Mirador: A Simple Fast Search Interface for Global Remote Sensing Data Sets

A major challenge for remote sensing researchers is searching and acquiring relevant data files for their research projects based on content, space, and time constraints. Several structured query (SQ) and hierarchical navigation (HN) search interfaces have been developed to satisfy this requirement. However, the popularity of free-text (FT) search in the general domain led the Goddard Earth Sciences Data and Information Services Center to develop an FT search interface named Mirador that supports space-time queries, including a gazetteer and geophysical event gazetteer. In order to compensate for a slightly reduced search precision relative to SQ and HN methods, Mirador uses several search optimizations to return results quickly, enabling iterative search strategies.

I. Introduction

The goddard earth Sciences Data and Information Services Center (GES DISC) has served as an archive and distribution center for global remote sensing data since 1994. Over that time, one of the most challenging aspects of data distribution has been providing an efficient search interface for diverse data sets and users. Science researchers and other users of remote sensing data spend significant effort identifying data sets supporting their research topic and reducing relevant inventories down to the files within spatial and temporal ranges of interest. This paper describes a simple free-text (FT) search tool called Mirador for searching GES DISC data using practical optimizations to return both data-set- and file-level results quickly.