Forest fires: management, characteristics and prediction

By Miltiadis ATHANASIOU

Synoptic description of the fire problem

Forest fires1 are part of the ecological circle of Mediterranean ecosystems. However, during the last decades, they have become a problem due to the damages they cause. In the second part of the 20th century, the fire problem in the countries of Southern Europe has got worse. Abandonment of rural areas, long-term fire exclusion practices and expansion of fast growing species that are highly flammable (e.g. pines), have increased fire hazards.

Fire hazard is directly related to fuel flammability and is a measure of that part of the fire danger contributed by the fuels available for burning (FAO, 2006). The most commonly accepted definition of fire danger is “the resultant descriptor of the combination of both constant and variable factors that affect the initiation, spread and difficulty to control a wildfire on an area” (Deeming et al., 1972; 1977).

The Mediterranean countries (Portugal, Spain, the south departments of France, Italy, Greece and Cyprus) face a profound forest fire problem (Xanthopoulos et al., 2006) and in common with Australia, Canada and U.S.A. they face a growing wildfire risk from Wildland Urban Interface (WUI) expansion (Fig. 1 & 2).

Forest fire risk can be defined as a function of the wildfire hazard agent, the exposure of the subject (e.g. forest or structures) reduced by the capacity of the responsible organizations to mitigate and recover from loss. It is growing exponentially as a result of high and increasing population density (Tedim et al., 2015) and the complexity of the WUI and Rural Urban Interface (RUI) (Fig. 3). In Greece, a significant and recent WUI fire event was the large and destructive wildfire of NE Attica, on 21-24 August 2009 (Xanthopoulos & Athanasiou, 2013) (Fig. 2).

Fire risk is socially constructed and needs to be resolved by social means (Pyne, 2007). Fire planning and risk assessment are concerned with how often fires burn, what effects they have on wildland and urban values, and what opportunities exist to improve the situation through management actions (Finney, 2005).

In Greece, fires started becoming a problem in the 1970s. Until the 1960s they were considered a relatively minor problem, and were mainly fought by the people in the villages, without even fire trucks until 1970 and with the guidance of the Forest Service personnel. A few fire engines of the urban Fire Service were occasionally available and there were no available aerial means (Xanthopoulos, 2008).

1 The term “forest fires” is mostly used in Europe. The terms “wildfires” or “wildland fires” and “bushfires” are equivalent to the term “forest fires” and are commonly used in United States of America and in Australia, respectively.

Fig. 1: A moderate intensity WUI fire.

Fig. 2: A settlement - Wildland Urban Interface (WUI) after the NE Attica wildfire of 2009.

Fig. 3: A low intensity RUI fire.

Forest fuels were controlled through forest biomass utilization (for construction, cooking, heating, animal feed) and agricultural cultivations were kept free of fine fuels in the summer in order to reduce risk of fire damage. There were many cultivated fields around the villages, interrupting forest continuity (Xanthopoulos, 2008). Grass was removed from vineyards that functioned as greenbelts (Fig. 4) and olive groves were also properly cultivated and played the role of fuel breaks (Fig. 5).

Fig. 4: Vineyards function as green belts under correct maintenance (i.e. periodical removal

of grass) and may be used as safety zones by the firefighters.

In the 1950s and 1960s, the population of the villages was decreased sharply due to the immigration abroad and migration to the big cities, resulting in a significant forest

fuel built-up (accumulation). The abandonment of rural areas (Fig. 6) and the poor forest management by the Forest Service that further contributed to fuel build-up, affected the wildfire problem.

Fig. 5: This olive grove plays the role of a fuel break.

Fig. 6: A not properly managed olive grove will facilitate fire spread.

Forest vegetation nowadays reaches the homes at the perimeter of villages allowing fires to reach there as well (Fig. 7). Unfortunately, most of the people that live in WUI and RUI areas do not understand the importance of fire prevention and they don’t prepare their homes for the event of a forest fire (Fig. 7, 8 & 9). Most of them have little knowledge about firefighting and fire safety and they often make wrong decisions leaving their homes in panic and/or late, when they are about to be hit by the fire. Errors in citizens’ response that have led to disasters, can lead to lessons learned for dissemination to firefighters and the public (Xanthopoulos, 2015).

Forest fire (wildfire) management

Forest fire management is not only a technical matter of fire suppression but has also important social, biological, political and financial aspects. It is a complex problem that has to do with factors affecting the occurrence of fires, their characteristics and their destruction potential.

Fig. 7: The grass reaches the homes of this settlement allowing a fire to reach there as well.

Fig. 8: The wildfire reached the village, destroyed several houses and caused damages to

others.

