The Climate Action Monitor 2024

Annex A. Data gaps, methodology and limitations

Chapter 1

GHG emissions estimates

Official country-level data are based on emissions inventories reported to the United Nations Framework Convention on Climate Change (UNFCCC). These inventory data are compiled using territory-based and production-based principles following the Intergovernmental Panel on Climate Change (IPCC) guidelines. Since data are not available for all countries or all years, missing data are estimated to determine global aggregates for some countries. The data sources and methodology used in this publication are summarised below:

• GHG emissions data from national inventories are available for all OECD countries that report annually to UNFCCC for the period 1990-2022 (i.e. all Annex 1 countries and South Korea).

• Data for other OECD countries (formerly referred to as “non-Annex 1”) are obtained through the OECD GHG emissions questionnaire. However, the time coverage is not complete. For example, Colombia covers data up to 2018, Costa Rica up to 2017 and Mexico up to 2019. There are also gaps for Israel before 2002.

• For OECD partner countries, many gaps remain. For example, official emissions data for 2020 are not available on the UNFCCC GHG emissions data interface. Major gaps are also present for large emitters, such as the People’s Republic of China (hereafter “China”) and India. China has provided official data for only five years (1994, 2005, 2010, 2012 and 2014), while India has presented data for only four years (1994, 2000, 2010 and 2016). There are also important gaps for Peru (for 1990-2010), Saudi Arabia (presenting only four years between 1990 and 2012), South Africa (for 1990-2000) and Indonesia (for 1990-2000).1

• In this report, when official data were not available, estimated data are used to compile global aggregates.

To deal with data gaps, the OECD developed a methodology to impute missing values in a consistent manner maintaining coherence with official reports (Cardenas et al, 2024). The methodology uses PRIMAP-hist national historical emissions time series adjusted to official estimates to impute values. The methodology is as follows:

• If all official series are missing, the PRIMAP series is used.

• The interpolation between two official data points is carried out using PRIMAP trends and rescaled ensuring the compiled series lies through all the official data points.

• If at least two non-missing official values are available, linear interpolation is performed.

• Extrapolation is executed by adding the PRIMAP level difference to the latest (or earliest) official point.

• Total, including land use, land-use change and forestry (LULUCF), is obtained by adding LULUCF and Total, excluding LULUCF estimates. The other remaining category levels are computed by multiplying the sectoral share with the total emissions estimates.

The methodology allows for the construction of a complete GHG emissions dataset from 1990 to 2022 covering the 198 UNFCCC Parties. It includes CO₂, N₂O, CH₄, SF6, NF3, HFCs, PFCs. Total GHGs are computed as a sum of the estimates of all gases and a breakdown by the six main IPCC categories: Energy, Industrial Process, Agriculture, LULUCF, Waste and Other is provided. Finally, the estimates are expressed in terms of several Assessment Reports: SAR, AR4, AR5, AR6. This was made possible by recording the Global Warming Potential (GWP) in which the data were reported, and by applying a mapping table containing conversion factors to calculate levels in all possible GWPs of individual gases.

Table A A.1. Official GHG emissions data availability per year, country level

Countries	Official data
Annex I OECD countries	1990-2022
Chile	1990-2020
Colombia	1990-2018
Costa Rica	1990-2017
Israel	1996, 2000, 2002-22
Korea	1990-2021
Mexico	1990-2019
OECD partner countries	Complete official data only for 2010
China (P.R. of)	1994, 2005, 2010, 2012, 2014
India	1994, 2000, 2010, 2016
Argentina	1990-2020
Peru	2008-19
Saudi Arabia	1990, 2000, 2010, 2012
South Africa	1990, 1994, 2000-2020
Brazil	1990-2016
Bulgaria	1990-2022
Croatia	1990-2022
Indonesia	1990-1994, 2000-16
Malta	1990-2022
Romania	1990-2022

Source: UNFCCC, GHG emissions inventory, BURs and (OECD, 2023[1]).

GHG emissions targets

To compare NDC emissions targets of countries, the OECD developed a methodology to estimate GHG emissions targets for 2030 as physical emissions targets (Pizarro et al., 2024[2]). The methodology is as follows:

If the country NDC target is defined for 2030 in physical terms (i.e. in tonnes of CO₂e), the target is maintained.

