The Application of a Smart Nexus for Agriculture in Korea for Assessing the Holistic Impacts of Climate Change

Abstract

Sustainable development involves maximizing the benefits of development while minimizing its consequent effects on the environment. This study uses a water–energy–food nexus framework, the Smart Nexus for Agriculture in Korea (SNAK), to assess the impact of climate change on sustainable resource management in agriculture. The nexus database applied in this study comprises three individual databases related to resources, interlinkages, and resource management scenarios, which include all variables and scenarios of the framework. Different resource management scenarios were evaluated via investigating the interlinkages between resources and quantifying resource consumption and sustainability. The variable selection and application module uses the interlinkage database to quantitatively model how the production and supply of one resource affects the consumption of other resources. The scenario analysis module involves the identification and application of resource management scenarios based on policies for individual resources and climate change. The sustainability evaluation module links the previous two modules to quantify food production, the consumption of food and energy resources, carbon (CO₂) emissions, and land use in each scenario. Finally, resource security and economic benefits were considered when estimating the sustainability index of each scenario. The SNAK platform is anticipated to possess the ability to analyze environmental, social, and economic systems grounded in water, energy, and food. It is believed that the platform can optimize the timing and allocation of agricultural resources, leading to the derivation of optimal management scenarios. Furthermore, the platform will utilize water–energy–food linkage assessments to formulate scenario-based policies addressing food demand, water resource utilization, and energy consumption.

1. Introduction

Water, energy, and food (WEF) are the material basis and vital resources for socioeconomic development [1]. Population growth, industrialization, urbanization, and changing lifestyles pose threats to WEF security [2]. Climate, with its regional and temporal variability, is a major determinant of agricultural production. All agricultural production is related to the performance of (cultivated) species, which are bound to particular environmental conditions. As climatic conditions change, production conditions are likely to change, with possible positive or negative implications for agricultural production [3,4,5,6]. The effect of climate change on crop yields varies according to the area and irrigation application. Crop yields can be increased by expanding irrigated areas, which can have a detrimental effect on the environment [7,8]. Kang et al. (2009) estimated the impacts of climate change on crop productivity using climate, water, and crop yield models and reported that climate change was a significant threat to global food security. They forecasted a significant reduction in global crop yield in the future due to fluctuations in weather [7,9]. In accordance with this, it is essential to implement appropriate responses and measures for climate factors affecting agricultural production. Additionally, efforts to establish sustainable agricultural systems are crucial.

Due to their vital roles and interdependencies, addressing the interplay and trade-offs between water, energy, and food systems, often described as the water–energy–food (WEF) nexus, has risen to prominence in policy and development discourses as a framework for addressing scarcities and achieving sustainable development in WEF sectors [10,11,12,13]. In particular, rural areas serve as crucial regions for food security, directly linked to urban areas’ food consumption, as they play a key role in providing food for urban areas. The utilization of resources for food production in rural areas has a direct impact on the food security of urban areas. Farmland abandonment in the rural–urban context may significantly alter food security with local and distal implications regarding land-use sectors [14,15,16,17].

The concept of the nexus emerged during the 2008 World Economic Forum organized by the United Nations. The nexus is defined as a system where the production, consumption, and management of water, energy, and food are interconnected, and the impact on one resource can affect others. At the 2011 Bonn conference in Germany, there was a strong emphasis on the need for decision making via the interaction of water, food, and energy security to promote sustainable development in society [18]. The Water Security report introduced a novel methodology, the Nexus approach, which allows for simultaneous consideration of the interactions among various factors [19]. Meanwhile, the OECD, via its report on “Innovation, Agricultural Productivity, and Sustainability in Korea”, has provided various policy recommendations for achieving sustainable agriculture and fostering agricultural innovation via analyzing the agricultural conditions and policies in South Korea [20]. However, in South Korea, research related to the water–food–energy nexus is currently concentrated on conceptual aspects, and there is a lack of technological development in interpreting the interrelationships among these factors. Given the recognized importance of integrated resource management both domestically and internationally, there is an urgent need for the development of technologies that can interpret these interconnections. In 2023, the World Economic Forum announced that the failure in climate change mitigation is the most significant global risk, as per The Global Risks Report 2023. Climate change has triggered natural disasters such as water scarcity, droughts, floods, and rising sea levels. It has adverse effects on food and energy production and has the potential to exacerbate social conflicts and instability. Additionally, both natural disasters and abnormal climate conditions are evaluated as major global risks. The frequency and intensity of natural disasters are increasing due to climate change, causing harm to human lives and property and potentially leading to socioeconomic disruption [21]. Climate change encompasses not only changes in specific weather conditions but also an increase in climate variability. This inherent uncertainty in securing resources in future periods emphasizes the growing importance of resource security. As the significance of resource security rises, the importance of securing critical resources and practicing sustainable resource management becomes increasingly vital [22].

The nexus can be a crucial tool for achieving sustainable development. System resilience and the metabolism of a system are at the “core of nexus security” since they highlight the diversity of options within a system [23]. Furthermore, recent scientific debates emphasized various risks to sustainability due to increasing demands but also from resource intensifications in the quest to increase resource use efficiency. The WEF nexus as a “sustainability innovation” faces the challenge of providing better solutions toward “sustainable intensification” [24]. Analyzing interdependencies among resources via nexus approaches can support policy development to address sustainability issues such as water scarcity, food insecurity, and climate change. Moreover, integrated nexus assessments serve as powerful tools to understand the interdependence of resources and support policy formulation for sustainable development. In the context of the water–energy–food nexus, assessing water scarcity is crucial for understanding the causes and impacts of water scarcity, considering the interdependence of water, energy, and food, and supporting policy development to address these issues.

