Informal Settlement Fires in Colombia

Colombia is an intertropical country embedded in a complex geography located in the Northeast of South America. Three parallel Andean mountain chains crossing the country create ranges of elevations and different ecosystems, which define five natural regions: the Pacific coast, Caribbean coast, “Llanos” (Orinoco), Amazon and Andean [19]. The five natural regions also define social, cultural, and economic differences in the country. In contrast to other countries in Latin America, Colombia concentrates its population in several major cities. The main factors associated with the widespread development of informal settlements include (i) six decades of armed conflict that implied 9.2 million events of forced displacement, (ii) volatile economic policies which led to different urbanisation patterns, and (iii) the rapid urbanisation in the 1930s due to massive migrations from rural areas to cities that drove the society towards rapid industrialization and concentration of inhabitants in cities [20, 21].

According to the latest national census in 2018, the Colombian population living in suburbs or surrounding communities raised up to 77% [22]. Dense cities have high pressure to offer citizen livelihoods to either settled or floating inhabitants, such as housing, three water systems (i.e., stormwater, drinking water, and wastewater), power, and other essential services that imply infrastructure provision (e.g., sanitation, hospitals, schools). However, housing infrastructure in dense cities is getting expensive and unaffordable as a result of a mix of macroeconomic factors such as unemployment, economic deficit, acceleration of demographic growth with social inequality, and fiscal deficit that increase low-income families out of real estate market led them to illegal occupation of land [4]. This is evidenced by the last report from the National Department of Statistics of Colombia (DANE for its acronym in Spanish—Departamento Administrativo Nacional de Estadística) showing that 5.5 million people (more than 11% of the population of the country) live in inadequate housing, informal settlements, or slums [23, 24]. Only in the capital city (Bogotá) the number of informal settlements reported in 2018 was 214 (Fig. 1). This number represents 18% of the total neighbourhoods in the capital of the country [4, 24].

Informal settlements in Colombia are known by multiple names, including “informal human settlements”, “precarious settlements”, “settlements of illegal origin”, “mixed settlements”, “slums”, “unconsolidated settlements”, “in process consolidation”, “invasion”, “marginal neighbourhoods” or “pirate neighbourhoods” [24]. There is an overlap between these definitions as they all suggest settlements operating outside of the formal legal and regulatory systems (e.g., planning, land use, building standards, health, and safety) [4]. However, in Colombia, these definitions are commonly associated with two main groups: informal settlements which relate to settlements with no secure tenure for residents, and slums making refer to settlements with inadequate construction and services. In this paper, the term “informal settlement” covers both groups of settlements. It is important to note that not all informal settlements are slums, as some informal settlements have well-developed housing and services without being legally recognized. Similarly, not all slums are informal settlements, as they may be legal housing that has deteriorated due to poor service delivery [6]. According to Zarate 2019 [25], the way informal settlements are named implies negative connotations that often relate their inhabitants to suffering discrimination, displacement, and even persecution. This cultural phenomenon increases the social and economic gap leading to a loop of underestimation of the problem in the country, including their vulnerability and risk associated with natural and technological hazards such as fire.

In general, the status of legality, physical conditions, and provision of services in informal settlements has implications for the quality of life and health of residents [26]. Informal settlements in Colombia have precarious access to roads leading to chaotic transportation systems, challenging emergency services provision (e.g., health, firefighters, among others) [20] and limiting access to infrastructure and social resilience assets [27]. For example, informal settlements in Bogota commonly share their space with a high-voltage tower, resulting in large ignition risk associated with unsuitable electrical connections and limited access to essential services [20]. The limited access to electrical services and power supply in informal settlements often leads inhabitants to find different unsafe ways to connect to the network, including multiple dwellings irregularly connected to a single power tower. These practices involve high physical risk that increases the risk of fire in dwellings of informal settlements (e.g., electrical arcing ignition). According to Carvajalino et al. [20], by 2015 at least 43% of all informal settlements in Bogota had one irregular connection to essential services, and 13% had no access to formal and legal essential services. These limitations hinder the ability of fire emergency services to apply effective suppression protocols aiming to mitigate the development of the fire, thereby increasing the level of vulnerability [20].

