Chapter 13: Terrestrial Wetlands

Overview/Main Issues

This chapter focuses on carbon cycling in terrestrial wetlands (that is, non-tidal freshwater wetlands), providing information about area, carbon pool size, and fluxes of CO₂ and CH₄. It discusses carbon stocks and fluxes separately for peatlands (organic soil wetlands) and mineral soil wetlands. The chapter also discusses lateral carbon fluxes from terrestrial wetlands to aquatic/coastal systems. The main findings are that terrestrial wetlands continue to be a large carbon reservoir, and they have been a CO₂ sink and CH₄ source.

Some or most data used in the assessment were derived from the new compilation by the authors and are first presented in this chapter— a different approach from all other chapters, which mostly present and assess data and modeling syntheses in the peer-reviewed literature.

The Committee identifies several areas for improvement in this chapter, discussed below.

Key Findings issues. Four key findings focus on wetland area and carbon stocks, CO₂ sequestration and CH₄ emissions, wetland loss and carbon sequestration function, and future research needs. Some suggestions on improving/rephrasing the key findings include the following:

• Key findings 1-3 only present single values on wetland carbon stocks and CO₂ sink and CH₄ source, without any indication of uncertainties and range. Uncertainty statements are needed.
• The values presented in the Key Finding 2 (18 Tg CH₄/per year) are inconsistent with values presented in Executive Summary (21 Tg CH₄/yr). Likewise, the carbon sink value of 53 Tg/yr is not consistent with the value stated in Executive Summary (nonforested wetlands 36 + forested wetlands 28 = 64 TgC/yr), or with the value of presented in Chapter 2 (36 TgC, on p.78, line 7, and in Figure 2.3). Please update the values and make sure the values are consistent throughout the report.
• Findings 1 and 2 appear to be based on the new compilation by the chapter authors, as indicated in the Description of Evidence section, but without comparison to the estimates from top-down and bottom-up approaches (as presented in Saunois et al. (2016) for the period 2003-2012, among other papers. This chapter is supposed to provide an assessment of current knowledge on wetland CO₂ sink and CH₄ emissions. At minimum, the authors need to put the new estimate presented here in the context of what is available in the peer-reviewed literature. Another concern is that the values derived by the chapter authors do not correspond with a clear time period, as these values represent the mean of various individual measurements collected from different wetland sites over different time periods.
• Chapter 2 does not cite the CH₄ results from this chapter, but instead summarizes the results from both top-down and bottom-up approaches with appropriate uncertainty ranges as in Saunois et al. (2016).
• Key Finding 1 of Chapter 9 (Forests) states that net carbon uptake by North American forests is 217 TgC/yr, with 80% in the U.S. This chapter (Wetlands) claims a carbon uptake by forested wetlands of 39 TgC/yr (Table 13.1). It is not clear if forested wetlands are included in the Chapter 9 uptake estimates. The authors need to coordinate with Chapter 9 authors and state clearly what is included in which chapter. Furthermore, it is not clear if CO₂ fluxes associated

Data compilation issues. Key data representation and data quality control issues related to this chapter, include the following:

• The Committee has concerns that 11 measurements of CH₄ emissions spanning three orders of magnitude, and two values on NEE from MN, WI, WV, MD and West Siberia (Table 13B.1), are averaged together to yield representative fluxes for forested peatlands in Canada and Alaska.
• In Table 13B.2, the 53 measurements on non-forested peatlands include wetlands with diverse hydrologic and biogeochemical conditions as well as sites from a coastal marsh, an estuary, and a tidal creek of Chesapeake Bay in VA (The Terrestrial Wetlands in this chapter should not cover tidal wetlands).
• At least two measurements appear to be from experimental study sites (poor fen – ammonium sulfate added in MN site, and poor fen with water table drawdown at Quebec, Canada site). The CH₄ measurements included in the table range over four orders of magnitude, from 0.0002 to 1.2 Mg C in CH₄/ha/yr. How would this range translate to uncertainties for scaled-up results for North America?
• It appears that the raw measurements/data haven’t gone through quality control evaluation, and as a result, the robustness of the new data compilation results so heavily relied upon in this chapter should not be assessed in the context of SOCCR2 review. Such a new compilation would be better presented as a new study in a peer-reviewed venue, so that the site selection criteria and individual data sets used would be critically evaluated by peer reviewers.

