Relationships of fire, biomass and nutrient dynamics along a vegetation gradient in the Brazilian Cerrado

1 The variable structure and mass of vegetation, nutrient pools, and responses to fire were quantified along a common vegetation gradient from grassland (campo limpo) to open evergreen woodland (cerrado sensu stricto) in the Brazilian Cerrado. 2 The biomass of the fuel loads (i.e. that proportion of the above-ground biomass susceptible to combustion) ranged from 7128 kg ha-1 in campo limpo to 10031 kg ha-' in cerrado sensu stricto. Grasses comprised 91-94% of the total above-ground biomass in the campo limpo and campo sujo grasslands. In the campo cerrado and cerrado sensu stricto communities, graminoids comprised only 27% of the fuel load biomass; the remainder was composed of deadwood (18%), dicot leaf litter (36%), dicots and shrub leaves (18%). 3 Fires consumed ?97% of the above-ground biomass in the grasslands, but only 72 and 84% of the fuel load in campo cerrado and cerrado sensu stricto, respectively. The total residual mass (uncombusted fuels and ash combined) was < 8% of the prefire mass in campo limpo and campo sujo but was 2 36% in the campo cerrado and cerrado sensu stricto. A circular relationship existed between vegetation composition, fuel composition, and fire behaviour. For example, fire-line intensity was significantly greater in grasslands, which would facilitate grass dominance by increasing above-ground tissue damage of trees and shrubs. (4) Within the fuel load, a consistently increasing pool size of N, P, C, and S was measured along the gradient from campo limpo to cerrado sensu stricto. Total mass of N increased from 24 to 55 kg ha-', total mass of P increased from 1.7 to 3.5 kg ha-' , and total mass of C increased from 3389 to 4657 kg ha-'. This trend was reversed for K (i.e. 18.5 kg ha-' in Campo limpo to 13.8 kg ha-' in cerrado sensu stricto). 5 Carbon, N, and S were the nutrients that were consistently lost in the highest quantities during fire. Losses of P were intermediate, and K and Ca were negligible. 6 The percentage of the above-ground N, C, and S pools that were lost by fire decreased along the vegetation gradient from campo limpo to cerrado sensu stricto. For example, 2 90% of the N pool was lost by fire in the grasslands while < 56% of the N pool was lost in the tree-dominated communities. Total site loss (kg ha-') of all nutrients was not significantly different among the vegetation types. However, significantly greater quantities of nutrients were lost as particulates in Cerrado sensu stricto while significantly greater quantities of nutrients were volatilized during fire in Campo limpo. 7 Nitrogen pools located in soils to a 10 cm depth ranged from 656 to 1670 kg ha-'. Of the total N pool consisting of fuel loads and soils, the proportion in fuels ranged from 1.4% in campo limpo to 4.0% in cerrado sensu stricto. Fires resulted in ecosystem losses of < 3.8% of this pool. Ecosystem losses of this magnitude are likely to be compensated for by natural nutrient inputs in one to three years.

[1]  J. Kauffman,et al.  Fire in the Venezuelan Amazon 2: Environmental conditions necessary for forest fires in the evergreen rainforest of Venezuela , 1988 .

[2]  R. Goodland A Physiognomic Analysis of the `Cerrado' Vegetation of Central Brasil , 1971 .

[3]  D. Ward,et al.  Smoke emissions from wildland fires , 1991 .

[4]  P. Crutzen,et al.  Biomass Burning in the Tropics: Impact on Atmospheric Chemistry and Biogeochemical Cycles , 1990, Science.

[5]  Van Wagner The Line Intersect Method in Forest Fuel Sampling , 1968 .

[6]  J. Walker,et al.  Fire in Australian Tropical Savannas , 1982 .

[7]  W. M. Haynes CRC Handbook of Chemistry and Physics , 1990 .

[8]  J. Kauffman,et al.  Deforestation, Fire Susceptibility, and Potential Tree Responses to Fire in the Eastern Amazon , 1990 .

[9]  B. E. Davies,et al.  Loss-on-Ignition as an Estimate of Soil Organic Matter1 , 1974 .

[10]  V. Pivello,et al.  Transfer of macro-nutrients to the atmosphere during experimental burnings in an open cerrado (Brazilian savanna) , 1992, Journal of Tropical Ecology.

[11]  E. J. Fittkau,et al.  Biomass and Structure in a Central Amazonian Rain Forest , 1975 .

[12]  C. E. Van Wagner,et al.  Height of Crown Scorch in Forest Fires , 1973 .

[13]  J. M. Bremner,et al.  A Simple Turbidimetric Method of Determining Total Sulfur in Plant Materials1 , 1970 .

[14]  R. Sanford,et al.  Seasonally Dry Tropical Forests: Nutrient cycling in tropical deciduous forests , 1995 .

[15]  Martin E. Alexander,et al.  Calculating and interpreting forest fire intensities , 1982 .

[16]  I. Salcedo,et al.  Biomass and nutrient dynamics associated with slash fires in neotropical dry forests , 1993 .

[17]  J. Levine Biomass Burning in West African Savannas , 1991 .

[18]  M. Lamotte The Structure and Function of a Tropical Savannah Ecosystem , 1975 .

[19]  S. R. Olsen,et al.  Test of an Ascorbic Acid Method for Determining Phosphorus in Water and NaHCO3 Extracts from Soil , 1965 .

[20]  Kenneth P. Davis Forest Fire: Control and Use , 1959 .

[21]  Robert L. Sanford,et al.  Nutrient Cycling in Moist Tropical Forest , 1986 .

[22]  L. Coutinho,et al.  Fire in the Ecology of the Brazilian Cerrado , 1990 .

[23]  R. J. Raison,et al.  Mechanisms of element transfer to the atmosphere during vegetation fires , 1985 .

[24]  P. Crutzen,et al.  Estimates of Annual and Regional Releases of CO2 and Other Trace Gases to the Atmosphere from Fires in the Tropics, Based on the FAO Statistics for the Period 1975–1980 , 1990 .

[25]  A. Page Methods of soil analysis. Part 2. Chemical and microbiological properties. , 1982 .

[26]  L. Coutinho,et al.  Ecological Effects of Fire in Brazilian Cerrado , 1982 .

[27]  J. Ewel,et al.  Slash and burn impacts on a Costa Rican wet forest site , 1981 .