Previous Page Table of Contents Next Page

Annex II - Summaries of selected wood fuel surveys

1. National surveys
2. Regional surveys
3. Sectoral surveys
4. Urban surveys
5. Village surveys


Two main problems confront anyone searching for systematic studies of fuel use in rural communities. First, few such studies exist. Second, many are not easily accessible. This Annex contains succinct descriptions of some important fuel surveys. Where the surveys have already been completed, the summary accounts have generally been prepared from published material reporting on theme For ongoing work, the descriptions have been contributed by persons involved in their execution. Another major source of description is a study commissioned by the United States Agency for International Development, "The Socio-Economic Context of fuelwood Use in Small Rural Communities." This 1980 report provides a broad and comprehensive review of the literature, with emphasis on socio-economic and environmental aspects. Separate sections deal with firewood and charcoal, each section considering, on a global basis, such topics as - source, type and availability; access to land; harvest/collection and transport; distribution; consumption; and changes in local fuel systems. Community forestry programmes are examined in another section, and the report ends with "Conclusions" and a series of specific "Issues and Recommended Guidelines." The report includes a 261 item bibliography, and is available at most USAID missions.

1. National surveys

1.1 Surveying wood fuels in Bangladesh
1.2 The Malawi rural energy survey

We include two recent and representative surveys, one from Bangladesh and one from Malawi. Many other developing countries are in the midst of conducting national surveys, or have already done so, India has led in the numbers of surveys at different levels. In the last few years, there has been much activity in Kenya, with the newly created Ministry of Energy co-operating with the Beijer Institute and with the United States Agency for International Development in surveying energy consumption and needs at different levels, including, of course, the national level. As Kenya was the location of the 1981 United Nations Conference on New and Renewable Sources of Energy, it is appropriate that. Kenya should be heavily involved in an impressive range of surveys of energy.

1.1 Surveying wood fuels in Bangladesh

by J.J. Douglas1

1 James J. Douglas was Chief Technical Advisor for FAO/UNDP on Project BGD/78/010 and coordinator of the national survey he describes. He is now with the Australian Bureau of Agricultural Economics.

The survey is being carried out ad a Government of Bangladesh/UNDP/FAO Project, BGD/78/010 -Supply and Demand of Forest Products and Future Development Strategies. This has, as its major information gathering objective, the formulation of estimates of consumption of all forest products in Bangladesh. Some information on the supply side has also been collected, but it is intended that the resource data will eventually be finalized on the basis of information from the sister FAO/UNDP project - Village Forest Inventory (BGD/78/020), and from national forest inventory data developed under a new FAO/UNDP Project - Assistance to the Forestry Sector (BGD/78/017).

Information on the production and use of wood fuel, in both the urban and rural enviroments, has been collected as part of the overall data gathering exercise for forest products.

Wood Fuel Usage and Supply Estimation

Wood fuel is the predominant forest product of Bangladesh, and therefore major attention was given to it in the survey procedures. Two major surveys were done:

(a) Urban Fuel Usage Survey

A special purpose survey was mounted in Dacca, the capital city, to measure the fuel consuming characteristics of the urban population. The city was divided into five major Strata: a sixth, located on the fringe of the metropolis, was included to measure the gradation from urban to rural fuel consuming patterns. About 2,000 households were covered - something in the order of 10,000 people in all. The objective was to measure usage of wood fuels, and all available alternatives, across income strata.

The Project designed the questionnaire, and Subcontracted the Institute of Statistical Research and Training (ISRT) of Dacca University to carry out the pre-testing, enumerator training, sample design and administration of the survey in the field. Householders were asked to estimate usage of various fuels. Also, where possible, enumerators measured fuel stocks on hand. In the case of natural gas usage, householders were obviously unable to estimate volume of usage - in that case, a Simple recording of the fact that gas was supplied to the household was made.

The survey was applied at a single point in time: no follow-up was judged necessary, given the results, which indicated that, where gas is available (as it already is in most of Dacca, and soon will be in the other major city, Chittagong) it is used to almost the total exclusion of all other fuels. Some 80 enumerators were in the field for a month or so gathering the data. Results were processed manually by IRST. The basic results from the survey have been listed and discussed in the Field Document No. 2 from the project (Douglas, 1981a). They have been used as part of the data for forecasting future wood usage in Bangladesh and for designing sector strategies thereupon. They have also been lodged with the Bangladesh Bureau of Statistics for inclusion in statistical bulletins.

(b) Rural Wood Consumption Survey

Prior to this project little was known about consumption and supply of wood products in rural Bangladesh. Some previous studies of wood fuel use have been made but these were either confined to small regions, or judged by the project to be highly inaccurate.

Accordingly, a major survey of consumption, (with some supply side estimation included, as a temporary data base pending results from the Village Forest Inventory Project) was mounted for the country as a whole. Details of the design of this survey are given in the Field Document No. 1 (Asrat, 1980). Basic frame ideas from FAO's food Survey methodology (see Clayton, 1978) were used. The basic framework of the survey was a stratified, self weighted sample of about 6,000 rural households, selected from 430 villages in 47 Thanas (districts) of the country. Six major zonal stratification were used, on the basis of ecological or geographical factors, and results were recorded within defined landholding (income) strata.

Considerable pre-testing of the questionnaire was done, and questions on known factors were included, to allow testing of the survey design for both sample and non sampling error. Standard error calculations for all major data items were done. These matters are discussed at length in Field Document No. 2. Questions and measurement on all aspects of wood usage in the rural area were included. Extremely detailed measurement of wood in place in house buildings, furniture, implements, etc., were necessary to formulate estimates, but it was impossible to carry out this measurement for all 6,000 households. A 20% sub-sample was therefore drawn for this purpose which duplicated all the questions from the main survey (for testing of the sub-sample), and included a detailed measurement schedule.