In 2023, a methodology for assessing agricultural climate adaptation from the perspective of the water–energy–food nexus [25]. They conducted an evaluation of the impacts of climate change on water, energy, and food, scrutinizing the effects of agricultural technologies like crop variety improvement, irrigation technology enhancement, and farming method improvement, along with related policies for adaptation [25]. In the same year, Javan et al. analyzed the water–energy–food nexus and resource sustainability in Ardabil Plain, a pivotal agricultural region in Iran, proposing optimization strategies. Their findings provide valuable information for policymakers to endorse sustainable agricultural development in Ardabil Plain, with potential applicability in similar regions [26]. Furthermore, Li et al. (2022) innovatively developed an optimization approach for the nexus of agricultural water, food, and energy, enhancing the efficiency of agricultural resources via multi-energy coordination [27]. This approach introduces a fresh perspective on optimizing agricultural resources within the water–energy–food nexus, with the potential to contribute to policy and technological development for sustainable agricultural production [27].

The nexus is intricately tied to sustainability, and given the inherent uncertainties, it becomes crucial to address external factors such as climate change that affect vital resources when implementing a policy framework or resource management in sustainable development. Moreover, in agriculture, where individual or dual-resource connections fall short, an integrated approach becomes imperative, encompassing the interconnections among pivotal resources like water, food, and energy, and acknowledging the substantial role of carbon emissions within the context of climate change. Hence, conducting research is essential to comprehensively assess simulations based on scenarios and the overarching impact of climate change, considering both resource utilization and carbon emissions.

The agricultural sector requires diverse resources as inputs, and output is directly connected to food security, which is fundamental to national competitiveness. Thus, it is necessary to examine the effectiveness of different policies in considering interlinkages across different resources and securing agricultural sustainability in a changing climate. Thus far, WEF-Climate nexus research in Korea has mainly focused on analyzing individual factors and evaluating climate change effects and has failed to assess the interlinkages across resources in combination with the effects of climate change. To overcome this limitation and assess interlinkages across a WEF nexus, a systematic approach is required. A WEF nexus based on a system dynamics approach can be used to develop a platform that links diverse resources, such as water, energy, food, greenhouse gases, and economics, to model different climate change and food security scenarios. This platform will act as an important basis for agricultural policymaking in Korea by suggesting an integrated WEF management system and infrastructure for sustainable agriculture under different climate change scenarios.

This study applied a WEF nexus assessment technique based on climate change scenarios. The aims were (1) to establish a database and simulation modules to analyze dependency and trade-offs across the production and utilization of resources within the nexus platform and (2) to assess the interlinkages among water, energy, and food under climate change impacts using the Smart Nexus for Agriculture in Korea (SNAK).

2. Methodology

2.1. The Water–Energy–Food Nexus System

Sustainable development seeks to maximize the efficient use of resources while minimizing the consequent negative effects on the environment. The WEF nexus model is an ideal tool for such sustainable development. A WEF nexus model is composed of an input database, a simulation module, and a scenario evaluation output. The nexus database includes all the variables and scenarios for the nexus framework and is further differentiated into a resource, interlinkage, and scenario databases. The resource database contains the availability and average consumption of each resource considered, and the interlinkage database contains quantified data on the interrelationships between resources, such as the amount of water required to produce 1 ton of crops or the energy input in 1 ha of cropland. The scenario database contains data on the national food self-sufficiency rate, alternative renewable energy supply rate, and water consumption rate of different water sources, which were used to build management scenarios for each resource sector in the system. The interlinkages across resource sectors are investigated via the simulations of different resource management scenarios in which resource consumption and sustainability potential are quantified. The simulations are based on a system dynamics approach, in which the interlinkage database is utilized in the variable selection module to quantify how production and supply from one resource sector affect consumption in other interlinked resource sectors. In the scenario module, resource management scenarios based on policies for individual resources are modeled to obtain the national food self-sufficiency rate and alternative renewable energy consumption rate. Finally, the resource security and economic benefits of each scenario are analyzed to calculate a sustainability index.

Based on this concept, several global nexus models have been developed. The Climate, Land-use, Energy, and Water (CLEW) nexus model is a framework developed based on quantitative tools to comprehensively evaluate resource dynamics. The purpose of the CLEW model is to provide solutions for issues related to food, energy, and water resource security while considering how the consumption of resources affects climate and how climate change affects the environment. In general, the CLEW nexus model requires extensive input data, including parameters related to the technical and economic aspects of resources, such as power plants, agricultural equipment, irrigation systems, and fertilizer production. These parameters are used to calculate the energy balance, including the energy required for food production, water balance, including water consumption in agricultural irrigation and hydropower plants, and to support agricultural decision making, such as fertilization rate and equipment usage. The model user interface was developed mostly for graphic information systems-based parameters rather than allowing for the manual manipulation of inputs.

