Consumption‐Based Carbon Emissions of 85 Federal Entities in Russia

As the fourth largest CO2 emitter, Russia's constituent entities collectively contribute with vast territory and regional heterogeneity. Existing studies only present production‐based inventories; state‐level consumption‐based emissions patterns and driving forces remain rare. Here, we built the Russia state‐level Multi‐regional input‐output table to ascertain heterogeneity in consumption‐based emissions and track carbon flows in the inter‐state. We found that 60% of consumption‐based emissions coming from affluent areas (top 20% of the GRP), including Moscow (139.1 Mt) and St. Petersburg (50.5 Mt). Energy‐intensive regions also had huge consumption‐based emissions (31.6 Mt in Khanty‐Mansi Autonomous Okrug and 29.4 Mt in Republic of Tatarstan). Household consumption emissions accounted for 41%–73% of consumption‐based emissions in the 83 regions, except for Tula and Lipetsk, where fixed capital formation dominated. In addition, the major contributor of embodied emissions in households were power and services sectors, which contributed about 8%–61% (0.03 Mt–12.7 Mt) and 12%–40% (0.1 Mt–20.1 Mt). In Russia's low carbon transition, policymakers should not only focus on a local mitigation policy in developed states (such as Moscow and St. Petersburg), but on key sectors to curb the consumption. Regions with high carbon intensity should switch to renewable energy and implement cleaner production techniques in high‐emission industries.


Introduction
Russia is the fourth largest CO 2 emitter in the world (Rüstemoğlu & Andrés, 2016).It revealed its Nationally Determined Contributions to meet its emissions mitigation goals: reducing greenhouse gas emissions by 70% from 1990 levels by 2030 and achieving carbon neutrality by 2060 (UNFCCC, 2022).To effectively meet mitigation targets, it is necessary to formulate emission mitigation policies tailored to state-level conditions (Gopalakrishnan et al., 2021;Tian et al., 2022Tian et al., , 2023)).Russia is a vast country comprised of 85 constituent entities in terms of development stages and population.It implies that there is a significant heterogeneity in emission characteristics and mitigation strategies.However, the existing emission inventories only measure CO 2 emissions at national level (Korppoo & Kokorin, 2017;Yang et al., 2017) or at production-based (Xiao et al., 2021), with consumption-based CO 2 emissions data missing.Therefore, it is urgent to build a Russian state-level consumption-based emissions inventory to formulate more tailor-targeted mitigation policies.
There are two methodologies for quantifying carbon emissions, including production-based and consumptionbased accounting (Zhang et al., 2016).Production-based emissions are based on territorial production, which ignores where goods are consumed or who ultimately consumes them (Atkinson et al., 2011;Peters, 2008).Nonetheless, it is vital to note that significant trade flows occur between regions to meet development needs, resulting in a transfer of embodied emissions between regions (Feng et al., 2013;Guan & Reiner, 2009;Su & Ang, 2014).However, while considering the expansion of trade, it is vital to note that the direct emissions accounting, which is based on the responsibility of producers, fails to consider the impact of trade (Meng et al., 2016).This may trigger carbon leakage problems and undermine overall regional emission mitigation efforts.To guarantee complete and equitable carbon mitigation measures, it is essential to consider both the direct emissions as well as embodied emissions (Guan et al., 2014;Jakob et al., 2014;Long, Jiang, et al., 2021, Long, Guan, et al., 2021;Steininger et al., 2014).Many scholars begin to quantify outsourced carbon emissions embodied in trade led by the domestic consumption (Nansai et al., 2009;Wood & Dey, 2009).Hertwich and Peters (2009) calculated greenhouse gas emissions from a consumption perspective in 73 countries and 14 global regions.They found that 72% of global greenhouse gas emissions were dominated by household consumption, 10% by government consumption, and 18% by investments.Davis and Caldeira (2010) have established carbon inventories for 57 sectors in 113 countries and territories from a consumer perspective.The results indicated that consumption-based emissions demonstrated the potential for international carbon leakage.Feng et al. (2014) also calculated the consumption-based emissions of four Chinese megacities and found that addressing consumption patterns is crucial to China's low-carbon development.Barrett et al. (2013) and Wiedmann et al. (2010) both calculated consumption-based emissions and found that there was a necessity for consumption-based emissions to be used as a complementary indicator for measuring geographical emissions.A range of studies have demonstrated that consumption-based emissions accounting provides a solid basis for climate change solutions (Girod et al., 2014;Long et al., 2018;Peters & Hertwich, 2008;Steininger et al., 2016).
Nevertheless, previous studies of consumption-based carbon emissions in Russia have focused on the national level, failing to capture the heterogeneity among constituent entities.Sun et al. (2019) used the input-output method to analyze the carbon emissions embodied in Russian international trade.They discovered that Russia was a net exporter of carbon emissions in 2013, with 10.7% of Russia's emissions generated for the consumption of other countries.Another study compared the carbon emissions characteristics of 14 major economies, including Russia, from the consumption and production side (Fan et al., 2016).It demonstrated that allocating responsibility solely on the basis of production-based emissions was unjust to countries like Russia and China.At the regional level, single research has been conducted to quantify the carbon emissions at the regional level in Russia.Xiao et al. (2021) quantified the carbon emissions at the sectoral and regional level in Russia from production-based perspective, while the consumption-based emissions remained limited.
In this study, we build up the first Russia state-level Multi-regional input-output analysis (MRIO) table to estimate the consumption-based emissions of 85 Russian Federation entities.Then we analyze the main driving factors of emissions change at regional scales.In addition, this study also shows the embodied emissions transfer paths among federal entities.And it clarifies the different positions of constituent entities in the embodied carbon emissions transfer network.Our findings will aid in the creation of targeted carbon mitigation strategies in Russian.

