Contents - Previous - Next

This is the old United Nations University website. Visit the new site at http://unu.edu

Losses and impairments due to fire

Uses of Fire

Generally, the use of fire in wildland areas has been limited to swidden agriculture, maintenance of hunting or grazing ground, and in forestry activities. In swidden agriculture, it has been an effective traditional way of clearing the land for crops. To maintain hunting or grazing ground, rural people usually use fire to clear old growths and, at the same time, to enhance new growth of young plant materials that are more palatable to wildlife or their cattle.

Although not widely used in South-East Asian forestry (Matthews; 1989), fire does play a role, particularly in silviculture and forest protection. By burning the forest floor periodically to reduce accumulation of litter and slashes, the occurrence and spread of a bigger and more devastating blaze can be avoided. Burning clears the litter and humus layers and exposes the mineral soils in preparation for the regeneration of the plant/timber species; and it also stimulates the shedding and germination of seeds. Fire can eliminate or reduce the weedy herb and shrub layers that otherwise use soil nutrients needed by timber. Infestation by insects and the incidence of diseases caused by fungus and others can be lowered by burning.

While the practice of burning may be of benefit, extensive wildfires that escape controlled burns or that may be caused by natural agents or by careless smokers are becoming increasingly frequent. Regardless of the scale, the burning of plant biomass or vegetation causes losses and impairments to the properties of soil and water bodies, the biological resources, the atmospheric conditions, and ultimately also the economy and welfare of the people.

Impacts on Soil and Water

In the tropical rain forest, the greater part of mineral nutrients are locked up in the plant biomass. Fire releases most of these nutrients for return to the soil, where they become readily available for plant growth. Burning also converts litter and humus of low pH to ash of higher pH which encourages nitrogenfixing bacteria to produce more nitrogen (Matthews, 1989). While these processes improve the soil fertility, this benefit may not last long. Most of the nutrients will soon be washed away by subsequent rains.

Other processes, however, may be quite contrary to intention. Fire can damage the soil fungal flora that help roots of plant species (such as dipterocarps and many other tropical rain-forest species) absorb nutrients from the soil and litter (W. Smits, personal communication). As indicated by Bormann and Likens (1979), fire can also remove certain nutrients by volatilization. Oxidation by fire, which is very rapid, promotes the decomposition of organic matter remaining in the soil, and the high intensity of heat may damage the structure of the soil. With the elimination of the plant biomass, much-if not all-of any subsequent rainfall will be directly hitting the bare ground surface, thus, dispersing the soil structures into finer particles, which then clog the soil pores. Under such conditions, much of the rain-water will be deflected as surface runoff: eroding the soil and the relatively fertile ash left by the fire. Thus, the burning of vegetation will greatly reduce the availability of nutrients in the ecosystem and increase the incidence and intensity of flash floods as well as accelerating soil erosion and siltation.

Following the 1982-3 forest fire in East Kalimantan, serious floods were experienced on major rivers in the province. Houses along the tributaries of the Mahakam River, for example, were submerged for several months during the 1983-4 rainy season (Wirawan, 1984c). A bridge along the Bengalon River, just north of the Kutai National Park, was reported to have been washed away. Furthermore, a study by Shimokawa (1988) within the Park area showed that soil erosion accelerated more than tenfold, from 0.13 - 0.35 millimetres per year in the unburned forest area to 2.30 - 4.65 millimetres per year in the burned forest area.

The Effect on Atmospheric Conditions

The burning of vegetation releases soot and gases into the atmosphere. These gases include carbon dioxide (CO2), carbon monoxide (CO), nitrogen oxides (NOx) such as nitric oxide (NO) and nitrogen dioxide (NO2) and hydrocarbons such as methane (CH4). In addition, soils exposed after the removal of forest cover also produce nitrous oxide (N2O), just like the nitrogen-rich fertilizers spread over fields (Graedel and Crutzen, 1989). Most of these gases-which are quite volatile, particularly in the presence of the hydroxyl radical (OH)- play a very Important direct or indirect role in regulating the climate of the earth (McElroy and Wofsy, 1986). The reactions of these and other gases in the atmosphere vary according to the local mixture of gases and particles, the temperature, the intensity of the sun's radiation, the presence of different kinds of clouds or precipitation, and the patterns of airflow (Graedel and Crutzen, 1989). Figure 11.1 shows the results of these reactions and their related problems.

