Impacts of Energy Transition on Life Cycle Carbon Emission and Water Consumption in Japan’s Electric Sector

Abstract

The United Nations (UN) proposed the Sustainable Development Goals (SDG 17) in 2015, together with The Paris Agreement endorsed by 195 signatories and state parties, to address sustainable development-related issues through ambitious and dynamic actions. The transition of the energy system is at the heart of greenhouse gas (GHGs) mitigation which is required to achieve those goals and the electric sector is the core of energy system of interest while the GHG with the largest contribution to rising temperature is carbon dioxide. However, in addition to being centrally relevant for carbon emissions, the electric sector is also an important water consumer. This study applied a hybrid life cycle assessment (LCA) model with the disaggregated electric sector to investigate the impacts on carbon emission and water consumption of the energy transition in Japan under the Sixth Strategic Energy Plan. The results indicate that the electricity mix under the Nationally Determined Contribution (NDC) scenario can cut 50% of existing carbon emissions while intensifying the water consumption by 36% from the life cycle perspective in which 30% are foreign water footprints. The Kaya identity analysis confirmed this conclusion and explained the impacts of four driving factors (population, economy, electricity intensity, and electricity mix) qualitatively and quantitatively showing that the development of technologies and continuous efforts in energy saving can provide a substantial contribution to sustainable development. The results confirmed that the efforts proposed by Japan’s NDC for emission reduction through an energy transition in the electric sector can meet the expectation of achieving the Paris Agreement goals but will also pose greater challenges to the future global water demand in the energy system and regional water stress.

1. Introduction

Climate change is one of the most serious challenges facing human society today [1]. The global average temperature of 2020 is 1.2 °C above pre-industrial levels (1850–1900 average), making it one of the three warmest years since complete meteorological observations were recorded and 2011–2020 represents the warmest decade since 1850. In 2020, the average Asian land surface temperature was 1.06 °C above the normal value, making it the warmest year since the beginning of the 20th century [2]. The 2015 Paris Agreement endorsed by 195 signatories and state parties aims to prevent the increase in global average temperature reaching 2 °C (ideally 1.5 °C) above pre-industrial levels and energy transition is one of the most significant parts to effectively fulfill the requirements of the Paris Agreement and achieve the goal of carbon neutrality [3,4]. The implementation of carbon emission reduction policies to address climate change will lead to comprehensive impacts from multiple dimensions. Since energy transition is the core of the global response to climate change, and low-carbon technologies and renewable energy promoted by climate change policies have different initial inputs and needs than the traditional energy systems dominated by fossil energy, their large-scale promotion will bring diversified impacts on other systems such as resources, the environment, economy, and society among which water resources are also greatly affected.

Water systems, energy systems, and climate change are closely linked [5]. On the one hand, the massive emissions from the energy system directly contribute to global climate change which in turn has adverse effects on the energy system [6,7]. In recent years, extremely high temperatures, as well as extreme precipitation, and snow disasters have occurred more frequently because of climate change, which places higher demands on the security and stability of energy systems [8,9]. In 2008, the snow disaster in South China paralyzed power systems in more than 20 provinces [10]. Pašičko [7] found that wind power will increase with decreasing hydropower due to climate change by 2040 in Croatia. Craig [11] reviewed climate change impacts on the planning and operation of the bulk power system in the United States due to higher power demands and declining transmission capacity. On the other hand, both water system, and energy systems play an inevitable role throughout the production and operation of each other [12]. The energy sector is the second largest water consumer after agriculture, with the global energy sector accounting for 15% of total water consumption each year, and in some countries (e.g., the United States) its share is up to 45% [13]. Therefore, the energy supply is constrained by water resources, and in some areas may be affected by water scarcity [12,14]. With the future of energy transition and an increasing energy demand, research has shown that the water consumption of the global energy system is thought to rapidly increase [15,16]. Mekonnen [17] conducted a global water footprint assessment of electricity and found that biomass, hydropower, and nuclear are the top three water consumers in electricity generation and have a much higher water intensity than that of fossil fuels, while solar and wind do not demand much water. The policy scenario of IEA [18] suggested that the share of non-fossil energy in global electricity generation reached almost 29% in 2020 and will increase to 60% by 2030 to meet Net Zero Emissions by 2050. The total electricity generated by biomass, hydropower and nuclear sources is about the same as the sum of wind and solar. Even though the water intensity of wind and solar remains low, it cannot compensate for the exorbitant demand created by biomass and hydropower. In addition, the production, transportation, and use of water resources also requires the support of an energy system. The annual energy demand of the water system accounts for about 3% of the world’s total primary energy consumption [19]. Studies have shown that with the large-scale application of cutting-edge water treatment technologies, the energy consumption of global water systems will double in 25 years [20].

