Report Overview Analysis

The SOCCR2 assessment is a valuable overview of the available data and the current state of knowledge about the global and the North American carbon cycle, drawing upon a very large and diverse body of scientific research. The Committee found that many aspects of this draft report were well done, and also found that many aspects could be improved. Many of these issues are highlighted below, framed around the specific questions in the Statement of Task. Further details and examples of these issues are provided in the chapter-specific reviews later in this document.

The audience for SOCCR2 is described in the Preface as “a diverse audience that includes scientists, decision makers in the public and private sectors, and communities across the United States, North America, and the world”. This definition could potentially encompass just about anyone, so in that sense is not clearly described. That said, it is a standard audience definition for these sorts of assessment reports, thus refining further may not be a critical priority.

The primary goals for SOCCR2 are articulated in the three questions listed below (taken from the Summary, p.21). The Committee finds that the draft report has mixed success in responding to these questions.

Geographic limitations. The draft report is ambiguous as regards its geographic scope. The assessment aims to address North America as a whole—including U.S., Canada, and Mexico—but the actual report content addressing Canada and Mexico is very spotty and inconsistent. This inconsistency apparently stems from the decision of Canada and Mexico to pursue their own independent assessment efforts this time, and the limited participation of Canadian and Mexican scientists, at least for some of the chapters. Furthermore, it is stated in the Summary [p.21] that “the geographic scope of the U.S. analysis includes the conterminous United States, Alaska, Hawaii, and U.S. territories”; and this idea is reinforced in Figure ES1. Yet Hawaii and the U.S. territories are not mentioned anywhere else in the document. These regions should either not be called out as an explicit part of the assessment scope, or they should be discussed in appropriate places throughout the assessment.

On the whole, the draft report does a reasonably good job of reflecting the relevant scientific literature; the studies that are cited throughout the report seem well-chosen and accurately described. One issue worth consideration is the balance of attention given in SOCCR2 to terrestrial versus aquatic sciences. A main goal of this assessment work is to advance accounting of carbon sources and sinks of North America, in order to facilitate engagement in policy frameworks that address greenhouse gas emissions (such as the UN Framework Convention on Climate Change). It is thus reasonable that the bulk of attention be paid to continental carbon sources and sinks, especially those that can potentially be controlled through various policies, practices, and technological applications. Carbon sources and sinks in coastal waters and the open ocean cannot be claimed as part of the “emissions inventory” of any one nation—and one could thus argue this topic is relatively less important to the SOCCR assessment efforts. However, ocean system dynamics are such a critical part of the global carbon cycle that they must be carefully assessed and understood. In particular, accurately quantifying cumulative ocean carbon uptake is critical for constraining the carbon budget overall, since the magnitude of the terrestrial sink is not independently constrained.

There are also a number of topics that the Committee sees as “critical content” that are not well captured in this assessment, listed below.

• Because the field has been evolving so rapidly, the treatment of methane should be expanded to include 2016 and 2017 papers on trends in U.S. methane sources and sinks.
• The report lacks coverage of relevant economics research, and of integrated modeling studies that encompass emission drivers, climate and other physical and biological consequences, and resulting economic effects (discussed below).
• The report needs more discussion of how climate change, especially of changes in precipitation patterns (seasonality, intensity, duration) may influence carbon cycle dynamics.
• Studies of trends in atmospheric O2 concentration should be added as an “incontrovertible line of evidence” regarding the role of fossil fuel burning in atmospheric CO₂ trends.
• Arctic region coastal zones (Alaskan/Canadian) should be given attention, given their potential importance as a major biogenic carbon source.
• The report would benefit from a clear explanation of how humans can influence carbon dynamics at global scales—since this is a question often raised in public debate about climate change (i.e., the argument that humans cannot possibly cause global-scale changes).
• The report needs more discussion about the practical opportunities for effective management of carbon sources and sinks (discussed below, under question #7).
• Given the paucity of peer-reviewed publications regarding the state of the carbon cycle on tribal lands, Chapter 7 could instead explore how to more actively engage and support indigenous communities, in harnessing their traditional practices and knowledge to inform polices and programs that affect carbon flux.

Most of the chapter findings are well-documented, in a reasonably transparent and credible way. However, in many places the authors simply offer a citation as justification for a particular conclusion or important analysis result. To strengthen and clarify these points, wherever possible (when the statement is an important one), the report should include a brief summary of the critical evidence in that citation—given that very few readers will chase down the original publications.

