Individual Based Modeling and Visualization of Forest Succession and Biomass within the Prospect Hill Tract of Harvard Forest in Petersham, Massachusetts

Biomass;Forest Succession;Harvard Forest;Natural Resources and Environment

Sims, JoshuaCurrie, William ;

University of Michigan

关键词: Biomass;    Forest Succession;    Harvard Forest;    Natural Resources and Environment;   

Others  :  https://deepblue.lib.umich.edu/bitstream/handle/2027.42/97372/joshsims_thesis_final.pdf?sequence=1&isAllowed=y

瑞士|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

The carbon storage and dynamics of terrestrial forest vegetation will play a major role in determining the degree of effects of climate change. As the response to climate change increases, the mapping and modeling of forest carbon stocks and dynamics will become of increasing importance at scales ranging from the individual forest stand, to the landscape level and ultimately global forest inventories. Individual-Based Modeling offers a way to investigate the above ground biomass distribution within a forest stand based upon differential ecological and historical phenomena which had led to a specific distribution of individual trees. This analysis will help to inform us of both how different model parameters affect biomass distribution and the influence of changing model parameters on model outcomes. We have parameterized and evaluated the SORTIE model of forest succession to reflect the effects of spatially explicit historical events upon the current biomass distribution across ten Prospect Hill plots at the Harvard Forest, MA, USA. From data recorded during the NASA DESDynI field campaign (Cook, 2010), biomass in each of these plots were estimated using the Jenkins set of allometric equations (Jenkins, 2003). Each initialization simulates a history of one particular plot. The most important parameters to calibrate were the species specific Asymptotic Diameter Growth and Slope of Diameter Response Curve. To further this investigation, we have used the output from these model runs to produce three dimensional representations of the forest structure within these plots. This methodology can be used to improve biomass remote sensing techniques. Sensitivity analysis using Monte Carlo methods to test the effects of randomly perturbing the yearly growth rate and intrinsic mortality probability parameters shows the model to be functioning properly. Futureareas of research include investigating the nature of the path dependency inferred by natural and anthropogenic disturbance and more realistic methods of creating three dimensional forest structures.