The hazard component uses data from the Ministry of Energy and Mineral Resources. This hazard is identified as a natural process that can cause disasters such as earthquakes, volcanic eruptions, floods, droughts and landslides. Thus, natural hazards are natural events arising from specific geophysical (Liu 2015).

As a natural event, the essential characteristics of hazards consist of space, time and magnitude (Rockefeller Foundation 2015; Ruggerio 2021; Ruiter & Loon, 2022). The spatial attribute identifies the location of the geophysical disaster and the extent of the affected area. The time scale also consists of two attributes: the time of occurrence and the duration of the event. Magnitude indicates the strength of the hazard event, including intensity, frequency and speed. The conclusion of the compilation of hazard components in this study is as follows:

The hazards considered in the risk calculation of Palu City comprise five types: liquefaction, tsunami, active fault, landslide and flood. Additionally, there are five disaster exposure classifications: very low, low, medium, high and very high (Figure 6).

The horizontal and longitudinal fault movement in the southern part of Kulawi, which crosses Palu City, caused a tsunami anomaly on 28 September 2018. The horizontal fault movement caused a tsunami wave that hit Palu City, resulting in a tsunami height of 6 m. The distribution of the tsunami flow depth varies. At the bottom of Palu Bay, along Talise Beach and the coastal tourist area of Palu City, the tsunami height ranges from 1.67 m to 5.55 m (Geological Agency 2018).

It impacted economic activity in Palu City because the dominant activities on Palu City’s coast were commercial, both formal and informal (Palu City Government, 2022). In addition, social facilities were also affected by the tsunami disaster. In addition, social facilities were also affected by the tsunami disaster.

Tectonic evolution on Sulawesi Island has resulted in a complex geological structure characterised by faults, folds and fractures at both sea and land levels. The Palu–Koro Fault is one of the geological structure patterns on land. The Palu–Koro Fault is divided into five segments from north to south: the Palu Segment, Gumbassa Segment, Saluki Segment, Moa Segment, and Meloi Graben Segment.

Liquefaction disasters are triggered by earthquakes, especially in areas such as the vicinity of the Purba River, which has a high potential for liquefaction (Geological Agency 2018). In 2018, massive liquefaction occurred in the Balaroa and Petobo areas, resulting in the emergence of an underground waterway.

Dhungana et al. (2025) stated that exposure is part of disaster risk. Exposure can be defined as the number, type and value of properties, infrastructure, environment, economy or population at risk under a hazardous event’s threat (Gill et al. 2022; Ortwin Renn et al. 2019). According to the UN Office for Disaster Risk Reduction (UNDRR 2017), exposure is a risk element that consists of individuals, groups of people or elements located in the vicinity of where the hazard event occurs (Jones & Tanner, 2015; Lindstädter et al. 2016; Otsuki, Jasaw & Lolig 2017). Castro et al. (2016) utilise elements such as population, buildings, infrastructure and economic centres to assess the level of exposure. In this study, population and building elements will be used to assess the level of exposure. This exposure component can be assessed based on the building plots exposed to the disaster, which are also used to determine the sample in this study. This exposure analysis is crucial for understanding the regional units or populations that are exposed to damage and both material and non-material losses.

The conclusion of the development and justification of the exposure components in this study is:

Figure 7 shows the delineation of exposure to settlement plots, public and social facility plots, economic activity plots, office plots and non-built-up areas. Thus, the level and type of exposure on each plot will differ and further differences will also be found in the sensitivity component. Other community groups also experience each exposure. For example, exposure to settlement plots is experienced by individual actors or households. At the same time, investor actors experience exposure to economic activity plots.

The exposure component utilises 30-m resolution Landsat imagery map data, building distribution maps and 1:5000-scale topography maps. From the 30-m resolution Landsat map, built-up and non-built-up areas can be identified. The building distribution map provides information on buildings in Palu City, including their functions. On the topography map, land use in Palu City can be identified. Based on the interpretation of the three data sets, land use synchronisation in Palu City can be implemented, supported by both secondary data and primary surveys. Land use reclassification in Palu City was conducted based on the results of field interpretation and observations. Then, the score of the reclassification results will be determined based on interviews with key stakeholders.

The exposure components are classified into five. The highest score indicates a high level of exposure. Likewise, the lowest score indicates a low level of exposure.

This study determines the highest score, namely settlement, with a score of five. This determination is due to the dominance of land use in Palu City, which is primarily characterised by settlement. Additionally, the 2018 disaster had the most significant impact on settlement areas. The lowest value is one in non-built areas. This determination is based on the fact that there are only a few settlements in non-built-up areas. So, it can be concluded that no one is exposed to disasters in non-built areas.

