The geographical distribution of biomass burning

Water Freedom System

Survive Global Water Shortages

Get Instant Access

The locations of biomass burning are varied and include tropical savannas, tropical, temperate and boreal forests, and agricultural lands after the harvest. The burning of fuelwood for domestic use is another source of biomass burning. Global estimates of the annual amounts of biomass burning from these sources are estimated in Table 7.2 (Andreae, 1991).

Table 7.2 Global estimates of annual amounts of biomass burning and of the resulting release of carbon to the atmosphere

Source

Biomass burned (Tg dm yr-1)

Carbon released (Tg C yr-1)

Savanna

3690

1660

Agricultural waste

2020

910

Fuelwood

1430

640

Tropical forests

1260

570

Temperate/boreal forests

280

130

World totals

8680

3910

Note: dm = dry matter (biomass matter). Source: Andreae (1991)

Note: dm = dry matter (biomass matter). Source: Andreae (1991)

As already noted, biomass matter is composed of about 45 per cent carbon by weight. Table 7.2 gives estimates of the carbon released (Tg C yr-1) by the burning of this biomass (the total biomass burned is multiplied by 45 per cent to determine the amount of carbon released into the atmosphere during burning). Combining estimates of the total amount of biomass matter burned per year (Table 7.2) with measurements of the gaseous and particulate emissions from biomass burning (Table 7.1) permits estimates of the global production and release into the atmosphere of gases and particulates from biomass burning, which will be discussed in more detail.

Biomass burning in the boreal forests

In the past, it was generally assumed that biomass burning was primarily a tropical phenomenon. This is because most of the information that we have on the geographical and temporal distribution of biomass burning is largely based on tropical burning. Very little information was available on the geographical and temporal distribution of biomass burning in the boreal forests, which cover about 25 per cent of the world's forests. To illustrate how our knowledge of the geographical extent of burning in the world's boreal forests has increased in recent years, consider the following.

Early estimates based on surface fire records and statistics suggested that 1.5 million hectares (1 hectare = 2.47 acres) of boreal forests burn annually (Seiler and Crutzen, 1980). Later studies, based on more comprehensive surface fire records and statistics, indicated that earlier values underestimated burning in the world's boreal forests and that an average of 8 million hectares burned annually during the 1980s, with great year-to-year fluctuations (Stocks et al, 1993). One of the largest fires ever measured occurred in the boreal forests of the Heilongjiang Province of Northeastern China in May 1987. In less than four weeks, more than 1.3 million hectares of boreal forest were burned (Levine et al, 1991; Cahoon et al, 1994). At the same time, extensive fire activity occurred across the Chinese border in Russia, particularly in the area east of Lake Baikal between the Amur and Lena rivers. Estimates based on National Oceanic and Atmospholic Administration (NOAA) Advanced Very High Resolution Radiometer (AVHRR) imagery indicate that 14.4 million hectares (35.7 million acres) in China and Siberia were burned in 1987 (Cahoon et al, 1994), dwarfing earlier estimates of boreal forest fire burned area.

While 1987 was an extreme fire year in eastern Asia, the sparse database may suggest a fire trend. Is burning in the boreal forests increasing with time, or are satellite measurements providing more accurate data? Satellite measurements are certainly providing a more accurate assessment of the extent and frequency of burning in the world's boreal forests. As global warming continues, predicted warmer and dryer conditions in the world's boreal forests will result in more frequent and larger fires and greater production of CO2 and CH4 by these fires, such increased burning could therefore represent a significant positive climate change feedback.

Calculations using the satellite-derived burn area and measured emission ratios of gases for boreal forest fires indicate that the Chinese and Siberian fires of the 1987 contributed about 20 per cent of the total CO2 produced by savanna burning, 36 per cent of the total CO produced by savanna burning, and 69 per cent of the total CH4 produced by savanna burning (Cahoon et al, 1994). Since savanna burning represents the largest component of tropical burning in terms of the vegetation consumed by fire (Table 7.2), it is apparent that the atmospheric emissions from boreal forest burning must be included in global emissions budgets.

