount of useful energy from
kerosene vis-a-vis fuelwood. One litre of kerosene contains approximately 10,500
kilocalories of energy (Christophersen and Karch 1988). One ton of dry wood contains
approximately 4.7 million kilocalories (Earl 1975). It is assumed that kerosene used for
cooking has a thermal efficiency of approximately 50%, i.e., half of the energy
produced actually cooks the food, and half is dissipated. It is also assumed that the
thermal efficiency of a typical improved wood-buming stove is approximately 25%.

Hence, the kerosene stove is considered to be twice as efficient as the wood-burning

 

 

stove.
Table 4.3 - Shadow Price for Fuelwood Based on Kerosene Substitute
Assume:
W helm
Retail price (FCFA/litre) 120 n.a.
Calorific value (kcal) 10,500/litre 4.7 * 106/ton
Kerosene equivalent/ton fuelwood (litres) 448 n.a.
Kerosene equivalent price (FCFA) 53.760 n.a,
Shadow price (FCFA/ton) 26,880"
Shadow price assumin g kerosene *
retailing @ FCFA ISO/litre 33,600

* n.b.: Since the kerosene stove efficiency is twice that of the wood-buming stove,
it is necessary to divide the kerosene equivalent price by two to determine
the fuelwood shadow price.

Sources: Christophersen & Karch 1988, and Earl 1975.

 

53

The increased estimated market price for fuelwood reflected in its shadow price
(Table 4.3) should make peri-urban plantation production more economically attractive

with respect to the production cost figures in Table 4.2.
Calculation of Rent Gradients

Using the location rent equation outlined in chapter three, (R = E(p-c) - Efd),
rent gradients for the five agricultural cr0ps are calculated. Table 4.4 gives the
maximum R values (maximum net revenue when distance equals zero), and the
maximum distances from the market at which profitable production takes place

(maximum distance when R = 0, i.e., the margin of profitable production).

Table 4.4 - Maximum Rent and Distance for Cr0ps

 

Crop Max. Rent (CFAZha) Max. Dist. (m)
Cotton 14,400 480
Groundnuts 1 1,000 440
Maize 1,600 40
Millet 3,200 160

Rice 1 1,900 680

 

The results obtained in Table 4.4 are used in constructing the rent gradients in
Figure 4.1. When the rent gradients are examined for these crops in competition with
each other, it is noted that only two of the five crops (cotton and rice) would actually be

cultivated according to the Thunen principle of highest and best use. Groundnuts, .

54

millet, and maize are all effectively ‘outbid’ by either cotton or rice.3 In this particular

case, cotton would be the most profitable land use of all crops up to a distance of 200

nub
l
1

Location Rent (fetal ha)

     
  

 

 

sooo - l
l
l
we Millet I Groundnuts
o r l r r .
0 100 200 300 400 500 600 700

Distance ( km )

Figure 4.1 Rent Gradients for Agricultural Crops

km from the market. At this point, the ‘break-even’ point (or margin of transference)
between cotton and rice is reached and the location rent for the two crops is exactly the
same. Beyond 200 km from the market, rice would become the most profitable crop to
cultivate, able to outbid all other crops up to a maximum distance of 680 km from the

market.

The crop zonation pattern suggested by Figure 4.1 does not accurately reflect
that of current agricultural production in the Mopti administrative region. Cotton and
rice are not, of course, the only crops produced. Groundnuts, millet, and maize are also
produced, the former as an important cash crop and the latter two as significant food

crops.

 

3 Rent gradients are sometimes referred to as ‘bid-rent’ curves.

55

There are many factors that influence the spatial pattern of crop production in
the Mopti region (Thom & Wells 1987, pp. 333-338). For example, rice cultivation
takes place predominantly in the inland delta area of the Niger River, where production
depends highly on seasonal flooding of the river valley. The cultivation of groundnuts,
millet, and maize is dependant for the most part on rain-fed production systems, and is
marked by extensive agricultural practices at the margins of the inland delta. Thus, the
spatial variability of production factors, e.g., soils, drainage, relief, etc., have a major

impact on the actual patterns of agricultural activity in the Mopti region.

In addition to the physical factors affecting crop production patterns, certain
aspects of the Malian agricultural economy can also influence the outcome of which
crops are ultimately produced. For example, the predominant position of cotton in the
modeled agricultural space economy (Figure 4.1) is explained in part by its high market
price relative to the other commodities (cotton is Mali’s most important agricultural
export). The fact that rice figures as an important staple food in cities and towns in
Mali helps to explain its relatively important position as the next most competitive crop

after cotton in the zone less than 200 km from the market.

Although the modeled spatial zonation of crop production does not match that of
the actual patterns observed in the Mopti region, the theoretical rent gradients are still
useful in the process of determining the best location for fuelwood production in the
agricultural space economy. Although based on assumptions of the Thunen model
which do not hold in the Mopti region, e.g., equal soil fertility in all areas, the rent
gradients provide useful information regarding how these crops would compete
economically with fuelwood. The analysis now turns to examining the rent gradients

for several fuelwood plantation scenarios.

56

Based on the data for estimated fuelwood plantation costs in a Taungya
production system (Table 4.2) and fuelwood shadow prices (Table 4.3), there are five
possible conditions under which plantation production would be profitable. Whether or
not production is viable is determined by various combinations of yield, market price,
and production cost factors. The five cases in which production is economically

feasible are summarized below in Table 4.5.

Table 4.5 - Conditions for Profitable Plantation Production

 

 

 

 

Qse No. Yield (ton/ha) Price (CPA/ton) Cost (CPA/ton)
1 3.0 26,880 22,500
2 3.0 33,600 22,500
3 5.0 26,880 13,500
4 5.0 33,600 27,000
5 5.0 33,600 13,500

 

In Table 4.5, cases one and two are based on the lower expected fuelwood yield
(3.0 tons/ha) and the low production cost (scenario #3 @ 22,500 FCFA/ton) from the
range noted in Table 4.2 above. The first two cases differ in the market price for
fuelwood, i.e., the shadow price. Case one is based on an assumed retail price for
kerosene of 120 CFA/litre (26,880 FCFA/ton) - consider this the ‘base shadow price’,
whereas in case two the retail price is based on 150 CPA/litre (33,600 FCFA/ton),
representing a 25 percent increase. Case three is based on the higher fuelwood yield
(5.0 tons/ha), the base shadow price of 26,880 FCFA/ton, and low production cost
(scenario #3 in Table 4.2). Finally, cases four and five are similar in terms of yield (5.0
tons/ha) and fuelwood price (33,600 FCFA/ton). However, the last two cases differ in

terms of plantation costs. Case four is based on the medium plantation cost, whereas

57

case five reflects the low plantation cost (scenarios #2 and #3 respectively from Table

4.2).

Based on the various factors of yield, price, and costs summarized in Table 4.5,
maximum rent and distance figures are calculated and listed in Table 4.6 below. The

results are used to construct rent gradients for each fuelwood plantation scenario. The

rent gradients are plotted in Figure 4.2.

Table 4.6 - Maximum Rent and Distance for Fuelwood Plantations

 

 

Case No. Max. Rent (CFA/ha) Max. Dist. (m)
1 13,140 1 10
2 33,300 278
3 66,900 335
4 33,000 165
5 100,500 503

 

According to the rent gradients in Figure 4.2, the fuelwood plantation cases with
the highest net revenues and most extensive production ‘cones’ (rings circumscribed
around the market center by the rent curves) are, in decreasing order, numbers five,
three, two, four, and one. Higher expected yield (5.0 tons/ha) and associated low
production costs explain the positions of the margin of production for cases five and
three, at 503 and 335 km from the market respectively. Case number five also displays
the effect of the higher fuelwood shadow price. Despite the low yield figure for case
two (3.0 tons/ha), a higher shadow price and low production cost effectively extend the
margin of production for this case to 278 km from the market. Due to a higher
production cost, case number four has a smaller production cone, profitable up to 165

km from the market. Finally, although case number one reflects a low production cost,

58

the lower expected yield and base shadow price restrict profitable production to a

distance of only 110 km from the market.

