The Bor Forest Island Fire Experiment
Fire Research Campaign Asia-North (FIRESCAN)
I. Fire in Ecosystems of Boreal Eurasia
The worlds total boreal forests and other wooded land within the boreal zone cover 1.2×109 ha of which 920×106 ha are closed forest. The latter number corresponds to ca. 29% of the worlds total forest area and to 73% of its coniferous forest area (ECE/FAO 1985). About 800×106 ha of boreal forests with a total growing stock (over bark) of ca. 95 billion m3 are exploitable (41% and 45% respectively of the world total). The export value of forest products from boreal forests is ca. 47% of the world total (Kuusela 1990, 1992).
The vast majority of the boreal forest lands (taiga) of Eurasia are included in the Russian Forest Fund, covering ca. 900×106 ha. Depending on the criteria used to define “boreal forest”, the area of closed boreal forest in the Russian Federation varies from 400 to 600×106 ha (Pisarenko and Strakhov 1993). These numbers correspond to a 43-65% share of the worlds closed boreal forest.
1. Disturbances in Transition: Natural to Anthropogenic
Among natural disturbances fire (lightning fire) is the most important factor controlling forest age structure, species composition and physiognomy, shaping landscape diversity, and influencing energy flows and biogeochemical cycles, particularly the global carbon cycle, since prehistoric times (cf. monographs and synopses e.g. by Sofronov 1967; Slaughter et al. 1971; Zackrisson 1977; Sherbakov 1979; Viereck and Schandelmeier 1980; Alexander and Euler 1981; Heinselman 1981; Wein and MacLean 1983; Kurbatsky 1985; Johnson 1992; Sannikov 1992; Furyaev 1994; Shugart et al. 1992; Goldammer and Furyaev 1996). Small and large fires of varying intensity have different effects on the ecosystem. High-intensity fires lead to the replacement of forest stands by new successional sequences. Low-intensity surface fires favor the selection of fire-tolerant trees such as pines (Pinus spp.) and larches (Larix spp.) and may repeatedly occur within the lifespan of a forest stand without eliminating it.
Large-scale forest disturbances connected with drought and fires are familiar from recent history. The Tunguska Meteorite Fall near Yenisseisk (ca. 60° 54’N-101° 57’E) on 30 June 1908, a cometary nucleus explosion at ca. 5 km altitude, was one of the more exceptional events which caused large-scale forest fires in the region of impact (see Grishin 1996).
Several years later, from June to August 1915, the largest fires ever recorded occurred as a consequence of an extended drought in Central and East Siberia (Tobolsk, Tomsk, Yeniseisk, NE Irkutsk, S Yakutsk regions). Shostakovich (1925) estimated that the fires were burning ca. 50 days in the region between 52-70° N and 69-112° E. The main center of fires was between Angara River and Nijnya Tunguska, and the total area burned was estimated at 14.2×106 ha. However, the smoke of these fires covered the region between 64-72° N and 61-133° E, corresponding to ca. 680×106 ha. Shostakovich estimated continuous smoke (visibility ca. 100 m) on 284×106 ha, heavy smoke (visibility 25-100 m) on 215×106 ha and thick smoke (visibility 5-20 m) on ca. 181×106 ha.
It is not clear, however, whether lightning, humans or a combination of the two were the primary cause of the extended fires of 1915. In Eurasia fire has been for a long time an important tool for land clearing (conversion of boreal forest), silviculture (site preparation and improvement, species selection), and in maintaining agricultural systems, e.g. hunting societies, swidden agriculture, and pastoralism (Viro 1969; Pyne 1996). In addition to the natural fires, these old cultural practices brought a tremendous amount of fire into the boreal landscapes of Eurasia. In the early 20th century, the intensity of fire use in the agricultural sector began to decrease because most of the deforestation had been accomplished for agriculture, and traditional small-sized fire systems (treatment of vegetation by free burning) was replaced by mechanized systems (use of fossil-fuel driven mechanical equipment). Despite the loss of traditional burning practices, however, humans are still the major source of wildland fires; only 15% of the recorded fires in the Russian Federation are caused by lightning (Korovin 1996).