In terms of the socioeconomic conditions of informal settlements in Bogota, according to Carvajalino et al. [20], inhabitants lack formal jobs or have temporal positions with low- or no income, which leads to an absence of health insurance and social protection. Unemployment reflects violence and insecurity in the informal settlements in Bogota. According to social community leaders, the critical socioeconomic issues that inhabitants living in informal dwellings in Bogota usually undergo are: (a) access to a safe environment, (b) food security, (c) public health, and (d) proper roads [20]. Drug traffic, gang groups and urban guerrillas are additional problems that can take place in this type of settlements [28]. In Colombia, illegal conditions of informal settlements limit the range in which authorities and local governments can operate to help inhabitants under disaster-related challenges, such as fire or fire-related catastrophes. Informal settlements have physical barriers and limit human resources for emergency management response that confine community resilience building to insufficient strategies and adaptation options [7, 24].

It is well known that the location and the morphological characteristics of informal settlements play a major role in fire risk. In Colombia, informal settlements are often located in fragile ecosystems such as riverbanks, coastlines, protected mountain lands, railway lines, and isolated forestry areas of the suburbs in the cities. Thus, the use of resources for the construction of dwellings and land occupation fosters ecosystem degradation, which had a negative impact on the soil conditions and the transformation of the natural landscapes. In Colombia, the diversity of natural regions defines the diversity of informal settlements and their dwelling types. Different building materials, access to critical infrastructure, provision of essential services (e.g., water, gas, power), and different fire ignition drivers expose dwellings in informal settlements to multiple hazards such as floods, landslides, and fires [29]. Figure 2 illustrates common types of dwellings located in informal settlements in Colombia

After a comprehensive review of the different types of informal settlements in Colombia, it was found that the five natural regions previously presented generally define different types of settlements and fire risks. The informal settlement shown in Fig. 2a is located in the Pacific region of Colombia. This region has tropical rainforests (northern ecosystems) and coastal mangroves (southern ecosystems). Inhabitants in the pacific region have diverse origins (30% Indigenous and 40% African) [30]. This region has been neglected due to armed conflict and forced displacement, which increase poverty, inequality, and social vulnerability [30]. The most common type of informal settlements located in this region is the so called “palafitos”, located over coastal mangroves and usually multi-story structures built from native wood (e.g., Rhizophora mangle, Calophyllum mariae Planch and Camnosperma panamensis) and zinc sheets for roofing purposes [31]. Dwellings in the Amazon region (Fig. 2c), which constitute around 40% of the country's surface area, is the less populated and home to more than 60 ancient indigenous groups [32]. Dwellings in the Caribbean region-Guajira (Fig. 2e) are desertic settlements surrounded by the ocean, characterised by elevated temperature climate and flat topography inhabited by ancient indigenous communities (Wayuu, Arhuaco, Chimila and Kogui) [33, 34]. The Orinoco or Llanos region is characterized by wide flatlands. It is known for its cattle raising, agriculture, natural gas and oil reserves, and many water resources (e.g., Meta, Guaviare) [35]. The dwellings in the Orinoco region do not have different structural characteristics than other regions, and the region has a low-density population compared to other regions [35].

Informal settlements in the Andean region, shown in Fig. 2b and d, are often located in sloped areas with pasture and forest ecosystems in city suburbs, usually surrounded by water resources and prominent rivers (Magdalena and Cauca) [36]. Informal settlements in this area are typically built from easily accessible materials and labour, usually involving self-build houses. According to Torres Tovar 2009 [37], they are progressively built following a similar pattern across the country, starting with a single space for all the functions of the home (i.e., the bed, the kitchen, the bathroom, and other items are located in the same place) and followed by an expansion including separate rooms (i.e., bedroom), kitchen and finally the restrooms. Concerning their structural characteristics, the most commonly used materials are wooden-based organic materials (often native bamboo or “Guadua”), zinc sheets and plastic sheets (often polypropylene fabric), resulting in highly flammable and extremely fire-prone structures. In addition to the fire risk associated with their structural and morphological characteristics, it is common to find informal dwellings surrounded by disposable and other types of highly combustible materials (e.g., tires, paper, plastics, etc.), which can considerably increase the fire risk.

