Chapter 2

Fire in the Fossil Record: Recognition

2.1 Fire Proxies: Fire Scars and Charcoal

While fire may be considered a destructive force, there are a number of ways through which fire history may be interpreted. In the more recent past, an increasingly important method has been to study fire scars as part of a tree ring analysis (Figure 2.1). When a surface fire passes through a forest, the outer part of the trunk may be partially burnt or destroyed in part, but not sufficiently to kill the tree. The tree then will resume its normal growth. In temperate environments where trees exhibit growth rings, this may be very useful. If a tree is felled and a cross-section of a tree made, then the age of the tree may be calculated through counting the rings. However, it is possible to calculate the age of any fire affecting the tree through the occurrence of a fire scar. A single tree may have many scars and, hence, a fire return interval may also be calculated.

Figure 2.1 Tree rings and fire scars.

The occurrence of fire scars provides researchers with three important pieces of data (Roos and Swetnam, 2012):

The spatial extent of a fire can be considered at two main scales. If trees are studied within a forested area - perhaps a few square kilometres - then the size of an individual fire might be calculated. If trees are studied over a much larger region - perhaps over hundreds or thousands of square kilometres - then it is possible to interpret high and low fire years (Figure 2.2). Such an approach has been found useful in the study of the relationship between fire and climate (Figure 2.3).

Figure 2.2 Fire scars across an area; south-western US fire scar network. This is a composite of fire scar chronologies from 55 forest and woodland sites in Arizona, New Mexico and northern Mexico, AD 1600 to Present. (From Swetnam et al., 1999). Reproduced by permission of Elsevier.

Figure 2.3 Fire scars and fire history; LMF-3 fire scar sample from Limestones Flats, AZ, USA. 42 fire scars on this one tree. (From Dieterich and Swetnam, 1984). Reproduced by permission of the Society of American Foresters.

In climate studies, the fire scars and tree ring data can be used to link, for example, the occurrence of fire in relation to El Niño and La Niña. This relationship has been proven to be very strong, with more fires occurring in the south-west of the USA during the La Niña years (Swetnam and Betancourt, 1990). In addition, fire records can also be combined with a range of climate records, so that it is possible to demonstrate that more fires occur in warmer years. This relationship between increasing temperature and increasing fire is of particular significance and concern. Recent studies in the western USA have demonstrated an increasing occurrence of larger fires with increasing temperature (Figure 1.27). It should be noted that the effects of El Niño and La Niña would be different in different regions.

Tree-ring fire scar analysis is a very powerful tool in unravelling climate and human impacts. In a detailed study of fires in forests in the south-western USA, using data from trees up to 1500 years old, Roos and Swetnam (2012) were able to show that recent mega-fires in the region were truly unusual. The data from the tree rings provides important information about the climate, and particularly about the occurrence of wet and dry years.

The fire scar data has been used to test whether today's hot-dry climate alone is causing mega-fires that destroy large tracts of forest. The researchers compared fire frequency during hotter and drier periods with that during cooler and wetter periods. Interestingly, they showed that the fire frequencies were similar. However, over the past 100 years, they noted a reduction in fire as a result of human fire suppression.

The historic data has demonstrated that frequent surface fires were important in keeping fuel loads low. The fires were predominantly low-temperature surface fires. The suppression of fire in such forests, however, leads to a build-up of fuel so that, during the increasingly warm arid phase of climate, any fire becomes a devastating high-temperature crown fire that destroys the forest (Figure 2.3).

Many tree-ring/fire scar studies are limited to data from less than 1500 years. In some vegetation types, much longer fire records can be obtained. In the Sequoia redwood forests of north-west USA, fire scar records have been obtained from trees more than 3000 years old (Swetnam, 1993). In addition, a link can be made between data from fire scars and data from charcoal in varved lake sediments (Figure 2.4). In the lake records, there may be less precision of exact fire years, as some charcoal may be reworked. The linkage of fire scar data and charcoal data has proven very important in the interpretation of local and regional fire events (Swetnam et al., 2009).

Figure 2.4 Types of fire data from maps to fire scars to charcoal in varved lake sediments. (from M. Power). Courtesy H. D. Grissino-Mayer.

2.2 The Problem of Nomenclature: Black Carbon, Char, Charcoal, Soot and Elemental Carbon

As discussed in Chapter 1, a range of products is formed as a result of a fire, and one significant problem is that of nomenclature. When plant material is subjected to fire, it undergoes both pyrolysis and combustion. During the initial heat increase, pyrolysis takes place in the absence of oxygen and, for example, wood may be converted to charcoal. This, however, is not a single stage process. The process of charcoalification has been called 'carbonization' by some authors. This may cause some misunderstanding, as this term has also been used to describe the thermal upgrading of coal.

Some authors prefer to refer to a combustion continuum model (Figure 2.5). This considers a continuum ranging from slightly chared biomass through to soot (see Hammes and Abiven (2013), for example).

Figure 2.5 Plate of SEM images - soot to charcoal.

This concept is basically a chemical one, reflecting an increase in aromaticity as the charring temperature increases. Plant material may be charred. Some authors distinguish a low temperature char to a higher temperature charcoal, but there is disagreement as to the identification of this transition. A significant problem results from the different methods used to study such material. Geochemists often refer to the spectrum of char to soot as black carbon (BC), but concentrations of such material in sediments may differ considerably, depending on the methods used for their detection (Hammes and Abiven, 2013). In some cases, discrepancies are a result of differing nomenclature. In others, it is the failure to detect some or part of the fire-derived material. In still other cases, over-detection by some methods means that some non-fire derived material is also detected (Masiello, 2004).

Soot (Figure 2.5) has often been included in some definitions of black carbon. It differs from charcoal, for example, in that it represents a condensation product. It may form from the combustion of any fuel - not only vegetation, but also from coal, oil and gas. The abundance of soot in sediment may not, therefore, be solely a function of biomass burning. In other cases (e.g. Bond et al., 2013), the two terms are considered to be synonymous.

Charcoal is the main residue from biomass burning (Figure 2.5; and see Scott, 2010). It is clear that chemical changes occur during the charring process. It has been shown that charcoal may retain the anatomy of the plant, so that it can be recognized visually (Figure...