Fire suppression oriented policy seems to be effective in Central European countries where forest fires are not a main environmental factor. In spite of that, some recent large fires could be considered as early warnings or signs of climate change that tends to affect forest fire risk in Central and Northern Europe. In 2008, a wildfire spread across Southern Norway burning 2,600 ha mainly of pine forest and in 2014 a wildfire burned 15,000 ha in Central Sweden (Tedim et al., 2015).

Fig. 9: The 2009 WUI fire is burning down a house in Palaia Penteli (Attica).

In Southern Europe, suppression is successful in “easy” fire seasons which means that under “easy” conditions (Fig. 10), most fires are stopped immediately. The doctrine “keep every fire as small as possible” which was the main objective of the forest fire policy in Europe during the twentieth century, combined with the poor forest fuel management, have led to fuel accumulation so when fire danger is extreme (that is under hot, dry and windy conditions) the stage is usually set for disaster (Fig. 11).

It is obvious that such an approach is not sustainable over the long run. What is needed is well designed, comprehensive forest fire policies based on a good understanding of all the interacting factors and, of course, adapted to each country’s environment and conditions (Xanthopoulos, 2007b). Those policies should take into consideration the underlying causes of fire incidence, the fire regime (frequency of fire occurrence, fire intensity, etc.) and the ecological role of the fire in the ecosystem.

In the Mediterranean region, wildfires occasionally erupt from natural causes (mainly lighting) but the majority of the wildfires, are human-caused (Leone et al., 2009;

Tedim et al., 2015), being also associated with the increasing WUI (Martınez et al., 2009; Badia et al., 2011; Marino et al., 2014). The human-caused starts, fall into three categories: a) negligence, b) accident and c) intentional arson.

Fig. 10: Firefighting in low maquis under relatively “easy” conditions.

Fig. 11: Extreme fire behavior in a lowland pine forest under hot, dry and windy conditions.

Negligent behavior causes the majority of the wildfires in the Mediterranean countries. Wildfires often start from unattended campfires in forested areas, irresponsibly discarded cigarettes, cars parked on grasses in WUI areas and burning debris in agricultural or RUI areas (e.g. annual crop residues which are mainly dead fine fuels such as wheat stubble or pruning). Sparks by train brakes, powerlines and machinery as well as car accidents, often cause accidental fires. Regarding arsonists, some of them are pyromaniacs, some others want to cause terror and some others focus on relatively small areas (Xanthopoulos, 2010).

The common Fire Cause classification that was established in 2012 and can be used by all European countries to report national fire causes to the European Fire Database (Camia et al., 2013), is expected to improve knowledge on this subject (Tedim et al., 2015). The Mediterranean countries contribute 94% of the total burned area in Europe, according to an analysis of the 1975-2000 statistics by the European Forest Institute. Analysis of fire statistics (causes, burned areas, etc.) shows that a small number of fires contribute most to the total area burned (Xanthopoulos, 2010).

Wildfire Prevention aims at minimizing the incidence of destructive fires. Forest fuel management plays a key role in successful wildfire prevention (Marino et al., 2014) and reduces wildfire hazard. The need for reducing fire hazard through active fuel management is becoming more and more obvious (Xanthopoulos, 2006) and should be integrated in sustainable long-term forest planning and in other rural development activities, taking into consideration fire risk, land use and fire causes. Apart from reducing fire potential, fuel management activities and agroforestry practices can directly involve and benefit the rural population (Marino et al., 2014).

As there is no single ideal technique for wildfire prevention, fuel management plans should commonly combine different practices (Fernandes & Botelho, 2003) and techniques, including mechanical treatments, prescribed burning and controlled grazing depending on the particular characteristics of each forest or rural area.

Applying fuel treatments at an appropriate landscape scale is critical to reducing wildfire damage (Agee & Skinner, 2005), in the context of prevention and pre-suppression planning. Areas with low surface fuel loads (Fig. 12) and properly cultivated agroforestry fields (Fig. 5) that may function as fuel breaks, as well as constructed fuel breaks along strategic locations (Fig. 13), can support suppression actions. Fuel break construction is common in Spain, France and Italy, while it is less common in Portugal and quite uncommon in Greece (Xanthopoulos et al., 2006). Nevertheless, vegetation removal is not a panacea as fuel breaks and fire breaks don’t stop a fire but they only provide a starting line for firefighters because they create an area of reduced fire intensity.

Estimation of fire hazard in broad areas may provide general guidance on forest type selection when planning re-vegetation and restoration of a burned site. (Xanthopoulos et al., 2012). One such example is that of the site of ancient Olympia, in Peloponnese Greece which was burned in August 2007 (Xanthopoulos et al., 2012). In that case,

the scientists of Mediterranean Forest Ecosystems and Forest Products Technology in Athens, re-introduced the broadleaved species that occupied the site in ancient times, reducing fire hazard for the regenerating forest (Lyrintzis et al., 2010).

