As a large archipelago country consisting of thousands of islands, Indonesia faces various types of water-related disasters. This Chapter illustrates how strong institutional arrangements between a central disaster control tower and line ministries can lead to efficient disaster prevention. It also recommends land use system adjustments as a disaster prevention measure. Lastly, the Chapter covers several non-structural measures that can further reduce flood disaster risk, including flood forecasting combined with an early warning system and an integrated water information system.
Water Financing and Disaster Risk Reduction in Indonesia
Highlights of a National Dialogue on Water
OECD Studies on Water
5. Disaster risk reduction through non-structural measures
Abstract
Indonesia is vulnerable to water-related disasters that cannot all be mitigated through infrastructure such as dams, embankments, retention basins or channels. This chapter presents options for reducing disaster risks through non-structural measures. Non-structural measures are measures not involving physical construction using knowledge, practice or agreement to reduce disaster risks and impacts, in particular through policies and laws, public awareness raising, training and education. Common non-structural measures include building codes, land-use planning laws and their enforcement, research and assessment, information resources and public awareness programmes (UNDRR, 2022[1]).
This chapter analyses policy and institutional frameworks, illustrates water resource information systems including monitoring stations, and identifies areas for improvement in early warning systems in Indonesia. It reviews the meteorological and hydrological data for early warning and techniques related to flood forecasting. It also discusses the potential of satellite data in land use planning to reduce the risk of water-related disasters such as floods, and the establishment of an integrated information system that makes knowledge of the impact of land use change on exposure and vulnerability.
This policy analysis focuses on Indonesia's inherent problems and presents realistically feasible alternatives using modern Information and Communication Technology (ICT). As has already been proven in many cases, it is important to recognise that policies to prevent natural disasters are always insufficient with the efforts of the government alone: successful policies require engagement at all levels of government and civil society, including the private sector.
5.1. State of play: water disaster risk management in Indonesia
5.1.1. Natural disaster profile of Indonesia
A report by IBRD and WB (World Bank, 2021[2]) confirms that Indonesia is one of the most disaster-prone countries in the world. More than 75% of disasters are classified as meteorological or hydrological disasters. Indonesia loses more than 1% of its forests every year and ranks low behind India and China on the global environmental score. Thus, if action is not taken by 2045, the decrease of rice and palm oil production is expected to reduce GDP by 3.4%. If land degradation continues, inland flood disasters can cause a GDP reduction of 0.11%. It is also expected that GDP will decrease by up to 1.65% in the case of floods with a frequency of 50-year return period, and by 2.4% of land subsidence if excessive groundwater intake and SLR (Sea Level Rise) continue.
In Indonesia, various water-related natural disasters are hampering development benefits and affecting the population and landscape. Indonesia is vulnerable to 'slow onset disasters' such as coastal erosion, inundation, and delta subsidence, along with floods, droughts, landslides, tsunamis, tidal waves, and earthquakes (Republic of Indonesia, 2020; IBRD and WB, 2021). These risks have been greatly increasing due to anthropogenic factors such as deforestation, environmental degradation, and unsustainable water use, along with rapid urbanisation, agricultural activity and industrial development. Other disasters such as earthquakes, volcanic eruptions and tsunamis are also affecting water resources and water infrastructure (Figure 5.1).
Figure 5.1. Trends in disaster occurrence in the last 10 years
Source: International Bank for Reconstruction and Development and World Bank, Indonesia Vision 2045: Toward Water Security, 2021 (World Bank, 2021[3]).
In Indonesia, water-related disasters cause economic losses of 2-3 billion US dollars annually. More than 100 million people, or about 38% of the population, are exposed to flood risk, and 325 cities and regions are classified as high-risk areas. The number of flood events almost tripled, from 50 events in 2006 to 146 events in 2017 (World Bank, 2019[4]). The increasing flood risk, along with the lack of disaster-related information, in the downstream areas of rivers with growing urbanisation exposes urban areas to fatal actual risks such as dam collapses in the upper reaches. Flooding in Jakarta in 2020 affected 173,000 people. The flood risk burden is unevenly allocated to large cities and the poor and vulnerable (Government of Indonesia, 2020[5]).
Climate change is expected to increase the risk of natural disasters. Overall, the impact of climate change on water resources is expected to be significant, and it is projected to have a strong negative impact on society. Artificial disasters caused by unplanned developments in urban areas is prevalent and rampant. Additionally, climate change will affect water quality. But the biggest climate threat is SLR (sea level rise), which is expected to inundate extensive coastal areas and reduce GDP by up to 2.4% by 2045 (World Bank, 2021[3]).
5.1.2. Policies and practices of water and disaster risk management in Indonesia
The Indonesian government has developed policies to strengthen preventive measures and increase disaster resilience to protect people's lives and property from natural disasters. In particular, it focuses on strategies that maximise the effects of non-structural measures to compensate for the limitations of structural measures. Resilience-enhancing measures proposed by ‘Vision 2045: Toward Water Security’ (World Bank, 2021[3]) include: (a) optimising water usage and related development by aligning spatial planning with available resources; (b) reducing groundwater over-abstraction and consequent land subsidence, especially in urban and lowland areas; (c) reducing water pollution that is worsening with rapid urbanisation, industrialisation, and agricultural runoff; (d) protecting ecosystems, including watersheds and peatlands; (e) improving resilience to water-related disaster risks.
Indonesia's water authorities are responsible for surface water (Ministry of Public Works and Housing, MoPWH) and groundwater management (Ministry of Energy and Mineral Resources, MoEMR); water quality and catchment management (Ministry of Environment and Forestry, MoEF); spatial planning (Ministry of Agrarian Affairs and Spatial Planning, MoASP); provision of agricultural, residential and industrial water services (MoPWH); economic activities affecting water resources (MoEMR, Ministry of Finance, MoEF, MoMAF); water-related disaster prevention and management (National Disaster Management Agency, BNPB); drinking water quality standards (Ministry of Health, MoH). It is underlined that disaster-related policies must be integrated or coordinated to efficiently implement the national plan, and information dissemination and early warning systems must be strengthened, and community participation must be encouraged (World Bank, 2021[3]).
Due to its geographical characteristics, combined with high exposure to natural disasters along with the effects of climate change, Indonesia may suffer greater losses and damages in the future if proper management and response measures are not taken. In relation to this awareness of the situation, the Indonesian government has prioritised improving resilience to natural disasters and climate change in the Medium-Term Development Plan 2020-2024 (Government of Indonesia, 2020[5]). The country’s national policy priorities are (1) improving environmental quality; (2) increasing resilience to natural disasters and climate change; (3) the application of low-carbon development approaches.
Structural and long-term strategies for protecting and managing watersheds and extensive collaborative efforts are being emphasised by the Indonesian Government. Currently, there is a lack of coordination and cooperation between upstream and downstream. Therefore, water resource management tasks and disaster management measures should be further integrated. A priority should be the integration of land and water planning and management, integrating planning and management across the water sectors, and integrating agricultural development, water planning and wildfire management in natural land planning. It is necessary to strengthen information exchange among relevant ministries, central, provincial, and local governments. Policy coherence should be maintained across climate change adaptation, water and land management, spatial planning, ecosystem and biodiversity protection, and disaster risk reduction.
Asia Water Council (AWC) identified key factors hindering the implementation of disaster risk reduction and disaster management in Indonesia. These are: capacity enhancement, inter-agency disaster management cooperation, disaster-resilient land use planning, and impact-based early warning system. Considering the various issues pointed out in Indonesia's disaster management system, practical non-structural measures such as reducing disaster risk and improving disaster and climate resilience will be recommended. Specifically, related technologies and interpretation methods for meteorological and hydrological forecasting and disseminating disaster risk information will be also suggested in Chapter 5.2, 5.3 and 5.4. The National Dialogue on Water pointed to the following barriers in the implementation of disaster risk reduction measures:
1. There is a lack of momentum to promote a series of institutional efficiencies leading to disaster prevention, management, response, and recovery. As highlighted in various documents such as the responses to the questionnaire (Government of Indonesia, 2022[6]) and the above-mentioned national plans, severe restrictions on information sharing exist. Cooperation between the central and local governments and NGOs that play a key role in the value chain of disaster risk management, such as coordinating work in disaster prevention and management and sharing information or data, is also limited.
2. Effective service provision for disaster prevention and reduction are not systematic in terms of scope, content, scale and method. Moreover, hardware infrastructure, such as measurement and monitoring, and software infrastructure, such as data and information analysis and communication, are insufficient.
3. In Indonesia, the technical complexity of hydromet services, remains a significant challenge. Robust hydromet services are the most basic and essential requirement for the prevention and management of natural disasters.
4. Since the information integration of disaster-related inter-agency depends on inter-agency coordination, a lesson must be learned from the lack of a centralised medium to facilitate data and information.
