june 2014

From Indonesia to Botswana, from Scotland to North Carolina, peat mega-fires burn for months, destroy habitat, clog the air with haze, and self-accelerate climate change impacts.

By Guillermo Rein

Smouldering combustion is the slow, low temperature, flameless burning of porous fuels. It is especially common in wildland fuels which are thermally thick and form a char on heating. In the natural environment, smouldering fires burn two types of biomass: thick fuels like tree branches or logs, and organic soils like the duff layer or peat. These are characterized by having a significantly greater thermal time compared to fine fuels like foliage. The persistent smouldering of thick fuels is typically observed for a few days after a flaming wildfire has passed, and it is often referred to as residual combustion. This can make residual smouldering be responsible for the majority of the biomass burned during a wildfires.

Peat soils are made by the natural accumulation of partially decayed biomass and are the largest reserves of terrestrial organic carbon. Because of this vast accumulation of fuel, once ignited, smouldering peat fires burn for very long periods of time (e.g. months, years) despite extensive rains, weather changes or fire-fighting attempts. Indeed, smouldering is the dominant combustion phenomena in mega-fires of peatlands which are the largest fires on Earth

in terms of fuel consumption. Smouldering fires contribute considerably to global greenhouse gas emissions, and result in widespread ecosystem destruction. Moreover, because peat is ancient carbon, and smouldering is enhanced under warmer and drier climates, it creates a positive feedback mechanism in the climate system, a self-accelerating global process [Rein, 2013].

Reports of peat fires lasting for several months are not unusual, like the 1997 Borneo fires in Indonesia, 2000 Okavango delta fires in Botswana or 2008 Evans Road fire in North Carolina, USA. Peat fires occur with some frequency worldwide in tropical, temperate and boreal regions. Droughts, drainage and changes in land use are thought to be main causes leading to the high flammability conditions of dry peatlands. Possible ignition events can be natural (e.g. lightning, self-heating, volcanic eruption) or anthropogenic (land management, accidental ignition, arson).

The most studied peat mega-fire took place in Indonesia in 1997 and led to an extreme haze event (see Figure 1). The smoke covered large parts of South-East Asia, even reaching Australia and China, and induced a surge of respiratory emergencies in the population and disruption of shipping and aviation routes for weeks. It was estimated that these fires released the equivalent to 13-40% of global man-made emissions of the year 1997 [Page et al, 2002]. This mega-fire was not an isolated case in the region, haze episodes have drifted to South East Asia once every three years on average. Rough figures at the global scale estimate that the average greenhouse gas emissions from peat fire is equivalent to >15% of manmade emissions.

Due to its complexity and coupling of heat and mass transport with chemical processes inside a reactive porous media, and despite its broad implications to the environment, current understanding of smouldering combustion is limited, and considerably less advanced than flaming combustion [Rein, 2013].

The characteristic temperature and intensity of smouldering combustion are low compared to flaming combustion. Because of these characteristics, smouldering spreads in a creeping fashion, typically around 1 mm/min, which is two orders of magnitude slower than flame spread. A smouldering front can be initiated with weaker ignition sources and is more difficult to suppress than flaming combustion. This makes it the most persistent combustion phenomena.

Because the water content of wildland fuels like peat can vary naturally over a wide range of values (from dry to flooded), and because water represents a significant energy sink, moisture content is the single most important property governing the ignition and spread of smouldering wildfires. The critical moisture content for ignition (related to the moisture of extinction) of boreal peat has been measured around 120% in dry basis [Frandsen, 1997], although exact value depends on mineral content and density. Peat drier than this is susceptible to smouldering. The prominent

role of moisture is such that natural or anthropogenic-induced droughts are the leading cause of smouldering mega-fires.

The second most important property that affects ignition is the mineral content. As experimentally found by Frandsen [1997], there is a decreasing linear relationship between the mineral content and the critical moisture content: higher mineral loads mean soil can only ignite at lower moistures. This is because the inert content is a heat sink to the fire. This rule can be applied to most organic soils or fuel beds to determine if they are susceptible to smouldering. Any soil which has a composition that is more than 80% mineral, cannot sustain a smouldering fire. After moisture and mineral contents, other important properties are bulk density, porosity, flow permeability and composition.

Organic material located close to the surface of the soil burns in shallow fires (roughly <1 m under the surface). They propagate laterally and downwards along the organic layers of the ground and leave voids or holes in the soil. This pattern allows fuel consumption to be using the depth of burn to calculate the volume of the void. Depth of burn is the vertical distance between the original soil location and the post-fire soil location. A typical value for the depth of burn reported in several field studies is around 0.5 m, which means that the average fuel consumptions per unit area is around 75 kg/m2 [Rein, 2013]. The depth of burn and amount of fuel consumption, along with the resulting impacts of peat fires, support their classification as mega-fires.

The prominent role of moisture is such that natural or anthropogenic-induced droughts are the leading cause of smouldering mega-fires.

Smouldering fires have detrimental effects on the forest soil, its microflora and microfauna. This is because it consumes the soil itself and also because the long residence time of smouldering means that heat penetrates deep into the soil layers. On the contrary, flames produce high temperatures above the ground for short periods of time (in the order of 15 min). This results in minimal heating of the soil below depths of a few cm and can leave the soil system relatively unharmed. However, smouldering fires lead to enhanced heat transfer into the soil for much longer durations (i.e. in the order of 1 h). As a comparison, the thermal conditions in smouldering are more severe than medical sterilization treatments, and mean that the soil is exposed to conditions that are lethal to biological agents [Rein, 2013].

Flaming wildfires may attract more attention than smouldering fires, but the study of flameless fires will contribute forward-looking ideas to understanding and managing a form of combustion that demands our focus.


Frandsen, W.H. (1997) Ignition probability of organic soils. Canadian Journal of Forest Research 27: 1471–7. http://dx.doi.org/10.1139/x97-106

Page, S.E., Siegert, F., Rieley, J.O., Boehm, H.D.V., Jaya, A. & Limin, S. (2002) The amount of carbon released from peat and forest fires in Indonesia during 1997. Nature 420: 61–65. http://dx.doi.org/10.1038/nature01131

G Rein, Smouldering Fires and Natural Fuels, Chapter 2 in: Fire Phenomena in the Earth System – An Interdisciplinary Approach to Fire Science, C Belcher (editor). Wiley and Sons, 2013. http://onlinelibrary.wiley.com/doi/10.1002/9781118529539.ch2

About the author

Dr. Guillermo Rein ([email protected]) is Senior Lecturer in Mechanical Engineering at Imperial College London, and Editor-in-Chief of Fire Technology. His professional activities focus on research in fire and combustion, and teaching of thermofluid sciences toengineers. He has studied a wide range of fire dynamics topics in the built and natural environments, including pyrolysis, fire modeling, wildfires, structures and fire, and forecasting techniques. Over the course of the last 15 years he has also specialized in smouldering combustion, conducting both computational and experimental studies on a variety of fuels like polyurethane foam, cellulose, peat and coal.