Previous Page Table of Contents Next Page


5.1 The Use of Tree Legumes for Fuelwood Production

P.A. Ryan


Introduction
Fuelwood Properties
Drying Fuelwood
Managing Trees for Fuelwood Production
Nutritional Considerations
Control of Wildfires
Conclusions
References


Introduction

Fuelwood is the cheapest fuel available per unit of heat in most developing countries. The annual use of fuelwood has been estimated at 1,200 million cubic metres worldwide (Arnold and Jongma 1978). Fuelwood can be harvested on demand and is easily stored and dried. It can be produced from most tree species and from a wide range of silvicultural systems. However, if fuelwood production is a primary management aim of tree planting, a variety of factors needs to be considered to optimise both the quantity and value of fuelwood produced. For example, a species with high volume production is of little fuelwood value if the wood is very light or if the burning wood produces toxic smoke.

In this section, some of the factors that determine the suitability of tree legumes for fuelwood will be discussed and some of the principles of plantation management and how these apply to fuelwood production will be considered.

Fuelwood Properties

Most wood burns relatively easily, though the wood of some species (e.g. species of Syncarpia) is fire retardant and will not burn except in hot fires in mixture with more flammable woods. Other woods, while they may burn readily, may not be suitable because of excessive spark production or odorous, toxic or irritating smoke. The wood of Sesbania grandiflora for example is not highly regarded as fuel because of the excessive smoke it produces when burning (Hegde 1990). The importance of these factors for domestic fuelwood use depends on the type of stove used, cooking methods and the adequacy of ventilation. Local preferences may also be important depending on the effects of the fuelwood on the flavour of the cooked food.

Other fuelwood properties include calorific value, wood density and wood moisture content. Gross calorific value is the total energy content per unit weight of wood determined from bomb calorimetry tests in the laboratory. Wood density and specific gravity are expressions of how much wood substance is present in a given volume of wood, i.e. how 'heavy' or 'light' a wood is. Although this is a simple relationship, it is complicated by the number of ways it can be derived. Three common expressions are basic density (the mass of oven dry wood per unit of green volume) air-dry density (the mass of air dry wood per unit of air dry volume) and specific gravity (the dry weight of a given volume of wood divided by the weight of an equal volume of water). Specific gravity of wood ranges from about 0.1 to 1.4 (Zobel and van Buijtenen 1989) (Table 5.1.1).

The significance of these factors in determining the utility of particular species for firewood is frequently misunderstood. Gross calorific value is not an important property since there is little variation among species (mean ± SD = 19.73 ± 0.98 MJ/kg for hardwood species) (Harker et al. 1982). However, the total energy contained in wood is not converted completely to available heating energy since this includes heat generated by the combustion of hydrogen (about 6% of wood mass). In practice, fires are open to the atmosphere and the heat generated by the combustion of hydrogen is lost as latent heat of vaporisation of the water. This loss is equivalent to about 1.4 MJ/kg (Harker et al. 1982). Heat is also lost in vaporising moisture contained in the wood. Thus the moisture content of fuelwood is the most significant factor affecting the production of usable heat when wood is burnt (Table 5.1.2). The initial moisture content of a tree when cut (i.e. when the wood is green) and the rate at which wood dries are important factors to be considered in assessing the potential utility of a species for fuelwood. Some of the characteristics of fuelwood of a range of tree legumes are presented in Table 5.1.1.

While there is little variation among species in total energy content per unit weight of wood, differences in wood density result in substantial differences in energy content per unit volume of wood. Consequently, wood density of a species may be an important consideration where the bulk of a load rather than its weight limits the amount that can be transported. The rate of heat production during burning is dependent on the piece size of the fuelwood but is also influenced by wood density. Where cooking requires slow steady concentrated heat production, fuelwood needs to be relatively dense and piece size needs to be relatively large. Small pieces of light wood burn too quickly to be suitable for this type of cooking. However, low density woods can be used for quick hot fires while high density woods are suitable if cut into small pieces.