So far as tree fuel is concerned, questions and measurement procedures were applied to ascertain householders' usage of fuel over the last day, the last week, and last twelve months - an attempt to pick up the seasonal effect. Supplementary questions asking whether the householder believed he used more, the same, or less fuelwood during the wet season were asked. An attempt was made, by questioning and enumerator observation and measurement, to break down tree fuel usage into categories of stem or main branch volume, and smaller wood, leaf and twig volume - so as to gain some idea of how much fuel came from fellings, as opposed to litter from standing trees. (However, see item (vi) under final remarks below.) Since tree fuel is only one of many used in Bangladesh, questions and measurement of use of other traditional (and commercial) fuels were also included.

An intensive, long-term study of energy use in a single village in Bangladesh by Briscoe (Briscoe, 1979) proved particularly useful for interpreting the national data. The first reason for this is that Briscoe's fuel usage measurement data provided an alternative estimate, with which the appropriate regional data from the national study could be compared. Secondly, because of the very concentrated nature of Briscoe's work, he was able to provide an extensive analysis of the socio-economic implications of the fuel situation, as it is developing in the village. Since Briscoe's findings on impending shortage of fuel for his study village were in fact found in the national study to be a widespread problem, it was possible to generalize parts of his analysis to the national, policy level, on the basis of the data obtained.

Some further validation of all overall survey results was possible using figures provided by Islam (Islam, 1978). Islam measured fuel consumption in a small region of Barisal, in Southern Bangladesh, comprising 23 villages. Islam's study was basically designed to investigate the feasibility of the bio-gas alternative in Southern area, but his very extensive fuel usage data, collected by direct measurement and questionnaire, provide a useful benchmark for comparison with appropriate regional subset from the national data. When this was done, a remarkable similarity in overall energy usage (on a per capita basis), and in the wood fuel group in particular within this, arose between the two sets of data. Less similarity was apparent when the figures were further disaggregated into "fuelwood" and "other tree fuel" components, emphasizing the difficulty of accurately drawing this distinction in the field.

Because of the detail with which Islam reported his individual village results, it was possible to test his results for internal consistency, using regression analysis. This was done, and the results as given in the Project study confirm the reliability of the figures.

Detailed questions and measurement of household tree holdings, and fellings over the previous 12 month period, were included in the survey process, to gain some perspective on standing volume and rate of felling in the village homestead forest resource.

It had been intended to carry out seasonal follow-up to the consumption estimates, but unfortunately the organization sub-contracted to do this was unable to carry out the task. However, it was possible, by using householders opinion response on seasonal variation, combined with some regional studies (for which equivalent figures from the main survey could be extracted), to make a reasonable assessment of the matter.

The survey questionnaire was designed in the Project. The sample design and fieldwork were carried cut by the sub-contractor (ISRT), with considerable input from the projects. Some 120 field enumerators and supervisors were employed, for approximately three months. A further 20 or so laboratory staff were employed. The results were entered via cards onto magnetic tape, and processed on the IBM 370 computer at the Bangladesh University of Engineering and Technology.

Major results from the survey are given in Field Document No. 2, and forecasts based on them in Field Document No. 4 (Byron, 1981). The results will be a major input into forestry planning and strategy design in Bangladesh, and have formed an important part of the data base for a large scale ADB farm forestry-project currently under consideration in Bangladesh. Results have also been conveyed for use by the Bureau of Statistics, and some of the methods and survey sampling frame will be used for regular updating information.

The cost of the rural and urban surveys discussed above combined war about US $ 50 000

Final Remarks

Some lessons were learnt in the process of the exercise discussed above.

(i) The results show that a previous attempt to measure tree fuel use in Bangladesh by supply calculations (aerial photo interpretation, combined with field measurement of some trees) was highly misleading. Such methods are not a viable substitute for direct consumption estimation. (See Montreal Engineering et al, 1976).

(ii) There is considerable regional and socio-economic variation in fuel consumption patterns, even in such a small and geographically homogeneous country as Bangladesh. For forecasting purposes in particular, stratification of the sample to measure such variation is necessary.

(iii) For household use, which is the predominant form of energy consumption, wood fuel and other traditional fuels are mutually substitutable to a wide extent. It is pointless, therefore, in a country such as Bangladesh, to assess -usage of tree fuel in isolation: the substitutes must also be included. However, this creates an analytical difficulty: the reason for doing such surveys in the first place is, usually, to assess whether energy consumption is in danger of outstripping supplies. But it is extremely difficult - perhaps impossible - to measure the stock of all traditional fuels, with any accuracy. It is difficult enough to do so for trees alone in Bangladesh. The Project has drawn conclusions on the basis of the assumption that the use of trees, the one fuel source not produced annually, as are the residues, can be used as indicator: in other words, over-use of trees to some extent indicates a shortage of all fuels. This is a matter that should be given some analytical attention in general, in cases such as Bangladesh where wood fuel is an ingredient in a complex mixture of traditional energy supply.

(iv) In general, the survey exercise demonstrated the importance of proper pre-testing, or pilot surveys: in the case of rural survey discussed above, a number of questions included in the original questionnaire design had to be discarded, or changed radically: in their original form they were answered by respondents extremely poorly, and were clearly not understood.