The next step of the analysis entails the determination of the best location for the
siting of the peri-urban fuelwood plantation according to the results obtained to this
point. In order to do this, the rent gradients for agricultural crops and fuelwood are

plotted onto the same graph (see Figure 4.3).

It is apparent from the results of combining the two sets of rent gradients that,

with the exception of case one, all fuelwood plantation scenarios would be able to

 

 
   
  
  
  
   

 

 

 

 

 
 

 

 

100500
Plantation Assumptions
Yield Price Coat
(tons/ha) (tcla) (feta)
Caaet 3.0 26880 22500
Caae2 3.0 33600 22500
g Caae3 5.0 26880 13500
g Caae4 5.0 33600 27000
1: 5.0 33600 13500
E
E
,8
i
0 100 200 300 400 500 600 700

Distance ( km )

Figure 4.2 Rent Gradients for Fuelwood Plantations

Location Rent (lcla/ ha)

59

100500 ,

 

  

 

 

 

 

 

Distance ( km )

Figure 4.3 Combined Rent Gradients

Plantation Assumptions
Yield Price Coat
(tone/ha) (feta) (lcl‘a)
3.0 26880 22500
3.0 33600 22500
5.0 26880 13500
5.0 33600 27000
5.0 33600 13500
Rice
500 600 700

 

6O

compete effectively with all other agricultural land uses presented in this market area.
Case one would not be able to compete with cotton production at any location, but
could outbid rice, groundnuts, and millet for distances of up to approximately 12, 23,

and 100 km from the market respectively.

Among the plantation scenarios that compete successfully with the other
agricultural crops, it can be argued that either case four or case one would be the most
plausible outcome. Cases two, three, and five are all predicated on the lowest possible
production costs, in addition to either a higher yield (case #3), market price (case #2), or
both (case #5). For these reasons it can be assumed that cases two, three, and five are

all unrealistic outcomes considering Sahelian conditions.

Although case four is based on b0th higher yield and price, it may be more
realistic to assume this outcome due to the higher production cost figure (scenario #2
from Table 4.2). Calculations also show that case four could compete at a lower
shadow price of at least 29,880 CFA/ton (approximately 11 percent increase over the
assumed base price), albeit over a smaller area. Despite the fact that case one (as in
cases 2, 3, & 5) is also based on a low production cost assumption, it may be even more
realistic to expect this outcome over that of case four based on lower yield and price
factors. However, in order for case one to be able to outbid cotton, and rice for more
that just the area within the closest proximity to the market (approximately 0-12 km), its
rent gradient calculation would require an increase in the shadow price to a minimum
level of 27,300 CFA/ton, an increase of approximately 1.6 percent above the assumed
base shadow price (see gradient 1’ in Figure 4.4). Nevertheless, based on these
assumptions cases one and four appear to be more likely outcomes than the remaining

C3565.

Location Rent (lcia / ha )

 

61

 

 

 

 

 

35000 ..
30000 --
Plantation Assumptions
Yield Price Coat
25000 -- (tons/ha) (iota) (feta)
c... a Case 1 3.0 26880 22500
Case 1' 3.0 27300 22500
Case 1" 3.0 28224 22500
Case 4 5.0 33600 27000
Case 1"
15000 --
10000 -
5000 -
Groundnuts Cotton Rice
0 i i l
0 300 400 500 600
Distance ( km)

Figure 4.4 Optimal Fuelwood Plantation Rent Gradients

 

62

Assuming that case one, if modified by an increase of 5 percent above the base
shadow price to'28,224 CPA/ton (in order that it might compete more effectively with
cotton - see rent gradient 1"), is the most likely fuelwood plantation to succeed, it can
be concluded that plantation-produced fuelwood, in this particular space economy,
would be the optimal land use in the annulus located between 0 and approximately 31
km from the market.4 In the remaining zones of this modeled market area cotton and
rice would be produced. Cotton production would be the best land use in the annulus
located between 31 and 200 km from the market. Following cotton, rice would be the
best suited crop for production in the annulus located between 200 and 680 km from the

market.

In light of this. hypothetical analysis, it should be recalled that the actual
locations of crop production under real-world conditions will be greatly influenced by
physical and socio-economic factors not accounted for by the basic model conditions.
As has been noted for the case of agricultural production in the Mopti region, different
crops tend to be produced in their own specific ‘spatial niches’. Therefore, the optimal
location for fuelwood production as determined by this analysis may vary substantially

depending upon prevailing local conditions.

Locating the Firewpo_d[Charcoal Threshold

As mentioned earlier in this study, detemrining the location of an economically
viable peri-urban fuelwood plantation is only one element in the analysis of this type of

urban energy production system. A second element that needs to be addressed is the

 

4 The 31 km limit of fuelwood production is determined by calculating the "break-even
point" between fuelwood production and the next most profitable land use, which in this
case is cotton. The break-even point represents the distance from the market at which
adjacent land uses generate equal location rents.

63

forrrr in which fuelwood will ultimately be produced, i.e., firewood or charcoal. This
entails determination of the firewood/charcoal calorific threshold.

This location is important in that, due to energy loss in the carbonization process
and the wood-to-charcoal conversion factor, charcoal production can contribute
significantly to increased rates of deforestation in the source area (or increased rates of
harvested acreage in plantations). It should be pointed out, however, that locating the
firewood/charcoal threshold is of greater importance when dealing with traditional
methods of forest clearance for fuelwood production. For example, in cases where the
cost of wood does n_ot reflect its intrinsic value, the transportation costs of bringing
fuelwood to market constitute a major percentage of final market price - up to 50
percent in some instances (Leach and Meams 1988). In plantation situations, the fact
that wood bears a higher, more realistic production cost reduces the percentage of the
final retail price attributable to transport costs. Nevertheless, in situations where
transport costs are relatively high, it is important to determine the location of the
firewood/charcoal threshold and attempt to plan for the production of wood within this

limit.

To determine the location of the firewood/charcoal threshold, data are required
on the costs of producing firewood and charcoal in terms of equal amounts of energy
content. In addition, a freight rate for transporting the fuelwood to market must be
known. Using the data for production cost and freight rate from plantation case number
one from the previous section, the firewood/charcoal threshold for this best-case

scenario is determined below.

The first step in determining the firewood/charcoal threshold is to calculate the
production costs of both firewood and charcoal in terms of a common energy

equivalent. Table 4.7 gives these costs per ton of coal equivalent (CE), based on a

54

conversion factor of three tons dry wood to one ton of charcoal (Earl (1975) estimates
this ratio at 4:1; Poulsen (1978) at from 2:1 to 4:1, depending on tree species and

moisture content).

Table 4.7 - Production Cost of Firewood and Charcoal Per Ton CE Energy,“

 

Production Costs Firewood Charcoal
CFA/ton (raw wood) 22,500 67,500
CFA/ton CE 33.075 65.475

 

* Based on the following energy equivalents:
1 ton CE = 6.9 * 106 kcal
1 ton dry wood = 4.7 * 106 kcal
1 ton charcoal = 7.1 * 106 kcal

note: 1.47 ton firewood/ton CE and .97 ton charcoal/ton CE

 

Based on the costs of firewood and charcoal per ton CE, and using the freight
rate of 40 CFA/ton/km (from Table 4.2), the firewood/charcoal calorific threshold
would be located at a distance of 1620 km from the market.5 To demonstrate how this
threshold point is sensitive to changes in transport costs, a second threshold value is
calculated based on a doubling of the freight rate. In this case, a fi'eight rate of 80
CFA/ton/km would result in a firewood/charcoal threshold located approximately 810
km from the point of consumption. Hence, a doubling of the freight rate results in

halving the distance between the market and the firewood/charcoal threshold.