Fig. 12: A Pinus brutia stand which can function as a fuel break.

Fig. 13: A stripe-shaped fuel break in Chios, constructed by the volunteer team Omikron.

Post fire forest regeneration depends on meteorological conditions, soil type and slope steepness, number of seeds, mortality of seedlings, competition among species, fire

regime (frequency of fire occurrence, fire intensity and the amount of fuel consumed), grazing intensity and land use changes. Restoration strategies may consist of a combination of techniques that may be used both immediately after a fire, and may continue to be necessary for years’ post-fire. Soil stabilization to control erosion and flood events is crucial and can be applied by utilizing suitable mechanical techniques.

Fire Danger A Fire Danger Rating level takes into account: current and antecedent weather, fuel types, and both live and dead fuel moisture. Fine dead fuel moisture content and wind speed are major components in the fire danger rating calculation. Since these components change daily, the fire danger rating of an area also changes daily, giving the fire manager a tool to help with the day-to-day “fire business” decisions (Fig. 14). Following are the terms and definitions for adjective fire danger.

Low (L) (Green) Fuels do not ignite readily from small firebrands although a more intense heat source, such as lightning, may start fires in duff or punky wood. Fires in open cured grasslands may burn freely a few hours after rain, but woods fires spread slowly by creeping or smoldering, and burn in irregular fingers. There is little danger of spotting.

Moderate (M) (Blue) Fires can start from most causes but, with the exception of lightning fires in some areas, the number of starts is generally low. Fires in open cured grasslands will burn briskly and spread rapidly on windy days. Timber fires spread slowly to moderately fast. The average fire is of moderate intensity, although heavy concentrations of fuel, especially draped fuel, may burn hot. Short distance spotting may occur, but is not persistent. Fires are not likely to become serious and control is relatively easy.

High (H) Yellow All fine dead fuels ignite readily and fires start easily from most causes. Unattended brush and campfires are likely to escape. Fires spread rapidly and short distance spotting is common. High intensity burning may develop on slopes or in concentrations of fine fuels. Fires may become serious and their control difficult unless they are attacked successfully while small.

Very High (VH) (Orange) Fires start easily from all causes and, immediately after ignition, spread rapidly and increase quickly in intensity. Spot fires are a constant danger. Fires burning in light fuels may quickly develop high intensity characteristics such as long range spotting and fire whirlwinds when they burn in heavier fuels.

Extreme (E) (Red) Fires start quickly, spread furiously, and burn intensely. All fires are potentially serious. Development into high intensity burning will usually be faster and occur from smaller fires than in the very high fire danger class. Direct attack is rarely possible and may be dangerous except immediately after ignition. Fires that develop headway in heavy slash or conifer stands may be unmanageable while the extreme burning conditions last. Under these circumstances the only effective and

safe control actions are on the flanks until the weather changes or the fuel loading decreases.

Fig. 14: An updated Fire Danger Prediction Map is issued and posted every day during the fire season (from 1 May to 30 October) by the General Secretariat for Civil Protection of

Greece.

Wildfires behaviour descriptors and prediction

Success in pre-suppression planning and actual suppression of wildfires is directly related to how well Fire Managers understand and are able to predict fire behavior. The safety of all firefighting personnel also depends on this knowledge and all actions taken on a fire during suppression, depend on how it “behaves”.

Fire behaviour is the manner in which fuel ignites, flame develops, and fire spreads and exhibits other related phenomena as determined by the interaction of fuel, weather and topography.

Fire managers use various fire behavior descriptors such as rate of spread, flame length, fire line intensity, spotting, the onset of crowning (or crowning initiation and/or propagation), etc. The term Rate of Spread (ROS, km/h or m/min) refers to the linear rate of advance of a wildfire either it is a head fire or not (Fig. 15). Head of the fire is the segment of the fire perimeter oriented in the direction of maximum spread (fanned by the wind or/and burning upslope) whereas heel of the fire is that segment of the fire perimeter that spreads against the wind or/and burning downslope (Fig. 15).

Flame length (FL, m), has been used to describe suppression difficulty (Alexander, 1982; Anderson et al., 2006). Flame size is directly related to the frontal fire intensity (Forestry Canada Fire Danger Group, 1992) or fire line intensity (Byram, 1959; Albini, 1976) (I, kW/m) which is defined as the energy output rate per unit length of fire front (Fig. 16). Depending on the fuel biomass which is available for

consumption, the general weather conditions and the time of the year, the fire intensity may vary, influencing the heat per meter of fire line that the soil is subjected to.

Fig. 15: The anatomical parts (segments) of a wildfire.

