Climate change risk assessment for the Republic of Korea: developing a systematic assessment methodology

Climate change risks have become a major concern of climate change adaptation, and a systematic risk assessment is required as the first step as well as a key principle of national adaptation policy processes. Although many countries conducted risk assessments, a debate over a systematic assessment process continues, and results of the risk assessment provide limited information to making adaptation policies. Based on a case study of South Korea, this research aims to establish a national-level risk assessment process which includes systematic methodologies given the current limited time/resource and insufficient climate change information. A four-step risk assessment process is proposed: (1) collecting scientific evidence, (2) making list of preliminary risks, (3) making lists of risks and prioritising the risks, (4) categorising the risks. Enough scientific evidence and data about climate change risks of Korea were retained through first two steps, and three components of risk (hazard, exposure, vulnerability) are systematically involved by assessing the magnitude and adaptive capacity of risks. As results of the risks assessment, 93 national-level climate change risks of Korea are identified, and most high priorities in risks have high risk magnitude but low adaptive capacity. This research provided insights for direction of national adaptation policy of each sector by categorising the risks into four categories.


Introduction
Climate change gives rise to cascading risks in human and natural systems (IPCC 2014;Adger et al. 2018). Given that the inevitable impacts of climate change caused by greenhouse gases (GHGs) already emitted, a need for adaptation has increased and the climate change risks have become a major concern of the adaptation (CCC 2017a). Since the Intergovernmental Panel on Climate Change (IPCC)'s Special Report on Extreme Events (SREX) (2012), the concept of climate change risk involves the climate science aspect that projects the probability of a hazard and the dynamic socio-economic aspects that drive exposure and vulnerability (IPCC 2012;Adger et al. 2018). IPCC (2014) defines risk as below.
"The potential for consequences where something of value is at stake and where the outcome is uncertain, recognising the diversity of values. Risk is often represented as a probability of occurrence of hazardous events or trends multiplied by the impacts of these events or trends occur. Risk results from the interaction of vulnerability, exposure, and hazard." Based on climate scientific evidence, hazards (heavy precipitation, tropical cyclone, droughts, floods, heatwaves, sea level rise and etc.) are projected to increase, and the severity of the impacts of the hazards relies strongly on the level of exposure and vulnerability to the hazards (IPCC 2012). IPCC (2014) also highlighted the paradigm shift from vulnerability assessment to risk assessment in their fifth assessment report.
Risk assessment is a crucial source of information and a key principle of managing risks and adaptation policies 1 3 (European Commission 2010, 2013Papathoma-Köhle et al. 2016;EEA 2018). Decision-makers, under significant uncertainties, should make decisions to address climate change risks. As the process of examining available information to guide decision-making (WRI 2009), risk assessment systematically evaluates potential impacts of hazards and their societal consequences (Morgan et al. 1990;Brown 2015). It is a process to understands the nature and determines the level of risk (Byrd and Cothern 2000;Adger et al. 2018) and provides the basis of an analysis of risk reduction strategies. It is required that risks assessments are regularly conducted, reflecting the change of the risk as hazard, exposure, and vulnerability are continuously changing due to climate and social-economic circumstances' changes (Papathoma-Köhle et al. 2016). At the national level, climate change risk assessments usually aim to make adaptation policies (Brown et al. 2018), and findings of the assessments provide consistency of priorities and scope for adaptation to moderate risk factors (Brown et al. 2018). Risk assessments provide information on what risks are expected in the future and what risks should be addressed first with response measures. Therefore, based on risk assessment results, decision-makers can set directions of adaptation policies and make strategies to reduce the negative consequences of the risks.