 

5 The equation for determining the firewood/charcoal threshold is as follows: PROD.
COST WOOD/1' ON CE + (1.47 TON WOOD/T ON CE * 40 CFA/T' ON WOOD/KM *
KM) = PROD. COST CHAR/1' ON CE + (.97 TON CHAR/TON CE * 40 CFA/I‘ON
CHARJKM * KM). Solving for KM (distance) will detenrrine the location of the
threshold.

65

According to these results, it is abundantly clear that fuelwood produced under
the scenario presented by case number one would be more economical to supply in the
form of firewood rather than charcoal. This would be true at any location within the
zone of profitable production. The location of this firewood/charcoal threshold is
significant in that under this particular set of conditions for production and
transportation costs, the harvesting of fuelwood should take place in a sustainable
fashion without adverse effects on the surrounding environment. However, in some
circumstances consumer preference for charcoal over firewood, particularly in crowded,
enclosed urban conditions, can influence the form in which the fuel is actually

marketed.

Preliminary Interpretation of Results

The primary question posed in this study relates to the determination of where,
in terms of the agricultural space economy, peri-urban fuelwood plantations could be
successfully located. According to the Thunen analysis for agricultural crop data and
the hypothetical fuelwood plantation considered for the Fifth Region of Mali, the results
demonstrate that, under certain assumed conditions, plantation-produced fuelwood can
compete with other agricultural land uses and be cultivated in an area proximate to the
consuming market. This suggests that it may be possible to alter the production system
from the current practice of fuelwood mining in remaining forested areas distant from
the market, to one in which fuelwood is produced on a competitive, sustainable yield

basis in the urban periphery.

Before any definitive conclusions are reached, however, it is imperative to point
out two critical sets of issues. First, the Thunen analysis in this case study is capable

only of providing a general sense of where, in terms of location rent, fuelwood could be

66

produced competitively in the area around the market center. There is no accounting
for site-specific factors, e.g., soil quality, water availability, etc., related to the
production of a given selected tree species. In addition, there are other competing uses
of land on the urban periphery which have not been dealt with in this analysis. The
potential for truck gardening, housing development, and other commercial uses of land

must be assessed in addition to the agricultural land uses discussed in this study.

A second set of issues, closely related to the first, concerns the assumptions used
for the fuelwood data in this analysis (as well as the agricultural crop data). The
success of any given fuelwood plantation scenario will ultimately depend on the
reliability and accuracy of the corresponding data. The data used for the hypothetical
Eucalyptus plantation are at best rough estimates, and may not adequately reflect the
local conditions of this study area. For example, the low production cost figure used for
plantation case one would most likely require that optimal site-specific physical
requirements for tree growth be met. In addition, favorable labor and capital costs
would be necessary in order to attain such low total production costs. The rate of
compensation provided by the implementation of a Taungya system is also a point of
concern. Although considered a valid and tested procedure, the benefit from this type
of agroforestry intervention is highly dependent on site-specific conditions. Whether
these requirements and assumptions are realistic for the difficult conditions extant in the

Sahelian environment must be carefully scrutinized.

Another assumption made in the analysis of plantation case number one (i.e.,
rent gradient 1" in Figure 4.4) regarding the estimated market price must be considered.

The value used in the location rent equation for case 1" is based on a value 5 percent

67

M the estimated base shadow price of fuelwood.6 Depending on the price stability
of kerosene, it is by no means certain that fuelwood priced at this level would still be
competitive. Although the fuelwood price figure used is 5 percent above the shadow
price, plantation case number one could still compete with cotton production (although
over a much smaller area) at a market price of just over 1.56 percent above the base
shadow price (see gradient 1’). In addition, any incremental increase in the production
cost and/or reduction in price figures would effectively decrease the competitiveness of
the fuelwood crop in this scenario. The low yield in the rent equation is probably the
most reliable of the figures used; however, this too depends on the site-specific
characteristics of the production system. Because of imperfections in available
fuelwood data, an attempt has been made to maintain as conservative an approach as
possible in selecting price, cost, and yield criteria upon which the decision of the best
plantation scenario is made. Avoiding the selection of plantation parameters that would
be unlikely to obtain in the Sahelian environment provides for as realistic an outcome as

possible.

Noting these provisos and caveats, one could cautiously state that there is
potential for the establishment of a peri-urban fuelwood plantation at some location in
close proximity (0-31 km) to the urban market. This would be considered a desirable
improvement over a fuelwood production system represented by the collection of ‘free’
resources at greater distances from the point of consumption. However, the benefits
obtained from reduced environmental degradation in rural forested areas (coupled with
any downstream benefits to the national economy through improved agricultural
production) would need to be estimated and compared to the increased costs to

consumers of plantation-produced fuelwood (coupled with any other costs, e.g., benefits

 

6 Whether or not this price for fuelwood would engender a switch to kerosene would
still depend on other factors, such as the regular supply of this alternative fuel and the
users’ ability to purchase the appr0priate cookin g devices.

68

foregone from alternative land uses) in order to arrive at some reasonable societal-level
benefit-cost analysis of such as spatial reorganization of the fuelwood production
system. An attempt must be made to establish some quantitative valuation of these
extemalities and incorporate them into the overarching social benefit-cost analysis. It
should be understood, however, that the data collection task thus required would be

extraordinary in most deve10ping-country situations.

The results of the data analysis given in this chapter have shown that
establishing fuelwood plantations in proximity to urban centers, rather than mining
forested rural areas to meet urban demand for energy could, under certain
circumstances, be economically feasible. The next chapter will proceed to evaluate the
use of the Thunen model in this particular application. In addition, Chapter V will
elaborate on the brief interpretation of the results discussed up to this point,
emphasizing the secondary questions of this study concerned with the economic impact,
ecological ramifications, and social constraints and opportunities of a different spatial

pattern of fuelwood production.

CHAPTER V
DISCUSSION

The first section of this chapter evaluates the use of the Thunen model as a
method of analysis for determining the possible location of peri-urban fuelwood
plantations. Following this, the evaluation of the results obtained in the previous
chapter is elaborated upon in terms of the secondary questions raised by this study. The
discussion explores in greater detail the economic impact, ecological ramifications, and
social constraints and opportunities that could be expected to follow from a potential
spatial reorganization of fuelwood production. Finally, policy problems associated with
such a change in the location of fuelwood production are briefly examined. These
issues are discussed in the context of the ‘regional political ecology’ framework

mentioned earlier in the study (see page 7).

Evaluation of the Thunen Model Application

In evaluating the use of the Thunen model as a method of analysis for
determining the location of peri-urban fuelwood plantations, it is important to realize

certain key limitations imposed by the model’s assumptions.

First, as was demonstrated by the analysis of crop data for the Mopti region of
Mali, factors related to the natural environment have a major impact on the spatial
pattern of agricultural land use. Whereas the Thunen model assumes uniform fertility
of the land, the actual pattern observed in the Mopti case is determined in greater part
by variability in land suitability rather than comparative location rent values. In the
same way that rice, millet, and maize occupy certain locations due to physical factors,

the siting of a given tree species in a plantation situation will also depend on specific

69

70

natural conditions. Obviously, the Thunen model, as it is commonly applied, does not
take into consideration the important effects on land use imposed by the physical
characteristics of land resources. It is only when assumptions of the model are relaxed

that influences of real-world conditions can be examined.

Second, the crop zonation patterns observed in T'hunen’s base model result from
the interaction of a single market center with its complementary agricultural region.
The possibilities, however, for trade and exchange outside the "isolated state" may
strongly influence actual crop production patterns. In the case of the M0pti region,
linkages with the greater Malian and West Afiican markets would appear to have an
affect on the type and extent to which certain crops are produced within the region. In
the case of fuelwood, it would be necessary to attempt to determine the degree to which

production might be influenced by these external linkages.