Data Source
This study used the 2016 Russian state-level carbon emissions inventory from the China Emissions Account and Data sets (Xiao et al., 2021).The inventory contains data on carbon emissions from 89 socio-economic sectors in 82 constituent entities of Russia.We use an entropy-based approach to create the Russian MRIO (Multi-Regional Input-Output) table using the available data for the year 2016 (Table 1).The construction of the Russian MRIO table involved state-level trade data, regional Gross Domestic Product (GDP) data, the Russian national Inputoutput table, and sector-specific manufacturing value-added data.Among them, the trade data is state-wise with HS6 code product which was obtained from the Observatory of Economic Complexity (OEC).The rest of data was sourced from the Russian Federal State Statistics Service.

10.1029/2023EF004323
Since the regions and sectors of the carbon inventory and the MRIO table do not match exactly, but in order to ensure consistency of data quality, the carbon inventory was processed as follows: we divided the 82 regions in the list into 85 regions, matched 89 sectors to 59 sectors in the MRIO table.

Construction of MRIO Table
The construction of Russia's multi-regional input-output table can be divided into five steps: (1) Estimate the domestic supply and domestic demand of each region.Domestic supply equals total supply minus foreign exports.Similarly, domestic demand equals total demand minus foreign imports, where total demand is the function of intermediate demand and final demand.(2) Decompose the domestic supply and domestic demand of each region.In this study, domestic supply is decomposed into self-supply and supply to other provinces, while domestic demand is decomposed into locally supplied demand and demand from other provinces.Additionally, to ensure that total domestic exports equal total domestic imports for any given product, a cross-entropy model addresses inconsistencies in trade flows between regions.The cross-entropy model ensures maximum similarity between the target and known distributions.(3) Adjust the regional single region input-output (SRIO) table.Since Step 2 realigns domestic supply and domestic demand, this study uses the generalized RAS (GRAS) model to update total demand across regions and calibrate it to adjusted domestic exports and imports.The matrix balancing of the SRIO tables for each region must meet two conditions.In row terms, the row sum of total demand should be equal to gross output minus net exports, while in column terms, the column sum of intermediate demand should be equal to gross output minus value added.(4) Estimate interregional trade matrix.To obtain the trade matrix, we combine the gravity model with trade data between regions to improve the data's accuracy and reliability.However, the initial trade matrix does not conform to the row and column constraints of the updated state SRIO tables.As a result, we use the RAS model to balance the trade matrix so that it is consistent with the state SRIO tables.(5) Link the adjusted state-level SRIO table with the trade matrix.According to the above steps, the diagonal matrix generates intermediate and final demands, whereas the non-diagonal matrix generates intermediate and final demands.We splice the intermediate and final demands from the diagonal and non-diagonal matrices separately to form the state-level MRIO table.The methodology employed in this study aims to construct a multiregional input-output table.It has been successfully implemented in the compilation of regional and city-level MRIO tables in China.Further details regarding this methodology can be found in our previous works (Zheng et al., 2021;Zheng, Long, et al., 2022;Zheng, Többen, et al., 2022).

