Applying Community Standards for Domain-Relevant Metadata to Enhance Data Product FAIRness

The reusability of data from airborne field campaigns should be supported by detailed metadata relevant to the measurements reported. The Atmospheric Composition Variable Standard Names Convention (ACVSNC) is a community standard which can provide domain-relevant metadata embedded within the data file, e.g., ICARTT data files. The structured format and controlled vocabulary of the ACVSNC is consistent with FAIR principles, enabling data to be Findable, Interoperable, and Reusable. To further make data findable, the data must be human and machine-actionable. One approach to do this is through the I-ADOPT ontology-based framework, which was designed to “harmonize the way observable properties are named and conceptualized across scientific domains”. The Aerosols, Clouds and Trace Gases Research Infrastructure (ACTRIS) vocabulary has successfully implemented the I-ADOPT approach. We have mapped the ACVSNC into the ACTRIS vocabulary to demonstrate the compatibility of ACVSNC with an ontology-based framework, which suggests the potential contribution to the United Metadata Model for Variables (UMM-Var) that NASA has developed. Mapping between ACTRIS and the ACVSNC has highlighted the ambiguity and subjectivity in assigning ontology components. For example, many variables within the field of atmospheric composition describe a process, not a physical being, which can be difficult to decompose into components. Mapping the ACVSN to ACTRIS has provided insight to improve the implementation of ontological approaches to make accurate variable metadata more machine-actionable and human-readable to a broad spectrum of data users/researchers. This process will involve the synergistic collaboration between data scientists and physical scientists.机载野外观测实验所产生的数据的可复用性，需依托与所报告测量数据配套的详细元数据予以保障。大气成分变量标准命名规范（Atmospheric Composition Variable Standard Names Convention, ACVSNC）是一项社区通用标准，可在数据文件（如ICARTT数据文件）中嵌入领域相关元数据。该规范的结构化格式与受控词汇表符合FAIR原则，可实现数据的可发现性、互操作性与可复用性。为进一步提升数据的可发现性，需确保数据兼具人类可读性与机器可操作性。实现该目标的一种途径是采用基于I-ADOPT本体的框架，该框架旨在“协调各科学领域内可观测属性的命名与概念化方式”。气溶胶、云与痕量气体研究基础设施（Aerosols, Clouds and Trace Gases Research Infrastructure, ACTRIS）已成功落地I-ADOPT方案。我们已将ACVSNC映射至ACTRIS词汇表，以验证ACVSNC与基于本体的框架的兼容性，这表明其可为美国国家航空航天局（NASA）开发的变量统一元数据模型（United Metadata Model for Variables, UMM-Var）提供潜在助力。ACTRIS与ACVSNC之间的映射工作揭示了本体组件分配过程中存在的模糊性与主观性。例如，大气成分领域的诸多变量描述的是过程而非物理实体，此类变量难以拆解为本体组件。将ACVSNC映射至ACTRIS的工作，为优化本体论方法的落地提供了思路，可让准确的变量元数据更易于被广大数据用户与研究人员理解，同时具备机器可操作性。这一过程需要数据科学家与物理科学家开展协同合作。