The problems above on data representation and unsuitable sites were noted by simply looking at the tables. Other issues may exist in other sites/data sets. For the reasons discussed above, the Committee suggests that the authors do not present the new compilation results, but instead focus on the available information in the peer-reviewed literature (such as Saunois et al., 2016 on wetland CH₄ emissions).

Scientific clarity and accuracy. Some statements in this Chapter may not be scientifically accurate. Below are a few examples, with more details offered in the line-by-line comments.

• The statement on p.503, lines 30-31 (“In undisturbed wetlands, carbon stocks are relatively stable over time…”) is wrong. Many published papers have documented that peatlands have continued to accumulate carbon since at least the end of last ice age, so the carbon stocks continue to grow over time (e.g., Gorham, 1991; Harden et al., 1992; Loisel et al., 2014; Yu et al., 2010, among others).
• On p.503 lines 37-38 the statement “Similarly, both carbon stocks and fluxes are very sensitive to disturbance” is redundant, as carbon stocks and fluxes are not independent of each other but closely related. A clear discussion and statement should be made earlier in the chapter that change in carbon stocks, or in any carbon reservoirs for that matter and wetlands included, would be induced by imbalance in carbon fluxes (uptake and release). A disturbance may increase carbon emissions, which in turn may affect carbon stocks, depending on other flux terms. Such a statement would guide reader to have clear understanding of the critical processes.
• At the beginning of subsection 13.3, the first sentence about “rooting zone” is problematic, as Sphagnum and all mosses have no roots, while these moss-dominated peatlands (Sphagnum-dominated bogs and brown moss-dominated rich fens) are widely distributed in Canada and

Global context. The chapter should provide a proper global context to discuss wetland carbon stocks and fluxes in North America. For example, regarding global or northern peatland carbon stocks, some seminal, recent synthesis papers (e.g., Gorham, 1991; Yu et al., 2010) are not discussed. Regarding global and North American wetland CH₄ emissions, many pertinent publications are not discussed (e.g., Bloom et al., 2017; Melton et al., 2013; Tian et al., 2015); Chapter 2 provides a more comprehensive synopsis of the topic (p.80).

Modeling discussion. The evidence for Key Finding 4 about the uncertainties appears to rely mostly on a 10-page USDA Forest Service report (Trettin et al., 2001) (p.525, lines 15-16). Also, the subsection 13.6.3 (Are Current Models Adequate?) barely mention many recent efforts on simulating wetland CH₄ emissions, such as models evaluated and compared in Melton et al. (2013) and Saunois et al. (2016).

Organization and writing. Some parts of the chapter could benefit from reorganization. For example, it may be more effective to divide sub-subsection 13.2.1 into several sub-subsections— including ones that focus on Historical Regulation and Policies, Change in Wetland Area, and Carbon Stocks and Fluxes. Within this chapter, the usage of words and technical terms is inconsistent and lacks clarity. For example, the uses of “carbon fluxes/emissions/update/release” and “annual accretion” sometimes lack clarity and at other times are used inaccurately. There is also some repetitive text within the chapter. It is clear that the chapter was written by a team of authors but hasn’t been edited thoroughly. Numerous small editorial suggestions are listed in line comments below.

Statement of Task Questions

• Are the goals, objectives and intended audience of the product clearly described in the document? Does the report meet its stated goals?

The goals and objectives are not clearly described in the chapter. Presumably the goals are to provide an updated assessment of available literature on carbon cycling in terrestrial wetlands; yet it appears that the authors of this chapter instead conducted their own new data analysis and relied on these new results to reach conclusions.