Within days, the NOx can be converted by the OH radical into nitric acid (HNO3), which is readily dissolved in water or water droplets to form acid rain. As the reaction occurs soon after the NOx become available, the atmospheric life of this gas is relatively short and the deposition of the resulting acid is of regional significance only. This acid deposition, which can occur everywhere-including the Tropics-contributes to costly corrosion problems of outdoor equipment, buildings, etc. In addition, it may also increase the acidity of lakes, leading to the reduction in size and diversity of fish populations (Graedel and Crutzen, 1989), and creating local drinkingwater problems.

In the Tropics, the strong and abundant solar radiation stimulates a quick photochemical reaction to the NOx. According to Graedel and Crutzen, this reaction makes the ozone concentration near the ground five times higher than nominal. Newell, Reichle and Seiler (1989) further state that the formation and accumulation of ozone near the ground is encouraged by the increase of CO. Ozone is a toxic gas which causes eye irritation, impairment of lungs, and damage to trees and crops. Together with NOx and soot, ozone reduces visibility through creation of smog.

FIGURE 11.1 Problems Related to the Effects of Fire on Atmospheric Conditions

Graedel and Crutzen also show that an increase in atmospheric CO encourages the accumulation of methane. Because they react readily with CO, fewer OH radicals remain in the atmosphere to break down CH4 and other molecules. With continuing production from rice fields, cattle, landfill and burning of biomass and fossil fuel, more CH4 will be accumulated in the atmosphere. The increase of this and other gases (CO2, N2O, O3) will add to the greenhouse effect and thus contribute to atmospheric warming. An analysis by Palm et al. (1986) shows a net release of between 8 and 19 x 10 15 grams of carbon from 1860 to 1980 and between 0.15 and 0.43 x 10 15 grams in 1980, caused mainly by changes of forests to shifting cultivation and other permanently cleared land. A detailed study by Newell, Reichle and Seiler (1989) shows that the combustion in tropical rain forests and savannas generates at least as much CO as in fossil fuels. Thus, through intricate chain reactions, the burning of vegetation could cause local or regional problems, such as poor health, acid rain, corrosion, lower fishery and crop production, and smog, as well as augmenting the global warming problem.

Except for the problems related to breathing, drinking water and smog, there are few written reports on the atmospheric effects of the 1982-3 forest fire in East Kalimantan. However, the increase of acidity (acid rain), floods, and other changes in water chemistry brought about by the ash were apparently the factors that triggered the population explosions of Aeromonas hydrophila, Staphyloc c us sp. and Pseudomonas sp. that attacked and caused an epidemic of skin infection among the fish population of the Mahakam River (Tempo, 4 August 1984). In addition, the disease also attacked and caused serious health problems to the protected freshwater dolphin (Orcaella brevirostris). which is unique to the middle Mahakam areas (Wirawan, 1986).

One other aspect that received wide attention during and soon after the burning period was the smog which seriously affected transportation systems. Widodo and Rahman (1984) noted that thick smoke covered areas near the source of the fire for more than three months. Elsewhere, airline pilots reported that dense smog reached up to 5 000 metres. Further away, the sky was hazy, and noontime resembled dusk, with the sun as red as at sunset. Flights to inland areas were cancelled and, at limes, Balikpapan airport was closed or used only between 2 and 5 p.m. when convection lifted the smog. Similar conditions affected other airports in Kalimantan, as well as impeding traffic at airports further away in Surabaya and Singapore. The same happened again in 1991.

Poor visibility also affected travel by land and water. Although there is no record of accidents directly related to this problem, travelling by road between Balikpapan and Samarinda was very dangerous, not only because of poor visibility but also because flaring fires often jumped across the road. Furthermore, ships had to wait at Balikpapan harbour for improved visibility, and travelling upstream along the Mahakam River took almost twice as long as normal. Water travel by night was impossible. No deaths were attributed to hunger, but food shortages caused by transportation problems were reported from a number of places.

Impacts on Biological Resources

The effects of fire on biological resources relate to the losses of type and number of species of both plants and animals. The number of individuals and the volume or dominance of each species is affected, and their regeneration and maintenance are impaired. These impacts vary according to the size of the affected area as well as the intensity and frequency of fire.

