Using CO2 emission stability to assess energy system resilience during COVID-19

The COVID‑19 pandemic abruptly disrupted economic activity and daily life, reshaping global fossil‑fuel CO₂ emissions. We use CO₂ emission stability to characterize how national energy systems absorbed and recovered from this common shock, providing an empirical view of energy system resilience. We compile annual fossil‑fuel CO₂ emissions and energy‑system indicators for 64 economies over 2015–2019 and 2019–2023. Emission stability is defined as the absolute difference between pre‑ and post‑COVID compound annual growth factors. Spatial clustering is assessed with Getis–Ord Gi* statistics. An entropy‑based resilience index is built from indicators of fuel mix, low‑carbon electricity, energy diversity, import dependence, and per‑capita energy use and income. From 2015–2019 to 2019–2023, Europe consolidates its role as a CO₂ emission cold‑spot region, while parts of Latin America and the Caribbean emerge as new hot spots. High‑income economies generally maintain declining or stable emissions and higher resilience scores, with limited rebound. Upper‑middle‑income countries, especially in East Asia and the Pacific, remain the main contributors to emission growth, driven mainly by coal. In most countries, emission stability and resilience fall into the same or adjacent classes. Using CO₂ emission stability as a proxy offers a tractable way to compare energy system resilience under a shared global shock. Yet stability can also reflect non‑resilience factors such as lockdowns, macroeconomic contractions, fuel price shocks, or hydrological variability. Our results therefore characterize structural, annual‑scale resilience and highlight differentiated policy needs across income groups.

1 Introduction

The energy system is the entire process of transforming natural energy resources in the natural world into specific forms of energy required for human social production and living (Hughes, 2012). An efficient and reliable energy system is crucial for the normal operation of modern society (Ahmadi et al., 2021). However, unexpected natural events (such as earthquakes, hurricanes, floods, etc.) or human-made events (such as terrorist attacks, strikes, COVID-19, etc.) could cause disturbances in the operation of the energy system, resulting in significant financial and societal damages (Schlör et al., 2018; Wainstein et al., 2019; Han et al., 2021). Hence, scholars have extensively studied the capacity of the energy system to withstand risks, also known as the energy system resilience (Mochizuki and Chang, 2017). Following the declaration of COVID-19 as a global pandemic by the World Health Organization on 11 March 2020, countries around the world, regardless of their level of development, have introduced unprecedented measures to contain the virus and halt daily activities (Pan, 2025), leading to changes in energy usage and CO₂ emissions (Liu et al., 2020). The COVID-19 pandemic provided a rare opportunity to examine how national energy systems respond to abrupt disruptions in demand and supply (Liu et al., 2022a).

The energy system resilience falls within the realm of resilience theory, a concept originating from engineering to describe the compressive strength or elasticity of materials such as wood or steel (Wears, 2006). In 1973, ecologist Holling introduced this concept into the field of ecology, focusing on the long-term survival strategies and operational mechanisms of populations, species, and ecosystems in ever-changing and fluctuating natural environments (Holling, 1973). Over time, research on resilience has expanded from a singular ecological focus to fields such as engineering, sociology, and economics. In disaster studies and social sciences, concepts like resilient cities and resilient communities highlight the ability of urban and community infrastructures to recover from disaster impacts. Moreover, the interdisciplinary integration endows resilience with a richer meaning, making resilience research a current focal point in academia.

Scholars have different perspectives on the concept of resilience in energy systems. According to the IEA, energy resilience is the capability of the energy system and its parts to withstand dangerous occurrences or patterns, while preserving their fundamental functions, qualities, and configurations, along with the ability to adjust, acquire knowledge, and evolve (IEA, 2015). On the other hand, Gatto and Drago (2020) argue that energy resilience refers to the energy system’s ability to preserve, adapt, respond to, and solve the impacts brought about by economic, social, environmental, and institutional factors, and this ability stems from the capacity to adapt to change. Some scholars (Faturechi and Miller-Hooks, 2015; Hosseini et al., 2016) also believe that resilience encompasses various aspects of a system’s abilities, such as predictive ability, the capacity to absorb shocks, and the ability to maintain functionality. Furthermore, factors such as system reliability, robustness, risks, stability, survivability, flexibility, agility, fault tolerance, and fragility can all influence the resilience of energy systems (Jasiūnas et al., 2021). Recent research has explored the pandemic’s impacts on energy demand, emissions reduction, and recovery dynamics (Le Quéré et al., 2021; Liu et al., 2023). However, few studies have quantitatively linked CO₂ emission stability to energy system resilience, especially from a systemic and comparative perspective across multiple countries.

Examining the concept of resilience in energy systems from various disciplinary perspectives leads to different understandings. For instance, from an engineering standpoint, Roege et al. (2014) suggest that energy system resilience includes four characteristics: preparatory, recovery, absorptive, and adaptive. Phillips et al. (2016) propose that energy system resilience features preparatory, recovery, mitigation capacity, and responsiveness. After summarizing a significant number of features and factors related to energy system resilience in the literature, Ahmadi, Saboohi, and Vakili (2021) identify four key features of energy resilience: anticipation, absorption, adaptation, and recovery. Looking at it from a management perspective, key features of energy resilience include energy diversity (Sato et al., 2017), energy productivity and sustainability (Schlör et al., 2018). Gatto and Drago (2020) conduct global resilience assessments of relevant countries using the Global Energy Resilience Index (GERI), which comprises indicators such as energy availability, energy efficiency, and renewable energy. Additionally, many scholars view resilience as a part of energy security (Cherp and Jewell, 2011; Manshadi and Khodayar, 2016; Azzuni and Breyer, 2018). Existing resilience frameworks often focus on technical recovery or energy security indicators, while emission dynamics are rarely employed as a direct proxy for resilience (Byrom et al., 2020; Friedlingstein et al., 2022). Yet, CO₂ emission stability—defined as the temporal persistence of emission levels under external shocks—can serve as a systemic measure of an energy system’s ability to absorb and adapt (Association for Computing Machinery, 2012; Wang et al., 2019). Recent progress in near-real-time CO₂ emissions monitoring — such as the high-frequency global estimates from International Energy Agency (IEA) showing a 1.1% rise in energy-related CO₂ in 2023, despite the pandemic’s disruption (IEA, 2024) — and the evolving energy-system resilience frameworks that integrate both technical and socio-institutional dimensions (Jasiūnas et al., 2025; Hotchkiss, 2023; Gong, 2024) provide the methodological foundation for the current study. By merging high-resolution emission data with systemic resilience theory, this paper operationalizes “emission stability” as a proxy for energy-system resilience and quantifies how national energy systems maintained (or failed) stability under the shock of COVID-19 (Figure 1). This study aims to (1) quantify the short-term emission stability of national energy systems during COVID-19, (2) evaluate the relationship between emission volatility and energy structure characteristics, and (3) propose a stability-based framework for assessing energy resilience in future global shocks.