Fig. 16: Flame characteristics of a head grassfire on a level terrain: Flame length (FL), flame

height (hF), flame tilt angle (AT), flame angle (A).

The effects of fire on soil depend on the temperature and the residence time of the fire as well as the soil moisture content (Canu et al., 2007). The degree of change in both the chemical and biological properties of soil are strongly linked with fire intensity (Flinn et al., 1984). Exposure of the mineral soil may lead to loss of nutrient (Chesterfield, 1984) and organic matter. Some of these events may be further aggravated by the weather conditions following fire (Canu et al., 2007).

Spotting occurs when a fire produces sparks or embers that are carried by the wind and start new fires (spot fires) beyond the zone of direct ignition by the main fire (Fig. 17, 18, 19, 20 & 21). This phenomenon often threats the safety of human life, exacerbates fire suppression activities and causes structural losses in WUI fires (Alexander, 2009). Spotting can jeopardize the safety of the firefighters (Athanasiou & Xanthopoulos, 2013) (Fig. 19).

Fig. 17: Massive spotting in phrygana

Fig. 18: A downwind spot fire in a pine forest

Fig. 19: Spotting in shrub lands jeopardizing the safety of firefighters

Fig. 20: Spot fires spreading through a rough landscape

Analysis of field observations and measurements of spotting on maquis, phrygana and grass, that were made during the evolution of a large number of wildfires in Greece, in eight fire seasons (2007-2014), has led to some preliminary conclusions. There was no spotting at air relative humidity (RH) values higher than 40,3 % but it was documented that for RH values lower than 17% there was massive spotting that triggered extreme fire behavior (Athanasiou, 2015). Those results represent the first approach of this type to the spotting phenomenon in Greece.

Fig. 21: Surface spot fires spreading rapidly upslope late in the evening

Surface fires (Fig. 22, 23 & 24) burn through surface fuels (the maximum height of surface fuels is roughly 2.5 meters).

Fig. 22: A surface fire spreading in phrygana, downslope, fanned by strong wind

Fig. 23: A surface fire spreading in short maquis, upslope

Fig. 24: A low intensity surface fire spreading beneath a pine forest canopy

The phenomenon that occurs when a fire transitions from a surface fire into the crowns of individual trees or small groups of trees and burns briefly and vigorously but not necessarily from one crown to another is called torching (Albini, 1983; Andrews, 1996) (Fig. 25). Tree flammability, presence of ladder fuels, the height of lower branches (i.e. Canopy Base Height, CBH) and the entire surface fire behavior, may help torching.

Fig. 25: A small group of pine trees torching

A fire in which the crowns of individual trees or small groups of trees torch and burn, but solid flaming in the canopy cannot be maintained except for short periods is called a passive crown fire: (Scott & Reinhardt, 2001) (Fig. 26).

As burning conditions get worse and winds increase, a passive crown fire may become an active crown fire (Fig. 27 & 28) in which the surface and crown “phases” of the fire travel together as a linked unit (Van Wagner, 1977).

A preliminary finding about active crown fires in Aleppo pine forests with tall maquis understory in Greece, is that active crown fire ROS tends to be two times greater than surface fire ROS (Athanasiou & Xanthopoulos, 2010; Athanasiou, 2015). This finding is in general agreement with the suggestion of Cruz et al. (2005) that rates of spread in crown fires are often twice the spread of the surface fire.

Fig. 26: A passive crown fire in a pine forest

Fig. 27: An active crown fire in a pine forest

Fig. 28: Strong winds drive the flames of an active crown fire. It is a wind driven (Rothermel,

1991) active crown fire

The fires that are associated with the development of a strong convection column or plume, that towers above the fire rather than leaning over before the wind, are called plume dominated fires (Rothermel, 1991) (Fig. 29 & 30). The power of a plume dominated fire is higher than the power of the wind, and spot fires are produced massively and are drawn back towards the huge convection column by strong indrafts. In Greece, it has been found that the behavior of the powerful, plume dominated fires is affected by the conditions that the plume generates rather than by the characteristics of the fuels that the fire spreads in (Athanasiou, 2015).

Forest fuels and fire behavior in Greece Vegetation in Greece mainly consists of Mediterranean shrubs (phrygana and maquis), and pine, oak and fir forests. In Greece, wildfires mainly occur in the lower elevation Mediterranean vegetation. In 2007, the worst fire season in its recent history, many fires burned high elevation and low flammability forests of Northern Greece, due to the extreme drought. They kept burning for more than ten days until that part of Greece received a significant quantity of rain (Athanasiou & Xanthopoulos, 2010). Phrygana (Fig. 31 & 22) are low, often thorny, xeric shrubs up to 0.5 m height (Sarcopoterium spinosum, Phlomis fruticosa, Cistus spp., etc.) which are adapted to high fire frequency and occupy lower elevation areas.

flank heel

head

Fig. 29: A plume dominated fire heading towards Ancient Olympia, in 2007

Fig. 30: A plume dominated fire spreading vigorously in mountainous Arcadia, in 2007

Fire in phryganic areas is characterized by low to medium fire intensity and can reach very high rates of spread under high wind conditions. Phrygana have been the cause of many firefighting accidents in Greece They are fine, quite flammable, flashy fuels

and respond very quickly to changes of the environmental conditions (Xanthopoulos 2007a).

