This article presents a discussion on wildfire and the technical and organizational efforts to mitigate wildfire disasters. Wildfire not only causes loss of life, property, and ecosystems but creates other disasters such as erosion and economic degradation. There is a global awareness of the catastrophic consequences caused by wildfire. Accordingly, there is an ongoing effort to mitigate wildfire by early detection and preparedness planning, which translates into the ability to respond quickly. The logic is the sooner a wildfire is detected, the sooner resources can be activated to respond to the emergency and prevent an incident from becoming a disaster. This article explores why wildfire has become a globally recognized natural hazard and explores recent technological advances to detect wildfire. The human equation and its relation to preparedness planning is also explored. In conclusion, wildfire threats are globally recognized as having significant detrimental economic and environmental impacts. The fact that preparedness planning efforts are not subjugated by media exposure usually reserved for tsunami and earthquake disasters is a testament of the global commitment to prioritize wildfire mitigation.
Wildfire causes loss of life, property, and ecosystems. There is a bona fide awareness and desire to mitigate wildfire by early detection, which translates into the ability to respond quickly and prevent an incident from becoming a disaster. This article explores why wildfire has become a globally recognized natural hazard and depicts recent technological advances to detect wildfire. The human equation is also explored.
This article is not intended to describe, endorse, or discuss any particular type of technology but to describe the scientific approach to wildfire preparedness planning. Wildfire mitigation and detection is very, very complex. It includes data supply and analysis, ground-based and satellite-based fire data, spatial fire risk modeling, mapping, information dissemination, scientific calibration according to global regions, and risk response training just to name a few.
For those of us who know little of the subject matter but strive to attain excellence in disaster and emergency management, it is necessary to learn the fundamentals of wildfire mitigation and planning, not wildfire science. The purpose of this research is to bring to the emergency management professional a sense of knowledge that will sufficiently identify this hazard, highlight the significance of this particular type of hazard by qualifying and quantifying its consequences, note technological advances that are expected to become a major player in wildfire mitigation, and describe the human element and its necessity to ensure any mitigation efforts are successful. This article serves as a primer for the emergency management professional to further explore and develop their planning skills on the subject of wildfire preparedness planning to prevent an incident from becoming a disaster.
What is Wildfire?
Wildfire (sometimes referred to as wildland fire) is uncontrolled burning on grasslands, brush, or woodlands.
Wildfire is classified into three categories:
- Surface fire is the most common type that burns along the floor of a forest, moving slowly and killing or damaging trees;
- Ground fire that usually starts by lightning and burns on or below the forest floor; and
- Crown fire that spreads rapidly by wind and moves quickly by jumping along the tops of trees.
Significance of Wildfire
Lightning is the prime source of fire not ignited by people. The majority of uncontrolled and destructive wildfire is caused by humans as a consequence of inappropriate use of fire in agriculture, pastoralism, and forestry.
Improved property is at risk when:
- Situated along or within a preserve area;
- Thick tree canopy and/or brush understory are present;
- Located on a ridgeline or cliff (a fire traveling up a slope will move faster and have longer flames than a fire traveling on flat terrain);
- Has only one escape road or has steep access roads that may slow down heavy response vehicles;
- Windows overlook slopes or vegetation (heat can cause them to break leaving an opening for flames);
- Extensions overhang slopes like eaves, room pushouts, or bay windows;
- Wooden structures are attached or nearby such as decks, porches, fences, and playscapes; and
- Susceptible to windy updrafts along a ridge (soaring vultures are tell-tale signs).
Fire is the most important disturbance agent in global vegetation worldwide, affecting between 3 to 4 million square kilometers (1.86 to 2.49 million square miles) annually. Secondary effects of fire include sudden onset disasters such as landslides, mudslides, rock falls, and flashfloods. Creeping disasters triggered by fire include post-fire soil erosion, ecosystem degradation, and reduced carrying capacity for human populations and their livelihood. Haze and smoke from wildfires are major threats to aviation, limiting visibility and leading to engine failure.
Global vegetation fires, including burning of peatlands, constitute a significant source of greenhouse gases and aerosols. Burning of forests and other vegetation is a major driver of transferring carbon from the terrestrial sphere to the atmosphere. Fires globally consume about 5% of net annual terrestrial primary production per year and release about 2 to 4 billion metric tons of carbon per year. Approximately 0.6 billion tons of carbon emitted to the atmosphere come from tropical deforestation and peat fires (the global figure is equivalent to about 20-30 percent of global emissions from fossil fuels).