Furthermore, the uneven distribution of global water resources in total and geographically has brought certain constraints to the implementation of carbon emission reduction policies. Only 2.5% of total global water, excluding the storage capacity of oceans and other brackish water bodies, is freshwater among which, excluding frozen water in glaciers and ice caps and groundwater, only about 1.2% of surface water can be directly used [21]. However, river water, which accounts for 0.0002% of the global water volume, is the most important water resource supporting most of the water demand for human development. In terms of geographical distribution, most of the world’s water resources are distributed in the Americas, accounting for 45% of the total; followed by Asia, 28%; Europe, 15.5%; and Africa, which accounts for only 9% [22]. According to research by the World Resource Institute (WRI) [23], among the top 10 countries in terms of total water resources in 2013, India and Indonesia faced the highest water stress (40–80%), while the United States and China also have moderate to high water stress (20–40%). Therefore, the virtual water embedded in the global trading system, for example, the tropical fruits purchased in Japan which actually involves importing virtual water from countries mainly in the South, is widely discussed to better understand and address the issue of an uneven distribution of water. Virtual water trade accounts for 22–30% of total global water use of which 32% is scarce water [24,25]. Lenzen [24] used an input–output model to simulate the global virtual flow and found that developed country are importing virtual water at an increasing rate to lessen their water stress; however, this could also intensify the uneven distribution of global water resources.

Japan, one of the top energy consumers and GHGs emitters, has announced its sixth strategic energy plan [26] with an ambitious energy mix outlook to cut approximately 45% of energy related CO₂ emission, as well as 46% of total GHGs emission compared with 2013 (25% and 26%, respectively, as former plan) and to raise energy self-sufficiency to reach approximately 30% (25% as former plan) by 2030. Figure 1 explains the detailed electricity demand and mix in 2030 compared with the former plan, which is also treated as the NDC scenario of Japan in this research. With strenuous efforts in energy saving, the total electricity demand in 2030 is considered to have a 230 TWh cut, which is 20% more than the former plan. The ratio of fossil energy in the new energy mix decreases by 41% while renewables account for 36–38%. Besides, 1% hydrogen and Ammonia is projected, which is the first time it has appeared in the energy plans. Besides, the electricity loss is also expected to decrease from 85 TWh to 60–70 TWh, which suggests that the efficiency of electricity transmission and distribution will further increase. There is not yet sufficient research regarding the environmental impacts brought by the energy transition of the whole electric sector. Hienuki calculated the impacts on carbon emission, GDP, and employment of Japan’s future electricity mix based on the input–output table of 2005 and found that emissions will be reduced by 8%, 16% and 16% in 2020 and 29%, 38% and 44% in 2030 for each respective scenario compared with that in 2012 [27]. Although the Japanese government has proposed such a strategic plan to show their determination to achieve the carbon neutral goal through continuous energy transition, the life cycle carbon emissions and water consumption, especially of different energy sources within Japan’s electric sector with regard to the energy transition have not yet been well understood. It is of great importance to understand this key information through the energy transition process so that policymakers can propose and implement comprehensive and visionary plans to better adapt to the further impact from various dimensions brought by climate change.

