Postdoctoral Research Fellowships in Biology for FY 2009

这一行动为NSF 2009财年博士后研究奖学金提供了资金。该奖学金支持丽莎·帕特里克的一项研究和培训计划，该计划名为“通过新的生物信息学方法将植物比例模型与植物生理模型联系起来，以估计不同植物的碳通量和水通量”。这项研究的主办机构是亚利桑那大学，赞助科学家是Brian Enquist。这项研究将代谢标度理论与植物生理过程相结合，为理解和预测全植物和生态系统动力学提供了一个更稳健的框架。具体的研究目标是(1)将植物生理模型与代谢标度理论相结合，以估计各种物种、生态系统和环境条件下的植物碳通量和水通量；(2)在量化模型误差的同时，统计评估耦合模型在每个水平(例如，叶、植物、物种、生态系统)考虑生物学意义参数变化的能力。该研究方法包括利用信息并将信息提供给不同的数据库；将模型集成在分层贝叶斯框架内；使用现有数据来参数化耦合的生理尺度模型，并随后估计各种物种、植物功能类型和环境中的植物水平的碳和水通量；以及对新的耦合生理尺度模型进行统计验证，以量化模型误差以及估计的模型参数和模型预测中的不确定性。这项研究利用尖端的统计工具和新的生物信息学方法，在半机械生态模型的背景下合成了不同的解剖、建筑和生理植物数据。结果被用来改善陆地尺度关系，并有助于提高估计木本植物和以木本植物为主的生态系统的区域和全球碳和水收支的预测能力。培训目标是通过现代统计方法、模型实施和验证技术以及生态学中的可伸缩性应用方面的高级培训，获得进行跨学科研究的能力。

This action funds an NSF Postdoctoral Research Fellowship for FY 2009. The fellowship supports a research and training plan entitled "Linking plant scaling models with plant physiological models through novel bioinformatics approaches to estimate carbon and water fluxes of diverse plants" for Lisa Patrick. The host institution for this research is the University of Arizon, and the sponsoring scientist is Brian Enquist.This research combines metabolic scaling theory with plant physiological processes to produce a more robust framework for understanding and predicting whole-plant and ecosystem dynamics. Specific research objectives are to (1) integrate plant physiological models with metabolic scaling theory to estimate plant carbon and water fluxes across a wide variety of species, ecosystems, and environmental conditions and (2) statistically evaluate the ability of the coupled model to account for variation in biologically meaningful parameters while quantifying model error at each level (e.g., leaf, plant, species, ecosystem). The research approach involves utilizing and contributing information to diverse databases; integrating models within a hierarchical Bayesian framework; using existing data to parameterize the coupled physiological scaling model and subsequently estimate plant-level carbon and water fluxes across a variety of species, plant functional types, and environments; and statistically validating the new coupled physiological scaling model to quantify model error as well as uncertainty in estimated model parameters and model predictions. The research makes use of cutting-edge statistical tools and novel bioinformatics approaches to synthesize diverse anatomical, architectural, and physiological plant data within the context of semi-mechanistic ecological models. Results are used to improve terrestrial scaling relationships and contribute to improved predictive power for estimating the regional and global carbon and water budgets of woody plants and ecosystems dominated by woody plants. The training goals are to gain the ability to conduct interdisciplinary research through advanced training in modern statistical methods, model implementation and validation techniques, and scaling applications in ecology.