Carbon Capture in Indonesia’s Energy Sector: A Least-Cost Optimization Approach

Abstract

Indonesia’s power sector is heavily reliant on coal, making it a major contributor to greenhouse gas (GHG) emissions. This study evaluates the role of carbon capture (CC) as a transitional mitigation strategy using the Low Emissions Analysis Platform (LEAP) for least-cost optimization. Five scenarios up to 2060 are assessed: Business as Usual (BAU), a renewables-only pathway (NRE), two carbon-capture strategies (CALL and CNEW), and a hybrid scenario (COMB). Results show that NRE eliminates fossil power plants but increases system costs by 3.2% and raises reliability challenges due to the variability of solar generation. CALL achieves the lowest abatement cost (USD 0.93/tCO₂e) but leaves 105 Mt CO₂e residual emissions by 2060. COMB provides the most balanced outcome, cutting emissions by 96% (40 Mt CO₂e), increasing costs by only 1.9%, and ensuring energy security by combining CC with renewable expansion. These findings highlight that a hybrid strategy offers a pragmatic, least-cost pathway for Indonesia to align its power sector with net-zero targets while maintaining grid adequacy.

1. Introduction

Indonesia’s energy sector plays a pivotal role in the nation’s economic development but remains heavily reliant on coal-fired power, which supplies more than 60% of the country’s electricity [1]. This dependence on coal poses significant challenges for Indonesia’s pledge to reduce greenhouse gas (GHG) emissions by 31.89% domestically (or up to 43.2% with international support) by 2030 [2]. Although renewable energy (RE) sources such as solar, wind, hydro, and geothermal are seen as vital to long-term decarbonization, the country continues to face multiple obstacles in scaling them up. Recent analyses highlight grid constraints, financing barriers, and intermittency concerns as critical bottlenecks [3,4,5]. These hurdles slow Indonesia’s transition away from high-emission baseload power, limiting the feasibility of relying on renewables alone to meet its growing electricity demand.

Despite its vast renewable potential, particularly in solar, hydro, and geothermal, Indonesia continues to face multiple challenges in integrating RE into the national grid. Technologically, grid infrastructure remains underdeveloped, especially outside of Java and Sumatra, limiting the capacity to absorb variable renewable inputs [6]. Economically, high upfront costs and the absence of long-term power purchase agreements (PPAs) have discouraged private investment in utility-scale RE projects [3]. On the policy side, regulatory inconsistencies, local content requirements, and land acquisition delays often hamper project implementation [5]. These barriers have slowed the RE rollout despite ambitious targets, suggesting that relying solely on renewables may be insufficient to meet Indonesia’s emission reduction goals. As such, carbon capture emerges as a transitional technology to decarbonize the country’s still-dominant fossil plants while systemic barriers to RE are addressed.

Carbon capture and storage (CCS) has emerged as an increasingly prominent strategy to bridge near-term energy security requirements and climate goals. Globally, large-scale CCS facilities, such as Sleipner in Norway, Boundary Dam in Canada, and others in the United States and China, collectively capture millions of tonnes of CO₂ annually, illustrating the technology’s technical maturity [7,8]. One study indicates that CCS can reduce emissions by approximately 90% in fossil-fueled power generation, with potential cost savings for nations where coal or gas remains dominant [9]. Integrated assessment models (IAMs), such as those summarized by the IPCC and the Stanford Energy Modeling Study 27 (EMF27), confirm that pathways to 1.5 °C or 2 °C targets are significantly costlier, or infeasible, without large-scale CCS deployment [10,11].

Recent studies emphasize the growing role of CCS in low-carbon transitions, particularly its integration with renewable systems. Bioenergy with Carbon Capture and Storage (BECCS) is being piloted in the US and UK, offering both power generation and negative emissions [12]. Blue hydrogen, produced from natural gas or biomass with CCS, is advancing in Europe and Asia, with capture rates up to 90% and growing project pipelines [13,14]. In Southeast Asia, interest in CCS is gaining traction: Indonesia leads with over 15 CCS projects and newly established regulations for CO₂ storage and imports [15,16], while Malaysia is developing legal frameworks to host international CO₂ [17]. These trends are further supported by expanded global incentives, such as the U.S. Inflation Reduction Act and the EU Innovation Fund [18,19].

These developments confirm CCS as a critical complement to renewables in achieving global and regional decarbonization targets. While much of the recent literature focuses on technology-specific assessments or pilot-scale deployment, the present study contributes by evaluating the system-wide impacts of integrating CCS across different power sector scenarios in Indonesia, thereby complementing and extending existing analyses. This focus is particularly timely given that Indonesia has begun to translate such global momentum into domestic policy action.

The Indonesian government’s stance and policies on CCS have evolved significantly in recent years. CCS has been identified in national plans as one of the “technology options” to achieve deep emissions cuts, especially for the continued use of fossil fuels. For instance, Indonesia’s Enhanced Nationally Determined Contribution (ENDC) and long-term strategy documents mention “clean coal technology” and Carbon Capture (CC)/Carbon Capture, Utilization, and Storage (CCUS) as part of the toolkit for reducing emissions from power plants [2,20], alongside renewables and energy efficiency. Policy-wise, the government has begun to put enabling regulations in place. A key step was the issuance of Ministerial Regulation No. 2/2023 by the Ministry of Energy and Mineral Resources (MEMR), which specifically governs the implementation of CC/CCUS in the upstream oil and gas sector. This rule provides technical and legal guidance for injecting and storing CO₂ in oil/gas fields (e.g., how projects can recover costs, obtain permits, etc.). While aimed at oil and gas operations, it establishes a regulatory precedent for CCS in Indonesia. Building on this foundation, Presidential Regulation No. 14/2024 outlines a broader national framework for CC, including licensing and oversight mechanisms across multiple ministries. These policy developments create the institutional foundation for large-scale deployment models. Within Southeast Asia, Indonesia’s archipelagic geography and existing oil and gas basins present opportunities to develop CCS “hubs,” allowing multiple emitters to share pipeline networks and geological storage infrastructure [6,21].

Several studies have examined CCS and renewable pathways in Indonesia. Rahmanta et al. [22] and the World Bank [23] provide detailed simulations of post-combustion CCS retrofits for individual coal plants, quantifying energy penalties and cost increases but without system-level integration. Lau [24] and Putra et al. [25] extend the scope to retrofit feasibility and CO₂ transport optimization yet stop short of linking these insights to national power system transitions. Ramadhan et al. [26] identify geological formations for long-term storage, but do not evaluate their role in shaping system costs or emissions trajectories. Kanugrahan and Hakam [27] used LEAP to model Indonesia’s long-term renewable expansion under multiple scenarios, demonstrating feasibility but excluding CCS entirely. In a regional context, Chhay and Limmeechokchai [28] assessed Thailand’s power sector using LEAP, finding renewables and carbon pricing more effective than CCS, while Handayani et al. [29] projected that ASEAN could reach a least-cost net-zero pathway through renewables-dominant expansion by 2050. Although valuable, these studies remain fragmented: some are site-specific, others regional, and most omit a direct CCS–renewables comparison at the national scale. This study addresses that gap by evaluating CCS, renewables, and hybrid pathways for Indonesia’s power sector using a national-scale least-cost optimization framework. To further clarify the differences,

Table 1. Gap analysis of prior studies and contributions of this study.

