Page 1 of 6 Original Research Disaster risk assessment at Roburnia Plantaion, Mpumalanga, South Africa Authors: Rudzani A. Makhado1,2 Amani T. Saidi3 Ailiaions: 1 Research and Evaluaion Secion, Limpopo Legislature, South Africa Disaster Management and Educaion Centre for Africa, University of the Free State, South Africa 2 South African Environmental Observaion Network, Naional Research Foundaion, South Africa 3 Correspondence to: Rudzani Makhado Email: makhado2002@yahoo.com Postal address: Private Bag X9309, Polokwane 0700, South Africa Dates: Received: 26 Oct. 2012 Accepted: 25 Apr. 2013 Published: 15 July 2013 Keywords: Disaster; Risk Assessment; Risk Equaion; Fire; Roburnia Plantaion How to cite this aricle: Makhado, R.A. & Saidi, A.T., 2013, ‘Disaster risk assessment at Roburnia Plantaion, Mpumalanga, South Africa’, Jàmbá: Journal of Disaster Risk Studies 5(1), Art. #64, 6 pages. htp://dx.doi.org/10.4102/ jamba.v5i1.64 Copyright: © 2013. The Authors. Licensee: AOSIS OpenJournals. This work is licensed under the Creaive Commons Atribuion License. Read online: Scan this QR code with your smart phone or mobile device to read online. This study reports about disaster risk assessment undertaken at Roburnia Plantation, Mpumalanga Province, South Africa. Both quantitative and qualitative approaches were followed to collect data. A total of eight experienced foresters and ire ighters were purposively sampled for interview at Roburnia Plantation. A questionnaire survey was also used to collect the data. Risk levels were quantiied using the risks equations of Wisner et al. (2004) and the United Nations International Strategy for Disaster Reduction (UNISDR 2002). Data were analysed using descriptive and inferential statistics. Analysis of variance (ANOVA, single factor) was also applied. This study found that Roburnia Plantation is highly exposed to ire risks. The mean (± s.d.) output from the Wisner risk equation shows that ire is the highest risk at 7.7 ± 0.3, followed by harsh weather conditions at 5.6 ± 0.4 and least by tree diseases, pests and pathogens at 2.3 ± 0.2. Similarly, the mean (± s.d.) output from the UNISDR risk equation also shows that ire is the highest risk at 2.9 ± 0.2, followed by harsh weather conditions at 2.2 ± 0.3 and least by tree diseases, pests and pathogens at 1.3 ± 0.2. There was no signiicant deference in the risk analysis outputs (p = 0.13). This study also found that the number of ire incidents were low during summer, but increased during winter and spring. This variation is mainly due to a converse relationship with rainfall, because the availability of rain moistens the area as well as the fuel. When the area and fuel is moist, ire incidents are reduced, but they increase with a decrease in fuel moisture. Introducion Disasters are a complex mix of natural hazards and human actions, and their increasing occurrence hinders economic and human development across the world (Wisner et al. 2004). The United Nations International Strategy for Disaster Reduction (UNISDR 2005) revealed that disaster loss is on the rise, with grave consequences for the survival, dignity and livelihoods of individuals, particularly the poor in developing countries. It is generally believed that the increasing risk and vulnerability to disasters is a result of changing demographics, technological, political and socio-economic conditions, as well as unplanned urbanisation, development in high risk zones, environmental degradation, and climatic variability (UNISDR 2005; Wisner et al. 2004). The multiplicity of the causal factors of disasters shows the complexity associated with their prevention and management. Commercial plantations in South Africa are frequently exposed to disaster risks such as ires, harsh climatic conditions, tree diseases, pests and pathogens. The occurrence of these disasters causes much damage to plantations. Records show that ires pose the highest risk to the sustainability of the forestry sector in South Africa. For instance, in a survey carried out by Forestry South Africa on behalf of the then Department of Forestry and Water Affairs in 2007/2008 found that of the 77 150 ha of plantation that had been lost or destroyed, 70 812 ha or 92% had been lost to or destroyed by ires (DWAF 2009). A combination of adverse weather conditions, diseases, insects and animals were responsible for the loss or destruction of the remaining 6338 ha or 8%. It is similarly reported that about 994 008 