Otherwise, targets are put into physical emissions levels for 2030 by applying transformations to country-level data to derive 2030 emissions targets in physical terms; these include adjustments for GDP, population and other transformations presented in OECD (2024[2]).

Countries' NDC targets are not directly comparable due to differences in coverage of sectors, gases and GWP. For comparison, emissions and targets were adjusted to a common scope and AR6, and covering all GHGs as follows:

• For any given country, only the GHG emissions for the gases and sectors implied in its NDC are considered and converted into AR6.

• Physical targets in the scope reported by an individual country in their NDC are converted into physical AR6 targets by multiplying the target by the ratio of the country’s total GHG emissions for 2022 in AR6 over the GHG emissions in the scope of the original target.

• GHG emissions outside of scope are imputed. This applies to China only as it uniquely covers CO₂ in its NDC. To this end, non-CO₂ gases are projected within a range. A lower bound assumes these gases are reduced at the same percentage as CO₂. An upper bound assumes these gases continue to maintain the same trend as in 2010-22.

• For non-IPAC countries, 2030 targets are calculated as a residual from global emissions (from UNEP, 2023), while 2050 targets are calculated as a residual considering only countries with no net-zero target.

To calculate emissions based on countries’ net-zero targets, a complete achievement of these targets as presented was assumed. Global emissions for 2030 were taken from UNEP (2023[3]), with the 2030 value used for emissions under unconditional NDCs. In addition, 2050 emissions were projected based on the assumption of countries achieving their net-zero targets as proposed. Emissions for countries without a net-zero target were assumed to be at the same level as 2022, with a range of +/- 10% (e.g. Iran, Azerbaijan and others).

Moreover:

• EU-27 countries were considered as a bloc that reaches climate neutrality by 2050, as per the EU Law; individual countries’ targets differ since some do not have a net-zero target.

• Emissions for countries with a net-zero target by later than 2050 are estimated based on a linear trajectory to their target year (e.g. China by 2060 and India by 2070).

• Emissions were adjusted based on differences in LULUCF estimates (according to (UNEP 2023[4]), differences among the bookkeeping method and national inventories were about 6.4 Gt CO₂e, assumed to remain constant by 2060). Projections are also adjusted by international aviation and shipping, which account for 4-5% of global emissions (IEA, 2023[10]). International aviation and shipping are assumed to be 0 by 2050 as per the international agreements by the International Air and Transport Association and the International Maritime Organization; this reduction is assumed to be linear from 2022 to 2050.

• GHG emissions for 2022 are calculated using data from PRIMAP and refined by the OECD. They refer to aggregate emissions for all 198 UNFCCC parties, including LULUCF; 2030 and 2050 emissions consistent with 1.5°C increase pathways were taken from UNEP (2023[3]).

Table A A.2. Detailed estimates of GHG emissions targets for 2030 and 2050 (Gt CO2e)

	2022	2030	2050
UNEP estimated global emissions under current policies*		57 (53-59)	56 (25-68)
Global emissions if targets achieved	55.5 (52-63)	55 (54-57)*	21.1 (18-24.7)
- LULUCF adjustments (outside inventories)	6.4 (5.2-7.6)	6.4 (5.2-7.6)	6.4 (5.2-7.6)
- International bunkers (aviation and maritime)	4.1	5.2 (4.7-5.7)	0
Total countries (in GHG invent., incl. LULUCF) if targets achieved	45.1	43 (41-45)	14.7 (12.6-16.8)
IPAC	37.3	32.2 (31.6-32.8)	7.8 (7-8.6)
i. OECD	13.3	9.4	0.2 (0.1-0.3)
ii. IPAC non-OECD	24.2	22.7 (22.2-23.4)	7.5 (6.8-8.2)
Rest of the world	7.6	10.7 (9.4-12.2)	7.0 (5.8-8.3)
1.5 °C consistent*		33 (26-34)	8 (5-13)
1.5 °C Emissions gap		22	13.2
2 °C Consistent*		41 (37-46)	20 (16-24)
2 °C Emissions gap		14	1.1

Source: *Projections taken from (UNEP, 2024[4]).

