Title: Semantic Metadata for Scientific Data Access and Management

We are developing and applying techniques for representing semantic
metadata to various problems in scientific data access and management,
including integrating scientific data from heterogeneous sources,
automating the capture of science metadata, and organizing complex
interconnected scientific information.

Heterogeneous Scientific Data Integration

To make sense of mission data, scientists must integrate it with data
collected in other contexts (e.g., during prior missions or as part of
previously conducted scientific experiments). Accessing and assembling
distributed, heterogeneous scientific data is difficult and
time-consuming. We are developing semantic data models to enable
automated integration of information across disparate data sources. Our
current work focuses on integrating various sources of MER data, as well
as Mars analog field study data.  However, the techniques are applicable
independent of the scientific domain.
Automated Science Metadata Capture and Management

Adequate scientific metadata – including information on data format and
content, collection context, provenance, data generation processes, etc.
– is essential for proper interpretation of scientific results. Yet
metadata is difficult to capture and typically requires costly human
effort to manage. As a result, most scientific data is minimally
annotated and cannot be used without significant consultation. We have
developed techniques that capture important contextual metadata at the
point of data generation, and store the metadata using rich semantic
annotations to facilitate reuse by people and automated tools.

Collaboratories for Remote Instrument Experimentation and Integrated
Data Management

Working with microbiology and exobiology colleagues in the Astrobiology
Institute (principally, Brad Bebout), we have developed a collaboratory"
in a greenhouse atop Building 239.  In this greenhouse, microbial algae
mats collected from various astrobiology field sites are incubated and
maintained in aquarium tanks.  Using our collaboratory software,
scientists can conduct remote experiments with individual mats,
measuring oxygen concentration levels at various depth levels using a
microsensor probe mounted on a motor above the aquarium.  Our team has
developed an interface to remotely schedule, control, and monitor these
oxygen microsensor experiments over the web.  The remote control setup
interfaces with the ScienceOrganizer information repository, and the
microsensor data is deposited directly into the system, where it can be
remotely accessed with suitable permissions.

Scientific Knowledge Management with ScienceOrganizer

The ScienceOrganizer system is a specialized science data and knowledge
management tool designed to enhance the information storage,
organization, and access capabilities of distributed NASA science teams.
ScienceOrganizer provides a common electronic repository in which
science team members can store and share project information. Team
members access the system through an intuitive Web-based interface that
enables them to upload, download, and organize project information --
including data, documents, images, and scientific records associated
with laboratory and field experiments. Information in ScienceOrganizer
is "threaded", or interlinked, to enable users to locate, track, and
organize interrelated pieces of scientific data. Linkages capture
important semantic relationships among information resources in the
repository, and these assist users in navigating through the information
related to their projects. ScienceOrganizer has served over 250 NASA
scientists working in teams at ARC, JSC, and throughout the agency,
including teams in astrobiology, astrobionics, ecogenomics, electron
microscopy, microbiology, epidemiology, and ecosystem sciences. In
addition, the system was adapted and used for evidence collection and
analysis by the Columbia Accident Investigation Board.