Wildfire occurs annually on every continent except Antarctica, and most global fire is unmonitored and undocumented. Increasing trends in wildfire activity have been reported in many global regions. Wildfire has many serious negative impacts on human safety, health, regional economies, and global climate change. Developed countries spend billions every year in an attempt to limit the impact of wildfire. In contrast, developing countries spend little money to control fire, yet they are often the most susceptible to the damaging impacts of fire because of increased vulnerability of human life and property (due to limited fire suppression capability), increased risk due to high fire frequency (often caused by the cultural use of fire), and sensitive economies (tourism, transport).
Instances of Death, Destruction, and Costs from Wildfires
Three years of drought in 2000, 2001, and 2002 resulted in several large wildfires in the South Dakota Black Hills forests. In the summer of 2002, one of those fires erupted in the northern Black Hills near the city of Deadwood, burning more than 700 acres and destroying 10 homes.
In the summer of 2009, the “Station” fire was the largest wildfire in Los Angeles County history scorching 160,000 acres, killing two firefighters, and destroying dozens of structures. The cost of fighting the blaze, caused by arson, was more than $95 million.
In the winter of 2009, firestorms in Australia killed 173 people and destroyed more than 2,000 homes (and entire towns). There were suspicions that arson may have played a role in the 400 blazes. The destruction caused by these fires prompted a judicial inquiry as to how the blazes raced out of control so quickly and the death toll so high. The ensuing report indicated that sources of information did not cope with the level of demand, resulting in the deliberate blocking of warnings in one area. Instances of communication problems causing delays and overloading of information-containing web sites were also cited.
As it pertains locally, Austin is at risk for wildfire year-round. The recent growth explosion has increased the threat from wildfire, especially on the edges of the city where homes border grassland and wooded areas, called the "urban/wildland interface." Emergency response is not as fast and water supplies are not as readily available in these outskirt areas, so advanced preparation is crucial. The Austin Fire Department has determined that the conditions in west Austin and Travis County are similar to those that were present before the 1991 Oakland, California disaster (according to the U.S. Fire Administration, the Oakland wildfire caused the largest dollar fire loss in United States history).
The Case for Early Warning Systems
Over the past decade, many regions of the world have experienced a growing trend of excessive fire application in land-use systems and land-use change, and an increasing occurrence of extremely severe wildfires. Some of the effects of wildfire are transboundary such as smoke and water pollution and its impacts on human health and safety and the loss of biodiversity or site degradation at a landscape level leading to desertification or flooding. The depletion of terrestrial carbon by fires burning under extreme conditions in some vegetation types, including organic terrain in peatland biomes, is one of the driving agents of disturbance of global biogeochemical cycles, notably the carbon cycle. This trend is causing the international community to address the problem collectively and collaboratively.
Currently, average fire detection time is five minutes in manned lookout towers. Guards have to work 24 hours in remote locations under difficult circumstances. They may get tired or leave the lookout tower for various reasons. Therefore, computer vision-based video analysis systems capable of producing automatic fire alarms are necessary to reduce the average forest fire detection time. Early warning with high spatial and temporal resolution that incorporates measures of uncertainty and the likelihood of extreme conditions allows managers to implement fire prevention, detection, and pre-suppression plans before the fire problems begin.
Some computer vision-based algorithm approaches are directed towards detection of the flames using infra-red and/or visible-range cameras, whereas some others aim at detecting the smoke due to wildfire. Infrared cameras and sensor-based systems have the ability to capture the rise in temperature. However, they are much more expensive compared to regular pan, tilt-zoom cameras. It is almost impossible to view flames of a wildfire from a camera mounted on a forest watch tower unless the fire is very near to the tower. Smoke rising up in the forest due to a fire is usually visible from long distances.
Wildfire detection algorithm is developed to recognize the existence of wildfire smoke within the viewing range of the camera monitoring forest regions. Smoke at far distances (>100 meters to camera) exhibit different temporal characteristics than nearby smoke and fire. This demands specific methods explicitly developed for smoke detection at far distances rather than using nearby smoke detection methods.
A wildfire smoke detection algorithm may consist of four main steps: (1) slow moving video object detection; (2) gray region detection; (3) rising video object detection; and (4) shadow detection and elimination.
If technology exists to detect a fire more quickly, “that’s an advantage to everyone,” says Elaine Aguilar, city manager of Sierra Madre, California. “Fires don’t always start somewhere where somebody sees it by driving by; they can happen in remote areas and travel quickly.” “Does the technology even exist to do this kind of thing?” asks Max Moritz, co-director of the University of California Berkeley’s Center for Fire Research and Outreach. “I think the question is open.”