5. In the case of Indonesia, various negative impacts such as extreme weather events due to climate change are expected to occur in the future, so stable finances and investment towards an information-oriented society are essential to ensure improved quality of climate data and flood forecasting.
6. To increase the effectiveness of a timely and sustainable “last mile” in disaster warnings, it should be carefully considered that even if a national disaster early warning system exists, the “last mile” will not necessarily reach communities or translate into effective early action. This can be seen in the recommendations of international organisations related to overcoming disasters, such as WMO and UNDRR, and the four priority measures of the Sendai Framework. Since people-centred early warning systems are based on risk knowledge, it is necessary to develop ideas to strengthen and improve knowledge management.
Indonesia’s unique geographic location makes flood risk management even more complex. Indonesia is located in a tectonic zone characterised by intense seismic and volcanic activity on the south-eastern edge of the geologically Eurasia continent. A long period of tectonic, seismic, and volcanic activity has created the topography of Indonesia today. Its features include long and narrow islands, high mountain ranges formed by active volcanic and orogeny activity, and rivers and coastal zones formed by erosion and deposition. These topographical features have become a factor that determines the way of life of the Indonesian people in terms of meteorology and hydrology, and natural disasters are part of life. Moreover, the topographic characteristics are untraceable and difficult to manage in terms of disaster risk management.
5.1.3. Review of international lessons learnt on disaster risk management
A survey of the Disaster Risk Governance (DRG) initiative found that there is a global shift from top-down disaster response to a more inclusive form of DRG. This trend has been observed in policies and organisations (Srikandini, Hilhorst and Voorst, 2018[7]). Especially in Indonesia, the role of civil society is well recognised. UNDRR (UNDRR, 2022[1]) provides clues to the direction of disaster management policy in Indonesia through its extensive reflection on the limitations and opportunities of systematic risk governance. It points out the need to improve the existing approach by reflecting on various disaster events in Indonesia.
In particular, community responses and systemic inefficiencies during disaster hazards have been identified as “last mile”, which is why the end-to-end early warning system must be strengthened by bridging the gap between the upstream and downstream of the early warning chain to ensure timely action (UNDRR, 2022[1]; UNDRR, 2021[8]). There have been several definitions of “last mile”: one is “last mile” as a challenge for rural communities to access media, which can be overcome by supplementing traditional media channels for warning dissemination with additional technologies (LIRNE Asia, 2008[9]); another one is ‘last mile’ as the capacity of the community to take action in response to a received warning, which supports the development of the capacities of local institutions (Bedi, 2006). In this report, the definition of “last mile” can be applied in a comprehensive sense of both definitions from the perspective of advanced information sharing in consideration of hazard arrival time or lead time for response.
The policy reform for reducing natural disaster damage and improving climate resilience in Indonesia still needs to be initiated. Disaster response capabilities can be assessed in 5 major sectors. These are 1) governance, 2) information/data and analysis, 3) strategic planning, 4) resources, and 5) monitoring and evaluation. Institutional capabilities can expect greater effects when all these areas are properly developed, and experiences and know-how are collected. Especially, institutional capacity is important not only for comprehensive resources within the organisation but also for digital resources that support them.
One example of the disaster response capability is to improve the accuracy of forecasts and warnings through technological advancement in the field of hydrometeorology and improvement of flood prediction models by means of various information provision methods such as satellite images and rainfall radar. Mobile devices, smartphones, and social network (SNS) have progressed from a one-way information delivery system (based on public media such as TV, radio, village speakers or direct delivery means) to a smart environment where information can be shared in two-way or even multiple-way communication. Smart communication methods fit very well with the four priorities for disaster risk reduction of the Sendai Framework for Disaster Risk Reduction.
As the World Meteorology Organisation (WMO) emphasises, flood forecasting and warning services should operate in conjunction with the National Weather and Hydrology Service (NMHS), which supports civil protection and emergency response services (WMO, 2010[10]); (Hodgkin, 2016[11]) (APFM, 2022[12]). Prediction and warning services are in most cases based on information systems developed to provide reliable and timely information to the public as well as to private protection services. It should consist of institutional and technical functions that ensure sufficient lead time for stakeholders to take measures to protect themselves from floods or to take appropriate measures and enable disaster management agencies to intervene and take appropriate measures.
5.2. Recommendations to integrate disaster management and land use planning
The integration of disaster management functions is intended to be reviewed with emphasis on coordinating and managing overlapping or conflicting functions in the control tower, while responding to disaster management by individual departments, rather than managing all disasters in one control tower. Disaster prevention through land use planning suggests disaster-resistant land use planning based on foreign cases and Korea's experience.
5.2.1. Coordination of disaster management functions
More than ten percent of Indonesia’s surface area is subject to a relatively high mortality risk from multiple hazards. However, the percentage of Indonesian people living in these areas is close to sixty percent (Dilley, 2005[13]). this is mostly due to the concentration of risks on the island of Java, which has by far the highest population density and bears a high natural disaster risk (Rossum and René, 2010[14]).
Under these difficult circumstances, the Indonesian government has made various efforts to improve disaster management. The Indonesian government plans to move the capital to East Kalimantan on the island of Borneo by 2045 as part of a comprehensive strategy of balanced national development and reduction of the impact of disasters on Java Island.
According to the special law on Disaster Prevention (Undang-Undang Penanggulangan Bencana No. 24: UUPB) in 2007, Indonesia is a region prone to various natural disasters such as earthquakes, tsunamis, volcanic eruptions, floods, droughts, typhoons, and landslides.
After the 2004 Sumatra tsunami, the Indonesian government realised the need for a stronger and more integrated disaster management organisation. In 2008, President Susilo B. Yudhoyono established the National Agency for Disaster Countermeasure (BNPB), a Ministry dedicated to disaster prevention. It is implementing effective disaster prevention policies by linking to various ministries within the government. However, it is evaluated that many Indonesians are not protected from disasters due to the wide spread of Indonesian territory, the occurrence of various disasters, and the lack of capacity and financial resources of related organisations including, BNPB (Park and Lim, 2017). It can be concluded that Indonesia's related laws and systems have been established to some extent with the cooperation of international organisations and many foreign specialised organisations.
The tsunami shifted the disaster management paradigm towards prevention and mitigation and led to the establishment of a National Disaster Management Authority and Regional Disaster Management Authorities. The law is a significant improvement, but many problems remain, including the overlapping jurisdictions of the central and regional governments as well as difficulties in coordinating disaster management activities. These must be overcome to prevent avoidable loss of life and property resulting from natural disasters.
Even though BNBP plays a role as control tower in the disaster management system, Indonesia is still vulnerable to disasters due to several reasons:
Although BNPB is in charge of all disaster operations, it still seems to lack the ability to coordinate and cooperate with the work of individual ministries directly responding to disasters. BNPB is directly responsible to the President of Indonesia and the chairman is directly appointed by the President. It provides guidance and direction on disaster management efforts that include disaster prevention, emergency response, rehabilitation, and reconstruction in a fair and equitable manner and assigns the standardisation and implementation of disaster management needs based on laws and regulations.
Promoting disaster preventive projects remains a challenge as the necessary financial resources have not been secured. Still, the priority of financial resources invested in disaster management is falling behind from the national financing priorities. Experience from Korea shows that significant investment can tremendously reduce disaster risks. Damage from typhoons and floods continued to occur in Korea until the end of the 1990s. To overcome flood disaster risk, the government dramatically increased the annual budget related river management from of 0.23 billion U.S. dollars, to 0.77-1.2 billion U.S. dollars since 2000(Internal data of Ministry of Land, Infrastructure and Transport of Korea). Owing to the increased fiscal budget the damage caused by river flooding has decreased, although the risk of disasters due to climate change has recently increased.
BNBP must make efforts to secure a stable budget for disasters. To this end, it should be known that disaster budget investment is an important factor in creating more national wealth in the long run. If there are difficulties in securing the government budget, it is necessary to seek ways to actively attract private capital. In general, there are limits to attracting private capital in disaster-related projects due to their strong nature as public goods. For example, in order to attract private capital to invest in river flood prevention, one way is to give development rights around rivers to private businesses. Land Value Capture as set out in Chapter 4 could be a relevant financing tool in this context.
Since Indonesia is a country composed of many islands, the central government has limitations in preventing and responding to all disasters. To overcome this issue, the BNPB's regional office’s function in each major region must be strengthened, and continuously establish a rolling plan for disasters through communication with the local government and community. Local office of BNPB also continuously promote the implementation and evaluation of the established plan.
Considering the overall situation, it is most important for BNPB to strengthen its role in coordinating and evaluating the work of various ministries related to disasters. At the same time, the local organisations of the BNPB - the BPBDs - should provide technical support for disaster-related tasks of local governments so that the local governments perform comprehensive functions for various projects, whether these come from central government and local government.