PLANT RESOURCES

Small-scale, low-intensity, low-frequency fires, such as those which commonly occur in traditional shifting agricultural systems, have been considered by Connell (1978) and Huston (1979) as an example of the moderate-level disturbance factors that stimulate maintenance of high biodiversity in the tropical rain forest. These fires help create a mosaic of diverse vegetational stages that allow great numbers of plant and animal species, with varied ecological requirements, to live and grow together within the area.

In contrast, large-scale devastating fires nullify such a possibility. When they recur frequently within the same area, even small-scale fires can change the floristically rich, multi-layered forest vegetation into a single-layer grass vegetation dominated by the fireresistant alang-alang (Imperata cylindrica ). This phenomenon, for example, occurs in the Riam Kanan Reservoir area in South Kalimantan, where sporadic small fires occur every year and maintain the extensive alang-alang grassland in the area.

In the 1982-3 drought and fire in East Kalimantan, the loss of biological resources, though uncertain, was far from uniform. Based on a relatively short period of ground and aerial survey in the affected area, Lennertz and Panzer (1984) recognized three classes of damage. Class I areas received only drought damage, and 10 per cent of canopy trees died; while class 2 areas had both drought and fire damage, and 10 50 per cent of canopy trees died; and class 3 areas suffered severe burns, and more than 50 per cent of the canopy trees died. They reported that the area affected by the fire was 3.5 million hectares, including 800000 hectares of primary forest, 1 400000 hectares of logged-over forest, 750000 hectares of secondary forest, shifting cultivation and settlements, as well as 550 000 hectares of peat swamps and peat-swamp forest.

Although the 1983 4 ground and air surveys suggested that most of the affected areas had class 2 or 3 damage, Wirawan (1984a) found that the number of dead canopy trees was not always correlated with the intensity of the burn. He described three large segments of forests (totalling about 380 000-400000 hectares) that were not affected by the fire, but were identified by Lennertz and Panzer as severely burned. Here, the high percentage of dead canopy trees (up to 71 per cent) was found to be the result of drought only. Wirawan (1985) also reported that after the fire, live remnants of the forests varied from scattered individual trees, through pockets or stands of a few unburned hectares, to largely undamaged areas several thousand hectares in size.

Partly because of these findings, Schindele, Thoma and Panzer (1989) undertook a more detailed study using satellite imagery, aerial video recordings and various maps. After mapping the burned and unburned forest areas and other preliminary investigations, a detailed forest inventory was carried out in the burned areas through a stratified, random, 'double-cluster' sampling design. As these findings basically support Wirawan's observations, and are directly related to the subject of this chapter, it is important to review this study in some detail. Table I 1.1 presents a summary of the results. Their map of the burned and unburned areas is reproduced in generalized form in Figure 11.2.

With reference to species survival, the unburned amount is more important. Relatively high percentages of the limestone and rock forests (44 per cent), the kerangas or heath forest (55 per cent), the brackish swamp vegetation (77 per cent) and the tidal forests (100 per cent) were unaffected by the fire. Consequently, most of the plant species found in these forest types should still he represented in the unburned area. Although the proportions are somewhat lower in the peat/freshwater swamp forest (28 per cent) and other swamp vegetations (18 per cent), because of the relatively uniform habitat and species distributions as well as the low species diversities, a similar situation may also be true for these forest types. However, a greater loss of species could occur from the mixed lowland rain forests.

TABLE 11.1 Burned and Unburned Areas of Vegetation and Land-use Types within the 4733 000-hectare Study Area Affected by the 1982-1983 Forest Fire in East Kalimantan, Indonesia

Source: Modified from Schindele. Thoma and Panzer (1989).
a Refers to percentage of the total study area of 4 733 hectares.
b Refers to percentage of that vegetation type: for example in row 1, of 3 244 hectares. 67 per cent was burned and 33 per cent, unburned.

