← Return to previous page

Colombia

Main findings

2050 projections of power and industrial high temperature heating sectors:  

  • Projected no-action emissions based on current energy mix: 51.7 MtCO2e /year+ 

  • Maximum mitigation potential: 51.7 MtCO2e /year 

  • Most important needle-mover are onshore wind and ammonia storage. 

  • Average cost of achieving zero emissions with only domestic measures is USD 26.5/tCO2e (shown as a negative “additional cost”) compared to BAU no action scenario. 

  • In international collaboration, onshore wind is a key element for reducing system costs. Ammonia storage also plays a significant role, providing flexibility in reducing reliance on costlier domestic resources.§ 

+Emissions are the annualized value for a 30-year project starting in 2050 (i.e., divided by 30 from the total project emissions) 

§A negative value for cost of achieving net zero in collaboration configurations implies that BAU-No Action scenario for the set of countries assuming current energy mix leads to a more expensive system than when investing in zero-carbon power generation for those countries. Please see Country-specific notes on Methodology - National and Collaboration Modelling (STEVFNs) section below for detailed information on assumptions leading to this result. 

Country context

Parameters 

Socioeconomic indicators 

  

Geographical opportunity/limitation 

  

Fossil fuel dependency 

GDP: Current USD 363.54 billion (2023)1 

GDP/capita: Current USD 6969.7 (2023)2 

Maximum resource potential*/**:  

  

PV rooftop: 38 GWp 

PV open field: 6012 GWp 

Wind onshore: 1089 GWp (329.7 GWp for STEVFNs modelling)^ 

Wind offshore: 157 GWp

  

Energy imports: 15% of the primary energy supply 3 

Population: 52.09 million (2023)4 

Population density: 46/sq km (2021)5 

  

Fossil fuel rent: 

Gas: 0.2 % of GDP6 

Coal: 0.7 % of GDP7 

Oil: 3.4 % of GDP8 

in 2021. 

Emissions: 139.1 MtCO2e (2023)9 

  

Carbon intensity of energy: 0.17 kgCO2/kWh10 

* The technical potential is a first order estimate calculated based on a generalised set of land / ocean area exclusion constraints and technical parameters for each technology. Differences to other literature can occur due to different modelling assumptions.  

** The technical potential for renewable energy sources used in STEVFNs including wind onshore, wind offshore, open field and residential rooftop solar PV is estimated by a Python-based simulation pipeline11. The pipeline applies temporally and spatially-resolved simulation models of the open-source python packages GLAES (Geospatial Land Eligibility for Energy Systems) and RESKit (Renewable Energy Simulation Toolkit)12

^For onshore wind technology, only the first (best) two “bins” from the simulation pipeline were used and averaged for CAPEX, capacity factor time series and maximum capacity as inputs to STEVFNs modelling

Country context  

Colombia, with a population of around 52 million people, recorded a GDP of $363.54 billion and a per capita income of approximately $6,980 in 2023. As one of South America’s largest and most biodiverse nations, Colombia has a significant stake in climate efforts. The country’s varied topography—from the Andes mountains and Amazon basin to the northern deserts and Caribbean coastline—supports a broad range of renewable energy resources. Hydropower is the dominant renewable energy source, particularly in the Andean region, while the La Guajira Peninsula in northern Colombia presents significant potential for wind and solar energy. Despite these resources, fossil fuels continue to play a role in Colombia’s economy. In 2020, coal and gas supplied about respectively 10% and 16% of the national power generation. Furthermore, coal accounted for roughly 13% of total exports, while petroleum made up 45%. As a fossil fuel-dependent developing country, Colombia faces challenges in transitioning to a sustainable economy amid global efforts to reduce carbon emissions. 

Politically, climate action has gained traction with the election of President Gustavo Petro in August 2022, who has committed to steering Colombia toward a low-carbon economy. His administration has highlighted the need to reduce deforestation, transition from fossil fuels, and build a sustainable energy system, though this ambitious agenda is dependent upon international cooperation. Early moves by his government include a draft legislative ban on fracking. However, while Petro’s administration seeks to advance a green agenda, the country’s reliance on coal and oil presents challenges, especially for exports as the global demand for coal declines and the international community aims for accelerated decarbonization. 

