A user-centered design for a spatial data warehouse for data exploration in environmental research
详细信息 查看全文
- 作者：Michael McGuire ; Aryya Gangopadhyay ; Anita Komlodi ; Christopher Swan
- 关键词：Spatial data warehouse ; Spatial OLAP ; Data exploration ; Ecological research ; User-centered design
- 刊名：Ecological Informatics
- 出版年：2008
- 出版时间：October 2008
- 年：2008
- 卷：3
- 期：4-5
- 页码：273-285
- 全文大小：2089 K
文摘
The integration of data from diverse fields of ecological research is paramount in the discovery of new ecological patterns and processes. The spatial exploration of an integrated dataset that spans multiple studies and disciplines can allow researchers to gain unforeseen insight into their data, spawn new research questions and hypotheses and identify data gaps. A user-centered approach was taken to design a spatial data warehouse and online analytical processing (OLAP) tools for data exploration in ecological research. The users in this study had diverse needs and current methods of data management do not easily allow for integration and exploration of data in multidimensional space. A generalizable data warehouse design methodology was created based on the results of a user study. This methodology was then demonstrated in the design of a data warehouse for data exploration in stream ecology resulting in a multidimensional data model with a fact table representing biological stream survey measurements and dimension tables representing spatial and categorical site and landscape variables. A generalizable extraction transformation and loading (ETL) workflow was created to integrate data across spatial dimensions before it was loaded into the data warehouse. A prototype data warehouse was implemented using biological stream survey, hydrologic, and vegetation data to observe spatial patterns in biological community distributions. Based on the exploration requirements identified in the user study, prototype OLAP queries were designed to facilitate spatial data cube exploration. Finally, a web-based interface was implemented to allow for multidimensional spatial visualization of biological stream survey data. The data warehouse and interface will allow researchers to explore biological assessment data at multiple spatial scales across many dimensions.