Fig. 31: Phryganic area where the dominant species is Sarcopoterium spinosum

Maquis (Fig. 32 & 33) are tall or short typical Mediterranean evergreen shrublands (Arbutus unedo, Pistacia lentiscus, Pistacia terebinthus, Phillyrea latifolia, Quercus coccifera, etc.). Those shrublands consist of drought-resistant broadleaved species, their fuel height ranges from 0.5 to more than 2.5 m and fuel load varies accordingly. Fire behavior varies significantly and depends on species composition, site characteristics, fuel loads, intensity of livestock grazing, etc. Most maquis species re-sprout vigorously after fire.

Fig. 32: Tall maquis

Most of the phryganic species (e.g. Sarcopoterium spinosum, Phlomis fruticosa) and maquis (Quercus coccifera, Pistacia lentiscus, Arbutus unedo, Phillyrea sp. Erica arborea) resprout after the fire while some phryganic species (e.g. Cistus spp.) are obligate seeders.

Low-elevation pine forests mainly comprise of Pinus halepensis or Pinus brutia, usually with evergreen shrub understory that is short or tall maquis (Fig. 34). In those lowland coniferous forests, high intensity crown fires occur under high fire danger conditions, and quite often their behavior is extreme (Fig. 35 & 36).

Fig. 33: Short maquis

Fig. 34: Aleppo pine forest (Pinus halepensis) with tall maquis understory

Fig. 35: A high intensity active crown fire is spreading in an Aleppo pine forest (Pinus

halepensis)

Fig. 36: High rates of energy release as shown by the flame lengths in an Aleppo pine forest

Deciduous and semi-deciduous oak forests (Quercus spp.) (Fig. 37 & 38), at altitude 800 m, and fir forests (Fig. 39) at higher altitude, usually burn with less intensity than the other types of Mediterranean vegetation. High elevation forests may consist of deciduous broadleaved species (Fagus orientalis, etc.) or cold tolerant conifers (Pinus

nigra, Pinus sylvestris, Abies spp., etc) which are not adapted to fire (Fig. 40). Fire events are rare in the high elevation forests but they are more destructive.

Fig. 37: Oak forest in Southern Peloponnese

Fig. 38: Oak forest in Central Peloponnese

Fig. 39: A moderate intensity passive crown fire in a fir forest

Fig. 40: Black pine forest (Pinus nigra) in Northern Greece

Weather (meteorological conditions) and topography When the fire spreads through a rough landscape (steep slopes, ravines, gorges, narrow valleys, canyons, saddles, ridges) the interaction between the general wind, the terrain and the energy that is released by the fire, affects its propagation, leading to erratic fire behaviour and making firefighting very difficult and dangerous (Athanasiou & Xanthopoulos, 2016) (Fig. 41, 42 & 43). The local wind field at the fire area often reflects the interaction between the topography and the general wind.

Fig. 41: Extreme fire behavior was documented in this gorge

Fig. 42: A wind-driven fire is spreading rapidly downslope, through an agroforestry mosaic

Fig. 43: Eruptive fire behaviour in a box canyon

Wildfire behaviour prediction Wildfire behaviour models can be adopted in wildfire management (forest management, fire prevention and suppression) only if their degree of reliability is well known. Broad operational adoption of such systems can be achieved and benefits can be maximized only if their strengths and generalizations, assumptions, weaknesses or limitations are well known. Continuous and extensive testing of fire behaviour prediction models, is necessary since it shows their limitations, documents their proper use and increases, eventually, their contribution to firefighting safety and efficiency (Athanasiou & Xanthopoulos, 2014; Athanasiou, 2015). Minor or significant disagreements between fire behaviour observations and predictions, are often attributed to the inadequacy of stylized fuel models to represent the existing forest fuel conditions. In those cases, specific custom fuel models can be developed to better describe heterogeneous forest fuel situations and mixed fuel complexes.

A significant number of studies have tested the prediction performance of BehavePlus (Andrews et al., 2005) for various types of fuels (Athanasiou & Xanthopoulos, 2014). It is based on Rothermel’s model (1972) which is the most widely used model of this kind in the world. It is practical, documented and the most comprehensive and robust to date. It’s not a spatial system but its ‘point-based’ fire modeling approach, provides a quick and easy way to do initial fire behavior assessments and allows easier ‘what if’ gaming. Measurements of weather conditions and topography as well as information about the forest fuels (fuel models), are necessary as inputs for using the BehavePlus prediction system.