Country target and policy

Colombia’s climate policy framework centers around its NDC and its Climate Action Law. The updated 2020 NDC aims to cut national GHG emissions by 51% by 2030 compared to a business-as-usual scenario. Additionally, Colombia submitted a 2050 Long-Term Strategy (E2050) to the UNFCCC, committing to a net-zero GHG emissions target by mid-century. The country formalized these commitments through the Climate Action Law (Law No. 2169) in December 2021, establishing Colombia’s NDC and carbon neutrality goals as legally binding. However, achieving these targets requires substantial efforts across sectors. 

Colombia has made good progress implementing renewables in the power sector in 2020, over 70% of Colombian power supply comes from hydropower. However, there are no further targets for rapidly upscaling renewables beyond 2022. Measures within Colombia’s NDC include improving energy efficiency standards for cooling and heating by implementing resolution N°0549 relating to sustainable construction. Colombia has committed to a 15% reduction in fossil fuel use in the cement industry by 2030.  

Although Colombia has set an ambitious 50% EV target by 2030, it has not adopted a full EV pledge. Efforts to expand EV infrastructure remain limited by support for natural gas in transportation, underscoring a complex energy transition.  

In 2021, Colombia signed the Global Methane Pledge at COP26, aiming for a 30% reduction in methane emissions from 2020 levels by 2030, focusing on energy, waste, and agricultural emissions. Methane, which accounts for roughly 40% of Colombia’s GHG emissions (excluding LULUCF), mainly originates from agriculture. While the NDC includes measures to reduce methane, achieving the pledge will demand further policy efforts and full implementation of planned initiatives in waste management. 

Outlook mitigation potential 2050 - STEVFNs

Summary: 

Up to 51.7 MtCO2e/y     Technical Potential at no additional cost^^ 

Up to 51.7 MtCO2e/y     Technical Potential at additional average mitigation cost^ of USD10/MtCO2

Up to 51.7 MtCO2e/y     Technical Potential at additional average mitigation cost^ of USD20/MtCO2

Up to 51.7 MtCO2e/y     Technical Potential at additional average mitigation cost^ of USD50/MtCO2

Up to 51.7 MtCO2e/y     Geographic potential# 

51.7 MtCO2e/y              Technical potential – domestic* 

Key elements: Onshore wind. Under STEVFNs modelling, Colombia is able to reduce emissions by 100% from the no-action scenario. This differs to the estimates under OSeMOSYS detailed national modelling due to the different sectors and technologies modelled. 

 

51.7 MtCO2e/y               Technical potential – International collaboration  

Key elements: Green electricity imports are key to reduce the cost of mitigation for Colombia in collaborations, with possibilities to reach net savings. 

^^This refers to a change in the technology mix that would result in the same system cost as the current policy scenario. It does not take into account costs associated with transiting to a different technology mix. 

^This refers to the additional average system cost with reference to the current policy scenario, costs expressed in USD 

#Geographical potential is estimated only for the sectors considered in GMPA. GMPA tries to consider the cheapest and biggest mitigation options/sectors, however other mitigation options/sectors also exist so actual geographical potential is larger. As GMPA adds more sectors, this number will get closer to matching the actual theoretical limit.  

* Refers to the emissions reductions that can be achieved by implementing the full set of available options in a given year. 

 

Dynamic Pareto Abatement Cost Curves (D-PACC) in Autarky and International Collaborations* 

*The following stills of the D-PACC show bar charts for the annualized cost of the main technologies in the highest mitigation scenario for each case study. For higher detail, please see the interactive mitigation potential diagrams when exploring the map. 

 

 

 

 

 

 

Summary of modelling results: 

  • Even in autarky, using only domestic measures, Colombia would be able to reduce to zero all power and industrial high temperature heating related emissions.  

  • Under the least cost model, Colombia is able to reduce emissions to 30% of a no-action scenario, mainly by installing a high share of onshore wind, requiring an annualised investment of USD 7 Billion.  

  • The total system costs increase accelerate when annual emissions budget in 2050 is reduced to below 8% of the level in a no-action scenario.  The first 90% of emissions reductions are much cheaper than the final 10%, seen as an “elbow” in the D-PACC near zero emissions. 

Choice of Distributed Energy Resource:  

  • Onshore wind is the dominant energy generation technology used from emissions reductions around 30% of BAU no action scenario, and increases in importance until reaching zero emissions. 

  • At abatements over 90%, the least cost system utilises a combination of onshore wind and open field PV as DER, coupled with battery and ammonia storage as well as ammonia generation. 

Role of energy storage technologies:  

  • Battery storage is deployed from over 30% of BAU emissions, while ammonia storage is needed for the last 8% of mitigation, driving mitigation costs higher. 

