5

Recommendations

Three converging trends motivated this report: (1) rapidly increasing demand from a range of users for trusted information about greenhouse gas (GHG) emissions across multiple sectors and geographic scales; (2) development of many new approaches for quantifying GHG emissions information that aim to address this increasing demand; and (3) a growing and rapidly evolving institutional landscape, including public, private, and academic entities seeking to provide better GHG emissions information. As described in the previous chapters, these trends have cultivated an exciting moment of innovation and improved capabilities. At the same time, this evolving landscape can be complicated and confusing to navigate. In the following recommendations, the Committee strives to find a reasonable balance between more effectively coordinating the large amount of GHG emissions information that currently exists, encouraging the creative energy and urgency to quickly advance new capabilities, and establishing the need for approaches and institutions that will lay the groundwork for trusted and useful information.

Current approaches for quantifying GHG emissions—mostly activity-based approaches that utilize activity data as representative indicators to calculate GHG emissions and atmospheric-based approaches that use measurements of atmospheric concentrations to infer emissions information—have been well developed in the research community and are being used, to some extent, to inform decision making. However, there is an opportunity to improve current approaches to maximize the adoption and usefulness of the information in response to needs from decision makers. For example, some activity-based approaches have been developed and refined for years and thus have well-established protocols that have been repeatedly documented, evaluated, and validated; however, some long-standing data gaps and uncertainties have proven hard to resolve. Although some approaches are well developed, they may not be easily adopted due to various constraints. For example, stakeholders in countries within Latin America, Asia, Africa, and Oceania (excluding Japan, South Korea, Singapore, and Israel) often lack sustained funding to develop and maintain technical capacity for regular GHG emissions information development and reporting.

In this chapter, the Committee introduces recommendations to both enhance current GHG emissions information development capabilities and strive for hybrid approaches that would optimize the integration of individual activity- and atmospheric-based approaches in order to provide

the best available GHG emissions information for users. The six pillars introduced in Chapter 4 underpin the recommendations and their applications, described below. The previous chapters have considered the full range of spatial scales relevant to users of GHG emissions information: global, national, regional, local, and facility. The Committee’s recommendations are intended to apply to this spectrum of geographic scales with a minimum common focus on the national to global scale. Furthermore, hybrid approaches, detailed below, could have the benefit of integrating GHG emissions information across all scales, improving both the provisioning of GHG emissions information and decision making. Lastly, the committee recognizes the varying capabilities and constraints of different members of the global community and that the different decision-making phases—planning, tracking, and assessment and verification—will have different needs.

Figure 5-1 presents a conceptual overview of how the pillars described in Chapter 4 provide a framework for the evaluation of GHG emissions information and approaches underlying their development, and how this relates to informing different stages of decision making in the context of a policy goal. Decision makers often have different GHG emissions information requirements and needs in each phase. While much of the scientific research in GHG emissions information focuses on method development for assessment and verification, many decision makers across scales are in the planning phase. Decision makers are assessing what they need to do, rather than what they have done. In this phase, they need to understand the GHG emissions information available to them so they can adequately plan and execute appropriate climate mitigation actions. The pillars underlie the knowledge-to-action circle depicted in Figure 5-1 in which the development of GHG emissions information would strive to satisfy the pillars in order to be actionable by decision makers in support of their policy goal. Furthermore, different pillars may be more important during different decision-making phases. Thus, the framework allows decisions makers to consider what phase they are in and which pillars will best suit their needs.

In general, inclusivity and equity remain an issue to decision makers at all spatial scales. Although low-income countries and communities share minimal responsibility for current and his-

toric anthropogenic GHG emissions, in general, the impacts of climate change disproportionately affect countries within Latin America, Asia, Africa, and Oceania (excluding Japan, South Korea, Singapore, and Israel). Thus, these countries must often prioritize climate adaptation rather than mitigation, including the quantification and attribution of GHG emissions. In building a globally coordinated GHG emissions information system, attention and investments to engage the global community are vital. Centrally, implementation of the recommendations that follow should include participation, communication, and knowledge exchange among scientists, users, and decision makers globally, particularly from Latin America, Asia, Africa, and Oceania. While these recommendations include specific points about enhancing global inclusivity in the development of GHG emissions information, ultimately, the needs and priorities for adapting to and mitigating climate change must be determined by countries and regions themselves.