• Does the report accurately reflect the scientific literature? Are there any critical content areas missing from the report?

Some key findings/conclusions appear to derive from the authors’ own new data analysis results, without proper assessments and comparison with abundant recent peer-reviewed literature. The chapter lacks a proper discussion of global context in terms of peatland carbon stocks and wetland CH₄ emissions. The model section is weak in discussing available models and simulations.

• Are the findings documented in a consistent, transparent and credible way?

Two main findings mostly rely on the new data compilation analysis by the authors of this chapter. As there are several issues related to the data representation and data quality control, the Committee

cannot fully evaluate the credibility of the results/values and findings as commented on above. A more credible, transparent assessment would consider other peer-reviewed literature (e.g., Saunois et al., 2016 and Bloom et al., 2017 on wetland CH₄ emissions) and provide appropriate uncertainty statements.

• Are the report’s key messages and graphics clear and appropriate? Specifically, do they reflect supporting evidence, include an assessment of likelihood, and communicate effectively?

Some Key Findings only present results from the chapter authors’ own new data compilation and analysis. For example, the “description of evidence base” for Key Findings 1 and 2 only refer to Appendices 13A and B, and are not compared explicitly with the peer-reviewed literature. Also, there are many discrepancies with wetland carbon fluxes values cited in Executive Summary and Chapter 2. Many statements are inaccurate or confusing, and there is repetition of some material.

Tables 13.1 and 13.2 provide a useful summary of wetland data in each country and territory in North America. However, the values presented here should be consistent with the values discussed elsewhere in report (Chapter 13, other chapters, Executive Summary: see comments above). The scaling-up of results on NEE and CH₄ emissions presented in Table 13.1 have not been demonstrated to be credible due to the issues with data representation and data quality control of the individual measurements listed in Table Annex 13B (see general comments above). Also, Table 13.2 uses different units for CO₂ flux (Tg CO₂/yr here vs. Tg C/yr elsewhere) and CH₄ emissions (Tg CO₂e/yr here vs. Tg CH₄/yr mostly elsewhere).

Figure 13.1 is unclear and inappropriate. The ranges shown for CH₄ emissions are fundamentally unsupported by evidence. As noted above, the raw measurements of CH₄ emissions presented in table Annex 13B have a range spanning four orders of magnitude, and mean values were used for scaling up to the wetland-type specific total emissions shown in Table 13.1. The scaled-up values were divided by wetland areas to derive CH₄ emissions per unit area, and the CH₄ emission ranges apparently only show the ranges from various countries/territories per wetland type as in Figure 13.1, all within one order of magnitude. The presentation vastly underestimates the uncertainties.

In Figure 13.1, there is no unit specified for carbon pools, and the ranges indicated for four wetland types are large and do not seem to reflect the values presented in Table 13.1. The apparent vegetation carbon pools shown in the figure do not appear in Table 13.1. The units for carbon fluxes are Mg C/hectare/yr while Table 13.1 shows units as Tg C/yr for NEE and Tg CH₄/yr for CH₄. The graphic quality can be improved. For example, the blue wavy line near the top of soils appears to show the water table, but this is not explicitly indicated. Also, some trees appear to grow in air.

• Are the research needs identified in the report appropriate?

The stated research needs in Key Finding 4 appear to be based mostly on an old 10-page government report (Trettin et al., 2001).

The stated research need on model improvement is too general to be useful and does not appear to adequately consider much pertinent literature on wetland carbon models.

The stated research need on evaluating carbon sequestration and flux differences between restored and natural wetlands is a valid one but is too narrowly focused, without adequate justification why this stands out as a key finding. For example, how would the uncertainties in these differences impact the wetland carbon pool and flux assessments?

• Are the data and analyses handled in a competent manner? Are statistical methods applied appropriately?