ha of plantation were damaged from 1980 to 2011 in South Africa and that the damage of 579 728 ha or 58% of the total was caused by ires, whilst other causes accounted for the damage of the remaining 414 280 ha or 42% of the total (Forestry & Forest Product Industry 1980–2011). Disasters have the potential to cause loss of livelihoods and services. Conducting disaster risk assessments is critical in order to reduce the possible effects of those disaster risks. As indicated by the UNISDR (2009), disaster risk assessments assist in determining the nature and extent of risk by analysing potential hazards and evaluating existing conditions of vulnerability that together could harm exposed people, property, services, livelihoods and the environment on which people depend. The most critical part of a disaster risk assessment is that data and information generated are useful in developing disaster risk prevention and control strategies. As also indicated by Jordaan (2006), scientiic risk identiication and analysis is crucial in order to provide accurate information for policy-making, risk prioritisation and the development of a reliable disaster risk index. htp://www.jamba.org.za doi:10.4102/jamba.v5i1.64 Page 2 of 6 The concept of risk analysis has evolved over time, and there is improvement in the recent risk analysis equations. However, a reliable and standardised risk equation is necessary in order to obtain similar results at international level. Wisner et al. (2004) indicated that: the risk of disaster is a compound function of the natural hazard and number of people, characterised by their varying degrees of vulnerability to that speciic hazard, who occupy the space and time of exposure to the hazard event. (p. 49) There are basically two elements in the risk equation, namely hazard and vulnerability, as expressed in the following Wisner et al. (2004:49) risk equation: R=HxV [Eqn 1] Where R = risk; H = hazard; V = vulnerability. However, the Wisner et al. (2004) risk equation does not consider the capacity to deal with hazard. That gap is addressed in the UNISDR (2002) risk equation, which considers capacity in the equation. The UNISDR (2002:36) risk equation is expressed as follows: R=HxV/C [Eqn 2] Where R = risk; H = hazard; V = vulnerability; C = capacity. Various risks analysis equations have been developed around the world (e.g. Jordaan 2006; Kaji 2002; Morimiya 1992; Van Westen, Van Asch & Soeters 2006), but they lack a standardised methodology. Jordaan (2006) indicated that the most commonly used formula in disaster risk assessments in South Africa is one that considers hazard, vulnerability and capacity in the risk equation. The risk equations presented in equations 1 and 2 above were developed by Wisner et al. (2004) and UNISDR (2002) respectively. This study applied both equations in order to assess the extent of ire disaster risk at Roburnia Plantation. Problem statement Fire is one of the major risks that negatively affect the sustainability of plantation forests in South Africa. Fire is one of the dominant disturbances in forest ecosystems (Flannigan, Stocks & Wotton 2000), and its occurrence has the potential to destroy large numbers of hectares of plantation forests. Fire occurs almost every year in the plantation forests in South Africa, and causes immense damage. However, the magnitude of ire risk and its impact has not been adequately addressed, particularly at a localised scale. The problem that this study sought to address was that of determining the extent of ire risk and its impacts at the Roburnia commercial forestry plantation in the Mpumalanga Province of South Africa. Methodology The study was conducted at the Roburnia Plantation, which is one of the Komatiland Forest Plantations, and is located at 26°40′04″S and 30°44′18″E near a small town called Amsterdam, Mpumalanga Province of South Africa. The htp://www.jamba.org.za Original Research types of tree species grown at Roburnia Plantation include pine, wattle and gum trees. The plantation is divided into two sections: the Roburnia and Blairmore sections. The total area in both sections is about 23 000 ha. The area planted with commercial forests is about 16 000 ha, whilst about 7000 ha are under indigenous vegetation cover. The reasons for choosing Roburnia forest plantation as the study area were twofold: irstly, Roburnia Plantation is located in the Mpumalanga Province – the province which has the highest number of commercial