The Committee expects that these recommendations will be implemented by a wide diversity of actors in the global community: local municipalities to national-scale governments with varying capacities, the United Nations Framework Convention on Climate Change (UNFCCC) and other international bodies, individual companies seeking to understand their own emissions, and large philanthropies and nongovernmental organizations seeking to develop crosscutting products. With this in mind, recommendations are offered that set clear directions that would be relevant across multiple audiences, and empower different entities to identify specific implementation steps within their means and to meet their needs.

Advancing Greenhouse Gas Emissions Information Capabilities, Trust, and Accessibility

Greenhouse gas emissions information development and evaluation should strive to align with the six pillars: usability and timeliness, information transparency, evaluation and validation, completeness, inclusivity, and communication.

The “pillars” presented in Chapter 4—usability and timeliness, information transparency, evaluation and validation, completeness, inclusivity, and communication—provide a way to evaluate individual emissions datasets and approaches, as illustrated through the case studies. Not surprisingly, there is a lot of variation among different approaches in meeting these criteria. More established protocols—for example, emissions reporting as part of the UNFCCC process—have put in place some standards around transparency and evaluation. New innovative ways are continually being identified to address challenges with data access or uncertainty quantification, but most of these approaches have not yet been rigorously evaluated. Thus, the pillars provide guidance for improving GHG emissions information development and products.

These same pillars embody the desired attributes for the institutions that develop GHG emissions information and the broader aspirations for the global, multiscale endeavor of understanding the sources and sinks of GHGs. Strengthening GHG emissions information to satisfy these pillars at the local and subnational level would help to build international coordination and support. A good practice would be for GHG emissions information providers to self-assess their adherence to the pillars. Such self-assessment would be particularly valuable in cases where multiple sources of GHG emissions information developed for different intended purposes are being used or integrated into a different product with a different intended use. The application of the six pillars of the framework to both individual datasets and approaches as well as the structures that support the development, provision, and exchange of GHG emissions information would advance the current complex GHG emissions information landscape toward one that could more comprehensively meet the needs of users and decision makers.

Greenhouse gas emissions information should be better coordinated (e.g., through the creation of a coordinated repository or federation of repositories) across the global community, enabling adherence to a set of minimum common pillar attributes.

As discussed in Chapter 2, many different GHG emissions information sources have been developed for both scientific and decision-making needs. Currently, some information can be accessed through individual data portals, but is often provided in different formats and varying levels of completeness, largely driven by the specific purpose initiating the data collection or model estimation effort; in some cases, the information is behind paywalls. As a result, it can be challenging to integrate and compare multiple data sources. A coordinated repository or federation of repositories where GHG emissions information can be hosted, documented, and clearly characterized would be a critical step forward in maximizing use and understanding of GHG emissions data products. A mechanism that brings different types of information together could facilitate the integration of multiple types of data at various spatial scales and make the information accessible in a timely manner to decision makers in ways that meet their needs. Such a clearinghouse or federated data center could establish standards and practices that enable decision makers and other data users to clearly and quickly grasp individual characteristics, comprehensiveness, transparency and traceability to source material, and citations for the wide range of GHG emissions information. It could also facilitate input from decision makers on future GHG emissions information development and specifications. To maximize adoption and equity, such a coordinated effort should support and integrate information from existing national and international programs—for example, the UNFCCC reporting program—to leverage, rather than replace, both established and emerging efforts. Countries, particularly developing countries, have spent time and effort to build the capacity needed to calculate reliable inventories using the IPCC guidelines, and there is an opportunity to build on this existing capacity. As a case in point, the International Methane Emissions Observatory is designed to gather information on methane emissions from multiple sources, including national inventories reported to the UNFCCC, corporate reporting, remote sensing data, and measurements of emissions through scientific studies (UNEP, 2021a).

Critical characteristics and functions of a coordinated repository or clearinghouse would operationalize each of the six pillars by including

• Timely information that is transparent and traceable to primary, supporting, and derived datasets;
• Standardization of data formats and metadata to facilitate comparability and interpret-ability across scales;
• Descriptive documentation of models and estimation procedures in nontechnical language, and in multiple languages, that can be used to enable the use of diverse information, including novel observations and methods from the research community;
• Qualitative (e.g., caveats and limitations) and/or quantitative (e.g., uncertainties, error characterization) evaluation metrics;
• Databases of key input data and information (e.g., emission factors, activity data, atmospheric observations, models) that would be regularly updated and widely accessible to facilitate information exchange;
• Governance mechanisms that are coordinated, trusted, and designed to be inclusive of the global community and built on the best practices of data governance and information quality;
• Education modules and capacity building for using GHG emissions information and contributing data and estimation results; and

• Mechanisms to support stronger collaborations between GHG, air quality, and meteorological science communities and stakeholders.