The development of the footprint theory provides a framework for describing the resource consumption and pollutant emissions directly and indirectly driven by the entire life cycle of specific products and services or technologies and industries [28]. As a resource-poor island country, Japan’s energy self-sufficiency rate slumped drastically after the Great East Japan Earthquake, from 20.3% in 2010 to only 7.4% in 2015. Although the government has exerted unremitting efforts in restarting nuclear power plants and the promoting renewable energy, the self-sufficiency rate reached 12% in 2019, meaning Japan still ranks second from the bottom among all OECD countries. Therefore, there is no denying that the foreign carbon and water footprints of Japan’s electric sector are also nonnegligible in the study of the environmental impacts of energy transition. A life cycle assessment (LCA) is a widely used method in studies related to footprint analysis. Based on the different perspectives of products and industries, the specific implementation of LCA can be divided into the following two categories: Process Analysis (PA), a bottom-up approach, and Environmentally Extended Input-Output Analysis (EEIOA), a top-down approach [29,30]. The electricity mix transition will exert different impacts on carbon emissions and water consumption due to various demand from different energy sources and linkages with other sectors. The PA approach is a standard bottom-up LCA method, which starts from the direct environmental impact of the most basic technical unit in the production process, such as resource input and pollutant discharge, and considers all environmental impacts in the multi-stage production process of the product [31]. However, such artificially defined boundaries can cause significant truncation errors, and their application to the entire economy among supply chains is also a tedious task that is almost impossible to complete. The EEIOA approach can fully examine the impacts of energy transition under the interaction of all industrial chains, however, because of high integration in the sectoral classification of the EEIOA model, it is difficult to identify the impacts brought by each energy source within the electric sectors, and to assess the impacts brought by structural changes in subsectors [32,33]. Combining the detailed description in the PA approach with the full sector-wide description in the EEIOA approach as a hybrid model will effectively take advantage of the strengths of both models. The hybrid model can integrate the micro-technological process and macro-sectoral linkages, and thus can more accurately analyze the life cycle impact brought by the industry [34]. In this research, we will construct a hybrid model to assess the carbon emission and water consumption of the life cycle of the electric sector in Japan quantitively under the official NDC scenario. Results could provide reference for government policy maker or practitioner in the electric sector to better understand the environmental impacts brought by energy transition.

2. Materials and Methods

2.1. Data Description

2.1.1. Input-Output Data

This research uses the official Japan input-output table of 2015 which is a 37-sector table. The target electric sector is an aggregated sector combining gas and heat supply and thus sectoral disaggregation is needed. The method proposed by Lindner, (2013) [35] is applied and a reallocated 46-sector classification can be found in Table S1. The gas and heat supply sector remains while the electric sector is divided into nine sub-sectors as shown in

Table 1. Disaggregation of the original sector.

Original Sector	First Disaggregation	Final Disaggregation
Electricity, gas, and heat supply	Electricity	Coal
		Oil
		Gas
		Bio
		Nuclear
		Hydro
		Geo
		Solar
		Wind
	Gas and heat supply	Gas and heat supply

.

2.1.2. Direct Carbon Emission and Water Consumption Data

In this research, the sectoral direct carbon emission data of 37 non-electric sectors were calculated by the work of the Embodied Energy and Emission Intensity Data (3EID), National Institute for Environmental Studies, Japan [36,37] which has been widely used in various studies [38,39,40,41]. The 390-fundamental-sector data were reallocated to fit our classification. The sectoral direct water consumption data of non-electric sector were calculated by the work of Ono [42]. Despite the water consumption data being based on the 2005 input-output table, the result obtained from the direct water consumption intensity in 2005 with the total sectoral output in 2015 did not show a significant bias from the official total water consumption data and thus the data are considered applicable to this research. All data were pre-processed to align with the disaggregated 2015 input-output table classification. The direct carbon emissions and water consumption data of different energy sources within the electric sector were calculated with the intensity data from the report of the Central Research Institute of Electric Power Industry, Japan [43], and other research data [17,44,45], which are collected in Table S2. Note that the boundary of direct carbon emission and water consumption from the electric sector is that carbon emitted by direct combustion of fossil fuel to generate electricity (direct emission from bio is considered as net zero since it is offset by the carbon it absorbed) and the amount of water deprived from, but not returned to, the same drainage basin during the power generation process such as water used for cooling and evaporated from the dam [42].