Some synthesis results and findings are presented here have not gone through normal peer review processes. More justification is needed to average CH₄ emission rates that vary over four orders of magnitude from individual measurements/studies for the scaling-up used in the chapter, in particular to assess how these very different values would impact the uncertainties of CH₄ emissions (see comments above).

• Are the document’s presentation, level of technicality, and organization effective? Are the questions outlined in the prospectus addressed and communicated in a manner that is appropriate and accessible for the intended audience?

The writing and organization can be much improved. There are many instances of scientifically inaccurate statements and lack of clarity in the use of terms and concepts. (Several examples are cited earlier in other sections of this chapter review).

• Are the key findings well stated and supported by the detail provided in the chapter?

Key Finding 1.

• The wetland carbon stock of 178 PgC is mostly from Canada at 130.5 PgC of wetland carbon stocks, including 116 PgC in peatlands (39.3 PgC in nonforested peatlands, plus 76.7 PgC in forested peatlands), as shown in Table 13.1. This peatland carbon stock in Canada is very different from the value used elsewhere—such as 153.7 PgC stated in Chapter 12 on P469 line 7 (cited Tarnocai, 1997). The chapter should consider all these estimates and come up with an assessment of likely range.
• Need to explain how the development of a U.S. soils database would improve greatly the assessment of carbon stocks in North America, including Canada and Mexico. Was the lack of a U.S. soils database a major uncertainty previously during SOCCR1 assessment?
• All figures/values stated here lack uncertainty statements.
• There is an uneven and inconsistent treatment of numerical values; for instance, why focus on the global proportion of wetland area in North America, but not global percentage of total wetland carbon stocks in North America?
• Why does the last sentence focus on wetland area?

Key Finding 2.

• This key finding relies only on the authors’ own new data compilation and analysis. The chapter should consider and assess other estimates in the peer-reviewed literature, as discussed above.
• These figures and values on CO₂ sinks and CH₄ sources are inconsistent with the ones in Executive Summary and Chapter 2.

• Need some explanation of the significance of 43% and 40% mentioned here – why readers should consider this important?

Key Finding 3.

• It is unclear what rates were during historical times and what historical period this refers to.
• The evidence for this key finding appears to derive from a 2011 government report (USFWS 2011), at least for the United States. The URL cited for the report appears to link to more than 20 web pages or documents of approximately 100 MB.
• There is no traceable evidence presented in the Key Finding or Description of Evidence Base.

Key Finding 4.

• The stated research needs appear to be based mostly on an old 10-page government report (Trettin et al., 2001).
• This stated research need on model improvement is too general to be useful, and does not appear to adequately consider the pertinent literature on wetland carbon models (e.g., Melton et al., 2013; Saunois et al., 2016).

• Are there other key findings missing? Any critical literature missing?

It would be helpful for this chapter to include a key finding related to the impacts of climate change and natural disturbance (wildfires, permafrost thaw, drought and water-table drawdown) on wetland carbon fluxes (CO₂ sink and CH₄ emissions). That is, how could the flux rates stated in Finding #2 change in the future? There is abundant literature, at least on CH₄ emissions, that can help inform projections of future wetland carbon balance under a warming climate and changing disturbance regimes.

• Are there any broader questions, such as the selection of the evidence and findings, weight of evidence, or the consistency of the application of uncertainty language?

Evidence heavily relies on the authors’ own new data compilation and analysis, and these new results have not been adequately evaluated in the context of available peer-reviewed literature. The quantitative statements in Key Findings lack uncertainty and range.

Line-Specific Comments

P503, Line 24-25

Perhaps state the wetland area first and then indicate the percentage of the global total. Why only indicate 42% of area, but not % of global total carbon stocks?

