HR: 0800h
AN: B41B-0189    [Abstracts]
TI: Spatial Patterns of Carbon Residence Time and Sequestration Capacity in Terrestrial Ecosystems of the Conterminous USA
AU: Zhou, T
EM: tzhou@ou.edu
AF: University of Oklahoma, 770 Van Vleet Oval, Norman, OK 73019
AU: Xu, T
EM: Tao.xu-2@ou.edu
AF: University of Oklahoma, 770 Van Vleet Oval, Norman, OK 73019
AU: * Luo, Y
EM: yluo@ou.edu
AF: University of Oklahoma, 770 Van Vleet Oval, Norman, OK 73019
AB: To model carbon (C) sequestration and its spatial pattern, three key parameters need to be quantified: (1) canopy carbon influx; (2) carbon residence time in an ecosystem; and (3) initial values of pool sizes. While spatial distributions of canopy carbon influx have been extensively studied, spatial patterns of carbon residence times have not been carefully characterized. In this study, we conducted an inverse analysis to estimate the carbon residence times in ecosystems of the conterminous US from 12 data sets. The 12 data sets are three NPP data sets (i.e., NPP in leaves, stems, and roots), five biomass data sets (i.e., biomass of leaves, stem, and roots in three soil layers), one litter data set (i.e., fine litter mass), and three SOC data sets in the three soil layers. The inverse analysis was based on a process-based Terrestrial ECOsystem Regional (TECOR) model and used the genetic algorithm for optimal parameter estimation. The inverted residence times and increase trends of net primary production (NPP) were then fed into a forward modeling analysis to map spatial patterns of carbon sequestration capacity. Our analysis estimated that the mean residence time for the whole conterminous US is 46 years with a range from 10 to 150 years. The central Great Plains have the lowest residence times (mean = 28 years, std = 13 years) and the west regions have the highest ones (mean = 64 years, std = 34 years) with the east regions in between (mean = 41 years, std = 20 years). When a 0.5 percent increase of NPP per year was uniformly applied to the whole conterminous US, our forward modeling showed that most of the eastern regions and some of the northwest regions have large carbon sequestration capacity. When a satellite-data-derived spatial distribution of NPP was applied in the forward modeling, it was estimated that the cropland has the largest carbon sequestration capacity followed by the deciduous broadleaf forest, grassland, wooded grassland, and mixed forest. The cropland and grassland store most of the sequestered carbon in soil whereas the forests and woodland store most of the sequestered carbon in plant tissues. We estimated that the whole conterminous US continent sequesters approximately 0.26 Pg C yr-1, similar to the estimate of carbon sequestration from inventory data (about 0.3 Pg C yr-1).
DE: 0414 Biogeochemical cycles, processes, and modeling (0412, 0793, 1615, 4805, 4912)
DE: 0426 Biosphere/atmosphere interactions (0315)
DE: 0428 Carbon cycling (4806)
DE: 0430 Computational methods and data processing
DE: 0466 Modeling
SC: Biogeosciences [B]
MN: Fall Meeting 2005