Accordingly, the research addresses the following questions:

i. How did national CO₂ emissions fluctuate during the pandemic, and what patterns of stability emerged across different economies?

ii. To what extent can emission stability serve as a robust indicator of systemic resilience?

iii. What implications do these findings hold for designing resilient low-carbon transitions?

Figure 1

Figure 1. Study workflow.

2 Materials and methods

2.1 Data source

This study is based on annual data for 64 economies. Relative to higher-frequency statistics, the annual scale offers broader cross-country coverage and greater consistency in definitions, which supports our focus on structural recovery before and after the pandemic. This paper primarily focuses on the CO₂ emissions resulting from the consumption of fossil fuels, including coal, crude oil, and natural gas. The data used in this study and its sources are presented in Table 1.

Table 1

Table 1. Data sources.

2.2 Methods

2.2.1 Spatial autocorrelation

Analyzing cold and hot spots can indicate how closely individual units are clustered spatially (Ord and Getis, 1995). The study utilizes hot and cold spot analysis to detect notable spatial clustering of high and low CO₂ emission compound annual growth rates calculated using Getis-Ord Gi*, which are denoted as hot spots and cold spots, respectively. The global G index formula is provided below:

$\mathit{G}=\frac{{\displaystyle {\sum }_{\mathit{i}=1}^{\mathit{n}}}{\displaystyle {\sum }_{\mathit{j}=1}^{\mathit{n}}}{\mathit{W}}_{\mathit{i}\mathit{j}}{\mathit{X}}_{\mathit{i}}{\mathit{X}}_{\mathit{j}}}{{\displaystyle {\sum }_{\mathit{i}=1}^{\mathit{n}}}{\displaystyle {\sum }_{\mathit{j}=1}^{\mathit{n}}}{\mathit{X}}_{\mathit{i}}{\mathit{X}}_{\mathit{j}}}(1)$

The value of the G_i* index for sample i is determined in the following manner (Equation 2):

${\mathit{G}}_{\mathit{i}}^{*}=\frac{{\displaystyle {\sum }_{\mathit{j}=1}^{\mathit{n}}}{\mathit{W}}_{\mathit{i}\mathit{j}}{\mathit{X}}_{\mathit{i}}}{{\displaystyle {\sum }_{\mathit{j}=1}^{\mathit{n}}}{\mathit{X}}_{\mathit{j}}}(2)$

The Z-score for G_i* is utilized to evaluate the statistical significance of G_i*. The equation can be expressed in the following manner (Equation 3):

$\mathit{Z}\left({\mathit{G}}_{\mathit{i}}^{*}\right)=\frac{{\mathit{G}}_{\mathit{i}}^{*}-\mathit{E}\left({\mathit{G}}_{\mathit{i}}^{*}\right)}{\sqrt{\mathit{V}\mathit{A}\mathit{R}\left({\mathit{G}}_{\mathit{i}}^{*}\right)}}(3)$

X_i and X_j represent the compound annual growth rate for country i and country j, respectively. W_ij is the spatial weight matrix, while E (G_i*) and VAR(G_i*) denote the mathematical expectation and coefficient of variation of G_i*. Z (G_i*) is the standardized statistic of the G_i* test, and its significance can determine the spatial distribution of hot and cold spots in various regions.

2.2.2 CO₂ emission stability

CO₂ Emission Stability (ES) of an economy is calculated as follows Equations 15–18:

1. Calculate CO₂ emissions compound annual growth rate α:

${\mathit{E}\mathit{M}}_{\mathit{y}}={\mathit{E}\mathit{M}}_{{\mathit{y}}_0}{\mathit{\alpha }}^{\left(\mathit{y}-{\mathit{y}}_0\right)}(4)$

where, EM_y represents the yearly base period CO₂ emissions, EM_y0 represents the CO₂ emissions at time y, y is the calendar year, y₀ is the base year, and α is the mean annual growth factor over the period. In this study, we partition the timeline into the pre–COVID-19 period (2015–2019, α_a) and the post–COVID-19 period (2019–2023, α_b).

Accordingly, from Equation 4 we obtain Equations 5, 6:

${\mathit{\alpha }}_{\mathit{a}}={\left(\frac{{\mathit{E}\mathit{M}}_{2019}}{{\mathit{E}\mathit{M}}_{2015}}\right)}^{\frac{1}{4}}(5)$

${\mathit{\alpha }}_{\mathit{b}}={\left(\frac{{\mathit{E}\mathit{M}}_{2023}}{{\mathit{E}\mathit{M}}_{2019}}\right)}^{\frac{1}{4}}(6)$

2. Calculate CO₂ emission stability ES as Equation 7:

$\mathit{E}\mathit{S}=\left|{\mathit{\alpha }}_{\mathit{b}}-{\mathit{\alpha }}_{\mathit{a}}\right|(7)$

We believe that this approach can better reflect the status of a country’s CO₂ emissions pre- and post- the COVID-19 pandemic.

2.2.3 Energy system resilience

Energy system resilience (ESR) of an economy is calculated as follows:

1. Calculate Energy System Score (ESS):

Here, we define the resilience of energy systems as their ability to reliably supply green energy, meaning that the energy system performance should remain stable before the pandemic (2015–2019) and after the pandemic (2019–2023) (Ahmadi et al., 2021; Gatto and Drago, 2020; Dong et al., 2021; Jasiūnas et al., 2021). This focuses on dynamic performance throughout the shock, rather than a static state at a point in time (Zhao et al., 2022). Among these, all indicators will be categorized into benefit-type indicators (positive indicator, +) and cost-type indicator (negative indicator, -). After standardization of all indicators, the energy system score for each economy in 2015, 2019 and 2023 is calculated through entropy method. Indicators in Table 2 are re-positioned as resilience enablers/determinants, explaining why some systems better sustain supply continuity.

Table 2

Table 2. Indicator framework of energy system score.

Mapping indicators to resilience dimensions and sensitivity note. Our indicator set primarily captures absorptive and adaptive capacities through energy diversity (+), low-carbon electricity share (+), fossil share (−), import dependence (−), and system scale proxies (per-capita electricity and primary energy, +). While we lack direct engineering reliability measures (e.g., SAIDI/SAIFI, reserve margins), energy diversity and import dependence partially proxy infrastructure robustness by reflecting redundancy and exposure, respectively. To assess construction robustness, we examined alternative entropy weightings (e.g., equal-weight baseline; excluding GDP per capita), finding that cross-country rankings and the qualitative association between ESR and emission stability remain consistent. We discuss implications and future integration of engineering reliability metrics in Limitations and future research.