The Water Evaluation and Planning Long-Range Energy Alternatives Planning System (WEAP-LEAP), developed by the Stockholm Environment Institute, is an energy-based model that combines water and energy resource planning tools. The model requires extensive technical and economic input data from the energy sector and provides energy balance as an output that includes the current energy supply and demand. In addition, it models the supply and demand of water resources in detail and provides information on groundwater, water quality, and hydropower supply. WEAP-LEAP version 3.4 is a user-friendly software tool that considers diverse resources for water resource planning and provides a flexible and holistic framework for policy planning and assessment. Water resource experts are increasingly using and extending the WEAP-LEAP model by including other models, databases, spreadsheets, and software. However, the WEAP-LEAP model is PC-based with low accessibility to web-based users and presents some limitations in the sharing of results. The WEF Nexus Tool 2.0, a platform developed by Texas A&M University to overcome the limitations of existing models and enable scenario-based analysis to support decision making, simplifies the flow between resources and does not consider interactive feedback. Nevertheless, this web-based system is highly accessible and presents an effective user interface for providing information to support decision-making processes. The Nexus Calculator is another tool developed jointly by the University College London of Newcastle University and Open Lab. The calculator integrates aspects of water, energy, food, and waste data to support bottom-up innovation for the joint design of technology and infrastructure, demonstrating a new paradigm for a supply system that directly meets household demands. In particular, an indirect scenario application on a simple user interface enables the easy representation of interactions among water, energy, food, waste, and data components. NexSym integrates ecosystem models with production and consumption components via modeling the interactions between the technology and ecosystem of a region to simulate and analyze the flow within a local system. This provides knowledge and understanding of important interactions among the components to balance supply and demand, as well as to increase synergies to sustain local ecosystems. NexSym is a PC-based program that provides various scenarios and results. AgroNexus, crafted by the Consultative Group on International Agriculture Research (CGIAR), serves as a platform employed to fathom and scrutinize the intricate interrelationships between agriculture and climate, water, energy, and ecosystems. It amalgamates data and models from the agricultural sector, providing simulation and decision-support tools. Platforms of this nature aim to tackle nexus issues in agriculture by presenting diverse data and models alongside simulation and decision-support tools. The Agri-Food Systems Nexus, devised by the United States Department of Agriculture (USDA), assesses and enhances the sustainability of agri-food systems, offering analytical tools that consider environmental, economic, and social aspects. The Water–Food–Energy Nexus in Agriculture, developed by Wageningen University & Research in the Netherlands, aids in comprehending and managing the relationships between water, food, and energy in agriculture. It integrates data and models from the agricultural sector and furnishes simulation and decision-support tools. The utilization of these platforms is anticipated to improve sustainability in agriculture, contributing to the resolution of challenges such as climate change, food security, and water security.

However, while nexus systems predominantly focus on analyzing present interconnections, it is crucial to consider how climate change can be incorporated into these systems for the future sustainability of agriculture. Therefore, this study seeks to answer the research question of how the quantitative assessment of the impact of water management on food, energy, and environmental factors under future climate change conditions can be conducted. To accomplish this, we plan to employ a climate-agriculture-specific nexus system known as SNAK, grounded in climate change and water management scenarios. The objective is to evaluate the combined impact of climate change and resource management methods on agriculture.

2.2. Application of SNAK as an Analytical Model

2.2.1. Structure of SNAK

In this study, Smart Nexus for Agriculture in Korea (SNAK) was applied to investigate the effects of climate change on agriculture via the nexus concept. In principle, SNAK analyzes the trade-offs among the water, energy, and food sectors to evaluate resource sustainability. Thus, the influence of individual resource consumption on the status of other interlinked resources, as well as the effects of external factors, such as climate and population, on resource consumption can be understood.

Figure 1 shows the structure of SNAK, including the flow of data and linkages of modules to outputs. The data used for analysis in SNAK differ in format, size, and time, depending on the resources, systems, and external factors considered. For example, the water footprint, which is required for resource interlinkage analysis, can be obtained via modeling the water requirements for crop production or from existing research data. Furthermore, data on region-specific water, land, and energy resource capacities, as well as on cultivation and irrigation systems, are required for regional analysis. Appropriate data were identified and categorized into resource, interlinkage, and scenario databases. The resource database mainly comprised data on the availability of individual resources and current consumption rates, especially the production and demand for food-related resources. The interlinkage database included water and energy footprints, which represented the water and energy consumption per unit of agricultural production, energy consumption per unit of irrigation water supplied, and carbon output per unit of energy consumption. Finally, the scenario database mainly comprised data related to resource management, such as food self-sufficiency and alternative energy consumption rates.

In the simulation modules, a system dynamics approach was used in the analysis tool to explore the complexity of the influence on the system of a change in each individual element. The system dynamics applied in the nexus model simulated resource consumptions in each production, water, and energy supply system, as well as carbon emissions. Particularly for resource consumption simulations, land productivity, and water and energy footprints were crucial input data that were obtained as either field measurements or simulated values from sub-models. For example, resource interlinkage analyses for different crop cultivation methods showed that water requirements for field crops could be partially met by precipitation, which reduced irrigation water consumption; however, this led to the sensitivity of the open-field cultivation method to changes in precipitation due to climate change. Agricultural production, cultivation area, and water requirements could be estimated using the production rate per unit area of cultivation, irrigation water consumption per unit production, and energy consumption per unit area of cultivation (or production).

Finally, water and energy consumption, food production, and carbon emissions were simulated for climate change and water management scenarios. Climate change was a critical factor influencing the water–energy–food nexus, and one of the main objectives of the SNAK model platform was to comprehensively analyze various scenarios, including climate change impacts. Therefore, a system in which a multiple general circulation model (GCM)-based scenario applied to a biophysical model was used in the nexus platform to analyze how resources are affected by climate change and to support decision making in resource management in response to climate change. Any changes in crop demand and production, which are critical factors in nexus analysis due to climate change, can yield different results from a single food-related scenario. The nexus system was constructed to allow changes in scenarios to be preset by constraining the fluctuations in the parameters caused by uncertainties in climate change.