In the East Kalimantan area, 2.2 million hectares or 69 per cent of the floristically rich 3.2 million hectares of lowland rain forests was affected by the fire. The remaining I million hectares of unburned forests, which may have been affected by the drought and logging, consists of 89 per cent of the 410 000 hectares of undisturbed forest, 25 per cent of the 2 807 000 hectares of disturbed or logged forest, and only 4 per cent of the 27 000 hectares of unclassified forests that were converted to plantations. As the fire in the burned undisturbed forest (11 per cent) was limited to surface fire that damaged only the undergrowth, and no discernible differences between the burned and unburned forests were any longer observable on 1988 satellite imagery, the forest was considered to have recovered. Therefore, the 410 000 hectares of undisturbed forest can be considered as floristically unaffected by the fire.

FIGURE 11.2 Burned and Unburned Areas in the 1982-1983 Forest Fires in East Kalimantan

However, in the disturbed or logged forest, the fire damage varied with the degree of disturbance that existed before the fire. In the lightly disturbed forest, the effects were primarily in the lower and middle layers only. In spite of fire, the multi-layered structure of the forest was basically maintained. Before the fire, 75 per cent of the forest was logged and 25 pet cent was still primary forest. In this category of forest, the diversity of tree species remaining after the fire was considered still quite high. From the 223 most important tree species recognized in the study area prior to the field survey, 155 were recorded during the sample inventory. Together with species of Lauraceae, Moraceae, Myrtaceae, Verbenaceae and Caesalpiniaceae, dipterocarps occurred in lower and middle layers as well as in the upper layer, where they formed the dominant species with a relative basal area of 24 per cent and a relative abundance of 8 per cent. The disturbed parts were taken over by dense populations of Euphorbiaceae, particularly the pioneer species Macaranga gigantea and M. triloba, which have a relative basal area of 19 per cent and a relative abundance of 33 per cent.

In the moderately disturbed forest, the lower and middle layers were seriously damaged and to some degree so was the upper layer. Before the fire, 80 per cent of the forest was logged. Here, although 141 of the 223 important tree species were still observed during the inventory, dominance had shifted towards the pioneer species. While the upper layer was still recognizable and was dominated by the dipterocarps and species of Lauraceae, the middle layer was nearly absent, and the lower layer dominated by members of the Euphorbiaceae particularly M. gigantea, M. triloba and M. hypoleuca, Moraceae, Verbenaceae and Leeaceae. While the relative abundance of Euphorbiaceae remained the same, its relative basal area was increased to 21 per cent. In contrast, the relative abundance of the dipterocarps was reduced to 4 per cent and the relative basal area to 17 per cent.

All layers of the heavily disturbed forest were seriously damaged. Although prior to the fire, 80 per cent of the forest was also logged, the logging activities in this forest category were carried out within eight years of the fire. Here, although 125 of the 223 important species were still observed, the whole forest structure was destroyed by the fire and the surviving trees occurred primarily as scattered individuals, or groups of individuals, in the matrix of a single, relatively uniform layer of pioneer species formed by members of the Euphorbiaceae. While the relative abundance and dominance of the Euphorbiaceae increased to 30 and 27 per cent, respectively, the relative abundance of the fire-resistant Eusideroxylon zwageri (Lauraceae) and certain species of Shorea, which form the second most important species after the Euphorbiaceae, was between 3.2 and 3.7 per cent only. In addition, as shown in Table 11.2, the increasing loss of species diversity and dominance of the 'climax' primary rain forest species are further indicated by the reductions of the 'importance value' of timber species, and the volumes of standing stocks of timbers with diameters larger than 20 centimetres.

TABLE 11.2 'Importance Value' of Timber Species and Volumes of Standing Stocks of Exported, Local and Other Timbers Larger Than 20 cm dbh after the Fire, in the Three Categories of Forest Disturbance

Source: As for Table 11.1

From these findings, Schindele, Thoma and Panzer (1989) concluded that the lightly disturbed forest should be able to recover naturally to its undisturbed condition within a few decades, while the moderately and heavily disturbed forest will take much longer. Combining the total area of lightly disturbed forest with undisturbed forest, in a few decades, almost half of the mixed lowland rain forest will be back to its original undisturbed condition. While many species have certainly been lost during the fire, there is no doubt that these areas contain a significant representative sample of the mixed lowland rain-forest flora. Together with what is left in the limestone and rock forests, the kerangas or heath forest, the peat/freshwater swamp forest, brackish and other swamp vegetation and also the tidal forest, they all form a significant representative sample of the remaining flora after having been affected by drought and fire.