As risk assessment gets more attention, the importance of national-level risk assessment has also been emphasised, and many nations have conducted it as the first step of their national adaptation policy process (Brown 2015). For example, the UK established the Climate Change Act 2008 to provide legal foundations for risk assessments, and the UK government have reported their Climate Change Risk Assessments (CCRAs) in 2012 and 2017. The CCRAs provide the evidence base of climate risks that are expected to encounter and analyse the magnitude of the risks. Based on the assessment results, national-level adaptation measures for the next 5 years are developed (Brown et al. 2018). Germany established the Vulnerability Network, which consists of 16 federal departments and 9 departmental research institutes, to assess climate change vulnerabilities. They reported Germany's vulnerability to climate change (2015) and Climate change in Germany: trends, impacts, risks and adaptation (2017), which investigate climate signal, sensitivity and each sector's adaptive capacity. Based on a common understanding of vulnerability and continuous communications in the Vulnerability Network, Germany develops national-level response measures (Buth et al. 2015;Buth et al. 2017).
In 2014, the Republic of Korea (Korea) conducted a qualitative risk assessment (Korea Government 2015). In 2015, based on the risk assessment, the Korean government established the second National Climate Change Adaptation Policy (NCCAP). However, the risk assessment poses several limits. First, a lack of scientific evidence of risks is pointed out as a problem. As the risk assessment was conducted through experts' discussions and workshops, scientific evidence of risks was unlikely to sufficiently collected, and the results were deeply dependent on experts' opinions. Second, the results of the assessment may have limitations in developing adaptation measures. The list of priority risks provided to government departments lacked detailed information on how and to what extent risks should be addressed and responded to (Song et al. 2019). Lastly, there have been no official processes or systematic procedures for national-level climate change risk assessment in Korea, and relevant terms (risk, vulnerability) have been complexly used. It leads to different understandings of risk assessment, its results, range, and application according to sectors, departments, and actors. Making a common and clear understanding of relevant concepts and establishing an official systematic process of climate change risk assessment at the national level are required to use the risk assessment results and to address the risk effectively and systematically at the national level (i.e. Chapter 2 of UK CCRA evidence report 2017).
In this regard, this research aims to establish a nationallevel risk assessment process and methodologies, given the limited time/resource and insufficient climate change information. Also, the results of this risk assessment will inform to making the third NCCAP. From collecting scientific evidence to risk categorisation, this research set a whole process of risk assessment and detailed criteria for assessing climate change risks of Korea. Based on the process and criteria, we also identify national-level climate change risks of eight sectors: health, land, agriculture, water, forest, industry/energy, ecosystem, and ocean/fishery/coast. In addition, as the second NCCAP will be over at the end of 2020, these research results will play an essential role to develop and establish the third NCCAP of Korea.

Establishing a systematic process of national-level risk assessment
This research proposes a systematic climate change risk assessment for NCCAP of Korea using the current limited information and resources. Considering the limitation of the previous risk assessments, the proposed systematic risk assessment aims to collect scientific evidence of climate change risks, provide clear criteria of assessments, prioritising risks through systematic assessments, informing key points to making national adaptation policies, and apply consistent methodologies and criteria in national climate change risk assessments.
Thus, to meet the purpose, it sets four key considerations for the assessment through literature reviews, case studies, a review of the previous risk assessment of Korea and discussions with experts.
• To make scientific evidence about climate change impacts (national-level) through reviewing reference (Buth et al. 2017;CCC 2017b;Brown et al. 2018). • To utilise relationships between risks and between risk factors to identify major risks (Papathoma-Köhle et al. 2016;Buth et al. 2017;CCC 2017b). • To correspond to the concept of "risk = f(hazard, exposure, vulnerability)" by conducting assessments about adaptive capacity and adaptive measures (Preyssl et al. 1999;IPCC 2007IPCC , 2014Tonmoy et al. 2018). • To present urgency and category of each risk to increase the usability of the result of the risk assessment (CCC 2017b(CCC , 2021. Although there are a variety of approaches and methods for national risk assessments, the choice of assessment approaches and methods needs to take into account the particular information needs and the purpose of each national risk assessment (EEA 2018). The major purposes of a climate change risk assessment of Korea are: (1) retaining scientific evidence of climate change risks of Korea, (2) identifying national-level climate change risks, and (3) prioritising identified climate change risks to provide information to national adaptation policy, rather than calculating related numbers concretely. Based on the considerations and the purposes, the risk assessment method consists of four steps: (1) collecting scientific evidence, (2) making lists of preliminary risks, (3) making lists of risks and prioritising, and (4) categorising the risks.