plantations in South Africa; and secondly, the same province annually experiences the highest number of ires that destroy plantations. Both epistemological and scientiic research designs were used in this study in order to assess the level of disaster risk. Quantitative data sets on factors that influence the occurrence of ire were collected. These include data on ire frequency, intensity, impact and the level of risk, and data on the number of hectares burnt by ire and economic losses. Qualitative data sets on perceptions of people towards the extent of ire were collected using questionnaire survey and interview methods. Secondary data sets were also collected from literature, including oficial documents and reports obtained from the management of the plantation. A purposive sampling method was followed in order to identify individuals to be interviewed at Roburnia Plantation. This sampling technique was chosen in order to ensure that data and information were obtained from knowledgeable and experienced participants. The participants interviewed were foresters and ire ighters and had more than 10 years of experience in the forestry sector. A total of eight people were sampled for interviews, which included five forest managers and three ire ighters. They were interviewed in order to determine the extent and impact of disaster risks at the plantation. Risk assessment score sheets were circulated via email, and also distribution to the ofices for completion. A risk assessment matrix or score sheet was used to capture collected data. The matrix used was developed by the Ekurhuleni Metropolitan Municipality: Disaster Management Centre (2002) Risk Assessment Matrix, adapted by Maryna StrydomStorie. Hazard assessment was determined through scoring the frequency, intensity and overall rank of the identiied risks (scale: from not likely to occur = 1 to certain = 3). The scores were added and then divided by 3 to give the hazard level. Vulnerability assessment was determined through scoring the impact of identiied hazard on socio-economic and environmental issues (scale: from low = 1 to high = 3). The scores were added and then divided by 5 to give the vulnerability level. Capacity assessment was determined through scoring the capacity to respond to the hazard (scale: from poor = 1 to good = 3). The scores were added and then divided by 5 to give the capacity level. Secondary data were gathered through literature reviews in books, journals, reports and websites which are available and accessible in the public domain. Data on the number of hectares burnt, number of ires and ire extinguishing costs were downloaded from Disaster Management System (DMS) at the Roburnia Plantation. doi:10.4102/jamba.v5i1.64 Page 3 of 6 The datasets collected through the questionnaire and the risk assessment matrix were captured in Ms Excel 2007 and then analysed using descriptive and inferential statistical methods. The risk equations 1 and 2 presented earlier were then used to determine the levels of risk. Analysis of variance (ANOVA, single factor) was also applied in order to examine the relationship between the two equations. Results and discussion Before the risk could be determined, it was essential to irst conduct a detailed hazard, vulnerability and capacity assessment as required in the risk assessment equations. The results of these assessments are presented below. Hazard assessment Major hazards identiied at Roburnia Plantation include ires, harsh weather conditions, diseases, pests and pathogens (Table 1). If these hazards are not properly prevented and/ or managed, they have the potential to negatively affect the sustainability and productivity of the forest at the plantation. However, the degree to which the Roburnia Plantation is exposed to those hazards varies. As reflected in Table 1, the mean (± s.d.) shows that ire is the single major hazard to forest sustainability at the plantation, at 3 ± 0.0 (41%), followed by harsh weather conditions at 2.7 ± 0.4 (36%) and lastly tree diseases, pests and pathogens at 1.7 ± 0.3 (23%). The hazard assessment results suggest that there is high probability for ires to occur at Roburnia Plantation, and that when such ires occur they have potential to cause severe negative impacts. This implies that the management at Roburnia Plantation needs to develop plans and strategies to prevent the occurrence of these hazards. This supports the observation made by Wisner et al. (2004) that