Table 5.1.1. Fuelwood characteristics for a range of tree legumes.

Table 5.1.2. Effect of moisture content on the heating value of fuelwood.

Moisture (%)1

0

10

25

50

75

100

150

250

400

Heating value (%)2

100

90

78

63

52

44

33

20

10

1 Moisture content is the weight of moisture as a percentage of wood oven dry weight for a fixed weight of green fuelwood

2 Heating value is the amount of usable heat produced by wood at a given moisture content compared with that produced by oven dry wood

Drying Fuelwood

Wood begins to dry out as soon as it is cut and progresses through several stages. In the first stage of drying, free water is lost until fibre saturation point (about 24% moisture content) is reached. Drying beyond this point to equilibrium moisture content (about 15%) takes progressively longer since it involves the removal of bound water. The moisture content at these levels and the rate of drying vary depending on humidity and temperature.

In practice, drying to about 24% moisture (which provides 80% of the energy yield of totally dry wood) is adequate. Cutting the wood into small lengths and splitting while wood is green will increase the rate of drying. This is especially important for those species which are slow to dry.

Managing Trees for Fuelwood Production

The way in which trees are managed for fuelwood production will depend on the aims and priorities of the particular land management system in use. For example, in mixed tree/pasture systems, the management of trees where wood production is the primary aim and pasture secondary will be different from that where pasture is of major importance and wood is a by-product. While the basic silvicultural principles are the same, their application varies and providing the principles are understood, management systems appropriate to any system can be devised.

Spacing and thinning

The general sigmoidal pattern of growth relative to age applies both to individual trees and to populations of trees grown together in stands. Growth may be measured in a variety of ways: stem diameter (usually measured at 1.3 m), tree height, stand basal area (sum of the cross-sectional area of all the stems in the stand), tree or stand wood volume, and tree or stand biomass. The harvesting age at which long-term yields are maximised (the silvicultural rotation) is the age at which the current annual increment (e.g. of volume) equals the mean annual increment. At this stage, cutting and replanting the stand will provide a greater yield in the long term than allowing the stand to continue to grow even though tree size will continue to increase. This principle also applies to maximising total volume, merchantable volume or biomass production, or the profitability of a commercial plantation.

'There are remarkably few data on fuelwood yields from dense plantations of trees of any kind and even fewer for leguminous trees. Forest mensuration data commonly focus on timber volumes of widely spaced trees not on total biomass of dense plantations.' The values for leucaena in Table 5.1.3 indicate that dense spacing is a necessary criterion for maximal early mean annual increments (Brewbaker et al. 1984). Data from other trials (Brewbaker et al. 1982) confirm these observations and suggest that a 4-6 year harvest of trees at 1 x 1 m or 1 x 2 m spacings may provide maximum yields when moisture is not limiting.

In the Philippines, Gliricidia sepium is grown in woodlots to produce fuelwood for curing tobacco. Trees are spaced at 1.5 x 2.0 m or 2.0 x 2.0 m and fuelwood size is ideal for the kilns. Annual yields of up to 20 m3/ha have been recorded from these lots. Living fences with spacings as low as 50 cm produce much smaller sized wood (Glover 1989).

An additional factor to be considered in the management of fuelwood plantations is the relationship between age, stand density and wood density. As a general rule for mature trees, wood density increases from the pith to mature heartwood and then declines from heartwood to sapwood. These properties tend to be related to age and to position within the tree. Young trees or trees harvested from dense plantings will tend to have a lower specific gravity than those that are older or that have grown more quickly at lower stand densities. Consequently, young trees may provide suitable firewood for quick hot fires but the firewood will not be satisfactory for slow, sustained heat production.

Wood density is especially important where wood is grown for charcoal production. While low density woods can be used to produce charcoal, it tends to be of poor quality and crumbles easily. The best charcoal is derived from high density woods. Most tree legume species produce good charcoal and some, including species of Acacia, Leucaena and Prosopis, are widely renowned (Brewbaker et al. 1984) (Table 5.1.1).