(v) It is essential, where nationwide, or large scale regional surveys are intended, to include information items which are already known, so that the survey procedure can be tested against a datum. In the case of Bangladesh, good prior data on the distribution of landholding (in itself good proxy for rural income) was available. Questions on this were therefore included in the survey, and the resulting distribution statistically compared to existing census data on this matter. It was possible, in this way, to check the extent of sampling bias present in the survey. This is important, because in areas where field operational difficulties can be severe, the probability of significant bias in results is always high.

(vi) It proved impossible, under field conditions, to obtain a useful breakdown of tree fuel into its felled (stem/main branch) and standing (leaves, twigs, etc.) components: assessment was done at the point of usage, where the distinction is difficult to make. Under the circumstances, the best that could be done to obtain an aggregate division was to apply some published empirical biomas results for tropical tree species to the overall figures. To make a more certain estimate of the breakdown of tree fuel would necessitate observation and measurement over time of fuel gathering in the rural area.

1.2 The Malawi rural energy survey

by David French1

1 David French is Senior Energy Adviser in the Energy Unit in the Ministry of Forestry and Natural Resources, Malawi.

The Energy Unit of Malawi's Ministry of Forestry and Natural Resources has carried out a rural energy survey as part of a nationwide sample survey of agriculture. Survey results are principally being used to assist in the planning of the Wood Energy Project, which is establishing government firewood plantations and supporting reforestation activities at farm and community level.

The Questionnaire

The energy survey questionnaire asked which jobs are done that require energy (e.g. cooking, heating water for washing, providing light, making beer or bricks, smoking fish), and what fuels are used to do these jobs. In addition, several questions dealt with firewood collection (perceived difficulty, distance to wood source, whether wood is collected or purchased). Other questions covered planting of-trees, the number of meals cooked each day, and types of stoves used. Finally, to provide basic data about other major tasks that consume energy within the household, questions were included on ways in which flour is prepared and water collected.

In addition to illustrating basic patterns of energy use, the survey questions were intended to show the consequences of wood scarcity. The extent of scarcity in a given area was judged by the distance to the fuelwood supply and by the degree of difficulty experienced by the household in collecting wood. In the tabulations, these measures of scarcity were related to such variables as the use of crop residues or charcoal for cooking, the number of meals cooked each day, purchases of wood, the use of commercially valuable trees (mango) for fuelwood, and planting of trees.


The energy survey was added to a package of surveys comprising the National Sample Survey of Agriculture 1980/81 (NSSA), a periodic survey effort about the farming and household activities of smallholders, including household composition, income and expenditures, garden cultivation practices, nutrition, and related concerns. To carry out these surveys, 344 enumerators were stationed throughout the country for a one-year period.

For the energy survey, each enumerator interviewed seven sample households, for a national total of 2,408 interviews. Since the survey was primarily concerned with domestic jobs, the questions were asked of the central woman in each household. In order to find out how energy use varies from season to season, the survey was administered both in March-April 1981 (at the end of the rains) and in September-October 1981 (at the end of the dry season).

While NSSA personnel distributed, administered, and collected the survey schedules, the Energy Unit itself assumed the rest of the survey's organizational responsibilities and costs. The Unit's staff drafted and tested the survey, conducted training sessions for enumerators, checked the completed schedules, arranged for computer processing of the data, held workshops throughout the country to discuss results with NSSA field staff, and wrote a final report on survey findings.

By joining an established survey process, the Energy Unit was able to carry out the survey at minimal cost. Total direct costs were only about U.S. $9 000, half for data processing and half for survey preparation, training courses, and post-survey workshops. Excluded from this figure is the cost of time spent on the survey by the Unit's Senior Energy Officer (about six weeks) and the Energy Survey Officer (about six months). Also excluded are minor costs incurred by NSSA in absorbing the energy survey into their existing work programme.


According to the survey, firewood is the dominant energy source for most fuel-consuming jobs done by the rural population, including cooking, keeping warm, heating water for washing, smoking food, and brewing beer. The only non-wood fuels used in significant amounts in rural areas are paraffin for lighting and diesel for commercial maize milling. Secondary fuels such as crop residues, charcoal, and dung are used very little.

Firewood collection is perceived as an increasingly difficult task, and many people are walking lengthy distances for wood. The problem is especially acute in the central and southern regions and lakeshore districts.

So far, however, scarcity of wood has had little measurable impact on rural life. Jobs requiring fuelwood (cooking, brewing beer, heating water) are done as commonly in wood scarcity areas, as in areas where wood is more abundant. In scarcity areas, people are somewhat more likely to purchase some of their wood or to use inferior wood and crop residues for cooking; but these tendencies are not yet pronounced. Although a significant number of farmers plant trees, these are primarily for building poles and fruit rather than for firewood.

The Wood Energy Project has been considering the implications of these findings for the location of tree nurseries, species to be grown, strategies for marketing seedlings, and so on. To pursue such questions further, the Energy Unit will use its own enumeration team to carry out detailed studies of specific issues. For example, one follow-up survey deals with farmers' motivations, preferences, and problems with respect to tree planting.


Copies of the energy survey final report and related reports can be obtained from the Energy Unit, Ministry of Forestry and Natural Resources, Lilongwe 3, Malawi.

2. Regional surveys

2.1 The rural energy system project, Nigeria

We include just one example of a regional survey, the comprehensive and carefully thought out recent exercise in southwestern Nigeria. There is certainly a need for such surveys, which are intermediate between the costly and time-consuming national surveys, and the locally specific village studies. All levels have their uses and it in a matter of balancing conflicting needs and limited budgets. But, because of regional differences, which are often very marked, it often makes sense to invest at least part of the funds available in a regional energy surveys.