effective execution of disaster prevention plans and strategies could enhance resilience to hazards. Vulnerability assessment The Roburnia Plantation is highly vulnerable to the damaging effects of hazards such as ires, harsh weather conditions, tree diseases, pests and pathogens (Table 1). As relected in Table 1, the mean (± s.d.) shows that the Roburnia Plantation is highly susceptible to ire hazards at 2.6 ± 0.5 (43%), followed by harsh weather conditions at 2.1 ± 0.5 (35%) and lastly tree diseases, pests and pathogens at 1.4 ± 0.2 (23%). This means that if these hazards are not properly prevented, the plantation will be exposed to the hazards’ potential negative impacts, which in turn would have detrimental effects on people’s livelihoods. Other likely negative impacts include Original Research loss of life as a result of ires, destruction of infrastructure, loss of forest resources, and loss of employment. The results obtained from the vulnerability assessment therefore suggest that management at Roburnia Plantation needs to address the factors that increase levels of vulnerability in order to reduce the exposure and susceptibility to ire hazard risks. As also indicated by Wisner et al. (2004), reducing vulnerability is a daunting task and all resources and efforts need to be mobilised and marshalled towards this goal. Capacity assessment As relected in Table 1, the mean (± s.d.) results obtained from the capacity assessment show that the management at Roburnia Plantation have the necessary capacity required to deal with risks such as ires at 2.7 ± 0.4 (38%) and harsh weather conditions at 2.6 ± 0.4 (37%). The management personnel at the plantation indicated that they annually conduct public awareness campaigns to sensitise their oficials and members of the public on the dangers of ire. The idea is to inform them on how they can deal with ire incidents. They also indicated that they have a ire management plan which is reviewed annually in order to address gaps identiied during the year. The results show, however, that the management at Roburnia Plantation does not have suficient capacity at 1.7 ± 0.3 (25%) to deal with other hazards such as tree diseases, pests and pathogens (Table 1). These results therefore suggest that the management at Roburnia Plantation also need to re-work their plans in order to address other risks besides ire. Applicaion of Wisner and United Naions Internaional Strategy for Disaster Reducion risk equaions As stated earlier, the risk equations developed by Wisner et al. (2004) and UNISDR (2002) were used in the study in order to determine levels of risk at Roburnia Plantation. Applicaion of the Wisner equaion (Equaion 1) Hazard and vulnerability variables are considered in the Wisner equation as the important variables in determining the level of risk (Wisner et al. 2004). The study used hazard and vulnerability variables following the Wisner equation in order to determine the extent of the risks at Roburnia Plantation. The mean (± s.d.) analysis output shows that the level of risk varies. Fire is rated as the highest risk at 7.7 ± 0.3 (49%), followed by harsh weather conditions at 5.6 ± 0.4 (36%) and least by tree diseases, pests and pathogens at 2.3 ± 0.2 (15%) (Figure 1). TABLE 1: Hazard, vulnerability and capacity assessment at Roburnia Plantaion. Disaster risk Fires Hazard assessment Vulnerability assessment Capacity assessment mean s.d. % mean s.d. % mean s.d. % 3.0 0.0 41 2.6 0.5 43 2.7 0.4 38 Harsh weather condiions 2.7 0.4 36 2.1 0.5 35 2.6 0.4 37 Trees, diseases, pests and pathogen 1.7 0.3 23 1.4 0.2 23 1.7 0.3 25 s.d., standard deviaion. n = 8. htp://www.jamba.org.za doi:10.4102/jamba.v5i1.64 Original Research Page 4 of 6 Applicaion of the United Naions Internaional Strategy for Disaster Reducion equaion (Equaion 2) sources, namely experienced foresters and ire ighters. It follows therefore that possible errors or gaps in risk analysis can be reduced by interviewing experts, experienced and knowledgeable people. The UNISDR (2002) equation, like the Wisner et al. (2004) equation, also considers hazard and vulnerability variables as important in risk analysis, but differs in the sense that it also includes capacity to respond to the identiied hazards. The mean (± s.d.) analysis output shows that the level of risk varies (Figure 2). Figure 2 shows that ire is rated as the highest risk at 2.9 ± 0.2 (45%), followed by harsh weather conditions at 2.2 ± 0.3 (34%) and