P503, Line 25

It is not clear if 178 PgC represents both soils and vegetation, as Figure 13.1 appears to show vegetation/forest carbon stocks as well? The value needs an uncertainty statement/range. There are two ways to distinguish wetlands: forested vs nonforested, and organic soil (peatlands) vs mineral soil wetlands. But this key finding only lists non-forested wetlands by area, and peatlands for both area and carbon stocks. There is no distinction made between soils and vegetation/biomass carbon stocks.

P503, Line30

All estimates need an uncertainty range. The stated CO₂ sink (53 Tg C/yr) is inconsistent with the value used in the Executive Summary (64 TgC/yr, (nonforested 36 + forested 28), p.37). Likewise, the CH₄ emissions value (18 Tg CH₄/yr) is inconsistent with the value in the in Executive Summary (21 Tg CH₄/yr, p.37). Also, Chapter 2 uses wetland CH₄ emission estimates with ranges in the peer-reviewed literature, rather than the values in Chapter 13.

P504, Line 1-6

Quantification is needed: What is the current wetland loss rate? What is the time period for “historical rates”? Has there been any change in wetland loss rates in the last 10 years since SOCCR1? It seems imbalanced to focus on wetland loss and wetland restoration/creation without commenting on impacts of natural disturbance on wetlands and their carbon dynamics.

P504, Line 7-10

This finding is too general to be useful. Either indicate the specifics/nature of improvements or delete.

P504, Line 13-36

Clarify wetlands definition (probably better to call the subsection “Terrestrial Wetlands”).

P504, Line 13

Perhaps begin with a description of the chapter goals and objectives—to clarify if this is an assessment of available peer-reviewed literature or a presentation of a new data analyses.

P504, Line 16

A general scientific definition of wetlands seems more appropriate than “The United States defines…”? Would it be better to define wetlands as ecosystems that include soils and vegetation? Also, from this definition, how does one distinguish “peatlands” and “mineral soil wetlands”—based on water table/hydrology?

P504, Line 25

The sentences are confusing. “all peatlands are… classified as wetlands in Canada”? Why is Gorham et al., 2012 used as reference for definition of wetlands? Would “Wetlands of Canada (Canada Committee on Ecological Land Classification; National Wetland Working Group, 1988)” be a better reference?

P504, Line 31-32

Peatlands are ecosystems while Histosols are soil type. They should not be used interchangeably.

P504, Line 33

40 cm is repeated here, which may not be necessary.

P504, Line 41

Chapter 11 also discusses boreal carbon, so not only Arctic as stated. This subsection/paragraph should be rewritten in a clearer manner. Peer-reviewed scientific literature is available to cite in addition to government agency documents.

P505-507, Section 13.2

This subsection discusses historical views on wetlands and regulation/policy, carbon stocks and fluxes, and wetland area change; yet it has only one labeled heading [13.2.1 on change in Wetland Area]. It may help to instead re-organize the paragraphs under three sub-subsections as follows:

• 13.2.1. Regulations on wetlands (including lines 6-33, P506 and lines 17-26, P505).
• 13.2.2. Change in Wetland Area
• 13.2.3. Carbon Stock and Fluxes (the paragraph on P505-506)

This subsection does not adequately reflect the literature, especially the paragraph on wetland carbon stocks and fluxes.

P505, Line 30-31

The statement that “In undisturbed wetlands, carbon stocks are relatively stable over time…” is inaccurate. Many published papers document that peatlands have continued to accumulate carbon since at least the end of last ice age, and so carbon stocks continue to become larger over time (e.g., Gorham, 1991; Harden et al., 1992; Loisel et al., 2014; Yu et al., 2010, among others).

P505, Line 32-34

There are more data available that have been synthesized than just the single site in Roulet et al. (2007). For instance, Yu (2012) and Ratcliffe et al. (2018) both summarize net carbon balance data from several sites in Canada/North America.

P505, Line 34-36

It is confusing to describe CO₂ fluxes as “CO₂ emissions”: do you mean C release/respiration? If so, what about C uptake/photosynthesis/GPP? Similar wording appears on line 42. Would be better to replace “emissions” with “fluxes” in this context.