Study	Scope/Method	Key Findings	Limitations	Gap Addressed in This Study
Rahmanta et al. [22]	CCS retrofit, IECM model	Capture efficiency, costs	Single-unit focus, no system perspective	Extend to national-scale impacts with LEAP
World Bank [23]	CCS in coal plants	Techno-economic penalties	Lacks multi-scenario policy evaluation	Integrates CC into system-wide pathways
Lau [24]	Retrofit feasibility	Prioritizes Java/Sumatra	Does not assess renewables trade-offs	Combines CC with RE expansion
Putra et al. [25]	Pinch design, CO2 transport	Optimized storage routes	No demand–supply integration	Adds system cost + emissions modeling
Kanugrahan & Hakam [27]	LEAP, RE-only scenarios	Shows RE feasibility	Excludes CCS	Provides CC vs. RE vs. hybrid comparison
Chhay & Limmeechokchai [28]	Thailand LEAP scenarios	CCS marginal, RE + carbon tax more effective	Country-specific, not Indonesia	Tests CC vs. RE trade-offs in national model
Handayani et al. [29]	ASEAN-wide, Applied Energy	RE-dominant pathway least-cost to 2050	Regional scope, lacks Indonesia detail	Provides Indonesia-specific CC–RE hybrid insights

summarizes prior studies, their limitations, and the specific contributions of this work.

Rather than focusing solely on a single unit or a small set of plants, this study evaluates CC across Indonesia’s national power mix under multiple scenarios. Specifically, it examines both standalone applications of CC in fossil-fueled power plants and its integration within a broader context of renewable energy expansion. This dual approach allows for a comprehensive assessment of CC’s potential role in Indonesia’s energy transition, addressing gaps identified in existing literature where most studies on CCS in Indonesia are either site-specific or considered only as part of broader energy system scenarios. By situating carbon capture within Indonesia’s broader context of rising electricity demand, fossil-fuel dependency, and ongoing challenges in renewable integration, this study employs LEAP to test multiple national-scale CCS scenarios, providing an evidence-based framework that aligns with national policy direction and offers actionable insights for policymakers, industry stakeholders, and researchers aiming to develop a balanced and economically viable strategy for deep power sector decarbonization.

This study makes several key contributions to the literature:

• National-scale CCS vs. RE vs. hybrid comparison: Provides the first least-cost optimization of Indonesia’s power sector that directly contrasts CCS-only, renewables-only, and hybrid pathways using the LEAP.
• System-wide evaluation of trade-offs: Goes beyond single-plant or regional studies by quantifying emissions, system costs, and reliability implications of alternative decarbonization strategies at the national scale.
• Integration of policy and storage context: Incorporates recent Indonesian regulations and geological storage estimates to ground results in the country’s evolving institutional and technical landscape.
• Policy-relevant insights: Identifies hybrid CCS–renewables strategies as the most balanced pathway toward net-zero, offering actionable guidance for policymakers seeking to balance affordability, energy security, and emissions reduction.

The remainder of this paper is organized as follows: Section 2 describes the methodology and modeling framework, including data sources and parameter settings; Section 3 presents the results of scenario simulations; Section 4 discusses power sector dynamics, climate alignment, energy security implications, and situates the findings within the broader literature; and Section 5 concludes with key insights and policy recommendations.

2. Materials and Methods

2.1. Overview

This study employs LEAP to model Indonesia’s energy system, with a specific focus on carbon capture (CC) as a mitigation strategy. While the overall modeling framework encompasses Indonesia’s entire energy system, the cost optimization was applied exclusively to the power sector, as outlined in Figure 1. The framework integrates government roadmaps, macroeconomic and demographic assumptions, historical energy data, and technology performance metrics to generate demand and supply projections. Although Indonesia spans multiple islands with diverse supply–demand dynamics, the analysis treats the country as a single regional system. A 2019 baseline was selected to ensure sufficient historical data for calibrating the demand and supply modules, with projections extending from 2022 to 2060 in alignment with Indonesia’s net-zero commitments [1]. Emissions are estimated using Intergovernmental Panel on Climate Change (IPCC) Tier-1 emission factors [30]. This framework enables a holistic assessment of outcomes such as emissions, system costs, and the roles of fossil versus renewable energy resources.

The analysis was conducted in optimization mode using the NEMO (Next Energy Modeling system for Optimization) framework within LEAP, linked to the CPLEX solver. Optimization was applied only to the power generation module (partial optimization), while other system components, demand projections, were modeled using LEAP’s simulation mode. The optimization simultaneously determines capacity expansion and dispatch, subject to annual electricity demand and technology-specific constraints. The objective function minimizes the net present value of system costs, including capital, O&M, and fuel expenditures, discounted at 10%. Constraints include demand-supply balance, capacity factors, plant efficiencies, and scenario-specific policies such as CC retrofits, renewable expansion targets, and fossil retirement schedules.

To enhance clarity, the main variables and scenario acronyms used in this study are summarized in

Table 2. Nomenclature of variables and scenarios used in the study.

Symbol/Abbreviation	Definition	Unit
Scenarios
BAU	Business as Usual	–
NRE	Non-CC + Renewable Expansion	–
CALL	Carbon Capture All	–
CNEW	New Carbon Capture-Equipped Plants	–
COMB	Combined CCS–Renewables	–
Model Variables
EL	Electricity demand	TWh
A	Activity level: Gross Domestic Product (GDP), households, vehicles, etc.	–
I	Energy intensity (electricity per unit of activity)	kWh/unit
$C_{j,t}^{cap}$	Capital cost of power plant	USD/MW
$C_{j,t}^{OM}$	O&M cost	USD/MW-year
$C_{j,t}^{fuel}$	Fuel cost	USD/year
r	Discount rate	%
t	Simulation year	Year
j	Power plant type (coal, gas, solar, etc.)	–

.

2.2. Demand and Supply Modeling

On the demand side, electricity consumption is disaggregated into four sectors: industry, residential, transportation, and commercial. The demand projection in the model follows a standard formulation where the Electricity demand (EL) is a product of activity (A) and energy intensity (I), written as follows:

$${EL}_{sector}=A_{sector}\times I_{sector}$$

(1)

This structure allows for a disaggregated, sector-specific approach to projecting future electricity demand. Each demand sector in the model is associated with a unique activity indicator, as seen in

Table 3. Activity data.

Demand Sector		Activity
Industry		Sectoral GDP
Transportation	Road	Number of vehicles
	Railway	Passenger-km or Tonne-km
Residential		Household number
Commercial		Sectoral GDP

. For example, the activity levels of the industry and commercial sectors are linked to GDP, reflecting economic output as the primary driver of demand. In the residential sector, activity is measured by the number of households, which is influenced by population growth and household size. The activity indicator for the transportation sector is captured through passenger kilometers traveled. Particularly for road transport, it is further disaggregated by the number of vehicles.

The projections for energy intensity were developed by extrapolating historical data. The model incorporates several key macroeconomic and demographic assumptions from official sources [31,32,33,34]. GDP growth is projected at an average of 5.4% per year. Population growth is expected to average 0.56% annually, reflecting a shift toward an ageing population. The average household size is assumed to remain around 3.8 persons per household.