Among these, the energy diversity (coal, crude oil, natural gas, and other primary energy) is quantified using the Shannon-Wiener index in Equation 8.

$\mathit{D}=-\sum \left({\mathit{p}}_{\mathit{i}}\mathit{ln}{\mathit{p}}_{\mathit{i}}\right)(8)$

where p_i represents the share of energy type i in total primary energy consumption.

The specific steps of the entropy technique are outlined in the following sections (Amiri et al., 2014; Han et al., 2015):

i. Standardize the original value of indicators as Equations 9, 10:

Positive indicator:

${\mathit{X}}_{\mathit{i}\mathit{j}}^{\prime }=\frac{{\mathit{X}}_{\mathit{i}\mathit{j}}-\mathit{min}\left({\mathit{X}}_{\mathit{j}}\right)}{\mathit{max}\left({\mathit{X}}_{\mathit{j}}\right)-\mathit{min}\left({\mathit{X}}_{\mathit{j}}\right)}(9)$

Negative indicator:

${\mathit{X}}_{\mathit{i}\mathit{j}}^{\prime }=\frac{\mathit{max}\left({\mathit{X}}_{\mathit{j}}\right)-{\mathit{X}}_{\mathit{i}\mathit{j}}}{\mathit{max}\left({\mathit{X}}_{\mathit{j}\mathit{t}}\right)-\mathit{min}\left({\mathit{X}}_{\mathit{j}\mathit{t}}\right)}(10)$

where, X^’_ij represents the standardized value of the ith object being evaluated on the jth indicator, with X_ij being the initial value; max(X_j) and min(X_j) denote the highest and lowest values, respectively.

ii. The proportion of the ith evaluating object on the jth indicator is calculated as Equation 11:

${\mathit{Y}}_{\mathit{i}\mathit{j}}=\frac{{\mathit{X}}_{\mathit{i}\mathit{j}}^{\prime }}{{\displaystyle {\sum }_{\mathit{i}=1}^{\mathit{m}}}{\mathit{X}}_{\mathit{i}\mathit{j}}^{\prime }}(11)$

iii. The entropy of each evaluating indicator can be defined as Equation 12:

${\mathit{e}}_{\mathit{j}}=-\mathit{k}{\displaystyle \sum _{\mathit{i}=1}^{\mathit{m}}}\left({\mathit{Y}}_{\mathit{i}\mathit{j}}\times \mathit{ln}{\mathit{Y}}_{\mathit{i}\mathit{j}}\right),{\mathit{Y}}_{\mathit{i}\mathit{j}}>0(12)$

where, k = 1/lnm, and m is the number of evaluating object. The evaluating object i on the indicator j is excluded if Y_ij = 0.

The redundancy of the entropy is computed as follows Equation 13:

${\mathit{d}}_{\mathit{j}}=1-{\mathit{e}}_{\mathit{j}}(13)$

iv. The weight of entropy of each evaluating indicator in t year could be expressed as Equation 14:

${\mathit{w}}_{\mathit{j}}={\mathit{d}}_{\mathit{j}}/{\displaystyle \sum _{\mathit{j}=1}^{\mathit{n}}}{\mathit{d}}_{\mathit{j}}(14)$

1. Refer to Equations 4–7, Energy System Resilience (ESR) of an economy is calculated as follows:

${\mathit{E}\mathit{S}\mathit{S}}_{\mathit{y}}={\mathit{E}\mathit{S}\mathit{S}}_{{\mathit{y}}_0}{\mathit{\beta }}^{\left(\mathit{y}-{\mathit{y}}_0\right)}(15)$

${\mathit{\beta }}_{\mathit{a}}={\left(\frac{{\mathit{E}\mathit{S}\mathit{S}}_{2019}}{{\mathit{E}\mathit{S}\mathit{S}}_{2015}}\right)}^{\frac{1}{4}}(16)$

${\mathit{\beta }}_{\mathit{b}}={\left(\frac{{\mathit{E}\mathit{S}\mathit{S}}_{2023}}{{\mathit{E}\mathit{S}\mathit{S}}_{2019}}\right)}^{\frac{1}{4}}(17)$

$\mathit{E}\mathit{S}\mathit{R}=\left|{\mathit{\beta }}_{\mathit{b}}-{\mathit{\beta }}_{\mathit{a}}\right|(18)$

ESS_y0 represents the yearly base period energy system score, while ESS_y represents the energy system score at time y. We believe that this approach can better reflect the energy system resilience of a country pre- and post- the COVID-19 pandemic.

3 Results

3.1 Evolution of fossil-fuel CO₂ emissions across income groups

This section presents the evolution of fossil-fuel CO₂ emissions, spatial clustering patterns, and the relationship between emission stability and energy system resilience across 64 economies. Each subsection begins with a summary statement to guide readers.

Figure 2 illustrates the CO₂ emissions from coal, crude oil, and natural gas in major global economies in 2015, 2019 and 2023. These 64 countries account for over 80% of the global CO₂ emissions. From 2015 to 2019, fossil fuel CO₂ emissions from 64 major economies increased from 29.37 Gt to 30.52 Gt, representing a growth rate of 3.92%. During this period, emissions associated with coal usage decreased, while CO₂ emissions related to natural gas increased by 0.79 Gt. From 2019 to 2023, fossil fuel CO₂ emissions from the same 64 major economies rose from 30.52 Gt to 31.38 Gt, with a growth rate of 2.82%. Unlike the previous period from 2015 to 2019, the increase in CO₂ emissions during this timeframe was primarily driven by emissions associated with coal, whereas emissions related to crude oil experienced a decline.

Figure 2

Figure 2. Changes in CO₂ emissions from different fossil fuels in world’s major economies in 2015, 2019 and 2023.