It has been found that for Greece, BehavePlus can be a useful tool for predictions of surface wildfire ROS in tall maquis (Fig. 44), short maquis (Fig. 45 & 23), phryganic areas where the dominant species is Sarcopoterium spinosum (Fig. 22), and grass (Fig. 46) (Athanasiou & Xanthopoulos, 2014; Athanasiou, 2015).

The analysis of the flame length (FL) for the same fuel types in Greece, showed that BehavePlus predictions are not reliable. The finding that FL is seriously under-predicted when using BehavePlus to predict fire behaviour in Sarcopoterium spinosum dominated phrygana fields, is an important result that can be very useful for the safety of firefighters (Fig. 47 & 48). It should be seriously taken into consideration in operational firefighting in the country as the underestimation takes place in a narrow band of FL values that includes the FL threshold value of 1.2 m which is considered as the limit for direct attack on the flames with hand tools. The reliability of FL predictions is crucial since FL affects extinguishing capacity with hand tools and inaccuracies in FL predictions could jeopardize the safety of the firefighters (Athanasiou & Xanthopoulos, 2014; Athanasiou, 2015).

Fig. 44: A passive crown fire spreading in tall maquis, downslope

Fig. 45: A surface fire spreading in short maquis, upslope

Fig. 46: A surface fire spreading in grass, on flat terrain

Fig. 47: Plot of observed (FLobserved ) and predicted flame length for Sarcopoterium spinosum

dominated phrygana fields. The pairs of values are sorted in ascending order of FLobserved.

Fig. 48: A surface fire spreading in phrygana, downslope and against the wind

References

Agee, J., Skinner, C., 2005. Basic principles of forest fuel reduction treatments. Forest Ecology and Management 211(1-2), 83-96.

Albini, F.A., 1976. Estimating wildfire behavior and effects. Gen. Tech. Rep. INT-30. Ogden, UT: USDA, Forest Service, Intermountain Forest and Range Experiment Station, 92 p.

Albini, F.A., 1983. Potential spotting distances from wind-driven surface fires. Res. Pap. INT-309. Ogden, UT: U.S. Department of Agriculture, Forest Service, Intermountain Forest and Range Experiment Station, 27 p.

Alexander, M.E., 1982. Calculating and interpreting forest fire intensities. Canadian Journal of Botany 60, 349-357.

Alexander, M.E., 2009. Some pragmatic thoughts on the prediction of spotting in wildland fires. MITACS/GEOIDE Conference on Forest Fire Modelling, June 22-23, 2009 – Hinton, AB.

Anderson, W., Pastor, E., Butler, B., Catchpole, E., Dupuy, J.L., Fernandes, P., Guijarro, M., Mendes-Lopes, J.M., Ventura, J., 2006. Evaluating models to estimate flame characteristics for free-burning fires using laboratory and field data. In: Viegas, D.X. (Ed.), ‘Proceedings, V International Conference on Forest Fire Research’, 27–30 November 2006, Figueira da Foz, Portugal (CD-ROM). Forest Ecology and Management 234S, S77.

Andrews, P.L., 1996. Fire behavior. In: Pyne, S.J.; Andrews, P.L.; Laven, R.D. Introduction to Wildland Fire. 2d ed. New York, NY: John Wiley and Sons: Chapter 2.

Andrews, P.L., Bevins, C.D., Seli, R.C., 2005. BehavePlus fire modeling system, Version 3.0: User’s Guide. General Technical Report RMRS-GTR-106WWW revised. Ogden, UT: U.S. Department of Agriculture, Forest Service, Intermountain Forest and Range Experiment Station, 132 p.

Athanasiou, M., Xanthopoulos, G., 2010. Fire behaviour of the large fires of 2007 in Greece. In: Viegas, D.G. (Ed.), Proceedings of the 6th International Conference on Forest Fire Research. 15-18 November 2010, Coimbra, Portugal. ADAI/CEIF, University of Coimbra, Portugal. Abstract p. 336, full text on CD.

Athanasiou, M., Xanthopoulos, G., 2013. Observations of the spotting phenomenon, in wildfires in Greece. Proceedings of the 16th Hellenic Forestry Conference, October 6-9, 2013, Thessaloniki, Greece, pp. 30-40. Hellenic Forestry Society.

