HR: 0830h
AN: B51E-1019 [PDF]
TI: Recent geographic variations in terrestrial carbon cycle based on new production efficiency
model
AU: Sasai, T
EM: sasai@system.eps.nagoya-u.ac.jp
AF: Department of Earth and Environmental Sciences, Graduate School of Environmental Studies,
Furo-cyo,Tikusa-ku, Nagoya, 464-8602
Japan
AU: Ichii, K
EM: ichii@system.eps.nagoya-u.ac.jp
AF: Department of Earth and Environmental Sciences, Graduate School of Environmental Studies,
Furo-cyo,Tikusa-ku, Nagoya, 464-8602
Japan
AU: * Yamaguchi, Y
EM: yasushi@turnip.eps.nagoya-u.ac.jp
AF: Department of Earth and Environmental Sciences, Graduate School of Environmental Studies,
Furo-cyo,Tikusa-ku, Nagoya, 464-8602
Japan
AB:
The terrestrial carbon budget must be understood more accurately for the prediction of future changes in climate and carbon
cycle. The goal of this study is to estimate spatial and temporal patterns of the carbon fluxes more accurately using the
newly developed terrestrial biosphere model and satellite data. Our model consists of terrestrial carbon cycle and hydrology
submodels. An advantage is a new approach in the LUE (Light Use Efficiency) concept, which calculates temperature and water
stress factor in LUE model from a photosynthetic model and stomatal conductance formulation. In carbon cycle model, GPP is
calculated from the LUE concept and satellite-based fPAR dataset. The soil carbon cycle model is based on CENTURY model with
optimized water and temperature factor. Hydrological submodel is based on BIOME3, calculating ET is used by Penman-Monteith
method. The model was run for 18 years (1982-1999) on a global scale, and we simulated the geographic distributions of the
terrestrial carbon fluxes.
We have checked simulated vegetation growth limiting factor with stress factor of MODIS NPP algorithm. Large differences
were found in the northern mid and high latitude forests because soil moisture stress is not incorporated into MODIS NPP
algorithm. Although responses of stress factors in MODIS NPP algorithm are mostly similar to our theoretically based one, our
model works well in the soil moisture limited regions.
Global total NPP was estimated at 61.7GtC/yr, and total NEP variations are strongly related with ENSO. Validation using
measured values from the GPPDI database showed that our NPP estimation was within a reasonable range. The temporal patterns
of the terrestrial carbon flux showed that NPP increased in the northern middle/high latitudes, central Africa, and India. In
contrast, NPP decreased in the south Amazon region, the middle latitudes of the southern hemisphere, a part of North
America, and Southeast Asia. Sensitivity analysis indicated that NPP variations were largely affected by fAPAR, solar
radiation, temperature, and precipitation variations. Our new model was shown to be an appropriate tool to estimate the
spatial and temporal patterns of the terrestrial carbon fluxes.
DE: 0400 Biogeosciences
DE: 1600 GLOBAL CHANGE (New category)
DE: 1615 Biogeochemical processes (4805)
DE: 1640 Remote sensing
DE: 4805 Biogeochemical cycles (1615)
SC: Biogeosciences [B]
MN: 2003 Fall Meeting