Consumption-Based Emissions Based on the Multi-Regional Input-Output Method
Input-output (IO) analysis is a top-down method for calculating consumption-based emissions.Many studies use the SRIO method to estimate carbon emissions (Long, Jiang, et al., 2021, Long, Guan, et al., 2021), but this model assumes that production technology is homogeneous and may bring some errors (Wiedmann, 2009).Multi- Earth's Future 10.1029/2023EF004323 regional input-output analysis (MRIO) can not only distinguish regional differences in production efficiency (Lin et al., 2017;Yu et al., 2010), but provide spatial linkages between industries (Blair & Miller, 1983;Jiang et al., 2019).In addition, the MRIO model can capture the relationships between direct and indirect emissions at sectoral level (Jiang et al., 2022).It can also overcome the problem of duplication or omission in carbon emissions due to sectoral complexity (Södersten et al., 2018).Therefore, more and more studies use multi-regional inputoutput (MRIO) models to calculate carbon footprints (Mi et al., 2016(Mi et al., , 2017;;Liu et al., 2016;Ou et al., 2019;Zheng, Wood, et al., 2023;Zheng, Zhang, et al., 2023).
The MRIO model is widely used to calculate consumption-based emissions, which can track trade relationships among regions and sectors (Guan et al., 2014;Meng et al., 2017Meng et al., , 2018)).It is also an important tool to track spillover effects and identify regional heterogeneity through regionally dispersed supply chains (Long et al., 2019;Zheng et al., 2019).
In the MRIO model, stimulated by the final demand vector Y, the total output X of each industry in each region can be expressed by the following formula: where I is the identity matrix; where A is the technical coefficient which is calculated as a rs = z rs ij / x s j .To calculate the consumption-based emissions, we extend the MRIO model with environmental extension (Guan & Barker, 2012;Jiang et al., 2023;Zheng, Long, et al., 2022;Zheng, Többen, et al., 2022): Where C is the total carbon dioxide emissions.Where E is a vector of emission intensity for all industries in all regions which is calculated as e r i = e r i / x r i and e r i indicates the emissions emitted from sector i in region r.
Evidently, C is the summation of the carbon emissions of every sector in the whole area.To get more detailed and meaningful data, we can use the following formula: Where CE denotes the matrix, whose elements represent the emissions from one producer sector of a region to another sector of the same or different region.

Consumption-Based Emissions of Federal Entities in Russia
In 2016, Russia's 85 federal subjects' total consumption-based emissions were 983.5 Mt.The overall emissions of Russia's 10 regions with the highest consumption-based carbon emissions were 452.9 Mt, accounting for 46% of all consumption-based carbon emissions in the country.However, these 10 regions' population accounts for 31% of the national population, while their GDP contribution accounts for 47% of the country.The emissions of the remaining 75 federal entities account for 54% of total national emissions, which is just 8% more than the emissions of these 10 regions.This not only verifies that substantial emissions are emitted in only a few regions (Qian et al., 2022), but also demonstrates the unequal distribution of carbon emissions throughout the constituent entities of the Russian Federation.
The carbon footprints of the constituent entities of the Russian Federation vary considerably due to the gap in economic growth and population distribution.Total consumption-based emissions are highest in the more influential federal entities (Moscow, St. Petersburg, Moscow Region) and the more industrially developed federal entities (Republic of Tatarstan, Lipetsk Region), as indicated in Figure 1.The federal entities with the lowest consumption-based emissions, on the other hand, are the least economically developed and least populous, such as Chukotka Autonomous Okrug, Republic of Ingushetia, and Republic of Kalmykia.This disparity in carbon emissions reflects the disparity in wealth distribution and responsibilities for emission mitigation (Zheng, Wood, et al., 2023;Zheng, Zhang, et al., 2023)  Earth's Future 10.1029/2023EF004323 139.1 Mt, and the smallest being Chukotka Autonomous Okrug, which is 0.5 Mt, with a difference of more than 270 times.Many variables contribute to this large disparity, including economic growth, population, technology, industry structure, and purchasing preferences.As an illustration, Moscow's population is 247.6 times larger than that of the Chukotka Autonomous Okrug, and its GDP is 209.5 times larger than that of the Chukotka Autonomous Okrug.
Per capita consumption-based emissions vary greatly across federal entities (Figure S1).Nenets Autonomous Okrug is the region with the highest per capita carbon emissions (36.7 t/capita).Chechen Republic has the lowest per capita carbon emissions (1.1 t/capita).Figure 1 shows that per capita consumption-based emissions are higher in federal entities with higher per capita GDP and abundant energy.For example, Nenets Autonomous Okrug has the highest per capita carbon emissions (36.7 t/capita).This is due to its location in the oil and gas-rich Timan-Pechora Basin, and Nenets Autonomous Okrug is Russia's least inhabited territory (Table S1 in Supporting Information S1).Furthermore, Yamal-Nenets Autonomous Area and Khanty-Mansi Autonomous Area are Russia's primary gas and oil producers, respectively.Because of their abundant energy resources, these two regions rank second (28.7 t/capita) and fourth (19.2 t/capita) in Russia in terms of per capita carbon emissions.The per capita carbon emission in the above-mentioned locations is higher than the national level in Russia (6.7 t/capita), meaning that residents of these regions have a heavier burden of emission reduction.On the other hand, most regions with the lowest per capita carbon emissions are federal entities with the lowest per capita GDP, such as Chechen Republic and the Republic of Ingushetia (Figure 1, second row, right).
Clarifying the carbon emission intensity of each region will assist the government in developing more accurate emission reduction policies based on local factors.Carbon intensity (emissions per unit of GDP) based on consumption is higher in places with more developed industries or energy-intensive activities (such as mining).
The carbon intensity calculated based on consumption in the aforementioned regions was higher than the Russian national level (13.3 g/RUB).This may be due to the economic development of these areas depends heavily on activities related to carbon-intensive industries, which leads to their carbon intensity being higher than the Russian average.For example, the economy of Omsk Region is highly industrialized and carbon emissions are concentrated in the industrial sector.This means that each unit of economic output in the Omsk Region is accompanied by higher carbon emissions.By contrast, some regions with a larger share of service sector value added have lower carbon intensity.The carbon intensity of the Chechen Republic, for example, is 7.6 g/RUB, which is just 14% of the carbon intensity of the Lipetsk Region.In 2016, the value added of services in Voronezh region, Republic of Adygea and Chechen Republic accounted for 62%, 63% and 74% of the total value added in each region, respectively.Regions with more developed service sectors tend to have very low carbon intensities because they are concentrated in less polluting sectors such as catering, tourism and education.