2.1 The rural energy system project, Nigeria

by W.B. Morgan

The survey was jointly sponsored by the University of Ife and the United Nations University and was directed by Professor Go J. Afolabi Ojo of the University of Ife. The project is located in southwestern Nigeria and concentrates on three urban areas and their rural hinterlands: Ibadan (estimated urban population 1,500,000), Ogbomosho (estimated urban population 300,000) and Ile-Ife (estimated population 150,000). Two of these locations are rainforest and one (Ogbomosho) is in relatively moist savanna with a markedly different range of tree species for fuelwood. In scale it is regional, covering an area with a total population of approximately 5 million. Survey work began in January 1979 and continued until September 1980. A number of related studies have either been completed since or are still continuing. It is hoped that energy study, including rural energy systems and the problems of urban energy supply, will become a permanent and major aspect of research at the University of Ife, so that the project will have become the nucleus of a major academic development. Computer analysis of the data has been completed and the reports of the organizers of each programme are currently being written. A discussion workshop with government energy and forestry staff is planned.

The five programmes of the project included surveys of various forms of energy consumption and distribution together with rural energy production. They consisted of:

1. a study of socio-economic factors affecting rural energy preferences;

2. a study of urban market Influences on rural energy supply, including a survey of dealers and of households, industries, offices, shops and institutions;

3. a land use mapping survey based largely on aerial photography;

4. a study of the ecology of fuelwood production on farms;

5. a group of technology studies, including an attempt to design and test fuelwood stoves, the determination of the specific gravity and caloric values of fuelwood samples, and some preliminary work on appropriate energy devices.

Of a total of probably just under a million households in the region 11,300 were surveyed and questionnaires for 10,800 were successfully completed. 7,437 of these were rural. Additional small surveys were made of transport, dealers, the role of women and children and local wood stove manufacture. Sampling techniques were varied. Rural surveys were based on geographical sectors for the selection of villages whilst an attempt at socio-economic stratification was made in the towns. Only a limited attempt was made to measure quantities of wood. Much more attention was devoted to expenditures and preferences. The survey occupied just over a year and a half, but most places were visited only for a few days. A distinctive feature of the survey was that all 22 members of academic staff and some 30 enumerators (at any one time) belonged to the region. The survey had the advantage also of a considerable existing fund of knowledge of local society and the economy. Preliminary results indicate widespread use of fuelwood and kerosene in the rural areas and of kerosene, electricity and gas with some wood, particularly as a stand-by, in the urban areas. Charcoal is used mainly for a few special purposes, but, wood is used extensively in informal industries. Wood cutting is concentrated close to the towns and in certain village areas near main roads and there is marked species selection. There are indications of overcutting and of some pressure on local resources despite the apparent resource riches of the rainforest. Two preliminary collections of papers are available from Ife workshops held in 1979 and 1980 (Morgan et al, 1980; Ojo et al, 1981). Publication of the final report for the questionnaire survey will eventually follow.

3. Sectoral surveys

3.1 Wood fuel utilization by small-scale industry in Nepal
3.2 Charcoal in a Guatemalan community

Our two examples here include small-scale industry and charcoal, and indicate the value of concentrating on specific sectors. Depending on the nature of energy use, almost all countries could profitably conduct surveys on specific sectors of economic and energy significance to the particular country.

3.1 Wood fuel utilization by small-scale industry in Nepal

by D. G. Donovan1

1 Deanna G, Donovan carried out this survey while she was a Forest and Man Fellow of the Institute of Current World Affairs (ICWA), Hanover, New Hampshire. Concurrently, she was affiliated as a Research Scholar with the Research Centre for Applied Science and Technology (RECAST) of Tribhuvan University, Kirtipur Campus, Kathmandu. During the final four months of the project, expenses were covered by the Agricultural Development Council (ADC), New York.

Nearly 87 percent of total energy consumed in Nepal is derived from wood. Although the majority of fuelwood goes to domestic users, many cottage and small-scale industries also depend heavily on wood fuel. Indeed, with the progressive price increases and supply irregularities of conventional fossil fuels, it is not uncommon to find fledgling industries turning to often cheaper and more readily available traditional fuels such as wood. In order to highlight this widespread dependence on forest resources by much of Nepal's small-scale industry, an investigation of wood fuel use by manufacturing, agricultural processing and service sectors was initiated in 1979.

Preliminary investigations confirmed that, to date, very little information has been systematically recorded regarding energy use by the various cottage and small-scale industries in Nepal. Traditionally, industrial feasibility studies have focused on requirements for raw materials, labour and capital; energy needs have been only briefly discussed, seldom seen as a constraint. Discussions with factory owners prove the opposite. The purpose of this research project, therefore, was to develop an information base on the energy requirements of small-scale industries, especially wood-fueled industries, in order to assist policymakers and planners in designing both environmentally and economically sound development programmes for rural areas.

The major research objectives were to identify those industries using wood fuel, to note the type of energy technology employed and to determine the amount of wood required annually and per unit of finished product. Secondarily, the project sought to assess the economic and environmental consequences of wood fuel use, especially on household consumption patterns. Finally, a brief survey of the energy options available to small-scale industry in rural Nepal and of the feasibility of introducing alternative energy technologies was carried out.

