Methodology

The HPA scenario was developed by Climate Analytics in collaboration with the Potsdam Institute for Climate Impact Research (PIK). This scenario explores how, starting from 2025, the world can limit the duration and magnitude of overshoot above 1.5ºC, and bring global warming back below 1.5ºC well before the end of the century.

The HPA scenario shows how transformative climate action, by scaling up renewables and electrifying the global economy, can slow and then halt global warming before 2050, and limit peak warming to around 1.7ºC. After peaking at around 1.7°C, warming then falls to approximately 1.2°C by 2100, with the overshoot period lasting roughly 40 years.

The HPA is built on four key levers – rapid renewables driven electrification, a fossil fuel phaseout, strong reductions in methane and other non-CO₂ emissions, and scaling-up carbon removal technologies such as direct air capture. Pursuing these four levers in line with the highest possible ambition can help keep 1.5ºC alive. More details on the HPA can be found in the accompanying report.¹1

The HPA scenario is produced by the REMIND 3.5 integrated assessment model.²2

In the HPA scenario, we constrain global emissions to follow the current policy trajectory until 2025. This means the scenario accounts for the failure to peak and reduce emissions rapidly over the first half of the 2020s. After this, we make the REMIND model meet the lowest possible carbon budget within the model’s techno-economic constraints. We are therefore exploring the feasibility range of the model to look for the lowest possible temperature outcomes that are techno-economically and geo-physically feasible.

We note that the feasibility of a scenario is a broader concept that includes socio-political and cultural dimensions.³3 While a model may identify a scenario as feasible or infeasible based on its underlying constraints and assumptions, this does not imply that the outcome is definitely feasible or infeasible in the real world.

We then apply a range of additional constraints and developments to the REMIND model to produce the HPA scenario through:

• Regionally differentiated carbon prices: Rather than applying a globally uniform carbon price, there is a spread of carbon prices across regions, which converge to a globally uniform price by 2070. Regions with higher GDP per-capita have higher near-term carbon prices and therefore cut emissions faster.

• Improvements in addressing energy equality: We model demand-side action in advanced economies which reduces energy demand relative to a current policy reference scenario. In parallel, the scenario enables a faster scale-up of energy service demands in low-income countries, helping reduce interregional inequality in energy service demands.

• Inclusion of sustainability and feasibility bounds for biomass, CCS and CDR. Many global scenarios have been criticised for relying too heavily on technologies such as biofuels, CCS and CDR, and using this to enable continued fossil fuel consumption. In the HPA, biomass availability is limited to ~80 EJ/yr, total underground carbon sequestration is limited to 8.6 GtCO₂/yr (across fossil, biomass, process emissions and direct air capture), and individual CDR methods broadly align with literature defined constraints

Under the HPA scenario, electricity meets almost two-thirds of all energy demand by 2050, with electrification outperforming other options on cost, scalability and energy efficiency. CO₂ emissions reach global net-zero before 2050, and the scenario also reaches net-zero greenhouse gases in the 2060s, roughly a decade earlier than the average 1.5ºC compatible scenario in the IPCC’s AR6 database. Methane emissions fall 20% by 2030 and 30% by 2035, relative to 2020 levels, and removal technologies scale to remove over 5 billion tonnes of CO₂ per year by 2050.

All scenarios produced by models are imperfect representations of the real world, which cannot fully capture all relevant dynamics that will guide the energy transition.

The scenario data underlying 1.5°C compatible pathways specify how future energy consumption and emissions should be composed in different regions of the world. Typically, this data is only available for regional aggregates called macro regions. In the integrated assessment model REMIND, there are 11 macro regions: the United States, Latin America, European Union (+ United Kingdom), Europe non-EU28, Middle East and North Africa, Sub-Saharan Africa, India, China, Japan, Other Asia and a region for the rest of the OECD regions (Australia, New Zealand and Canada) . This means that in order to determine national energy consumption and emission pathways, the data of the macro-regions needs to be downscaled to the national level.

The downscaling process itself can be broken down into several sub-steps:

1. Defining the macro-region(s) in which the country of interest is located.

2. Country’s historical emissions and energy consumption are determined for all countries in the macro-region(s).

3. Future emissions and energy consumption are obtained from the scenario data underlying the to be downscaled 1.5°C compatible pathway.

