In this chapter the authors demonstrate that grasslands typically take up more carbon from the atmosphere than they release (i.e., a sink) and that, unlike forests where carbon is in wood, much of this carbon is stored in soils. Carbon storage in grasslands is sensitive to climate, operating primarily through variation in the length of the growing season; and while there is unrealized potential to store additional carbon in these systems through proper management, with current practices this system is expected to become less of a sink with time.
The authors define grasslands in part as ecosystems that occur in areas where average annual evapotranspiration is greater than precipitation. While appropriate for most grasslands in North America, this definition misses the grasslands of central Florida. These grasslands occupy relatively small area but have a rather large economic impact through beef production. This chapter would benefit by expanding the discussion of southeastern grasslands.
In discussing processes affecting grassland carbon stocks (section 10.3.2), precipitation is identified as very important. Over the past few decades, there have been demonstrable changes in the timing and intensity of precipitation. While the effect of changes in the amounts of precipitation is discussed, a bit more attention to the role of intensification of hydrologic cycle would be useful.
The encroachment of woody vegetation into grasslands is increasing as the climate warms and as fire is suppressed. The authors acknowledge this trend, but it was unclear how woody encroachment affects carbon stocks and fluxes in grasslands.
Statement of Task Questions
- Does the report accurately reflect the scientific literature? Are there any critical content areas missing from the report?
The statement that “… moisture availability exerts more control over variability in productivity and carbon storage in grasslands than does grazing.” (p.386, ln 10) is overstated, and this statement would benefit from appropriate citations.
The authors may wish to include one or more of the following references in their discussion of how cheatgrass affects biogeochemistry and hydrology (paragraph beginning on p.385, ln 12): Obrist et al. (2003); Prater and DeLucia (2006); Prater et al. (2006).
- 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?
Do the areas in Table 10.1 represent total area? If so, the area of grasslands would be more appropriate.
- Are the research needs identified in the report appropriate?
In the discussion of knowledge gaps (section 10.6.2), another source of uncertainty is the interaction between changes in land use and climate change. There are very few studies that investigated how changes in land use (grazing) in concert with changes in climatic factors (precipitation) will alter carbon processes. This lack of knowledge hinders our capacity to predict the response of carbon storage in grasslands to future climate changes, as we know that ecosystem responses derived from knowledge of single-factor experiments are likely to be misleading. See for instance: Norby and Luo (2004); Templer and Reinmann (2011).
In the discussion of “major uncertainties” (p.395, ln 21), another major uncertainty is how much carbon sequestration in grasslands can be increased through management practices, plant breeding, or genetic modified organisms.
- Are the data and analyses handled in a competent manner? Are statistical methods applied appropriately?
Most of the uncertainty and statistical analyses are presented in the primary literature cited in this chapter, rather than being applied directly in this synthesis.
P380, Line 13
Would it be advisable to present the key findings in order of confidence?
P380, Line 32
Start with areal extent of grasslands in the U.S.
P380, Line 33
In discussing the areal extent of grasslands it would be good to open with their coverage in North America rather than globally.
P381, Line 19-20
Using the term “C uptake” without at least specifying if this refers to net or gross carbon uptake is misleading. See for instance, comments under section 10.3.2. It is unclear if references to C uptake always refer to net carbon sink or they are referring to GPP. Need to define what GPP is and that GPP-Reco is NEP. This way one could avoid using less-specific terms such as “all C uptake”.
P381, Line 19-21
Should this convention be adopted in all chapters?
P381, Line 20 (and elsewhere)
Phrases such as carbon uptake and loss are used rather loosely. Do the authors mean net or gross? Where possible, it would be best to use standard ecosystem carbon cycling terminology, e.g. GPP, Reco, NEP, NPP, etc.
P381, Line 32-34
One would think this would be extensive to changes in precipitation such as drought. Arid systems will be more vulnerable to reductions in precipitation.
P383, Line 16-28
Does woody encroachment affect carbon cycling in grasslands?
P383, Line 34
Standardize units across chapters.
P385, Line 1-11
One could use this study, along with others, to strengthen the idea (raised in section 10.5) that grasslands have potential to sequester additional carbon if managed properly.
P385, Line 12-26
It would be good to comment on precipitation changes expected in the future, and the fact that these changes will have a marked seasonality.
P385, Line 14
The discussion of how growing season “plasticity” and corresponding variation in productivity (NPP?) responds to climate would benefit from consideration of how this variation would affect carbon losses by plant and soil respiration.