Third, Thunen’s model is based on land use around a static urban market center
with relatively stable boundaries. However, with urban expansion as a dynamic process
occuning over space and through time, the spread of non-agricultural land uses around
the urban fringe influences the uses to which such land will be put. Hence, land in peri-
urban environs has an opportunity cost associated with it that is not taken into account

when determining its best agricultural use based on Thunen’s model.

This is an important factor to consider when attempting to determine the best
location for peri-urban fuelwood plantations. Plantation development requires a long-
term investment in land, which is subject to various other non-agricultural uses as urban
centers continue to expand rapidly as is the casein the Sahel today. However, Earl
offers some thoughts regarding the decision making process involved in weighing the
opportunity cost of land against the consideration of plantation establishment. He states

that if the provision of fuelwood is regarded as a

71

social function of the State (through the medium of the forest service), an
expected financial return from plantations approaching the marginal cost of '
capital should be sufficient justification for the establishment of fuelwood
plantation schemes....[since] the economic or social rate of return would be
normally expected to be much higher (Earl 1975, p. 91).

This outlook on the role of the state is consistent with the regional political ecology

concept as it relates to the allocation of land-based resources for societal benefit.

Despite the limitations noted above, applying Thunen’s model to the locational
analysis of peri-urban plantations does provide useful information for establishing
fuelwood as a crop in the agricultural space economy. This is particularly the case with
regard to determining the parameters under which fuelwood plantations could compete
with other agricultural land uses. Based on the analysis of the data for the Mopti
region, the results give a fairly clear indication of the plantation yield, market price, and
production cost conditions that must be met in order for plantations to be viable and

occupy a location in proximity to the urban center.

In addition to the information provided on yield, cost, and price factors, the
T'lrunen model is particularly well-adapted to explain patterns of agricultural land use
based on transportation costs. Indeed, Thunen’s arguments for the agricultural zonation
patterns he observed in his time were based on transportation costs, as a function of

distance from the market, as the primary force determining such patterns.

The importance of transportation costs as a factor in fuelwood exploitation is
clearly stated in this study. Under the present conditions of fuelwood collection in
remote forested areas, transportation costs bear directly on increasing market prices for
energy in the urban centers of the Sahel. In addition, it has been demonstrated how
transportation costs as a function of distance affect the form in which fuelwood is

marketed. Greater distances from market and higher transportation costs eventually

72

result in crossing the firewood/charcoal threshold, which in turn results in accelerated

rates of deforestation and environmental degradation.

As Sinclair (1967, p. 76) has noted, where the condition of transportation
infrastructure in the developing world today resembles that of Europe and North
America in the last century, it is likely that the Thunen principles of agricultural land
use are still applicable. Therefore, where transportation costs are essentially
pr0portional to distance and a factor of bulk, as they are for fuelwood, one could expect
to discover zones of forest products in relative proximity to the urban market. This
pattern is evident in Thunen’s observations as reported in the Isolated State (Hall 1966),
as well as in Horvath’s (1968) contemporary study of the zonation of agricultural land
use around Addis Ababa, Ethiopia. According to the steeper rent gradients obtained in
this study for fuelwood plantations vis-a—vis other crops (see Figure 4.3), it is evident
that higher transportation costs for fuelwood would restrict this land use to areas closer

to the market.

It is important, however, to note that factors associated with the dynamics of
urban land use may reduce the attractiveness of establishing peri-urban plantations.
Although siting plantations close to urban markets is expected to reduce the transport
costs of fuelwood, plantations may lose their economic feasibility as urban areas
continue to grow. Rising land values, wage rates, and the profitability of alternative
uses of land can quickly erode the feasibility of peri-urban plantations. The opportunity
costs of other non-agricultural land uses also need to be considered for the zone

determined best-suited for fuelwood production.

Finally, in much the same way that T'hunen himself proceeded, it would be
interesting to attempt to deterrrrine how relaxing the model’s assumptions would affect

the location of peri-urban plantations in the Sahel. For example, in Thunen’s

73

consideration of a navigable river (or other transportation improvement) traversing the
flat plain, he concluded that the concentric ring pattern would be altered to the effect
that land use zones would be elongated roughly parallel to the direction of the
transportation artery. In many instances, Sahelian cities are located such that they are
positioned along rivers (e.g., Bamako, St. Louis (Senegal), Mopti, Niamey) or at
various points along the region’s rail lines (e.g., Dakar, Bamako, Ouagadougou).

Economic Impacts of Fuelwood Plantations

Economic analysis of the establishment of peri-urban fuelwood plantations
should attempt to take into consideration the cost and benefit flows attributable to this

type of energy production and resource management intervention.

In terms of the economic impact resulting from the establishment of a peri-urban
fuelwood plantation, the most obvious statement that can be made is that fuelwood
prices in the urban market must rise in order for plantations to be viable.1 Changing
from a system of depletive resource ‘mining’ to more sustainable, regenerative
production practices will require that the price paid for this renewable energy resource
reflects more accurately the cost of its replacement. The ability of urban fuelwood
consumers to pay more for their energy needs is, of course, related to the broader issue
of increased economic development in a given country. It is beyond the scope of this
study to investigate the dynamics of national economic growth in developing countries.
However, the spatial reorganization of fuelwood production systems in such countries

will likely impose additional economic (and perhaps political) difficulties on the urban

populace.

 

1 Prices should be expected to rise with or without plantations, as distance to the market
increases. However, overall societal costs due to degradative ‘mining’ could be much
higher than the cost of adjusting to higher-priced, plantation-produced fuelwood.

74

In considering the energy needs of the Sahelian population, fuelwood should be
thought of as a locally available renewable energy for these resource-poor countries
which lack the financial means to make the switch to fossil fuels. Although rural
populations may satisfy their energy requirements from their immediate surroundings,
urban dwellers must acquire their fuel needs from the land-based capital stocks of forest
resources in rural areas. However, energy derived from these natural resources obviates
the need to depend upon imported petroleum products, which are acquired only at great
expense to the national treasury. Therefore, the development of sustainable supplies of
wood-based fuels could form the basis for meeting the short-to-medium term energy
requirements of Sahelian countries. Despite the image of fuelwood use in developing
countries as representing a "bondage to a past that is unacceptable” (Goodman 1987, p.
112), such use may be ultimately easier to secure than fossil fuels which require
substantial expenditures of foreign exchange. Hence, developing systems of fuelwood
production in the urban periphery could be a cost effective means of securing necessary

energy supplies.

If the above statements are considered from the perspective of the regional
political ecology concept, then one can observe the linkages that exist between the
political economy of energy supply and the issue of effective natural resource
management. Unable to increase their consumption of fossil fuels due to economic
constraints, Sahelian countries continue to rely on fuelwood. As tree stocks are
depleted in the areas closest to growing urban centers, the increasing demand for fuel
must be met by harvesting the extant supply of energy in the forests of the hinterland.
Rather than making an investment in ‘landesque capital’, i.e., reforestation, the
remaining resources are mined with little consideration for the effects this has on the
ecological destruction of productive agricultural lands, increasing energy costs to urban

consumers, or the overall stability of the national economy.

75

As an alternative to depletive resource use, peri-urban plantations, where
feasible, should be encouraged as a means of meeting energy demand. Situated in a
suitable location proximate to urban markets, such plantations would have the potential
to 1) correct the inequities of supplying urban areas with energy at the expense of rural
populations, 2) stabilize the processes of environmental degradation, 3) improve the
potential productivity of predominantly rural-based economies, and 4) reduce
dependence on prohibitively expensive petroleum imports. It would appear that such an

alternative system distributes its benefits in an equitable fashion.

As a final note on the economic impact of peri-urban fuelwood plantations, it
should be pointed out that it remains unclear as to exactly who will be able, or willing,
to make the required investment in this type of energy production system. One might
consider it to be the role of the state to assure the establishment of such landesque
capital through forestry interventions. However, much remains to be investigated
regarding this point. Questions on issues such as subsidies to potential plantation
investors, taxation policy, and what, if any, pricing mechanisms should be considered
will need to be explored. To date, the public and parastatal sectors in Africa have not
performed very well. Nevertheless, due to the natural resource and planning policy
decisions required in order to establish peri-urban plantations, there will inevitably be a

role for the state to play in such undertakings.