There are several reasons that burning in the world's boreal forests is very important:

1 The boreal forests are very susceptible to global warming. Small changes in the surface temperature can significantly influence the ice and snow albedo feedback (a change in how reflective the land surface is). Thus, infrared absorption processes by fire-produced greenhouse gases, as well as fire-induced changes in surface albedo and infrared emissivity in the boreal forest regions, are more environmentally significant than in the tropics.

2 In the world's boreal forests, global warming will result in warmer and drier conditions. This in turn may result in enhanced frequency of fire and an accompanying enhancement in the emissions of CO2 and CH4 that will then amplify the greenhouse effect.

3 Fires in the boreal forests are the most energetic in nature. The average fuel consumption per unit area in the boreal forest is in the order of 25,000 kg ha-1, which is about an order of magnitude greater than in the tropics. Large boreal forest fires typically spread very quickly, most often as 'crown fires', causing the burning of the entire tree up to and including the crown. Large boreal forest fires release enough energy to generate convective smoke columns that routinely reach well into the upper troposphere, and on occasion, may directly penetrate across the tropopause into the stratosphere. The tropopause is at a minimum height over the world's boreal forests. As an example, a 1986 forest fire in northwestern Ontario (Red Lake) generated a convective smoke column 12 to 13km in height, penetrating across the tropopause into the stratosphere (Stocks and Flannigan, 1987). There is a strong link between boreal forest fires in Canada and eastern Russia in 1998 and increased stratospheric aerosols during the same period (Fromm et al, 2000).

4 The cold temperature of the troposphere over the world's boreal forests results in low levels of tropospheric water vapour. The deficiency of tropospheric water vapour and the scarcity of incoming solar radiation over most of the year result in very low photochemical production of the hydroxyl radical over the boreal forests. The OH radical is the overwhelming chemical scavenger in the troposphere and controls the atmospheric lifetime of many tropospheric gases, including CH4. The very low concentrations of the OH radical over the boreal forests result in enhanced atmospheric lifetimes for most tropospheric gases, including CH4 produced by biomass burning. Hence, gases produced by burning, such as CO, CH4 and the oxides of nitrogen, will have enhanced atmospheric lifetimes over the boreal forest.

New information about burning in the world's boreal forests, based on satellite measurements, was reported by Kasischke et al (1999). Some of the findings reported in this study are summarized here:

1 Fires in the boreal forest covering at least 100,000 hectares are not uncommon.

2 In the boreal forests of North America, most fires (>90 per cent) are crown fires. The remainder are surface fires. Crown fires consume much more fuel (30 to 40 tonnes of biomass material per hectare burned) than surface fires (8 to 12 tonnes of biomass material per hectare burned).

3 The fire record for North America over the past three decades clearly shows the episodic nature of fire in the boreal forests. Large fire years occur during extended periods of drought, which allow naturally ignited fires (i.e. lightning-ignited fires) to burn large areas. Since 1970, the area burned during six episodic fire years in the North American boreal forest was 6.2 million hectares per year, while 1.5 million hectares burned per year in the remaining years. There is evidence that a similar episodic pattern of fire may also exist in the Russia boreal forest.

4 The fire data in the North American boreal forest show a significant increase in the annual area burned over the past three decades, with an average 1.5 million hectares per year burning during the 1970s and 3.2 million hectares per year burning during the 1990s. This increase in burning corresponds to rises of 1.0 to 1.6°C over the same period (Hansen et al, 1996). The projected 2 to 4°C increase in temperature due to projected increases in greenhouse gases during the 21st century may result in high levels of fire activity throughout the world's boreal forests in the future (Stocks et al, 1993).

5 During typical years in the boreal forests, the amounts of biomass consumed during fire ranges between 10 and 20 tonnes per hectare. During the drought years with episodic fires, the amounts of biomass consumed during biomass burning may be as high as 50 to 60 tonnes per hectare. Assuming that biomass is about 50 per cent carbon by mass, such amounts would release 450 to 600Tg C globally. These amounts are considerably higher than the often-quoted value for total carbon released by biomass burning in the world's boreal and temperate forests of 130Tg C globally (Andreae, 1991; see Table 7.2).