In particular, community-based disaster management, where disasters actually occur, can minimise human casualties through rapid evacuation and efficiently execute disaster recovery in coordination with the central government, local governments, and foreign aid. What is more, coping with disasters by relying on the ability of the central government alone is no longer effective because of resource limitations. Therefore, involvement of the entire community in disaster management is essential. Section 5.4 provides some suggestions on ICT applications for community-based disaster management.
5.2.2. Disaster-resistant land use planning
Land use planning is a major tool for reducing risks from natural hazards, and aiding sustainability and increasing resilience. Risk-based land use planning not only minimises damage in the event of a disaster, but also provides rapid resilience.
Conventionally, planning relies on the probability of occurrence of a disaster. However, this method tends to underestimate the ability to assess and quantify the consequences of various disasters, including the number of fatalities per year. If the probability of a disaster comes out low, it may encourage decisions that put society at risk, such as over-developing disaster-prone areas. This is because even if the probability of occurrence of a disaster is low, its impact may be very serious.
The key to risk-based planning is to be able to distinguish between different levels of risk (e.g., acceptable, tolerable, intolerable) and link them to appropriate land use policies. Acceptable risk levels should be based on measurable levels (see also Figure 5.3). Indicators to monitor risk levels ensure that acceptable levels of risk are not exceeded, and efforts are made to mitigate existing risks so that towns and cities can achieve sustainable development. The risk-based planning approach (RBPA) proposed by Saunders and Kilvington (Saunders and Kilvington, 2016[15]). has many implications for planned land use planning.
Risk-based land use planning has been evaluated as very encouraging by the United Nations Office for Disaster Risk Reduction (UNISDR) (Saunders and Kilvington, 2016[15]), and it is considered as an essential technique for mitigating risks in existing development and prohibiting or strengthening development in extreme risk areas.
Risk-based land use planning can also be used for risk-based planning assessments of the impact of a particular natural disaster that triggers another disaster (e.g., a tsunami or landslide caused by an earthquake). It can be quantified as part of future planning decisions about the economy, society, culture, infrastructure, and security resulting from a particular development.
5.2.3. International experiences with risk-based planning approaches
Risk-based planning is used all over the world, including Australia, Canada, European Union, Hong Kong, United Kingdom and the United States. During the development of the risk-based planning approach (RBPA) in New Zealand (between 2007 and 2013, see Box 5.1), other countries were developing similar risk-based frameworks. For example, Australia's Queensland Reconstruction Authority published a risk-based approach to flooding in 2012. The Geological Survey of Canada prepared a risk-based land use guide for risk assessment in 2015.
Box 5.1. Disaster-based land use planning in New Zealand: a matter of coordination
In New Zealand, multiple authorities are responsible for natural disasters. Ministry for the Environment (providing national regulatory and non-regulatory guidance), local councils (responsible for regional or watershed-scale policy frameworks), and land authorities (responsible for specific land-use designs and decisions are city and zoning committees), civil defence emergency management groups (disaster prevention and response), and national civil groups (social infrastructure management) share responsibilities complementary to each other.
Cooperation between these agencies is key to ensuring a streamlined and holistic national approach to planning for natural disasters and catastrophes. In New Zealand, which is surrounded by seas like Indonesia, there are four important laws that contribute to natural disaster management: the Resource Management Act (1991), Building Act (2004), Civil Defense Emergency Management Act (2002), and Local Government Act (2002). These laws are intended to operate in an integrated way to achieve the common goal of sustainability. The Resource Management Act is an important law for regulating land use planning.
Under the Resource Management Act, sustainable management enables people and communities to provide for their social, economic, and cultural well-being and for their health and safety while sustaining the potential of natural and physical resources to meet the reasonably foreseeable needs of future generations. The law deals with governance at three levels: national, regional and district. Risk-based planning helps regional and regional authorities mitigate the impact of natural disasters through land-use planning. The mitigation measures required by the district level under current resource management laws are those that can result in a reduction in disasters.
RBPA comprises five stages. The key is that stakeholders and affected communities must participate at each stage of the risk analysis, evaluation, and decision-making process. The five steps are to recognise the disaster risk in the area, determine the degree of disaster impact, evaluate the probability of occurrence of a disaster, select a risk-based approach, and finally go through the stages of monitoring and evaluation Figure 5.2. Communication and engagement about the disaster is essential to risk assessment and decision-making. In a broader sense, once the risk management process is established, it becomes a series of processes in which a risk-based planned approach is made. Figure 5.2 compares the standard risk management process with the risk-based planning approach, and shows that each step of the risk-based approach is consistent with the risk management standard (Saunders and Kilvington, 2016[15]).
A risk-based land use plan is finalised through the creation of a matrix on the likelihood and consequences of disaster occurrence Figure 5.3 shows how land use planning should be controlled by creating a matrix for the order of probability of occurrence (very high: 5) and the magnitude of the impact according to the severity level (largest: 5).
RBPA is a robust, transparent, and participatory framework for decision makers to determine risk levels. Policies are drafted on a risk basis, evaluated over time, and ultimately reduce the loss of people and property. RBPA is a method that considers the theoretical and practical challenges of adopting a risk-based approach to land use planning.
Figure 5.2. Comparison of risk-based planning and risk-based management process
Source: Saunders and Kilvington, 2016, Innovative land use planning for natural risk reduction: A consequence-driven approach from New Zealand, International Journal of Disaster Risk Reduction (Saunders and Kilvington, 2016[15])
Figure 5.3. An example of Levels of Risk and Associated Levels of Land Use Control
Source: Saunders and Kilvington, 2016, Innovative land use planning for natural risk reduction: A consequence-driven approach from New Zealand, International Journal of Disaster Risk Reduction (Saunders and Kilvington, 2016[15])
5.2.4. Nature-based solutions as low-impact development
Low Impact Development (LID) is a term used to describe a land planning and engineering design approach to manage storm-water runoff as part of green infrastructure. LID emphasises conservation and use of on-site natural features to protect water quality. This approach implements engineered small-scale hydrologic controls to replicate the pre-development hydrologic regime of watersheds through infiltrating, filtering, storing, evaporating, and detaining runoff close to its source.
In areas with high disaster risk, land is best used as a nature-based solution that can act as a low-impact development, wetland or floodplain to prevent flooding, improve water quality, and create walking trails and wetland trails for leisure activities of residents.
As a result of applying the LID technique to the housing development site of Korea's LH Corporation, it shows that the efficiency of urban water circulation increased by 41 per cent (Lee, 2016[16]). Bai et al. (Bai et al.[17]) also show a 32.5 per cent reduction in runoff at peak times during rainfall and a 31.8% reduction in overall flooding.
5.2.5. Some proposals related to institutional improvement
Reinforcing disaster response governance capabilities
Even if BNPB is the control tower, it is part of a cooperative governance system among various disaster management entities. Large-scale disasters lead to a wide variety of disasters. Since various disaster management entities have no choice but to participate, cooperative work among them is more important than anything else.
The UUPB is the highest law for overcoming disasters in Indonesia, and it is necessary to clearly specify the authority and responsibility of the central and local governments in each stage of disaster management, such as prevention, emergency response, and recovery. At the same time, it is necessary to explicitly stipulate in the law how to urgently secure financial resources for disaster recovery in the event of a disaster.
It is necessary to establish regular (5-year) plans for various sectors to prevent disasters and strengthen the system for evaluating the implementation results. Guidelines that local governments can refer to should be presented, and based on these, specific comprehensive measures should be established by the local government.
Strengthening the capacity of local governments
Local governments must establish disaster prevention measures at the local level and establish consistent measures with central government’s plan. The Japanese model could be relevant for Indonesia. In Japan, regional disaster prevention plans are established every year for the purpose of improving regional disaster prevention capabilities. When a disaster occurs, it is emphasised to support the vulnerable by utilising private resources (manpower, organisation, resources, logistics, etc.) within the region (Tokyo Disaster Management Council, 2014[18]).
Box 5.2. Participation of the private sector to improve Tokyo’s disaster preparedness capacity
The Tokyo Metropolitan Government, the capital of Japan, aims to improve Tokyo's disaster preparedness capacity in order to protect the lives and properties of citizens and maintain urban functions. To this purpose, relevant measures have been prepared with the goal of creating a city that is resistant to earthquakes, creating a risk management system that protects the lives of citizens and urban functions, and creating a system that supports the lives of victims and regenerates Tokyo in a short period of time. The foundation is based on self-help, mutual assistance, and public assistance based on the participation of the private sector.
