Parallel Computing for Terrestrial Ecosystem Carbon Modeling

• Main Authors: Wang, Dali ORNL, Post, Wilfred M ORNL, Ricciuto, Daniel M ORNL, Berry, Michael University of Tennessee, Knoxville (UTK)
• Language: unknown
• Published: 2013
• Subjects: 54 ENVIRONMENTAL SCIENCES; 72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS; 99 GENERAL AND MISCELLANEOUS//MATHEMATICS; COMPUTING; AND INFORMATION SCIENCE; CARBON; CARBON CYCLE; CLIMATES; DISTURBANCES; ECOSYSTEMS; LAND USE; PERIPHERAL MODELS; PLANTS; SOILS; STORAGE; TERRESTRIAL ECOSYSTEMS; PARALLEL PROCESSING; CLIMATIC CHANGE; COMPUTERIZED SIMULATION; Tundra
• Online Access: http://www.osti.gov/servlets/purl/1003746/

Terrestrial ecosystems are a primary component of research on global environmental change. Observational and modeling research on terrestrial ecosystems at the global scale, however, has lagged behind their counterparts for oceanic and atmospheric systems, largely because the unique challenges associated with the tremendous diversity and complexity of terrestrial ecosystems. There are 8 major types of terrestrial ecosystem: tropical rain forest, savannas, deserts, temperate grassland, deciduous forest, coniferous forest, tundra, and chaparral. The carbon cycle is an important mechanism in the coupling of terrestrial ecosystems with climate through biological fluxes of CO{sub 2}. The influence of terrestrial ecosystems on atmospheric CO{sub 2} can be modeled via several means at different timescales. Important processes include plant dynamics, change in land use, as well as ecosystem biogeography. Over the past several decades, many terrestrial ecosystem models (see the 'Model developments' section) have been developed to understand the interactions between terrestrial carbon storage and CO{sub 2} concentration in the atmosphere, as well as the consequences of these interactions. Early TECMs generally adapted simple box-flow exchange models, in which photosynthetic CO{sub 2} uptake and respiratory CO{sub 2} release are simulated in an empirical manner with a small number of vegetation and soil carbon pools. Demands on kinds and amount of information required from global TECMs have grown. Recently, along with the rapid development of parallel computing, spatially explicit TECMs with detailed process based representations of carbon dynamics become attractive, because those models can readily incorporate a variety of additional ecosystem processes (such as dispersal, establishment, growth, mortality etc.) and environmental factors (such as landscape position, pest populations, disturbances, resource manipulations, etc.), and provide information to frame policy options for climate change impact analysis.