4. The macro-region’s scenario data is adapted to match the country’s historical data in a base year. This process is called harmonisation and it is required to update the pathways to the latest available historical data.

5. The macro-region’s emissions and energy consumption are downscaled to its countries. They are distributed to the countries in an internally consistent way, which preserves total values and matches the historical value of each country.

Different methods are deployed for the downscaling process, depending on the sector/emissions to be downscaled.

Input data

Historical primary, secondary and final energy consumption are obtained from the International Energy Agency (IEA) World Energy Balances (WEB).⁴4

Category	Consumption by fuel	Determination method
Primary energy mix and power sector	Non-biomass renewable	the sum over the consumption of wind, solar, geothermal and hydro energy
	Biomass	the difference between renewable energy consumption and the aforementioned non-bio renewable energy consumption
	Coal	the sum over peat and coal
	Oil	the sum over oil, oil products and natural gas liquids
	Gas and Nuclear	directly obtained from the database
End-use sectors (buildings, transport, industry)	Oil and e-fuels	the sum over primary and secondary oil and bio jet kerosene
	Coal	the sum over coal, peat and oil shale and non-renewable waste
	Natural gas	directly obtained from the database
	Biomass	the sum over primary solid biofuels, renewable waste, charcoal and non-specified primary biofuels and waste
	Biofuel	the sum over biodiesels, biogasoline and other liquid biofuels
	Biogas	directly obtained from the database
	Hydrogen	assumed to be zero in the base year
	Electricity	directly obtained from the database
	Heat	the sum over heat, geothermal and solar thermal

It should be noted that in the IEA database, secondary energy consumption is assigned to both electricity and heat generation. For downscaling purposes, we assume that the majority of the consumption is dedicated to the power sector with heat generation being a by-product of centralised electricity generation.

Historical emissions are obtained from the Climate Action Tracker (CAT) as a primary source where available. For countries not covered by the CAT, the PRIMAP dataset is used for non-CO₂ emissions and CO₂ emissions from the non-energy sectors and from an IEA dataset for energy sector CO₂ emissions, again with the caveat of including combustion related CO₂ emissions from both electricity and heat.

The consumption and emissions projections for the macro-region(s) are obtained from the integrated assessment models REMIND.

Energy supply and consumption, and energy-related CO₂ emissions

Harmonisation

Before the downscaling, we perform a harmonisation of all the variables, to align the projections from REMIND with historical data. We do that using the Aneris package,⁵5 which determines a default harmonisation method per variable depending on the difference between historical and modelled emissions, the level of variation in the historical data, whether the variable becomes negative and more. In all cases, harmonised data starts from the historical data, but converges to the long-term original results of the scenario at a given point in time, usually 2080. This attempts to capture a balance between accurately capturing historical data and calibrating near-term variables on the basis of historical data, while still representing the key dynamics in the scenario.

For the macro-regions that are full countries and don’t need downscaling, United-States, China, India and Japan, the process stops here and the analysis is done on these data.

DSCALE

The downscaling is performed using DSCALE (Downscaling Scenarios to the Country level for Assessment of Low carbon Emissions), an open-source algorithm designed to downscale regional IAM outcomes to the country level.⁶6 DSCALE builds on two separately derived paths:

• A National Data-Driven ("NAT") path, where country-level data and derived historical trends are the main driver, harmonised for consistency with regional IAM scenarios. This path is particularly relevant in the short term as it facilitates a smooth transition toward future low-carbon scenarios and accounts for the different starting point of countries in terms of their levels of electrification, energy intensity, fuel mix, resource availability and more.

• An IAM-Driven ("IAMatt") path, where IAM scenarios are translated to the country level using country-specific GDP and population pathways from the Shared Socioeconomic Pathways (SSPs). This path serves as a long-term attractor toward which national paths gravitate.

These two paths are merged into a "composite" path using a weighted average, where the weight shifts over time from the NAT path (at the base year) to the IAMatt path (at the time of convergence). The timing of convergence is linked to the scenario narrative: for 1.5°C compatible scenarios, a faster convergence is assumed. In our scenarios the convergence is set at 2100 for the final energy sectors and 2200 for primary and secondary energy sectors. That means out to 2050 the driving force is the nationally data-driven pathways, which account for historical and current country context to inform how the energy transition can be downscaled from the regional to the national level.