Calculation of gaseous and particulate emissions from burning

To assess both the environmental and health impacts of biomass burning, information is needed on the gaseous and particulate emissions produced during the fire and released into the atmosphere. The calculation of gaseous emissions from vegetation and peat fires can be calculated using a form of an expression from Seiler and Crutzen (1980) for each burning ecosystem/terrain:

where M = total mass of vegetation or peat consumed by burning (tonnes), A = area burned (km2), B = biomass loading (tonnes km-2), and E = burning efficiency (dimensionless). The total mass of carbon (M(C)) released to the atmosphere during burning is related to M by the following expression:

where C is the mass percentage of carbon in the biomass. For tropical vegetation, C = 0.45 (Andreae, 1991); for peat, C = 0.50 (Yokelson et al, 1996). The mass of CO2 (M(CO2)) released during the fire is related to M(C) by the following expression:

The combustion efficiency (CE) is the fraction of carbon emitted as CO2 relative to the total carbon compounds released during the fire. For tropical vegetation fires, CE = 0.90 (Andreae, 1991); for peat fires, CE = 0.77 (Yokelson et al, 1997). The biomass load range and the burning efficiency for tropical ecosystems are summarized in Table 7.3.

Once the mass of CO2 produced by burning is known, the mass of any other species, X; (M(XJ), produced by burning and released to the atmosphere can be calculated with knowledge of the CO2-normalized species emission ratio (ER(Xj)). The emission ratio is the ratio of the production of species X; to the production of CO2 in the fire. The mass of species, Xb is related to the mass of CO2 by the following expression:

Table 7.3 Biomass load range and burning efficiency in tropical ecosystems

Vegetation type

Biomass load range (tonnes km-2)

Burning efficiency

Peat'

97,500

0.50

Tropical rainforests2

5000-55,000

0.20

Evergreen forests

5000-10,000

0.30

Plantations

500-10,000

0.40

Dry forests

3000-7000

0.40

Fynbos

2000-4500

0.50

Wetlands

340-1000

0.70

Fertile grasslands

'50-550

0.96

Forest/savanna mosaic

'50-500

0.45

Infertile savannas

'50-500

0.95

Fertile savannas

'50-500

0.95

Infertile grasslands

'50-350

0.96

Shrublands

50-200

0.95

Source: From Scholes et al (1996) except1 Brunig (1977) and Supardi et al (1993); ;

2 Brown and Gaston (1996)

where X; = CO, CH4, NOx, NH3, and O3. It is important to re-emphasize that O3 is not a direct product of biomass burning. However, O3 is produced via photochemical reactions of CO, CH4 and NOx, all of which are produced directly by biomass burning. Hence, the mass of ozone resulting from biomass burning may be calculated by considering the ozone precursor gases produced by biomass burning. Values for emission ratios for tropical forest fires and peat fires are summarized in Table 7.4.

To calculate the total particulate matter (TPM) released from tropical forest fires and peat fires, the following expression is used (Ward, 1990):

where P is the conversion of biomass matter or peat matter to particulate matter during burning. For the burning of tropical vegetation, P = 20 tonnes of TPM per kilotonne of biomass consumed by fire; for peat burning, we assume P = 35 tonnes of TPM per kilotonne of organic soil or peat consumed by fire (Ward, 1990).

Arguably, the major uncertainties in the calculation of gaseous and particulate emissions resulting from fires involve poor or incomplete

Table 7.4 Emission ratios for tropical forest fires and peat fires

Species

Tropical forest fires (%)

Reference

Peat fires (%)

Reference

CO2

90.00

Andreae (1991)

77.05

Yokelson et al (1997)

CO

8.5

Andreae et al (1988)

18.15

Yokelson et al (1997)

CH4

0.32

Blake etal (1996)

1.04

Yokelson etal (1997)

NO,

0.21

Andreae etal (1988)

0.46

Derived from

Yokelson et al (1997)

NH3

0.09

Andreae etal (1988)

1.28

Yokelson etal (1997)

O3

0.48

Andreae etal (1988)

1.04

Derived from

Yokelson et al (1997)

TPM1

20 tkf

Ward (1990)

35tkf

Ward (1990)

Note:1 Total particulate matter emission ratios are in units of t kt-of biomass or peat material consumed by fire).

(tonnes of total particulate matter per kilotonne

Note:1 Total particulate matter emission ratios are in units of t kt-of biomass or peat material consumed by fire).