least by tree diseases, pests and pathogens at 1.3 ± 0.2 (21%). Impact of ires at Roburnia Plantaion Fires occur annually, affecting many hectares of plantation forest at Roburnia Plantation. The major causes of ires at the plantation include arson, honey gathering, heavy machinery and lightning. However, arson appears to account for more than 70% of ires at Roburnia Plantation. Most of these ires occur at midday and in the evening. The results of this study concur with other studies which indicated that human activities are mainly responsible for major ires (e.g. Granger 1984; Leistikow, Martin & Milano 2000; Working on Fire 2009). The study found that the two risk equations gave more or less the same results. The ANOVA single factor probability value between the two risk equations output was not signiicant (p = 0.13). This means that the meaning and interpretation must be similar. This is against the indings of Nogueira, Luqi and Nada (2000), who indicated that the outcome of risk analysis would not necessarily be consistent because different experts could arrive at different conclusions from the same scenario. The most likely explanation why in this study the two models gave similar results was that the data sets used in both models were obtained from the same This study has found that the tree species that are mostly affected by ires are gums, followed by pines. The gums are most affected by ires because they are highly lammable. They can burn even when the leaves are wet and green because the leaves of gum trees contain volatile substances which increase the intensity of ire. Number of ires and hectares burnt annually Riskraing rating at at Roburnia Risk Roburnla 9.0 9.0 The number of ires and hectares burnt in South African commercial forest plantations varies annually, but they are on the rise (DWAF 2008). This study found that the number of ires and hectares burnt at Roburnia Plantation varies annually. Such variation also had results in differences in annual ire extinguishing costs (Table 2). The most ire incidents, amounting to 218, were recorded during the year 2010. However, the mean (± s.d.) number of ires that occurred between 2007 and 2011 was 165.40 ± 56.19 (Table 2). The cause of the high number of ire incidents in 2010 is not clearly known by the management at Roburnia Plantation, but from the indings of this study it can be deduced that it could have been a result of the low amount of rainfall received in 2010. The total recorded rainfall for 2010 was 689 mm, which was far below the average of 908 mm calculated from 1975 to 2011. Low rainfall in that year could have created conditions suitable for more outbreaks of ires. 8.0 8.0 7.0 7.0 6.0 6.0 5.0 5.0 4.0 4.0 3.0 3.0 2.0 2.0 1.0 1.0 0.0 0.0 Fires Harsh weather Fires Tree diseases, pests & Tree diseases, pests & pathegens Harsh weather pathogens Disaster risk n = 8. FIGURE 1: Extent of risk at Roburnia Plantaion using the Wisner equaion. Riskraing rating at at Roburnia Risk Roburnla 3.5 3.5 3.0 3.0 2.5 2.5 2.0 2.0 Between 2007 and 2011, the largest area of forest destroyed by ire at Roburnia Plantation was 309.57 ha, which occurred in 2008. The mean (± s.d.) number of hectares burnt between 2007 and 2011 were 126.45 ± 110.81 (Table 2). 1.5 1.5 1.0 1.0 0.5 0.5 0.0 0.0 Fires Harsh weather Fires Tree diseases, pests & Tree diseases, pests & pathegens Harsh weather The highest cost incurred for extinguishing fires was R725 969.92 and this was incurred in the year 2010. The mean (± s.d.) amount incurred for extinguishing fires at Roburnia Plantation between 2007 and 2011 was R471 903.01 ± R256 330.32 (Table 2). The data presented in pathogens Disaster risk n = 8. FIGURE 2: Extent of risk at Roburnia Plantation using the United Nations Internaional Strategy for Disaster Reducion equaion. TABLE 2: Impact of ires at Roburnia Plantaion between 2007 and 2011. Impact of ires 2007 2008 2009 2010 2011 Min Max Average s.d. Number of ires 94 200 116 218 199 94 218 165.40 56.19 Hectares burnt Exinguishing costs (Rands) 25.98 309.57 91.97 62.72 141.99 25.98 309.57 126.45 110.81 587 143.84 93 710.15 331 567.84 725 969.92 621 123.30 93 710.15 725 969.92 471 903.01 256 330.32 s.d., standard deviaion. htp://www.jamba.org.za doi:10.4102/jamba.v5i1.64 Table 2 implies that management at Roburnia Plantation needs to effectively implement ire prevention plans and strategies in order to ensure that the number of ires, hectares burnt and costs incurred due