P505, Line 37-38

“Similarly, both carbon stocks and fluxes are very sensitive to disturbance.” This should clarify that carbon stocks and fluxes are not independent, as a change in carbon stocks would be caused by imbalance in carbon fluxes (uptake and release).

P505, Line 41

Among many references available, why cite here a study (Drexler et al., 2009) on California Delta about wetland drainage impacts on wetland decomposition?

P506, Line 1-5

References are needed here. Note this is Key Finding #3.

P507, Line 9-18

Provide discussion on oil sands exploration impact on wetlands/peatlands in Western Canada, especially since 2007.

P507, Line 39

The first sentence about “rooting zone” is problematic. What about Sphagnum or other moss-dominated peatlands in Canada and northern U.S. states? These moss-dominated peatlands are major C storage and sink, but they have no roots and rooting zone. This statement is not general enough as an opening sentence for the subsection; This paragraph overall is rather loose and lacks a single citation.

P508, Line 10

Change “Methane flux” to “Methane emission”?

P508, Line 19-21

The sentence is unclear. The sentence could mean large CO₂ flux dynamics, CO₂ uptake, or CO₂ release.

P508, Line 21

Change “from the perspective of“ to “considering organic and mineral soils wetlands separately”

P508, Line 22

Delete “quite.”

P508, Line 32

Change “reported literature” to “reported values in the literature.”

P508, Line 40

The appropriate terms here should be “CO₂ uptake” and “CO₂ release”, not “CO₂ sequestration and emissions.”

P509, Line 4-9

The authors attempt to define the net ecosystem carbon balance (NECB) concept as defined by Chapin et al., 2006, but this paragraph lacks clarity. A distinction between respiratory carbon loss and non-respiratory loss (due to disturbance) is needed.

P509, Line 7

It should be clarified that carbon monoxide is due to fires.

P509, 13.3.1. Peatlands C stocks and fluxes

There are several improvements needs for this sub-subsection:

• Compare their “new” estimates of C stocks and fluxes with what is in the peer-reviewed literature
• Address the poor organization, lack of a reasonable global overview, and lack of proper references.
• Be more consistent use of CH₄ units (Tg C as CH₄/year on p.509, line 24 vs. Tg CH₄/yr in Key Finding #2)
• Address the lack of proper documentation of the value 20-30 gC/m²/yr.

P509, Line 11-12

The distinction of fens and bogs as described here (based on water source and pH) is incomplete and inaccurate. As this chapter is about wetland carbon, it should state the difference in dominant plants in fens (sedges, and brown mosses mostly) and bogs (dominated by peat moss Sphagnum).

P509, Line 21-22

The 116 PgC in Canadian peatlands are inconsistent with the value of 153.7 PgC used in Chapter 12 and other literature. This difference needs to be discussed.

P509, Line 19-11

It would better (here and throughout the chapter) to present the values and then indicate the percentage.

P509, Line 24

The CH₄ unit is inconsistent with elsewhere in the chapter and report.

P509, Line 19-35

The values as presented in Table 13.1 should be discussed and compared with the peer-reviewed literature, such as Tian et al. (2015), Saunois et al. (2016), and Bloom et al. (2017) – all these references were cited and discussed in Chapter 2.

P509, Line 39

It is unclear what is meant by “mode of primary production”. The two references cited here appear to focus narrowly on specific macromolecules in peat, but this paragraph is supposed to talk about decomposition in general. Some more general discussion is needed to provide that unstated macromolecule examples.

P509, Line 40

The term “carbon density” is unclear. Does this mean carbon concentration (% carbon) or soil carbon density (kg C/m2) or bulk carbon density (g/cm3)?

P509, Line 41

The term “peat accretion” is not commonly used outside discussion of mineral soil wetlands, such as salt marsh, as in these mineral-rich systems mineral sediment transport and deposition are important part of peat and carbon accumulation process. “Peat accumulation” is a better term, as most or all materials are derived from dead plant litter.