Table 5.1.3. Effect of plant density on wood yields of Leucaena leucocephala (after Brewbaker et al. 1984).

Location

Age
(years)

Density
(plants/ha)

Annual increment
(m3/ha/year)

Waimanalo Hawaii


1

40,000

87

4

20,000

70

Kauai Hawaii


1

40,000

71

4

20,000

93

Molokai Hawaii


1

40,000

97

3

20,000

72

Taiwan


1

40,000

20

4

5,000

41

Pruning, lopping, pollarding and coppicing

Trees can be managed by applying various cutting treatments to provide a sustained source of firewood while allowing the tree to continue to grow. Some of these treatments are particularly suitable for systems where the trees fulfil multipurpose roles.

Pruning to remove the lower branches flush with the stem (to c. 5 m above ground level) may provide useful though relatively small amounts of firewood and will not affect the growth of the tree unduly providing about 67% of the green crown is retained. Pruning, particularly of trees which carry a low, dense crown, could be carried out in agroforestry systems to increase the amount of light reaching the pasture underneath or to increase the amount of clear, knot-free wood and thus improve the value of the tree for sawn timber or veneer.

Lopping (cutting off branches but leaving stubs of 30-100 cm, frequently as high as can be reached by climbing) and pollarding (cutting out the top of the tree) are common management practices in agroforestry. Frequently both lopping, leaving branch stubs to climb on, and pollarding are carried out at the same time, the branches and upper stem being used for firewood.

Coppicing is the most extreme of these treatments and, in forestry terms, generally involves cutting the tree at about 10 cm above ground level. In many cases, continual harvesting is possible over a number of rotations without the need to replant before yields begin to decline. Gliricidia has been cut at 20-30 cm above ground on a 2-4 year cycle for over ten cycles in Timor, Indonesia (A.P.Y. Djogo, unpublished data). However, tree species vary in their capacity to produce coppice shoots when cut; some (like leucaena and gliricidia) reshoot readily while others (e.g. Sesbania grandiflora) will die. Individual species may vary also in their response to cutting depending on their age and size. Species that do not sprout readily from stumps may do so when cut at a greater height above the ground (say at 100 cm) with the retention of a live branch. Coppice regrowth, even from those trees that sprout readily, may be more vigorous when a live branch is retained. Cutting should be carried out when trees are not under stress (e.g. from drought or waterlogging) to maximise the survival of stumps and the vigour of coppice.

Nutritional Considerations

Regular harvesting of tree legumes for fuelwood may result in a substantial removal of nutrients depending on management. The amount of nutrients immobilised in the biomass of a tree may vary among species and sites and also with age and stocking density (Table 5.1.4). Nutrients are distributed unevenly throughout a tree, the concentration generally being highest in the foliage component and decreasing in the order: foliage >> stem bark > branches > stemwood. The amount of nutrient removed at harvest depends on the biomass and the nutrient concentration of each component removed. Significant nutrient removals can result from harvesting branches for firewood because of the higher nutrient content in branch wood and bark. However, this depends on both the age of the trees and the branch and stem diameter. As stem diameter increases, the proportion of bark in the biomass declines and consequently, the relative loss of nutrient due to bark removal declines. Nutrient concentration in wood also declines with increasing stem diameter due to withdrawal and translocation of nutrients from older wood. These factors need to be taken into consideration in determining the spacing at which trees are planted (since diameter tends to increase as spacing increases) and also the age at which trees are harvested. Leaves (either live or as ground litter) should never be harvested for burning as this leads to substantial depletion of nutrients from the site.

Control of Wildfires

Brewbaker et al. (1984) reviewed wildfires and found that 'fire is probably the major cause of loss of forests planted for fuelwood in the tropics'. Fuelwood lots established without adequate control of perennial grasses are especially prone to fire damage. These authors state that 'most newly planted forests are adjacent to agricultural land and fire is the principal weapon of subsistence farmers against insidious grasses and other weeds. It is also their major tool in revitalising rangelands and preparing farmlands for planting.'