2.1.3. Scenario Data

Total electricity demand under NDC scenario in 2030 from different energy sources was calculated with the data proposed in Japan Sixth Strategic Energy Plan [26]. The portion of Hydrogen and Ammonia were assigned to other renewable energies since they are not considered in this research. Population, GDP, and other related data under NDC scenario were also calculated based on the reported data.

2.2. Hybrid LCA Model

We applied the model framework of Wan [46] and optimized sector classification as well as the disaggregation coefficient for Japan’s circumstances which are listed in Table S3. The disaggregation in upstream is divided into three categories based on the different demand of each of the power generation technologies. The sectors that provide the electric sector with the fuels to generate various types of electricity are defined as fuel-related sectors which are further divided into two categories. The fuel-related_A sectors’ input to electric sector all goes to bio while the fuel-related_B sectors’ input goes to fossil fuels by a ratio that is based on fuel cost per unit of electricity generated by each source. Another is defined as capital-related sectors which provide investment in machinery equipment and plant construction, etc. The disaggregation of capital-related sectors is based on the overnight investment cost of each power generation technology. The remaining upstream sectors are defined as other sectors, which generally only have an indirect input to the electric sector or whose inputs to the electric sector do not vary significantly by different power generation technology, such as food manufacturing, non-metal, and metal products, and other third industries (commerce, medical service, etc.). They are disaggregated with reference to the operation and maintenance cost (O&M cost) of each technology and the share of generation capacity of each power generation technology. The downstream disaggregation is relatively easier since they can be determined by the power generation share of each technology.

Take the disaggregation of two sectors as an example. Assume that n is the original sector (electric sector) and a_i,n is the direct consumption coefficient from the original sector n to upstream sector i; a_n,j is the direct consumption coefficient from sector j to the original sector n. Therefore, we have A_i,n and A_i,n+1 which are the direct consumption coefficient from the new disaggregated sectors n and n + 1 to upstream sector i and so are A_n,j and A_n+1,j. a_n,n is the direct consumption coefficient within the original sector and s and 1−s are the share of the two disaggregated sector in the original sector.

Disaggregation in upstream:

a_i,n = (1 − s)·A_i,n + s·A_i,n+1

(1)

Disaggregation in downstream:

a_n,j = A_n,j + A_n,j+1

(2)

Disaggregation within the original sector:

a_n,n = (1 − s)·(A_n,n + A_n+1,n) + s·(A_n,n+1 + A_n+1,n+1)

(3)

The import rate of each power generation technology in 2015, as listed in

Table 2. Import rate of each power generation technology.

Coal	Oil	Gas	Bio	Nuclear	Hydro	Geo	Solar	Wind
99.3%	99.7%	97.5%	67.2%	100%	0%	0%	63.3%	81.1%

, is calculated by government reports. Some import components for the energy systems are processed individually in the model and thus not included in the table. Since this paper only considered the impacts brought by electricity mix change, other variables in 2030 such as total input and output and import rate are considered the same as base year.

Table 3. Environmentally Extended Input-Output Table.