Ecological Considerations

~The establishment of fuelwood plantations in areas surrounding Sahelian urban
centers has the potential to produce a dual response in environmental rehabilitation. A
shift in the spatial pattern of production for urban supply from a primarily rural-based

operation to a peri-urban one could generate ecologically beneficial results for both

76

environments. This change would represent a major improvement over the currently
expanding limits of degraded lands surrounding urban centers, and the increasingly

vulnerable resources of rural areas.

Assuming that plantations can be developed in areas of the urban periphery
previously devoid of forest cover, their establishment should produce some regenerative
benefits. Erosion control, soil stabilization, and microclimate improvement are some of
the secondary benefits that can be expected to accompany plantation development.
While such benefits accrue to the areas in proximity to the peri-urban plantations,
concurrent ecological improvements could be realized in the previously exploited rural
areas. Assuming that the amount of fuelwood produced in plantations is sufficient to
meet most of the urban demand, forested zones in rural areas should benefit from lower
rates of deforestation than are currently experienced. A situation in which the M.A.I.
(mean annual increment) of forest growth approaches the annual harvested cut required
to meet rural needs would be a positive sign of moving toward achieving an
ecologically stable situation. Under current conditions, according to Anderson (1987,
p.7), demand exceeds M.A.I. of growth by approximately 30 percent in most Sahelian
countries; however, severe levels of off-take are evident in Niger where demand

exceeds M.A.I. by 200 percent!

Experiments with greenbelt projects around Ouagadougou, Burkina Faso
(Weinstabel 1984), demonstrate that increasing the amount of woody biomass around
the city contributes to soil erosion control and the reduction of wind-blown sand and silt
pollution in the built-up areas. The reestablishing of vegetation cover on degraded land
is aided by the development of a humus layer resulting from tree litter. The root
systems of trees and shrubs also contribute to better water retention in the surface layers

of the soil.

77

Additional ecological benefits would most likely accrue from the application of
agroforestry techniques in pen-urban plantations. A groforestry systems seek to
optimize the positive interactions between the various components (e.g., crops, trees,
animals) in order to achieve a more productive and protective, i.e. sustainable, output
from the land than would be achieved through conventional agricultural practices. As
already mentioned in this study, agroforestry techniques are virtually an a priori
requirement for achieving economic feasibility in this type of land use investment.
Several studies support this position (Arnold 1983; Harou 1983; and Hoekstra 1983a,
1983b). In addition, the contributions of an agroforestry approach to ecologically
sustainable land use and fuelwood energy production have been the focus of numerous
studies (NAS 1983b; Winterbottom & Hazlewood 1987; Repetto 1987; Cook & Grut
1989; and Weinstabel 1982, 1984).

An example of the ecological benefits of an agroforestry intervention in the
Sahel is the intentional planting of Acacia senegal, more commonly known as gum
arabic, within a traditional farming system (Pearce 1988). This leguminous species is
well-adapted to the semi-arid ecological Sahelian zone, and provides a range of
products and services. For instance, fodder is provided for domestic animals, fuelwood
for cooking, and gum arabic as a marketable commodity. In addition, soil stabilization
and nitrogen fixation help to improve field conditions and yields for the crop

component of the system.
Social Constraints and Opportunities
Relocating the production of fuelwood energy for urban areas from distant rural

locations to plantation schemes surrounding towns and cities will impose certain social

opportunities and constraints on the nations of the Sahel. The composition of groups

78

currently involved in fuelwood exploitation may change considerably with a shift in the
focus of activity from rural to peri-urban zones. These changes would most likely have
the greatest effect on employment and investment patterns in the fuelwood business.
The focus of activity would of necessity shift from rural forest collection areas to more
formally controlled peri-urban production sites. As it currently stands in many Sahelian
countries, much of the fuelwood industry is considered to belong to the informal or
‘dual’ sector of the economy. As a part of the informal sector, there are large numbers
of people employed as occasional, seasonal workers in this activity. This is significant
for Sahelian countries in which unemployment and underemployment are relatively
high. A change in the location of fuelwood production (with its opportunities and
constraints) must take into consideration the impact such a change will have on the

status of the employment picture.

The changes that take place when the provision of fuelwood energy switches
fiom household collection to that of a monetized good were discussed earlier in this
study. Formerly a ‘free good’ (opportunity costs aside), fuelwood becomes a
marketable commodity for which city dwellers are required to devote a portion of their
cash expenditures. As this transformation takes place, a complex production and
marketing apparatus forms to supply growing urban centers. The stages in the process
include the production (felling of trees and conversion to charcoal), transportation to
market, and distribution of product from fuelwood wholesalers to retailers throughout

the neighborhoods of Sahelian cities and towns.

Studies related to urban fuelwood markets in the Sahel have looked at the nature
and composition of the marketing and distribution channels which have developed to
supply urban centers (Ribot 1989; Republique du Burkina Faso 1987). Ribot’s study of

the charcoal market in Senegal indicates that a significant number of people in rural

79

areas are employed (mostly seasonally) in the fuelwood production stage, while large

numbers in urban areas are involved in the wholesale and retail stages.

Shifting fuelwood production activities to peri-urban locations will obviously
reduce off-farm employment possibilities for rural villagers. Whether this change
should be looked upon as depriving the villager of opportunity is debatable.
Involvement in fuelwood production does represent an occasion for earning cash to
supplement farm incomes. However, wages paid to rural workers are comparatively
low, and the fuelwood merchant is not particularly concerned with reforestation
activities once all the harvestable wood has been collected. In effect, rural farmers are
cutting down trees for a relatively small profit, while contributing to environmental
degradation and their own impoverishment in the process. Low rural wages and the
frequent evasion or non-payment of stumpage fees2 result, in effect, in a subsidy on the

price of fuelwood to urban consumers made at the expense of the rural population.

According to the regional political ecology framework, this exploitation of the
rural environment for the benefit of providing energy to urban areas is a good example
of an urban bias in the dialectic between society and land-based resources. The
allocation of forest resources to meet urban needs, without compensatory action, i.e.,
reforestation, results in the reduction of the productive capacity of rural populations.
The service functions of forested land which benefit rural agriculture, e.g., erosion
control and nitrogen fixation, are foregone when there is no effective management

policy in place (or enforced) to protect the land-based capital of rural populations.

 

2 Stumpage fees are intended as a ‘clearance fee’ or tax, payable by those involved in
the exploitation of fuelwood resources. The official government agency responsible for
regulating forestry-related activities is usually charged with collecting the stumpage fee.
However, payment of this tax is frequently circumvented (see Ribot 1989).

 

8O

Peri-urban fuelwood plantations would indirectly serve to correct some of the
urban bias in resource allocation mentioned above. The establishment of a sustainable
production system closer to urban markets could result in a stage of environmental
recuperation in degraded rural areas, while at the same time contribute to the restoration
of previously degraded areas surrounding urban centers. Peri-urban plantations would
also generate employment opportunities for some of the unemployed now crowding into

the already saturated urban job market.

Policy Problems

In proposing an alternative to fuelwood mining in rural areas, this study has
shown that any new system must be economically viable, technically feasible, and
socially acceptable. In addition, the alternative system must be institutionally
supported. The neglect of this last factor can easily lead to the failure of the proposed
alternative. Bromley, as cited in Southgate and Hitzhusen (1987, p. 77), states that

the disappointing performance of many Third World environmental projects is

explained by the fact that those projects usually do little to alter the incentives

existing under prevailing institutional regimes to manage resources in a

depletive fashion.