2.2.2. Water–Energy–Food–CO₂ Simulations

For food production simulations, crop production requirements mediated by the food self-sufficiency rate were linked to food consumption and population data using rice as the representative crop. Specifically, future food consumption was estimated based on changes in population and per capita consumption, which were then used to calculate the crop production required to meet the predetermined self-sufficiency rate. Therefore, the self-sufficiency rate could be manipulated for each scenario in the model. Food production may also be affected by the fertilization rate and agricultural management methods such as continuous or intermittent irrigation, organic matter input, inorganic nitrogen fertilization, and green manure. The amount of land required for agricultural production was estimated as follows:

$$Population\times Consumption per capita(\mathrm{ton}/\mathrm{capita}/\mathrm{yr})=Consumption \left(\mathrm{ton}\right)$$

(1)

$$\{Consumption\left(\mathrm{ton}\right)\times Self-sufficiency rate(\%)+Export \left(\mathrm{ton}\right)=Production\left(\mathrm{ton}\right)$$

(2)

$$Production\left(\mathrm{ton}\right)÷Productivity\left(\mathrm{ton}/\mathrm{ha}\right)=Land use\left(\mathrm{ha}\right)$$

(3)

Water requirements in open-field cultivation vary depending on the crops and cultivation methods, and details of cropping systems, stages of crops, and days since seeding or transplanting are required to calculate the water requirement for a particular crop. Water requirements may also vary depending on the irrigation method, such as continuous or intermittent. Therefore, the water consumption for different irrigation methods and crop production were considered together in the estimation of water requirements as follows:

$$Water footprint ({\mathrm{m}}^3/\mathrm{ton})\times Production \left(\mathrm{ton}\right)=Irrigation water requirement ({\mathrm{m}}^3)$$

(4)

$$Irrigation water requirement \left({\mathrm{m}}^3\right)\times \left\{1+Irrigation efficiency \left(\%\right)\right\}=Total water supply \left({\mathrm{m}}^3\right)$$

(5)

Within the water–energy interlinkage, energy consumption can be divided into energy consumed during agricultural equipment operations and energy consumed during irrigation. Further energy consumption during irrigation may occur when an additional pumping step, either via a pumping station or mobile pump, is required rather than supplying water directly from a reservoir. The electrical energy supply required to operate irrigation systems in pumping stations, reservoirs, and mobile pumps was calculated as follows:

$$Electricity footprint for water supply\left(\mathrm{GJ}/{\mathrm{m}}^3\right)\times Total water supply\left({\mathrm{m}}^3\right)=Electricity use for water supply\left(\mathrm{GJ}\right)$$

(6)

The energy and interlinkages between the resources used in agricultural facilities showed that the main source of energy consumption during field crop cultivation was the operation of agricultural equipment. The relationship between the energy sector and food production in the SNAK model was as follows:

$$Fuel footprint by food systems (\mathrm{GJ}/\mathrm{ha})\times Land use \left(\mathrm{ha}\right)=Fuel use for food \left(\mathrm{GJ}\right)$$

(7)

$$Electricity footprint for food systems (\mathrm{GJ}/\mathrm{ha})\times Land use\left(\mathrm{ha}\right)=Electricity use for food \left(\mathrm{GJ}\right)$$

(8)

Carbon emissions from energy consumption were distinguished and calculated as direct CO₂ emissions and indirect CO₂ emissions. Direct CO₂ emissions were calculated using CO₂ emissions from fossil fuel (diesel and kerosene) consumption during the operation of agricultural equipment (fuel used for food) and CO₂ emissions per unit of fossil fuel consumption (direct CO₂ footprint). Indirect CO₂ emissions include CO₂ emissions during electricity production in power plants and CO₂ emissions per unit of electricity produced in different power plants such as coal-fired and nuclear power plants. Therefore, carbon emissions resulting from water–energy–food interlinkages were expressed as follows:

$$Direct {CO}_2 footprint (\mathrm{ton} \mathrm{of} {\mathrm{CO}}_2/\mathrm{GJ})\times Fuel use for food \left(\mathrm{GJ}\right)=Direct {CO}_2 emission (\mathrm{ton} \mathrm{of} {\mathrm{CO}}_2)$$

(9)

$$\begin{gathered} Indirect footprint (\mathrm{ton} \mathrm{of} {\mathrm{CO}}_2/\mathrm{GJ})\times \{Electricity use for water supply \left(\mathrm{GJ}\right)+ \\ Electricity use for food \left(\mathrm{GJ}\right)\}=Indirect {CO}_2 emission (\mathrm{ton} \mathrm{of} {\mathrm{CO}}_2) \end{gathered}$$

(10)

The carbon footprint represents the total greenhouse gas emissions during agricultural activities in open fields, and the emission factors differ depending on the type of fuel. Agricultural management practices such as continuous or intermittent irrigation also yield different emission factors. For irrigation water supply, carbon emissions differ depending on whether energy is required in the process. Furthermore, organic matter input and fertilizer type yield different emission factors that must be considered in CO₂ emission estimations.

3. Results and Discussion

3.1. Application of Climate Change and Management Scenarios to SNAK

The main function of SNAK is to assess holistic impacts. Thus, we applied the climate change scenarios shown in

Table 1. Analysis model and scenario conditions.