ANIMAL RESOURCES

To a certain degree, the same situation may also exist with the animal community. Somewhat similarly to the pioneer species of Macaranga that took over the barren lands after the fire, Wirawan and Hudiyono (1983) noted population explosions of butterfly and moth species in the Kutai National Park areas soon after the fire. While these insects apparently were making the best use of the rich minerals that became available from the ash, an epidemic of caterpillars was also attacking the dry logs and dead trees created by the fire (Asiaweek, 13 July 1984). In addition, Leighton and Wirawan (1986) noted the abundance of spiders building their webs between the standing dead trees. Aside from these phenomena, the change in animal life was less conspicuous.

A study on the mammals of Bukit Soeharto Reserve by Yasuma and Alikorda (1990) confirmed that 1 of 8 Insectivora, 4 of 10 Scandentia, 1 of 1 Dermoptera, 8 of 92 Chiroptera, 9 of 13 Primata, I of 1 Pholidota, 18 of 61 Rodentia, 12 of 28 Carnivora, 0 of 1 Proboscidea, 0 of 1 Perissodactyla, and 6 of 12 Artiodactyla species recorded for Borneo are still present in this logged and burned forest. Considering that many of these Bornean species (such as the orang-utan (Primate), Sumatran rhinoceros (Perissodactyla), elephant (Proboscidea) and others) did not occur in the area even before the forest fire, the survival of these mammals is relatively high. This is probably also the case for other parts of the burned area.

KUTAI NATIONAL PARK

The eastern coastal half of the affected area is the most intensively studied pan of the province, and perhaps of all Kalimantan. The richness of wildlife long ago led Witkamp (1932) to propose a 2-million-hectare reserve bordered by natural boundaries, which happened to be also the distributional limits of some of the key animals (the orang-utan and Sumatran rhinoceros). The area was defined by the Miau and Karangan Rivers in the north, the Makassar Strait in the east, the Mahakam River in the south, as well as the Rantau Kedang, Ngayau and Telen Rivers in the west. Other key animals that Witkamp considered important in this area were the Malayan sun bear, banteng, sambar deer and proboscis monkey. Approval was given by the Sultan of Kutai and the Dutch government in 1936 for a greatly reduced 306000-hectare reserve, but all these mammals (except for the Sumatran rhinoceros whose presence was uncertain) were still present in this reserve when the Ministry of Agriculture reduced it in 1971 to 200000 hectares. All the studies carried out in this reserve up to 1985 are listed by Wirawan (1985).

Before the drought and forest fire, Pearson (1975) found this reserve (now known as the Kutai National Park) rich in bird species. Based on a survey in the area, he listed some 300 species belonging to 47 families and sub-families, including 239 species of 33 families and sub-families or 83 per cent of the forest birds of Borneo. Cockburn and Sumardja (1979) observed 7 species of ungulates (including the Sumatran rhinoceros, banteng, sambar deer, barking deer, mouse deer and bearded pig), 12 species of carnivores (including the Malayan sun bear, leopard cat, binturong, civets, mongooses, weasel and otters), 11 species of primates (including the orang-utan, macaques, proboscis and leaf monkeys, Bornean gibbon, slow lords and tarsier) and 25 species of rodents (including the porcupine, mice, rats and 15 species of squirrels).

After the fire, studies by Wirawan (1985), Leighton and Wirawan (1986), Azuma (1988), Doi (1988) and Suzuki (1988) indicated that most of the big mammals (except the Sumatran rhinoceros) were still in the area. While smaller animals are more difficult to see and therefore need more intensive fieldwork, major concerns were raised on the population status of the frugivorous birds and rodents, as noted by Leighton and Wirawan (1986) in the Mentoko study area. The fire killed 52 per cent of the fruit trees belonging to the Meliaceae and Myrtaceae, which are highly preferred by the 6 species of hornbills; 2 of the 5 territorial hornbills were no longer observed in the area during the fieldwork in September 1983 and August 1984. With the loss of many of the 50 species of Lauraceae trees that produce lipid-rich fruits, and other trees that produce sugar-rich fruits, both preferred by frugivorous birds such as pigeons, barbels, hill myna, green broad-bill and others, these bird species could also be disappearing.

