CGV: Small: Quantification and Visualization of Ensemble Uncertainty

CGV：小：集合不确定性的量化和可视化

• 批准号: 1117871
• 负责人: Song Zhang
• 金额: $ 47.52万
• 依托单位: Mississippi State University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2011
• 资助国家: 美国
• 起止时间: 2011-08-15 至 2017-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1117871&HistoricalAwards=false
• 关键词: CGV; Small; Quantification; Visualization; Ensemble

Uncertainty is inherent in scientific data. Quantification and interpretation of the uncertainty are essential to a holistic understanding of the data. Uncertainty representation is often overlooked in high-dimensional and multivariate data due to a lack of effective methods and a lack of guidance on appropriate existing methods. This project systematically investigates uncertainty visualization in high-dimensional data, within the application domain of meteorology. The interdisciplinary research team spans different areas such as visualization, atmospheric science, and statistics. The research team is investigating (1) quantifying uncertainty by using bootstrap mean confidence intervals, which relax assumptions about the underlying distribution; (2) designing a number of uncertainty visualization methods and implement these methods in an integrated user interface; and (3) employing both qualitative and quantitative user studies to evaluate the uncertainty visualization methods.The resulting tools and methods are deployed to the Environmental Modeling Center. Although the primary application area of this project is meteorology, the principles learned have the potential to be applied to a variety of scientific fields, including hydrology, geospatial science, medical imaging, and biology. The research work of this project is integrated into Visualization and Meteorology classes at Mississippi State University.

不确定性是科学数据固有的。对不确定性的量化和解释对于对数据的整体理解至关重要。 由于缺乏有效的方法以及对适当的现有方法的指导，在高维和多元数据中通常会忽略不确定性表示。 该项目系统地研究了气象的应用领域内的高维数据中的不确定性可视化。跨学科研究团队跨越不同的领域，例如可视化，大气科学和统计。研究团队正在研究（1）使用自举平均置信区间来量化不确定性，这些间隔放松了对基础分布的假设； （2）设计多种不确定性可视化方法并在集成的用户界面中实现这些方法； （3）采用定性和定量用户研究来评估不确定性可视化方法。所得的工具和方法已部署到环境建模中心。尽管该项目的主要应用领域是气象学，但所学的原则有可能应用于各种科学领域，包括水文学，地理空间科学，医学成像和生物学。该项目的研究工作被整合到密西西比州立大学的可视化和气象课上。