(tonnes of total particulate matter per kilotonne information about four fire and ecosystem parameters: (1) the area burned (A); (2) the ecosystem or terrain that burned, i.e. forests, grasslands, agricultural lands, peatlands, etc.; (3) the biomass loading (B), i.e. the amount of biomass per unit area of the ecosystem prior to burning; and (4) the fire efficiency (C), i.e. the amount of biomass in the burned ecosystem that was actually consumed by burning.

A case study of biomass burning: The 1997 wildfires in Southeast Asia

Extensive and widespread tropical forest and peat fires swept throughout Kalimantan and Sumatra, Indonesia, between August and December 1997 (Brauer and Hisham-Hishman, 1988; Hamilton et al, 2000). The fires resulted from burning for land clearing and land use change. However, the severe drought conditions resulting from El Niño caused small land-clearing fires to become large uncontrolled wildfires. Based on satellite imagery, it has been estimated that a total of 45,600km2 burned on Kalimantan and Sumatra between August and December 1997 (Liew et al, 1998). The gaseous and particulate emissions produced in these fires and released into the atmosphere reduced atmospheric visibility, impacted the composition and chemistry of the atmosphere, and affected human health. Some of the consequences of the fires in Southeast Asia were: (1) more than 200 million people were exposed to high levels of air pollution and particulates produced during the fires; (2) more than 20 million smoke-related health problems were recorded; (3) fire-related damage cost in excess of US$4 billion; (4) on 26 September 1997, a commercial airliner (Garuda Airlines Airbus 300-B4) crashed in Sumatra due to very poor visibility due to smoke from the fires on landing with 234 passengers killed; (5) on 27 September 1997, two ships collided at sea due to poor visibility in the Strait of Malacca, off the coast of Malaysia, with 29 crew members killed.

International concern about the environmental and health impacts of these fires was great. Three different agencies of the United Nations organized workshops and reports on the environmental and health impacts of these fires: the World Meteorological Organization (WMO) 'Workshop on Regional Transboundary Smoke and Haze in Southeast Asia', Singapore, 2-5 June 1998, the World Health Organization (WHO) 'Health Guidelines for Forest Fires Episodic Events', Lima, Peru, 6-9 October 1998, and the United Nations Environmental Programme (UNEP) Report on 'Wildland Fires and the Environment: A Global Synthesis', published in February 1999 (Levine et al, 1999). The Indonesian fires also formed the basis of an article in National Geographic magazine, entitled 'Indonesia's plague of fire' (Simons, 1998).

Indonesia ranks third, after Brazil and the Democratic Republic of the Congo (formerly Zaire), in its area of tropical forest. Of Indonesia's total land area of 1.9 million km2, current forest cover estimates range from 0.9 to 1.2 million km2, or 48 to 69 per cent of the total. Forests dominate the landscape of Indonesia (Makarim et al, 1998). Large areas of Indonesian forests burned in 1982 and 1983. In Kalimantan alone, the fires burned from 2.4 to 3.6 million ha of forests (Makarim et al, 1998). It is interesting to note that there is an uncertainty of 1.2 million ha.

Liew et al (1998) analysed 766 Satellite Pour l'Observation de la Terre (SPOT) 'quicklook' images with almost complete coverage of Kalimantan and Sumatra from August to December 1997. They estimate the burned area in Kalimantan to be 30,600km2 and the burned area in Sumatra to be 15,000km2, for a total burned area of 45,600km2 (this is equivalent to the combined areas of the states of Rhode Island, Delaware, Connecticut and New Jersey, in the US). The estimate of Liew et al (1998) represents only a lower limit estimate of the area burned in Southeast Asia in 1997, since the SPOT data only covered Kalimantan and Sumatra and did not include fires on the other Indonesian islands of Irian Jaya, Sulawesi, Java, Sumbawa, Komodo, Flores, Sumba, Timor and Wetar, or the fires in the neighbouring countries of Malaysia and Brunei.