DSCALE ensures that the sum of country-level results align with regional IAM outcomes through iterative structure adjustments: the sum of sectors and energy carriers is aligned with total final energy demand within each country, and the sum across countries is equal to the regional IAM results.

Downscaling the end-use sectors

Total final energy demand is driven by the energy intensity (final energy divided by GDP), GDP per capita and population, using exogenous projections from the SSP framework. The relationship between energy intensity and GDP per capita is estimated using a log-log regression model, based on historical data at the country level for the NAT path and on regional IAM scenario results for the IAMatt path. This conditional convergence approach ensures that convergence in energy intensity across countries is tied to the level of economic development.

Final energy is downscaled hierarchically: first total final energy is downscaled using GDP as the main driver, then energy carriers (electricity, solids, liquids, gas) as shares of total final energy, and finally sectoral breakdowns (buildings, transport, industry) as shares within each energy carrier.

Downscaling the power sector

When downscaling the electricity sector specifically, the algorithm considers additional criteria beyond the general approach:

• the remaining technical lifetime of fossil fuel-based power plants (data from the Platts World Electric Power Plants Database),

• supply cost curves for renewable energy allocation,⁷7

• and governance indicators.

These criteria are combined with historical data using configurable weights to determine the NAT path for the electricity mix.

Ensuring consistency across the energy flow

After each step of the downscaling and at the end of the pipeline, DSCALE performs a thorough consistency analysis to ensure that the energy flow is consistent. It first ensures that in the end-use sectors, the fuels sum-up together is equal to the carriers, and that carriers sum-up together is equal to the sectoral total. It also looked at consistency between the power sector and the electricity consumption in the end-use sectors, and creates a T&D losses, conversion to synthetic fuels and net exports to capture all electricity which is generated but not directly consumed as final energy, whether due to T&D losses, conversion to indirect electrification (e.g. hydrogen), or exported.

Deriving CO₂ emissions for the power and end-use sectors

Country-level CO₂ emissions for key demand and supply sectors are first derived from IAM regional data using downscaled energy consumption as the primary proxy. For each sector, country-level energy by fuel type (coal, gas, oil) produced in the energy downscaling steps is combined with fuel-specific emission and efficiency factors to estimate an initial country-level emissions profile. These country estimates are then scaled proportionally so that their sum within each IAM region matches the regional IAM trajectory, preserving the top-down regional constraint while distributing emissions in proportion to each country's energy mix.

The derived country-level emissions are then harmonised to match IEA historical data: for each country and sector, the pathway is shifted by the absolute difference between the IEA observation and the model value at the harmonisation year, then blended back to the downscaled trajectory by 2050. Fuel sub-sectors are subsequently rescaled proportionally to preserve the original fuel mix shares while remaining consistent with the adjusted sector total. It is worth noting that here, we prioritise country-level historical accuracy over exact IAM regional consistency.

Other emissions

Agriculture sector

The agriculture sector emissions on the macro-region level are collected from variables in the scenario data and harmonised to historical data. The emissions for individual subregions are determined by assuming their shares in the base year are constant over the whole scenario period, a simple downscaling methodology called base-year pattern.

Industrial processes, waste and energy emissions

The macro-region emissions from industrial processes, waste and non-CO₂ emissions in the energy sector are taken from variables in the IAM scenario data and harmonised to historical data in a base year, as for the agricultural sectors described in the previous section. To perform the downscaling from macro-regions to the subregions, a methodology based on intensity convergence is used; more specifically the Impact, Population, Affluence, and Technology (IPAT) method as developed by Van Vuuren et al. (2007)⁸8 and extended by Gidden et al. (2019).⁹9 It assumes that emissions intensities (i.e. the ratio of emissions to GDP) will converge from their values in the historical base year to the macro region intensity in the last year of the scenario data, in the year 2100. This is made possible by an exponential interpolation of emission intensities from the base-year to the convergence year. Together with the annual GDP by the given scenario, this interpolation defines how the emissions of the macro-region are shared amongst the countries.

Land-use, land-use change and forestry

For further details on our previous work in the land-use, land-use change and forestry sector, see here.