For consistent understandings, this assessment sets key terms. Following the definition in IPCC (2014), 'risk' results from the interaction of hazard (h), exposure (e), and vulnerability (v); risk = f (h, e, v). 'Climate impact' refers to the consequences of climate change, it is a concept that excludes vulnerability from risk, consisting of hazard (h) and exposure (e); climate impact = f(h, e). 'Risk magnitude' refers to the sum of risk consequences and likelihood, which is measured in terms of the severity of its consequences (s) and its probability of occurrence (p) (Preyssl et al. 1999); risk magnitude = f(s, p). 'Adaptive capacity' refers to "the combination of the strengths, attributes, and resources available to an individual, community, society, or organisation that can be used to prepare for and undertake actions to reduce adverse impacts, moderate harm, or exploit beneficial opportunities" (IPCC 2014(IPCC , p. 1758. Thus, this assessment considers that the adaptive capacity can represent the vulnerability (v) in the function of risk, as the vulnerability consists of sensitivity (s) and adaptative capacity (ac) (IPCC 2007(IPCC , 2014. To measure adaptive capacity, this assessment includes institutional capacity (i), actor capacity (a), infrastructure capacity (f), technological capacity (t); adaptative capacity = f(i, a, f, t).

Collecting scientific evidence
Scientific evidence plays an important role in risk assessments, providing scientific grounds for making adaptative actions and helping to devise proper adaptation actions with various evidence and approaches. Thus, this risk assessment introduces a systematic literature review (SLR) to collect scientific evidence. SLR is a literature review methodology following a clearly defined protocol or plan where the criteria are set before the review is conducted (Dewey and Drahota 2016). It makes it possible to collect data systematically and comprehensively, as well as reduces subjective errors or bias of selecting literature to review (Petticrew and Roberts 2006). As a summary and assessment of the status of knowledge on a given topic or research question, SLRs have been increasingly used in the climate change field Ford and Berrang-Ford 2011;Spires et al. 2014). Thus, this risk assessment conducts an SLR following the seven stages of SLRs (see Petticrew and Roberts 2006, p.27) to enhance scientific evidence of climate change impacts on Korea. It sets three research questions for an SLR: (1) What are climate change risks of each sector in Korea (including national and local levels)? (2) What are the research results on the risk's factors (hazard, exposure, vulnerability) and impacts? (3) What are Korea national climate change risks drawn through analysing the risk factors and impacts? To make a clear and objective data range focusing on climate change impacts on Korea, this risk assessment chooses two web databases (KISS (http:// kiss. kstudy. com) and DBpia (https:// www. dbpia. co. kr)) that are the biggest and most frequently used in Korea. In order to collect objectively verified data, at first, it searches only peer-reviewed articles in the databases. Both Korean ([기 후변화]) and English ([climate change]) terms were used to search. It is supposed that studies on climate change in Korea published before 2014 were reviewed and involved in Korean Climate Change Assessment Report (MoE and NIER 2014); thus, we focused on studies published between 2014 and 2019 in this risk assessment searches. In May 2019, total 20,518 articles are retained (KISS: 1,952; DBpia: 18,566). The criteria for inclusion and exclusion are summarised in Table 1. Based on the criteria, a total of 565 articles were identified and analysed (Table 2).