Greenhouse gas emissions information providers should clearly communicate underlying data, methods, and associated uncertainties.

While the information clearinghouse or federated repository effort described above would be a longer-term undertaking for the global community, actionable steps can be taken by data providers in the short term to enhance the transparency of GHG emissions information where feasible. Focused resource allocation or government purchases aimed at bringing data and methods into the public domain with standards on transparency and open principles (e.g., FAIR data principles [findable, accessible, interoperable, and reusable]) could have substantial near-term impacts on the utility of GHG emissions information. Following many of the same guidelines outlined for a clearinghouse and aligning with the pillars, data providers have the opportunity to facilitate comparability and verification of their data and methods to foster trust between information providers and users. In addition, communication is an important consideration in the presentation of datasets to potential users.

As the international community more seriously takes up the need to reduce GHG emissions, decisions based on GHG emissions information will have significant implications for governmental policy making and regulation, business and financial outcomes, and community and household planning. Arguments about the validity of data sources could lead to delay in action and inaccurate information could lead to costly errors in determining how to mitigate emissions. To foster trust in GHG emissions information, transparency of the process and methods of data collection, transformation, analysis, evaluation, and validation is of utmost importance. In particular, transparency is essential for knowledge and resource sharing in the global community, and, specifically, to build capacity in regions with less GHG emissions information development capabilities (e.g., the Global South).

Many atmospheric- and activity-based GHG emissions information products and research efforts span a range of methodological and data transparency standards. While some approaches are documented in the peer-reviewed literature and many scientific journals encourage or require authors to make the data publicly available, the highly technical language used is complex, limiting the transparency of methodologies for users of GHG emissions information. Many atmospheric observations are publicly available, but the analytical tools necessary to use the data may not be. Modeling codes are not often available, and datasets, including those produced by the private sector, are often behind paywalls and may not include information about the sources of the data.

Addressing Key Data and Information Gaps and Uncertainties

Greenhouse gas emissions information (e.g., observations, data analysis, activity data, emission factors) development at more granular temporal and spatial scales with source-level detail should be accelerated to meet the rapidly increasing needs of cities, states, and provinces for managing their emissions.

Beyond the traditional global decision makers—most commonly identified at the international/national scale—are a large and growing community of decision makers and users that have emerged as critical policy actors in the last two to three decades. Cities, states, provinces, landowners, and the business community, among others, are organizing collectively, enacting mitigation policies, and in critical need of consistent, standardized, trusted GHG emissions information. For example,

for governments or the private sector to meet their planned emission mitigation targets, they will need to quantify their baseline GHG emissions, identify the most effective mitigation strategies, and track progress toward meeting their targets.

Where possible, there are substantial gains associated with extending the GHG emissions information development identified thus far in this report to these finer space and time scales, enhancing completeness. Not only does this dramatically increase the user base for the GHG emissions information but it can also improve the accuracy and robustness of GHG emissions estimation at all scales. Furthermore, enhancing source-level detail—for example by characterizing the entire distribution of emission sources—would strengthen completeness of GHG emissions information and better inform mitigation decisions. This is particularly important for methane that has a mixture of point and nonpoint sources, with varying degrees of temporal and spatial emission distributions. Enhancing GHG emissions information itself and communication among decision makers at multiple scales could have a considerable policy benefit in countries with multiscale governance systems. Finally, information at finer space and time scales has the potential to improve methods and pilot new observational capabilities that could be scaled up. Information about GHGs at these more granular spatial scales may also enhance efforts to evaluate and validate national and global scale information. The clearinghouse or federated repository recommended by the committee would be intended to bring information across spatial scales together in a central location or coordinated set of locations.

Currently, data available on granular spatial and temporal scales are insufficient and there is a need to expand the necessary data resources. This includes activity data, emission factors, and atmospheric observations. Methods catered to smaller scales that utilize various datasets to generate GHG emissions information also need to be developed and best practices established. Pilot projects including atmospheric observations currently running at city scales could be extended and expanded to other locations, with particular attention paid to inclusivity and usability to maximize effectiveness and impact.

The accuracy and representativeness of all underlying data used to estimate greenhouse gas emissions (e.g., emission factors, activity data) should be further improved.