P510, Line 2-4

The sentence is confusing. The term “carbon stocks” is unclear, as the subheading (Peatlands – Carbon Stocks and Fluxes) indicates it refers to the size of carbon pools in PgC. In this context, it may be better to refer as “soil carbon density” (kgC/m2). It is so obvious the values will depend on peat depths. Also, what is the range of 200-3000 MgC/hectare represented by peat depths? Finally, do these values represent North America or other geographic regions? Yu (2012) provides a range of peat carbon density values from the literature that were used in peatland carbon stock estimates.

P510, Line 4

Time frames need to be indicated for the carbon accumulation rates of 7-300 gC/m2/yr. Also, Loisel et al. (2014) presents the synthesis carbon accumulation data from a large database from northern peatlands that discuss the change in carbon accumulation rates over time.

P510, Line 5-6

The geographic region for the conclusion that bogs accumulate carbon faster than fens should be clearly indicated here to evaluate the relevance and applicability. Does this conclusion refer to peatlands in Finland (Tolonen and Turunen, 1996)? A recent study on several peatlands in western Canada show the opposite conclusion that fens accumulate the same or more peat than bogs (Yu et al., 2014).

P510, Line 9-22

The terms describing CH₄ fluxes are confusing in this paragraph. CH₄ emissions are the difference between CH₄ production and CH₄ oxidation in soil column. So their uses should be clear. CH₄ effluxes in line 14 should be replaced by “CH₄ emissions”.

P511-512, Section 13.3.2

This subsection has the similar issues as for 13.3.2 on Peatlands.

P514, Line 11-13

The 52.5 Tg/yr carbon sinks are not consistent with the value stated in Executive Summary on page 37 (nonforested wetlands 36 + forested wetlands 28 = 64 TgC/yr) and with the value of 36 TgC as presented in Chapter 2 (page 78, line 7 and Figure 2.3). Also, it appears that all these individual values (NEE, CH₄ and DOC) use different units (Tg C, Tg CH₄, or Tg DOC, respectively).

P514, Line 16

The phase “carbon accretion in biomass” should be changed to “carbon accumulation”.

P514, Line 16-18

This is a too simplistic an approach to estimate peatland carbon sequestration. First, the rates of 20-30 gC/m2/yr are likely apparent rates of peat carbon accumulation, rather than actual carbon accumulation rates (see Turunen et al., 2002 and Yu, 2011 for discussion). So these rates cannot be directly used to estimate contemporary peatland carbon sequestration rates. Second, it should be indicated for what time periods (for example over the last several thousand years) these rates were derived and applicable. The apparent rates (see Loisel et al., 2014) and modeled actual rates (Stocker et al., 2017) of peat carbon accumulation show highly variable values throughout the Holocene (the last 12,000 years). The Holocene means in two recent large-scale syntheses on northern peatlands are approximately 20 gC/m2/yr (Loisel et al., 2014; Yu et al., 2010).

P514, Line 21-24

This sentence is confusing. More discussion on vegetation/biomass is needed to make this simple calculation credible.

P514, Line 25

The authors realize that 120% value does not make sense, because the approach described here is not scientifically reasonable.

P514, Line 28

Is this (13.3.5) still under subsection 13.3?

P515, Line 6-7

“moist soil management”: The wording is awkward. Is that wett soil management?

P517, Line 31 – P518, Line 32

The discussion in this subsection is inadequate.

P518, Line 34

This statement requires a reference citation.

P518, Line 35

This is inaccurate. Many models as discussed in Melton et al. (2013) use observational wetland areas, rather than simulate wetland extent directly.

P518, Line 36-37

Change to “between … and ….”

P519, Line 2

It would be useful to provide a global context of peatland and mineral soil wetland areas separately.

