Climate and Chemistry Modeling of Forest-atmosphere Feedbacks in the Great Lakes Region.

[摘要] Vegetation impacts air quality and climate by emitting ozone precursors, known as volatile organic compounds (VOC), and water vapor. The influence of these natural emissions depends on how efficiently they escape the forest layer. This dissertation examines our understanding of this escape using multi-scale models. First, sonic anemometer data and an updated chemical mechanism are incorporated into a 1-D canopy model to improve the representation of near-canopy turbulence and VOC chemistry. The revised turbulence improves the vertical gradients in VOC, suggesting better escape, while the new chemistry scheme exacerbates VOC overestimates. The influence of canopy heterogeneity is evaluated by comparing vertically uniform and variable VOC emission potential distributions. The variable case con- strains light-induced VOC emissions to the upper canopy where more light is available, leading to higher emissions. As a practical implication of this result, accounting for heterogeneity enhances changes in VOC escape following succession. Lastly, land- and lake-atmosphere feedbacks in the Great Lakes region are simulated with a 3-D regional climate model (RCM) coupled with a state-of-the-art land surface model(LSM). A moisture budget analysis reveals that evapotranspiration accounts for up to one-quarter of the local precipitation. However, the RCM does not fully capture the spatial variability in evapotranspiration, estimated by eddy covariance measurements at three field stations, potentially due to missing complexity in its response to surface conditions in LSM parameterizations. In conclusion, the forest-atmosphere exchange of biogenic VOC and water vapor, and thus the impact of vegetation on air quality and climate, is highly sensitive to complexities in surface layer processes, which are not adequately represented in current atmospheric models at both local and regional scales. Additional eddy covariance measurements at multiple canopy depths are needed to better characterize and improve model parameterizations of canopy turbulence and evapotranspiration. Land surface descriptions require more detailed vegetation data (e.g., stand height, age, and density) to account for vertical heterogeneity in VOC emissions.

[发布机构] University of Michigan

[效力级别] atmospheric chemistry

[关键词] land-atmosphere interactions;atmospheric chemistry;climate;Great Lakes region;forests;Natural Resources and Environment;Science;Atmospheric, Oceanic and Space Sciences