In recent years wildfires were more or less eliminated in Western Eurasia. The average annual area affected by fire in Norway, Sweden and Finland is less than 4,000 ha. Thus, the major occurrence of Eurasian fires is in the territory of the Russian Federation and other countries of the Commonwealth of Independent States. Statistics compiled by the Russian Aerial Fire Protection Service Avialesookhrana show that between 10,000 and more than 30,000 forest fires occur each year, affecting up to 2-3×106 ha of forest and other land (Korovin 1996). Since fires are monitored (and controlled) only on protected forest and pasture lands, it is estimated that the real area affected by fire in Eurasia’s boreal vegetation is much higher. For instance, satellite-derived observations by Cahoon et al. (1994) indicate that during the 1987 fire season approximately 14.5×106 ha were burned. In the same fire season ca. 1.3×106 ha of forests were affected by fire in the montane-boreal forests of Northeast China, south of the Amur (Heilongjiang) River (Goldammer and Di 1990; Cahoon et al. 1991). Fires in boreal North America in the past decade affected, on average, 1-5 x 106 ha per year. An exceptional year was 1987 in which 7.4 x 106 of forests were burned in Canada (FIRESCAN Science Team 1994).
2. Concerns: Global Change and Fire
Expected global warming over the next 30-50 years, as predicted by Global Circulation Models, will be most evident in the northern circumpolar regions (Bolin et al. 1986; Maxwell 1992; Shugart and Smith 1992; Shugart et al. 1992). As Wein and de Groot (1996), Fosberg (1996), Stocks (1993), and Stocks and Lynham (1996) underscore, fire may be the most important (widespread) driving force in changing the taiga under climatic warming conditions. The prediction of increasing occurrence of extreme droughts in a 2xCO2 climate indicates that fire regimes will undergo considerable changes. Increasing length of the fire season will lead to a higher occurrence of large, high-intensity wildfires. Such fire scenarios may be restricted to a transition period until a new climate-vegetation-fire equilibrium is established.
Regional warming may also lead to the shift of vegetation zones, e.g. the boreal forest shifting north ca. 500-1000 km (Kauppi and Posch 1988). The shift of ecosystems will have considerable impacts on the distribution of phytomass. Estimates of carbon stored in above- and below-ground live and dead plant biomass (without soil organic matter) in the global boreal forest area range between 66 and 98 Gt (66-98×1015 g) (US Department of Energy 1983; Apps et al. 1993). Additional large amounts of carbon are stored in boreal forest soils (ca. 200×1015 g) and boreal peatlands (ca. 420×1015 g) (Apps et al. 1993). There is concern that changing fire regimes due to climate change will affect the balance of the boreal carbon pool and lead to additional release of carbon into the atmosphere, thus acting as temporary positive feedback loop to global warming.
Changing forestry practices in boreal Eurasia, stimulated by increasing national and international demands for boreal forest products, have resulted in the widespread use of heavy machinery, large-scale clearcuts, and, with this, in the alteration of fuel complexes. The opening of formerly closed remote forests by roads, and subsequent human interferences bring new ignition risks. Additional fire hazards with little predictable environmental consequences, are created on forest lands heavily damaged by industrial emissions (severe damages in the Russian Federation affect ca. 9×106 ha). Radioactive contamination on an area of ca. 7×106 ha creates considerable problems because it redistributes radionuclides through forest fires (Dusha-Gudym 1992). These direct effects on the ecosystem are added to the indirect effects of climate change, and both will almost certainly lead to an unprecedented era of fire.
3. Objectives of Cooperative Fire Research in Boreal Eurasia
Jointly with the first East-West conference entitled “Fire in Ecosystems of Boreal Eurasia” (Goldammer and Furyaev 1996), the Fire Research Campaign Asia-North (FIRESCAN) was prepared under the co-sponsorship of the International Boreal Forest Research Association (IBFRA) and the IGBP/IGAC subprogram “Impact of Biomass Burning on the World Atmosphere” (Biomass Burning Experiment [BIBEX]; for details cf. FIRESCAN Science Team (1994) and Goldammer and Furyaev (1996)).