Economy-wide GHG and CO₂ emissions pathways

After the emissions projections of the illustrative pathways have been downscaled to the individual countries using the methods described in the previous sections, the 4 sectors (Agriculture, Energy, IPPU and Waste) are aggregated to consistent economy-wide emissions pathways. For each country, we show the total GHG emissions in CO₂e emissions based on 100-year global warming potentials from the Sixth Assessment Report of the IPCC and the Highest Possible Ambition Scenario, as well as the CO₂ only emissions.¹⁰10 Total CO₂ emissions cover energy CO₂ emissions and non-energy CO₂ emissions. We also allow a toggle to include or exclude LULUCF emissions, for countries where 1.5°C compatible LULUCF pathways have been produced. Currently this data is only provided out to 2035, but we plan on extending this time horizon as we continue to develop downscaling methodologies to produce country–level LULUCF pathways.

Current policy projections

For all countries assessed by the Climate Action Tracker, current policy projections data are from the Climate Action Tracker’s country updates, along with historical emissions data. See captions under graphs for each country for the full reference.

For countries not assessed by the Climate Action Tracker, current policy projections are taken from national documents and modelling or the European Environment Agency as available. For other countries, see captions under graphs for each country for the full reference.

NDCs and domestic targets

For all countries assessed by the Climate Action Tracker, unless otherwise specified, Nationally Determined Contributions or domestic targets are from the Climate Action Tracker. See the captions under graphs for each country for the full reference.

For countries not assessed by the Climate Action Tracker, please see the information box under the graph for a complete reference.

Historical emissions

For all countries, a historical time series of emissions excluding LULUCF from 1990 to the latest available year is provided, expressed in global warming potentials from the IPPC’s Fifth Assessment Report.¹¹11

For countries assessed by the Climate Action Tracker, the time series is taken from the latest available update and extended using growth rates from the dataset ‘PRIMAP-Hist’ when data years are missing (see captions under graphs for specific dataset version).¹²12 For other countries, historical emissions are taken either from country reported data to the UNFCCC, the dataset ‘PRIMAP-Hist’ based on country reported emissions, or using a combination of sources.

Additionally, in the current situation section, a snapshot of the sectoral and gas breakdown of a countries’ emissions is provided for a single year. The following decision tree is applied for the selection of the underlying data:

• For Annex I countries, we use the dataset provided by the country reported data to the UNFCCC which allows for more granularity in the breakdown of emissions per sub-sector.

• For non-Annex I countries, we use PRIMAP-Hist dataset for non-LULUCF emissions. Combustion emissions are then broken down to sub-sectors using data from the IEA. LULUCF emissions are provided by the Climate Action Tracker, when available, and from country reported sources when available.

• When available, for some countries, country-specific data is used as specified in captions under country graphs.

The table below shows how emissions reported in national inventories are mapped to different sectors in the 1.5NPE.

IPCC 2006 guidelines	1.5 national pathway explorer sectoral emissions
1. Energy	Energy
1.A. Fuel combustion activities
1.A.1. Energy industries
1.A.1.a. Public electricity and heat production	Energy/Power
1.A.1.b. Petroleum refining	Energy/Fossil fuel industry and others
1.A.1.c. Manufacture of solid fuels and other energy industries	Energy/Fossil fuel industry and others
1.A.2. Manufacturing industries and construction	Energy/Industry (energy-use)
1.A.3. Transport	Energy/Transport
1.A.4. Other sectors
1.A.4.a. Commercial/institutional	Energy/Buildings
1.A.4.b. Residential	Energy/Buildings
1.A.4.c. Agriculture/forestry/fishing	Energy/Industry (energy-use)
1.A.5. Other	Energy/Other
1.B. Fugitive emissions from fuels	Energy/Fossil fuel industry and others
2. Industrial processes and product use	Industry (processes)
4. Land use, land-use change and forestry	LULUCF
5. Waste	Waste
6. Other	Other

For all countries, full references are provided in captions under graphs on country profiles.

Historical energy data

Historical primary energy consumption, power sector generation and final energy consumption in end-use sectors (both in absolute and shares) are taken from the International Energy Agency (IEA) World Energy Balances.¹³13 The version of the data set used varies across country updates, indicated in captions under the respective energy mix graphs for each sector.

Historical sectoral CO₂ emissions and power carbon intensity

Historical CO₂ emissions from power and end-use sectors are based on IEA datasets, specified in the captions under graphs for each sector. Power sector carbon intensity is calculated using power sector generation from the IEA World Energy Balances and power sector CO₂ emissions from other IEA sources (see power energy mix graphs for specific datasets).

Projected emissions and energy consumption

See the sections above on the HPA scenario and downscaling method.