Figure 3 shows an example of a common type of informal settlement in the Andean region of Colombia, as well as an illustration of some of their main structural characteristics. Generally, the structure is composed of Guadua (Guadua Angustifolia Kunt), which is a giant native bamboo of Colombia widely spread to Andean regions of Latin America (where most of the biggest cities are located), and the exterior walls use what is referred to as "Esterilla" (derived from dried Guadua). Historically, Guadua has been used as housing material by indigenous communities and farmers in rural communities because of its rapid growth, easy forestry management, and relevance for the cultural landscape [38, 39], and it is typically readily available in the suburbs of the cities. Currently, the fire behaviour of Guadua and “Esterilla” is not well understood, and limited information is available in the literature. Mena et al. [40] investigated the ignition properties of round and laminated dried Guadua using the LIFT apparatus and reported critical heat flux values for piloted ignition around 14 kW/m², which is similar to other wooden materials [41, 42]. These results suggest that Guadua behaves as highly flammable and charring organic material, which is in accordance with its documented behaviour during informal settlement fires discussed in the following session. Polypropylene (PP) fabric and/or polyethylene plastic layers are also typically used to cover and insulate the dwelling, considerably increasing the fire hazard level in terms of ignition, fire spread and toxicity. Thin layers of galvanized steel (or zinc) sheets are generally used for roofing and walling purposes, usually allowing for large effective leakage surface areas with a high potential of modifying the development of compartmentalized fires, as well as hindering the shielding protection from external fire exposure. The main unknowns associated with the fire dynamics of informal dwellings in Colombia are related to the material used and the construction style (i.e., round Guadua and Esterilla sheets). The round Guadua has cavities and separated internal sections that could change its dynamics with respect to a wooden stick. The Esterilla has multiple openings that can change the ventilation dynamics during a fire, and their combination with plastic coatings has the potential to further increase the risk of spread.

In Colombia, information about informal settlements' fire events is available in newspapers (local or national) and in the database maintained by both Fire Department and the National Unit for Disaster Risk Management -UNGRD-, which are national emergency response units. The UNGRD is the highest Colombian organisation responsible for the identification, analysis, evaluation, and monitoring of disasters, including structural fires and informal settlements [43]. After a fire event, the fire emergency department shares with the UNGRD relevant information regarding the response and consequences of the fire, aiming to maintain a unified database, which also includes other types of disasters, such as flooding and earthquakes. Although it is possible to find reports from other non-governmental and not-for-profit organisations such as TECHO Colombia [44], UNGRD is the only organisation with an updated digital database readily available, including information at the national level relevant to the study of informal settlement fires in recent years. It is worth mentioning that the information available in digital format is not complete, as there might be reports physically stored in some of the fire departments of the country (more than one thousand offices). In this research, data collection of informal settlement fire cases was made by setting and filtering the types of events reported in the UNGRD annual report. This information is open access to the public on the website through spreadsheets [45].

Table 1 shows a summary of search terms used to filter the database and to access the relevant information concerning informal settlement fires used in this work.

It was found that the descriptions of some events were inconsistent and incomplete, which is one important limitation of this study. For example, in some cases, although the number of people and families affected during the event was available, the information about the number of dwellings affected (i.e., dwelling partially damaged by fire) and destroyed were not. This situation is evident, for example, in the informal settlement fire that took place in Moravia, Medellin (0.8 ha fire and 108 dwellings destroyed) [46], where some relevant information, such as dwellings destroyed, was reported in the local newspaper but was not included in the UNGRD database. Another relevant limitation of the study is the lack of clarity in the description of the events. Some descriptions of informal settlements fires did not report possible fire causes, type of dwellings involved in the fire or further analysis of the facts, despite the fact that this information is essential to improve response protocols and community recovery strategies. Additionally, it was found that fire cases involving one or two dwellings often lacked detailed information required to identify informal settlement fires.