Many of the data elements, observations, and models used to estimate GHG emissions rely, sometimes by necessity, on large spatial averages or averages that represent well-observed or high-capacity parts of the globe. The Committee recognizes the need to improve both the specific representativeness and resolution across the globe of these key underlying data drivers to strengthen the completeness and accuracy of GHG emissions information. Examples include emission factors that are often fuel averages (i.e., missing true coal quality variation or biomass variation) or tied to countries with well-quantified fuel characteristics; activity data collected for particular countries but used to calculate emissions for other countries even if the data are unrepresentative; and atmospheric monitoring that may only reflect large-scale integration of information or is biased to locations with high scientific capacity.

Similarly, while emission factors for fossil fuel combustion CO₂ emissions are better known due to more predictable fuel sources and infrastructure types, emission factors for non-energy-sector emissions (e.g., agriculture, forestry, and other land use and waste) or non-CO₂ GHGs such as methane are less well known due to the heterogeneity of the sources and emissions processes. Additionally, emission factors are typically determined for countries in the Global North and may represent averages over large spatial scales (e.g., national, regional), and thus are not representative of locations in the Global South and specific local conditions more broadly. Although activity data may be a larger source of uncertainty for national and subnational activity-based emissions

estimates, particularly in the Global South, unrepresentative emission factors can also result in large uncertainties and errors in emissions inventories developed using activity-based approaches.

Finally, widely used emission factors from databases are often out of date and not regularly updated with new information, and the use of generic emission factors in activity-based approaches can lead to overestimates of emissions. Countries and localities that do not have the resources to invest in updating emission factors specific to their local emission sources are relegated to using outdated information that is biased toward Global North¹ countries. On the other hand, the use of more refined emission factors can lead to underestimates in emissions estimates by not accounting for abnormal conditions. The lack of dynamic updates to emission factor information may also limit opportunities to take advantage of novel approaches being developed. The current emission factor workflow and infrastructure has not facilitated the use of the best available information to date, and there is an opportunity to devote resources in a way that would improve the quality of emissions estimates. For example, at the facility scale, emissions estimates could be improved by employing supplier-specific data rather than employing national- or regional-level data.

Operationalizing Current Capabilities

Greenhouse gas emissions estimation research efforts should transition with urgency to operational capabilities with institutions to maintain and ensure longevity.

As the urgency to immediately reduce GHG emissions is increasing, decision makers likewise need the best-available information about emissions as soon as possible. The typical pace of research and development for new approaches, and for bringing approaches from research into operations, is too slow to meet this demand. Accelerating the transition of research to operations will require scientists, research funders, and data users to identify ways to lower existing barriers to that transition and ways to make new data products more immediately usable. The clearinghouse or federated repository recommended above, along with alignment with the pillars, should help make new GHG emissions information usable more quickly.

Some of the approaches for generating GHG emissions information that have been developed within the scientific community are, or are approaching, a level of readiness that would allow for operationalization, resulting in consistent, reliable information that could be better embraced and exploited by decision makers. New highly granular activity-based approaches and much of the work on atmospheric-based approaches to date have largely been limited to the expert scientific community and have not yet been fully embraced in the decision-making process, even though their potential has been recognized (e.g., IPCC, 2019a).

Some countries already use atmospheric-based approaches to assess their non-CO₂ GHG emissions annually, and the Copernicus Atmosphere Monitoring Service and Global Carbon Project use atmospheric-based approaches to assess global non-fossil fuel CO₂ fluxes annually. Limitations to incorporating the atmospheric-based component of estimating fossil fuel CO₂ emissions and other gases to the operational stage on a global scale are the lack of sufficient observations and a specific infrastructure at one or more institutions that perform the calculations operationally.

Advanced activity-based research efforts have remained in the research realm but similarly show promise within operational systems. Many of these new efforts are availing of new data, sometimes specific to high-income countries, but there is also use of globally available, though potentially costly, remote sensing data. Barriers to operational application include efforts to collect more activity data in countries and regions where limited research into advanced techniques has

___________________

¹ Note: The IPCC Emission Factor Database was created using data based on OECD (Organisation for Economic Cooperation and Development) countries.

occurred. Enhancing inclusivity by enlisting stakeholders to help develop methods and priorities, training and building capacity of underrepresented communities, using citizen scientists to make observations, and leveraging local expertise to design and optimize atmospheric measurement networks would improve atmospheric-based approaches. Furthermore, infrastructure on high-volume data ingestion and processing is essential. Finally, increased granularity is needed in much of the globe in order to achieve data completeness, which, in turn, places emphasis on both ground-based and remotely-sensed information on emitting infrastructure with greater coverage.