Industry Structure and Driving Factors of Regional Consumption-Based Carbon Emissions
All 85 regions are categorized by trading patterns: 45 are net consumer regions in which consumption-based emissions exceed production-based emissions, whereas 40 are net producer regions, with the opposite emissions pattern.Figure 2a   Petersburg are the power sector in St. Petersburg (10.8 Mt), the power sector in the Sverdlovsk region (4.7 Mt), and high-end heavy industrial manufacturing in the Vologda region (2.9 Mt).The above-mentioned differences reflect the heterogeneity of carbon emission caused by consumption structure and industry structure in different regions.Consequently, different emission mitigation policy must be implemented in different regions based on local characteristics.
Khanty-Mansi Autonomous Okrug (31.6 Mt) and Lipetsk Region (26.4 Mt) rank seventh and ninth respectively in terms of consumption-based carbon emissions, but they are net producer regions.Their sectoral structure is dominated by industry, with energy production dominating in the Khanty-Mansi Autonomous Okrug and manufacturing in the Lipetsk Region.In terms of industry structure, the power sector (11.1 Mt) and basic services (6.5 Mt) account for the largest consumption-based emissions in Khanty-Mansi Autonomous Okrug.Meanwhile, high-end heavy industrial manufacturing (10.8 Mt) and construction (6.3 Mt) contribute the most to consumptionbased emissions in Lipetsk Region.Moreover, high-end heavy industrial manufacturing (43.6 Mt) stands as the primary contributor to Lipetsk's production-based emissions (Figure 2a).This is because iron processing and mechanical engineering are the most prominent industrial sectors in the Lipetsk Region.And Lipetsk Region has a federal special economic zone, several local special economic zones and industrial parks, with a favorable investment environment and complete supporting facilities.The park's construction regularly requires the consumption of carbon-intensive materials like cement and steel, resulting in significant carbon emissions in construction sector.Household final consumption (20.2 Mt) and fixed capital formation (6.3 Mt) are the main drivers of emissions in Khanty-Mansi Autonomous Okrug.The principal driver of carbon emissions in the Lipetsk Region is the formation of fixed capital formation (11.0 Mt), which is closely followed by household final consumption (8.2 Mt).A potential explanation is that the Lipetsk region receives a disproportionately large share of total fixed capital investment in Russia as well as the world.