Table 5.1.4. Biomass and nutrient content of a number of tree legume species at different locations and of different ages.

Species


Country


Age

Total above ground biomass

Total nutrient content
(kg/ha)

Source


(years)

(t/ha)

N

P

K

Acacia dealbata

New Zealand

8.0

182

1006

43

638

1

Albizia lebbeck

Puerto Rico

3.0

29

534

33

149

2, 4

Albizia procera

Puerto Rico

5.5

124

540

102

370

2, 4

Leucaena leucocephala (K-8)

Puerto Rico

5.5

47

370

39

220

3, 4

Leucaena leucocephala

Puerto Rico

5.5

33

210

23

127

2, 4

Prosopis juliflora

Kenya

8.0

216

1674

116

1219

3

References: 1. Frederick et al. (1985); 2. Lugo et al. (1990); 3. Maghembe et al. (1983); 4. Wang et al. (1991)

'Escaped fires may account for as much as half the loss of newly planted forests in the tropics' (Brewbaker et al. 1984). These authors documented fireburns in planted pine and eucalypt forests in Western Samoa of 5,000 ha, in the Philippines of 3,000 ha, in Sri Lanka of 2,000 ha, in Nepal of 1,000 ha and in Thailand of 500 ha. None of these forests was planted with firebreaks and Brewbaker et al. (1984) suggested that future reforested areas must either be planted with fire tolerant species, or be provided with firebreaks or buffer zones of multipurpose species for community use.

Among leguminous trees, 'many Acacia and Prosopis species have fire retardant foliage but are often planted in grasslands that are prone to fire. Firebreaks of leucaena and gliricidia have proved highly effective in the Philippines provided the break is densely planted (1x2 m) and adequate in width (10-20 m)' (Brewbaker et al. 1984). Both species grow back from the root crown after fire unless burned thoroughly to the earth line. Similarly, some Casuarina and Acacia species could be used as firebreaks and will regenerate from the crown after fire.

Conclusions

Most tree legume species are potentially suitable for fuelwood production providing that the wood is not fire retardant and that levels of sparks, smoke, odours and tastes are acceptable. The rate of combustion depends on wood density (specific gravity) and firewood piece size. There is little variation among species in the gross calorific value of wood and this is not an important property. Effective heating value, however, depends on the moisture content of the wood when burnt. The rate at which wood dries is therefore a very important property. Tree legumes which have been identified as outstanding fuelwood species include Acacia auriculiformis and A. mangium for acid soils, A. mearnsii and Leucaena diversifolia for highland sites, and L. leucocephala and Gliricidia sepium for lowland tropical sites (NAS 1980, 1983b, Brewbaker et al. 1984).

References

Anonymous (1983) Leucaena Research in the Asian-Pacific Region. Proceedings of a workshop, 23-26 November 1982, Singapore. IDRC, Ottawa, Canada, 192 pp.

Arnold, J.E.M. and Jongma, J. (1978) Fuelwood and charcoal in developing countries; an economic survey. Proceedings of the VIII International Forestry Congress, Jakarta, Indonesia.

Brewbaker, J.L. (1987) Prosopis pallida - pioneer species for dry, saline shores. NFT Highlights 87-05, NFTA, Hawaii.

Brewbaker, J.L., Van den Beldt, R. and MacDicken, K. (1982) Nitrogen fixing tree resources: potentials and limitations. In: Graham, P.H. and Harris, S.C. (eds), BNF Technology for Tropical Agriculture. CIAT, Cali, Colombia, pp. 413-425.

Brewbaker, J.L., Van den Beldt, R. and MacDicken, K. (1984) Fuelwood uses and properties of nitrogen fixing trees. Pesquisa Agropecuaria Brasileira 19, 193-204.

