Skip to main content
ARS Home » Research » Publications at this Location » Publication #191639
Title: MODELING SOIL CARBON SEQUESTRATION AND ANALYSES OF POTENTIAL ERRORS

Author
item Doraiswamy, Paul
item AKHMEDOV, B - SSAI
item McCarty, Gregory
item Hunt Jr, Earle

Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 1/19/2006
Publication Date: 4/20/2006
Citation: Doraiswamy, P.C., Akhmedov, B., McCarty, G.W., Hunt, E.R. 2006. Modeling soil carbon sequestration and analyses of potential errors [abstract]. Workshop: Emerging Modalities for Soil Carbon Analysis: Sampling Statistics and Economics. 2006 CDROM.

Interpretive Summary:

Technical Abstract: Evidence for global climate change has intensified international efforts toward reducing anthropogenic carbon dioxide sources and increasing sink capacities of the terrestrial biosphere, particularly for agriculture, since agriculture comprises a large part of the biosphere. Agricultural ecosystems are thought to have potential as substantial carbon dioxide sinks when management practices are optimized for carbon sequestration. Estimating soil organic matter change across large areas such as the U.S. Corn Belt is complex because of spatial variability in climate, soil properties, and crop management practices. Direct measurement of soil carbon changes at regional scales requires enormous resources and is impractical. Biogeochemical models offer a valid alternative for estimating long-term changes in soil carbon changes. Numerous biogeochemical models capable of simulating changes of soil organic matter have been reported and evaluated. A potential approach is described in which a biophysical model, is combined with soil sampling and remotely sensed data to achieve reliable and verifiable estimates of soil carbon over time and space. There are uncertainties associated with input data and model predictions. The reliability of changes in soil organic matter estimates is improved by using observations to minimize the uncertainties of model predictions. An overall framework is suggested for providing reliable estimates of soil carbon by reducing error propagation using the Ensemble Kalman Filtering technique.