Embodied Emissions in Interregional Trade
The trade-related embodied emissions of Russia amounted to 510 Mt, or 52% of the nation's overall consumptionbased carbon emissions.This observation highlights the potential occurrence of carbon leakage when solely emphasizing the direct carbon emissions of certain regions.However, the estimation of carbon emissions under consumer responsibility allows for a comprehensive calculation of both direct and indirect carbon emissions associated with each specific location.This approach facilitates a more equitable distribution of the responsibility for emissions reduction across various regions.In other words, regions that are net importers of embodied carbon emissions should bear more responsibility for reducing emissions.
Figure 3a illustrates the primary emissions transfers observed among the 85 federal subjects in Russia.The background color assigned to each federal subject corresponds to the net emissions value resulting from trade.We found that more developed regions tend to import more embodied emissions.In 2016, the regions like Moscow City, St. Petersburg, and the Moscow Region exhibited the highest GDP levels within the Russian Federation.They all belong to the net consumer region, which is the top three regions of Russia with the largest imports of embodied emissions.As shown in Figure 3a, they imported 115.5 Mt, 34.0 Mt and 27.7 Mt of carbon dioxide from other parts of Russia, respectively.In the case of Moscow, it is notable that the Chelyabinsk Region contributed 6.6 Mt of emissions to the city.Additionally, Lipetsk Region, Sverdlovsk Region, and Vologda Region transported 6.1 Mt, 5.4 Mt, and 5.1 Mt of emissions to Moscow, respectively.Regarding the industry structure depicted in Figure 3b, basic services (37.5 Mt), construction (20.7 Mt), high-tech services (14.1 Mt) are the main contributors to Moscow's embodied carbon emissions imports.This may be because Moscow is a net consumer region dominated by services (the value added of services in Moscow accounted for 85% of total value added in 2016) and does not have abundant energy resources.Hence, it is imperative to engage in the importation of substantial quantities of steel, coal, and other energy-intensive items from other regions to sustain social development.The provision of these products will result in a heightened level of production activities in other federal subjects, particularly in the more industrially advanced regions surrounding Moscow.From the perspective of final demand, the carbon emissions associated with imported goods primarily stem from household consumption, accounting for 61.2 Mt.This phenomenon could potentially be attributed to the comparatively elevated per capita consumption observed among inhabitants residing in developed areas, such as Moscow.The sectors that exhibit the largest levels of embodied emissions from imports related to household final consumption in Moscow are basic services (20.1 Mt) and light industrial manufacturing (11.2 Mt).
Earth's Future 10.1029/2023EF004323 Sverdlovsk Region (35.7 Mt) and Chelyabinsk Region (29.3 Mt) are the regions with the highest embodied emissions in exports.The reason for their higher carbon emissions can be attributed to the more advanced state of their steel sector and the greater volume of carbon-intensive items they export.As shown in Figure 3b, the sectors with the highest embodied emissions exported from Chelyabinsk Region are construction (8.9 Mt) and basic services (6.3 Mt).This may be because Chelyabinsk Region have developed mainly carbon-intensive industries such as metallurgy and car manufacturing, which have led to higher emissions embodied in the goods they export.Furthermore, Russia's main embodied emissions export locations are energy-rich regions like as Khanty-Mansi Autonomous Okrug (18.1 Mt).Hence, by optimizing the energy consumption structure in these locations and implementing more energy-efficient technologies, it is possible to yield a favorable outcome in the region's carbon reduction endeavors.