We classified 64 major countries into high income, upper middle income, and lower middle income categories according to the World Bank’s classification, as illustrated in Figure 3a. For consistency and comparability in this analysis, we used the 2023 classification as a fixed reference point throughout the paper. Among these, there are 38 high income countries, 19 upper middle income countries, and 7 lower middle income countries. High income countries are mainly concentrated in North America and Europe. Upper middle income countries are primarily distributed in Asia and the Pacific region, while lower middle income countries are mainly concentrated in South Asia and West Asia. Figures 3b–d present the CO₂ emissions from coal, natural gas, and crude oil for countries of different income levels in 2015, 2019 and 2023. From the Figure 3, it can be observed that the CO₂ emissions from upper middle income countries are nearly equivalent to the combined emissions of high income countries and lower middle income countries. Upper middle income countries account for approximately 49% of the CO₂ emissions in the study area in 2023, with high income countries and lower middle income countries accounting for 38% and 13% respectively. In 2015, high income, upper middle income, and lower middle income countries accounted for 45%, 45%, and 10%, respectively. By 2019, these figures changed to 41%, 47%, and 12%. From 2015 to 2019, fossil fuel CO₂ emissions from high income countries decreased from 13.18 Gt to 12.76 Gt, representing a reduction of 3.19% (Figure 3b). During this period, only emissions associated with coal usage declined, while CO₂ emissions related to natural gas increased by 0.39 Gt. From 2019 to 2023, fossil fuel CO₂ emissions from high income countries rose from 12.76 Gt to 31.38 Gt, reflecting a decrease of 7.37%. Notably, between 2019 and 2023, CO₂ emissions associated with all three fossil fuels experienced a decline. From 2015 to 2019, fossil fuel CO₂ emissions from upper middle income countries increased from 13.33 Gt to 14.30 Gt, representing an increase of 7.28% (Figure 3c). In contrast, from 2019 to 2023, this figure rose by 9.37%. Furthermore, in both periods—2015 to 2019 and 2019 to 2023—the extensive use of coal was the primary factor contributing to the increase in fossil fuel CO₂ emissions in upper-middle-income countries. From 2015 to 2019, fossil fuel CO₂ emissions from lower middle income countries increased by 0.6 Gt, representing a growth rate of 20.98% (Figure 3d). From 2019 to 2023, this figure changed to an increase of 0.46 Gt, with a growth rate of 11.51%. The growth rate during this period remained significantly higher than that of high income and upper middle income countries. Therefore, the characteristics of CO₂ emissions from the 64 major economies during these two periods (2015–2019 and 2019–2023) are that upper middle income countries held a significant share and were the primary contributors to the increase in CO₂ emissions, while lower middle income countries experienced rapid growth in their CO₂ emissions.

Figure 3

Figure 3. Changes in CO₂ emissions from different fossil fuels in countries with different income levels from in 2015, 2019 and 2023. (a) Income group. (b) High ncome. (c) Upper middle income. (d) Lower middle income.

3.2 Spatial clustering and significance of CO₂ emission changes

Hotspot analysis reveals significant clustering patterns in East Asia and the Pacific, with cold spots persisting in Europe.

As depicted in Figure 4a, the CO₂ emission compound annual growth rates of the world’s major economies are divided into five classes according to the natural breaks. During the period 2015–2019, European and American countries predominantly exhibited emission reductions, while Asian nations, particularly those in Southeast Asia, and African countries were characterized by emission increases. As illustrated in Figure 4b, during the period 2019–2023, countries in Latin America and Caribbean, as well as those in East Asia and Pacific region, experienced a continuous rise in CO₂ emissions. In contrast, a majority of European countries witnessed a decline in CO₂ emissions. Overall, both before and after the COVID-19 pandemic, most countries in East Asia and Pacific region sustained emissions growth, whereas the United States and a majority of European countries consistently reduced emissions. Certain nations, primarily in Latin America and Caribbean region, shifted from emission reduction to increase—a development that warrants attention.

Figure 4

Figure 4. CO₂ emission compound annual growth rates distribution in 2015–2019 and 2019–2023. (a) 2015–2019. (b) 2019–2023.

Figure 5 depict the cold and hot spots of the compound annual growth rates of CO₂ emissions among various economies during the periods 2015–2019 and 2019–2023, respectively. One of the most notable features is that many secondary cold spot transitioned into cold spot after the pandemic. As evident from Figure 5a, during 2015–2019, with the exception of Russia, Europe emerged as a cold spot for the compound annual growth rates of CO₂ emissions, while countries in East Asia and Pacific region were identified as hot spots. In Figure 5b, it is observed that during 2019–2023, a majority of European countries continued to be cold spots, and there was a reduction in hotspots in East Asia and Pacific region, with Brazil emerging as new hot spots.

Figure 5

Figure 5. CO₂ emission compound annual growth rates cold and hot spots distribution in 2015–2019 and 2019–2023. (a) 2015–2019. (b) 2019–2023.

3.3 Using CO₂ emission stability to assess energy system resilience

High income countries maintain stable low-carbon transitions; upper middle income economies exhibit rebound-driven growth; lower middle income groups remain volatile.

As shown in Figure 6, the values of CO₂ emission stability and energy system resilience of the world’s major economies are divided into five classes according to the natural breaks. Overall, the changes of CO₂ emission in East Asia and Pacific, North America and most of Europe is relatively stable. Energy systems in economies other than Central Asia and Africa are well resilient. It is noteworthy that the changes in CO₂ emissions in Latin America and Caribbean countries are less stable, but they still have some resilience in their energy systems. Overall, the resilience of energy systems and the stability of CO₂ emissions are at the same or relatively similar levels in most countries.

Figure 6

Figure 6. Spatial distribution of CO₂ emission stability and energy system resilience of world’s major economies. (a) Emission stability. (b) Engery system resilience.

As shown in Table 3, 42 economies (indicated by the underlined numbers in the Table 3) have CO₂ emission stability and energy system resilience that are at the same or similar levels. Among the remaining 22 countries, 12 countries also did not exceed two levels in either CO₂ emission stability or energy system resilience.

Table 3

Table 3. CO₂ emission stability (ES) and energy system resilience (ESR) matrix.

4 Discussion

4.1 Clarifying the boundary between stability and resilience

Emission stability is informative of resilience when shocks are exogenous and systems maintain core functions without extraordinary restrictions. However, stability (or instability) can also result from non-resilience drivers (i) policy suppression (e.g., stringent lockdowns) can transiently stabilize emissions irrespective of system robustness; (ii) macroeconomic contractions can reduce emissions absent structural resilience; (iii) fuel price/security shocks and hydrological variability may destabilize emissions in otherwise robust systems through short-term switching; and (iv) inventory/trade adjustments can shift emissions across borders. Our annual-scale approach is thus best viewed as a measure of structural resilience, to be complemented by higher-frequency dynamics and sectoral context.

4.2 Integrating absorptive, adaptive, and transformative resilience perspectives

This study shows that the COVID-19 shock reconfigured the geography of CO₂ emission dynamics across 64 major economies, with Europe and parts of North America maintaining declines or stability while several economies in East Asia and the Pacific and in Latin America and the Caribbean exhibited renewed growth. Interpreted through the lens of energy transition and resilience research, these patterns are consistent with structural differences in energy systems, economic composition, and policy responses.

First, countries with higher pre-existing decarbonization momentum and electricity mixes with larger low-carbon shares tended to avoid persistent rebounds and kept emissions closer to pre-pandemic trajectories. This aligns with work highlighting that diversified, low-carbon power systems and demand-side flexibility enhance the capacity to absorb shocks and maintain core functions, i.e., resilience (IEA 2015; Jasiūnas et al., 2021; Heffron et al., 2021).