What is the nature ofthe ecosystem/terrain that burned in Kalimantan and Sumatra? In October 1997, NOAA satellite monitoring produced the following distribution of fire hot spots in Indonesia (UNDAC, 1998): agricultural and plantation areas: 45.95 per cent; bush and peat soil areas: 24.27 per cent; productive forests: 15.49 per cent; timber estate areas: 8.51 per cent; protected areas: 4.58 per cent; and transmigration sites: 1.20 per cent (the three forest/timber areas add up to a total of 28.58 per cent of the area burned). While the distribution of fire hot spots is not an actual index for area burned, the NOAA satellite-derived hot spot distribution is quite similar to the ecosystem/terrain distribution of burned area deduced by Liew et al (1998) based on SPOT images of the actual burned areas: agricultural and plantation areas: 50 per cent; forests and bushes: 30 per cent; and peat swamp forests: 20 per cent. Since the estimates of burned ecosystem/terrain of Liew et al (1998) are based on actual SPOT images of the burned area, their estimates were adopted in our calculations.

What is the biomass loading for the three terrain classifications identified by Liew et al (1998)?

Values for biomass loading or fuel load for various tropical ecosystems are summarized in Table 7.3. The biomass loading for tropical forests in Southeast Asia ranges from 5000 to 55,000 tonnes km-2, with a mean value of 23,000 tonnes km-2 (Brown and Gaston, 1996). However, in our calculations we have used a value of 10,000 tonnes km-2 to be conservative. The biomass loading for agricultural and plantation areas (mainly rubber trees and oil palms) of 5000 tonnes km-2 is also a conservative value (Liew et al, 1998). Nichol (1997) has investigated the peat deposits of Kalimantan and Sumatra and used a biomass loading value of 97,500 tonnes km-2 (Supardi et al, 1993) for the dry peat deposits 1.5m thick as representative of the Indonesian peat in her study. Brunig (1997) gives a similar value for peat biomass loading.

The combustion efficiency for forests is estimated at 0.20 and for peat is estimated at 0.50 (Levine and Cofer, 2000). Values for emission ratios for tropical forest fires and peat fires are summarized in Table 7.4. Inspection of Table 7.4 indicates that the emission ratio of CH4 from the burning of underground peat is three times larger than the emission ratio of CH4 from burning of above-ground vegetation. Based on the discussions presented in this section, the values for burned area, biomass loading and combustion efficiency used in the calculations are summarized in Table 7.5.

Table 7.5 Parameters used in calculations

1 Total area burned in Kalimantan and Sumatra, Indonesia

in 1997:45,600km2

2 Distribution of burned areas, biomass loading and combustion efficiency

A Agricultural and plantation areas 50%

5000 tonnes km-2

0.20

B Forests and bushes 30%

10,000 tonnes km-2

0.20

C Peat swamp forests 20%

97,500 tonnes km-2

0.50

Results ofcalculations: Gaseous and particulate emissions from the fires in Kalimantan and Sumatra, Indonesia, August to December 1997

The calculated gaseous and particulate emissions from the fires in Kalimantan and Sumatra, from August to December 1997, are summarized in Table 7.6 (Levine, 1999) (it is important to keep in mind that wildfires continued throughout Southeast Asia from January through to April 1998 and that the fires covered much more of the region than Kalimantan and Sumatra).

Table 7.6 Gaseous and particulate emissions from the fires in Kalimantan and Sumatra in 1997

Agricultural/

Forest fire

Peat fire

Total fire

plantation fire

emissions

emissions

emissions

emissions

CO2

9.234

11.080

171.170

191.485

(4.617-13.851)

(5.54-16.62)

(85.585-256.755)

(95.742-287.226)

CO

0.785

0.942

31.067

32.794

(0.392-1.177)

(0.471-1.413)

(15.533-46.600)

(16.397-49.191)

CH4

0.030

0.035

1.780

1.845

(0.015-0.045)

(0.017-0.052)

(0.89-2.67)

(0.922-2.767)

NOx

0.023

0.027

0.921

0.971

(0.011-0.034)

(0.013-0.040)

(0.460-1.381)

(0.485-1.456)

NH3

0.010

0.012

2.563

2.585

(0.005-0.01 5)

(0.006-0.018)

(1.281-3.844)

(1.292-3.877)

O3

0.177

0.213

6.710

7.100

(0.088-0.265)

(0.106-0.319)

(3.35-10.06)

(3.55-10.65)

TPM

0.460

0.547

15.561

16.568

(0.23-0.69)