Discussion
The inter-state carbon emissions and internal transfers in Russia have multiple impacts on the global landscape.
To reflect Russia's position in the global economy and its impact on the environment (Jiang et al., 2020), we considered the embodied emissions in Russia's international trade.This study revealed that Russia's embodied emissions in exports amounted to 463.8 Mt, accounting for 31.8% of its production-based emissions.Emissions from the high-end heavy manufacturing sector amount to 208.5 Mt, which is the main contributor to the embodied emissions in Russian exports.The high-end heavy industrial manufacturing sectors in Chelyabinsk Region (62.5 Mt), Lipetsk Region (35.8 Mt), Sverdlovsk Region (24.2 Mt), and Vologda Region (20.9 Mt) collectively contribute to 68.8% of their combined embodied emissions in export.Chelyabinsk Region is the region with the highest embodied emissions from Russia's exports (65.1 Mt), with industry being the driving force behind its economic development and machine building being its dominant industry.In 2016, the embodied emissions in export of Chelyabinsk Region's high-end heavy industrial manufacturing accounted for 96.0% of the total embodied emissions in export in the region.For Lipetsk Region (36.2 Mt), it is one of the top three regions with embodied emissions in Russia's exports, with the metal industry and machinery manufacturing being its leading industries.High-end heavy industrial manufacturing (35.8 Mt) has become the main contributor to Lipetsk Region's embodied emissions in exports due to the sector's high carbon emission intensity and exports.In addition, Khanty-Mansi Autonomous Okrug (20.7 Mt) and Yamal-Nenets Autonomous Okrug (10.1 Mt) are the top two regions of Russia with the highest embodied emissions in exports of the oil sector.The main reason is that the Yamal-Nenets Autonomous Area's natural gas and oil mining output account for 90% of Russia's total annual natural gas production and 19% of Russia's total oil production, respectively.A large amount of oil and gas extraction and exports have led to higher embodied emissions in exports in the Yamal-Nenets Autonomous Area.
For Khanty-Mansi Autonomous Okrug is the largest oil-producing region in Russia and the world and is an export-oriented region (crude oil accounts for 99% of exports).In addition, its oil mining output and power generation capacity rank first in all regions of Russia.Therefore, Khanty-Mansi Autonomous Okrug (43.6 Mt) ranks as the second-largest contributor to Russia's embodied emissions in exports.The petroleum sector leads the pack with 20.7 Mt, followed closely by the electricity, gas, steam, and water sector, which contribute 14.1 Mt.However, according to OECD data, the embodied emissions in imports in 2016 totaled 136.1 Mt (OECD, 2021), accounting for only 13.8% of Russia's carbon footprint (983.5 Mt).Further decomposing these carbon emissions into 85 regions, each region's embodied emissions in imports will be smaller.As a result, this study overlooked the analysis of embodied emissions in Russia's imports.
Specifically, the findings of this study confirm significant variations in emissions across different regions of Russia.Due to the significant heterogeneity in different regions, a one-size-fits-all policy is not applicable to all regions of Russia, and it can reduce emission reduction efficiency in each region (Bai et al., 2021).In this research, the 85 regions of Russia were categorized into two distinct groups based on their placement along the supply chain: net consumer regions and net producer regions.For net producer regions, we recommend gradually promoting energy transition, industrial upgrading and technological improvement to achieve emission reduction targets.Given Russia's reliance on fossil fuels and their cost-effectiveness, transitioning to renewable energy is a formidable challenge.Hence, Russia should strategically utilize its ample renewable energy resources while still utilizing fossil fuels, progressively amplifying investments in clean energy.The industrial and energy sectors in Russia hold significant influence over domestic carbon emissions, with most productive regions exhibiting wellestablished industrial infrastructure.An illustrative instance can be observed in the Lipetsk Region, which is characterized by the significant development of iron processing and mechanical engineering.This region can be classified as a net producer region.The sector with the highest production-based carbon emission of Lipetsk Region is high-end heavy industrial manufacturing (43.6 Mt).Hence, it is recommended that the high-end heavy industrial manufacturing sector of Lipetsk Region should consider the adoption of production methods that are more energy-efficient and environmentally sustainable.As an illustration, Novolipetsk Steel (NLMK), Russia's largest steel producer, implemented innovative iron ore beneficiation technologies in its mining and beneficiation facilities from 2020 to 2023.This endeavor is designed to improve the grade of iron ore, leading to an annual reduction of 0.8 Mt of carbon emissions.Additionally, we suggest advocating for the advancement and utilization of environmentally friendly energy sources, with a gradual transition toward wind, solar, and other sustainable forms of energy.For example, encouraging NLMK to construct a new power plant utilizing secondary energy sources will assist in reducing its reliance on third-party energy, consequently decreasing carbon emissions.The new technologies employed in the power plant will allow the utilization of converter gas, partially substituting natural gas consumption.
However, the above strategies may not be effective for net consumer regions.For net consumer regions, we suggest paying more attention to optimizing consumption demand and consumption structure to achieve emission mitigation goals.Like most vast countries, Russia's regional emission differences also indicate that the wealthiest regions dominate the total consumption-based emissions (Huang et al., 2021).For example, in Moscow, which has the highest GDP in Russia, consumption-based emissions account for 14% of the country's total consumptionbased emissions.Among the several sectors in Moscow, it is seen that the basic services sector, with a carbon Earth's Future 10.1029/2023EF004323 emission of 43.4 Mt, exhibits the greatest level of carbon emissions.In order to address this issue, one potential solution is the implementation of a carbon tax policy specifically targeting high-emission products and services within the service sector.Despite significant obstacles posed by Russia's current overall consumption levels and international sanctions to the implementation of tariffs, the adoption of the Carbon Border Adjustment Mechanism signals that the establishment and imposition of carbon taxes will likely become a future trend.In addition, the implementation of a carbon tax may increase the price of energy and products, adversely affecting low-income households and energy-intensive industries (Guan et al., 2023).Furthermore, the implementation of a carbon tax has the potential to result in a rise in the cost of energy-related commodities, consequently constraining household spending and impacting the overall welfare of individuals (Zhang et al., 2023).Hence, the adoption of a carbon tax policy necessitates the incorporation of certain social security measures aimed at alleviating the negative consequences.For example, individual or household rebates(also called dividends)have increased public support for carbon taxes in Canada and Switzerland (Mildenberger et al., 2022).Additionally, it is worth noting that the construction sector in Moscow accounts for a substantial amount of carbon emissions, making it the second greatest contributor in the city, with a total of 22.5 Mt.Our recommendation entails enhancing the energy efficiency of buildings and infrastructure by implementing energy-saving lighting systems and implementing measures to increase building energy efficiency.Furthermore, it's suggested to factor in future climate change in architectural design to ensure that current buildings meet future performance needs and cut carbon emissions (Gesangyangji et al., 2022).For the transportation sector in Moscow, we propose to reduce the use of cars quickly and massively (Winkler et al., 2023), expand the subway and public transportation network and promote electric vehicles.
Russia is an expansive nation characterized by significant variations in socio-economic systems throughout its various regions.The decreasing of carbon emissions in various places is confronted with diverse risks and challenges.Certain places encounter the predicament of transitioning away from a state of reliance on energy sources.The development of clean energy implies the need to reduce dependence on these high-carbon energy sources, which may have a certain impact on the economy and employment in different regions (Jia et al., 2022;Tian et al., 2022Tian et al., , 2023)).In particular, the territories that rely on oil and gas as a means of fostering economic growth include the Yamal-Nenets Autonomous Area and the Khanty-Mansi Autonomous Okrug.Moreover, certain geographical areas are susceptible to detrimental consequences arising from severe climatic circumstances.The presence of extremely cold temperatures in many locations, like the Krasnoyarsk Region and the Republic of Sakha (Yakutia), necessitates a substantial amount of energy for heating purposes.Consequently, this heightened demand for energy can result in increased carbon emissions.Certainly, this may potentially incentivize Russia to enhance energy efficiency and allocate resources toward the development of renewable energy sources, thereby mitigating emissions.Not only that, the huge differences in economic conditions in different regions have also had an impact on the implementation of emission reduction policies.The Jewish Autonomous Region and the Republic of Altai, being economically disadvantaged regions, will face obstacles in their efforts to pursue technical advancements and implement emission reduction measures due to inadequate financial resources.It will be detrimental to the implementation of their emission reduction policies.Hence, it is imperative to allocate financial assistance and subsidies to economically disadvantaged and underdeveloped regions throughout the implementation of emission reduction measures.The ongoing conflict between Russia and Ukraine will likely have implications for Russia's carbon reduction ambitions and future industrial structure.For instance, the utilization of fuel in tanks, planes, and other machinery, the establishment of fortifications, and the manufacturing of weapons collectively contribute to the generation of additional carbon emissions (Climate Focus, 2022).This could delay the achievement of the goal of peaking carbon emissions in various regions of Russia, especially those with a more developed military industry.In addition, Russia's resource allocation and industrial structure will also be affected.It is anticipated that the government will enhance its backing for companies associated with warfare, so shifting the nation's focus toward the military sector at the expense of environmental and low-carbon industry.This shift is expected to impede the timely attainment of carbon peaking targets.