Making lists of preliminary risks
Preliminary risks refer to potential risks that can be drawn from related literature and data review, without adaptive capacity assessments. Acknowledging that academic articles in the SLR do not include every aspect of climate change risks of Korea, 'the climate impact database' is additionally reviewed to supplement the SLR results and make broader lists of preliminary risks. The climate impact database 1 3 was used to build causation maps to supplement the SLR results. The climate impact database was established (Sin et al. 2017), which consists of risk factors (hazard, exposure, impact, and risk). It provides information about climate impacts on the real-life of each sector. Based on the results of the SLR and the climate impact database review, lists of preliminary sectoral climate change risks are drawn. The preliminary risks are classified into eight sectors: health, land, agriculture, water, forest, industry/energy, ecosystem, and ocean/fishery/coast.

Making lists of risks and prioritising
In this step, every progress is conducted by sectoral expert groups; seven to ten experts from academia, research institutes, public organisations, etc. participated in each sectoral expert group. The preliminary risks, first, are reviewed and revised by sectoral expert groups. In this step, new risks that sectoral experts consider important risks but are not included in the lists of preliminary risks can be added. Also, risks that are not significant enough to be considered are eliminated through expert discussion in this step. Given that research articles cannot deal with all climate change risks and authors have bias to choose research topics, adding and deleting risks through sectoral expert discussions are essentially required. Then, the sectoral expert groups select each sector's climate change risks. The selected risks are prioritised by assessing risk magnitude and adaptive capacity. Following the definitions of key terms and concepts, the severity of the risk consequences (s) and the probability of the risk occurrence (p) are used to assess the risk magnitude. These are assessed with a three-point scale from low to high (Table 3). The adaptive capacity assessment consists of four sub-assessments, which assess with a five-point scale   (IPCC 2014) Hazard-The potential occurrence of a natural or human-induced physical event or trend or physical impact that may cause loss of life, injury, or other health impacts, as well as damage and loss to property, infrastructure, livelihoods, service provision, ecosystems, and environmental resources. In this report, the term hazard usually refers to climate-related physical events or trends or their physical impacts Exposure-The presence of people, livelihoods, species or ecosystems, environmental functions, services, and resources, infrastructure, or economic, social, or cultural assets in places and settings that could be adversely affected Vulnerability-The propensity or predisposition to be adversely affected. Vulnerability encompasses a variety of concepts and elements including sensitivity or susceptibility to harm and lack of capacity to cope and adapt. The four steps of national-level climate change risk assessment described above are summarised in Fig. 1.

Preliminary climate change risks of Korea
This risk assessment drew sectoral preliminary climate change risk lists through analysing retained 565 articles of the SLR and the climate impact database from Sin et al. (2017). In eight sectors, 204 climate impacts 1 and potential risks (without adaptive capacity assessments) which have scientific evidence were drawn: 10 for health, 23 for land, 32 for agriculture, 24 for water, 13 for forest, 52 for industry/ energy, 31 for ecosystem, and 19 for ocean/fishery/coast. For example, the preliminary risks of the health sector are presented in Table 5.

Climate change risks in Korea
Sectoral expert groups selected 95 climate change risks of Korea through reviewing and revising the preliminary risks. In this process, sectoral experts updated the risks by adding new important risks, removing or merging. For example, in the health sector, 'water-borne diseases increase due to temperature increase' and 'mediator diseases increase due to temperature increase' were added, and subsequently, 12 climate change risks for the health sector were selected. The number of risks in the industry/energy sector also decreased from 52 to 13 by removing inconsiderable risks or merging similar risks. The assessment results of risk magnitude and adaptation capability of the 95 risks were plotted in each sectoral quadrant sheet. In Fig. 2, the upper-right quadrant refers to a  Ozone concentration increase, temperature increase, precipitation increase Patients Early fatality increase due to air pollution Early fatality increase due to air pollution caused by temperature, ozone concentration increase 8 Temperature increase, high temperature Patients Renal colic increase Renal colic increase due to temperature increase 9 Temperature increase, CO 2 concentration increase Patients Asthmatic patient increase Asthmatic patient increase due to temperature and CO 2 concentration increase 10 Precipitation reduction Human Respiratory disease and skin disease increase Respiratory disease and skin disease increase due to dry air risk that has high risk severity and probability but low adaptive capacity, while the upper-left quadrant refers to a risk that has high risk severity and probability and high adaptive capacity. In contrast, the lower-right quadrant shows a risk that has low risk severity and probability and low adaptive capacity, and the lower-left quadrant presents a risk that has low risk severity and probability but high adaptive capacity. Lastly, 93 risks were selected as climate change risks of Korea through mediations between sectors (12 for health, 12 for land, 14 for agriculture, 10 for water, 12 for forest, 11 for ecosystem, 12 for industry/energy, and 10 for ocean/ fishery/coast). There were some similar risks between sectors, and two risks from water and industry/energy sectors were removed by comparing all selected risks.