  • In the autarky scenario, ammonia and battery storage are invested in similarly for the zero emissions case.  

Role of Fossil Fuel power plants:  

  • Fossil Fuel power plants in Colombia are used and needed for flexibility services until zero emissions with only domestic measures. 

  • While battery storage is invested in for this purpose for lower emissions reduction levels, fossil fuel power plants are still required at those higher levels. 

Benefits of International Collaboration:  

  •  Colombia is able to reach zero emissions with only domestic measures, but international collaborations could aid in reducing mitigation costs for Colombia, or even reach a net saving per ton of CO2e abated. 

  • Collaboration can make the “elbow” of increasing costs near zero emissions less pronounced compared to the countries in autarky 

  • Green electricity imports are the main source of this, with the presence of HVDC cables between Colombia and neighbouring countries reducing the total investment required for onshore wind in Colombia for the least cost zero emission system. 

  • There is some risk of lock-in for onshore wind if planning a system for autarky in Colombia and then attempting to collaborate with Chile for example, as the collaboration system has a reduction in onshore wind investment for Colombia near zero emissions. 

We have included some illustrative case studies of effective policy interventions in particular countries and cities. 

  1. Power
  2. Energy
  3. Industry
  4. Transport
  5. Agriculture

National and Collaboration Modelling (STEVFNs) 

National and international modelling was performed using STEVFNs energy system model generator. Modelling was performed by setting annual emissions limits and finding the cost- 

optimal technology mix that meets all hourly demands with the emissions constraints. 

This modelling assumes a greenfield model built in 2050 to meet fixed electricity and high-temperature heating demand, minimizing the net present value of a 30-year project. 

First, a baseline case study for no action was set up for COUNTRY with its current energy mix, determining the no-action emissions on its own (autarky). The “current energy mix” is required to meet national electricity and industrial heating demand, and is assumed to include: 

  • Open field Solar PV – Constrained to a maximum capacity which projects current installed capacity linearly with demand growth 

  • Onshore Wind – Constrained to a maximum capacity which projects current installed capacity linearly with demand growth 

  • A generic powerplant asset that considers grid emission factors by country 

  • A fossil fuel industrial heater 

From this baseline, a linear reduction towards zero emissions was determined that would constrain the scenarios. A total of eleven scenarios with emissions constraints ranging from those obtained for BAU no action to zero were run to build the cost-optimal technology mixes shown in the Dynamic Pareto Abatement Cost Curve (D-PACC). 

To determine emissions reductions strategies in terms of technological investment, a wider set of technologies is given to the model in the autarky and collaborations scenarios. These include, in addition to the technologies modelled in BAU no action: 

  • Rooftop PV 

  • Offshore Wind 

  • Battery storage 

  • Electric industrial heater 

  • Ammonia-fueled industrial heater 

  • Electricity-synthesized ammonia 

  • Ammonia-fueled electricity generator 

  • Ammonia storage 

In this case, the renewable generation assets are constrained in capacity to the maximum technical limits per country as shown in the Parameters table under Country Context. 

 

After determining the total mitigation potential in autarky for COUNTRY, two additional case studies were created: 

  1. A combination of countries in autarky, with maximum total emissions defined as the sum of their individual autarky emissions  

  1. A combination of countries collaborating (adding electricity and ammonia transport between them in addition to the 8 assets listed above), with maximum total emissions defined as the sum of their individual autarky emissions. 

Then, the methodology implementing emissions constraints in eleven scenarios with respect to the sum of the set of countries no action emissions were set up. Results from these then build the D-PACC for collaboration and for the set of countries without collaboration. In some cases, the set will not be able to reach zero emissions in the modelled sectors when modelled independently, as their individual reductions make the problem infeasible at some level of constraint for total emissions. These, however, may reach zero (or at least higher emissions reduction) when energy trade is enabled through green electricity and ammonia trade. 

 

Under these assumptions, Colombia can reach net zero with an “Additional Average Abatement Cost” per ton of CO2e with only domestic measures. However, collaborating with 

other countries it may participate in systems with an “Additional Average Abatement Cost” 

per ton of CO2e that is negative. These numerical results show that the transition to cleaner technologies is cheaper than maintaining and operating a system with the current technology mix. However, it is important to emphasize some assumptions in this work and their influence on such results. 

  • The current version of GMPA represents only the electricity and industrial heating sectors. Furthermore, it only models a certain number of assets to meet demand, which reduces the detail of current and potential future energy mixes (e.g. no hydropower, nuclear, or other generation technologies are modelled, different types of fossil fuel powerplants are grouped into one with total grid emissions factors as assumptions, etc.) 