Second, economic structure matters. High-income economies with larger service sectors experienced sharper declines during lockdowns yet limited rebounds during recovery, consistent with observations that services are more amenable to remote activity and efficiency improvements, while manufacturing- and construction-heavy structures are more prone to cyclical rebounds (Ntounis et al., 2022; Le Quéré et al., 2021).

Third, fuel price and security shocks intersected with the pandemic. The 2021–2022 surge in natural gas prices and supply risks prompted coal substitution in several regions, contributing to coal-related emission increases despite ongoing recovery (Davis et al., 2022). Where import dependence is high and energy portfolios are less diverse, the capacity to buffer such price/supply shocks is reduced, which is in line with the energy security–resilience link (Kruyt et al., 2009; Stirling, 2010).

Comparison with high-frequency studies based on daily emissions (e.g., Liu et al., 2020; Liu et al., 2022b) suggests directional consistency at the regional and income-group levels. For instance, prior high-frequency work reports persistent declines in parts of Europe and rebounds in some East Asian and Latin America and the Caribbean economies following the initial shock, patterns that align with our annual-scale findings. While annual aggregation cannot reveal the amplitude or timing of weekly/monthly “V-shaped” fluctuations, it does capture structural recovery paths and relative repositioning across economies, thereby providing a consistent cross-country framework for assessing resilience at the annual scale.

Furthermore, in line with resilience theory (Holling, 1973; Ahmadi et al., 2021; Gatto and Drago, 2020), energy systems exhibit three complementary dimensions: absorptive capacity (the ability to buffer short-term shocks such as demand collapse or fuel price volatility), adaptive capacity (the ability to adjust operational structures and diversify energy sources under evolving constraints), and transformative capacity (the ability to reconfigure toward fundamentally lower-carbon and more flexible systems).

High income economies demonstrate all three capacities simultaneously, while upper middle income economies are largely in the adaptive phase—enhancing renewable penetration yet still dependent on legacy coal infrastructures. Lower middle income economies primarily display absorptive capacity, struggling to translate short-term stability into structural transformation.

Overall, our empirical patterns fit a broader narrative in the transition literature: path dependencies in fuel mix, infrastructure, and policy regimes condition how systems absorb, adapt to, and recover from shocks, and these path dependencies vary systematically across income groups and regions.

4.3 Comparative insights by income group and region

Most high income economies continued pre-pandemic declines through 2019–2023 or avoided sustained rebounds (e.g., the United States, Germany, the United Kingdom, France). This stability coheres with higher low-carbon shares, stronger market and regulatory instruments (e.g., carbon pricing, renewables targets), and greater deployment of flexibility resources. Consequently, the CO₂ emissions of high income countries have not experienced sustained rebounds. This reflects the better resilience of high income countries in terms of their energy and industrial structures when facing the impact of unforeseen events. Additionally, energy system resilient entities may find it easier to transition to a low-carbon economy.

As for upper middle income economies, aggregating emissions increased in both sub-periods, predominantly coal-driven, with China and India contributing the largest absolute increments. While resilience signals are present—emissions generally recovered swiftly—the system response is more sensitive to demand rebounds and fuel availability, consistent with ongoing coal reliance and rapid industrial expansion. At the same time, countries in this group have also expanded renewables, which helps explain why rebounds do not uniformly translate into high instability. But the energy and industrial structures of upper middle income countries need improvement.

Emission trajectories of lower middle income economies display greater volatility and less predictable reversion toward pre-pandemic trends, indicating higher vulnerability and thinner buffers (e.g., Pakistan, Egypt, Bangladesh). Constraints include limited fiscal space for green stimulus, greater exposure to international price swings, and infrastructure gaps.

Quantitatively, the average CO₂ emission stability (ES) values were 0.84, 0.80, and 0.20 for high-, upper middle-, and lower middle income groups, respectively, indicating progressively weaker stability with declining income levels. Correspondingly, the energy system resilience (ESR) index averaged 0.88, 0.78, and 0.13. Notable outliers include Brazil, where ESR is higher than predicted by ES, reflecting strong hydropower buffering, and India, where ES remains moderate despite rapid growth, due to expanding renewables. These cross-group patterns substantiate the absorptive–adaptive–transformative framework discussed above.

Brazil. Despite a structurally low-carbon power mix dominated by hydropower, Brazil experienced periods of elevated emissions during the study window due to hydrological droughts that necessitated hydro–thermal switching and emergency thermal dispatch. This dynamic illustrates how short-term volatility can coexist with medium-term resilience: the system’s flexibility and interconnections enable recovery, yet hydro variability can temporarily destabilize emissions independent of underlying robustness.

China. Stringent lockdowns in 2020 and localized measures in subsequent waves rapidly suppressed activity and emissions, followed by an investment-led recovery that lifted industrial energy use (e.g., cement, steel). These policy cycles can produce stability or volatility that does not solely reflect energy-system resilience. Sectoral signals indicate that power-sector decarbonization advances improved structural absorptive capacity, while industrial rebounds shaped aggregate volatility. Together, these cases show why emission stability must be interpreted alongside policy and resource contexts.

4.4 CO₂ emission stability as a proxy for energy system resilience

The primary contribution of this study lies in proposing an approach to assessing energy system resilience through the lens of CO₂ emission stability. Specifically, we use changes in CO₂ emissions as a proxy indicator of how resilient an energy system is in the face of external shocks, such as the COVID-19 pandemic. This innovative perspective enables a more straightforward and observable measure of resilience, providing new insights into the dynamic responses of energy systems under crises. By focusing on emission stability, our approach bridges environmental data with resilience evaluation, offering a practical tool for policymakers and researchers to monitor and enhance the robustness of energy infrastructures in challenging times.

4.5 Policy implications differentiated by country context

4.5.1 High income countries

As shown in Figure 6 and Table 4, policy priorities should focus on consolidating decarbonization momentum while safeguarding system adequacy and flexibility Table 5. This entails:

1. Sustained investment in clean electricity (renewables, storage, grid modernization, digitalization) to limit rebound effects;

2. Market and regulatory designs that remunerate flexibility, reliability, and demand response, thereby stabilizing emissions during demand shocks;

3. Credible, durable carbon-pricing and standards that reinforce long-term expectations and reduce policy risk. Given recent fuel price and security shocks, risk-management instruments (e.g., strategic reserves, diversified gas contracts, and hedging frameworks) are warranted to reduce short-term substitution back to high-carbon fuels. Where appropriate, scaling firm low-carbon capacity (e.g., nuclear, CCS-equipped plants, long-duration storage) can support adequacy without undermining long-run emission trajectories.

Table 4

Table 4. The standard deviation, average and Coefficient of variation of ES and ESR by income group.

Table 5

Table 5. Policy summary.