Conclusion
In this study, MRIO model is used to analyze the carbon footprint of Russian federal entities in 2016, which fills the data gap and provides data support for the development of carbon mitigation policies in the regions.The main results are as follows: First, we find that there are significant differences in consumption-based emissions across different regions of Russia.The emissions from the region with the highest consumption-based emissions were over 270 times greater than those from the region with the lowest consumption-based emissions.Many variables contribute to this inequality, including economic structure and consumption structure.Second, the carbon footprint of Russia is primarily influenced by final consumption in developed regions.Household final consumption plays a significant role in driving consumption-based emissions in various regions.In addition, fixed capital formation emerges as the second most influential factor contributing to consumption-based emissions.Third, carbon emissions in different regions not only occur within the territory, but also can be outsourced to other regions through interregional trade.In particular, carbon emissions in most parts of Russia are mainly driven by the demand from net consumer regions (such as Moscow and St. Petersburg).Consequently, it is imperative to accord greater consideration to the situation of each region throughout the supply chain when formulating mitigation policy.
To attain more precise carbon emission mitigation targets, it is imperative to conduct an analysis of the interplay between carbon emissions across different regions and sectors, considering both production and consumption sides.It is recommended that emission reduction strategies should be developed in a targeted and differentiated manners, considering the position of each region within the supply chain.Net producer regions should prioritize the enhancement of technological efficiency and the optimization of their energy structure.Net consumer regions should pay more attention to the optimization of consumption demand and consumption structure (such as encouraging low-carbon lifestyles).Furthermore, more ecologically friendly consumption habits might be encouraged, or carbon pricing could be introduced in key industries to discourage consumption.Implementing mitigation measures in net consumer regions has the potential to reduce carbon emissions not just at the local level, but also within net producer regions.