On the supply side, the model incorporates existing and potential electricity-generation technologies, ranging from coal and natural gas units to various renewable energy sources (solar, wind, hydro, geothermal), to simulate capacity expansions and retirements. The supply calculations are demand-driven with optimization only in the power sector, allowing the model to endogenously determine capacity expansion and dispatch of power plants that satisfy projected electricity demand at the lowest system cost. This is achieved using the CPLEX solver, which identifies the least-cost mix of technologies while minimizing the total net present cost of the electricity generation system over the study period, subject to constraints such as energy demand and fuel availability. The general formulation of the objective function used for optimization is:

$$min\sum _t\sum _j\left(C_{j,t}^{cap}+C_{j,t}^{OM}+C_{j,t}^{fuel}\right){\displaystyle \frac{1}{{(1+r)}^t}}$$

(2)

where

• $C_{j,t}^{cap}$ is the capital cost;
• $C_{j,t}^{OM}$ is the fixed and variable operating and maintenance (O&M) cost;
• $C_{j,t}^{fuel}$ is the annual fuel cost;
• r is the annual discount rate (10%);
• t is the simulation year (2022–2060);
• j is the power plant type.

Input parameters, such as capital and O&M costs, fuel prices, and plant efficiencies, are primarily sourced from the MEMR [35].

2.3. Carbon Capture Parameter Settings

The capital and operational cost assumptions for carbon capture in

Table 4. CC data relative to conventional plant baselines [35].

Parameter	Coal Power Plant		Gas Combined Cycle Power Plant
	2030	2050	2030	2050
Electricity efficiency drop 1 (%)	−8	−8	−8	−8
CO2 emission reduction (%)	−90	−90	−90	−90
Capital Cost (Million USD/MW)	+1.79	+1.42	+0.97	+0.75
Fixed O&M (USD/MW/year)	+40,500	+39,300	+8700	+8500
Variable O&M (USD/MWh)	+3.01	+2.91	+1.16	+1.13

¹ Percentage ratio of electricity output (not including any co-product energy recovered) to fuel consumption.

, covering both coal and gas-fired power plants, are based on the Technology Data for the Indonesian Power Sector catalogue, published by the MEMR [35]. The reported figures represent the additional investment costs relative to standard, non-CC power plants, not total plant costs. For coal-fired units, integrating CC increases capital costs by approximately USD 1.79 million per MW in 2030, declining to USD 1.42 million per MW by 2050. For gas combined-cycle plants, the added cost is USD 0.97 million per MW in 2030, decreasing to USD 0.75 million per MW in 2050. These cost reductions follow a one-factor learning curve with a 12.5% rate, aligned with global expectations for gradual cost decline as deployment scales up.

In terms of performance, the catalogue assumes a capture efficiency of 90% for both coal and gas applications, alongside an 8-percentage-point net efficiency penalty due to energy required for solvent regeneration and CO₂ compression. This reflects commercially available post-combustion capture systems, primarily using amine scrubbing. Fixed and variable O&M costs are also increased, with adjustments for solvent use, auxiliary power loads, and system integration complexity. The investment costs include the full engineering, procurement, and construction (EPC) scope for CC integration but exclude CO₂ transport and storage infrastructure, which is highly location-dependent and beyond the scope of this model. Cost projections assume average global deployment levels based on IEA’s Stated Policies and Sustainable Development scenarios and have been validated through international literature comparisons.

2.4. Scenario Configurations

Five scenarios were developed to explore how Indonesia’s electricity mix might evolve under varying levels of CC and renewable adoption. Each scenario aligned with existing policy directives such as the National Electricity Supply Business Plan (RUPTL) [36] and ENDC [2]. They include a Business as Usual (BAU) trajectory, an aggressive renewables-focused pathway, two CC-specific approaches, and a hybrid scenario combining CC and a renewables strategy.

The BAU scenario represents a straight-line continuation of Indonesia’s current electricity policy, as outlined in the 2021–2030 RUPTL. Fossil units retire only at the end of their technical lifetimes, and no new emissions constraints are applied. Capacity additions of all power plants after 2030, including renewables, are determined endogenously by LEAP’s least-cost optimization. The model can select any combination of coal, gas, or renewable technologies that minimizes discounted system cost while meeting projected demand and policy constraints already embedded in the RUPTL. The BAU scenario is the reference case against which all other scenarios’ economic and environmental impacts are measured.

By contrast, the NRE (Non-CC + Renewable expansion) scenario imposes an aggressive clean-energy mandate. Starting in 2030, fossil plant retirements will significantly increase, with a complete phase-out of all fossil plants by 2060. In addition, carbon-capture technologies are prohibited. These restrictions force the optimization routine to rely exclusively on renewable options to satisfy demand, driving an accelerated renewable capacity expansion. As a result, the power system will become entirely renewable and achieve zero emissions by 2060.

Government briefings indicate that CCS deployment could begin as early as 2035 for coal plants and 2045 for gas plants, although no binding regulation has been issued. Reflecting this guidance, the three CC-focused scenarios adopt the following setting:

• CALL (CC All): A linear increase in CC installation on the BAU fossil plants, with CC-coal commencing in 2035 and CC-gas in 2045, reaching complete coverage by 2060. This approach includes retrofitting the existing fossil plants.
• CNEW (New CC-Equipped Plants): Introducing new CC-coal plants after 2035, and new CC-gas plants after 2045. These new CC-plants are different from BAU. This approach excludes retrofitting in the existing fossil plants. While not based on any specific national policy, this scenario is designed to isolate the systemic effects of relying exclusively on new CC capacity, and to contrast with CALL, where CC is introduced into the BAU mix via both retrofits and new builds. This approach enables a clean comparison between technology sequencing and capital investment strategies.
• The COMB (Combined CC-Renewables) scenario merges the CC installation on the BAU fossil plants of CALL with the new-build rules of CNEW while simultaneously accelerating renewable expansion. From 2035 onward, existing coal plants are progressively equipped with CC, and any additional capacity must be CC-equipped. After 2045, the same rule applies to gas plants. Non-CC fossil plants are forced to retire by 2060.

Details of all five scenarios and their key design features are presented in

Table 5. Scenario’s details.

Scenario	Description	CC Implementation	Fossil Fuel Power Plant Policy
BAU	Continuation of existing policies without significant changes	Not applied	Coal-dominated, with minimal retirements following RUPTL
NRE	Zero emissions in the power sector by 2060 with renewable expansion	CC implementation is prohibited	Enforce the retirement of all fossil plants starting in 2030 and complete the phase-out by 2060
CALL	New and existing coal and gas plants in BAU must progressively adopt CC technology over time, including retrofitting.	CC-coal share: 0% in 2035, increase to 100% in 2060. CC-gas share: 0% in 2045, increase to 100% in 2060	Enforce the retirement of non-CC plants starting in 2030 and complete the phase-out by 2060
CNEW	Deploying new coal and gas plants with integrated CC. The new plants are different from the ones in BAU. No retrofitting.	New CC-coal plants after 2035 and new CC-gas plants after 2045	Enforce the retirement of non-CC plants starting in 2030 and complete the phase-out by 2060
COMB	Integrates CC on BAU fossil plants (CALL) with new CC-equipped plants outside BAU (CNEW) and accelerates renewables (NRE)	Combination of CALL and CNEW, excluding NRE’s CC prohibition. CC-plants’ capacity is constrained by CALL parameters (for BAU plants) and CNEW parameters (for new CC plants)	Enforce the retirement of non-CC plants starting in 2030 and complete the phase-out by 2060

.