It can be argued that Bromley’s point has been proven correct in projects related
to the forestry sector in the Sahel, as well as in other parts of Afiica, particularly at the
village level. Village-level woodlots have been promoted in Sahelian countries as a
means of securing adequate fuel supplies as well as the ecological benefits accruing
from trees. However, these projects frequently fail to achieve their expected results
(Brokensha & Castro 1984; Fortmann 1985; and Skutch 1983). Many projects fail

because the intended beneficiaries, i.e. the rural population, are not convinced that the

investment required of them (financial or in-kind) is worthy of the risk. Under the

81

severe conditions of the Sahel, it is understandable that rural peasants would be risk"
averse. In addition, factors such as the lack of institutionally secured land tenure rights
(Thomson 1981), and the relatively poor relationship between rural farmers and the
police-like structure of many African forestry departments (Shaikh 1985/86, pp. 51-3)

contribute to project failure.

The issue of institutional support (i.e. political sanction) for peri-urban fuelwood
plantations will be as important here as in the case of the village-level situation
mentioned above. It is clear that in order to develop a sustainable energy supply system
based on fuelwood, the unpleasant political reality of higher prices will have to be
defended. In addition, stricter enforcement of measures designed to protect the
remaining stocks of forested land will be required. An increase in market price for
fuelwood is likely to encourage more tree cutting outside formal plantation channels.
The control of untaxed fuelwood transiting into Sahelian cities is already a major
problem. The potential for corruption and mismanagement will present difficult

challenges to the political structures in the Sahel.

The discussion in this chapter points out that, despite some of its limiting
assumptions, the T'hunen model of agricultural land use can provide useful information
when applied as a method of analysis to determine the best location for peri-urban
fuelwood plantations. Its application assists in determining the econorrric parameters
under which fuelwood production could compete with other agricultural crops, and it
gives some indication as to the location of such land use. Although the model does not
effectively consider the important physical requirements of plantation establishment,
nor the dynamic processes of urban expansion and their effect on peri-urban land use, it
does demonstrate how transportation costs would effect the location of various

agricultural land uses including fuelwood production.

82

This chapter has also considered the economic, ecological, social, and policy
implications of a locational shift in fuelwood production. Despite the high financial
cost involved in producing fuelwood in a plantation scheme, and the fact that these
costs would have to be borne ultimately by urban energy consumers, it would appear
that a spatial reorganization of wood-based energy production might result in greater
economic benefit to society at large. Provided that sustainable production systems of
this renewable energy resource can be established, with the requisite political support
such an endeavor would demand, a return on investment to the national econorrries of
Sahelian countries should be realized in the form of higher agricultural productivity and

a more stable ecological balance.

CHAPTER VI
SUMMARY AND CONCLUSIONS

According to the literature reviewed at the outset of this study it is clear that
Sahelian countries will continue to depend on wood-based energy for some time to
come. The fuelswitching strategy discussed in chapter two can only begin to take effect
when higher levels of economic growth are reached. The prospects for this taking place
in the debt ridden, resource poor countries of West Africa in the near future are dismal
at best. In addition, the implementation of energy conservation measures such as the
dissemination of improved cookstoves, while a positive step in the right direction, can
only be expected to contribute in an complementary fashion to reducing the rapidly

growing deficit between fuelwood production and consumption.

In contrast to fuelswitching and conservation measures, fuelwood supply
enhancement strategies may be able to effectively respond to meeting the energy
requirements of Sahelian cities and towns. Under certain conditions, such strategies
may also be able to contribute to stabilizing the degradative processes currently
afflicting the fragile ecological balance of the Sahel. The establishment of peri-urban
fuelwood plantations may be one option capable of producing wood-based energy in the

quantity that will be required to meet growing urban demands.

The actual establishment of peri-urban fuelwood plantations will depend on
meeting an array of prerequisite conditions. First, a consensus must be reached on the
part of political decision makers that such an approach should be carried out for reasons
that it is in the long-term best interests of society. The imperative of reversing the
pattern of depletive resource exploitation, and in turn investing in productive and
protective land uses must be a set precondition by which resource allocation decisions
can be guided. The idea of landesque capital investment, which seeks to safeguard the

83

84

productive capacity of land-based resources for future social and economic benefit, '
provides a useful framework from within which these land use decisions can be made.
From this perspective, plantation development must be seen as an investment geared
not only toward assuring a supply of energy, but also as an effective means of securing

future agricultural productivity and hence economic deve10pment.

Second, a decision to proceed with a peri-urban fuelwood plantation strategy
must be subjected to a method of locational analysis in order to determine where this
type of land use could be established under economically viable conditions. This study
has endeavored to demonstrate how the principles of the Thunen model of agricultural

land use could be employed to determine the best location for this use.

Third, a technical feasibility study will be required that is capable of combining
the information obtained from the economic-locational analysis with site-specific
physical requirements for plantation establishment. Although this point seems blatantly
obvious, there have been previous instances in which such feasibility studies failed to
generate a satisfactory physical data base (Freeman and Resch 1985/86). The results of
such poor project planning can be disasuous in terms of wasting scarce financial

TCSOUTCCS.

The location analysis undertaken in this study has generated some potentially
useful information regarding the siting of peri-urban fuelwood plantations. Although
the data used were for the most part hypothetical, and thus must be interpreted with
extreme caution, the results of the location rent analysis indicate fairly clearly the
conditions under which plantations would most likely be able to compete in the peri-

urban space economy. In light of the poor natural resource base and difficult prevailing

85

economic situation in the Sahel, the scope for viable peri-urban plantation development

may be very limited.

From the analysis it can be concluded that the economic feasibility of any
particular fuelwood plantation scheme will be subject to 1) future prices obtained for
fuelwood as well as alternative fuels such as kerosene, 2) attainable fuelwood yields, 3)
level of returns from associated agroforestry crop components (e.g., from a Taungya
intercropping system) which affect total production costs, and 4) potential competition
from other uses of land in the peri-urban environment. It is also apparent from the
analysis that, even under favorable price conditions, fuelwood production is likely to be
restricted to land relatively close to the market center. This may prove to be
problematic under conditions of dynamic urban growth, particularly in light of the long-

term investment required for establishing plantations.

Having established the important economic parameters associated with the
production of fuelwood in peri-urban plantations, this study proceeded to consider
several additional factors related to potential changes in the Sahelian urban energy
supply system. Tire present discussion has illustrated that fuelwood resource
exploitation is not restricted simply to an energy supply issue. It is a complex,
interrelated phenomenon that is best examined in the context of an energy-economy-
environment nexus incorporating societal and political factors of natural resource
management. Regional political ecology theory enables the complexity of the fuelwood
problem to be addressed in a systematic, integrated fashion because it provides an
holistic framework from within which analysis of the related social, political, ecological

and technical factors of fuelwood production can be undertaken.

LIST OF REFERENCES

LIST OF REFERENCES

Agarwal, Bina 1986. Cold hearths and barren slopes: The woodfuel crisis in the Third
World. London: Zed Books.

Allen, Julia C. and Douglas F. Barnes 1985. The causes of deforestation in developing

countries. Annals of the Association of American Geogzaphers v. 75, n. 2, pp.
163-84.

Anderson, Dennis 1986. Declining tree stocks in African countries. World
Development v. 14, n. 7, pp. 853-863.

1987. The econonrics of afforestation: A case study in Africa World
Bank Occasional Paper (new series no. 1). Baltimore, MD: The Johns Hopkins
University Press.

 

Anderson, Dennis and Robert Fishwick 1984. Fuelwood consumption and
deforestation in African countries. Staff Working Paper no. 704. Washington,
DC: World Bank.

Arnold, J.E.M. 1983. Economic considerations in agroforestry projects. Agroforestry
Systems v. 1, pp. 299-311.

Arnold, J .E.M., and J. Jongma 1977. Fuelwood and charcoal in developing countries.
Unasylva v. 29, no. 118, pp. 2-9.

