The Influence of Historical Land Use and Land Cover Change Assumptions, CO2 Fertilization, and Nitrogen Deposition on Global Carbon Balance in an Earth System Model

Abstract:

Land Use and Land Cover (LULC) are not consistent across the CMIP5 model simulations because LULC distribution was determined in each model by unique application of land use change to model-specific initial land cover states. Such differences impede understanding of the carbon cycle and climate because differences in LULC change can both dominate, via emissions, and alter, via biome extent, other sources of uncertainty in the global carbon cycle, such as the effects of CO₂ fertilization and nitrogen deposition on terrestrial ecosystems. For example, the Community Earth System model overestimates 2005 atmospheric CO₂ concentration by ~20 ppmv, and we explore the contribution of historical LULC trajectory to this bias in relation to the effects of CO₂ fertilization and nitrogen deposition on terrestrial carbon. Switching from the default, year-2000-referenced LULC trajectory to a chronological LULC trajectory actually increases the bias by about 70%, and the difference in atmospheric CO₂ concentration could be ~9 ppmv between a “pasture rule” scenario that preferentially uses non-forest for new pasture and a “crop rule” scenario that preferentially uses forest for new cropland. These atmospheric differences correspond to changes of 28 and 17 Pg of total land ecosystem carbon, respectively. We will compare these results to additional scenarios with static CO₂ concentration and nitrogen deposition to assess the relative contributions of each of these factors to carbon cycle and climate uncertainties. We will also evaluate extreme forest clearing and retention scenarios to determine a potential uncertainty envelope due to land use conversion assumptions. Our current results suggest that the initial forest area is too high and thus contributes to the 2005 bias in atmospheric CO₂ concentration. This highlights the need for more accurate LULC scenarios in earth system simulations to provide robust historical and future projections of carbon and climate.