While the large-bodied anthropoid primates were considered least affected by the fire, and they were able to adapt successfully to short-term habitat damage by shifting to less preferred food items and exploiting the insect outbreak (Berenstain, 1986; Kompas, 1984b; Suzuki, 1988); the populations of three species of diurnal seed-eating squirrels, common in the forest floor prior to the fire, were found to be definitely declining. Aside from losing their food supply, the loss of forest structure was also considered to have greatly impaired the movement of many of these tree-dwelling animals.

A preliminary study in the burned and unburned forest areas in the Park, by Yajima (1988), found that the terrestrial invertebrate communities (macro-soil animals, floor animals and phytal animals) of the tropical rain forests were recovering quite quickly after the fire. Despite losses of plants and animals during the drought and forest fire, the above studies suggest that significant biological resources still remain in this unique area. Even though half of its 200 000-hectare area was affected by the fire, Kutai National Park is still considered to be the only large reserve in the world that contains such an assortment of biological resources (Leighton and Wirawan, 1986; Ministry of Forestry, 1991; Petocz, Wirawan and MacKinnon, 1990; Wirawan, 1985).

Impacts on the People and the National Economy

Because the problems resulting from fire are complex, and very often difficult to quantify. complete economic assessment and valuation of the losses of crops, personal property and the diversity of biological resources, as well as impairment to the environment, is not simple. Nevertheless, tire greatly affects the welfare and the economy of the people, the country and even the world.

As previously described, some of the problems that affect the welt:are of the people, particularly local rural people, are related to health, sources of drinking water. food and cash income. While drought may have already reduced the supply of water before the fire, acid rain causes further damage to these water resources-damage which then affects not only the sources of drinking water but also the fish stocks on which many of the rural people, such as those living in the middle Mahakam, depend. Severe food shortages may also be caused by transportation delays brought about by poor visibility and the loss of bridges. Furthermore, the reduction of wildlife, due to the direct effect of fire or the loss of habitats, reduces the availability of meat. Damage to vegetation, affecting timber and non-thnber forest products (such as rattan, gaharu and natural latex), and the loss of crops (like food plants and pepper), eliminates sources of food and cash income. At villages surrounding the Bukit Soeharto Reserve, for example, farmers lost a great deal of their pepper crop. Because the crop is so valuable and important for their survival, some of them were reported to have suffered severe mental breakdown because of the loss.

All these effects were of both local and national significance. But, while loss of species might not immediately impact on decision makers, major concern was felt over the loss of timber and non-timber forest products. Attempts to evaluate these losses were made by Lennertz and Panzer (1984) and Schindele, Thoma and Panzer (1989). As noted earlier, the total forest area estimated by Lennertz and Panzer to have been affected by the drought and fire was 3.5 million hectares. Based on the estimated volumes of timber lost in the three classes of forest damage and the price of timber at that time ($80 per cubic metre), they calculated that the total loss of commercial timber from the area was $5.5 billion.

They went on, however, to look forward, comparing the present standing stock volumes with the volume that is expected to become available again 35 years after the first time the forest was logged. In this calculation, it was assumed that through the Indonesian Selective Logging System, the volume of timber recovered during each logging cycle is about the same as the volume of the first cut. Based on the standing stock volumes shown in Table 11.2, the loss of timber from the lightly, moderately and heavily disturbed forests was respectively estimated to be 25, 50 and 100 per cent of the timber otherwise available during the next cutting cycle. Although reliable data on the amount of commercial timber harvested during the first cycle are generally lacking, a volume of 80 cubic metres per hectare was considered a good estimate. Multiplying the area of burned, disturbed forests in each of the three classes with the respective percentage of timber loss by this volume estimate, and with the 1983 price of timber, they calculated the total loss of commercial timber to be $7.98 billion.

In addition, from the 405 000 of the 566 000-hectare swamp forests burned during that time, using the local market value and the Sarawak estimate of timber volume in this forest type, the timber loss was calculated at $348 million. Together with an estimated $373 million worth of nontimber goods (such as rattan, fruits, nuts, resins, honey and latex) foregone and the cost of rehabilitation at $250 per hectare of the moderately and heavily burned forests totalling another $373 million, the total loss of timber and non-timber goods from the area affected by the 1982-3 drought and fire was calculated to be $9.075 billion.

Contents - Previous - Next