Barlowe, Raleigh 1978. Land resource economics. 3d ed. Englewood Cliffs, NJ:
Prentice-Hall.

Bertrand, Alain 1985/86. Marketing networks for forest fuels to supply urban centers
in the Sahel. Rural Africana v. 23-24 (Fall 1985 - Spring 1986). PP. 33-47.

Blaikie, Piers, and Harold Brookfield 1987. Land degradation and socieg. New York:

Methuen.
Bogach, V. Susan 1985. Wood as fuel: Energy for deve10ping countries. New York:
Praeger Press.

Bouttoud, Gerard 1988. La crise du bois de feu en Afrique de l’Ouest. Afrique
Contemporaine n. 148, pp. 30-48.

Bowonder, B. 1985-86. Deforestation in developing countries. Journal of
Environmental Systems v. 15, n.2, pp. 171-192.

86

87

Brokenshaw, David and Alfonso P. Castro 1984. Fuelwood, agppforespy and natural
resource management. Binghampton, NY: Institute for Development '
Anthropology.

Catterson, T.M., F.A. Galick and T. Resch 1984. Rethinking forestry strategy in
Africa: Experiences drawn from USAID activities. Desertification Control
Bulletin no. 10, pp. 31-37.

Chauvin, Henri 1982. When an African city runs out of fuel. Unasylva v. 33, n. 133,
pp. 11-20.

Chisholm, Michael 1979. Rural settlement and land use: All essafln location. 3d ed.
London: Hutchinson University Library.

 

 

Christophersen, Kjeil A. and G. Edward Karch 1988. Financiriand economic analyses

pf reforestation, soil conservation and improved woodstove activities: Villagp
reforestation project. Washington, DC: Energy/Development International.

 

Conkling, Edgar C. and Maurice Yeates 1976. Man’s economic environmppt. New
York: McGraw-Hill.

Cook, Cynthia C. and Mikael Grut 1989. Agmforesg in sub-Saharan Afiica: A
farmer’s perspective. World Bank Technical Paper N o. 112. Washington, DC:

IBRD.

de Montalembert, MR. and J. Clement 1983. Fuelwood supplies in the developing
countries. FAO Foresu'y Paper No. 42. Rome: Food and Agricultural
Organization of the United Nations.

de Souza, Anthony 1990. A geogrgaphy of world economy. Columbus, OH: Merrill
Publishing Co.

Delwaulle, J. C. and Y. Roederer 1973. Le bois de feu a Niamey. Bois et Forets des
Tropigues n. 152, nov./dec., pp. 55-60.

Di-Meo, G., J.P. Jambes, and R. Guerrero 1985. Dynamigue socio-energetigue a
Dakar. Dakar: ENDA - Environnement et Developpement du Tiers Monde.

Digemes, Turi Hammer 1977. Wood for fuel: Energy crisis implying desertification -
The case of Bara, The Sudan. Ph.D. Thesis, University of Bergen, Norway.

Digemes, Turi Hammer 1979. Fuelwood crisis causing unfortunate land use - and the
other way around. Norsk Geogr. Tidsskr. v. 33, pp. 23-32.

Dunn, ES. 1954. The location of agriculturagrroduction. Gainesville, Fla.:
University of Florida Press.

Earl, DE. 1975. Forest energy and economic development. Oxford: Clarendon Press.

Eckholm, Erik P. 1975. The other energy crisis: Firewood. Worldwatch Paper no. 1.
Washington, DC: Worldwatch Institute.

88

Eckholm, Erik P. et al. 1984. Fuelwood: The energy crisis that won’t go away.
London: Earthscan/IIED.

El-Hinnawi, Essam 1982. Fuelwood: The environmental dimension. In Proceedings
of the ESCAPzEAOZUNEP expert gmup meeting on fuelwood and charcoal.

Energy Resources Development Series N o. 24. New York: United Nations.

Elkan, Walter 1988. Alternatives to fuelwood in African towns. World Development
v. 16. PP. 527-533.

Floor, Willem and Jean Gorse 1988. Household energy issues in West Africa. Chapter
5 in Desertification Control apd Renewable Resource Management in the
Sahelian and Sudanian Zones of West Africa. eds. Francis Falloux and Aleki
Mukendi. World Bank Technical Paper N o. 70. Washington, DC: IBRD.

Foley, Gerald 1986. The economics of fuelwood substitutes. Unasylva v. 38, n. 151,
pp. 12-20

Fortmann, Louise 1985. The tree tenure factor in agroforestry with particular reference
to Africa. Agmforesg Systems v. 2, pp. 229-251.

Freemann, Peter H. and Tim Resch 1985/86. Large plantations of rapidly growing
exotic species: Lessons forom the Bandia, Senegal. Rural Africana v. 23/24,
fall 85/winter 86. PP. 87-93.

French, David 1986. Confronting an unsolvable problem: deforestation in Malawi.
World Development v. 14, n. 4, pp. 531-540.

Gill, Jas 1987. Improved stoves in developing countries: A critique. Energy Policy
(April). PP. 135-144.

Goodman, Gordon T. 1987. Biomass energy in developing counuies: Problems and
challenges. Ambio v. 16, n. 2-3, pp. 111-119.

Goor, Amihud Y. and CW. Barney 1976. Forest tree planting in arid zones. 2d ed.
New York: Ronald Press.

Gorse, Jean Eugene, and David R. Steeds 1985. Desertification in the Sahelian and
Soudian zone of West Africa. Washington, DC: World Bank.

Grainger, Alan 1982. Desertification. London: International Institute for Environment
and Development.

Griffin, Ernst 1973. Testing the Von Thunen theory in Uruguay. Geogmphical
Review v. 63, n. 4, pp. 500-16.

 

Gritzner, J. 1982. Deforestation and environmental change in the West African Sahel.
(commissioned paper). Washington, DC: Office of Technology Assessment.

1988. The West African Sahel: Human agency and environmental
change. Chicago Geography Reseacrch Paper No. 256. Chicago: University of
Chicago Press.

 

89
Hall, Peter (ed.) 1966. Von T'hunen’s isolated state. Oxford: Pergamon Press.

Harou, RA. 1983. Economic principles to appraise agroforestry projects. Agr_icultural
Administration v. 12, pp. 127-139.

Harrison, Paul 1987a. A tale of two stoves. New Scientist 28 May, pp. 4043.
1987b. Thp gr_eening of Africa. London: lTED.

 

Harvey, David 1966. Theoretical concepts and the analysis of agricultural land-use

patterns in geography. Annals, Association of American Geogzaphers v. 56, n.
2. PP. 361-374.

Henry de Frahan, Bruno et al. 1989. Etude de faisabilite de l’extension de la Division

d9 Recherches sur les systemes de production rurale en Cinguieme Region du
Mali. Bamako: Ministere de l’Agriculture, Republique du Mali.

Hoekstra, D.A. 1983a. The use of economics in agzoforesg. ICRAF Discussion
Paper No. 16. Nairobi: ICRAF (lntemational Council for Research in
Agroforestry).

1983b. An economic analysis of a simulated alley cropping system for
semi-arid conditions, using micro-computers. Agroforestry Systems v. 1, n. 4,
pp. 335-345.

Horvath, Ronald J. 1969. Von Thunen’s isolated state and the area around Addis
Ababa, Ethiopia. Anngp,_Association of American Geographers v. 59, pp. 308-
323.

 

Hosier, Richard H. 1988. The economics of deforestation in Eastern Africa. Economic
My v.64, n.2, pp. 121-136.

Hosier, Richard H. et al. 1982. Energy planning in developing countries: Blunt axe in
a forest of problems? Ambio V. 11, pp. 180-187.

Jayne, Thomas S., John C. Day, & Harold E. Dregne 1989. Technology and
agg'cultural productivity in the Sahel. Agricultural Economic Report No. 612.
Washington, DC: United States Department of Agriculture.