Evans, D.O. (1986) Sesbanias - a treasure of diversity. NFT Highlights, 86-04, NFTA, Hawaii.

Frederick, D.J., Madgwick, H.A.I. and Jurgensen, M.F. (1985) Dry matter, energy and nutrient contents of 8-year-old srands of Eucalyptus regnans, Acacia dealbata and Pinus radiata in New Zealand. New Zealand Journal of Forestry Science 15, 142-157.

Glover, N. (1989) Gliricidia Production and Use. NFTA, Hawaii.

Harker, A.P., Sandals, A. and Burley, J. (1982) Calorific values for wood and bark and a bibliography for fuelwood. Report of the Tropical Products Institute, London, UK, G162.

Hegde, N. (1990) Wood production systems. In: Evans, D.O. and Macklin, B. (eds), Perennial Sesbania Production and Use. NFTA, Hawaii, pp. 20-22.

Lowry, J.B. and Macklin, B. (1988) Calliandra calothyrsus - an Indonesian favourite goes pantropic. NFT Highlights, 88-02, NFTA, Hawaii.

Lugo, A.E., Wang, D. and Bormann, F.H. (1990) A comparative analysis of biomass production in five tropical tree species. Forestry Ecology and Management 31, 153-166.

Maghembe, J.A., Kariuki, E.M. and Haller, R.D. (1983) Biomass and nutrient accumulation in young Prosopis juliflora at Mombasa, Kenya. Agroforestry Systems 1, 313-321.

NAS (1980) Firewood Crops: Shrub and Tree Species for Energy Production. National Academy Press, Washington, DC.

NAS (1983a) Calliandra: a Versatile Small Tree for the Humic Tropics. National Academy Press, Washington, DC.

NAS (1983b) Firewood Crops: Shrubs and Tree Species for Energy Production. National Academy Press, Washington, DC.

NAS (1983c) Mangium and Other Fast-growing Acacias for the Humid Tropics. National Academy Press, Washington, DC.

Prinsen, J.H. (1988) Albizia lebbeck - a promising fodder tree for semi-arid regions. NFT Highlights, 88-03, NFTA, Hawaii.

Raintree, J. (1987) The multipurpose rain tree. NFT Highlights, 87-06, NFTA, Hawaii.

Reyes, G., Brown, S., Chapman, J. and Lugo, A.E. (1992) Wood Densities of Tropical Tree Species. General Technical Report, SO-88, USDA Forest Service Southern Forest Experimental Station, New Orleans, LA.

Rosecrance, R.C. (1989) The relationship between specific gravity and growth of 12 fast growing tree species. Nitrogen Fixing Tree Research Reports 71, 34-35.

Trurnbull, J.W. (1986) Multipurpose Australian Trees and Shrubs: Lesser-known Trees for Fuelwood and Agroforestry. ACIAR Monograph No. 1, Canberra.

Turnbull, J.W. (1987a) Acacia auriculiformis - the adaptable tropical wattle. NFT Highlights, 87-03, NFTA, Hawaii.

Turnbull, J.W. (1987b) Acacia mangium - a fast growing tree for the humid tropics. NFT Highlights, 87-04, NFTA, Hawaii.

Turnbull, J.W. (1987c) Australian Acacias in Developing Countries. Proceedings of an international workshop, Gympie, Australia. ACIAR Proceedings No. 16, Canberra.

Turnbull, J.W. (1991) Advances in Tropical Acacia Research. Proceedings of an international workshop, 11-15 February 1991, Bangkok, Thailand. ACIAR Proceedings No. 35, Canberra.

Wang, D., Bormann, F.H., Lugo, A.E. and Bowden, R.D. (1991) Comparison of nutrient-use efficiency and biomass production in five tropical tree taxa. Forest Ecology and Management 46, 1-21.

Zobel, B.J. and van Buijtenen, J.P. (1989) Wood Variation - Its Causes and Control. Springer Verlag, Berlin, 363 pp.


Previous Page Top of Page Next Page