For the purposes of this study, industry was defined as any enterprise engaged in the production of goods and services for sale or exchange in the market place. The scale of production varied widely within as well as between industries. Due to the scarcity of information regarding industrial location, the originally planned nationwide survey of a few selected industries was redesigned to a broader industrial scope, but a narrower geographic focus. Industries included in the investigation were those known to have used wood fuel in the recent past. In the final count, some 35 industries were identified as presently using firewood, charcoal or sawdust at some if not all production sites.

Given the distinct difference in socio-economic and ecological situations in the terai (Gangetic plain) and the hills in Nepal, Plus the disastrous ecological consequences of deforestation in the hill and mountain region, emphasis was placed on those industries operating in the hill districts. Additional points considered in the selection of suitable study areas were:

1. distribution and type of industry;
2. existence of previous socio-economic studies;
3. bioclimatic zone;
4. political development region designation; and
5. physical accessibility.

The seven areas ultimately surveyed - Surkhet valley, Mustang district, Pokhara valley, Trisuli valley, Hetauda valley, Ilam and Phikkal Panchayats in Ilam district and Kathmandu valley - present a cross-section of physical and cultural conditions in the hills of Nepal.

A combination of information-gathering techniques, including sample surveys, structured and unstructured interviews, direct observation and measurement, were employed in the data collection process. Among the information solicited by questionnaire were production levels, types of fuels used and the amount species, form, cost and source of wood fuel. Following the formal interview with a factory owner or manager, a visit was made to the actual production site where the various stages in the production process were recorded. Sketches and photographs were made as information was taken on the design as well as the construction materials of the wood-fired kiln or stove. The condition of the fuel was noted and a sample charge weighed when possible. Not only was the amount of wood used of interest, but also the method of handling and firing especially as related to the efficiency of energy conversion. Additional information regarding household fuelwood consumption and local economic and environmental conditions were secured through sample surveys and interviews with key informants, such as community leaders and forest department officers.

The interviews and production-site surveys were conducted by the author's Nepalese research assistant together with the author. An attempt was made to visit many different factories in order to gain an understanding of the diversity of technology employed throughout the country. Logistical constraints precluded repeated visits to all but a few factories. Support services, such as transport, typing, porterage, translation and graphic arts, were secured on contract as needed. Numerical data were processed with simple desk calculators.

The project final report, completed in November 1981, will review the rationale and research methods of the project as well as an analysis of the research findings with recommendations for future research and action programmes. Appendices to the main document will present brief profiles of various rural industries and discuss in greater depth economic and environmental conditions in selected study areas. Copies of this report will be submitted to the three sponsor agencies, Institute of Current World Affairs, Research Centre for Applied Science and Technology and the Agricultural Development Council. Formal publication and distribution are being considered. Interested parties should contact the following for copies:

(1) Peter Bird Martin
Executive Director
Institute of Current World Affairs
4 West Wheelock Street
Hanover, New Hampshire 03755, USA

(2) Theodore Smith, President
The Agricultural Development Council, Inc.
1290 Avenue of the Americas
New York, New York 10019, USA

(3) Kedar Lal Shreshta
Executive Director
Research Center for Applied Science and Technology
Tribhuvan University
Kirtipur Campus
Kathmandu, Nepal

3.2 Charcoal in a Guatemalan community

summarized from: "Charcoal: its Multifarious Effects in a Guatemalan Community," John G. Hehr, unpublished M.A. Thesis, Western Michigan University. Kalamazoo, 1967

The author spent three months in 1965 doing fieldwork in the highlands of Guatemala. The study concentrates on the charcoal industry of Cajolá, a small rural municipio (population 5000) located approximately 17 kilometres northwest from Quetzaltenango. Hehr examines the charcoal industry in relationship to rural employment problems and also in terms of its environmental impact. What emerges from his excellent description and analysis are bleak pictures of rural livelihood, impending fuel shortages, and environmental deterioration. Hehr's study is largely descriptive. Besides participant observation (e.g. accompanying charcoal vendors on their door-to-door sales in the city), he interviewed local charcoal makers, district forestry officers, charcoal vendors, municipal officials, a bus driver, a farmer who lives on the coast and the owner of a charcoal depository (15 names of persons interviewed are listed in an appendix). Although community economic (especially agricultural and wage labour) and fuel-related activities are well described, little information is given on the patterns of socio-economic stratification within this primarily Indian community.

The study in divided into five sections. The first places the community within its regional, environmental, socio-economic and historical context. Part two reviews historical aspects of charcoal production in Cajolá, and relates the incentives that exist for peasants to engage in commercial charcoal makings Especially highlighted is the lack of employment alternatives attributed to Guatemala's rigid class and ethnic stratification as well as local agro-environmental limitations, that leads poor villagers to cut down the few remaining trees for Quetzaltenango's expanding fuel market. Hehr also points out the competition for trees among different communities.

The third section describes the process of charcoal making, from the finding and purchasing of a tree to the firing process. The next section relates the cycle of production and price variation, and describes the spatial commercial movement of charcoal to consumers in Quetzaltenango and on the Guatemalan coast. Hehr's study concludes with an analysis of deforestation and its socio-economic implications.

Besides his description and analysis, the chief value of Hehr's study is its demonstration of how much important data can be gathered by a single field worker in a relatively short time period, using relatively few resources. Unfortunately, no budget is given, but it is obvious from his account that Hehr's intense, local-level focus required few resources.

The type of study presented by Hehr can be particularly important as a pilot study for a more comprehensive large-scale survey, or used for case studies that would supplement macro-surveys.