P519, Line 12-13

The value of 21 Tg CH₄ /year is not consistent with other values used in the report (see general comments above). The top-down estimates for North America in Saunois et al. (2016) are 17-52 Tg CH₄/year. The 21 Tg CH₄/year value should be considered along with top-down and bottom-up estimates as presented in Saunois et al. (2016). These estimates focus a specific relevant time period from 2003-2012, almost identical to the decade between SOCCR1 and SOCCR2, so the values are most relevant to the SOCCR2 assessments. On the other hand, the value of 21 Tg CH₄/year as derived by the chapter authors have no specific time period that can be assigned.

P519, Section 13.6.3.

The section is very weak. See general comments above.

P536, Table 13.1. Line 2

Change “CH₄ flux” to “CH₄ emissions.” Also, the table needs to provide a global context as well, including global wetland areas, carbon stocks, CH₄ emissions, etc.

P538, Table 13.2.

The units used here are inconsistent with ones in the text, such as Tg CO₂ per year, and Tg CO₂e per year for CH₄ flux.

P540-560, Appendix Tables

These tables and the appendix text are not necessarily useful or appropriate for this report. Considering all the issues with the approach, data representation, data quality control, and large range of individual measurements , the authors should consider take a different approach to assess the available peer-reviewed literature.

P541, Line 4-5

This statement is misleading, as both Dahl (2011) and this chapter use the same NWI database, so the tiny difference (of 0.09%) could be simply due to minor updates or rounding differences. Perhaps Chapter 13 can simply use the value by citing “Dahl (2011), with update NWI 2015)”. Reproducing the content of the database here, even in a summarized form, may not be necessary. The level of details presented in this chapter is incomparable to other chapters in the SOCCR2 report.

P545, Line 13

Are these two estimates independent? If not, why the difference? Are they supposed to be the same?

P546-547, Tables 13A.7 and 13A.10

These tables use different units for area (km² vs. ha). Need more consistency.

P556-558

See the general comments on the data representation, data quality control and data variability/range. Similar issues may exit on mineral soil wetlands in Table 13B.4. Also, these tables use different units on CO₂ and CH₄ fluxes (Mg C/ha/year).

P504, Line 3

What does “23 times less” mean? Do you mean 1/24 here? This should be modified to avoid confusion.

P511, Line 14-30

You may want to consider citing Lu et al. (2017), which synthesizes the annual carbon fluxes (GPP, ER, and NEP) for a number of wetland sites globally.

P514, Line 11-13

“a net sink for atmospheric CO₂” should be changed to “a net carbon sink”. The net ecosystem carbon balance here includes not only CO₂ but also CH₄.

P514, Line 16-18

It would be more reasonable and informative to use the range (20-30 g C/m²/year) to derive a range for the peat soil carbon annual accretion rate in Tg C, rather than a single value.

P517, Line 30-32

Surprisingly this section does not quantitatively assess the historical and future trends of wetland carbon fluxes/stocks. Qualitatively assessing the potential effects of climate change on wetland carbon dynamics is certainly informative. But it would be much better if new or even existing model ensembles could be used to assess the trends of wetland fluxes/stocks.

P521, Line 34 – P522, Line 29

It is unclear whether the current models are adequate or not. If not, in what aspects of these models should be improved?

P522, Line 23-29

The one sentence on data issues does not seem sufficient. It is better to have a separate section on data needs. For example, it would be useful to have spatially and temporally explicit, high-resolution datasets that characterize the type, extent, and seasonal dynamics of wetlands.

P525, Line 11-12

“a significant carbon sink” should be changed to “a significant CO₂ sink” to avoid confusion because wetlands are a source of CH₄ (as indicated by the first half of the sentence).

P525, Line 17-19

The authors should mention another source of major uncertainty - the imperfect model structure and underlying processes (or model uncertainty).

P525, Line 27-29

It is certainly important to evaluate models, but the community needs to move forward by improving these models. Improving models should be explicitly mentioned here.

This page intentionally left blank.