Measures for prevention emergency and recovery are: 1) improvement of disaster resilience of residents and regions, 2) realisation of safe city development, 3) securing of safe transportation networks and lifelines, 4) countermeasures against tsunami, etc., 5) enhancement of emergency response capabilities from a wider perspective 6) securing information and communication, 7) measures such as medical aid, 8) measures for those who are unable to return home, 9) measures for victims of the disaster, 10) promotion of logistics and transportation measures, 11) measures against radioactive substances, and 12) short-term reconstruction of residents' lives. In particular, as Tokyo, with a population of more than 13 million, is a large city with a very high population density, human casualties can greatly increase if a large-scale earthquake occurs. In order to minimise casualties, major activities that must be carried out within 72 hours after an earthquake are listed hourly in detail.
Pre-planning and post-development for urban land uses
In order to prevent reckless and disorganised development and to achieve environment-friendly disaster-safe development, it is necessary to first establish a specific plan for development before proceeding with development. To this end, vulnerability assessment according to changes in land use such as urban reorganisation, new town construction, redevelopment, etc., should be made mandatory.
In addition, in case of any land use changes due to infrastructure construction activities such as industrial bases, airports, and ports of all central and local governments, it is necessary to specify in individual laws to carry out disaster impact assessment in advance so that the principle of pre-planning and post-development can be settled down.
In Korea, the disaster prevention basic plan manages disaster risks based on districts. The plan is led by the National Emergency Management Agency and the Ministry of Environment. The main components of the plan Natural Disaster Countermeasures Act in Korea (2022) are:
Implementation of Disaster Impact Assessment for the land development over a certain scale
Designation of natural disaster risk improvement district and establishment of maintenance plan
Establishment and operation of disaster prevention performance targets by region considering regional characteristics
Disaster prevention training for the civil servants and technicians
Establishment of disaster information system & emergency support system
According to the National Territory Plan and Use Act in Korea (the top law for the use, development and preservation of national land), disaster vulnerability analysis should be included in urban management plans, which is established every 5 years. For example, Indonesia’s UUPB could specify that the disaster risk in disaster risk areas should be assessed every five years and the mitigation methods should be established.
Risk-based land use planning in river basins
Legal support is needed to conduct disaster impact assessment on existing land use, focusing on areas vulnerable to disasters. In order to induce innovation in land use (Saunders and Kilvington, 2016[15]), the probability of disaster and the size of the impact of disaster should be matrixed. Land development must be banned in areas with high frequency of disasters and severe consequences.
It is essential to promote the relocation support of residents in areas with high disaster risk through the support of the central government and re-designate the area as a floodplain or disaster mitigation area.
In Korea, under its River Act, a basic river plan is established every 10 years to reduce damage caused by flooding of rivers. River plans are established in consideration of changes in rainfall caused by climate change and urban development.
The following figure shows the designation of river areas through the river basic plan. In the past, it was agricultural land, but if the planned flood level of the river increases due to climate change, these areas are designated as rivers.
Figure 5.4. Adjustment of river areas by river basic plan
Source: Author(Asia Water Council internal paper)
As mentioned above, land use planning that reflects the probability of occurrence of disasters and the impact of disasters is very important in disaster management. Based on past disaster experiences, it is necessary to manage risk areas through this approach to disaster risk areas. In order to push forward more strongly, it is necessary to specify disasters in this direction in the relevant laws.
If various disasters such as floods, landslides, and earthquakes are expected in a specific area, first, each disaster impact matrix is created, second all matrices are summed with the same or different weights. It is desirable to weigh each disaster type through the AHP (Analytic Hierarchy Process) technique.
Planning based on zones or districts
At the same time, it is necessary to implement strict land use regulations by actively introducing the zoning system for the sustainable urban development and growth. A good example is New Zealand's planning system is grounded in effects-based Performance Zoning under the Resource Management Act (Memon and B. Gleeson, 1995[19])
In Korea, the National Territorial Planning and Use Law stipulates that the protection of people's lives and property in response to climate change and natural disaster is regarded as one of the important principles. In addition, the designation or change of land use districts is determined by urban and county management plans. This is because local governments are well aware of the disaster-prone areas in their area. In addition, disaster prevention districts are also designated for the prevention of storm and flood damage, landslides, and ground collapse.
In order to mitigate natural disasters caused by typhoons and floods, the Natural Disaster Countermeasures Act requires the head of a local government to conduct a disaster impact assessment before permitting development projects such as urban development, river use and development, and mountain development. At the same time, natural disaster risk improvement districts are designated and managed for habitually flooded areas and landslide risk areas.
Finally, it is necessary to increase the integrity of the water cycle by inducing development through the LID technique. The Seoul Metropolitan Government has been enforcing the Basic Ordinance on Water Circulation Recovery and Low Impact Development since 2020. The main purpose of this ordinance is to preserve the natural infiltration ability of rainwater based on related laws. Its purpose is to provide basic directions for LID development for the restoration of the natural water cycle deteriorated by urbanisation and the preservation of the water environment. It is necessary to enact laws or ordinances for efforts to restore the soundness of urban water circulation in Indonesia.
5.3. Recommendations to strengthen the flood forecasting system
This chapter recommends flood forecasting systems as one the most useful non-structural measures for flood disaster mitigation. Three shifts in the flood forecasting paradigm are discussed: 1) a shift from conventional technology to information and communications technology, 2) a shift from forecasting along the river to spatial forecasting, and 3) a shift from simulation models to the adoption of artificial intelligence.
5.3.1. Hydrological and meteorological data collection and management
For accurate flood forecasting, hydrological and meteorological data collection is essential. This includes data on rainfall, evapotranspiration, soil moisture, river stage, and discharge. Data monitored during a flood event are not only used for real time flood forecasting, but also for establishing or refining forecasting models. For this, the collected data must be properly managed. As much as the data quality, adequate frequency of data collection needs to be assured. Hourly data could be acceptable for flood forecasting for large rivers. However, data of every ten-minutes are required for small rivers and local urban floods.
With the advances in image processing and related measurement technology, stream discharge measurement is opening a new era. For example, Indonesia uses CCTV as a surveillance tool in several reservoirs. CCTV based automatic wireless discharge measurement technology enables unmanned, non-contact, automatic real-time monitoring of water level and discharge (see Figure 5.4). CCTV is a safe, non-contact method because no manpower is needed, and there is no flow disruption during the measurement. Since it allows continuous real time measurement, peak discharge measurement is always possible, and this is very useful if the measurement purpose is, e.g., to develop a stage-discharge relationship.
Indonesia could consider expanding the use of CCTV technologies for additional purposes, such as in an urban context. CCTV measurement technology is useful especially for small mountainous streams and urban streams. In view of flood forecasting standpoint, all the on-site information including video images can be transmitted through cell phone to the people downstream so that they can take necessary actions. Having suffered severe urban flooding in several cities including Seoul, Korean government is planning to install this kind of surveillance/measurement system in urban areas, and this could be also useful for flash flood forecasting/warning for Indonesian urban streams.
Table 5.1 shows data collection frequencies for hydrological and meteorological variables mostly related to flood forecasting. In Korea, the four Flood Control Offices (FCO) are mostly responsible for the collection of hydrological data, and Korea Meteorological Administration (KMA) collects and manages the meteorological data. Rainfall and stage data for rivers and dams are open to public real time through FCO and KMA websites and cell phone apps. Most of the discharge data are converted from the stage data using stage-discharge rating relations. However, FCO also collects so-called automatic discharge data for some locations, which is measured by electric wave surface current meters. All the collected hydrological and meteorological data are disseminated every year in the form of a yearbook. The yearbook from 1960 to the present can be downloaded from the website of the Han River Flood Control Office.
Table 5.1. Data collection frequencies for hydrological and meteorological variables
Source: Hydrologic Survey Works Handbook, internal document of Han River Flood Control Office (written in Korean)
Figure 5.5. Example of CCTV based Automatic Discharge Measurement Technology
Source: (Yoon. B, 2022[20])
5.3.2. Spatial flood forecasting based on flood hazard maps
Flood forecasting has been largely made for major locations (points) of interest along a river. However, the need for spatial flood forecasting, i.e., forecasting for regions (areas), is growing. Flood hazard maps can contribute to improving the flood forecasting system by presenting the flood forecast for an entire area based on real time data.
Flood hazard maps provide information of expected flood depth and extent of river flooding (over-flowing) or urban inundation in the form of paper or electrical maps. Flood hazard maps can be categorized into river flood hazard maps (caused by river flooding) and urban flood hazard maps (caused by rainfall events exceeding drainage capacity) as shown in Figure 5.5. The spatial range of river flood hazard maps includes main river reaches, tributaries and expected flooding areas depending on flood scenarios. The spatial range of urban flood hazard maps include total drainage zones and flooding areas due to exceeding rainfall events, water level rise in receiving water bodies, and pump failures. Flood hazard maps are an example of non-structural flood mitigation measures. They provide information on potential flooding areas to regional governments and relevant authorities which can support in effective disaster prevention such as evacuation, flood insurance, land use regulation, and so forth.