Sensitivity analysis applies to five types of disasters: liquefaction, tsunami, active fault, landslide and flood. The scoring and processing of spatial data on sensitivity components are based on the results of questionnaires distributed to the community, stakeholders and economic actors. The results of spatial data processing are in scoring reclassification into four reclassification (Figure 8).

The level of sensitivity can be identified from the characteristics inherent in each individual and society. These characteristics can be observed in various sectors, including economic, social, environmental, demographic and others. Sensitivity can be crucial in forming resilience, but it can also increase the risk of disasters. If individuals or communities lack the necessary character or resources to face disasters, then their sensitivity to disasters is likely to be higher. In other words, the risk experienced will be higher. However, if individuals and communities have the character or capital to face disasters, then their sensitivity will be lower. In other words, the risk experienced will be lower.

Figure 8 proves that the sensitivity of Palu City is high. Individuals and communities were not ready to face the catastrophic disasters. Even the community needed to recognise or understand the threat of catastrophic disasters in Palu City. They have yet to be informed about the disaster’s vulnerability in Palu City. So, they were unable to survive or recover quickly from the 2018 disaster. The Palu City area was also paralysed for several days after the disaster. The infrastructure is also not redundant, and there is no alternative. That is what caused the life of Palu City to come to a standstill after the disaster.

The highest sensitivity is in settlements’ land use, with an area of 6699.16 Ha. This classification is the widest, encompassing economic activities, public and social facilities and offices. This is because the number of exposed residents in settlements’ land use is greater than in other land uses. Settlements are also not ready to face the catastrophic disasters. As evidence of the 2018 catastrophic disaster, the worst damage was in the settlement areas.

Sensitivity analysis applies to five types of disasters: liquefaction, tsunami, active fault, landslide and flood. The level of resilience can be identified through the disaster risk reduction programme, specifically in the form of disaster mitigation. The Palu City resilience programme is based on the disaster risk assessment document, the Palu City spatial plan, the Palu City medium-term development plan and other relevant documents related to disaster mitigation in Palu City.

Researchers examine the resilience programmes to categorise them into resilience components: coping capacity, adaptive capacity, spatial planning and governance. The level of resilience is measured based on the plan and its implementation. If there is only a plan, the score is one; however, if there is both a plan and its implementation, the score is two. Seventy-six programmes have been grouped into resilience components. There are still programmes that need to be implemented, resulting in a low level of resilience. This affects the resilience score.

The classification of coping capacity, adaptive capacity, spatial planning and governance is based on several literatures. Coping capacity is an inherent characteristic or resource possessed by a system to deal with shocks resulting from disasters (Parsons et al. 2016). Adaptive capacity refers to the ability of a system formed through a learning process to withstand shocks because of disasters (Mentges et al. 2023). Spatial planning and governance are supported by policies and stakeholders to implement disaster risk reduction.

The following are some programmes as an effort towards resilience:

These five permanent dwellings are further discussed in the following section:

Figure 9 proves that Palu City’s resilience remains in the low classification. Disaster mitigation efforts aimed at achieving resilience are still limited and have yet to encompass the entire urban system.

Resilience with a high classification is associated with land use for economic activity. This is because the land use of economic activity has the funding to develop disaster mitigation infrastructure. In contrast, the ability of each individual to use settlement still needs to be higher.

Tabular and spatial data processing on hazard, exposure, sensitivity and resilience components are inputs for the Palu City Disaster Risk Index Analysis. The overlay is performed on the four components to produce disaster risk. The mathematical formula has been described. Figure 10 is a process assessment of the Palu City Disaster Risk Index.

As explained in point three, this Disaster Risk Index Calculation has different components from the initial calculation. The following are the differences.

Table 1 illustrates that the resilience-based Disaster Risk Index calculation is more detailed and accurate than the initial Disaster Risk Index calculation.

Disaster risk analysis goes through the stages of aggregation analysis and overlay of hazard, exposure, sensitivity and resilience components, as illustrated in Figure 10. High-classified risks are found in settlement land use and economic activities, the percentage of which is < 1% of the area. The Disaster Risk Index considering the resilience component is smaller than the Disaster Risk Index considering the capacity component, especially in coastal areas and city centres. The sensitivity level is included in the medium classification. This is because the government, the private sector and non-governmental organisations (NGOs) have initiated resilience programmes.

Figure 11 shows areas with high risk, such as: (1) industrial activities, containers and public facilities; (2) areas that are included in high-risk zones are open areas that function as tourist attractions, rather than settlement areas. The use of space with this design is developed by developers who have paid attention to disaster mitigation and (3) areas with multiple risks because there is more than one type of hazard in these areas with high sensitivity values, but low resilience values. The assessment of the multi-risk disaster index is also found in areas with multiple risks. This is due to the overlap of more than one type of hazard, for example, areas prone to liquefaction disasters overlap with areas prone to tsunami disasters and areas prone to active fault disasters.