  • STEVFNs model generator is used as a greenfield model. As such, no existing generation infrastructure is considered. This could increase costs in BAU-no action scenarios requiring capital investment for capacity that may already exist. Similarly, no costs for decommissioning such existing infrastructure towards net-zero emissions in the autarky and collaboration scenarios are considered. 

In summary, countries where renewable energy infrastructure is already considerably cheaper than fossil powerplants for example, and which have high technical capacities for these resources, are likely to find a cost-optimal system which saves money per ton of CO2e emitted by these two sectors in a net zero system compared to its BAU counterpart. 

Data 

Electricity and high-temperature heating demand are projected to 2050, following the estimates from high-resolution additional detailed modelling performed in OSeMOSYS. Technology capital and operational costs were translated from detailed OSeMOSYS modelling where available, and estimated based on literature figures for technologies, including high-voltage direct current (HVDC) submarine cables. See detailed methodology in for specifics on this data. 

For renewable generation technologies, IRENA13 projected CAPEX values to 2050 (medium scenario, in 2022 USD), were used for onshore wind, utility-scale (open field) PV, and rooftop PV. For both wind technologies, within the Python-based pipeline for renewable energy source analysis, separate “bins” were created to differentiate performance in different regions and using different turbines. For Onshore Wind, the best two bins (lowest LCOE) were used to estimate and average capacity factor timeseries, the average CAPEX and the maximum total capacity potential.  

For offshore wind, the CAPEX was estimated through a weighted average of fixed and floating offshore technologies based on the shares of their maximum potential.14  The baseline cost was IRENA’s medium scenario CAPEX, assumed for fixed offshore wind; a ratio of floating CAPEX to fixed CAPEX of 1.44 was used based on each technology CAPEX.15   

 

Additional Detailed National Modelling  

In the pilot, additional “detailed national modelling” was performed using OSeMOSYS energy system model-generator. These are supposed to emulate potentially different energy system models currently used by different countries. In future phases of GMPA, countries will be encouraged to share their national models and data. These will be translated to the generalized STEVFNs system-of-systems model-generator.  

  

This is done for the following benefits:  

  1. Reduce barrier to entry for countries to engage with GMPA.  

  1. Leverage on national modeling efforts around the world instead of repeating modelling efforts.  

  1. Have a scalable, distributed method of managing and updating national modelling data.  

  1. Ensure the input assumptions on GMPA are “owned” by countries so that they are comfortable enough with the results for the results to be useful as a basis to bring parties onto the negotiation table.  

In the pilot, a detailed national modelling was performed for Singapore by building a “0th order” OSeMOSYS “starter data kit” using the methodology developed by Climate for Compatible Growth (CCG) that is applied to more than 60 countries around the world. The model determines the least cost optimal technology mix pathway from 2015-2070 to meet all end-use energy demands given some emissions constraints. The sectors included are power, transport, industry, household, and commercial sectors.
  

If you would like to see the D-PAC curves for detailed national modelling, please contact GMPA. 

The D-PAC Curves for detailed national modelling were developed by setting annual emissions constraints for 2050. The annual emissions from 2025 to 2050 were constrained to reduce linearly to the 2050 values. The emissions were assumed to be constant from 2050-2070. The value on the x-axes is the annual emissions constraint at 2050. The value on the y-axes of the stacked charts and the bar charts is the net present value of the money spent on each technology from 2015-2070. Note that this means that there will be some investments in fossil fuel technologies even when the pathway leads to net zero by 2050. While this may seem less intuitive than the results of the greenfield models, they are more representative of the difference in cumulative pathway costs rather than the costs for a specific year.  

  

The results presented on the D-PACC for the detailed national modelling is slightly different from the data presented on the DAC Curves for the rest of the modelling. This is done in the pilot to demonstrate that different information can be presented using the same visualization. The aims of presenting different types of results with slightly different interpretations are as follows:  

  1. Get feedback from users to understand which kind of model results/ assumptions are more useful for planning and negotiations.  

  1. Provide alternative methods of looking at the problem to check if both methods give the same policy recommendation and insights. This is inspired by similar methods used by physicists to interpret results for “first picture of a black hole”.  

  1. Give country experts access to results from models that they are familiar with so that they can compare and be comfortable with the results for STEVFNs for individual countries before looking results from international collaboration models. 

Contact

For additional information, please contact the GMPA consortium at info@mitigationatlas.org