	Intermediate Demand	Final Demand			Total Output
		Domestic Final Demand	Export	Import
Intermediate Input	T	Y	EX	−IM	X
Value Added	V
Total Input	X
CO2 Emission	C
Water Consumption	W

depicts a basic structure of a typical EEIO table. Based on the method proposed by Hienuki [47], the direct and indirect carbon emission as well as water consumption of different energy sources can be calculated with the following equations:

M is the import coefficient vector consist of $m_j$,

$$m_j=\frac{\mathit{I}{\mathit{M}}_j}{{\mathit{T}}_j+{\mathit{Y}}_j+\mathit{E}{\mathit{X}}_j} \left(j=1, 2, \dots , n\right)$$

(4)

where $\mathit{I}{\mathit{M}}_j$, ${\mathit{T}}_j$, ${\mathit{Y}}_j$, $\mathit{E}{\mathit{X}}_j$ are the import, intermediate demand, domestic final demand, and export of sector j, respectively. The domestic impact ${\mathit{X}}_d$ is:

$${\mathit{X}}_d={\left[\mathit{I}-\left(\mathit{I}-\widehat{\mathit{M}}\right)\mathit{A}\right]}^{-1}\left[\left(\mathit{I}-\widehat{\mathit{M}}\right)\mathit{Y}\right],$$

(5)

where $\widehat{\mathit{M}}$ is the diagonal matrix of M, $\mathit{A}$ is the input-output coefficient vector.

The foreign indirect impact ${\mathit{X}}_f$ is:

$${\mathit{X}}_{\mathit{f}}=\mathit{X}-{\mathit{X}}_{\mathit{d}}={\left(\mathit{I}-\mathit{A}\right)}^{-\mathbf{1}} \mathit{Y}-{\left[\mathit{I}-\left(\mathit{I}-\widehat{\mathit{M}}\right)\mathit{A}\right]}^{-\mathbf{1}}\left[\left(\mathit{I}-\widehat{\mathit{M}}\right)\mathit{Y}\right]$$

(6)

Similarly, E is the environmental impact coefficient vector consist of $e_j$:

$$e_j={\mathit{E}}_j/{\mathit{X}}_j\left(j=1, 2, \dots , n\right)$$

(7)

where ${\mathit{E}}_j$ and ${\mathit{X}}_j$ are the direct environmental impact and total output of sector j, respectively, the total environmental impact is:

$$\mathit{E}=\widehat{\mathit{E}}{\left(\mathit{I}-\mathit{A}\right)}^{-\mathbf{1}}\left(\mathit{Y}+\mathit{E}\mathit{X}+\mathit{I}\mathit{M}\right)$$

(8)

The domestic environmental impact is:

$${\mathit{E}}_d=\widehat{\mathit{E}}{\left[\mathit{I}-\left(\mathit{I}-\widehat{\mathit{M}}\right)\mathit{A}\right]}^{-1}\left[\left(\mathit{I}-\widehat{\mathit{M}}\right)\mathit{Y}\right]$$

(9)

the foreign indirect impact ${\mathit{E}}_f$ is:

$${\mathit{E}}_{\mathit{f}}=\mathit{E}-{\mathit{E}}_{\mathit{d}}=\widehat{\mathit{E}}{\left(\mathit{I}-\mathit{A}\right)}^{-\mathbf{1}}\left(\mathit{Y}+\mathit{E}\mathit{X}+\mathit{I}\mathit{M}\right)-\widehat{\mathit{E}}{\left[\mathit{I}-\left(\mathit{I}-\widehat{\mathit{M}}\right)\mathit{A}\right]}^{-1}\left[\left(\mathit{I}-\widehat{\mathit{M}}\right)\mathit{Y}\right]$$

(10)

2.3. Kaya Identity

Kaya identity is a widely used method for analyzing the driving factors of carbon emissions, which was proposed by Yoichi Kaya in 1989 [48]. The original Kaya Identity is used to analyze the total energy system while in this study, we focus on the electric sector and thus some changes are made as Equation (8)

$${\mathit{E}}_{\mathit{e}}=\mathit{P}\times \frac{\mathit{G}\mathit{D}\mathit{P}}{\mathit{P}}\times \frac{\mathit{T}\mathit{E}\mathit{C}}{\mathit{G}\mathit{D}\mathit{P}}\times \frac{{\mathit{E}}_{\mathit{e}}}{\mathit{T}\mathit{E}\mathit{C}}=\mathit{p}\times \mathit{g}\times \mathit{e}\times \mathit{f}$$