2.5. Limitation

While the methodology provides a robust mechanism for comparing various carbon capture and renewable energy strategies, several limitations should be noted. First, the LEAP model used in this study simplifies regional heterogeneity by treating Indonesia as a single zone, potentially overlooking local infrastructure constraints and renewable resource variability. Second, the storage capacity of CC is not explicitly modeled, and the analysis omits the costs associated with transporting and storing captured CO₂. This omission may underestimate the full lifecycle cost of CC, particularly in regions requiring long pipeline routes or offshore storage. Future studies should incorporate detailed geo-spatial assessments of emitter-to-reservoir distances and infrastructure needs to refine these costs further. Third, while the main model scenarios use fixed capital cost projections and technical parameters (such as a 90% CO₂ capture rate and an 8% efficiency penalty for CC, and a 2% annual solar cost decline), the robustness of these assumptions has been tested through a dedicated sensitivity analysis. Fourth, local air quality impacts and the social cost of carbon are not considered. Despite these limitations, the modeling framework provides meaningful insights into how Indonesia’s power sector might optimize its electricity mix under diverse decarbonization pathways, offering valuable guidance for policymakers and energy planners.

3. Results

3.1. Electricity Demand Projection

To confirm the reliability of the model outputs, the BAU scenario’s short-term results are compared with historical data for 2019–2021. This process verifies that fuel consumption, capacity factors, and dispatch patterns remain consistent with actual records from MEMR [1]. Looking forward, as shown in Figure 2, Indonesia’s electricity demand is projected to grow rapidly. The total demand rises roughly 5.1% annually, exceeding 2000 TWh by 2060. This trajectory mirrors regional expectations: Southeast Asian demand is forecast to increase about 4% annually through 2035, driven by rising incomes and manufacturing expansion [37].

Data for 2020 underscore the sectoral starting point: residential consumption was 112.1 TWh, industry 89.5 TWh, commercial 58.2 TWh, and transportation only 0.28 TWh. Echoing regional trends, industrial demand has overtaken residential as the main growth engine, reflecting Indonesia’s shift toward higher-value manufacturing. Power-intensive subsectors, like food and beverages, chemical fertilizers, and textiles, are expected to expand significantly, pushing industrial electricity use to almost 900 TWh by 2060, or 5.6% per year.

Transportation electrification, by contrast, starts from a negligible base and, despite a 5.2% annual growth rate projected under the government’s Electric Vehicle (EV) roadmap, still reaches less than 2.5 TWh in 2060. Infrastructure and policy gaps thus keep transport demand marginal relative to other sectors [38,39]. These trends highlight the strategic importance of deploying CC-equipped coal and gas plants to supply the large, steady industrial load while curbing power sector emissions.

3.2. Scenario Outcome

A central insight emerging from Figure 3 and Figure 4, as well as

Table 6. Power generation cost.

Scenario	Cumulative Emission Reduction (A)	Cumulative Cost Increase 1 (B)	Average Emission Reduction Cost (B/A)
	Billion Tonne CO2e	Billion USD	USD/Tonne CO2e
NRE	11.35 (−54%)	41.42 (+3.2%)	3.65
CALL	7.11 (−34%)	6.63 (+0.5%)	0.93
CNEW	9.55 (−45%)	19.74 (+1.5%)	2.07
COMB	12.40 (−59%)	23.72 (+1.9%)	1.91

¹ Discount rate 10%.

, is the markedly different trajectories in emissions and associated costs among the modeled scenarios. On one end of the spectrum, the BAU scenario sustains a coal and gas-intensive mix with minimal retirements, where fossil plants’ output rises from around 240 TWh in 2020 to over 1700 TWh in 2060, preserving a 75% share of total generation. At the other extreme, NRE (Non-CC + Renewables) shows a swift and sustained transition away from fossil fuels. NRE is forcing the early retirement of fossil plants by 2060, replacing them mainly with solar photovoltaic (PV) (35%), other renewable (35%), and hydro (21%). The CC scenarios maintain the BAU dominance of fossil plant production in 2060, but CNEW has a bigger share of CC-coal power plant (40%) compared to CALL (22%). The hybrid COMB scenario yields the most balanced portfolio, with renewables at 64%, CC-coal at 4%, and CC-gas at 32%.

These divergent supply trajectories translate into different emission outcomes (Figure 4). The BAU scenario culminates in the highest CO₂ emissions, while NRE managed to reach zero emissions by 2060. Although they cannot reach zero emissions by 2060, the CC scenarios reduce CO₂ emissions significantly, 90% for CALL and 86% for CNEW. The combination of renewable expansion and CC implementation in the COMB scenario reduces 96% of emissions. Adding further nuance to these comparisons, recent data on Indonesia’s CO₂ storage potential provide a critical context for assessing the feasibility of large-scale CC deployment. Indonesia has an estimated CO₂ storage potential of approximately 680.57 billion tonnes in deep saline aquifers and 10.14 billion tonnes in oil and gas fields, yielding a total storage capacity of roughly 690.71 billion tonnes [40]. This high storage potential far exceeds the cumulative CO₂ captured in any of the modeled scenarios (Figure 4), which are 9.17 billion tonnes for CALL, 16.15 billion tonnes for CNEW, and 4.81 billion tonnes for COMB, respectively.

The total discounted power sector cost under the NRE scenario rises 3.2% from BAU, largely due to accelerated fossil plant retirements and substantial renewable additions. CALL and CNEW increase costs by 0.5% and 1.5%, respectively; COMB lies in between at 1.9%. The resulting average emission reduction costs per tonne CO₂e are USD 0.93 for CALL (the lowest), followed by USD 1.91 for COMB, USD 2.07 for CNEW, and USD 3.65 for NRE (Table 6).

3.3. Regression-Based Emissions Modeling

Regression analysis is performed for each scenario to investigate the relationship between total CO₂ emissions (dependent variable) and total electricity production (independent variable). The objective is to minimize the error term between the total simulated and predicted dependent variables [41], as represented by the following equation:

$$\mathit{min}Error=\sum _{t=2019}^{2060}{({C{O}_{2}emission}_t^{Simulation}-{C{O}_{2}emission}_t^{Estimated})}^2$$

(3)

where ${C{O}_{2}emission}_t^{Simulation}$ represents the total CO₂ emission data obtained from Figure 4, and the estimated CO₂ emission is modeled using a polynomial equation:

$${C{O}_{2}emission}_t^{Estimated}({MtCO}^{2}e)=\sum _{i=1}^4{a}_i\times {(ElectricityProduction\left[TWh\right])}^i$$

(4)

Figure 5 presents the total annual CO₂ emission values against total annual electricity production, with simulated values shown as dots and estimated values represented by curves. In all scenarios, the most accurate estimation models were those of polynomial order two and three.

Table 7. Estimation Model Parameters for Each Scenario.