Scenario	Period		GCM
Observation	Observation	1976–2018	-
RCP 4.5	Short term	2020–2059	bcc-csm1-1
			CanESM2
			GFDL-ESM2G
			GFDL-ESM2M
	Long term	2060–2099	HadGEM2-CC
			inmcm4
			MIROC-ESM
			MIROC-ESM-CHEM
RCP 8.5	Short term	2020–2059	bcc-csm1-1
			CanESM2
			GFDL-ESM2G
			GFDL-ESM2M
	Long term	2060–2099	HadGEM2-CC
			inmcm4
			MIROC-ESM
			MIROC-ESM-CHEM

and assessed the impacts on water, energy, and food in rice paddies, which are the main food production areas in Korea. To analyze food production requirements, weather data, including daily precipitation, temperature, humidity, mean wind speed and solar radiation, were obtained, and eight climate models with data projected up to 2099 were selected to establish climate change scenarios. The model and scenarios were built based on weather data measurements from 1976 to 2018 in the city of Suwon. Simulations were run for the short-term future (2020–2059) and long-term future (2060–2099) based on eight GCMs (bcc-csm1-1, CanESM2, GFDL-ESM2G, GFDL-ESM2M, HadGEM2-CC, inmcm4, MIROC-ESM, and MIROC-ESM-CHEM) and two representative concentration pathway (RCP) scenarios (RCP 4.5 and RCP 8.5). Future climate change scenarios were used to generate weather data that could be applied to the model.

In addition, we applied management scenarios for the analysis that were divided into business-as-usual (BAU) and climate change adaptation scenarios. In the BAU scenario, the source of water was 100% reservoirs, continuous irrigation was conducted, fertilization rate was 9 kg/10 a, no green manure was applied, and the mid-summer drainage period was one week. In the climate change adaptation scenario, the sources of water were 50% reservoirs and 50% pumping stations, intermittent irrigation was conducted, fertilization rate was the same as under BAU at 9 kg/10 a, no green manure was applied, and the mid-summer drainage period was two weeks. The two scenarios were simulated to analyze the maximum, mean, minimum, and 10-year frequency values of water consumption (m³), energy consumption (Gcal), and crop yield (tons).

3.2. Assessment of Holistic Impacts of Climate Change Using the SNAK

The climate change adaptation model resulted in a 12.5% reduction in the maximum water consumption, while the maximum energy consumption increased by 63.6%, and the maximum crop yield decreased by 2.1% compared with the maximum values observed during the field measurement period (1976–2018). Similarly, the mean water consumption decreased by 16.1%, while the mean consumption rate increased by 44.5%, and the mean crop yield decreased by 1.5%. For the minimum values, water consumption decreased by 18.8%, and energy consumption increased by 21.7%, whereas crop yield did not change (

Table 2. Maximum, mean, and minimum values during the observation period (1976–2018).

	BAU			Climate Change Adaptation Scenario
	Water Consumption	Energy Consumption	Food Production	Water Consumption	Energy Consumption	Food Production
	(1000 m3)	(Gcal)	(ton)	(1000 m3)	(Gcal)	(ton)
Maximum	11,310	870	6215	9893	1423	6084
Mean	8246	870	5693	6921	1257	5609
Minimum	4160	870	2773	3378	1059	2773

).

The future climate change RCP 4.5 scenario was simulated using eight GCMs for the short-term future (2020–2059) and long-term future (2060–2099) to analyze the maximum, mean, and minimum values of resource consumption and crop yield (

Table 3. Maximum, mean, and minimum values for the short-term (2020–2059) and the long-term (2060–2099) future under the RCP 4.5 scenario GCM.

GCM	Period		BAU			Climate Change Adaptation Scenario
			Water Consumption	Energy Consumption	Food Production	Water Consumption	Energy Consumption	Food Production
			(1000 m3)	(Gcal)	(ton)	(1000 m3)	(Gcal)	(ton)
bcc-csm1-1	Short term	Maximum	11,220	870	6215	9970	1427	6208
		Mean	7618	870	5495	6413	1229	5442
		Minimum	5250	870	4410	3978	1092	4411
	Long term	Maximum	11,400	870	5918	10,454	1454	5885
		Mean	7962	870	5570	6715	1245	5472
		Minimum	4710	870	4916	3504	1066	4936
CanESM2	Short term	Maximum	10,470	870	6154	9128	1380	6062
		Mean	7702	870	5540	6478	1232	5482
		Minimum	3650	870	4958	2952	1035	4939
	Long term	Maximum	10,580	870	6031	9186	1384	6035
		Mean	7115	870	5417	5957	1203	5375
		Minimum	3510	870	4297	2681	1020	4297
GFDL-ESM2G	Short term	Maximum	12,820	870	5993	11,451	1510	5991
		Mean	8231	870	5513	7015	1262	5404
		Minimum	3290	870	4847	2439	1006	4847
	Long term	Maximum	11,850	870	6201	10,551	1460	6211
		Mean	7784	870	5570	6594	1239	5460
		Minimum	4190	870	4653	2952	1035	4653
GFDL-ESM2M	Short term	Maximum	11,550	870	6127	10,203	1440	6057
		Mean	8516	870	5516	7235	1274	5430
		Minimum	2670	870	4038	2004	982	4038
	Long term	Maximum	13,130	870	6250	11,732	1526	6140
		Mean	8266	870	5577	7030	1263	5486
		Minimum	4190	870	4036	3475	1064	4037
HadGEM2-CC	Short term	Maximum	12,310	870	6061	10,987	1484	6023
		Mean	8446	870	5638	7173	1271	5545
		Minimum	4210	870	4817	3311	1055	4775
	Long term	Maximum	12,280	870	6068	10,948	1482	6025
		Mean	8251	870	5460	7058	1265	5398
		Minimum	3960	870	3647	2672	1019	3647
inmcm4	Short term	Maximum	12,340	870	6189	10,958	1483	6027
		Mean	8657	870	5409	7406	1284	5355
		Minimum	4910	870	4124	3688	1076	4021
	Long term	Maximum	11,990	870	6035	10,745	1471	5961
		Mean	8330	870	5345	7113	1268	5294
		Minimum	4610	870	3883	3301	1055	3883
MIROC-ESM	Short term	Maximum	11770	870	6213	10,396	1451	6196
		Mean	8492	870	5891	7236	1274	5846
		Minimum	5260	870	4938	3756	1080	4947
	Long term	Maximum	12,510	870	6197	11,103	1491	6210
		Mean	8491	870	5910	7257	1276	5869
		Minimum	2880	870	5368	2207	993	5311
MIROC-ESM-CHEM	Short term	Maximum	10,770	870	6207	9448	1398	6218
		Mean	7518	870	5727	6308	1223	5657
		Minimum	2730	870	4798	1994	981	4798
	Long term	Maximum	11,630	870	6255	10,329	1447	6264
		Mean	7637	870	5884	6491	1233	5833
		Minimum	4560	870	4841	3069	1042	4725