Note: LULUCF=land use, land-use change and forestry. Emissions are expressed in terms of Gt CO₂e, using AR6 GWP. Global emissions (in blue) equal total emissions of countries in national inventories (green) and emission adjustments for emissions not accounted in national inventories. Net-zero targets by 2050 assume the full implementation of the targets as presented. Emissions for countries without a net-zero pledge were calculated as a residual from business-as-usual emissions estimated by (UNEP, 2024[4]). EU-27 countries were considered as a bloc that reaches climate neutrality by 2050. Individual EU countries’ targets commitments differ, however, since some do not have a national level net-zero target. Further details are available in 0.

Chapter 2

The OECD set of indicators is based on historical observational data – collected and recorded – that go back as far as 1979. This time period is relatively short for analysing climate change events. Nevertheless, the data, while limited, still show the exposure of climate-related hazards to the population, croplands, forests and urban areas.2 These 44 years of data illustrate that climate change impacts are already visible by analysing even a short period of historical data [for a full discussion, see Maes et al. (2022[5])]. These indicators reflect what has happened, not what will happen (Box 2.1). Nevertheless, the dataset can support countries to understand the evolution and potential impact of climate-related hazards to guide policy choices.

Information on disaster events and their related costs has limitations due to inconsistent reporting by national governments and in international databases. The loss databases are essential to assess policy and monitor progress, but they are hardly ever mandated by national or supra‑national legislations. There are several supra-national framework directives, but they remain vague when it comes to recording losses from disasters; their implementation would hugely benefit from the availability of such information.

No single database has complete coverage of losses from disaster events. For example, the United Nations Disaster Risk Reduction (UNDRR) DesInventar-Sendai database provides a common platform for countries to collect loss data on a national level; however, only ten OECD and OECD partner countries use this database to date (UNDRR, n.d.[6]).3 It is especially difficult to quantify economic losses in harmonisation. Although definitions exist for calculating basic measurements of economic losses, such as affected buildings, agricultural assets and civil infrastructure, this is not consistently done for all disaster events across countries.4 The threshold employed in a database to determine whether an extreme weather event is recorded is also significant; it can generate different results and comparability issues.5

Finally, methods for calculating losses exist in the context of estimating damages in the immediate aftermath of a disaster to anticipate the level of support required by the international community such as the Post-Disaster Needs Assessment (PDNA). To mainstream and standardise the PDNA, the United Nations, the World Bank and the European Commission have jointly developed methodological guidelines. Physical damages and economic losses are evaluated using the Damage and Loss Assessment and human recovery needs are investigated through the Human Recovery Needs Assessment and a Recovery Framework. However, there is no central database to collect the results of PDNAs that had been conducted, except for the countries covered in DesInventar.

Chapter 3

The Climate Actions and Policies Measurement Framework (CAPMF) is an internationally harmonised climate policy database developed by the OECD. It is based on a structured policy typology that tracks a common set of policies with common definitions and harmonised policy attributes on an annual basis. The CAPMF complements other international policy tracking tools such as the reporting frameworks to the UNFCCC.

Based on 130 policy variables, the CAPMF tracks 56 climate actions and policies, which cover 75% of policies listed in the 2022 IPCC report, from 1990-2023 for 50 countries and the European Union. These countries are jointly responsible for over 63% of global GHG emissions. For each policy, the CAPMF measures policy stringency, defined as the degree to which policies incentivise emissions reductions. The CAPMF comprises climate-positive instruments (e.g. carbon taxes), as well as reform of climate-negative measures (e.g. reform of fossil fuel subsidies). The CAPMF also includes some climate-relevant policies such as air pollution standards, i.e. policies whose primary intent is not mitigation, but which have a material effect on emissions. While the focus of the CAPMF is on national climate action, it still includes key subnational policies such as subnational emissions trading schemes and renewable portfolio standards.

For The Climate Action Monitor 2024, missing policy data in 2023 were substituted by the last observed data in the last five years. This concerns 16% (9 of 56) of the policies captured by the CAPMF. Substituting the imputed data by the actual data once they become available is, unlikely to change the key results of Chapter 3. However, some messages may need to be refined, notably as country-level data on fossil fuel subsidies are not yet included.

The primary focus of the CAPMF is to track the evolution and stringency of mitigation policies over time (1990 to 2023). It collects a broad range of harmonised data that is internationally comparable and suitable for a broad-based quantitative and qualitative analysis. However, the CAPMF has some limitations as explained in the chapter and, hence, should be interpreted carefully (Nachtigall et al., 2022[7]).