Methodologies for Early Warning of Wildfire
Early warning of wildfire and related hazards include a variety of methodological approaches to identify precursor developments and assess/predict the escalation of the wildfire theatre including:
- Assessment of fuel loads: round measurements and available satellite-generated information allow estimation of the amount of fuels available for wildfire. This is important because dryness and fire risk alone do not determine the extent and severity of fire;
- Prediction of lightning danger: methods exist for observing/tracking lightning activities as a source of natural ignition (ground-based lightning detection systems and space borne monitoring of lightning activities);
- Prediction of human-caused fire factors: modeling/predicting is critical since wildfire is caused directly or indirectly by human activity. This field of research requires adequate consideration of socioeconomic factors (ownerships, land uses, unemployment problems, etc.);
- Prediction of fuel moisture content: this term is closely related to the readiness and ease of vegetation to burn. Early warning systems include meteorological danger indices and space borne information on vegetation dryness (intensity and duration of vegetation stress) and soil dryness. Prediction of inter-annual climate variability/drought is important for preparedness planning in many countries;
- Prediction of wildfire spread and behavior: airborne and space borne monitoring of active fires allows the prediction of movements of fire fronts to areas with values at risk. The technologies used include airborne instruments to monitor fire spread in situations of reduced visibility (smoke obscured) or to cover large areas. A large number of orbiting and geostationary satellites are available to identify active fires. Numerous wildfire behavior models are in place that allows the prediction of spread and intensity of wildfires; and
- Assessment of smoke pollution: in situ (i.e.-in the original position) air quality monitoring systems allows tracking of fire smoke pollution and issue alerts (warnings to populations). Surface wind prediction allows prediction of smoke transport from fire-affected regions to populated areas. Satellite imageries can detect smoke transport.
Types of Early Warning Systems
Early Warning System for Forest Fire Management in East Kalimantan, Indonesia
The early warning system developed for East Kalimantan uses two tools: 1) the National Fire Danger Rating Index (NFDRI) based on the Keetch-Byram Drought Index and 2) the Preparedness Level defined by the actual NFDRI, the actual fire occurrence, weather forecast, haze conditions, and hot spots (hot temperature events) detected by using National Oceanic and Atmospheric Administration-Advanced Very High Resolution Radiometer (NOAA-AVHRR) data. NFDRI for the province is currently based on only six meteorological stations along the east coast of the province. Due to the absence of weather data for remote areas in East Kalimantan, NOAA-AVHRR thermal channels will be used to include land surface temperature and rainfall probability data in dynamic risk mapping to assess the fire risks of remote areas more precisely. This method currently being tested could be a cheap and extremely valuable alternative for many countries facing the same problem of lacking sufficient meteorological stations to assess their fire risks adequately.
Mexico National Forest Fire Information and Early Warning System
Following the extreme fire events of 1998, fire management agencies in Mexico realized the need for an integrated fire information system. Therefore, at their request, the Canadian Forest Service developed an operational prototype system for Mexico. The Mexico Fire Information System (Sistema de Información de Incendios Forestales) was developed in cooperation with the Secretariat of the Environment and Natural Resources (SEMARNAT) of Mexico. This system was launched on the Internet in the spring of 1999 (http://fms.nofc.cfs.nrcan.gg.ca/Mexico/). The web site offers daily maps of fire weather and fire behavior potential for Mexico based on real-time weather data observed in Mexico and is disseminated by the World Meteorological Organization through a Canadian weather satellite. The production of the maps is accomplished with the Canadian Forest Fire Danger Rating System (CFFDRS), which is used by Canadian fire management agencies for preparedness planning and firefighting resource allocation.
The system serves to help coordinate fire management at a national level, reduce the risk of future fire disasters by developing and implementing an automated national system to monitor forest fire danger, to demonstrate an example of Canadian/Mexican cooperation in mitigating a natural disaster common to both countries, and demonstrate Canadian science and technology internationally.
Over a period of three years, the system was evaluated by SEMARNAT to determine its applicability to Mexican climatic and forest conditions. Various strengths and weaknesses were identified, but the system generally was deemed to be of significant value and worth pursuing.
European Union Sensor and Computing Infrastructure for Environmental Risks System
The European Union-funded program, Sensor and Computing Infrastructure for Environmental Risks (SCIER), took on the challenge of developing a state-of-the-art automated system to detect disasters in the making, forecast how an emergency is likely to unfold, alert authorities, and get them the information they need to respond effectively.