and Figure 2, Figure 3 and Figure 4). For water consumption, the maximum, mean, and minimum values were 11,789,000 m³, 8,063,000 m³, and 4,036,000 m³, respectively, under the BAU scenario, and 10,474,000 m³, 6,842,000 m³, and 2,999,000 m³, respectively, under the climate change adaptation scenario. Thus, water consumption decreased under the climate change adaptation scenario compared with the BAU scenario, which was attributed to different agricultural management practices. Energy consumption in the BAU scenario only occurred in the cultivation practices, which resulted in a maximum, mean, and minimum consumption of 870 Gcal. In the climate change adaptation scenario, where additional energy was consumed in acquiring water from pumping stations, the energy consumption values were 1456, 1252, and 1038 Gcal, respectively. Crop yield fluctuated the least relative to the other resource sectors, resulting in maximum, mean, and minimum values of 6,132, 5,591, and 4536 tons, respectively, under the BAU scenario, and 6095, 5522, and 4517 tons, respectively, under the climate change adaptation scenario.

In the short-term future period (2022–2059), RCP 4.5 scenario analysis with MIROC-ESM_CHEM under the climate change adaptation model decreased the maximum water consumption by 12.3% but increased the maximum energy consumption and crop yield by 60.7% and 0.2%, respectively. The mean water consumption decreased by 16.1%, the mean energy consumption increased by 40.6%, and the mean crop yield decreased by 1.2%. The minimum values in the climate change adaptation scenario in MIROC-ESM decreased by 28.6% for water consumption and increased by 24.1% and 0.2% for energy consumption and crop yield, respectively.

In the long-term future period (2060–2099) of the climate change adaptation scenario in the CanESM2 model, the maximum value of water consumption decreased by 13.2% and increased by 59.1% and 0.1% for energy consumption and crop yield, respectively. The mean values decreased by 16.3 for water consumption, increased by 38.3% for energy consumption, and decreased by 0.8% for crop yield. The minimum value for water consumption in the climate change adaptation scenario in the bcc-csm1-1 model decreased by 25.6% but increased the minimum energy consumption and crop yield by 22.5% and 0.4%, respectively.

The future climate change RCP 8.5 scenario was simulated using eight GCMs for the short-term future (2020–2059) and long-term future (2060–2099) to analyze the maximum, mean, and minimum values of resource consumption and crop yield (

Table 4. Maximum, mean, and minimum for short-term (2020–2059) and long-term (2060–2099) future under the RCP 8.5 scenario GCM.

GCM	Period		BAU			Climate Change Adaptation Scenario
			Water Consumption	Energy Consumption	Food Production	Water Consumption	Energy Consumption	Food Production
			(1,000 m3)	(Gcal)	(ton)	(1,000 m3)	(Gcal)	(ton)
bcc-csm1-1	Short term	Maximum	10,520	870	5881	9254	1387	5893
		Mean	7541	870	5409	6376	1226	5361
		Minimum	3010	870	3321	2101	987	3321
	Long term	Maximum	10,130	870	5835	8828	1363	5850
		Mean	7099	870	5044	5961	1203	5013
		Minimum	4240	870	3703	3407	1060	3703
CanESM2	Short term	Maximum	9620	870	6315	8451	1342	6111
		Mean	7290	870	5321	6033	1207	5261
		Minimum	4090	870	4116	2798	1026	4116
	Long term	Maximum	10,500	870	5738	9051	1376	5656
		Mean	7262	870	4816	6073	1209	4804
		Minimum	4360	870	3387	3640	1073	3387
GFDL-ESM2G	Short term	Maximum	12,500	870	6084	11,151	1493	5988
		Mean	8351	870	5501	7170	1271	5423
		Minimum	4160	870	4180	3446	1063	4242
	Long term	Maximum	13,900	870	5899	12,545	1571	5823
		Mean	7905	870	5284	6748	1247	5188
		Minimum	3040	870	4178	2004	982	4190
GFDL-ESM2M	Short term	Maximum	12,320	870	6182	10,967	1483	6188
		Mean	8478	870	5513	7239	1275	5404
		Minimum	5460	870	4511	4433	1118	4415
	Long term	Maximum	11,210	870	6127	9825	1419	6125
		Mean	7651	870	5371	6425	1229	5285
		Minimum	4820	870	3830	3688	1076	3848
HadGEM2-CC	Short term	Maximum	11,790	870	6164	10,454	1454	6066
		Mean	7771	870	5462	6556	1236	5401
		Minimum	5290	870	4244	4046	1096	4244
	Long term	Maximum	11,560	870	6160	10,280	1445	5822
		Mean	7937	870	4774	6729	1246	4728
		Minimum	4920	870	3024	4066	1097	3025
inmcm4	Short term	Maximum	12,710	870	6225	11,355	1505	6065
		Mean	8437	870	5387	7231	1274	5338
		Minimum	4980	870	3974	4095	1099	3974
	Long term	Maximum	12,120	870	5997	10,967	1483	5868
		Mean	8044	870	4823	6830	1252	4816
		Minimum	3970	870	3469	3262	1052	3469
MIROC-ESM	Short term	Maximum	12,100	870	6199	10,580	1461	6162
		Mean	8634	870	5751	7332	1280	5646
		Minimum	4190	870	4982	3427	1062	4869
	Long term	Maximum	12,050	870	6173	10,609	1463	6187
		Mean	8191	870	5483	6995	1261	5456
		Minimum	4000	870	3453	3020	1039	3454
MIROC-ESM-CHEM	Short term	Maximum	13,540	870	6255	12,168	1550	6259
		Mean	8074	870	5646	6809	1251	5567
		Minimum	4790	870	3591	3746	1079	3571
	Long term	Maximum	11,530	870	6070	10,241	1442	6090
		Mean	6876	870	5355	5671	1187	5327
		Minimum	2460	870	3149	1220	938	3150

