A PROPOSED GEOCHEMISTRY DATA MODEL

We propose a data model to be used as a framework for geochemical databases, as a data structure for geochemical modeling, and for use in geographic information systems. This model is an outgrowth of a Microsoft Access database structure developed for and successfully utilized in USGS Mineral Resources Program projects.

The model contains two primary tables—FieldSite and Sample—that are the bilateral core around which tables of environment descriptions and analytical results are associated (fig. 1). Information pertaining to the sample location and its physical environment is managed through the FieldSite table, which links to tables containing data about the environment at large; such as climate, topography, geology, hydrology, ground cover, and ecosystems. Information pertaining to the chemical analyses of samples is managed through the Sample table, which links to table groups recording analytical results and associated information; such as quantitative and qualitative values, laboratories, and analytical methods and parameters. Optional ancillary tables in the model include publication references for sample sites and environmental data, and tables recording quality assurance and quality control data. The data model also includes a Field Name Dictionary, a standalone metadata table incorporated into the general database structure that describes all the database fields and lists all the tables that contain a particular field.

The analytical data are provided in two different table groups to facilitate different data access and query approaches. The first group consists of tables that contain aggregates of all analytical results, and the second group comprises individual tables generated for each analytical method. This duplication was designed to provide flexibility in querying, and allow the export of tables to other tabular software formats that cannot manage large datasets.

The simple structure of this data model provides a flexible interface to (1) manage large amounts of multi-element geochemical data determined by many methods, (2) explore the relationships of geochemistry with various environmental factors, and (3) obtain information on the range of methods and quality of results. The architecture supports a full spectrum of geochemical and geospatial modeling and analyses, and can be implemented in several database platforms, including Microsoft Access, Oracle and ESRI Geodatabases.

Figure 1. Geochemistry data model structure. Rectangular boxes are single tables; rounded boxes are table groups; dotted groups are optional.