The first level of the group’s solution is to deploy networks of ground-based sensors such as video cameras, meteorological instruments, and river-level gauges in high-risk areas, especially the “urban-rural interface,” where homes and businesses lie close to undeveloped terrain.
The ground-based sensors are linked wirelessly into what the researchers call a local area control unit. This level of the system structures and compares the raw data; for example, checking to see if a temperature spike at one sensor is matched by similar changes at nearby sensors. The system reportedly filters out what is unrealistic to not trigger a false alarm.
When the local area control unit decides a threat is real, it activates the next level of SCIER’s computational armamentarium to forecast how the emergency is likely to develop during the crucial first hours. The researchers have implemented sophisticated mathematical models of how natural disasters unfold. Those models include detailed information about the local geography, plus real-time sensor data concerning wind, rainfall, temperature, and other variables. They found that in order to produce meaningful forecasts, they need to generate multiple simulations of a disaster. Only then can their models provide authorities with accurate and useful information, such as where a wildfire is most likely to threaten homes. The system uses the most likely simulations to generate detailed maps that authorities can use to manage the emergency.
Generating these complex simulations in real time demands enormous amounts of computing power. SCIER relies on the GRID to provide that computational clout. The GRID, sometimes known as the next-generation Internet, is a dedicated network that links thousands of computers via a fiber-optic network that is up to 10,000 times faster than the Internet. It allows researchers to perform calculations that could not be done otherwise.
Early Recognition System in Germany, Estonia, Mexico, Portugal, and Czech Republic
Tower-based, automatic forest fire early recognition systems are being utilized in Germany, Estonia, Mexico, Portugal, and the Czech Republic. The optical scanning system has automatic recognition of clouds by day and smoke at night. It incorporates local online data processing and utilizes a small band radio or cable transmission of alerts to a central office. It has an optimum coverage range detection of 15 kilometers. The time to alert is approximately four minutes for a single tower setup and approximately two minutes for a multi-tower system. The system has a detection accuracy for smoke clouds of 10 x 10 meters at a 10 kilometer distance. During a 360° rotation, the camera takes three photos every 10°. For a better presentation of the smoke clouds, 36 photos are combined to form a panorama view in the central office. Reported smoke areas are marked on electronic maps and an operator evaluates all events by means of the data transferred to the central office. The system installation, maintenance, and service require experienced personnel. The operator has to hire staff that is familiar with the local area to decide, in view of their knowledge of the area, if there really is a fire.
City of Deadwood, South Dakota Early Detection and Communication System
As mentioned earlier in this report, in the summer of 2002 a wildfire in the northern Black Hills near the city of Deadwood burned over 700 acres and destroyed 10 homes. The city elected to install a ground-based wildfire alert system of seven, heat-activated sensors. The sensors are mounted on 10 foot-tall fiberglass poles. In theory, activation sends a signal to the city’s email system that gives global positioning system (GPS) coordinates of the wildfire (within 3 meters) and then activates the alarm of the closet fire station. On March 9, 2010 this researcher contacted the fire department and talked to retired Chief Ken Hawki (telephone number 605-578-1212). Chief Hawki reported deficiencies with the system. He indicated that in order for Deadwood to obtain GPS coordinates upon activation of a sensor, the fire department had to call the systems manufacturer in California. The coordinates could only identify the angle at which the suspect wildfire may be located but could not determine the distance from the sensor; it could be six feet or six miles from the sensor. Another problem was the system persistently tripped false alarms. Chief Hawki told a story where one sensor location continually activated the alarm system. It turned out the problem had to do with semi-tractor trailers parking too close to the sensor. Drivers would park their rigs (and leave them idling) and walk across a road to use an automated bank teller. By the time the truck driver returned to the rig and drove away, the heat from the rig triggered the alarm. Chief Hawki reported the department literally stumbled on this occurrence. Chief Hawki’s analysis can be summed up in his own words, it’s “not worth a damn.” He reported the city is working with the manufacturer to repair the system.
The Human Element
All the technology in the world will not prevent or abate wildfire disasters without incorporating the human elements of awareness and participation. In fact, it could be said that all disaster-related activities are people-centered. The objective of people-centered early warning systems is to empower individuals and communities threatened by hazards to act in sufficient time and in an appropriate manner so as to reduce the possibility of personal injury, loss of life, damage to property, the environment, and loss of livelihoods. A complete and effective early warning system comprises four inter-related elements: risk knowledge, monitoring and warning service, dissemination and communication, and response capability (see figure below). A weakness or failure in any one part could result in failure of the whole system.