and Figure 5, Figure 6 and Figure 7). The water consumption maximum, mean, and minimum under BAU were 11,756,000 m³, 7,846,000 m³, and 4,236,000 m³, respectively (short-term future), and 10,421,000 m³, 6,636,000 m³, and 3,275,000 m³, respectively (long-term future), in the climate change adaptation scenario. Therefore, water consumption showed a trend similar to that under the RCP 4.5 scenario. Energy consumption in the BAU scenario resulted in maximum, mean, and minimum consumption of 870 Gcal, whereas in the climate change adaptation scenario, the values were 1453, 1241, and 1053 Gcal, respectively. Crop yield fluctuated the least relative to the other resource sectors, resulting in maximum, mean, and minimum values of 6082, 5309, and 3820 tons, respectively, under the BAU scenario, and 6010, 5251, and 3811 tons, respectively, under the climate change adaptation scenario.

In the short-term future period (2022–2059), RCP 8.5 scenario analysis with GFDL-ESM2M under the climate change adaptation scenario decreased the maximum water consumption by 11.0% but increased the maximum energy consumption and crop yield by 70.5% and 0.1%, respectively. Under the climate change adaptation scenario in CanESM2, the mean water consumption decreased by 17.1%, the mean energy consumption increased by 38.7%, and the mean crop yield decreased by 1.1%. The minimum values in the climate change adaptation scenario with GFDL-ESM2G decreased by 17.2% for water consumption and increased by 22.2% and 1.5% for energy consumption and crop yield, respectively.

In the long-term future period (2060–2099) of the climate change adaptation scenario in the bcc-csm1-1 model, the maximum value for water consumption decreased by 12.9% and increased by 56.7% and 0.3% for energy consumption and crop yield, respectively. Under the climate change adaptation scenario in MIROC-ESM-Chem, the mean values decreased by 17.5% for water consumption, increased by 36.4% for energy consumption, and decreased by 0.5% for crop yield. The minimum value for water consumption in the climate change adaptation scenario in the MIROC-ESM-CHEM model decreased by 50.4% but increased the minimum energy consumption by 7.8%, whereas the crop yield did not change.

In general, when the climate change adaptation scenario was compared with the BAU scenario, the maximum, mean, and minimum values for water consumption decreased, the values for energy consumption increased, and no significant change was observed in crop yield. The GCM that reduced water consumption the most under RCP 4.5 while increasing energy consumption slightly was MIROC-ESM-CHEM, whereas bcc-csm1-1 increased crop yield the most. Under RCP 8.5, MIROC-ESM-CHEM reduced water consumption the most and increased energy consumption slightly, whereas GFDL-ESM2G increased crop yield the most.

3.3. Assessment of Water and Land Footprint under Drought Conditions

To comprehensively assess the impact of future climate change, it is necessary to assess how climate change may affect water, energy, and food resources, as well as the potential trade-offs between resources caused by climate change. Trade-offs resulting from the application of the conventional and climate change adaptation scenarios suggest that under the climate adaptation scenario, water sustainability was higher, but variability in the sustainability of future food production was also higher than that in the BAU. Thus, a climate change impact assessment was performed on the nexus system under drought conditions. The 10-year drought frequencies under the conventional and climate change adaptation scenarios were calculated, as shown in

Table 5. Observation of short-term and long-term water and land footprints via the 10-year return period drought scenario GCM.