One place that raised concerns is Key Finding 2 in Chapter 13, which bases major conclusions on the results of original data analyses by the chapter authors. To maintain the credibility of the report as a review and summary of current knowledge, their results should be compared to related findings in the published literature, or else not presented.

The presentation of key findings throughout the draft report could be improved, for example:

• The findings are unbalanced among chapters in terms of their detail and clarity. Many of the findings seem “forced” in that they are obvious statements that do not offer specific new insights or information, or convey a clear message relevant to the topic of the chapter. (Several examples are cited in the chapter summaries). If this practice flows from an editorial

requirement that all chapters have some minimum number of key findings, the Committee suggests re-examining that policy.

• Some of the key findings are about methodology, and not about what has been learned. The Committee suggests emphasizing specific advances in understanding of the North American carbon cycle, rather than simply noting that “understanding has improved”. Also, establishing confidence levels should be limited to estimates of fluxes and inventories, or to advances in understanding; not to methodological issues (e.g. the fact that data and models do not agree). For example, Chapter 2 Finding 5 assigns high confidence to the finding that top-down and bottom-up estimates of carbon sinks are comparable, despite the large uncertainties of the two approaches. Similarly, high confidence is assigned to the poor characterization of anthropogenic urban methane emissions (Chapter 4, Finding 5), and to the divergence between inverse model and empirical estimates (Chapter 8, Finding 2; Chapter 11, Finding 4). These findings do not advance the readers’ understanding of the carbon cycle; they could be reworded or eliminated.
• There are several places where improvements and additions to the report graphics are needed. Specific suggestions are made in the individual chapter reviews.
• Research recommendations presented as key findings could be coordinated across the chapters to articulate a specific research agenda for the next decade.

The draft SOCCR2 assessment does not present many new statistical analyses or syntheses (with the exception of Chapter 13), but instead provides more of a synthesis of existing work in the published literature. Overall, this seems to be done reasonably well, but there are places where data are inadequately described or are used as the basis for questionable conclusions, and where their handling is inconsistent across chapters. For instance:

• Many of the carbon budget numbers presented in the draft report provide no uncertainty values. Furthermore, many of the numbers are presented with 3 or 4 significant digits, and thus overstate the confidence anyone should have in them. The authors must convey confidence and uncertainties more explicitly and more consistently across the report. Expressing uncertainties in model projections can be particularly challenging of course, and the variations among model results should be included together with the median where possible (e.g. Figure 19.8).
• A general problem throughout the draft report is the inconsistency in units. For instance, when discussing carbon fluxes and energy issues, the authors use Pg, Tg, and other units of carbon, CO₂, CO₂-e, and CH₄—making it difficult for most readers to compare values across different chapters, figures, etc. The Committee recommends this be standardized across chapters, with TgC as the key unit (or include TgC in parentheses) whenever possible.
• The draft report is inconsistent in terms of the time periods that are used in different figures and different parts of the discussion. In particular, the end date of observational records is highly variable across the chapters. Of course this stems largely from differences in the actual observational data records available, but the authors should use a consistent time period, where possible, across the critical data records shown in the report.

• Another inconsistency is found in the different references to soil depth, which changes from one chapter to another. This may be constrained by the observational data used in the syntheses. As there is little link across the chapters regarding soil depths, explicit mention of depth should be included where possible in the text, tables, and figures.
• It would be helpful to the reader if the discussion of carbon sources, sinks, and flows in individual chapters were placed into a common framework that provides context relative to the overall carbon cycle. Figure ES2 nominally provides this sort of overview, but it is not linked to information provided in the report chapters. Representatives from across the different chapter teams could be convened to construct a diagram that puts all the different types of flux estimates made in different chapters into one framework.

Most of the chapters are well written, comprehensive, and well organized, presenting useful information drawn from appropriate sources. In some chapters, however, it is hard to “see the forest for the trees” because small details are often mixed with major concepts with little to distinguish between them, and also because much information is not put into a useful context with respect to the overall carbon cycle. This points to a general concern of the Committee about insufficient integration of some key topics into the overall assessment. The Committee acknowledges the real challenges for fostering seamless integration among numerous related topics in an assessment process such as this one, where each chapter is produced by an independent team of authors, but greater attention to these integration concerns is needed to assure that the overall report is more than just the sum of its individual pieces.