Figure 1 .
Figure1.The top 10 in consumption (row 1), the bottom 10 in consumption (row 2), the top 10 in domestic exports (row 3) and the top 10 in domestic imports (row 4), all are expressed in terms of regional total (left column), emissions per unit of Gross Domestic Product (GDP) (middle column) and per capita emissions (right column).The color of the bar chart from green to red indicates that GDP per capita from low to high.
illustrates the 10 biggest contributors to consumption-based carbon emissions in Russia.Production-based and consumption-based emissions of the above-mentioned federal entities are significantly different.Obviously, the most developed federal entities in Russia, such as Moscow, Moscow Region, St. Petersburg, are net consumer regions.Most of the industrial-oriented federal entities and underdeveloped federal entities are net producer regions.Moscow and St. Petersburg are characterized as net consumer regions, exhibiting urban economies that are constrained by a scarcity of natural resources.As indicated in Figure2a, the top three sectors contributing to Moscow's consumption-based emissions are basic services (43.4 Mt), construction (22.5 Mt), and high-tech services (13.4 Mt).The consumption-based carbon emissions from basic services are mainly caused by the final consumption of households and public administration.Carbon emissions from construction are mainly caused by fixed capital formation.This is because capital formation requires large amounts of steel, cement and electricity to support production, and these raw materials are carbon-intensive products(Meng et al., 2017).Notably, the power sector in Moscow (12.8 Mt), high-end heavy industrial manufacturing in Lipetsk (5.6 Mt), and high-end heavy industrial manufacturing in Chelyabinsk (4,7 Mt) are the top three contributors to Moscow's consumption-based emissions.As shown in Figure2b, 53% (73.7 Mt) of Moscow's consumption-based carbon by household final consumption.Among them, basic services (20.1 Mt), light industrial manufacturing (11.2 Mt) and high-tech services (10.5 Mt) are the top three sectors of carbon emissions caused by household final consumption in Moscow.This could be owing to the local population's high standard of living, which creates a strong demand for a variety of services, including high-end services.Furthermore, carbon emissions caused by fixed capital formation (35.3 Mt) account for 25% of Moscow's consumption-based emissions, making it the second largest contributor to Moscow's emissions.The subsequent analysis of St. Petersburg, Russia's second-largest city, reveals that it secured the second position in total consumption-based emissions in Russia, amounting to 50.5 Mt.In contrast to Moscow, the major contributors to consumption-based emissions in St. Petersburg are the power sector (14.8 Mt) and basic services (11.4 Mt).Household final consumption is the primary driver of emissions in St. Petersburg, followed by fixed capital formation, which contributes to 29.5 Mt and 11.0 Mt of consumption-based emissions, respectively.Unlike Moscow, the power sector (12.7 Mt), basic services (5.8 Mt) and light manufacturing (3.5 Mt) are the main contributors to emissions due to household final consumption in St. Petersburg.In addition, the top three contributors to consumption-based emissions in St.

Figure 2 .
Figure 2. (a) Comparison of industrial structure between production-based emissions and consumption-based emissions (Republic of Crimea and Sevastopol City are not represented in the map).(b) Consumption-based carbon emissions and driving factors in various regions of Russia.The figure highlights the emission driving factor structures of 10 federal entities with the largest consumption-based carbon emissions, in which D1-D7 respectively represent seven final demand categories: (D1) Final consumption expenditure by households, (D2) Individual final consumption expenditure by Public Administration, (D3) Collective final consumption expenditure by Public Administration, (D4) Final consumption expenditure by non-profit institution serving households (NPISH), (D5) Gross fixed capital formation, (D6) Change in inventories, (D7) Acquisition less disposals of valuables.(Republic of Crimea and Sevastopol City are not represented in the map).

Figure 3 .
Figure 3. (a) Major carbon emission transfers among 85 federal entities in Russia (Republic of Crimea and Sevastopol City are not represented in the map).(b) Embodied emissions in trade at the sectoral level.

Table 1 A
Brief Schematic of Region-Level MRIO Table Construction