4.5.2 Upper middle income countries

The central challenge is to decouple post-shock activity rebounds from emissions in contexts with higher coal reliance and rapid industrial expansion. A pragmatic sequencing would:

1. Accelerate clean power deployment and grid integration (including transmission expansion and system flexibility) to reduce the marginal emissions intensity of growth.

2. Target the retirement or repurposing of the least-efficient coal assets while preventing new lock-ins through clear capacity-planning signals;

3. Advance industrial electrification and end-use efficiency (with attention to process heat and motors), supported by stable tariff, interconnection, and permitting regimes;

4. Strengthen fuel-price and import-risk management to mitigate volatility. De-risking instruments (e.g., guarantees, blended finance) and predictable procurement frameworks can crowd-in private capital at scale.

4.5.3 Lower middle income countries

Policy emphasis should balance reliability, affordability, and resilience, recognizing tighter fiscal constraints and infrastructure gaps. Priorities include:

Expanding least-cost renewables given local resource endowments, complemented by loss-reduction, grid reinforcement, and context-appropriate flexibility (including storage and demand-side measures);

Improving system operations and planning (data, forecasting, integrated resource planning) to reduce volatility and outage risks;

Leveraging concessional finance and risk-mitigation mechanisms to lower the cost of capital for clean energy assets. Where access deficits persist, modular solutions (mini-grids, distributed PV) can improve service quality while containing emissions growth. Diversification away from volatile imported fuels can further enhance resilience.

4.6 Limitations and future research

This study provides a novel perspective on assessing the relationship between CO₂ emission changes and energy system resilience, but several limitations remain.

First, the use of annual data constrains the temporal resolution of system dynamics. Annual averages may smooth short-term shocks and omit intra-year volatility that can be crucial for understanding resilience processes such as rapid absorption and rebound. Future work could incorporate higher-frequency datasets—including monthly or daily emission estimates, real-time electricity statistics, or mobility data—to capture dynamic adjustment patterns and temporal asymmetries in energy system responses.

Second, while our framework emphasizes structural stability and recovery at the national scale, energy system resilience is inherently multi-dimensional. Future research should integrate indicators reflecting institutional adaptability, technology innovation, and behavioral responses, alongside emission-based metrics. This could involve coupling CO₂ stability with socio-economic, governance, and infrastructure variables to construct a more comprehensive resilience index.

Third, the present analysis applies fixed 2023 income classifications to ensure comparability across time, but such static groupings may understate within-group heterogeneity and transitions between development stages. Incorporating dynamic classifications and policy-specific contexts would enhance the granularity of future comparative assessments.

Furthermore, infrastructure robustness is only indirectly proxied (via diversity and import exposure); direct engineering reliability metrics (e.g., outage indices, reserve margins, storage adequacy, interconnection capacity) are not included due to cross-country data gaps. Future research should integrate such indicators to more fully capture robustness. Sectoral heterogeneity is not quantified due to the lack of a harmonized, open-access global panel aligned with our coverage within the revision timeline. Mixing multiple sectoral definitions and licenses risks biasing cross-country comparisons. We therefore prioritize conceptual synthesis and case evidence here, and earmark a sectoral extension as future work.

Finally, while this study focuses on the COVID-19 shock, the proposed emission-stability framework could be extended to other systemic disruptions—such as geopolitical crises or extreme climate events—to evaluate how energy systems perform under diverse external stresses. Through these refinements, future resilience research can more effectively inform adaptive policy design and global energy transition strategies.

5 Conclusion

This study advances resilience research by introducing CO₂ emission stability as an operational proxy for national energy system resilience during a global crisis. The core finding is that, in response to this specific type of unexpected global shock, the stability of CO₂ emissions is associated with the robustness and adaptability of the under-lying energy system.

Our spatial-temporal analysis of CO₂ emission changes in 64 major economies before and after the COVID-19 outbreak revealed distinct regional and income-level patterns. High-income countries typically demonstrated greater emission stability, indicating stronger energy system resilience. In contrast, lower-income countries showed larger fluctuations in emissions, reflecting greater vulnerability. These variations highlight that emission stability can serve as a quantifiable, timely, and observable measure of energy system resilience in the face of sudden crises.

Moreover, the findings underscore the importance of integrating emission stability metrics into energy policy and resilience planning. As countries recover from the pan-demic and prepare for future shocks, the capacity to maintain stable carbon emissions — despite disruptions — can be a critical marker of sustainable and resilient energy systems.

Finally, this study provides a foundation for future research to apply emission stability as a practical resilience indicator in other types of crises. We observe that countries with more stable emissions tend to have higher resilience scores in our framework; thus, under similar public health shocks, emission stability may serve as a useful proxy. Our conclusions apply to the assessment of structural resilience at the annual scale. Analyses of short-term (weekly/monthly) shock absorption and rapid recovery dynamics require higher-frequency data and should be pursued in subsequent work.

Data availability statement

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

Author contributions

ZW: Software, Writing – original draft, Writing – review and editing, Visualization, Formal Analysis, Methodology. KY: Conceptualization, Writing – review and editing, Validation, Methodology, Visualization. BP: Writing – original draft, Conceptualization. YhZ: Writing – review and editing, Visualization, Formal Analysis, Software, Conceptualization. SC: Validation, Writing – review and editing, Writing – original draft. ZP: Writing – review and editing, Validation. YgZ: Writing – review and editing, Validation. MS: Methodology, Writing – review and editing. WD: Writing – review and editing, Visualization.

Funding

The author(s) declared that financial support was received for this work and/or its publication. This research was supported by National Natural Science Foundation of China (Grants Nos. 42530115, 42071381, 42161071).

Conflict of interest

The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Generative AI statement

The author(s) declared that generative AI was not used in the creation of this manuscript.

Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.

Publisher’s note

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.