(0.273-0.820)

(7.780-23.341)

(8.284-24.852)

Note: For total burned area = 45,600km2. For each species, the best estimate emission value is on first line and the range of emission values in parenthesis under best guess (see text for discussion of emission estimate range and uncertainty calculations). Units of emissions: million metric tonnes (Mt) of C for CO2, CO, and CH4; Mt of N for NOx and NH3; Mt of O3 for O3; MtC of particulates; 1Mt = 1012 grams = 1Tg. Source: Levine (1999)

Note: For total burned area = 45,600km2. For each species, the best estimate emission value is on first line and the range of emission values in parenthesis under best guess (see text for discussion of emission estimate range and uncertainty calculations). Units of emissions: million metric tonnes (Mt) of C for CO2, CO, and CH4; Mt of N for NOx and NH3; Mt of O3 for O3; MtC of particulates; 1Mt = 1012 grams = 1Tg. Source: Levine (1999)

For each of the seven species listed, the emissions due to agricultural/plantation burning (A), forest burning (F), and peat burning (P) are given. The total (T) of all three components (A+F+P) is also given. The 'best estimate' total emissions are: CO2: 191.485 million Mt of C (Tg C); CO: 32.794Tg C; CH 1.845Tg C; NOx: 5.898Tg N; NH3: 2.585Tg N; O3: 7.100Tg O3; and total particulate matter: 16.154Tg C.

The CO2 emissions from these fires are about 2.2 per cent of the global annual net emission of CO2 from all sources (see Table 7.2 for global annual production of CO2, which is ~8700Tg C). The percentage for other gases produced by these fires compared to the global annual production from all sources is: CO: 2.98 per cent; CH4: 0.48 per cent; oxides of nitrogen: 2.43 per cent; ammonia: 5.87 per cent; and TPM: 1.08 per cent.

However, it is important to re-emphasize that these emission calculations represent lower limit values since the calculations are only based on burning in Kalimantan and Sumatra in 1997. The calculations do not include burning in Java, Sulawesi, Irian Jaya, Sumbawa, Komodo, Flores, Sumba, Timor and Wetar in Indonesia or in neighbouring Malaysia and Brunei.

It is interesting to compare the gaseous and particulate emissions from the 1997 Kalimantan and Sumatra fires with those from the Kuwait oil fires of 1991, described as a 'major environmental catastrophe'. Laursen et al (1992) have calculated the emissions of CO2, CO, CH4, NOx and particulates from the Kuwait oil fires in units of Mt per day. The Laursen et al (1992) calculations are summarized in Table 7.7. To compare these calculations with the calculations for Kalimantan and Sumatra (Table 7.7), we have normalized our calculations by the total number of days of burning. The SPOT images (Liew et al, 1998) covered a period of five months (August-December 1997) or about 150 days. For comparison with the Kuwait fire emissions, we divided our calculated emissions by 150 days. The gaseous and particulate emissions from the fires in Kalimantan and Sumatra significantly exceeded the emissions from the Kuwait oil fires. The 1997 fires in Kalimantan and Sumatra were evidently a significant source of gaseous and particulate emissions to the local, regional and global atmospheres.

Table 7.7 Comparison of gaseous and particulate emissions: The Indonesian fires and the Kuwait oil fires

Species Indonesian fires Kuwait oil fires

Table 7.7 Comparison of gaseous and particulate emissions: The Indonesian fires and the Kuwait oil fires

Species Indonesian fires Kuwait oil fires

CÜ2

1.28 X

: 106

5.0 X

: 105

CO

2.19 X

: 105

4.4 X

: 103

CH4

1.23 X

: 104

1.5 X

: 103

NOx

6.19 X

: 103

2.0 X

: 102

Particulates

1.08 X

: 105

1.2 X

: 104

Note: Units of emissions: Mt per day of C for CO2, CO and CH4; Mt per day of N for NOx; Mt per day for particulates Source: Data on Kuwait oil fires from Laursen et al (1992)

Note: Units of emissions: Mt per day of C for CO2, CO and CH4; Mt per day of N for NOx; Mt per day for particulates Source: Data on Kuwait oil fires from Laursen et al (1992)

Was this article helpful?

0 0

Post a comment