Figure 5.6. Conceptualisation of river flood and urban flood inundation
Source: Lee et al., Catalogue of Hydrologic Analysis for Asia and the Pacific (2019[21])
Box 5.3. Flood hazard mapping in Korea
In Korea, the process of flood hazard mapping includes the following activities:
Site surveys that include collecting basic hydraulic and hydrologic information, relevant plans, flood damage history and existing survey data for the target river or drainage areas.
Topological data construction is a process to obtain topological information such as ground elevation for inundation analysis and, hence, it should guarantee the required accuracy and precision for inundation analysis.
Flood scenario development, consisting of catchment conditions, flood scale, and inundation scenarios.
Flood and inundation analyses are performed based on flood scenarios with a methodology determined by land use and land cover. Flood inundation results are obtained from one- and/or two-dimensional hydraulic modelling. In Korea, the HEC-RAS model is widely used for one-dimensional analysis, and the FLUMEN model is used for two-dimensional unsteady flooding simulation.
Data base system: these results are mapped as a flood hazard map and kept in a database system.
Flood hazard maps are distributed to central and regional governments for the purpose of flood prevention and mitigation policies and activities. They can be used for establishing emergency action plan, natural disaster insurance management, and integrated planning of natural disaster mitigation and management. The Ministry of the Interior and Safety (MoIS) utilises flood hazard maps as a basis for the hazard map, flood insurance map, life safety map, and so forth. Regional governments utilise flood hazard maps as a basis for flood insurance maps and for flood-prone area management. Flood hazard maps can provide basic information for natural disaster management policies and reduce the budget and effort at the same time.
Source: Lee et al., Catalogue of Hydrologic Analysis for Asia and the Pacific (2019[21])
Spatial flood forecasts can be based on scenarios and dynamic spatial flood forecasting or on real time flood forecast. Both methods are compared in Table 5.2. Flood hazard maps can contribute to improving the scenario-based spatial flood forecasts, because the development of flood hazard maps is based on a scenario-based procedure. The database built for flood hazard maps can be directly used to improve the spatial flood forecasting.
In addition, real time spatial flood forecasting can be obtained by real time flood simulation and long-term analysis. However, current computing power limits the application of real time flood simulation due to lengthy simulation time, which deteriorates the accuracy, proper timing and reliability of flood forecasting. It is expected that substantial amounts of technical advances and infrastructure are necessary to accomplish this. Currently, the scenario-based spatial forecasting system, combined with river flood hazard maps and urban flood risk maps, is regarded as the best alternative for improving flood forecasting systems. It is expected that real time dynamic spatial flood forecasting system would be possible in the near future with technical advances. Figure 5.6 shows an advanced example of a forecast system based on the flood risk map.
Table 5.2. Type of flood forecasts and their features
Source: Lee et al., Catalogue of Hydrologic Analysis for Asia and the Pacific (2019[21])
Figure 5.7. Korea’s flood forecasting system
Source: Lee J. Seo Y, Park S et al, 2019 Catalogue of Hydrologic Analysis for Asia and the Pacific | Volume 1 Flood Hazard Mapping Flood Hazard Map of Korea (Lee et al., 2019[21])
5.3.3. Artificial intelligence-based flood forecasting
Physically-based flood simulation models have been widely used for flood forecasting purposes. Physically-based hydraulic or hydrologic models are based on an understanding of the physical process that occurs within a natural system. The model reproduces the dominant processes through governing equations that are derived under a few simplifying assumptions. Typically, it is not possible to include each process within a simulation model owing to various simplifications and assumptions, and thus, the resulting model can only approximate a real-world system. Errors between the computed and observed values are inevitable due to various uncertainties including those of the model itself, hydrological initial and boundary conditions, and model parameters. All these uncertainties simultaneously affect errors in the model output that produce poor model performance despite a thorough understanding of the governing laws.
In contrast to physically-based models, data-driven (artificial intelligence, machine learning) models are not based on either a preconceived conceptualisation of the behaviour of the system or an explicit representation of discrete physical processes. Conversely, in data-driven models, system response is characterised by exploiting statistical information in a set of time-series data. Furthermore, data-driven models can use the advantages of whatever relevant data are available, and this allows such models to represent particular processes. With the accumulation of enormous amount of data due to the progress of sensor and Internet of Things technology, together with the computing power, deep learning models represented by artificial neural network (ANN) are widely used. The artificial intelligence (AI) based model is a powerful tool to address various practical problems, and it is extensively used for simulation and forecasting in diverse areas such as water resources, power generation, finance, and environmental science (Maier and Dandy, 2000[22])
Box 5.4. International experiences with AI flood forecasting
About 200 million Indian people and more than 40 million in Bangladesh are believed to receive flood information from flood forecasting system of Google AI for Social Good. Google also developed Nowcast, a deep learning weather forecasting model, which shortened the computation time to 5-10 minutes, and can forecast with lead time of up to 6 hours.
Japan also developed an AI based flood forecasting system, called WinmuSe, to provide forecasted stage and discharge information.
In Korea the top priority regarding the improvement of flood forecasting systems is to utilise big data collected from all sectoral sources. An AI based flood forecasting platform is expected to be developed in the very near future.
Source: Han River Flood Control Office (2022[23])
Figure 5.8. AI based Data-driven Model and Flood Forecasting
Source: Han River Flood Control Office (2022[23])
The AI based data-driven model shows excellent performance if the output variable(s) has certain correlations with input variable(s). If an artificial neural network is applied to forecast flood water level with 1 hour lead time, it shows much better performance than physically-based simulation models. However, as the forecast lead time increases, it shows poor performance because the output variable (stage after e.g., 3 hours) has less correlation with the input variables (current stages and discharges).
In order to improve the accuracy of flood forecasting, it is necessary to hybridise a physically-based model and a data driven model to extract complementary strengths and eliminate weaknesses in the respective methodologies rather than choosing between the individual modelling techniques. The fact that data-driven models exploit the relationship between data, whereas hydrodynamic flow models are based on physical principles, gives them a complementary nature. In order to provide more accurate information about flood flow, errors produced by the simulation model can be corrected by using postprocessor methods based on a data driven model. The selected error correction model should extract useful information about the physical process that was not considered during the construction of a physically based simulation model such that additionally obtained information can remove systematic biases of the calibrated simulation model.
Figure 5.8 illustrates the framework of a hybrid approach for river flow forecasting. The procedure shows that the hydrodynamic flow model is initially executed, and this is followed by the error correction model. The forecasts are improved by combining the results of the two models. An ANN model is used to estimate residual errors of the hydrodynamic model by considering time series that are closely related to the model errors. It is assumed that the structural problems in the hydrodynamic model and data can be reflected in the detected patterns. The results of previous studies indicate that aggregating a physically based simulation model and a data-driven model rather than only applying a simulation model or a data-driven model can fully exploit different aspects of the physical system to minimise the prediction errors between estimated and observed flood water levels.
Figure 5.9. Methodology for a hybrid approach for river flow forecasting
Source: Li, L. and Jun, K.S. (2022). A hybrid approach to improve flood forecasting by combining a hydrodynamic flow model and artificial neural network (Li and Jun, 2022[24])
Figure 5.9 illustrates another kind of hybrid approach where the ANN model is combined with a rainfall-runoff model such as HEC-HMS. The runoff forecasted by the rainfall-runoff model feeds into the ANN model as an input variable to enhance the prediction accuracy.
Figure 5.10. Illustration of a Hybrid Approach for Runoff Forecasting
Source: Professor Jun, Korea
Figure 5.9 illustrates another kind of hybrid approach where the ANN model is combined with a rainfall-runoff model such as HEC-HMS. Since the forecasted runoff becomes less accurate as the forecast lead time increases if the discharge and rainfall data observed only up to the current point in time are used as inputs of the ANN mode, the future runoff forecasted by a simulation model is adopted in addition to the observed data. In short, the runoff forecasted by the rainfall-runoff model feeds into the ANN model as an input variable to enhance the prediction accuracy.
5.4. Recommendations to integrate flood information systems and early warning
Among the ways to respond to climate change, adaptation is to minimize the burden imposed on both humans and nature by activating non-structural measures rather than structural measures. The early warning system is one of the core frameworks recommended by many international organisations. In this regard, the ICT-based smart early warning system in order to realize “last mile or must have” more effectively is commented. The other is a strategy to increase the efficiency of water resource and disaster management by integrating water-related information systems that are dispersed among various ministries and agencies in Indonesia.