In the expert forum, the selected risks were prioritised, based on the results of the assessments of the risk magnitude and adaptive capacity. Most high priorities in the risks commonly have high risk magnitude but low adaptive capacity, although a few risks were exceptionally added to the high priority through expert discussions. The priorities of the 93 risks are shown in Appendix, and high ranked risks have a high priority.

Risk categorisation
Based on the analysis of institutional capacity assessment and technological capability assessment, this risk assessment classified the identified 93 risks into the four categories (see "Categorising the risks").
In this categorisation, it showed very different results according to each sector (Table 6). For example, agriculture and industry/energy sectors have a high proportion of risks in the existing category. This suggests that most of the climate change risks in these sectors were already covered by existing measures. Subsequently, less new measures or additional research were added. In contrast, ecosystem and ocean/fishery/coast sectors had one or no risk in the existing category, and most risks were in the new and fundamental research category. It means that existing measures in these sectors need to be reconsidered and redesigned to address their identified risks.

Discussion and conclusion
As the importance of managing climate change risks has been widely acknowledged, risk assessment has been to the fore as the first step of risk management (Papathoma-Köhle et al. 2016;Adger et al. 2018;Dawson et al. 2018;  EEA 2018). In this respect, Korean government conducted a national-level climate change risk assessment in 2014 to establish the second NCCAP, but there were several limits that hinder from understanding the concept of climate change risks to the use of the risk assessment results in adaptation measures. Thus, this research aimed to establish a systematic national-level risk assessment process and methodology given the limited time/resources and insufficient climate change information. Also, through identifying risks, it purposed to provide essential insights into making the third NCCAP. The climate change risk assessment methods in other countries require a large amount of time and resources (human and financial). For example, the UK's CCRA takes at least 2 years to collect scientific evidence (in the first year) and to assess the risks (in the second year), and it also includes a number of policy stakeholders and sectoral experts. Also, significant costs are required in this process (CCC 2021). These risk assessment methods are not proper for countries that don't have sufficient resources, expertise or that need to immediately conduct a risks assessment for their national adaptation policy. Thus, in this research, a four-step climate change risk assessment is proposed that is considering the current limited time/resources and insufficient climate change information: collecting scientific evidence, making lists of preliminary risks, making lists of risks and prioritising, and categorising the risks. Based on a Korean case, the proposed risk assessment process took only 6-8 months to draw meaningful assessment results.
This risk assessment process retained scientific evidence and data about the climate change risks of Korea through the first two steps. Through an SLR and the climate impact database of Korea, this research drew 204 climate impacts on Korea and collects related scientific evidence. We found that industry/energy, agriculture, water, and ecosystem sectors relatively had more research evidence than other sectors. Although we collected plenty of scientific evidence of risks, there are still gaps. The gaps in data and the integration of quantitative and qualitative information are the common challenges of national-level risk assessments (EEA 2018). Thus, it is important that the risk assessment process has room for adding scientific evidence and data of risks in the next steps through additional analysis or expert discussions. However, it is emphasised that these early steps of the proposed risk assessment in this study can retain baseline scientific evidence and data of climate change risks for Korea, which were not made in the previous assessments.