Much like weather forecasting systems, optimal approaches will combine a more complex suite of observed data to drive optimal estimates of GHG emissions. For atmospheric-based approaches, this will combine ground- and aircraft-based observations that have high accuracy and sensitivity to surface emissions with satellite observations that have global spatial coverage. A key element may include the measurements of tracers that can attribute specific emission sources. Some of these tracers are already monitored by the air quality community, so the collaboration between the GHG and air quality institutions would support this effort. Current atmospheric observations are not sufficiently representative or dense. More representative and higher density of ground- or aircraft-based observations of GHGs and tracers such as isotopic measurements of CO₂ and methane and ancillary gases are needed to move toward useful and well-constrained national-, state-, or city-scale estimation. More observations of atmospheric transport and dynamics—specifically meteorological parameters and boundary layer height—in the low atmosphere are also needed and would be aided by stronger collaboration with meteorological institutions. Satellite observations also need improvement in resolution, accuracy, and precision and reduced sensitivity to clouds, aerosols, and land surface characteristics. Iteration between scientists and decision makers is needed to prioritize the continuity and deployment of observing systems and to ensure these approaches are integrated with the decision-making process.

Striving for Hybrid Approaches

Greenhouse gas emissions data collection, modeling, and information development should be designed and implemented to enable a fuller integration and “hybridization” of information and approaches.

Most of the current GHG inventory and information development to date has tended to use single methods or approaches with single-technique observations or data sources (e.g., CO₂ mixing ratios, energy consumption data). While this approach was warranted during the development of many of the state-of-the-art estimation techniques, going forward, a “cross-technique” or hybridization of (traditional) approaches and datasets would provide more accurate GHG emissions information by integrating different types of information with more granularity. Some of this work has begun and includes new machine learning and other nonparametric numerical techniques that leverage new data from private and public satellites, sensors, and other types of activities.

Efforts that more fully integrate the traditional activity- and atmospheric-based approaches also present a path toward a more integrated, complementary approach that overcomes gaps and weaknesses in each approach when used in isolation. For example, deeper integration of atmospheric transport models and emissions models or algorithms allow for more direct use of hybrid approaches that need further development and support. This approach is more akin to data assimilation techniques, which have matured in the numerical weather prediction community, among other fields. A single dynamical model can then directly use all observations at higher fidelity—for example, using atmospheric mixing ratios, observed traffic data, and fuel consumption data directly to produce the best estimate of fluxes, calibrated to the uncertainty associated with each observed quantity—is another example of such a hybrid approach. Greater synergy between air quality,

meteorology, and GHG emissions information data collection and analysis efforts would facilitate the development of these hybrid approaches.

Recent developments in multiple dataset integration through advanced ML and related numerical schemes can uncover new statistical relationships and patterns between predictors and GHG emissions data. These techniques also offer opportunities to integrate a wider and more diverse set of data relevant to GHG emissions and fluxes with activity- or source-based inventories. Enabling the development of greater integration and hybrid approaches would require designing data collection to fill the most needed gaps. To strive for hybridization is to holistically improve GHG monitoring across scales, approaches, and capacity. Hybrid approaches that mix complementary approaches, datasets, and models and that integrate the needs of end users promise to offer richer, more usable data outputs for decision making.

Ensuring Usability, Timeliness, and Effective Communication of Greenhouse Gas Emissions Information

Greenhouse gas emissions information generators, decision makers, and global stakeholders should engage in an iterative process in a timely manner to ensure the information provided is relevant and useful.

Incorporating decision-maker input is critical for information developed to respond to the policy needs of stakeholders and decision makers. The time lag to integrate relevant findings from new research into developing empirical- or measurement-informed inventories limits development and execution of sound mitigation policy, and delays transmittal of appropriate market signals for investments and technology development related to mitigation programs by various stakeholders. Usability and timeliness of GHG emissions information can be enhanced if data producers and users engage in an iterative process, which the clearinghouse or federated repository could support, to facilitate investments in systems that are focused on providing decision support and responsive to an evolving policy-making landscape.