(11)

Where ${\mathit{E}}_{\mathit{e}}$, P, $\mathit{G}\mathit{D}\mathit{P}$, and TEC represent the life cycle environmental impacts of electric sector, total population, Gross Domestic Product, and the total electricity consumption. The total population is defined by p = P; the GDP per capita is g = GDP/P; the electricity intensity of GDP is e = TEC/GDP and the total environmental footprint intensity of electricity is f = ${\mathit{E}}_{\mathit{e}}$/TPEC. Therefore, the total environmental impact can be divided into four parts:

$$∆\mathit{E}={\mathit{E}}_{\mathit{t}}-{\mathit{E}}_{\mathbf{0}}=∆{\mathit{E}}_{\mathit{p}}+∆{\mathit{E}}_{\mathit{g}}+∆{\mathit{E}}_{\mathit{e}}+∆{\mathit{E}}_{\mathit{f}}$$

(12)

where $∆{\mathit{E}}_{\mathit{p}}$, $∆{\mathit{E}}_{\mathit{g}}$, $∆{\mathit{E}}_{\mathit{e}}$, $∆{\mathit{E}}_{\mathit{f}}$ are changes brought by population effect, economy effect, electricity intensity effect, and electricity mix effect, respectively. GDP and population under the NDC scenario are calculated with the official plan. Carbon emission and water consumption data in 2015 and 2030 are derived from the results in Section 3.2.

3. Results

3.1. Electricity Generation

After the Great East Japan Earthquake in 2011, nuclear power use declined drastically and the dependance on fossil fuel increased significantly. Figure 2 shows the detailed electricity mix change of the base year 2015 and under the NDC scenario by 2030. The total electricity in 2015 was 1037.7 TWh which ranked fifth in the world among which over 83% of electricity was generated from fossil fuel. Hydropower contributed the largest share among all renewable energies. Under the NDC scenario, total electricity generation is expected to drop to 937.8 TWh, optimistically. The overall share of fossil fuels will decrease to 41.5% percent while nuclear power shows the greatest increase to 200TWh. Solar and wind power will also become major sources in the NDC mix that show a growth by 200% and 800%, respectively, while hydropower does not show significant change. Detailed electricity mix data can be found in Table S4.

3.2. Carbon Emission and Water Consumption

Figure 3 depicts the life cycle carbon emission and water consumption of Japan’s electric sector in 2015 and 2030 by the Japanese government. In the base year, total carbon emission was about 630 Mt where fossil energy produced over 98% of all energy. Nonetheless, given that Japan has most developed technologies and equipment in the fossil power industry, the emission intensity of fossil fuel is much lower compared to that of other countries, especially developing countries. Bio emission came first after fossil fuel, producing 5.2 Mt, and all non-fossil emission was about 10.7 Mt. It is clear that the renewables did not account for enough of the capacity that nuclear power had driven after the Great East Japan Earthquake and thus the re-emergence of fossil power plants did lead to greater emissions due to the absence of nuclear power. Under the NDC scenario, the life cycle emission is expected to decline by roughly 50% to 313 Mt by 2030 among which coal power represents the highest decrease of over 170 Mt CO₂ emissions, followed by gas power of 108 Mt which will become the highest emission source. Oil changes the least in fossil fuels, which is 51 Mt since the remaining capacity of oil electricity is only 2%. Emissions from bio and nuclear will double while renewables nearly tripling, which is still no comparison with that of fossil fuel. The energy mix transition in the electric sector is expected to have a significant contribution on the emission reductions under the NDC scenario and will play a vital role in achieving the future carbon neutral goal.