Scenario	Fitted Model Order (Without Intercept)	Fitted Model Parameters	Accuracy Measure (R-Squared)
BAU	Quadratic polynomial	a1 = 0.6764; a2 = −0.0001	0.9986
NRE	Cubic polynomial	a1 = 0.8851; a2 = −0.00074; a3 = 0.00000016	0.887
CALL	Cubic polynomial	a1 = 0.8375; a2 = −0.00062; a3 = 0.00000013	0.700
CNEW	Quadratic polynomial	a1 = 0.7731; a2 = −0.0003	0.9848
COMB	Cubic polynomial	a1 = 9480; a2 = −0.0009; a3 = 0.0000002	0.9096

provides the parameter details for the estimation models, where the BAU and CNEW scenarios are fitted with quadratic polynomial models, while the CALL, NRE, and COMB scenarios are fitted with cubic polynomial models. For all scenarios, the accuracy metrics (R-squared values) exceed 0.7, indicating a high degree of accuracy in the estimated models.

The regression illustrates how emissions respond to electricity generation under different policy scenarios. In the BAU case, emissions increase steeply with generation, reflecting a direct, near-linear relationship due to continued reliance on unabated fossil fuels. The CNEW scenario shows a similar trajectory but with a slight flattening, indicating that while new CC-equipped plants reduce emissions intensity over time, the lack of retrofits limits early reductions. CALL exhibits a more complex curve: emissions decline more significantly mid-century as retrofitting scales up, but the rate of reduction diminishes toward 2060, suggesting limited long-term gains once retrofit potential is saturated.

The NRE scenario, relying heavily on renewables, achieves a sharp early decline in emissions, but the curve exhibits mild fluctuations as generation increases, potentially due to residual fossil backup during peak demand. In contrast, the COMB scenario displays the most consistent downward trend in emissions relative to generation. The model suggests a strong decoupling between electricity growth and emissions, enabled by the combined effects of renewables and carbon capture. Overall, these results show that while each scenario affects emissions differently, only COMB maintains a steady emissions decline across increasing electricity demand, highlighting its robustness for long-term decarbonization. These fitted models can serve as compact mathematical representations of system-level emissions dynamics and may be useful for high-level policy simulations or integration with other decision-support tools.

The regression models presented above are intended primarily as illustrative, high-level approximations of the relationship between total CO₂ emissions and electricity generation across scenarios. Such functional forms can be useful for rapid, top-down policy analysis, scenario screening, or as modules in integrated assessment models when computational efficiency or data availability is limited. However, it is important to emphasize their limitations:

-

Temporal and Structural Oversimplification: These regressions abstract away the temporal dynamics, technology mix shifts, and system constraints that drive emissions in the full scenario simulations (as shown in Figure 4). They cannot capture year-on-year fluctuations, policy shocks, or inflection points due to technology breakthroughs or retirements.

-

Scenario-Specific Validity: The fitted polynomials are scenario-dependent and may not generalize to other policy cases or future energy mixes.

-

No Causality or Mechanistic Insight: The models are purely statistical and do not provide a mechanistic understanding of mitigation pathways.

While these regressions may facilitate rapid estimation or high-level sensitivity analyses, they are not substitutes for detailed system modeling when evaluating specific policy options or planning transitions. We present them here to illustrate broad trends and as potential tools for simplified policy analysis, but caution that they should be applied only with full awareness of their simplifying assumptions.

4. Discussion

Chapter 4 presents four interrelated analyses that contextualize the scenario outcomes within broader policy, technical, and scholarly frameworks. Section 4.1. (Power Sector Dynamics) examines how technology sequencing shapes the evolution of Indonesia’s generation mix, emissions trajectory, and mitigation costs. Section 4.2. (Climate Target Alignment) evaluates the extent to which each pathway aligns with international decarbonization benchmarks, highlighting technical levers for deeper emissions reductions. Section 4.3. (Energy Security) analyzes trade-offs between carbon mitigation and system reliability, emphasizing the role of domestic fuels, firm capacity, and infrastructure readiness. Section 4.4. (Comparison with Literature) situates this study within the existing body of national and regional research, clarifying its contributions and identifying areas for further investigation.

4.1. Power Sector Dynamics

4.1.1. Technology Sequencing and Trade-Offs

In 2060, the CNEW scenario exhibits a higher share of CC-coal relative to CC-gas than CALL, even though CC-gas is 10–15% cheaper on a levelized-cost basis [35]. This counterintuitive outcome is rooted in technology-sequencing assumptions. CNEW allows new CC-coal builds from 2035, a full decade before CC-gas (2045). By 2040, CC-coal will already supply a significant share of power generation, establishing supply chains and operator learning that reinforce its position. This early foothold endogenously influences subsequent expansion decisions, a classic illustration of path dependence and first-mover advantage in energy-system optimization. As noted by Davis et al. [42], early investments in specific technologies can create self-reinforcing mechanisms that make transitions to alternatives increasingly difficult. Similarly, Unruh [43] introduces the concept of carbon lock-in, where early adoption of a technology, such as CC-coal, can generate technological and institutional momentum through established supply chains, regulatory alignment, and operational experience, discouraging later alternatives. When CC-gas finally becomes available, much of the demand is already met, limiting its uptake. In CALL, CC-coal expansion is still limited by 2040, so the final shares of CC-coal and CC-gas reflect underlying cost differentials more closely.

Because of this higher coal share, CNEW’s residual emissions are more significant in 2060, even though total generation from CC-equipped units is almost identical in the two scenarios. Coal’s higher carbon content (94 kg CO₂/GJ) compared with natural gas (56 kg CO₂/GJ) means that 10% of flue gas that escapes capture leaves a larger emissions footprint per kilowatt-hour [9,30]. Consequently, CNEW records 142 Mt CO₂e, about 35% above the 105 Mt CO₂e in CALL by 2060. These dynamics highlight the policy importance of technology-timing rules: granting one option a long head start can crowd out cheaper, cleaner choices that arrive later, with lasting cost and emissions consequences.

Implementing CC in BAU’s fossil plants (CALL) delivers the lowest emission reduction cost (USD 0.93/t CO₂e). Reliance on the CC-equipped plants nonetheless carries three structural risks. First, it can impede the ramping flexibility required in a high-renewables system. Second, residual emissions expose plants to future carbon pricing. Third, continued coal use may invite border-adjustment duties in key export markets, mirroring the European Union’s Carbon Border Adjustment Mechanism (CBAM). The COMB scenario, with the second-lowest cost, mitigates these risks by pairing CC plants with rapid renewable expansion, creating a smooth path for coal retirements while preserving adequacy.

Additionally, the current model excludes costs related to CO₂ transportation and storage. Recent studies estimate pipeline-based CO₂ transport in Indonesia is on the order of a few dollars per tonne (for moderate distances), and geological CO₂ storage in suitable reservoirs costs on the order of $10–20 per tonne [44,45,46]. To illustrate the potential impact, we apply a conservative estimate of USD 20 per tonne CO₂ captured to each CC scenario. This adjustment raises the cumulative system cost in CALL from USD 6.63 billion to USD 189.94 billion, in CNEW from USD 19.74 billion to USD 342.71 billion, and in COMB from USD 23.72 billion to USD 119.87 billion. As a result, the average emission reduction cost increases substantially: from USD 0.93/tCO₂e to USD 26.72/tCO₂e in CALL, from USD 2.07/tCO₂e to USD 35.89/tCO₂e in CNEW, and from USD 1.91/tCO₂e to USD 9.67/tCO₂e in COMB.