The primary motivation for improving the quantification of GHG emissions is to guide decisions—across all scales—that can contribute to rapidly decreasing global emissions. Thus, information needs to be made available in ways that decision makers can readily use. This involves both what kind of information is developed and how the information is presented and described. Improved communication between scientists and researchers working on improving GHG emissions information for end users would enhance communication of results, guide updates to analytical tools to better address users’ needs, and inform priorities for new investments in measurements and analyses that align with users’ needs. Particular attention is needed to the communication and usability of newer atmospheric-based and hybrid approaches. While these approaches have demonstrated powerful examples of source attribution at local and regional scales, as well as the potential to evaluate and improve activity-based emissions information, they have had more limited utility in the emissions reporting and decision-making context.

As more users are utilizing and communicating information about GHG emissions, and as more data sources become available, clear expectations about how to evaluate information will help build literacy and shared understanding. Drawing on the pillars in the Committee’s framework, Box 5-1 outlines key questions that users and communicators of GHG emissions information should pose as they consider how to interpret and use the data.

Concluding Thoughts

The focus of this report has been on analyses and inventories of the anthropogenic emissions affecting atmospheric GHG gases (broadly examined here in terms of all gases and particles affecting the changing radiative forcing on climate). Existing development of GHG inventories has provided significant insights into human-related emissions across all scales. However, as highlighted in the report, there remain major uncertainties in current inventories that limit their use in meeting the extensive, and growing, needs of decision makers going forward. This report highlights a framework for addressing some of these needs, while also mentioning additional issues requiring further consideration. The report will conclude by highlighting some of these issues.

Changes in natural sources—such as those associated with wetlands, permafrost thaw, and wildfires and their associated emissions—can also affect the atmospheric concentrations of CO₂, methane, and other radiatively important gases and particles. In addition, removal processes, or sinks, affect atmospheric concentrations of GHGs. A variety of research studies cited earlier have shown that natural sinks can change spatially and temporally and that they affect the interpretation of emissions inventories. In addition, the changes in climate are themselves affecting natural sources and sinks, with the changing intensity of extreme events being a major factor in these effects. Enhancing understanding of the natural sources and sinks for these GHGs in relation to changing atmospheric concentrations is therefore fundamental to understanding climate change and to the policy responses made by decision makers.

Nonetheless, the integrated overall net emissions have received less focus historically. The focus has largely been on the reduction of fossil fuel emissions, industrially produced (e.g., fluorocarbon) emissions, and emissions from agriculture, land use, and deforestation. Given the importance of land use and land cover change globally, nature-based solutions may be an emerging mitigation and adaptation strategy. A better understanding of the net emissions and associated changes in atmospheric fluxes at all spatial scales is essential for analyzing atmospheric concentration changes for all GHGs and considerations of potential future policy actions. This enhanced understanding will be important to climate plans for all spatial scales, especially with many such planning analyses already beginning to integrate the natural and biogenic components into their inventories; those analyses need more guidance and an enhanced capacity for developing accurate estimates. The global financial sector has an emerging interest and quite possibly new requirements for reporting on GHG emissions and sinks at industrial and business sites and operations.

In addition to the importance of a better understanding of the natural emissions relating to the ocean and unmanaged lands, there is a need for a better understanding of the human-related emissions and sinks associated with managed vegetation. There are especially large uncertainties in net GHG emissions from land use and land-use changes. Improvements in modeling across scales are needed to enhance understanding of emissions and sinks from natural and managed landscapes. Despite an increasing emphasis on capturing these effects in Earth system models, these models still do not capture many important feedbacks and interactions affecting emissions and sinks, limiting understanding of net emissions and the resulting radiative effects.

Overall, significant uncertainties remain in GHG inventories for many parts of the world. Human-related emissions are especially poorly understood for many countries in the Global South. As mentioned earlier, additional observations and analyses are needed to improve emissions inventories for these regions. While low uncertainties may not be needed for every application and decision process, in general, across the planet, reducing uncertainties in the strongest climate forcers (i.e., CO₂, methane, N₂O, and fluorocarbons) is likely important for many policy considerations. A similar emphasis may eventually be needed in reducing uncertainties in inventories for shorter-lived gases and particles. Requirements will likely evolve as decision making evolves.

This report is aimed at analyzing and improving the evaluation of anthropogenic emissions information considered for decision making. The six pillars recommended here provide a framework to evaluate GHG emissions information that can be adapted as GHG emissions information systems become more complex and to serve a range of decision-making needs. In addition, the report includes a number of recommendations toward significantly advancing the accuracy of emissions inventories by better accounting for all available observations and information, while also improving the understanding of the role of human activities in GHG emissions. By examining existing inventories and future needs, the hope is that this report will help push us all forward in assisting the future decision-making process.