4. Urban surveys

4.1 Urban use of energy in Papua New Guinea.

One study from Port Moresby in Papua New Guinea is included. Towns and cities have been relatively neglected in wood fuel studies, there having been more emphasis on rural areas. Ideally, what is needed is not one or the other, but one study which examines the complex links between rural and urban areas, regarding them as one single social and economic and energy field. The wood fuels that are consumed by urban residents come either from peri-urban areas, or, in greater quantities, from rural areas, sometimes as far as 250 km away. Therefore, it is important to ask what are the effects on the rural areas of all this export of wood fuels to the towns, and this has seldom been critically examined.

4.1 Urban use of energy in Papua New Guinea.

summarized from "Household Energy Consumption in Port Moresby", by Matthew S. Gamser, Dept Minerals & Energy (Report 8.80), Papua New Guinea, Port Moresby, 1980.

The study surveyed over 1800 low cost, domestic quarters, urban village, and squatter settlement households that comprise 79 percent of the city's dwellings. The study was done in July/August, 1980, with the help of 57 student surveyors from the University of PNG.

97.8 percent of the homes surveyed used firewood or kerosene for cooking and lighting.

Electricity is rarely used for purposes other than lighting because of its high cost.

Almost everyone surveyed paid to obtain firewood, either to buy bundles from sellers or to travel great distances to gather it. Few areas in Port Moresby possess adequate firewood supplies within walking distance from their households.

The average price paid for kerosene in the survey was 39.7 Toea per litre, over 13 Toea, greater than the price controlled price for this fuel at the time of the survey.

Kerosene purchased in small containers from trade stores can cost 4 to 5 times as much as the controlled price charged at petrol stations.

Rapidly decreasing firewood supplies make this fuel more difficult and expensive to obtain each day, and many people are switching to cooking with imported kerosene.

Firewood and charcoal produced from sawmill wastes and sold at community distribution centres can undercut the present energy cost of firewood by 19 to 60 percent and the energy cost of kerosene by over 50 percent.

Port Moresby households pay far too much for domestic fuels and rely far too heavily on imported kerosene. The development of firewood, charcoal and other indigenous energy resources is essential for the city's future, and that of other areas in Papua New Guinea.

5. Village surveys

5.1 Energy use in a Bangladesh village
5.2 Fuelwood in a Nepalese hill village
5.3 Wood fuels in a small Sudanese town
5.4 Wood fuels in a marginal area of Kenya
5.5 Fuelwood in a Tanzanian village

We have here a representative sample of sound village studies from Bangladesh, Nepal, the Sudan, Kenya and Tanzania, all concentrating on wood fuels, but all with somewhat different approaches and methods and time periods involved. It is important, in doing a village study, to try to indicate how representative the selected village is, and what are the implications for the national or at least the regional energy picture. These are but a small sample of the hundreds of village studies that have been done in recent years, with many more currently going on.

5.1 Energy use in a Bangladesh village

summarized from "The Political Economy of Energy Use in Rural Bangladesh", by Jo Briscoe. (mimeo) Environmental Systems Program. Cambridge, Massachusetts: Harvard University, 1979, and "Energy Use and Social Structure in a Bangladesh Village" by J. Briscoe, Population and Development Review. Vol, 5, No. 4, 1979.

The Population and Development Review article is a briefer version of the study described in the August, 1979, unpublished paper. The study field work was done in the village of Ulipur, Bangladesh. The village is located in the western part of the Comilla district and in a deep-water flooding plain of the Meghna River. The study population consisted of eight Hindu fishing families and forty Muslim agricultural families (330 people, 48 cattle and 43 acres of agricultural land).

The study aimed at developing a detailed understanding of the natural resources availability and patterns of use with particular focus on energy resources. A major objective was to examine the issues of distribution and control of available resources and the structure of social organizations which govern the distribution of resources from owner to user.

It was an independent investigation, although for part of the period, Briscoe had an appointment with an international institution. The investigator incorporated all energy sources, both animate and inanimate; the latter was dominated by crop residues but also included firewood and other traditional fuels (bamboo, etc.) and very small amounts of conventional fuels.

Both supply and consumption patterns were studied. Field methodology utilized participant observation. Briscoe, who is fluent in Bengali, an important factor in his research, lived in the village for the better part of the year. Survey information was collected at different intervals: initially, a census to gather demographic and economic information on study population including energy-producing resources - land animals and land utilization, crop production and residue production were all estimated by the farmer. Information gathered fortnightly included weighing of overnight dung excretion, type and source of food consumed by animals, human and animal labour activities, crop and residue production (including weighing of residue before and after drying) and accounting for fuels obtained or put in storage). A quarterly close observation of family cooking, including weighing of fuels to be used and an estimate of quantities of food cooked and the number of people eating, was made.

The data collected over the period of about one year were utilized to study in detail the energy resource flows in this society. The analysis indicated the complex and frugal but inefficient nature of the system. The analysis indicated the importance the social structure has on energy use since villagers of different classes meet their energy (and other) needs in different fashions, given their access to resources. The mechanisms are changing, much to the detriment of the poorer people as traditional forms of social organizations characterized by client/patron relations give way to more market-oriented structures. The analysis indicates the impacts of the introduction of high-yielding varieties and income for non-agricultural sources and shows clearly how the impacts differ dramatically between classes.

Briscoe's study describes and analyses several facets of differential access to fuel. In the village he studied, he discovered that 16 percent of the population owned 80 percent of the trees, 55 percent of the cropped land, and 45 percent of the cattle. Moreover, traditional practices which had allowed the poor and landless access to crop residues and other fuel resources were being discontinued. Poor people were forced increasingly to purchase fuel, or else resorted to theft.

