Document: DAV-3-66-2

Carbon storage in the US 1895-1993.

SCHIMEL, D.* ², H.TIAN ², S.RUNNING ³, J.MELILLO ², R.NEILSON ⁴, W.PARTON ⁵, D.OJIMA ⁵, T.KITTEL ⁶, M.SYKES ⁷ and A.D.MCGUIRE ⁸

Max-Planck-Institute for Biogeochemistry, Jena Germany ¹
Ecosystems Center,Marine Biological Laboratory, Woods Hole MA ²
University of Montana, Missoula Montana ³
US Forest Service, Corvallis Oregon ⁴
NREL, Colorado State University, Ft. Collins CO ⁵
National Center for Atmospheric Research, Boulder CO ⁶
Plant Ecology, Lund University, Lund Sweden ⁷
Alaska Cooperative Fish and Wildlife Unit, University of Alaska, Fairbanks AK ⁸

Abstract:
North America plays a disproportionate role in global fossil fuel emissions, and some analyses of atmospheric carbon dioxide suggest it may also play a major role in ecosystem storage of carbon. The Vegetation and Ecosystem Modeling and Analysis Project used a set of state-of-the-art ecosystem models to simulate the response of carbon storage in the US to climate and increasing CO₂ from 1895-1993. We combined process models with extensive new data to analyze the carbon budget of the natural and agricultural ecosystems of the conterminous US. Models were integrated using a new historical climate database, historical and satellite-based land cover and agricultural statistics. We also synthesized the results of recent inventory-based estimates of carbon storage, which suggest a sink of approximately 0.3 Gt per year (+/- 100%). The results show that increasing CO₂ has led to modest increases in carbon storage (~0.1 Gt per year in the 1980s, +/-25%). The rate of carbon uptake varied significantly with climate over the century with periods of both loss and accumulation. Over the period 1980-1993 the flux varied by ~100% showing a strong influence of interannual climate variability. The drought of 1988 played a particularly significant role; El Nino effects are also evident. Short-term measurements could be in error in approximating long-term fluxes by 100% or more and could even infer the wrong sign. This is of great significance in using measurements of carbon fluxes in climate-change policy measures and land management. Ecological measurements sustained for ten or more years are required to detect critical biogeochemical phenomena. Agricultural lands played a modest role in recent decades, neither losing nor gaining substantial amounts of carbon, as is expected in the absence of positive measures to increase soil carbon storage. Inventory estimates of carbon storage are consistently larger than our best estimate of the CO₂ effect, suggesting that carbon storage due to forest regrowth is equal to 100-200% of the direct CO₂ effect. For the carbon cycle, the landscape effects of human disturbance may be larger than ecosystem physiological responses to the changing environment.

Keywords: carbon cycle, CO2 fertilization, disturbance, climate

This abstract is being presented at: 5:00 PM in session:
Oral Session #52: Carbon Storage in Ecosystems.