to extinguishing fires are drastically reduced. This clearly demonstrates the need to conduct risk assessment so that effective risk prevention plans and strategies can be developed and implemented. Seasonal variaion in the number of ires and hectares burnt Data acquired from the study shows that there were seasonal variations in the number of ires and hectares burnt in 2011 (Figure 3). The ANOVA single factor analysis conirms that the number of ires is strongly correlated to the number of hectares burnt (p = 0.66). This means that an increase in the number of ires also results in an increase in the number of hectares burnt, and conversely, that a decrease in the number of ires also results in a decrease in the number of hectares burnt. This study found that the number of ire incidents was low during the summer months, but increased during winter and spring (Figure 3). This variation is mainly due to a converse relationship with rainfall. The availability of rain moistens the area as well as the fuel. During summer, the number of ires and hectares burnt are low because of the high moisture content, but the number of ires and hectares burnt increases during winter due to its low moisture content. These conditions inluence ire ignition and complete burn during winter and spring (Figure 3). Although the number of ires and hectares burnt increases from March, the highest peak was recorded in September (Figure 3). The findings of this study concur with other studies conducted in southern Africa, which indicated that the highest peak of ires occurs in August/September (e.g. Scholes, Ward & Justice 1996; Van Wilgen, Trollope & Everson 1990). However, some odd high incidents of ires might also occur in summer and autumn (Forsyth & Van Wilgen 2008), mainly caused by lightning. Original Research Average monthly rainfall (mm) Page 5 of 6 Number of ires Hectares burt Average monthly rainfall (1975–2011) 180 180 160 160 140 140 120 120 100 100 80 80 60 60 40 40 20 20 00 Jan. Feb. Mar. Apr. May June July Aug. Sept. Oct. Nov. Dec. Jan Feb Mar Apr Number of fires May Jun Month Hectares burnt Jul Aug Sep oct Nov Dec Average monthly rainfall (1975-2011) The data were compared with the average monthly rainfall (mm) measured from 1975 to 2011. FIGURE 3: Number of ires and hectares burnt at Roburnia Plantaion in 2011. plantation managers re-focus their strategies and deal with risk before the actual disaster event occurs. An important inding of the study is that even if different risk equations are applied to analyse the risks, the outcome of such analysis provides similar results. Although this inding needs to be further tested, it nevertheless renders credence to the use of models and modelling techniques in risk assessment, particularly in situations where the sources of data used are reliable and trustworthy. Acknowledgements The authors acknowledge the inancial support provided by the Department of Science and Technology (DST) and the National Disaster Management Centre (NDMC) towards the study from which this paper is an output. The management of Komatiland Forest are thanked for granting permission to conduct this study at Roburnia Plantation. Mr Roche Olivier (plantation manager), Mr Sizwe Gama (senior forester) and Ms Lindiwe Mthalane (forester) at Roburnia Plantation are acknowledged for assisting with all the data and information required during the study. We also acknowledge valuable comments provided by two anonymous reviewers. Compeing interests Conclusion The study proved that possible risks can be identiied by conducting disaster risk assessments. It was found that Roburnia Plantation is highly exposed to ire risks. A record of high frequency of ire occurrence, high exposure and vulnerability to ires at Roburnia Plantation suggests that there are high probabilities that this trend will continue in the future. The study further conirmed that data and information generated through risk assessment are essential for developing effective disaster risk prevention and control strategies. The advantage of preventing risk is that it significantly reduces the cost associated with disaster response, recovery and development, particularly in the long term. The current challenge hindering effective risk prevention plans at Roburnia Plantation, and in most plantations in South Africa, is that most resources are allocated for ire ighting and control, whilst meagre resources are allocated for risk assessment and prevention. Such