P385, Line 19-23
It would be good to also mention carbon losses, and the fact that the inter-annual variation in ecosystem productivity reflects interactions between SM and temperature controls on both “all C uptake” (GPP?) and carbon losses (Reco) - not only on carbon uptake. This would help illustrate that the sensitivity of GPP and Reco to these climate factors will likely differ, and this determines the net carbon sink or source strength of grasslands. This exemplifies how using the term carbon uptake is misleading, as it is unclear whether the authors mean gross or net carbon uptake.
P385, Line 33-36
Might want to add the recent paper Gomez-Casanovas et al. (2016), which shows that grazing increased the carbon sink strength of subtropical pastures. Subtropical grasslands are very important for U.S. beef production (look up Florida in the rankings for beef production); and along with tropical pastures, they are one of the most abundant grassland types across the world.
P386, Line 9-11
This seems like an overstatement—at least if not accompanied by literature. If one of them exerts more control over productivity or carbon storage, it will depend on how much these factors change. Think about increasing the stocking rate from moderate to heavy. That will certainly affect carbon storage, which they acknowledge. Perhaps what is meant is that we can alter grazing intensity to a desired outcome—for instance, increased NEP-C storage in grasslands is more resilient to grazing than to precipitation because theoretically we cannot alter precipitation (although the management practice of “rain harvesting” may allow for this to some degree).
P386, Line 12-15
See papers by Prater cited above.
P387, Line 31
The authors mention that models predict an increase in forest land carbon stocks in the Great Plains by 2050. It would be useful to know what factors are predicted to drive this increase in forest area.
P387, Line 31
The driver for increasing forest is unclear.
P388, Line 35 – P389, Line 4
Is worth mentioning nitrogen (N) deposition in this context, as many grasslands are not fertilized and therefore their only N input comes from deposition.
P388, Line 38 – P389, Line 4
Accurately predicting the response of carbon sequestration to elevated CO2 and warming depends on the limitation or saturation of ecosystem processes to nitrogen. For instance, if a system is limited, one would expect CO2 to increase carbon sequestration as biomass increases— and one would expect the opposite if the system is saturated. Predicting if grasslands will be N limited or saturated therefore depends on N deposition rates these systems experience in the future (at least for grasslands that are not fertilized by humans). It is also worth acknowledging that there is a large uncertainty in carbon responses to N deposition, and that this is hindering our capacity to accurately predict grassland carbon response in the future (see for instance Gomez-Casanovas et al., 2016).
P389, Line 18-25
It may be worth mentioning that the responses of carbon storage to changes in precipitation will likely differ in xeric, mesic, and hydric systems, although we are not capable of accurately predicting the magnitude of the response. Need more discussion of timing of precipitation
P390, Line 25 – P392 Line 19
The discussion of “societal drivers” would benefit from a short paragraph stating which other practices along with changes in grazing management and fire regime could potentially increase carbon sequestration in grasslands. It would also be interesting if the authors could link this to the uncertainty in future carbon stocks in grasslands. Mainly, we don’t know which and how practices other than grazing and fire could affect carbon sequestration in grasslands.
P391, Line 8
The statement that removal of above ground biomass by grazing reduced soil carbon stocks would benefit from a reference or two.
P391, Line 26-31
How does woody encroachment affect som?
P392, Line 8
Specify annual crops. It is a different story for perennial biofuels.
P392, Line 1-19
The discussion of how converting grasslands to other vegetation types or management regimes (namely crops) needs to be a bit more specific. Some of the pressure on grasslands in the future will come from the expansion of perennial bioenergy crops. While replacing perennial grasslands with
annual row crops typically reduces soil carbon stocks, replacing these systems with high-yielding perennial grasses for energy production can have the opposite effect.
P392, Line 22-31
The synthesis section would be improved by including some mention of grasslands in Southeastern U.S. because these grasslands are grazed and globally, tropical and subtropical grasslands play an important role in the carbon cycle; they store vast amounts of carbon, some of which is emitted to the atmosphere as CH4. In addition, they are important from an economic perspective as the contribution of beef production in Florida is large.6
P393, Line 21
Change “easily” to “readily”.
P393, Line 32-33
Is this true even when considering row crops?
P395, Line 21-29
A major uncertainty is how much we can increase carbon sequestration in grasslands through management practices, plant breeding or genetic modified organisms—because of the lack of field data. This seems a crucial point to make in addition to the uncertainty in precipitation patterns.
P411, Table 10.1.
Specify that “approximate area” is referring to grasslands.
This page intentionally left blank.