During the analysis of the UNGRD database, it was found that prior to 1998, no descriptions or reports of structural fires were recorded digitally and nationally [45], and that the reports published between 1998 to 2011 presented incomplete or vague information. Although the information published between 2011 and 2020 is more detailed, some events do not provide complete indications. Consequently, knowing the limitations and lack of information in the database, only a 10-year study period is contemplated (from 2011 to 2020). Considering the aforementioned limitations and the nature of the database, it is expected that the statistical results discussed and reported in the next section do not account for the full magnitude phenomenon but provide a reasonable overview of the consequences associated with informal settlement fires in Colombia, being a complete study of this kind to the date. Newspapers and TV news (available on official websites and video databases such as YouTube) were used to cross-check some of the cases in order to increase the reliability of the analysis.

Results following the distillation of the database are presented in this section, with the aim of measuring and illustrating the magnitude of the problem of informal settlement fires in Colombia. From 2011 to 2020, 17% (432 out of 2540) of total structural fire events (including fires in informal settlements, residential buildings, factories, warehouse, etc.) were identified as informal settlement fires. The number of people affected and the number of informal settlement dwellings destroyed during that time were 22,563 (about 5349 families) and 3489, respectively. During the same period of time, the total number of people dead and injured was 35 and 275, respectively. The most devasting informal settlement fire event affected approximately 1400 people.

On average, every year in Colombia, 2256 people are directly affected by the occurrence of fire in informal settlements, 43 fires take place, and three people die due to this type of event (Table 2). In addition, about 349 homes are completely destroyed, and another 63 are partially affected during the spread of the fire each year. These numbers are of the same order of magnitude as informal settlement fires in Costa Rica, another country in Latin America. In Costa Rica, 229 informal settlements fires took place between 2015 and 2019, including a highly destructive case which affected 532 people and destroyed 189 dwellings [47]. Although the aforementioned figures are alarming, it is worth mentioning that the magnitude of the problem in Colombia and Costa Rica is considerably low compared to South Africa. Between 2003 and 2018, there were more than 64,000 cases of fires in informal settlements. In 2011 only 410 people died as a result of fires, and by 2018, the number of deaths was 289. In the case of South Africa, 10,000 people were affected during a single fire event [26].

Figure 4 presents the annual number of fires in informal settlements, people affected, people injured, people dead, dwellings destroyed, and dwellings affected in Colombia between 2011 and 2020. The highest number of informal settlements fires were registered in both 2014 and 2015, which are related to several individual cases of great magnitude. One of the most devasting fires occurred in the department of Nariño (Pacific region) in 2014, where 370 people were affected, and nearly 60 dwellings were destroyed. Another high-impact fire event took place in 2015, where 702 people were affected, and 180 dwellings were destroyed in the department of Cauca (Pacific region). Although the number of informal settlement fires progressively increased between 2011 and 2014, from 2015, the cases were found to vary without a consistent pattern. It is worth noting that in both 2013 and 2019, there were no reported deaths, even though the number of informal settlement fires exceeded 40 cases. However, the descriptions of the fire cases available in the database generally include fatalities that occurred during the fire (i.e., incident date) and not due to health complications associated with it sometime later. This might further contribute to underestimating the results reported in this work.