a challenge requires that htp://www.jamba.org.za The authors declare that they have no inancial or personal relationship(s) that may have inappropriately inluenced them in writing this article. Authors’ contribuions R.A.M. (Limpopo Legislature) was responsible for experimental and research design. He also collected the data, analysed them and wrote the article. A.T.S. (National Research Foundation) designed and supervised the study. References DWAF, 2008, ‘Report on commercial imber resources and primary roundwood processing in South Africa’, Department of Water Afairs and Forestry (DWAF), Pretoria. DWAF, 2009, ‘Report on commercial imber resources and primary roundwood processing in South Africa’, Department of Water Afairs and Forestry (DWAF), Pretoria. Flannigan, M.D., Stocks, B.J. & Woton, B.M., 2000, ‘Climate change and forest ires’, The science of the total environment 262, 221–229. htp://dx.doi.org/10.1016/S00489697(00)00524-6 doi:10.4102/jamba.v5i1.64 Page 6 of 6 Forestry & Forest Product Industry, 1980–2011, ‘Recently released data as contained in the 2009/10 and 2010/11 ediions of DAFF’s commercial imber resources and primary roundwood processing staisics’, Forestry South Africa, South Africa, viewed 08 January 2013, from htp://www.forestry.co.za Forsyth, G.G. & Van Wilgen, B.W., 2008, ‘The recent ire history of the Table Mountain Naional Park and implicaions for ire management’, Koedoe 50, 3–9. htp://dx.doi. org/10.4102/koedoe.v50i1.134 Granger, J.E., 1984, ‘Fire in forest’, in P.V. Booysen & N.M. Tainton (eds.), Ecological efects of ire in South African ecosystems, pp. 177–198, Springer-Verlag, New York. htp://dx.doi.org/10.1007/978-3-642-69805-7_8 Jordaan, A.J., 2006, ‘Disaster risk assessment: Thumb sucking or scientific risk quantification?’, Disaster Management Institute of Southern Africa (DMISA), Hartenbosch. Original Research Scholes, R.J., Ward, D.E. & Jusice, C.O., 1996, ‘Emissions of trace gases and aerosol paricles due to vegetaion burning in southern hemisphere Africa’, Journal of Geophysical Research 101, 23677–23682. htp://dx.doi.org/10.1029/95JD02049 United Nations International Strategy for Disaster Reduction (UNISDR), 2009, ‘Terminology on disaster risk reducion’, United Naions Internaional Strategy for Disaster Reducion, Geneva, Switzerland. United Naions Internaional Strategy for Disaster Reducion (UNISDR), 2005, ‘Hyogo framework for acion 2005–2010: Building the resilience of naions and communiies to disasters’, Proceedings of the 2005 World Conference on Disaster Reducion, 18–22 January 2005, Kobe, Japan. United Naions Internaional Strategy for Disaster Reducion (UNISDR), 2002, ‘Living with risk: A global review of disaster reducion iniiaives’, World Bank, Geneva. Kaji, H., 2002, ‘The successful experience in disaster risk management in Japan: Focus on earthquake disaster’, Regional Workshop on Total Disaster Risk Management, 07–09 August 2002, Faculty of Policy Management, Keio University, Japan. Van Westen, C.J., Van Asch, T.W.J. & Soeters, R., 2006, ‘Landslide hazard and risk zonaion – Why is it sill so diicult?’, Bullein of Engineering Geology and the Environment 65, 167–184. htp://dx.doi.org/10.1007/s10064-005-0023-0 Leisikow, B.N., Marin, D.C. & Milano, C.E., 2000, ‘Fire injuries, disasters, and costs from cigaretes and cigarete lights: A global overview’, Prevenive Medicine 31, 91–99. htp://dx.doi.org/10.1006/pmed.2000.0680, PMid:10938207 Van Wilgen, B.W., Trollope, W.S.W. & Everson, C.S., 1990, ‘Fire management in southern Africa: Some examples of current objectives, practices and problems’, in J.G. Goldammer (ed.), Fire in the tropical biota: Ecosystem process and global challenges, Ecological Studies 84, 179–215. Morimiya, Y., 1992, ‘What is risk?’, Yobo-jiho, Japan Associaion of Fire and Marine Insurance, No. 168. Nogueira, J., Luqi, V.B. & Nada, N., 2000, ‘A formal risk assessment model for sotware evoluion’, Proceedings of the 22nd Internaional Conference on Sotware Engineering (ICSE 2000), Limerick, 04–11 June 2000, 2nd Internaional Workshop on EconomicsDriven Sotware Engineering Research (EDSER-2), Limerick, Ireland. htp://www.jamba.org.za Wisner, B., Blaikie, P., Cannon, T. & Davis, I., 2004, At risk: Natural hazards, people’s vulnerability and disasters, 2nd edn., Routledge, London. Working on Fire, 2009, ‘Integrated ire management: Fire awareness’, viewed 08 June 2009, from htp://www.workingonire.org/ifm_introducion.php doi:10.4102/jamba.v5i1.64