References

Ahmadi, S., Saboohi, Y., and Vakili, A. (2021). Frameworks, quantitative indicators, characters, and modeling approaches to analysis of energy system resilience: a review. Renew. Sustain. Energy Rev. 144, 110988. doi:10.1016/j.rser.2021.110988

CrossRef Full Text | Google Scholar

Amiri, V., Rezaei, M., and Sohrabi, N. (2014). Groundwater quality assessment using entropy weighted water quality index (EWQI) in Lenjanat, Iran. Environ. Earth Sci. 72, 3479–3490. doi:10.1007/s12665-014-3255-0

CrossRef Full Text | Google Scholar

Association for Computing Machinery (2012). Resilience engineering: learning to embrace failure. Commun. ACM 55, 40–47. doi:10.1145/2076450.2076464

CrossRef Full Text | Google Scholar

Azzuni, A., and Breyer, C. (2018). “Definitions and dimensions of energy security: a literature review,”WIREs Energy and. Environ. 7. e268. doi:10.1002/wene.268

CrossRef Full Text | Google Scholar

Byrom, S., Bongers, G. D., Dargusch, P., Garnett, A., and Boston, A. (2020). A case study of Australia's emissions reduction policies - an electricity planner's perspective. J. Environ. Manag. 276, 111323. doi:10.1016/j.jenvman.2020.111323

PubMed Abstract | CrossRef Full Text | Google Scholar

Cherp, A., and Jewell, J. (2011). The three perspectives on energy security: intellectual history, disciplinary roots and the potential for integration. Curr. Opin. Environ. Sustain. 3, 202–212. doi:10.1016/j.cosust.2011.07.001

CrossRef Full Text | Google Scholar

Chuang, M. C., and Ma, H. W. (2013). Energy security and improvements in the function of diversity indices—Taiwan energy supply structure case study. Renew. Sustain. Energy Rev. 24, 9–20. doi:10.1016/j.rser.2013.03.021

CrossRef Full Text | Google Scholar

Dai, K., Bergot, A., Liang, C., Xiang, W.-N., and Huang, Z. (2015). Environmental issues associated with wind energy – a review. Renew. Energy 75, 911–921. doi:10.1016/j.renene.2014.10.074

CrossRef Full Text | Google Scholar

Davis, S. J., Liu, Z., Deng, Z., Zhu, B., Ke, P., Sun, T., et al. (2022). Emissions rebound from the COVID-19 pandemic. Nat. Clim. Change 12, 412–414. doi:10.1038/s41558-022-01332-6

CrossRef Full Text | Google Scholar

Dong, K., Dong, X., Jiang, Q., and Zhao, J. (2021). Assessing energy resilience and its greenhouse effect: a global perspective. Energy Econ. 104, 105659. doi:10.1016/j.eneco.2021.105659

CrossRef Full Text | Google Scholar

Faturechi, R., and Miller-Hooks, E. (2015). Measuring the performance of transportation infrastructure systems in disasters: a comprehensive review. J. Infrastructure Syst. 21, 04014025. doi:10.1061/(asce)is.1943-555x.0000212

CrossRef Full Text | Google Scholar

Friedlingstein, P., O'sullivan, M., Jones, M. W., Andrew, R. M., Gregor, L., Hauck, J., et al. (2022). Global carbon budget 2022. Earth Syst. Sci. Data 14, 4811–4900. doi:10.5194/essd-14-4811-2022

CrossRef Full Text | Google Scholar

Gatto, A., and Drago, C. (2020). Measuring and modeling energy resilience. Ecol. Econ. 172, 106527. doi:10.1016/j.ecolecon.2019.106527

CrossRef Full Text | Google Scholar

Gong, J. (2024). Framework for decentralized energy and enhanced resilience: islands and beyond. Stockholm, Sweden: Stockholm Environment Institute Report.

Google Scholar

Han, B., Liu, H., and Wang, R. (2015). Urban ecological security assessment for cities in the Beijing–Tianjin–Hebei metropolitan region based on fuzzy and entropy methods. Ecol. Model. 318, 217–225. doi:10.1016/j.ecolmodel.2014.12.015

CrossRef Full Text | Google Scholar

Han, P., Cai, Q., Oda, T., Zeng, N., Shan, Y., Lin, X., et al. (2021). Assessing the recent impact of COVID-19 on carbon emissions from China using domestic economic data. Sci. Total Environ. 750, 141688. doi:10.1016/j.scitotenv.2020.141688

PubMed Abstract | CrossRef Full Text | Google Scholar

He, P., Ng, T. S., and Su, B. (2019). Energy-economic resilience with multi-region input–output linear programming models. Energy Econ. 84, 104569. doi:10.1016/j.eneco.2019.104569

CrossRef Full Text | Google Scholar

Heffron, R. J., Körner, M.-F., Schöpf, M., Wagner, J., and Weibelzahl, M. (2021). The role of flexibility in the light of the COVID-19 pandemic and beyond: contributing to a sustainable and resilient energy future in Europe. Renew. Sustain. Energy Rev. 140, 110743. doi:10.1016/j.rser.2021.110743

PubMed Abstract | CrossRef Full Text | Google Scholar

Holling, C. S. (1973). Resilience and stability of ecological systems. Annu. Rev. Ecol. Syst. 4, 1–23. doi:10.1146/annurev.es.04.110173.000245

CrossRef Full Text | Google Scholar

Hosseini, S., Barker, K., and Ramirez-Marquez, J. E. (2016). A review of definitions and measures of system resilience. Reliab. Eng. and Syst. Saf. 145, 47–61. doi:10.1016/j.ress.2015.08.006

CrossRef Full Text | Google Scholar

Hotchkiss, E. (2023). Understanding resilience valuation for energy systems. Hamilton, United States: U.S. Department of Energy/NREL.

Google Scholar

Hughes, L. (2012). A generic framework for the description and analysis of energy security in an energy system. Energy Policy 42, 221–231. doi:10.1016/j.enpol.2011.11.079

CrossRef Full Text | Google Scholar

International Energy Agency (2015). Making the energy sector more resilient to climate change. Paris, France: IEA.

Google Scholar

International Energy Agency (2024). CO2 emissions in 2023 – analysis. IEA.

Google Scholar

Jasiūnas, J., Lund, P. D., and Mikkola, J. (2021). Energy system resilience – a review. Renew. Sustain. Energy Rev. 150, 111476. doi:10.1016/j.rser.2021.111476

CrossRef Full Text | Google Scholar

Jasiūnas, J., Lund, P. D., and Mikkola, J. (2025). Renewable and integrated energy system resilience – a review and generic resilience index. Renew. Sustain. Energy Rev. 150, 111476. doi:10.1016/j.rser.2025.115554

CrossRef Full Text | Google Scholar

Kemmler, A., and Spreng, D. (2007). Energy indicators for tracking sustainability in developing countries. Energy Policy 35, 2466–2480. doi:10.1016/j.enpol.2006.09

CrossRef Full Text | Google Scholar

Kharrazi, A., Sato, M., Yarime, M., Nakayama, H., Yu, Y., and Kraines, S. (2015). Examining the resilience of national energy systems: measurements of diversity in production-based and consumption-based electricity in the globalization of trade networks. Energy Policy 87, 455–464. doi:10.1016/j.enpol.2015.09.019

CrossRef Full Text | Google Scholar

Kruyt, B., Van Vuuren, D. P., De Vries, H. J. M., and Groenenberg, H. (2009). Indicators for energy security. Energy Policy 37, 2166–2181. doi:10.1016/j.enpol.2009.02.006