- Risk knowledge: risks arise from the combination of the hazards and the vulnerabilities to hazards that are present. Assessments of risk require systematic collection and analysis of data and should take into account the dynamics and variability of hazards and vulnerabilities that arise from processes such as urbanization, rural land-use change, environmental degradation, and climate change. Risk assessments and risk maps help to motivate people, prioritize early warning system needs, and guide preparations for response and disaster prevention activities.
- Monitoring and warning services: this is the core of the system. They must have a sound scientific basis for predicting and forecasting and must reliably operate 24-hours a day. Continuous monitoring of hazard parameters and precursors is necessary to generate accurate warnings in a timely fashion. Warning services for the different hazards should be coordinated where possible to gain the benefit of shared institutional, procedural, and communication networks.
- Dissemination and communication: warnings must get to those at risk. For people to understand warnings they must contain clear, useful information that enables proper responses. Regional, national, and community-level communication channels and tools must be pre-identified and one authoritative voice established. The use of multiple communication channels is necessary to ensure everyone is reached and to avoid the failure of any one channel, as well as to reinforce the warning message.
- Response capability: communities must also respect the warning service and know how to react to warnings. This requires systematic education and preparedness programs led by disaster management authorities. It is essential that disaster management plans are in place and are well practiced and tested. The community should be well informed on options for safe behavior and on means to avoid damage and loss of property. Strong interlinkages are required between all of the elements, underpinned by effective governance and institutional arrangements, including good communication practices. It also requires linking the subject, perceived as predominantly technical, to sustainable development and community development agendas and to disaster risk reduction agendas.
In the United States, working with local communities is a critical element to reduce fire hazards proximate to homes and communities and restore damaged landscapes. To accomplish this, the Forest Service and the Department of Interior recommends:
- Expanding the participation of local communities in efforts to reduce fire hazards;
- Improving local fire protection capabilities through financial and technical assistance to state, local, and volunteer firefighting efforts;
- Assisting in the development of markets for traditionally underutilized small diameter wood as a value-added outlet for removed fuels; and
- Encouraging a dialogue within and among communities regarding opportunities for reducing wildfire risk and expanding outreach and education to homeowners and communities about fire prevention.
Wildfire is a destructive force created by natural (lightning) and manmade causes. Wildfire creates harmful haze and smoke and can cause soil erosion, ecosystem degradation, and reduced carrying capacity for human populations and their livelihood. Secondary effects of wildfire include onset disasters such as landslides, mudslides, rock falls, and flashfloods. Today, there are numerous operational disaster-related detection systems, particularly in most industrialized countries. However, these systems are less well-developed for wildfire and there exists no internationally accepted fire-warning system.
Based on existing regional and global initiatives and partnerships, efforts are underway that support the development of intergovernmental agreements and resourcing necessary to implement a global wildfire monitoring and early warning system. Unfortunately, despite the efforts of governments and civil society, the majority of countries do not have sufficient human or technical resources for sustainable fire management. Even if worldwide funding were available through a transfer of wealth from rich nations to poor nations, it has been demonstrated that elites in poor nations have been adept at capturing aid flows for their own purposes, thereby creating situations that do not necessarily translate into helping the poor.
It is evidenced in this article that technological advances in detection are ongoing at an amazing rate. Whereas the need for a standardized global system of wildfire detection is desirable, it may not be practical at this time. Wildfire detection may best be served by allowing competitive entities (private enterprise, government consortiums, etc.) to continually engage in the development of more effective technologies allowing for the evolution of better and cheaper systems (similar to combat and passenger aircraft manufacturing, cellular telephone technologies, etc.).
The human factor is the lynchpin to success or failure of wildfire preparedness planning. It is here where commitment originates and response efforts hinge. Without good detection and early warning, response resources can be wasted and possibly result in needless environmental and human destruction. The desire to mitigate, plan, respond, and reevaluate is dependent on the willingness of government leaders and emergency response professionals to devote the time and physical and economic resources to achieve sustainability in our environment.
Bob Bridge is the principal of Hazard Assessment & Safety in Austin, Texas. He is a Certified Environmental Technician through the Texas Engineering Extension Service (TEEX) in the Texas A&M University System and is a Certified Infrastructure Preparedness Specialist and Registered Environmental Property Assessor through the National Registry of Environmental Professionals.