5.4.1. Latest trends of early warning systems
International framework for early warning systems based on the Sendai Framework
International frameworks for action to reduce risks from natural disasters are nothing new. The Sendai Framework is a voluntary and binding framework, recognizing that countries play a primary role in reducing disaster risk by 2030, but that responsibility must be shared with all stakeholders, including local governments, the private sector and other stakeholders. It aims to understand disaster risk at national, regional, local and global levels, and strengthen disaster risk governance for disaster risk management, and invest in disaster risk reduction for resilience, and more in effective response and recovery, rehabilitation and reconstruction. It sets priorities for action to be taken to "Build Back Better".
The gist of the Sendai Framework's action code emphasises that disaster prevention is the best strategy to reduce disaster risk. Disaster risk management is based on an understanding of disaster risk at all levels, disaster risk vulnerability, exposure of life and property, disaster risk characteristics and situations, and response capability. To this end, the framework recognises the shared responsibility between the central government, relevant sectors and stakeholders, and share disaster-related resources and data/information. Certainly, the national framework related to the natural disaster risk management has been presented in great detail from a theoretical point of view. However, as can be seen from the case of Indonesia, it is true that there is still much work to be done before implementing the policy and seeing the result.
One of the most representative elements is the multi-hazard early warning system (MHEWS) which addresses several hazards and/or impacts of similar or different type in contexts where hazardous events may occur alone, simultaneously, cascading or cumulatively over time (J Luther et al., 2017[25]). Another is the impact-based forecast and early warnings services (IBF-EWS) which is based on ensemble predictions to capture and exploit the uncertainty in the forecast to improve decision-making. The other is the end-to-end early warning service for flood forecasting (E2E-EWS-FF) which is the WMO framework to promote the enhancement of flood forecasting and early warning capabilities of NMHSs (National Meteorological and Hydrological Services), which is interoperable at all levels from data collection to informing users and decision support system (APFM, 2022[12]).
Even though those approaches slightly differ from one another, flood forecasting and warning as a focused activity are a subset of non-structural measures for flood management under the overall framework of Integrated Flood Management (IFM).
Table 5.3. Comparison of Three Flood Forecasting and Early Warning Services
Sources: WMO (2021[26]) ESCAP and WMO, 2021 (2021[27]) ; WMO, 2022 (2022[28])
Current situation and challenges with Indonesia’s Early Warning System
In Indonesia, the multi-hazard early warning system started off well, but stakeholders commented that the low accuracy of warnings and the need for cooperation between multiple agencies hinder the functioning of the early warning system. It is worth noting this point because the development process of the system needs a lot of information, cooperation with related agencies and the capability to develop it as described in the WMO’s guideline (WMO, 2010[10]).
Creating a common understanding of the disaster risks and building a culture of sharing risk knowledge is a key condition for any early warning system. In particular, it is important to change the perception of the people who directly or indirectly perform the relevant tasks in the field, because most of the actual damage is experienced by ordinary citizens and private business facilities rather than the national institutions that specialise in disaster risk management. Although this is being addressed through education or media, the quantity or quality of the information does not directly influence people living in disaster-prone areas. It has been proven by many examples that disaster warnings can significantly reduce losses when definitive risk warnings are available at the “last mile”, actually starting five days before the flood event occurs (UNDRR, 2019[29]).
A series of processes related to the floods early warning system can be divided into three categories: (1) weather forecasting; (2) flood forecasting; (3) and flood warning (MetMatters, 2022[30]). After all, if these three axes are the key factors that determine the ability of early warning, all sectors need to be improved in Indonesia. Weather forecasting usually involves the National Weather Service collecting atmospheric and hydrological data from satellites, radar, and rainfall observation facilities to commit rainfall forecasts using numerical weather forecasting (NWP) models. Flood forecasting is the estimation of future water levels or flows at a single or multiple sites of a river system for different lead times. Flood warnings use the forecasting results of the rainfall-runoff model to predict flooding or flood-affected areas due to river level rise and specify flood warning areas. Early warning for a specific spot must be carried out quickly and accurately and must be made in real time in consideration of ever-changing discharge conditions. The series of processes should be expanded into the concept of “Service Chain”, which actively conveys information by enabling two-way communication between information providers and users.
As pointed by Srikandini et al. (2018), a striking feature of disaster risk governance (DRG) is its heavy organisational set-up. The implementation of ‘decentralized DRR’ in Indonesia remains problematic because of the complexity of power sharing between the central and local governments and because of bureaucratic heaviness. The BNPB and BPBD are connected by a ‘coordination line’ rather than a ‘command line’. Local government acts as the frontline in formulating local policy, arranging resources and building community capacity. However, lack of budget, human resources and capacity as factors hampering its work in the region is still pervasive. In addition, intra-government coordination remains a major issue for DRG in Indonesia, where approximately 22 ministries and government agencies work on DRR related issues. Considering these points, it is necessary to further simplify the decision-making and cooperation system for disaster risk management through a barrier-free information sharing system.
5.4.2. ICT-based integrated Flood Information Systema and Smart-Early Warning Services
Under the Sendai Framework, an End-to-End Early Warning Systems for Flood Forecasting (E2E-EWS-FF) is a complete set of components that connects those who need to hear messages to others who compile and track the hazard information (Figure 5.11) (APFM, 2022[12]). It should be interoperable at all levels, such as information sharing and feedback, with a decision support system to provide an environment where residents can make their own decisions and take action through risk level warnings (APFM, 2022[12]).
Figure 5.11. End-to-end early warning systems for flood forecasting
Source: APFM, End-to-End Early Warning Systems for Flood Forecasting (2022[12])
In Indonesia, such an attempt has already been carried out or is underway. The structural EWS flourished after the 2004 tsunami in Aceh, and led to assorted laws and policies related to responsibilities of assorted government agencies at all levels, including information dissemination (Juwitasari, 2022). However, the evaluation these laws evoked several challenges (Government of Indonesia, 2020[5]) (Reni Juwitasari, 2022[31]). There are many factors that make this goal difficult, but technically implementing E2E-EWS-FF itself is not an easy task. In Indonesia, the information or analysis technology required to implement E2E-EWS-FF and the system supporting it are not sufficiently equipped. Particularly, building E2E-EWS-FF for 325 high-risk areas is not an easy job in terms of cost, manpower and equipment. As an alternative to cope with this task, adopting ICT-based and smart communication systems is recommendable.
The service chain of early warning systems has several challenges related to disaster risk knowledge, real-time monitoring and detection, data analysis and forecasting, early warning dissemination and communication, decision making, and preparedness and response to warning (Government of Indonesia, 2020[5]). Nationally developed contingency plans are common in many countries. Most of these are the responsibility of government agencies, but it is also the responsibility of ordinary citizens to respond to real situations. However, contingency plans are not customised to the target communities and integrated into emergency response plans due to a lack of participatory approaches in the planning and development of warning response measures. In the case of the Palu tsunami, BMKG successfully issued a tsunami early warning five minutes after the earthquake. This is in accordance with their SOP. However, the tsunami arrived sooner than that (UNDRR, 2019[29]).
ICT technologies can facilitate a human-centred early warning system, particularly in countries with high smart phone coverage like Indonesia. A concept to consider is the smart-early warning service concept, if ICT technology is applicable to the local context. A complete and effective smart-early warning system consists of four elements, namely risk knowledge, monitoring and warning service, dissemination and communication and response capability as decisively demonstrated in the Sendai Framework. An exemplary early warning system should have strong interconnections and effective communication channels between all elements (UNISDR, 2015[32]).
The four elements above-mentioned must solve the information sharing problem in each field in order to organise an effective people-centred early warning system. The Indonesian government also emphasises that data and information should be easily shared between government agencies and implements the 'one data policy'. Currently, government agencies do not have easy access to water-related data collected by other government agencies. The water-related information system managed by Han River Flood Control Office (HRFCO, 2022[23]) provides water-related data/information collected from 2 central ministries, 15 government-affiliated organisations, and 2 local governments. Table 5.4 lists the information systems used in the hydrologic and flood forecasting related data sectors currently in operation in Korea. These systems are based on ICT technologies and operated individually, but its accessibility and use are open systems for public and private users. An important condition is that anyone can access all data released by the government through the Open API.
Indonesia is modernising water monitoring, improving analysis tools, investing in water knowledge, and building open access and real-time centralized information systems (World Bank, 2021[3]). The Ministry of Health's “Smart for Public”, which currently monitors the progress of sanitation programs at the community level, is a good example. To promote public participation through clear roles, incentives and regulation in the city area, the private sector needs to bear its fair share of responsibilities for urban resilience. Incentives and clear regulations for development projects need to encourage developers to design risk-informed investments. Innovative programs for retrofitting of critical infrastructure, or land value capture opportunities in the floodplain management could be considered prosperity and liability (The World Bank, 2019).