Figure 5 shows the number of informal settlements destroyed between 2011 and 2020 by the department. The Pacific and Andean regions have the highest number of dwellings destroyed, accounting for 90% of the total cases. In Colombia, in the last ten years, the Pacific region accounted for 52% of the informal settlement fires, and the department of Valle del Cauca alone accounted for nearly 18% of the cases. One of the reasons related to the high presence of informality and the large number of informal settlement fires in these regions is the phenomenon of forced internal displacement. As indicated by Soler Barón et al. [24], around 1.4 million victims were forced to migrate from rural areas to informal settlements located in cities, sometimes within the same region. Therefore, most of the population currently living in informal settlements are located in the Pacific region (Quibdó—91.56% and Buenaventura—41%) [23].

Figure 5 emphasises why poor people are disproportionately impacted by natural disasters [8]. For example, Fig. 5 shows that the Pacific Coast Region has high risk and destructive consequences due to fire incidents in the decade analysed. Historically, this region manifests outstanding and generalised poverty [48]. The lack of resources and limited access to safe housing conditions suggests that vulnerability even increases when safety conditions are not appropriate to address the risk. We emphasized that the connection between fire risk and poverty depends on building conditions and high vulnerability with low fire safety. This evidence shows that poverty in informal settlements of Colombia exacerbates the vulnerability of inhabitants who are more exposed to fire risks. Finally, it is worth mentioning that reconstruction processes could be limited due to a lack of resources and local capacities to attend to emergencies [49, 50].

It was found that 88.5% of the causes of fires in informal settlements were reported as “undetermined” or “unknown” between 2011 and 2020. In some cases, words such as "apparently" were used to describe the causes of fires. This lack of detail can be related to the difficulty of carrying out adequate post-fire analysis suitable for determining possible causes of the fire, including forensic work and data collection from surveys. The other 11.5%, shown in Fig. 6, are mainly caused by electrical problems (42%), food preparation (26%) and storage of materials (12%). Similar causes of informal settlement fires are reported in other studies focused on other countries. In Costa Rica, for example, the main cause of informal settlement fires in 2019 was a short circuit, failure of electronic equipment and arson [47]. In South Africa, in 2018, 12% of the fires were related to electrical problems and 20% to open flames [26].

These high-risk ignition sources are the consequence of the previously mentioned common problems that communities of informal settlements undergo on a daily basis, including limited access to electrical connections, inadequate material for cooking (e.g., wood) and poor access to suitable waste management services. Once the fire started, the construction material and the proximity of the dwellings to each other considerably increased the impact of the fire and its ability to spread. For these reasons, and especially because of the unknown causes, there is an urgent need to develop better post-fire investigation protocols that can help to clarify possible fire causes, as well as to obtain additional information to help improve response, interventions, and methods of community involvement. Collecting reliable fire data (e.g., type of building, primary causal factor, incident time and date) has been helpful in reducing building fires and their associated deaths, injuries, and economic damage. Likewise, fire data have been important to guide the allocation of public resources to areas of greatest need and impact and to support safety interventions and education programs [51].

The structural characteristics and location of informal settlements, combined with the limited access to essential services (e.g., power and water), represent a great challenge to first responders and emergency protocols. Some of those challenges are discussed in this section and illustrated in Fig. 7, which shows the aftermath of two informal settlements in Colombia. Obstructed clearance, inadequate swept path width and unsuitable slope road gradients are common challenges hindering access to emergency services. These difficulties force emergency services to carry out suppression protocols far away from the fire source, commonly from the periphery of the settlement in areas such as highways or from other places of difficult access, as shown in Fig. 7a (Moravia fire 2017, 108 dwellings destroyed, four people injured, and 471 families affected). In consequence, some suppression actions are generally not effective enough to limit the spread of the fire, which is often stopped by fire-resistant nearby constructions (e.g., brick or concrete) or another non-flammable boundary. In some cases, such as during the fire in Riosucio, shown in Fig. 7b (82 dwellings destroyed, two deaths and more than 450 people affected), no local firefighting brigade was available and timely access to the fireplace was unachievable. Additionally, some areas of the city were only accessible via wooden footpaths without the structural capacity to carry loads from vehicles, which made it even more difficult to control and suppress the fire.