In accordance with the hypotheses of the IBFRA Stand Replacement Fire Working Group (Fosberg 1992), the objectives of the experiment were:
set a high-intensity stand replacement fire under controlled conditions, under conditions and with characteristics of an uncontrolled wildfire;
investigate all pre- and postfire characteristics of the site;
describe fire behavior and relate the findings to ecological and meteorological conditions before and during the fire;
analyze emissions of aerosols (characteristics and transport), the most important radiatively active trace gases, and trace gases with stratospheric ozone-depleting effects;
relate the fire experiment to the fire history of the site and the surrounding landscape;
set up an investigation area for long-term follow up research on ecosystem response (e.g. collection of data on mortality and recovery, succession, biological diversity nutrient cycling, soil respiration, and carbon accretion);
demonstrate and compare methodologies in fire research developed in the East and West.
To meet these objectives, the FIRESCAN Science Team, an international multidisciplinary research team, assembled in the summer of 1993 to investigate site characteristics, fire effects, fire emissions, and fire behavior on a 50 ha experimental stand-replacement fire in a typical boreal pine forest.
The experimental site is in the central part of the Krasnoyarsk Region of Siberia, about 28 km west of the Yenesei River and 28 km south of the Dubches River (60° 45’N, 89° 25’E) at an elevation of approximately 150 m above sea level (Figure1). The study site is a nearly level, slightly elevated, sandy island, about 50 ha in size, which is surrounded by bogs dominated by mixed-grass, sphagnum and tall sedge (Figure 2). The site was referred to as Bor Forest Island, after the town of Bor, 90 km to the North, which served as the transportation base for research activities. The Bor Forest Island study site is on the Sym Plain, in the Western Siberian Lowland – a large block of the earth’s crust characterized by past tectonic depression. The Sym Plain is an area of low relief, with sandy surface materials of glacial outwash and alluvial origin. Very deep, unconsolidated deposits are present, and there are numerous lakes and oligotrophic and mesotrophic bogs. Forests are dominated by pure pine stands of the Pinus sylvestris-Ledum-Vaccinium vitis idaea-Pleurozium schreberi, P.sylvestris-P.schreberi-Cladonia sylvatica (40%), and Pinus sylvestris-Polytrichum commune-dwarf shrub-Sphagnum (20%) forest types. Oligotrophic bog ridges with pools covered by P.sylvestris-dwarf shrub-Sphagnum forest cover 40% of the landscape. The forest on the experimental fire site is a typical middle taiga pine forest of the Sym Plain.
Fig.1. Location of Bor Forest Island near Bor, Krasnoyarsk Region, Russian Federation. All other names of locations are regional headquarters of the Russian Aerial Fire Protection Service Avialesookhrana or relevant services in other countries of the Commonwealth of Independent States.
Fig.2. Aerial view of Bor Forest Island immediately prior to the experimental fire.
Because Atlantic air masses are transformed to continental over the Western Siberian Lowland, zones and subzones are clearly discernible across the landscape. The climate is cool and moist. Average annual air temperature ranges from 3.2 to 5.7° C. Total annual precipitation is 450-500 mm, with wide year-to-year variations. Although most precipitation occurs in the summer, frequent dry periods are caused by dry cyclonic air masses coming from the south. In the past century, 26 droughts have occurred in the area (an average of 2-3 times per decade).
The fire season lasts from May to September, with most fires in June-July. In the Pinus sylvestris-lichen forest types, about 20% of the area is in even-aged stands that have regenerated from stand-replacement fires. Fire periodicity varies from 40-50 years in the north to 25-29 years in the central part of Krasnoyarsk Territory. For the P.sylvestris-V.vitis idaea-Sphagnum forest type characteristic of the central part of the area, forest fire periodicity is 10 to 80 years. As in the rest of Siberia, periodic extreme fire seasons are common. Those seasons are remarkable for long rainless periods (up to 38 days), with relative humidities down to 30%, and air temperatures up to 30-35° C. Until the end of the 19th century, extreme fire seasons in the central Krasnoyarsk Territory occurred from 3-4 to 7 times a century. This has increased up to 20-25 events in the 20th century. Most of these fires are human caused, as a result of intense forest exploitation in the area. For the past 50 years, extreme fire seasons associated with mass forest fires have occurred at least twice a decade, sometimes two years in succession (for more details on geology, climate and ecology of the region cf. Goldammer and Furyaev (1996).
Results of the first two years of investigation from the Bor Forest Island Fire Experiment, including pre- and post-fire studies, are given in the following two sections. Part II reports characteristics of the study site and its vegetation, and presents preliminary results on short-term fire effects. Part III describes fuel characteristics, fire behaviour, and emissions.