Briscoe's study is particularly important because he considers land tenure change in historical perspective. He related the village's energy situation to changes stemming from the Green Revolution which led to increased concentration of land ownership, a decline in the number of available agricultural jobs and in the prevailing farm labourer wage rate, and in the replacement of customary harvest and gleaning rights by commercialized relationships. Briscoe's analysis demonstrates the necessity of considering local fuelwood situations in terms of how different social and economic groups and classes have been affected by changes in both the local and wider society.

The study is a seminal one in its examination of the complex interrelations of the traditional fuel systems, demonstrating dramatically the importance of more micro level analysis of the different (by class) effects which may be hidden in village aggregate analysis. Its implications for energy policy point to the need to understand these relations.

5.2 Fuelwood in a Nepalese hill village

summarized from fuelwood and Food Needs Versus Deforestation: An Energy Study on a Hill Village Panchayat in Eastern Nepal" by D. Bajracharya, in Energy Analysis in Rural Regions: Studies in Indonesia, Nepal and the Philippines. Part III. Honolulu, Hawaii: East-West Center, 1980.

The study focused on Pangma village panchayat1 in the Sankhuwa Sabha district of Nepal. This area lies in the hill region of the Kosi zone of eastern Nepal.

1 A village panchayat in the smallest political unit in Nepal, an distinguished from a district panchayat, a larger unit.

The objective of the research was to examine at the micro scale the multiple fuel, food and fodder pressures on land resources that can lead to deforestation. Consequently, the investigation looked at both traditional fuel needs and supply potential (primarily fuelwood) as well as other food and fodder needs that draw from the same resource base. Land use was estimated from existing aerial panchromatic photographs; tree counts were utilized to establish forest inventory and total forest productivity (large stem and small twigs, etc.), fuelwood needs were estimated from household interviews of 181 (31%) of the 582 households in the study population Bajracharya conducted most of the interviews (with the assistance of three locally hired enumerators). Information was collected on the followings amount and type of fuelwood, time and place of collection and type of combustion (chula or stove versus agena or open hearth). Certain households were selected to cover variations in ethnic, altitude, settlement cluster and economic status. The interviews also gathered land tenure and use data, crop production data and food consumption data.

Bajracharya's analysis examines the fuelwood versus deforestation question by comparing the needs versus supply potential, taking into account the important differences in availability and access to resources for different classes/groups. The importance of location (altitude) and concomitant access to agricultural and forest lands of different productivity is shown to be important as is the need to simultaneously examine the food, fodder and fuelwood needs which all exert pressure on the land resources. He drawn implications for rural energy and technology policy arguing against simple solutions posed without real understanding of the physical, social and economic structures and against the frequent over-estimated focus on fuelwood demands as the source of deforestation. He suggests that the problem cannot be divorced from the need for better agricultural productivity and the process of innovation and learning that can foster development - a more holistic approach to the problem is necessary.

5.3 Wood fuels in a small Sudanese town

summarized from "Wood for Fuel - Energy Crisis Implying Desertification: The Case of Bara, The Sudan" by Turi Hammer Digernes. Thesis, Polit Geography Department, University of Bergen, November 1977. (Also Addendum, October, 1978).

The author spent six months (August 1976 to February 1977) doing her fieldwork in the Sudan. The location was a small town, Bara (population 10,750) in Kordofan Province, about 400 km southwest of Khartoum. Digernes set out to study the energy system, and concentrated on use of three major fuels, fuelwood, charcoal, and kerosene.

"During the interviews we talked about (the three fuels), how and where they were obtained, what kinds of trees are used, and where these trees are out, amounts used, prices - seasonal fluctuations and any recent changes". In addition to inquires made in Bara, visits were made to the ten main supply villages of fuel to Bara; these were determined according to information obtained from households and from the Forest Office. At the villages extensive interviews were carried out with sheikhs, wood-cutters and charcoal burners; "sellers of wood and charcoal met on the road were stopped occasionally and asked about their products and supply area".

Household interviews were held at 137 households using the following method of selection. The 1973 local rates list contained 1,620 households which were stratified according to amount of tax paid. A 10 percent random sample was drawn from each group, giving 162 households of which 25 had moved or dissolved since 1973, leaving 137 households which were visited of these,

- 86 were in Group I paying annual rates less than £1
- 29 were in Group II paying annual rates £1 to £7
- 22 were in Group III paying annual rates over £7

This is a simple and effective way of obtaining a stratified sample, so that comparisons can be made between poor, average, and rich households. The rates are fixed according to location and area of house, number of rooms, kitchens and cabinets, building materials, and numbers and economic status of household members. Each selected household was visited, with apparently no refusals, by Digernes and her interpreters (one male, one female). By the last two months of the study, the author had enough Arabic vocabulary to work on her own.

The most important chapter for present readers is likely to be Chapter 4, "Energy Situations in Bara", which provides extensive detail about the three fuels - where and how they are obtained, at what prices, who the suppliers are, what trees are used and major changes over last ten years. Much of this information is conveniently presented in tabular form (Table 10-98), including useful comparisons (e.g. on money spent on charcoal per household member, according to socio-economic group). The basic questions that were asked in interviews are included in an appendix.

Attention is paid to alternative energy sources, including electricity, biogas, windmills, and solar energy.

The report - as one would expect in a geography thesis - includes clear and comprehensive information on the physical environment. There are also good maps and photographs.

A short (15pp) "Addendum" presents the results of a brief follow-lip study made in mid-1978, when 32 of the original 137 households were re-visited, and changes noted, including "the rapidly diminishing supply of kerosene", and an increased awareness among women of the problem of desertification. Extra information is provided on supplies, prices, consumption and commercial use of fuels by schools, prisons, restaurants, bakers, etc.