Table 5.4. ICT-based Information Systems for Hydrological Data and Flood Prediction in Korea
Note: These systems are used in Korean and are categorized from the author's point of view.
Source: Authors
The application of ICT basically consists of a platform that provides hydrometeorological and flood analysis systems, as in the example of Korea above (Box 5.5). ICT is to provide a system that collects and analyses information from the central government's disaster management agency that is advantageous to manpower, technology and finance. For successful implementation of E2E-EWS-FF for 325 high-risk regions in Indonesia, the central government provides necessary expertise, resources, and information (e.g., demographic data, cloud sourcing/computing technology, GIS system, etc.) in order to realise ensure interoperability and data integration.
Box 5.5. Korea’s Smart Water City information system
Recent advances in ICT are contributing to the qualitative growth of the business environment as well as the scalability of information use. In Korea, local governments are implementing the concept of ‘Smart Water City’, which is centered around the drinking water system. Currently, the Han River Flood Control Office of the Ministry of Environment is promoting the improvement of the service that provides probabilistic rainfall and flood discharge at specific points in major watersheds as depicted below in terms of providing hydrological information (Development of Provision System for Map-based Probabilistic Rainfall and Flood Estimates) (see Figure 5.12 below). This system will be able to provide the information needed to determine urban flood drainage. For the Smart Water management projects in Korea, Songsan Green City and Busan Eco-Delta City, which are currently under construction, were designed as Smart Water Cities from the beginning, and in 2017, the first national Smart Water City project was launched in Sejong City. K-water has also been given approval to introduce the system in Dongducheon City, Yangju City, Jeongeup City, Goryeong County and Naju City in 2017 (K-water, 2022[33]).
This system enables impact-based flood forecasting of most river points nationwide and allows early warning of specific hydrologic events. Indonesia has great potential for smart-early warning system as it has high rates of smartphone possession (238 million), social network services (217 million) and internet accessibility (Statista, 2022[34]). Strong policy enforcement of ICT-based disaster risk management system support Indonesia’s agenda to become a flood-resistant country.
As shown in the figure, the map-based system for providing probabilistic rainfall and flooding forecasting is a structure in which the provision of probabilistic rainfall and flooding information is added by utilising already-built hydrologic data, river information, and GIS. In Korea, most of the basic data set for simulating flood event in major rivers have been well established. The information generated by the new system will be opened to the public, and the flooding at a certain site can be calculated according to the user's request.
Figure 5.12. Development of Provision System for Map-based Probabilistic Rainfall and Flood Estimates
Source: Han River Flood Control Office (2022) (not officially published)
Source: (K-water, 2022[33])
Local governments that should provide information to the public and interested parties can use the platform provided by central government agencies. Especially, local governments are strengthening cooperation during the “last mile” in predicting or forecasting, helping people understand the impact of risks on them and guiding them to take appropriate actions. Here, there are subjective measures in which stakeholders rely on their own judgement, or otherwise objective methods, that rely on risk knowledge information through impact prediction models using vulnerability data, exposure data, and weather information.
Meanwhile, the smart-based early warning system is an adaptive measure that uses an integrated communications system to help communities prepare for hazardous climate-related events (Climate ADAPT, 2022[35]). Indonesia's service chain of natural disaster management has multiple organisations (multi-institution), produces and distributes various information (multi-source), and has multiple information users (multi-user). The success factor of smart technology lies in the realisation of two-way communication. The system in place at the sub-regional and regional level should enable two-way communication to components for risk knowledge, monitoring and forecasting, information dissemination and warning. In addition, preparation for hazards, rapid response, monitoring results and immediate response should be possible when it works.
After all, it must be considered that the penetration rate of smartphones is much faster than the speed of building infrastructure for water supply or disaster management in lower-middle income countries, including Southeast Asia. Recognizing those implications can help shape how the early warning system in Indonesia needs to be transformed.
5.4.3. Establishment of a smart-early warning system based on ICT
Transforming a conventional approach into a smart approach for non-structural disaster risk management
Today, non-structural measures to deal with natural disaster puzzles can be a reasonable way to consider the natural and ecological environments. Nevertheless, if structural measures are the most reasonable alternative and their execution has sufficient meaning, it should be pursued to achieve obvious public goals. However, non-structural measures must be addressed from a longer-term perspective, such as sustainability or climate change mitigation and adaptation. What non-structural measures mean today is not simply the opposite of structural measures, but rather a smart approach. Information and communication-based solutions such as Smart Water, Digital River, and ICT-based solutions can be used for water resources, water disasters, aquatic ecology, and water conservation facilities.
Building an integrated information system based on the cloud computing platform concept
The platform concept is a solution through so-called next-generation ICT technology. It builds an integrated water resource management platform that secures the necessary resources and decision-making space by real-time monitoring of water sources, hydrological conditions, disaster situation management and related facility maintenance, equipment operation, and abnormal activities. It also aims to provide monitoring, analysis and early warning of water disasters based on integrated data and GIS maps to support flood and drought control, accurate early warning and efficient emergency response. In particular, comprehensive monitoring of major water resource conservation targets such as rivers, reservoirs, and dams are possible based on the IoT, video, satellite remote sensing, and 5G technologies in modern society (inspur, 2022[36]). In the case of Indonesia, in order to carry out disaster management tasks targeting 325 high-risk areas, it is reasonable to build a cloud computing system that can support all of these from the central agency, considering the geographical conditions, professional manpower, and budget.
Rethinking a smart water management system for enhancing efficiency of water related works
Smart water management systems can provide a more resilient and efficient water supply system while reducing costs and improving sustainability. Technically, smart water management systems can detect any abnormal factors of water resources and water facilities and provide an integrated platform that can monitor in real time, integrated hydrological monitoring based on abundant observation data, flood forecasting analysis and early warning. The concept of smart water management is to monitor water resources and facility maintenance to ensure water resource safety and normal operation and reduce abnormal and illegal activities. It is to support scientific decision-making on water resource management, regulation and allocation, ultimately ensuring the quality and safety of water resources, improving water resource utilisation (inspur, 2022[36]) (K-water, 2022[33]).
Linking information providers and users using two-way communication for smart early warning services and securing “last mile”
Early warning to hazards, such as flooding, provides lead time to take action to reduce the risk impact and ensure safety for people, industries and communities. Prediction of upcoming floods is an important part of this warning service. In Indonesia, smart early warning systems deal with riverine, coastal and surface floodplains, especially inundation caused by landslides in volcanic regions and tsunamis in coastal lowlands. It is to use flood forecast information to evaluate the area and scale of floods and to predict the degree of possibility of disaster occurrence. The way of delivering the most recent information corresponding to the “last mile” directly to the residents of the 325 disaster-prone areas in Indonesia should be reviewed in a consistent and sustainable way.
Unlocking the potential of ICT to improve water related disaster risk management and smart early warning service
As of October 2022, Indonesia's social media landscape has over 260 million active social media users, ranking it 4th among countries with the highest number of social media users worldwide. Currently, Facebook has 70% market share, and WhatsApp is rapidly taking up the market (Statista, 2022[34])Other social media platforms such as Instagram are also gaining popularity. Ultimately, a policy that actively utilises mobile phones and social media environments is needed for disaster forecasting and warning. ICT-based flood analysis information and smart-early warning services are based on these diverse information-communication environments. In addition, the effectiveness of the dual warning system that utilizes simultaneously both the structural early warning and the traditional method highlighted by Juwitasari (Reni Juwitasari, 2022[31]) can be more synergistic when both systems are combined with ICT.
Box 5.6. Water Disaster Response System in the Han River in Korea
The Han River is a representative river of South Korea and serves as a drinking water source for approximately 30 million people. It has a total length of about 480 kilometers and a watershed area of 35,000 square kilometers. Geographically, the Han River includes administrative areas such as Seoul, Gyeonggi Province, Gangwon Province, and Chungcheongbuk Province. The river flows as two separate rivers, the North Han River and the South Han River, in its upper watershed before merging at the Paldang Dam and flowing into the West Sea. In the past, frequent heavy rainfall and rainstorms have caused flood damage along the Han River, but relatively little drought damage has occurred due to the river's abundant water supply and dam structures. However, managing multiple water resource facilities requires a high level of operational management skills, and failure to do so could result in dam collapses in upstream areas and significant flood damage in downstream areas.
The flood response system for the Han River is centered around the Han River Flood Control Center operated by the Ministry of Environment, and it includes local governments, water agencies such as K-water, etc. The management system of water related disaster in Han River is shown in the following (Figure 5.13).