Scenario	GCM	Period	Water Footprint (1000 m3/ton)			Land Footprint (ha/ton)
			BAU	Climate Change Adaptation Scenario	Change Rate	BAU	Climate Change Adaptation Scenario	Change Rate
Observation			1627	1403	▼ 13.7%	20.890	20.621	▼ 1.3%
RCP 4.5	bcc-csm1-1	Short term	1524	1306	▼ 14.4%	22.114	22.011	▼ 0.5%
		Long term	1541	1351	▼ 12.3%	22.325	22.094	▼ 1.0%
	CanESM2	Short term	1406	1215	▼ 13.6%	22.613	22.397	▼ 1.0%
		Long term	1391	1181	▼ 15.1%	21.698	21.618	▼ 0.4%
	GFDL-ESM2G	Short term	1591	1391	▼ 12.6%	22.256	22.028	▼ 1.0%
		Long term	1520	1309	▼ 13.9%	22.460	22.245	▼ 1.0%
	GFDL-ESM2M	Short term	1541	1344	▼ 12.8%	21.798	21.565	▼ 1.1%
		Long term	1661	1475	▼ 11.2%	22.068	21.765	▼ 1.4%
	HadGEM2-CC	Short term	1589	1386	▼ 12.8%	22.565	22.346	▼ 1.0%
		Long term	1663	1434	▼ 13.8%	21.186	21.032	▼ 0.7%
	inmcm4	Short term	1694	1483	▼ 12.5%	21.797	21.409	▼ 1.8%
		Long term	1664	1438	▼ 13.5%	21.281	21.103	▼ 0.8%
	MIROC-ESM	Short term	1546	1316	▼ 14.8%	23.213	23.145	▼ 0.3%
		Long term	1480	1294	▼ 12.6%	23.669	23.577	▼ 0.4%
	MIROC-ESM-CHEM	Short term	1352	1159	▼ 14.3%	22.926	22.819	▼ 0.5%
		Long term	1476	1267	▼ 14.1%	23.199	23.015	▼ 0.8%
RCP 8.5	bcc-csm1-1	Short term	1509	1294	▼ 14.2%	20.605	20.514	▼ 0.4%
		Long term	1500	1283	▼ 14.4%	20.446	20.391	▼ 0.3%
	CanESM2	Short term	1380	1166	▼ 15.6%	21.782	21.445	▼ 1.5%
		Long term	1608	1376	▼ 14.4%	19.579	19.449	▼ 0.7%
	GFDL-ESM2G	Short term	1663	1465	▼ 11.9%	21.740	21.549	▼ 0.9%
		Long term	1671	1461	▼ 12.6%	21.237	20.959	▼ 1.3%
	GFDL-ESM2M	Short term	1668	1470	▼ 11.9%	22.256	21.904	▼ 1.6%
		Long term	1574	1342	▼ 14.7%	21.385	21.231	▼ 0.7%
	HadGEM2-CC	Short term	1587	1371	▼ 13.6%	21.932	21.715	▼ 1.0%
		Long term	1747	1552	▼ 11.2%	19.775	19.304	▼ 2.4%
	inmcm4	Short term	1711	1511	▼ 11.7%	21.597	21.326	▼ 1.3%
		Long term	1733	1523	▼ 12.2%	19.864	19.712	▼ 0.8%
	MIROC-ESM	Short term	1561	1369	▼ 12.3%	23.070	22.727	▼ 1.5%
		Long term	1646	1419	▼ 13.8%	21.385	21.332	▼ 0.3%
	MIROC-ESM-CHEM	Short term	1699	1479	▼ 13.0%	21.722	21.593	▼ 0.6%
		Long term	1478	1238	▼ 16.2%	20.835	20.790	▼ 0.2%

. Water and land footprints were analyzed for short-term (2023–2040) and long-term (2041–2060) future periods in GCMs under the RCP 4.5 and RCP 8.5 scenarios, which showed a decreasing trend for water and land footprints in both short-term and long-term future periods under RCP 4.5. In the short-term, water and land footprints decreased by 13.5% and 0.9%, respectively, and in the long-term, they decreased by 13.3% and 0.8%, respectively. Similarly, the water and land footprints also decreased under the RCP 8.5 scenario. In the short-term, the decreases were 13.0% and 1.1%, respectively, and in the long-term, 13.7% and 0.8%, respectively. Water and land footprints varied significantly depending on the GCMs applied, and the water footprint changed the most under RCP 8.5, with MIROC-SEM-CHEM for the long-term future, whereas the most changed land footprint was observed in the RCP 8.5 scenario with HadGEM2-CC for the long-term future.

4. Conclusions

This study applied SNAK, a platform for model user access. The final step in the nexus decision-making support platform of the SNAK system is to allow users to analyze the sustainability of resources and obtain graphics, trends, and comparisons via climate change scenario applications. To achieve this goal, the SNAK system was designed to assess trade-offs among water, energy, and food elements in the nexus; analyze complex factors using a system dynamics approach; visualize interlinkages across diverse factors related to water, energy, food, and climate; and determine how changes in individual factors influence the entire nexus system. The SNAK platform is crafted as a system that employs decision-makers scenarios grounded in modeling for water, energy, and food resources. It encompasses a resource modeling and interconnection analysis system that facilitates the simultaneous application and assessment of user scenarios and climate change scenarios from an integrated perspective. However, limitations arise due to the inherent uncertainty of climate change scenarios, given the application of climate change data in these scenarios. Additionally, the application of SNAK to specific regions may constrain its applicability to other areas. Nevertheless, as the proposed SNAK is constructed on a platform, integration with the platform is considered feasible, provided input data for the study area can be generated. In particular, the demonstration section of the SNAK system allows the input of environmental variables based on hypothetical information of scenarios to assess and provide details of water, energy, food, carbon emissions, and economic benefits and enables a review of changes in yearly trends in individual categories under a climate scenario. Thus, the SNAK system identifies the effects of resource management policies on individual resource consumption, carbon emissions, and the economy and allows users to use sustainability indices to consider external factors such as climate and population changes in assessing water, energy, and food policy scenarios.