The total life cycle water consumption of Japan’s electric sector will increase by about 36% to nearly 7300 Mt under the NDC scenario in 2030. Unlike carbon emissions, bio consumes the most water and will increase by 87% to over 4300 Mt due to a large demand throughout the growth of biomass. The shift from fossil energy will save up to 700 Mt of water while with the share of nuclear and renewables will continue growing, water demand for the construction of the power plant as well as and that of the production of renewable energy generation equipment will no longer be negligible. Nuclear is expected to consume 480 Mt of water, which ranks third after bio and hydro. Within other renewables, solar becomes the largest consumer by an increase of 300% to 60 Mt followed by geo’s 24 Mt. Wind power has the lowest water consumption intensity, requiring only 0.28 Mt.

As a resource-poor island country, almost all fossil fuels are imported in Japan and thus, the extraction, processing, and transportation of fossil fuels, which consume a large amount of water and produce objective carbon emissions are all borne by the exporting countries, the detailed domestic and foreign carbon and water footprints from nine energy sources were calculated as in Figure 4. On the one hand, the foreign carbon footprints share of all fossil fuels will increase with the significant fall in their capacity. Most non-fossil energies have a higher foreign carbon footprint than that of fossil fuels among which wind power reaches 77%. Bio and solar power will also have over 40% of a foreign carbon footprint between 2015 and 2030. Hydro and geo power will both increase by about 5% and 10%, respectively. Nuclear power shows the greatest decline from 59% to 40% due to the re-emergence of nuclear power plants. On the other hand, foreign water footprint share does not show significant change in fossil fuels among which natural gas becomes the lowest because of less foreign water footprint demand during the extraction process compared with coal and oil. Bio will remain over 40% due to the self-sufficiency rate of wood stabilizing at a fairly low level. Solar power is expected to have the most significant reduction from 38% to 13% since the growing capacity will accordingly raise the water demand in operation and maintenance.

3.3. Kaya Identity Analysis

This research calculated the impacts of Japan’s electric sector on carbon emission and water consumption from 2015 to 2030 with the Kaya Identity method. The results are listed in

Table 4. Kaya Identity analysis of environmental impacts change.

		Unit: Mt
Effects	Carbon Emission	Water Consumption
Population Effect	−20.996	−270.517
Economy Effect	201.562	2596.984
Electricity Intensity Effect	−251.947	−3246.155
Electricity Mix Effect	−240.408	2603.824
Total Change	−311.789	1684.135

. It is expected to have a total 311 Mt cut in emissions and to increase water consumption by 1684 Mt of water consumption. The population decline results in slight positive impacts on emission mitigation and water saving due to a decreased total electricity demand while continuous economic development contributes to the growth of both environmental factors. The electricity intensity effect is the most significant driver which means that the development in technology and in the industry structure plays a crucial role in the energy transition process and can make satisfactory contributions. Changes by the electricity mix effect indicate that a shift away from fossil energy can mitigate carbon emissions while intensifying the water consumption of the electric sector which, is consistent with the results that we obtained in Section 3.2 that some non-fossil energies have higher life cycle water consumption intensity than that of fossil fuels. The result well explains the impacts brought by each driving factor qualitatively and quantitatively.

4. Discussion

This research analyzed the impacts of an energy transition on the domestic and foreign life cycle carbon emissions and water consumption of different energy source in Japan’s electric sector by applying a hybrid LCA model based on EEIOA and Kaya Identity under the projected electricity mix of the latest Japan NDC scenario. The results showed that the shift away from fossil energy can greatly reduce total carbon emissions while leading to a considerable growth in total water consumption since the electricity mix effect mitigates emissions but also intensifies water consumption. The electricity intensity effect has the most significant impact on cutting carbon emissions and water consumption in the electric sector, which is one of the results of the ambition for continuous energy saving in the Sixth Strategic Energy Plan. Other research has also confirmed that improved energy intensity can contribute to considerable carbon emission reduction [49,50] and “In Japan, the energy efficiency of the energy conversion is assumed to be improved relatively well” [51]. The development of technologies and continuous efforts in energy saving can provide a substantial contribution to sustainable development.

