Effects of Fire on Soil Nutrients

Fires increase soil nutrient availability through three primary mechanisms: (1) nutrients added to the soil as ash, (2) heating of soil organic matter, and (3) increased rates of biological mineralization due to increases in soil pH, temperature, and moisture, and a reduction in C:N ratio (Wright and Bailey 1982, Pritchett and Fisher 1987). The degree of increase in nutrient availability after fires depends largely on fire intensity. Most studies of low- to moderate- intensity fires report increases in available nutrients (reviews by Dunn and others 1977, DeBano and others 1977, Wells and others 1979, Humphreys and Craig 1981, Wright and Bailey 1982, Hungerford and others 1990, Neary and others 1999). In contrast, intense fires may cause a net loss of nutrients (DeBano and others 1977, Giovannini and others 1990). Fires in forested systems generally decrease total ecosystem nitrogen (Neary and others 1999). Due to nitrogens low temperature of volatilization (200^o C; Weast 1988), nitrogen loss is linked with the consumption of organic matter (for example, see Dunn and others 1977). Where fuels are completely consumed and the surface layer of soil organic matter is destroyed, loss of nitrogen through volatilization can be substantial (Nye and Greenland 1964, Ewel and others 1981). Volatilization of phosphorus and cations is usually minor due to the high volatilization temperatures of these minerals (>760^o C; Weast 1988). Their loss from severely burned sites may be caused by surface erosion, leaching, or transport of ash (Wright and Bailey 1982). For example, annual and periodic burning on the Highland Rim in Tennessee had no effect on exchangeable phosphorus but did reduce soil potassium (Thor and Nichols 1974).

Nitrogen losses during and after prescribed fire in southern forests range from 20 kg ha^-1 for low-intensity understory burns (Kodama and Van Lear 1980) to >400 kg ha^-1 for high-intensity site preparation burns in heavy fuels (Vose and Swank 1993). After fires, the greatest proportions of C and N often are found on or within the forest floor. For example, Clinton and others (1996) found that 2 years after a fell-and-burn treatment in the Nantahala National Forest, aboveground C and N pools were well below pretreatment levels but forest floor N was 90 percent of its pretreatment amount. Hence, organic soil layers are an important source of available nutrients after burns, and rapid plant regrowth after fire can be aid in retention of N and other site nutrients (Boring and others 1988, Van Lear and others 1990).

Burning techniques and burning conditions have been shown to greatly influence the loss of organic matter and, subsequently, loss of N. For example, Clinton and others (1998) found that an understory burn in a mixed pine-hardwood stand in North Carolina consumed about 40 percent more humus mass than a fell-and-burn treatment would have. Although understory burns may potentially consume more humus than other techniques, understory burns have been shown to only remove part of the forest floor in long-term burning studies in South Carolina (Metz and others 1961) and in Virginia (Romancier 1960). Even high-intensity broadcast burns generally leave portions of the forest floor intact, due to the heterogeneity of burns and low consumption of large woody debris.

Vose and others (1999) evaluated the effects of a BROKEN-LINK stand replacement burn used to restore a pine-hardwood forest in the Wine Spring Creek, North Carolina. Impacts on the components of the biogeochemical cycle were minimal, especially when compared with the alternative fell-and-burn treatment. Nitrogen losses during the stand replacement burn were confined to where fire temperature were highest on a ridge, but losses were small enough (78 kg N ha^-1) to be rapidly replenished by atmospheric inputs and N fixation. Moreover, soils and streams showed no response to the burn. The authors concluded that the effects of the stand replacement burn were limited to the forest floor.

Some studies report an increase in soil C and N after fire. For example, Groeschl and others (1991) reported increased total carbon and nitrogen levels in the surface 10 cm (4 inches) of mineral soil following low-intensity prescribed burns in a fire-dependent Table Mountain pine-pitch pine community of the Shenandoah National Park in Virginia. These authors also reported increased mineral soil inorganic nitrogen levels in burned areas; these increases are conducive to increased plant uptake. Increases in soil C and N are often attributed to unburned slash fragments being incorporated into soils after fire (Buckner and Turrill 1999).

Although fires sometimes decrease total N, available forms of N usually increase after fires for a number of reasons. Initial increases in soil-extractable NH₄ concentrations are generally attributed to the downward movement of volatilized N and its subsequent condensation in the cool soil layers below. Nitrifying microorganisms are also particularly sensitive to increases in soil temperature, and heating the soil can decimate microbial populations at the soil surface, thereby increasing the availability of carbon and soil nutrients. Following burns, increased mineralization rates are favored by increased soil temperature, moisture, and pH. In pine-hardwood ecosystems in the southern Appalachians, Knoepp and Swank (1993) found that the fell-and-burn technique increased available soil N (NH₄) but caused little movement of dissolved inorganic N off site during the first year after burning.

Increased N mineralization rates can offset total N losses from fire. However, the degree to which N losses are ameliorated depends on the magnitude and source of lost N (forest floor vs. wood) and on subsequent N recovery. In southern Appalachian hardwood ecosystems, forest floor losses are particularly important because release of N from the forest floor provides approximately 50 pecent of the total available N (Monk and Day 1988, Clinton and others 1996).

Nitrogen inputs from precipitation or N-fixers may also offset losses from fire. Nitrogen inputs from precipitation approximating 5 pounds per acre per year have been measured in the southern Appalachians (Swank and Douglass 1977) and in the upper Piedmont (Van Lear and others 1983). Early stages of plant succession are often dominated by nitrogen-fixing species, especially in ecosystems with a high fire frequency (Gorham and others 1979). For example, Boring and Swank (1984) reported that 4-year-old stands of black locust fixed about 30 pounds of N per acre per year in the southern Appalachians (Van Lear and Waldrop 1988).