Figure 5.13. Water related Disaster Management System
Real-time weather and hydrological data collected from the Han River and its watershed are gathered through the National Water Resources Information Management System (www.wamis.go.kr). The Han River Flood Control Center calculates dam inflow and river flow rates and predicts water levels at major river sites based on this data. The center analyzes this data and prediction results comprehensively to predict the possibility of flood occurrence and disseminates the information to relevant agencies and citizens. In preparation for water disasters such as floods, local governments, and related agencies collaborate to plan and perform joint response operations from the beginning of each year. To support this effort, they strengthen management organisations, secure and provide budgets, and do not neglect administrative management.
Flood facilities around the Han River such as embankments are continuously maintained. In case of a disaster, the local government forms a task force for a rapid response, and the central government provides support when needed. Related agencies such as the Ministry of Environment educate local residents on flood response and preparation methods and continuously conduct safety campaigns to minimize flood damage. The Han River Flood Control Center has created and distributed flood risk maps indicating flood risk levels and flood depth around riverbanks to minimize citizens' flood damage, and the information is currently available on the Flood Risk Map Information System (www.floodmap.go.kr).
Source: Ministry of Environment of Republic of Korea (2022) (not officially published)
References
[40] (n.a.) (n.d.), WMO Guidelines on Multi-hazard Impact-based Forecast and Warning Services Part II: Putting Multi-hazard IBFWS into Practice.
[12] APFM (2022), End-to-End Early Warning Systems for Flood Forecasting (E2E-EWS-FF), https://www.floodmanagement.info/get-help/end-to-end-early-warning-systems-for-flood-forecasting-e2e-ews-ff/ (accessed on 22 December 2022).
[17] Bai, Y. et al. (2018), “Storm water management of low impact development in urban areas based on SWMM”, Water (Switzerland), Vol. 11/1, https://doi.org/10.3390/w11010033.
[35] Climate ADAPT (2022), Awareness campaigns for behavioural change, https://climate-adapt.eea.europa.eu/en/metadata/adaptation-options/awareness-campaigns-for-behavioural-change (accessed on 22 December 2022).
[13] Dilley, M. (2005), Natural Disaster Hotspots A Global Risk Analysis.
[6] Government of Indonesia (2022), Questionnaire for the National Dialogue on Water. Responses from the Government of Indonesia.
[5] Government of Indonesia (2020), THE NATIONAL MEDIUM-TERM DEVELOPMENT PLAN FOR 2020-2024 Narration REPUBLIC OF INDONESIA.
[38] Government of indonesia (2020), “2020_Indonesia_WASH_questionnaire_OHCHR”.
[11] Hodgkin, D. (2016), Emergency Response Preparedness in Indonesia A Consultation Report prepared exclusively for the Indonesia Humanitarian Country Team EMERGENCY RESPONSE PREPAREDNESS IN INDONESIA A CONSULTATION REPORT PREPARED EXCLUSIVELY FOR THE HUMANITARIAN COUNTRY TEAM.
[23] HRFCO (2022), Han River Flood Control Office, Han River Flood Control Office, http://www.hrfco.go.kr (accessed on 22 December 2022).
[36] inspur (2022), Smart Water: Intelligent supervision, protection and efficient utilization of water resources, https://en.inspur.com/en/smart_water/index.html (accessed on 22 December 2022).
[25] J Luther et al. (2017), Concerted International Efforts for AdvancingMulti-hazard Early Warning Systems, Springer International Publishing, https://doi.org/10.1007/978-3-319-59469-9.
[33] K-water (2022), Overview and current status of the Smart Water City projects of Korea, https://www.kwater.or.kr/eng/busi/water02/smartWater01Page.do?&s_%20mid=1186 (accessed on 22 December 2022).
[16] Lee, D. (2016), “Restoration of Water Cycle by a Rainwater Management System Applied to Low Impact Development (LID)”, Ecology and Resilient Infrastructure, Vol. 3/2, pp. 130-133, https://doi.org/10.17820/eri.2016.3.2.130.
[21] Lee, J. et al. (2019), Catalogue of Hydrologic Analysis for Asia and the Pacific | Volume 1 Flood Hazard Mapping Flood Hazard Map of Korea.
[24] Li, L. and K. Jun (2022), “A Hybrid Approach to Improve Flood Forecasting by Combining a Hydrodynamic Flow Model and Artificial Neural Networks”, Water (Switzerland), Vol. 14/9, https://doi.org/10.3390/w14091393.
[9] LIRNE Asia (2008), Evaluating Last-Mile Hazard Information Dissemination: A Research Proposal (HazInfo Project) SRI LANKA, http://www.lirneasia.net.
[22] Maier, H. and G. Dandy (2000), “Neural networks for the prediction and forecasting of water resources variables: a review of modelling issues and applications”, Environmental Modelling & Software, Vol. 15/1, pp. 101-124, https://doi.org/10.1016/S1364-8152(99)00007-9.
[19] Memon, P. and B. Gleeson (1995), “Towards a New Planning Paradigm? Reflections on New Zealand’s Resource Management Act”, Environment and Planning B: Planning and Design, pp. 109-124.
[30] MetMatters (2022), Flood forecasting factfile, https://www.rmets.org/metmatters/flood-forecasting-factfile (accessed on 22 December 2022).
[31] Reni Juwitasari (2022), Disaster Management in Indonesia: Complex Challenges of a Dual Early Warning System, Heinrich Boll Stiftung, https://th.boell.org/en/2022/03/23/disaster-management-indonesia (accessed on 22 December 2022).
[14] Rossum, J. and K. René (2010), “Disaster Management in Indonesia: Logistical Coordination and Cooperation to Create Effective Relief Operations”, Jurnal Teknik Industri, Vol. 12.
[15] Saunders, W. and M. Kilvington (2016), “Innovative land use planning for natural hazard risk reduction: A consequence-driven approach from New Zealand”, International Journal of Disaster Risk Reduction, Vol. 18, pp. 244-255, https://doi.org/10.1016/j.ijdrr.2016.07.002.
[7] Srikandini, A., D. Hilhorst and R. Voorst (2018), “Disaster Risk Governance in Indonesia and Myanmar: The Practice of Co-Governance”, Politics and Governance, Vol. 6/3, pp. 180-189, https://doi.org/10.17645/pag.v6i3.1598.
[34] Statista (2022), Number of smartphone users in Indonesia from 2017 to 2020 with forecasts until 2026, https://www.statista.com/statistics/266729/smartphone-users-in-indonesia/ (accessed on 22 December 2022).
[18] Tokyo Disaster Management Council (2014), Tokyo regional disaster prevention plan (Earthquake Section) - ｢東京都地域防災計画 (震災編), Tokyo Disaster Management Council, https://www.bousai.metro.tokyo.lg.jp/taisaku/topics/1000019/1001121/1001278.html (accessed on 12 May 2023).
[27] UN ESCAP and WMO (2021), Manual for Operationalizing Impact-based Forecasting and Warning Services (IBFWS).
[1] UNDRR (2022), Governance strategies in Indonesia for addressing systemic risks: where do we stand and the future outlook.
[8] UNDRR (2021), Scoping Study On Compound, Cascading And Systemic Risks In The Asia Pacific, https://www.undrr.org/contact-us.
[29] UNDRR (2019), Our world at risk transforming governance for a resilient future.
[32] UNISDR (2015), Sendai Framework for Disaster Risk Reduction 2015 - 2030.
[28] WMO (2022), Assessment Guidelines for End-to-End Flood Forecasting and Early Warning Systems.
[26] WMO (2021), WMO Guidelines on Multi-hazard Impact-based Forecast and Warning Services Part II: Putting Multi-hazard IBFWS into Practice.
[39] WMO (2011), World Meteorological Organization Communications and Public Affairs Office Hydrology and Water Resources Branch Climate and Water Department MaNual ON FlOOD FORECaSTiNG aND WaRNiNG P-ClW_102107.
[10] WMO (2010), GUIDELINES ON EARLY WARNING SYSTEMS AND APPLICATION OF NOWCASTING AND WARNING OPERATIONS, http://www.wmo.int/pages/prog/amp/pwsp/publicationsguidelines_en.htm.
[2] World Bank (2021), How Indonesia Strengthened its Disaster Response with Risk Finance and Insurance, https://www.worldbank.org/en/news/feature/2021/11/17/how-indonesia-strengthened-its-disaster-response-with-risk-finance-and-insurance.
[3] World Bank (2021), Indonesia Vision 2045 : Toward Water Security, The World Bank Group, Washington, DC, http://hdl.handle.net/10986/36727.
[4] World Bank (2019), Indonesian Disaster Resilience Project(P170874).
[37] World Bank & Ministry of National Development Planning(BAPPENAS), I. (2021), Indonesia Vision 2045:Toward Water Security, http://www.worldbank.org/gwsp.
[20] Yoon. B (2022), “Smart river flow measurement technology using CCTV”, Journal of Korea Water Resources Association.