These adjusted costs remain within or below global benchmarks. A review by Bui et al. [9] suggests that typical CC abatement costs range from USD 20–120/tCO₂e, depending on fuel type, plant configuration, and region. Similarly, the IEA [47] reports that current CC costs for coal and gas power plants fall in the USD 40–90/tCO₂e range under standard conditions. Compared to these international figures, the revised costs, particularly for the COMB scenario, indicate that carbon capture in Indonesia can be both cost-effective and globally competitive. The relatively low figures reflect Indonesia’s heavy reliance on existing fossil infrastructure, potential for shared CC hubs, and favorable site conditions for early-stage CC deployment. These findings highlight the feasibility and economic viability of scaling up CC in Indonesia’s power sector, especially when combined with strategic renewable expansion.

To test the robustness of the main findings, an optimistic sensitivity analysis was conducted with 20% lower CC capital costs by 2050, a higher CO₂ capture rate (95% versus 90%), and an accelerated annual solar capital cost decline (3% versus 2%). Under these assumptions, only minor decreases in cumulative system cost are observed for all scenarios, with the most noticeable reductions in NRE (from 3.2% to 2.1%) and COMB (from 1.9% to 1.6%). Cumulative emission reductions remain nearly unchanged. As a result, the average emission reduction cost declines more substantially, with NRE falling from 3.65 to 2.4 USD/tCO₂e and COMB from 1.91 to 1.68 USD/tCO₂e. These findings indicate that scenario rankings and qualitative conclusions are robust to improved technology and cost assumptions: technological advances primarily enhance cost-effectiveness without significantly altering overall mitigation outcomes or the relative performance of key pathways.

Cost trajectories for both carbon capture and renewable technologies play a crucial role in shaping long-term system outcomes. The current model uses official projections from the MEMR, which assume moderate cost declines for both CC and renewables. However, global experience suggests that cost reductions for renewables, especially solar and battery storage, are likely to outpace those for CC due to greater deployment scale and steeper learning curves [6,9]. For example, utility-scale solar PV costs have declined by over 80% since 2010, whereas post-combustion CC cost reductions have remained modest due to limited commercial rollout. Nonetheless, ongoing pilot projects and policy incentives, such as the U.S. Inflation Reduction Act and EU carbon pricing reforms, may accelerate cost declines for CC technologies in the coming decades. Future research could explore probabilistic cost modeling or sensitivity analyses to assess the robustness of scenario outcomes under different technology price trajectories.

4.1.2. Regional Feasibility and Infrastructure Constraints

While a national-scale LEAP model provides valuable system-wide insights, Indonesia’s pronounced regional disparities can significantly affect the technical and economic feasibility of CC deployment, RE integration, and grid reliability. Several studies highlight the importance of these regional differences. Java and Sumatra host over 80% of Indonesia’s coal-fired power capacity and are proximate to major sedimentary basins with significant CO₂ storage potential, such as South Sumatra, Central Sumatra, and West Java, each with capacities exceeding 7–40 Gt [40,44]. Recent World Bank and ERIA studies have shown that power plants in West Java and South Sumatra can be retrofitted with CC at relatively low Transport and Storage (T&S) costs, with pipeline distances often under 100 km and estimated CCS costs (capture plus T&S) of approximately US$71–100 per tonne of CO₂ [23,44]. In contrast, coal power plants in eastern Indonesia are more dispersed, and storage sites are much less accessible; for example, basins in Eastern Indonesia such as Bintuni and Salawati offer limited capacity and are often far from major emission sources [40,44]. Although eastern Indonesia possesses significant theoretical storage potential, there is little detailed T&S cost data available, and the region’s limited number of large emission sources means that transport distances from power plants to suitable storage sites would be much greater. This would likely require long-distance pipelines or even marine transport, making overall CCS costs in the eastern region substantially higher than in Java or Sumatra. As a result, the lack of cost-effective proximity between emitters and storage sites poses a major challenge for early or large-scale CCS deployment in eastern Indonesia, despite the headline storage volumes [40].

The large-scale deployment of CC scenarios in Indonesia also hinges on the development of T&S infrastructure, which is currently at an early stage. Unlike countries with established pipeline networks (e.g., the U.S. Gulf Coast), Indonesia would need to develop new CO₂ corridors to connect point sources, such as coal plants in Java and Sumatra, with geological storage sites in deep saline aquifers or depleted oil fields in East Kalimantan and South Sumatra [26,40]. This requires not only significant capital investment but also coordinated permitting across energy, land use, and environmental ministries. For CALL, the retrofitting of existing fossil plants may be constrained by the proximity to viable storage sites or right-of-way issues for pipeline development. CNEW has more spatial flexibility, as new CC-equipped plants could be strategically sited near reservoirs. COMB inherits challenges from both retrofits and new builds, but benefits from diversified infrastructure planning. Regardless of pathway, the creation of shared “CC hubs” with centralized storage and pipeline infrastructure, similar to models under development in Southeast Asia and Europe, could offer economies of scale. However, unlocking these benefits will require proactive regulation, long-term investment frameworks, and cross-sector coordination to minimize delays and cost escalation. Notably, these infrastructures and permitting challenges are likely to be even more pronounced in eastern Indonesia due to greater distances and less-developed infrastructure.

Renewable integration also faces pronounced regional differences. Grid capacity in Java-Bali enables absorption of higher shares of variable renewables (with up to 20 GW of planned solar and geothermal additions), while Eastern Indonesia’s grid remains fragmented, with limited interconnections and lower system inertia [6,36]. As a result, the renewable penetration limits in the east may be significantly lower without substantial new investment in transmission, storage, and grid stabilization technologies [6]. For instance, even under optimistic assumptions, the technical limit for solar PV in Sulawesi or Papua may be capped at 20–30% of local demand due to curtailment and balancing constraints, compared to 40–50% in Java-Bali [6,48].

By treating Indonesia as a single region, the model likely underestimates both the technical and economic challenges of CC deployment and renewable integration in less-developed areas. The direction of bias is toward optimistic estimates of nationwide feasibility and cost-effectiveness. In reality, achieving comparable decarbonization in all regions would require differentiated investment, targeted subsidies, and region-specific grid upgrades. Moreover, even in the most accessible western basins, local siting, social acceptance, and permitting issues may further limit the share of storage capacity that can be practically utilized by 2060, underscoring the need for proactive planning and stakeholder engagement alongside technical assessments. Future studies using a multi-region or nodal model could address these gaps and better inform the spatial allocation of CC and renewables investments.

4.1.3. Strategic Implications and Policy Priorities

A closer comparison of the three CC scenarios, CALL, CNEW, and COMB, reveals distinct trade-offs between economic efficiency and environmental effectiveness. CALL, which includes retrofitting existing fossil plants with CC, yields the lowest abatement cost at USD 0.93/tCO₂e (excluding T&S), due to its reliance on already-built infrastructure and phased implementation. However, it fails to meet zero emission by 2060, due to the residual emissions from retrofitted coal and gas units. CNEW, which builds new CC-equipped fossil plants, reduces emissions more effectively but at double the cost (USD 2.07/tCO₂e), as new plants incur higher capital costs. COMB offers a middle ground: by combining retrofits and new builds while scaling renewables, it achieves the highest emissions reduction (96%) with a moderate cost increase (USD 1.91/tCO₂e).