Athanasiou, M., Xanthopoulos, G., 2014. Wildfires in Mediterranean shrubs and grasslands, in Greece: In situ fire behaviour observations versus predictions. In: Viegas, D.G. (Ed.), Proceedings of the 7th International Conference on Forest Fire Research: Advances in Forest Fire Research, 17-20 November 2014, Coimbra, Portugal, pp. 488 (full text on CD). University of Coimbra, Portugal: ADAI/CEIF.

Athanasiou, M., 2015. Development of an optimal methodology for forecasting forest

fire behaviour in Greece. PhD dissertation, Department of Geology and Geoenvironment, National and Kapodistrian University of Athens, 408 p.

Athanasiou, M., Xanthopoulos, G., 2016. The wildfire of 17 July 2015, at the Faraklo village, Lakonia, Greece and its suppression. International Conference on Forest Fires and WUI Fires, May 25-27th 2016, Aix en Provence, France (in Press).

Badia, A., Serra, P., Modugno, S., 2011. Identifying dynamics of fire ignition probabilities in two representative Mediterranean wildland–urban interface areas. Applied Geography 31(3), 930–940. doi:10.1016/J.APGEOG.2011.01.016.

Byram, G.M., 1959. Combustion of forest fuels; Forest fire behavior. In Davis, K.P. (Ed.), Forest fire: Control and use, pp. 61-89, 90-123. New York: McGraw-Hill.

Camia, A., Durrant, T., San-Miguel-Ayanz, J., 2013. In: Harmonized Classification Scheme of Fire Causes in the EU Adopted for the European Fire Database of EFFIS Executive Report. Ispra, Italy: JRC, European Commission.

Canu, A., Arca, B., Ghiglieri, G., Pittalis, D., Deroma, M., Ventura, A., Arca, A., 2007. Fire intensity in moderate drought conditions: effect on topsoil properties in Mediterranean shrubland vegetation. Seventh Symposium on Fire and Forest Meteorology, Bar Harbor, ME American Meteorological Society.

Chesterfield, E.A., 1984. Effects of Fire on the Flora of dry Sclerophyll Forest. In: Ealey, EHM (Ed.), Fighting Fire with Fire, pp.129-145. Proc. Symposium on Fuel Reduction Burnibg, Monash University, Victoria, September 17-18, 1983.

Cruz, M. G., Alexander, M. E., Wakimoto, R. H., 2005. Development and testing of models for predicting crown fire rate of spread in conifer forest stands. Canadian Journal of Forest Research 35, 1626-1639.

Deeming, J.E., Lancaster, J.W., Fosberg, M.A., Furman, R.W., Schroeder, M.J., 1972. The National Fire Danger Rating System, Report No. RM-84. USDA, Forest Service, Ogden, UT.

Deeming, J.E., Burgan, R.E., Cohen, J.D., 1977. The National Fire-Danger Rating System -1978. USDA, Forest Service, General Technical Report INT-39, Intermountain Forest and Range Experiment Station, Ogden Utah, 63 p.

FAO, 2006. Fire management: voluntary guidelines. Principles and strategic actions. Fire Management Working Paper 17. Rome. Available at www.fao.org/forestry/site/35853/en.

Fernandes, P. M., Botelho, H.S., 2003. A review of prescribed burning effectiveness in fire hazard reduction. International Journal of Wildland Fire 12, 117–128. doi:10.1071/WF02042.

Finney, M.A., 2005. The challenge of quantitative risk analysis for wildland fire. Forest Ecology and Management 211, 97-108.

Flinn, D.W., Farrell, P.W., Steward, H.T.L., Leitch, C.J., Hopmans, P., 1984: The

Effects of Fire in Eucalypt Forest on Soil, Nutrient Cycling, Tree Growth and Catchment Hydrology: a Review with Particular Reference to Fuel reduction Burning. In: Ealey, EHM (Ed.), Fighting Fire with Fire, pp.146-185. Proc. Symposium on Fuel Reduction Burnibg, Monash University, Victoria, September 17-18, 1983.

Forestry Canada Fire Danger Group, 1992. Development and structure of the Canadian Forest Fire Behavior Prediction System. For. Can. Info. Rep. ST-X-3.

Leone, V., Lovreglio, R., Martın, M.P., Martınez, J., Vilar, L., 2009. Human factors of fire occurrence in the Mediterranean. In: Chuvieco, E. (Ed.), Earth Observation of Wildland Fires in Mediterranean Ecosystems, pp. 149-170. Berlin Heidelberg: Springer-Verlag.

Lyrintzis, G., Baloutsos, G., Karetsos, G., Daskalakou, E.Ν., Xanthopoulos, G., Tsagari, C., Mantakas, G., Bourletsikas, A., 2010. Olympic Rebirth. Wildfire 19(1), 12-20.

Marino, E., Hernando, C., Planelles, R., Madrigal, J., Guijarro, M., Sebastian, A., 2014. Forest fuel management for wildfire prevention in Spain: a quantitative SWOT analysis. International Journal of Wildland Fire,23, 373-384. http://dx.doi.org/10.1071/WF12203.

