HYDROSEEK

Abstract

Search engines have made considerable contributions to the overall web experience. However locating scientific data remains a problem since databases are not readily accessible by search engine bots. Considering different temporal, spatial and thematic coverage of different scientific data repositories, especially for interdisciplinary research it is typically necessary to work with multiple data sources. Today integration of hydrologic data sources are mostly at the level of content aggregation by providing links to several data providers on a web page. However being able to query multiple databases simultaneously is a feature that has been sought after since the first data repositories; USGS' National Water Information System (NWIS) and EPA's Storage and Retrieval System (EPA STORET) came online. This study examines the current state of hydrologic data availability and dissemination in the US. It identifies the data accessibility problem and suggests a data discovery mechanism named Hydroseek as a solution. Hydroseek enables querying multiple hydrologic data repositories through a single interface and effectively combines spatial, temporal and thematic aspects of search in order to make it possible to discover more of the desired data in less time. It provides a unified view despite heterogeneity issues within and among data repositories, allows data discovery using keywords which eliminates the need to know source specific parameter codes, improves data browsing capabilities by incorporating data classification based on conceptual hierarchy and has an interface design capable of providing access to a large data inventory without overwhelming the user. System's performance was evaluated based on statistical analysis of a user study in which users were asked to perform a certain data retrieval task using currently available systems and Hydroseek .