Comparing pasture c sequestration estimates from eddy covariance and soil cores

Authors: Skinner, Robert; Dell, Curtis

Submitted to: Agriculture, Ecosystems and Environment
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/27/2014
Publication Date: 9/7/2014
Citation: Skinner, R.H., Dell, C.J. 2014. Comparing pasture c sequestration estimates from eddy covariance and soil cores. Agriculture, Ecosystems and Environment. 199:52-57.

Interpretive Summary: Temperate pastures can either be sources or sinks for atmospheric carbon dioxide depending on management practices, climate, and pasture age. However, determining how much carbon is being gained or lost can be difficult due to large inherent variability in soil organic carbon (SOC) content and the relatively slow rate of change compared to the size of the SOC pool. We used two contrasting methods to determine changes in SOC over a nine-year period for two pastures differing in the frequency of grazing vs. haying and in the amount of applied nitrogen fertilizer. The first method used eddy covariance to measure the continuous gain and loss of carbon dioxide at 20 minute intervals. These measurements were combined with estimates of harvested forage removal and manure deposition to estimate changes in SOC. The second used soil cores taken at one to two year intervals to directly measure changes in SOC to a depth of 100 cm. Both methods suggested that SOC in the low fertility pasture was near equilibrium and was not significantly increasing or decreasing over time. However, the eddy covariance technique suggested that the high fertility pasture was also in equilibrium or perhaps a small net source of carbon dioxide, whereas the soil core method indicated that a large amount of SOC was being lost each year, especially from lower in the soil profile. We do not know the reason for the difference between the two methods but they suggest a pathway existed for SOC loss from the deeper soil in this pasture that was not captured by the eddy covariance method.

Technical Abstract: Temperate pastures in the northeast USA are highly productive and could act as significant sinks for soil organic carbon (SOC). However, soils under mature pastures are often considered to have reached equilibrium such that no further sequestration of SOC is expected. This study quantified changes in pasture SOC over nine years using micrometeorological (eddy covariance) measurements of net changes in ecosystem C and direct measurements of changes in SOC (soil cores). Eddy covariance estimates of C flux were made both with and without corrections for sensor self-heating. Change in SOC as measured by the soil core method was non-significant for the Low-N pasture (19 +/- 105 g C m**-2 yr**-1, P = 0.89). However, a significant loss of -504+/- 91 g C m**-2 yr**-1 (P = 0.01) occurred for the High-N pasture with the amount lost increasing with depth in the soil profile. Eddy covariance measurements without the self-heating correction differed only slightly between pastures with a net flux of -9 and -20 g C m**-2 yr**-1 for the Low- and High-N pastures, respectively. Applying the sensor self-heating correction to the eddy covariance data increased the estimated loss of ecosystem C by 94 g m**-2 yr**-1 for the Low-N and 102 g m**-2 yr**-1 for the High-N pasture. Both soil cores and eddy covariance suggest that the Low-N pasture was C neutral over the nine years of the study. A large amount of C was lost from deep in the soil profile from the High-N pasture which could not be explained by fluxes measured with the eddy covariance system. Comparison of eddy covariance and soil core data was not useful for determining the appropriateness of using the sensor self-heating correction at this location.