With most focus having been placed on carbon emissions, the water consumption of the whole energy system can also no longer be neglected. This research calculated a water demand increase of 36% in the electric sector and the total water consumption intensity. Although Japan is an island country with rich water resources, the water footprint of its international resources trade does intensify the unequal distribution of water resources, especially given that some of the oil-rich Middle Eastern countries are also among the world’s most water-scarce countries. Besides, biomass power also requires a great deal of firewood and wood waste from construction which largely depend on imports. According to the Ministry of Agriculture, Forestry and Fisheries (MAFF), the import rate of wood pellets increased drastically from 42.3% in 2012 to 91.6% in 2019 and the top two importing countries are Vietnam and Canada that together constitute over 83% of total imports. The massive water demand during the growth of wood has also resulted in the emerging water crisis among these wood export countries to some extent. Forests do not require watering, however, the growth of wood would consume a massive water through evapotranspiration which is a major green water flux (used by vegetation), while river discharge and groundwater are typical blue water fluxes (used by human). Tree growth can consume more water than other shorter vegetation [52,53]. According to the mass balance principle, if more water is used by trees (such as demand of biomass in this research), less water will flow into rivers and lakes or recharge the groundwater so that people can directly use water. Although the impacts of the low-carbon transition of the electric sector on water consumption is not as significant on the whole economy as it is for carbon emissions, with the share of electricity in the energy system continuing to increase and renewable energy technologies being promoted, the potential for elevating water consumption in the future cannot be ignored.

Furthermore, after the Fukushima nuclear disaster, the public trust and acceptance of nuclear energy also exacerbated the difficulty in restarting nuclear power plants. On August 11, 2015, the Sendai nuclear power plant in Kyushu resumed after the Atomic Energy Commission’s compliance review, which was the first nuclear power plant to be restarted, ending Japan’s nearly two-year “zero nuclear power” period. As anti-nuclear voices and forces grow stronger in Japan, nuclear restarts will also face additional legal risks. Due to different meteorological conditions and topography, the cost of renewable energy in Japan is much higher than the world average and the uncertainty of renewable power makes it hard to totally replace the traditional power generation technologies. However, research has illustrated the potential of renewable energies in Japan and 100% renewables could be possible [54,55,56]. The uncertainty of renewable energy may be greatly alleviated by the development of electricity storage technology, whose levelized cost has decreased and makes it possible to proliferates from 2030 [57,58]. Moreover, other forms such as pumped-storage power station can also act as an adjuster to make full use of the power [59,60]. Besides, Japan has planned a 1% share of Hydrogen and Ammonia in the NDC electricity mix. Even though there are still limitations on Ammonia as a Hydrogen carrier for transportation and storage such as evaporation and loss of energy content [61], researchers are still working on the optimization of this technology to prepare it for the market [62,63]. As one of the most active countries to promote hydrogen-powered vehicles, Japan is exploring the possibility of renewable energy sources in multi-energy systems. Nonetheless, if the Japanese government becomes overdependent on nuclear power, the possible return of fossil energy will be ever-present if nuclear power and renewable energy cannot meet the expected capacity in the near future, which could make it counterproductive.

Existing studies have suggested that the actions from 2020 to 2030 are somewhat inadequate for achieving long-term temperature control goals. More aggressive emissions reduction actions and ambitions are significantly important to successfully achieve the Paris Agreement goal [64,65]. Energy transition is surely an indispensable and important measure to face the challenge of climate change, however, current efforts in dealing with climate change do not show a satisfactory result and an optimistic future. Therefore, the low-carbon transition of the electric sector including the promotion of renewable energy in the future still has a large scope for development, which will also pose greater challenges to the future global water demand in energy systems and regional water stress. Finally, due to data limitation, we cannot construct a multi-region Input–Output (MRIO) model to assess the flow with other countries and thus the export part is included in the domestic sector, and it was not within the remit of this research to assess domestic exports.