Indonesia’s policy environment for CC has advanced with the issuance of Presidential Regulation No. 14/2024, which outlines the governance of CC and CCUS activities. This regulation establishes a basic framework for permitting, licensing, and project approval, but several concrete hurdles and enablers remain that directly affect the feasibility of the modeled scenarios:

-

Licensing and Permitting:

Presidential Regulation No. 14/2024 requires all CC/CCUS projects to secure permits from the MEMR, with input from the Ministry of Environment and Forestry and other relevant agencies. For retrofits (CALL), this means that existing power plants must undergo a separate, potentially time-consuming licensing process for each retrofit project, including detailed environmental and technical assessments. Scenario CNEW face similar approval steps but may benefit from streamlined permitting if projects are sited in government-designated CC “hubs” or industrial clusters.

-

Cross-Ministerial Coordination:

The regulation mandates coordination across MEMR, the Ministry of Finance, and the Ministry of State-Owned Enterprises, but does not yet provide a clear “one-stop-shop” for project developers. This fragmentation may cause delays in project timelines, especially for large-scale transport and storage infrastructure needed for shared CO₂ hubs (see Section 4.1.2).

-

Cost Recovery and Investment Incentives:

Presidential Regulation No. 14/2024 allows for the recovery of approved CC project costs, but the mechanisms for determining eligible expenses, returns on investment, and risk-sharing between government and private investors are not yet fully defined. For CALL, a lack of clarity in cost recovery may slow the retrofit rate, as private utilities are reluctant to make capital-intensive upgrades without guaranteed returns. For CNEW and hub projects, the regulation provides for special fiscal incentives (e.g., tax holidays, import duty exemptions), but these are not automatic and require additional applications.

-

Timelines and Long-Term Planning:

The regulation provides a high-level timeline for regulatory review (e.g., 60–90 days for initial assessment), but in practice, no detailed national roadmap with milestones for scaling CC retrofits, new builds, or transport/storage infrastructure exists. The absence of a binding timeline risks lagging implementation relative to modeled pathways.

-

Policy Gaps and Needs Revealed by Modeling:

The scenario results suggest several additional policy measures are needed:

• Minimum Carbon Price: Achieving robust CC deployment in CALL and CNEW requires a sufficiently high carbon price (≥USD 30–50/tCO₂e) to ensure project viability and competitive returns. Current Indonesian carbon pricing pilots (<USD 5/tCO₂e) are not sufficient.
• Guaranteed Offtake/Contracts-for-Difference: Long-term contracts or contracts-for-difference (CfD) for low-carbon electricity would de-risk private investment in CC-equipped power plants, similar to mechanisms used for renewables.
• Infrastructure Incentives: Targeted public investment or blended finance for CO₂ pipelines, shared storage, and grid upgrades will be critical for enabling both retrofits and new-build clusters. International climate finance, export credits, or multilateral support could accelerate hub development.
• One-Stop Permitting and Clear Lead Agency: Designating a single, empowered coordinating body for CC project approvals would streamline regulatory risk and lower transaction costs.

While COMB offers a balanced and cost-effective pathway, it is also important to recognize the implications of alternative strategies under specific constraints. The CNEW scenario, which relies solely on new CC-equipped capacity, may become relevant if technical, economic, or logistical barriers prevent large-scale retrofitting, particularly as some older plants may not be suitable due to layout, efficiency, or space constraints. International experience, such as Boundary Dam (Canada, retrofit) and Petra Nova (U.S., purpose-built), shows that both retrofitting and new builds are viable, and that siting new plants closer to geological storage or industrial clusters can reduce transport costs and facilitate shared infrastructure. Although CNEW incurs higher system costs than CALL, it offers greater spatial flexibility and long-term scalability, especially if policies support clustering of emitters and integrated CO₂ management. Thus, CNEW serves not as a forecast, but as an analytical boundary case for understanding a new-build-centric pathway.

4.2. Climate Targets Alignment

The IPCC AR6 [9] emphasizes that power-sector emissions must reach near-zero globally between 2050 and 2080 to stay within “well below 2 °C” pathways. The IEA’s Net Zero Emissions Roadmap for Indonesia [6] similarly targets net-zero or net-negative emissions by 2060, relying heavily on renewables and carbon removal technologies like BECCS. Against these benchmarks, the modeled scenarios display a clear hierarchy:

• NRE achieves near-zero emissions well before 2060 by phasing out all fossil generation.
• COMB reduces emissions to 40 Mt CO₂e (96% below BAU) but still exceeds full-alignment thresholds.
• CALL, limited to retrofitting existing fossil plants with CC, plateaus at 105 Mt CO₂e, insufficient for Paris-consistent pathways.

Closing this gap within a “CC-only” framework would require one or a combination of the following technical levers:

• Higher capture rates. Commercial amine systems typically remove 90% of flue-gas CO₂, but studies have investigated post-combustion capture efficiencies of up to 99% [49,50]. Raising capture performance to the upper end of this range would lower residual emissions and bring the scenario closer to the AR6 target.
• Low-carbon co-firing. Blending sustainably sourced biomass with coal can further abate stack emissions. Indonesian studies report life-cycle reductions of 70–80 kg CO₂ MW/h at a 20% biomass share, even before capture is considered [22]. Integrating such co-firing with CC would cut the remaining gap. Ammonia co-firing is also being explored as a strategy to reduce residual emissions from CCS-equipped power plants [51,52].

Indonesia’s biomass availability for energy is estimated at 130–137 million tonnes per year, translating to approximately 39.4 million tonnes of oil equivalent, primarily from agricultural and forestry residues such as palm oil waste, rice husks, and wood processing by-products [53]. While this presents substantial potential, the distribution of biomass is uneven, concentrated in regions like Sumatra and Kalimantan, and faces logistical, economic, and land-use challenges. Additionally, although Indonesia produces over 21 million m³/year of wood, actual utilization for co-firing or wood pellet production remains modest due to policy, infrastructure, and investment constraints.

In summary, while biomass and ammonia co-firing can complement CCS, their scalability is limited by supply, land-use, and infrastructure constraints. A phased and regionally targeted deployment of these strategies, alongside renewable expansion, could improve the alignment of CC-heavy scenarios with national climate goals. However, this requires further techno-economic evaluation to ensure long-term sustainability and system reliability.

4.3. Energy Security

Indonesia holds an estimated 33 billion tonnes of proven coal reserves and more than 90 billion tonnes of resources, enough to sustain current production levels for many decades [1]. Relying on that domestic fuel base insulates the power system from imported energy shocks. Conversely, high solar shares can weaken that security margin, especially with output volatility, where Java-Bali’s solar units recorded a 35% decline in average output during the 2023–2024 wet season [48]. Covering the fluctuations with battery storage or pumped hydro is technically feasible but remains capital-intensive [54,55,56]. Against this backdrop, the three CC pathways (CALL, CNEW, and COMB) offer low-carbon power generation that leverages abundant domestic coal and existing gas infrastructure. They therefore:

• Reduce import exposure. Fuel is sourced locally, limiting vulnerability to Liquefied Natural Gas (LNG) or oil price shocks.
• Provide firm capacity and support grid stability. CC-equipped units operate as baseload, covering renewable shortfalls without the need for large-scale storage in the near term.

Among them, the COMB scenario offers the most balanced approach. It combines retrofitted and new CC-equipped fossil plants with accelerated renewable deployment, achieving a lower emissions profile than CALL and CNEW while avoiding the operational constraints of a renewables-only system. Nevertheless, COMB and other CC pathways involve trade-offs, including exposure to potential future carbon pricing and the need for extensive investment in CO₂ transport and storage infrastructure.