Martınez, J., Vega-Garcıa, C., Chuvieco, E., 2009. Human-caused wildfire risk rating for prevention planning in Spain. Journal of Environmental Management 90, 1241–1252. doi:10.1016/J.JENVMAN.2008.07.005.

Pyne, S.J., 2007. Problems, paradoxes, paradigms: triangulating fire research. Int. J. Wildland Fire 16, 271-276.

Rothermel, R.C., 1972. A mathematical model for predicting fire spread in wildland fuels. Res. Pap. INT-115. Ogden, UT: U.S. Department of Agriculture, Forest Service, Intermountain Forest and Range Experiment Station, 40 p.

Rothermel, R.C., 1991. Predicting behaviour and size of crown fires in the Northern Rocky Mountains. Res. Pap. INT–438. Ogden, UT: USDA Forest Service, Intermountain Research Station.

Scott, J. H.; Reinhardt, E. D., 2001. Assessing crown fire potential by linking models of surface and crown fire behavior Res. Pap. RMRS-RP-29. Fort Collins, CO: U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. 59 p.

Tendim, F., Xanthopoulos, G., Leonne, V., 2015. Forest fires in Europe: Facts and Challenges. In: Paton, D., Mccaffrey, S., Tedim, F., Büergelt, P. (Eds.), Wildfire Hazards, Risks, and Disasters, pp. 77-93. Elsevier

Van Wagner, C. E., 1977. Conditions for the start and spread of crown fire. Canadian Journal of Forest Research 7, 23–34.

Xanthopoulos, G., Caballero, D., Galante, M., Alexandrian, D., Rigolot, E., Marzano,

R., 2006. Forest Fuels Management in Europe. In: Andrews, P. L, Butler, B. W. (Eds.), Proceedings of the International Conference on “Fuels Management - How to Measure Success”, March 28-30, 2006, Portland, Oregon, USA, pp. 29-46. Fort Collins, CO: USDA Forest Service, Rocky Mountain Research Station. RMRS-P-41.

Xanthopoulos, G., 2007a. Forest fire related deaths in Greece: confirming what we already know. Book of abstracts of the “IV International Wildland Fire Conference”, May 13-17, 2004, Seville, Spain, p. 339. Full paper on the CD accompanying the book of abstracts.

Xanthopoulos, G., 2007b. Forest fire policy scenarios as a key element affecting the occurrence and characteristics of fire disasters. Book of abstracts of the “IV International Wildland Fire Conference”, May 13-17, 2007, Seville, Spain, p. 129. Full paper on the CD accompanying the book of abstracts.

Xanthopoulos, G., 2008. People and the Mass Media during the fire disaster days of 2007 in Greece. Proceedings of the International Bushfire Research Conference on “Fire, Environment and Society” of the Bushfire Cooperative Research Centre and the Australasian Fire Emergency Service Authorities Council (AFAC), September 1-3, 2008, Adelaide, Australia, pp. 494-506.

Xanthopoulos, G., Viegas, D. X., Caballero, D., 2009. The fatal fire entrapment of Artemida (Greece) 2007. pp. 65-75. In: Viegas, D.X. (Ed.), Recent Forest Fire Related Accidents in Europe. European Commission, Joint Research Centre, Institute for Environment and Sustainability. EUR 24121 EN. 75 p.

Xanthopoulos, G., 2010. Examining the causes of large forest fires in mediterranean countries. Proceedings of the international workshop on “Assessment of Forest Fire Risks and Innovative Strategies for Fire Prevention”, 4-6 May, 2010, Rhodes, Greece, pp 22-23. Ministerial Conference on the Protection of Forests in Europe.

Xanthopoulos, G., Fernandes, P., Calfapietra, C., 2012. Fire hazard and flammability of European forest types In: Moreira, F., Arianoutsou, M., Corona, P., De las Heras, J., (Eds.). Post-Fire Management and Restoration of Southern European Forests, pp. 79-92. Heidelberg: Springer.

Xanthopoulos, G., Athanasiou, M., 2013. The evolution and the suppression of the NorthEast Attiki wildfire of August 21 to 24, 2009: Lessons for the future. Proceedings of the 16th Hellenic Forestry Conference, October 6-9, 2013, Thessaloniki, Greece, pp. 73-83. Hellenic Forestry Society (in Greek with English abstract).

Xanthopoulos, G., 2015. Wildfires and Safety Issues in Greece. In: Leblon, B., Alexander, M.E., (Eds.), Current International Perspectives on Wildland Fires, Mankind and the Environment, pp. 157-176. New York, USA: Nova Science Publishers.