CrossRef Full Text | Google Scholar

Le Quéré, C., Peters, G. P., Friedlingstein, P., Andrew, R. M., Canadell, J. G., Davis, S. J., et al. (2021). Fossil CO₂ emissions in the post-COVID-19 era. Nat. Clim. Change 11, 197–199. doi:10.1038/s41558-021-01001-0

CrossRef Full Text | Google Scholar

Liu, Z., Ciais, P., Deng, Z., Lei, R., Davis, S. J., Feng, S., et al. (2020). Near-real-time monitoring of global CO2 emissions reveals the effects of the COVID-19 pandemic. Nat. Commun. 11, 5172. doi:10.1038/s41467-020-18922-7

PubMed Abstract | CrossRef Full Text | Google Scholar

Liu, Z., Deng, Z., Davis, S. J., Giron, C., and Ciais, P. (2022a). Monitoring global carbon emissions in 2021. Nat. Rev. Earth and Environ. 3, 217–219. doi:10.1038/s43017-022-00285-w

PubMed Abstract | CrossRef Full Text | Google Scholar

Liu, Z., Deng, Z., Zhu, B., Ciais, P., Davis, S. J., Tan, J., et al. (2022b). Global patterns of daily CO2 emissions reductions in the first year of COVID-19. Nat. Geosci. 15, 615–620. doi:10.1038/s41561-022-00965-8

CrossRef Full Text | Google Scholar

Liu, Z., Deng, Z., Davis, S., and Ciais, P. (2023). Monitoring global carbon emissions in 2022. Nat. Rev. Earth and Environ. 4, 205–206. doi:10.1038/s43017-023-00406-z

PubMed Abstract | CrossRef Full Text | Google Scholar

Manshadi, S. D., and Khodayar, M. (2016). “Resilient operation of multiple energy carrier microgrids,” in IEEE-Power-and-Energy-Society General Meeting (PESGM), 1. doi:10.1109/tdc.2016.7519947

CrossRef Full Text | Google Scholar

Martinez, D. M., and Ebenhack, B. W. (2008). Understanding the role of energy consumption in human development through the use of saturation phenomena. Energy Policy 36, 1430–1435. doi:10.1016/j.enpol.2007.12.016

CrossRef Full Text | Google Scholar

Mochizuki, J., and Chang, S. E. (2017). Disasters as opportunity for change: Tsunami recovery and energy transition in Japan. Int. J. Disaster Risk Reduct. 21, 331–339. doi:10.1016/j.ijdrr.2017.01.009

CrossRef Full Text | Google Scholar

Ntounis, N., Parker, C., Skinner, H., Steadman, C., and Warnaby, G. (2022). Tourism and hospitality industry resilience during the Covid-19 pandemic: evidence from England. Curr. Issues Tour. 25, 46–59. doi:10.1080/13683500.2021.1883556

CrossRef Full Text | Google Scholar

Ord, J. K., and Getis, A. (1995). Local spatial autocorrelation statistics: distributional issues and an application. Geogr. Anal. 27, 286–306. doi:10.1111/j.1538-4632.1995.tb00912.x

CrossRef Full Text | Google Scholar

Pan, S. (2025). The short-term impacts of COVID-19 outbreak on carbon emissions: causal evidence from China. Appl. Econ. Lett. 32, 1–7. doi:10.1080/13504851.2023.2266566

CrossRef Full Text | Google Scholar

Patlitzianas, K. D., Doukas, H., Kagiannas, A. G., and Psarras, J. (2008). Sustainable energy policy indicators: review and recommendations. Renew. Energy 33, 966–973. doi:10.1016/j.renene.2007.05.003

CrossRef Full Text | Google Scholar

Phillips, J., Finster, M., Pillon, J., Petit, F., and Trail, J. (2016). “State energy resilience framework”.

Google Scholar

Roege, P. E., Collier, Z. A., Mancillas, J., Mcdonagh, J. A., and Linkov, I. (2014). Metrics for energy resilience. Energy Policy 72, 249–256. doi:10.1016/j.enpol.2014.04.012

CrossRef Full Text | Google Scholar

Sato, M., Kharrazi, A., Nakayama, H., Kraines, S., and Yarime, M. (2017). Quantifying the supplier-portfolio diversity of embodied energy: strategic implications for strengthening energy resilience. Energy Policy 105, 41–52. doi:10.1016/j.enpol.2017.02.024

CrossRef Full Text | Google Scholar

Scarlat, N., Dallemand, J.-F., Monforti-Ferrario, F., Banja, M., and Motola, V. (2015). Renewable energy policy framework and bioenergy contribution in the european Union – an overview from national renewable energy action Plans and progress reports. Renew. Sustain. Energy Rev. 51, 969–985. doi:10.1016/j.rser.2015.06.062

CrossRef Full Text | Google Scholar

Schlör, H., Venghaus, S., Märker, C., and Hake, J.-F. (2018). Managing the resilience space of the German energy system - a vector analysis. J. Environ. Manag. 218, 527–539. doi:10.1016/j.jenvman.2018.04.053

PubMed Abstract | CrossRef Full Text | Google Scholar

Shah, S. A. A., Valasai, G. D., Memon, A. A., Laghari, A. N., Jalbani, N. B., and Strait, J. L. (2018). Techno-economic analysis of solar PV electricity supply to rural areas of Balochistan, Pakistan. Energies 11. doi:10.3390/en11071777

CrossRef Full Text | Google Scholar

Stirling, A. (2010). Multicriteria diversity analysis: a novel heuristic framework for appraising energy portfolios. Energy Policy 38, 1622–1634. doi:10.1016/j.enpol.2009.02.023

CrossRef Full Text | Google Scholar

Vera, I., and Langlois, L. (2007). Energy indicators for sustainable development. Energy 32, 875–882. doi:10.1016/j.energy.2006.08.006

CrossRef Full Text | Google Scholar

Wainstein, M. E., Dangerman, J., and Dangerman, S. (2019). Energy business transformation and Earth system resilience: a metabolic approach. J. Clean. Prod. 215, 854–869.

CrossRef Full Text | Google Scholar

Wang, D., Wan, K., and Yang, J. (2019). Measurement and evolution of eco-efficiency of coal industry ecosystem in China. J. Clean. Prod. 209, 803–818. doi:10.1016/j.jclepro.2018.10.266

CrossRef Full Text | Google Scholar

Wears, R. L. (2006). Resilience engineering: concepts and precepts. Qual. Saf. Health Care 15, 447–448. doi:10.1136/qshc.2006.018390

CrossRef Full Text | Google Scholar

Zhao, R., Fang, C., Liu, J., and Zhang, L. (2022). The evaluation and obstacle analysis of urban resilience from the multidimensional perspective in Chinese cities. Sustain. Cities Soc. 86, 104160. doi:10.1016/j.scs.2022.104160

CrossRef Full Text | Google Scholar