Keita, MN. 1982. Les disponibilites de bois de feu en region Sahelienne de L’Afrigue
Occidentale. Rome: UN FAO.

Kostrowicki, A.S. 1986. Economic and non-econorrric estimation of losses incurred
due to the degradation of the natural environment. Geoggaphia Polonica v. 52,
pp. 5-10.

Krugmann, Hartmut 1988. Improved cookstove programs: Boon or boondoggle?
IDRC (Ottawa) Repprts (Jan.), pp. 21-22.

Leach, G. and R. Meams 1988. Beyond the fuelwood crisis: People, land and trees in
Africa. London: Earthscan Publications.

 

90

Matthews, William H and Tonfiq A. Siddiqi 1981. Energy and environmental issues in
the developing countries. Chapter 3 1n Energy and environment in the
developing countries, ed. Manus Chatterji. New York: John Wiley and Sons.

Meier, Peter and Mohan Munasinghe 1987. Implementing a practical fuelwood
conservation policy. Energy Policy (April), pp. 125-144.

Moran, Warren 1979. Spatial patterns of agriculture on the urban periphery: The

Auckland case. Tijdschrift voor Econ. en Soc. Geoggafie v. 70, n. 3, pp. 165-
76.

Moss, RP. and W.B. Morgan 1981. Fuelwood and rural energyproduction and supply
in Lire humid tropics. Dublin: Tycooly Press.

Munslow, Barry et al. 1988. The fuelwood trap. A study of the SADCC region.
London: Earthscan Publications Ltd.

National Academy of Sciences 1980. Firewood cr0ps. Washington, DC: National
Academy of Sciences.

1983a. Environmental change in the West African Sahel. Washington,
DC: National Academy of Sciences.

 

1983b. Agroforestry in the West African Sahel. Washington, DC:
National Academy of Sciences.

 

Nicholson, Sharon E. 1982. The Sahel: A climatic perspective. Sahel D(82) 187.
Paris: Club du Sahel.

Pearce, David 1988. Natural resource management and anti-desertification policy in
the Sahel-Sudan zone: A case study of gum arabic. Geo-journal v. 17, n. 3, pp.
349-355.

Peet, Richard J. 1969. The spatial expansion of commercial agriculture in the
nineteenth century: A Von Thunen interpretation. Economic Geography v. 45,
n. 4. PP. 283-301.

Pimentel, D. et al. 1986. Deforestation: Interdependency of fuelwood and agriculture.
Qk’ 05 v. 46, n. 3, pp. 404-412.

Poulsen, Gunnar 1978. Man and tree in tropical Africa: Three essays on the role of
trees in the African environment. Ottawa: IDRC - International Development
Research Centre.

Rasmusson, Eugene M. 1982. Global climate change and variability: Effects on
drought and desertification in Africa. In Drought and hunger in Africa, ed.
Michael H. Glantz. Cambridge: Cambridge University Press.

Redclift, Michael 1987. Sustainable development: Exploring the contradictions.
London: Methuen.

Repetto, Robert 1987. Creating incentives for sustainable forest development. Ambio
v. 16, n. 2-3, pp. 94-99.

 

91

Republique du Burkina Faso 1987. Enguete approvisionnement, consommation et prix
d; bois de chauffe dans la ville de Ouagadougou: Fevrier-Mars 1987.

Ouagadougou. Institut National de la Statistique et de la Demographic.

Republique du Mali 1989. Detemflfion des couts moyen de production des
m’ncipaux produits agrjcoles. Bamako: Ministere de l’Agriculture, Institut
d’Economie Rurale.

Ribot, Jesse 1989. Forestry Policy and the Charcoal Industry: A Case Study of
Environmental Decline in Senegal’s Villages. Energy and Resources Group,
University of California/Berkeley (unpublished).

Shaikh, Asif 1985. Forespy development options in the Fifth Region of Mali: The

economic tradeoffs. Washington, DC: Energy/Development International.

1985/86. Initiating forestry reforms in the Sahel: The role of transitional
financing. Rural Africana v. 23/24, (fall 85/winter 86), pp. 49-59.

 

Sinclair, A.R.E. and J. M. Fryxell 1985. The Sahel of Africa: Ecology of a disaster.
Canadian Journal of Zoology V. 63, pp. 987-994.

Sinclair, Robert 1967. Von Thunen and urban sprawl. Annals AAG v. 57, n.1, pp. 72-
87.

Skutch, Margaret McCall 1983. Why people don’ tplant trees: The socio-economic

impacts of existing woodfuel programs- village case studies, Tanzania.
Discussion Paper D-73P, Energy 1n Developing Countries Series. Washington,

DC. Resources for the Future.

Southgate, Douglas D. and Fredrick J. Hitzhusen 1987 . Economic analysis of
renewable resource conservation in the Third World. Chapter 6 in Sustainable
Resource Develo ment in the Third World. eds. Douglas D. Southgate and John
F. Disinger. Boulder, CO: Westview Press.

 

Stevens, Richard and Yuk Lee 1979. A spatial analysis of agricultural intensity in a
Basotho village of Southern Africa. Professional Geographer v. 31, n. 2, pp.
177-183.

Stryker, J. Dirck et al. 1987. Incentive System and Economic Policy Reform in Mali.
Sommerville, MA: Associates in International Resource Development.

Swift, Jeremy 1977. Sahelian pastoralists: Underdevelopment, desertification, and
famine. Annual Review of Anthropology v. 6, pp. 457-478.

Tanticharoen, Morakot and J. Gururaja 1987. Production, development and
management of short-rotation plantation of fuelwood. In Proceedings of the

ESCAPEAOZLJNEP expert ggoup meeting on fuelwood and charcoal. Energy

Resources Development Series No. 24. New York: United Nations.

Thom, Derrick J. and John C. Wells 1987. Farming systems in the Niger inland delta,
Mali. Geographical Review v. 77, n. 3, pp. 328- 342.

92

Thomson, James T. 1981. Public choice analysis of institutional constraints on
firewood production su‘ategies in the West African Sahel. In Public Choice and
Rural Development, eds. C.S. Russell and N.K. Nicholson. Washington, DC:
Resources for the Future.

Tibesar, Arthur and Rodney White 1985. An analysis of household energy use in
Dakar, Senegal. Project Ecoville Working Paper No. 27. Toronto: Institute for
Environmental Studies, University of Toronto.

Timberlake, Lloyd 1985. Africa in crisis: The causes. the cures of environmental
bankruptcy. London: IIED.

UNFAO 1986. Yearbook of forest products. Rome: United Nations Food and
Agriculture Organization.

1981. Map of the fuelwood situation in developing countries. Rome:
United Nations Food and Agriculture Organization.

 

Wardle, Philip and Massimo Palmieri 1981. What does fuelwood really cost?
Unasylva v. 33, n 131, pp. 20-23.

Weinstabel, Peter E. 1982. Forestry as a possible solution to the environmental and
energy difficulties facing semi-arid Africa: The case of Upper Volta. Applied
Geoggphy and Develcgrment v. 20, pp. 70-81.

1984. Agroforestry aspects of the establishment of the green belt around
Ougadougou, Upper Volta. In Social. economic. and institutional aappcts of
agi'oforesg, ed. J.K. Jackson. Tokyo: United Nations University.

 

Whitney, J .B.R., C.M. Dufoumaud, and B.W. Murck 1987. An examination of
alternatives to traditional fuelwood use in the Sudan. Journg of Environmental
Management v. 25, pp. 319-346.

Winterbottom, Robert and Peter T. Hazlewood 1987 . Agroforestry and sustainable
development: Making the connection. Ambio v. 16, n. 2—3, pp. 100-110.

 

World Bank 1985. Desertification in the Sahelian and Sudanian zones of West Africa.
Washington, DC: IBRD.

1988. World development report. Washington, DC: IBRD.

 

1989. Sub-Saharan Africa: From crisis to sustainable glowth.
Washington, DC: IBRD.