References

[8] Dottori, F. et al. (2021), “A new dataset of river flood hazard maps for Europe and the Mediterranean Basin”, Earth System Climate Data, Vol. 14/4, pp. 1549-1569, https://doi.org/10.5194/essd-14-1549-2022.

[5] Maes, M. et al. (2022), “Monitoring exposure to climate-related hazards: Indicator methodology and key results”, OECD Environment Working Papers, No. 201, OECD Publishing, Paris, https://doi.org/10.1787/da074cb6-en.

[7] Nachtigall, D. et al. (2022), “The climate actions and policies measurement framework: A structured and harmonised climate policy database to monitor countries’ mitigation action”, OECD Environment Working Papers, No. 203, OECD Publishing, Paris, https://doi.org/10.1787/2caa60ce-en.

[1] OECD (2023), “Air and climate: Greenhouse gas emissions by source”, OECD Environment Statistics, (database), https://www.oecd.org/en/data/indicators/air-and-ghg-emissions.html (accessed on 1 August 2024).

[2] Pizarro, R. et al. (2024), “GHG Emission Trends and Targets (GETT): Harmonised quantification methodology and indicators”, OECD Environment Working Papers, No. 230, OECD Publishing, Paris, https://doi.org/10.1787/decef216-en.

[10] UNDRR (2019), Global assessment report on disaster risk reduction 2019, United Nations Office for Disaster Risk Reduction, Geneva, https://www.undrr.org/publication/global-assessment-report-disaster-risk-reduction-2019.

[6] UNDRR (n.d.), United Nations DesInventar Open Source Initiative, website, https://www.desinventar.net/ (accessed on 28 July 2023).

[4] UNEP (2024), Emissions Gap Report 2024: No more hot air … please! With a massive gap between rhetoric and reality, countries draft new climate commitments, https://www.unep.org/emissions-gap-report-2024.

[3] UNEP (2023), Emissions Gap Report 2023: Broken Record – Temperatures hit new highs, yet world fails to cut emissions (again), United Nations Environment Programme, https://doi.org/10.59117/20.500.11822/43922.

[9] WMO (2021), Atlas of Mortality and Economic Losses from Weather, Climate and Water Extremes (1970–2019), https://library.wmo.int/index.php?lvl=notice_display&id=21930#.Y-5nmi_MLIU (accessed on 16 February 2023).

Notes

← 1.  Additional data may be available through other sources such as national statistics websites. IPAC is exploring different alternatives to fill data gaps, including desk research to examine different sources and developing statistical methods.

← 2.  The development of this set of indicators is informed by standards from the World Meteorological Organization, the US National Ocean and Atmospheric Administration, latest research and standard developed by well-recognised organisations and builds on international frameworks for assessing climate‑related hazards. Notwithstanding, over- or under-estimations of the actual exposure to climate-related hazards may occur. Further details on these limitations can be found in Maes et al. (2022[5]).

← 3.  Several databases gather secondary data on disaster occurrence and their human and economic cost. These include the EM-DAT database from the Centre for Research on the Epidemiology of Disasters’ Emergency, hazard-specific databases (e.g. the Dartmouth Flood Observatory), actuaries and re-insurers (e.g. MunichRe’s Natcat-SERVICE and SwissRe’s Sigma databases). A weakness of such databases is the heterogeneous nature of the secondary data it relies upon; it often has problems with comparability across countries.

← 4.  For example, EM-DAT collects data on costs that accrue directly to assets, but it does not account for costs that accrue from business disruption in areas directly affected by the disaster; conversely, this is included in SwissRe’s Sigma database. Beyond this, more intangible losses such as the impact on health or the environment, as well as cultural heritage loss and loss of reputation, are hardly ever accounted for due to difficulties in monetisation.

← 5.  For example, the WMO World Atlas report compares human loss figures from nationally reported data in two databases (DesInventar and EM-DAT) with different thresholds across Colombia, Ecuador, Indonesia and Niger (WMO, 2021[9]). It concluded that different threshold levels between the two databases did not affect the reporting of intensive (high-intensity, low-frequency) disasters (WMO, 2021[9]). However, the two databases had varying thresholds for extensive disasters (low-intensity, high‑frequency), responsible for most economic losses from disasters (68.5% between 2005‑17). Therefore, extensive disasters were sometimes included in one database but not the other (UNDRR, 2019[10]).