The process of assessing the magnitude and adaptive capacity of risks revealed that the risk assessment process could involve three components of risks: hazard, exposure, and vulnerability. This research suggested expert surveys and expert group discussions to identify climate change risks for Korea and to prioritise identified risks, not to calculate the magnitude of risks and vulnerability with statistical models. Expert survey and expert group discussion methods were commonly used to assign weights of indicators to identify high-vulnerabilities and to classify grades in other risk assessments (Feng and Chao 2020). Based on the criteria of risk severity and risk probability, sectoral experts assessed the 95 selected climate change risks from the preliminary risk lists. In addition, to assess vulnerability, this research suggested assessing an adaptive capacity for each selected risk in four aspects: institutional capacity, actor capability, infrastructure capacity, and technological capability. Although the assessments still relied on experts' subjective opinions, this assessment provided clear criteria and factors for assessing compared to the previous risk assessment.
The 93 risks were chosen as the final national climate change risks for Korea by displaying the risks on sectoral quadrant sheets consisting of risk magnitude and adaptive capacity dimensions. Also, the priority of each risk in each sector was given in this step. Most high priorities in risks have high risk magnitude but low adaptive capacity (in the upper-right quadrant).
This research provided insights for directions of national adaptation policy of each sector by categorising the 93 risks into four categories (existing, new, fundamental research, and new and fundamental research). The results showed that only one-third of national climate change risks of Korea can be dealt with existing measures or policies. Also, there were differences between sectors. In particular, agriculture and industry/energy sectors can deal with most sectoral risks with existing measures, whereas ecosystem and ocean/fishery/coast sectors cannot deal with any risks with existing adaptation measures. These results indicate that sectors like agriculture and industry/energy need to focus mainly on maintaining the current adaptation measures in the next NCCAP. Ecosystem and ocean/fishery/coast sectors have to check the problems or directions of the current adaptation measures first and then develop their measures based on identified risks for the next NCCAP. In addition, the results revealed that Korea still does not have enough fundamental research to address climate changes risks and need to invest in more to each sector's fundamental research, although the government has implemented the adaptation policy for about past 10 years. More than one-third of the national climate change risks for Korea requires new measures and fundamental research. The risks of sectors seem to need immediate actions to address them. However, policy-makers should pay attention to and invest in fundamental research for the risks with long-term views. Moreover, for the risks with a lack of research but requiring quick responses, it is necessary to prepare policies through a discussion process so that basic research and direct action can be carried out at the same time.
It is acknowledged that there were several limitations in this risk assessment method. First, this assessment had a limitation in dealing with cross-cutting risks, as we focused on sectoral risks separately. Secondly, this risk assessment did not involve a spatial concept; thereby, it did not involve how the national risk assessment link with and use sub-nationallevel climate change risk assessments. Thus, there is a need for research on risk assessments that include horizontal and vertical cross-cutting risk issues. Third, the SLR in collecting scientific evidence step reviewed only domestic journals to set subjective and clear inclusion criteria. If one can set subjective criteria that can include international journals, it would be helpful to collecting a wider range of scientific evidence related to climate change risks. In addition, because this risk assessment focused on establishing a systematic methodology that can be efficiently conducted with limited resources and time, it did not address every specific factor related to a climate change risk separately. For example, to assess a vulnerability, the proposed risk assessment focuses on adaptive capacity, excluding a sensitivity (such as elderly population density), to make the assessment more efficient and easier based on only related policy data, excluding demographic data. To address this limitation, it is required to develop an assessment method that can combine sensitivity with demographic data and adaptive capacity with related policy data.
With the Paris Agreement and Katowice Climate Package, it became a clear mandate for all parties to undertake adaptation progress and report their efforts to the international society (Berrang-Ford et al. 2019), and a climate change risk assessment is being essentially required for their national adaptation policy. In this context, this study will be of great help to countries where a climate change risk assessment needs to be conducted immediately but systematically. In addition, this assessment process can be used not only at the national level, but also at local or individual organisational level risk assessments. Crop productivity quality decrease due to temperature increase 14 Energy consumption of livestock barns increase due to heat wave and cold wave