Some topics seem to be treated as isolated subjects that are not connected clearly to the main focus on understanding the carbon cycle. Likewise, some biological, physical, and societal processes that are in fact highly coupled are treated as isolated subjects. For instance, the report should discuss how physical environmental changes can affect key biological systems (e.g., how warming temperatures and changing precipitation patterns might affect the carbon emissions from certain terrestrial and aquatic ecosystems). It is likewise important to illustrate how energy policy and technology decisions can affect other sectors (e.g., how expanding biofuel production can affect the management of grasslands, forestry, agriculture), and how energy use contributes to the carbon budget overall.

Integrated assessment analyses—which consider the social and economic factors driving greenhouse gas emissions, the biogeochemical cycles that determine the fate of those emissions, and the resultant impacts on climate and human welfare—provide a framework for looking more holistically at the pieces addressed in different parts of the SOCCR2 report. Adding discussion of this research literature could thus greatly help convey the crucial role of carbon cycle science in environmental management.

A particular concern is Chapter 6 (Social Science Perspectives on Carbon) and Chapter 7 (Tribal Lands) and their integration into the rest of the report. Carving out two separate chapters, with the exclusion of economics from social science, and the neglect of discussion regarding the particular challenges and opportunities pertaining to carbon fluxes on tribal lands, make the chapters read like “add ons” to the assessment. Also, useful social science insights are overlooked in other relevant parts of the report. For instance, the Energy chapter should include consideration of how social and behavioral science insights are critical for the design of measures to encourage the adoption of energy efficiency practices and technologies. Major changes to report organization to include these issues may be infeasible at this late stage, but in planning the next round of SOCCR assessments, the organizers should consider possible alterative models wherein social science research findings and needs that

consider the diversity of populations and institutions are woven throughout relevant chapters of the report.

Noted below are some other improvements that could be made to the report. Most of these issues are discussed in greater depth in the chapter-specific review material.

• The Committee recommends adding a figure to Chapter 3 (Energy Systems) that shows the U.S. energy flow chart and associated CO₂ emissions (e.g., see: http://flowcharts.llnl.gov).
• The Committee recommends reconsidering the idea of assigning confidence levels to direct factual information—such as the atmospheric concentrations of greenhouse gases. Saying there is “high confidence” that atmospheric concentrations of CO₂ and CH₄ are increasing actually undermines the fact that this is an incontrovertible observation. Where appropriate, confidence levels could be ascribed to specific parts of a finding, rather than to the finding overall.
• The use of terms “C uptake”, “C sequestration”, “C emission of –xx Tg”, “C sink of –xx Tg”, “C sink of xx Tg” are used variably (and defined very loosely) through the report, particularly with respect to discussion of forests, soils, and agriculture. Efforts should be made to standardize the definitions and usage of these terms across all chapters.
• In the discussions of future emissions scenarios (Chapter 19), it would be helpful to consider a wider array of scenarios, including ones that reflect major efforts to avoid a 2°C temperature increase—i.e., what sorts of changes to the carbon cycle (reducing particular emissions, enhancing particular sinks) would be needed to achieve this scenario.
• The draft report Executive Summary should be more concise and more accessible to a general audience. The many instances of technical jargon and confusing wording could be improved with the services of a good science writer.
• Some of the chapters seem longer than necessary, and this may stem in part from situations where the chapter authors did not have clear guidance on what information is most critical to convey in that chapter, and thus they included everything that seemed potentially relevant. For instance, Chapter 3 (Energy Systems) could be organized much more tightly by focusing more squarely on the issue of energy as a source of carbon emissions.
• Much of the draft SOCCR2 assessment focuses on elucidating fundamental physical and biological aspects the carbon cycle, and the challenges of accounting for all major carbon stocks and flows for the North American continent, which is appropriate given the mandate of the U.S. Carbon Cycle Science Program. A major concern of the Committee however, is that the report does not also provide adequate explicit discussion of carbon management issues—that is, the actions that can be taken to reduce carbon emissions (e.g., through new energy policies and technologies) and to enhance carbon sinks (e.g., through effectively managing certain terrestrial or aquatic coastal habitats). The draft report’s Executive Summary has some discussion about carbon management needs and challenges, but this does not seem to be based on material found in the body of the report. These shortcomings are important because a central reason why scientists study the carbon cycle is to help inform efforts to manage carbon source and sinks, and to identify the “levers” where effective changes can be made. The lack of

coherent discussion of such issues limits the usefulness of the report for people who are making governance and management decisions that can affect carbon sources and sinks.

• In the discussion of research needs, some of the chapters provide long generic laundry lists. It would be more helpful to instead offer some sense of prioritization, highlighting critical research advances that could most feasibly be made in the coming years.

This page intentionally left blank.