The NRE scenario, which aims to achieve near-zero emissions by 2060, raises concerns regarding system reliability and feasibility. The LEAP model does not explicitly simulate grid stability and curtailment. However, external assessments indicate that maintaining grid reliability in a high-renewable system would require 12–16 GW of battery storage and 5–10 GW of pumped hydro, along with elevated reserve margins of approximately 25%, significantly above Indonesia’s current average of 17–18% [6,37,48]. Without aggressive investment and regulatory support, the NRE pathway may face short- to medium-term reliability challenges.

All CC scenarios incorporate CC-gas to support dispatchable, low-carbon generation amid Indonesia’s rising LNG imports. Analysts project a shortfall of ~10 LNG cargoes in 2024, and the IEA forecasts a sharp increase in Southeast Asia’s LNG imports through 2050 [37,57]. This exposes the power sector to price volatility and supply risks. One mitigation strategy is the deployment of hydrogen-ready turbines, which can initially run on natural gas and later transition to low-carbon hydrogen as the domestic supply develops [37,58]. While hydrogen use in gas turbines will remain limited until the 2030s, the option of integrating new CC-gas units with hydrogen offers future flexibility and enhances energy security.

Accordingly, COMB serves as a transitional model that integrates fossil-based reliability with renewable growth while remaining adaptable to fuel diversification. As discussed in Section 4.1., successful implementation hinges on institutional readiness, including clear regulatory timelines for retrofits, investment incentives for dual-fuel infrastructure, and integrated planning across fossil, renewable, and CO₂ transport sectors. Such measures are critical to aligning energy security with Indonesia’s broader decarbonization agenda.

Beyond the national electricity mix, the findings of this study hold relevance for industrial sectors that contribute significantly to Indonesia’s emissions, particularly mining and resource extraction. These operations frequently depend on on-site fossil-fueled power generation, and incorporating CC technologies can significantly reduce their operational carbon footprint. Recent studies demonstrate that CC systems can capture up to 90–99% of CO₂ emissions from coal- and gas-fired power plants, making them a viable decarbonization option even in off-grid or hybrid applications [15,59,60]. Integrating CC into industrial supply chains, either through retrofits or hybrid CC-renewables strategies, not only supports companies’ alignment with evolving sustainability expectations and international emissions standards but also provides a path to deep emission reductions. Thus, the COMB scenario modeled in this study offers insights not only for national power planning but also as a template for energy-intensive industries aiming for decarbonization.

4.4. Comparison with Literature

This study contributes to the growing body of research on decarbonization pathways for Indonesia’s power sector by integrating CC technologies into a national-scale scenario framework using the LEAP model. In doing so, it complements and expands upon prior studies that have addressed decarbonization from narrower perspectives, whether focused on unit-level technology assessments, top-down macroeconomic projections, or renewables-only energy futures.

Rahmanta et al. [22] and the World Bank [23] provide detailed simulations of post-combustion CCS retrofits for individual coal-fired power plants using the IECM tool. These studies quantify energy penalties, cost increases, and CO₂ capture potential, offering useful micro-level benchmarks. However, they do not consider interactions with national energy systems or long-term scenario trajectories. In contrast, the present study scales these insights to assess their systemic impact under multiple national pathways, thereby linking technical feasibility with strategic policy implications.

Kanugrahan and Hakam [27] is another LEAP-based study that modeled Indonesia’s transition to 100% renewable electricity under five different scenarios. While their modeling is aligned in terms of software platform and long-term ambition, they exclude CC technologies from their analysis. Their results underscore the feasibility of a renewables-only path but do not explore the transitional role or comparative cost-efficiency of CC. This study addresses that gap by showing how combined pathways (COMB) can achieve comparable emissions reductions (96%) with moderate cost increases, offering an alternative route that aligns with Indonesia’s fossil-reliant starting point and infrastructure constraints.

Broader studies [5,6,61] provide valuable context for understanding Indonesia’s energy transition’s macroeconomic, policy, and investment dimensions. While these reports favor renewables-dominant pathways and emphasize system-level goals, they typically use top-down models that lack the technological resolution offered by LEAP. The present study complements these high-level perspectives by offering bottom-up, policy-relevant modeling grounded in Indonesia’s power sector structure.

In a regional context, a relevant comparison can be drawn with Thailand’s LEAP-based assessment, where CC retrofits were modeled alongside carbon tax and renewable scenarios [28]. The analysis found that while CC could marginally reduce emissions, carbon pricing and aggressive renewable expansion achieved far greater mitigation at lower cost. Similarly, an ASEAN-wide study projected that a renewable-dominant pathway, without reliance on CC, was the most cost-effective route to net-zero by 2050, highlighting the regional competitiveness of renewables over CC technologies in power sector transitions [29]. Our findings partly align with these studies: when full CO₂ T&S costs are included, the RE-only scenario in Indonesia becomes the most economical in terms of average abatement cost. However, our analysis highlights that such a pathway poses significant short-term reliability and flexibility risks due to geographic and institutional constraints. The COMB scenario addresses these by combining CC and renewables.

Despite its potential, CC deployment in developing countries must be approached with caution. Nasir and Go [62] highlight Malaysia’s limited readiness for CCS due to regulatory and financial gaps, challenges Indonesia also faces. Additionally, critiques from Nature Climate Change and One Earth warn that CCS can delay renewables, incur high costs, and risk stranded assets if over-relied upon [63,64]. Our findings reflect this caution: CC-only scenarios (e.g., CALL, CNEW) are less effective than the hybrid COMB approach. By combining CC retrofits and new builds with accelerated renewables, COMB provides a transitional strategy tailored to Indonesia’s infrastructure, cost, and reliability context, positioning CC not as a replacement for RE, but as a pragmatic complement in the near-to-medium term.

5. Conclusions

This study assessed the role of CC in Indonesia’s power sector using LEAP, evaluating five scenarios: BAU, NRE, CALL, CNEW, and COMB. Results show that CALL, which introduces CC to the BAU fossil plants, including retrofitting, offers the lowest abatement cost at USD 0.93/tCO₂e but falls short of deep decarbonization targets. COMB, combining retrofits, new CC builds, and renewable expansion, achieves the highest emissions reduction (96%) at a moderate cost of USD 1.91/tCO₂e, striking a balance between mitigation, cost, and system flexibility. Including T&S costs raises abatement costs, most notably for CALL (USD 26.72/tCO₂e) and CNEW (USD 35.89/tCO₂e). COMB remains relatively affordable at USD 9.67/tCO₂e, though NRE becomes the lowest-cost option. However, NRE faces challenges related to grid stability, storage needs, and high capital investment. COMB provides dispatchable, low-carbon power while supporting renewable growth, enhancing energy security and operational reliability.

Policy priorities should therefore center on front-loading CC retrofits, building new CC-units, accelerating renewable and grid upgrades, and developing a national CO₂ transport-and-storage network, alongside pilots for sustainable biomass and green-ammonia fuels. These measures can place Indonesia on a durable, least-cost pathway toward net-zero power-sector emissions while safeguarding energy security. Notably, the results underscore the critical role of technology timing: allowing one option a significant head start, such as early CC-coal deployment, can crowd out cleaner alternatives that become available later, locking in higher long-term costs and emissions. Future work should explore spatial modeling, storage integration, and sensitivity to cost trajectories and policy shifts.