5.4 Wood fuels in a marginal area of Kenya

summarized from "Forests, Foraging, Fences and Fuel in a Marginal Area of Kenya" by David Brokensha and Bernard Riley. (Unpublished paper, prepared for a United States Agency of International Development, Africa Bureau, Firewood Workshop, Washington, D.C. in June 1978).

The authors are, respectively, a social anthropologist and a geographer who have been collaborating since 1970 on an ongoing study of social and ecological change in Mbere Division, Embu District, Kenya. This paper is a village level study which examines the role of fuelwood and charcoal in the lives of the people of Mbere, with emphasis on recent changes.

Particular significance is attached to the local peoples perceptions of wood fuels, of their system of classification of tree fuel trees. Trees are classified according to overlapping criteria including size, inherent qualities (such as slow-burning, easy to out, smokeless), accessibility and seasonal availability and particular uses. The uses include cooking, brewing beer, log fires, smelting iron, night torches, fire making, roasting meat, simmering milk and the like. An important separate category is charcoal, production. For each use, local people have a clear ranking of preferred species, but because of shortage of many useful species, they are often forced to use third or fourth preferences. For fuelwood, accessibility is much the dominant consideration.

Much detail is given of the production, distribution and consumption of fuelwood and of charcoal, examining men's and women's roles, and looking at time and distance involved in collecting fuelwood. The average was 2 ½ trips per household per week, of 2 ½ hours each, with considerable variation.

"Fences" a major factor in fuelwood collection, referring to the granting of individual titles to land and the consequent restrictions on gathering fuelwood rapidly changed what had been a free good in most circumstances to a commercial commodity, with severe hardships for people such as elderly women living alone, fuelwood is now sold at all the markets.

Much charcoal is produced, mainly by poor people as a substitute cash crop, and is intended for sale in the towns. The demand for charcoal has accelerated the disappearance of valuable hardwood.

This paper stresses the social aspects of wood fuels, with many references to species of trees. The authors based it on their fieldwork carried cut 1970/71, 1974, and 1976/77, with use made of a survey conducted by 100 secondary school students in 1976. These young men and women questioned people at their homesteads about present and past perceptions and uses of fuelwood and charcoal.

5.5 Fuelwood in a Tanzanian village

summarized from "Fuelwood Use in a Peasant Community: Tanzanian case study" by Patrick C. Fleuret and Anne K. Fleuret, Journal of Developing Areas 123 (1978): 315-322.

The authors, both social anthropologists, did field-work in the Usambara Mountains, Lushoto District, in Tanzania in 1975/1977, and this article is part of the result of that research. They focus on one village, Kwemzitu population 173). In contrast to most other African fuelwood studies, this is an area of generally good rainfall, probably a little under 1,000 mm, with an elevation of 1,700 metres.

Farmland in Kwemzitu border a forest reserve, and the inhabitants are allowed to collect fallen trees and branches, and the pieces left over by licensed timbercutters. "It, takes about an hour to walk to the principal fuelwood gathering area, so Kwemzitu people have relatively free access to the wood."

Consumption patterns of fuelwood are outlined, beginning with the fire made at first light to cook breakfast, and continuing until 11 p.m. "if people stay late visiting or if the night is cool".

Data on households and total population are presented. While the former are accurate (33 households in three hamlets), the authors "suspect that population figures are inflated by the inclusion of people socially, but not physically, present". This is an important point, which should be taken into account in all village level surveys. The information provided indicated that the average household size (adjusted for household members not actually in residence) was 5.2 persons.

Fuelwood was measured in four households, two of five members, and two of three members, for comparison. The methods used, while subject to many possible sources of error, are ingenious, and demonstrate how useful information can be obtained in a short period from a small sample, providing that the investigator has a good general background knowledge of the community studies. "Household members were asked to display wood equivalent to that actually used - each piece was measured by length and circumference, and these figures were used to calculate volume". Measurements were taken in April and are believed to be not too unrepresentative of the true yearly average.

Daily household totals were estimated at 0.036 m3 (21.8kg) with a yearly per capita figure of 4.3 m3 (2,605 kg) for the three member households, and 2.7 m3 (19636 kg) for the five member households. These figures were checked by measuring three headloads, which showed the average to be 33 kg - these three loads took five hours to collect, and were estimated to be sufficient for a week, supplemented by wood brought back from agricultural trips. The authors also checked the figures by weighing the amount of wood collected by one household over a one week period, with close agreement with other estimates.

The report also deals with labour aspects - "The fuelwood task is disliked by Kwemzitu women (as) arduous, taxing and time-consuming". Labour surveys showed that women worked an average of 64 hours per week, with

- cooking and housework - 27 hours per week;
- agriculture - 20 hours per week;
- fuelwood collection - 11 hours per week.

Some families, especially wealthier ones, also burn charcoal and wood, and the report includes figures on wood out and wood actually Used. Total fuelwood consumption is estimated at an annual per capita rate of 2.6 m3, or 450 m3 for the village, plus 26 m3 for charcoal. In terms of tree-planting for fuelwood, the authors estimate that "Kwemzitu people would have to plant 1,360 trees every year" (using the locally popular wattle, Acacia mearnsii).

The study concludes with a consideration of total needs for Lushoto District, "tremendous" pressure on the forest reserve, and possible alternate energy sources: "massive reafforestation efforts are needed".

Previous Page Top of Page Next Page