Mining Spatio-Temporal Data
Earth Systems Science
John Vande Castle, Robert Waide, University of New Mexico, All Campuses
Last modified: 27-AUG-04

---

Contact Information:
John Vande Castle
phone: 505/272-2634
fax: 505/272-7080
jvandecastle at lternet.edu


The University of New Mexico’s NPACI/ESS effort has been to develop the core elements of the LTER Spatial Data Workbench (SDW). The goal of the SDW (http://www.lternet.edu/technology/sdw/ and http://sdw.sdsc.edu/) is to establish an integrated knowledge management system that uses extensible and scalable infrastructure to provide data and analytical services to the broad LTER Network community of field biologists and associated collaborators working at these sites (over 1500 individuals). The SDW is designed to provide simplified access to complex datasets such as hyperspectral remote sensing data to researchers who would not be able to use the data.

The current SDW consists of several components, including a SRB-based collection management system for remote sensing imagery, analytical pipelines (workflows) for high-throughput processing of remote sensing data, mining algorithms for analysis, band selection and classification, and dynamic Internet mapping of analysis results. Although the primary focus of the SDW is for improved access to hyperspectral remote sensing image data, the collection also includes time series data products as well as hyperspatial (image resolutions of 1 meter or less) data holdings of the LTER Network. The access menus include browse images of overflight data (Figure 1) and direct access to subset browse images (Figure 2). The current holdings of the SDW include over 0.75 terabyte of data in more than 48 thousand files of imagery and documentation. This includes hyperspectral AVIRIS data (http://aviris.jpl.nasa.gov/) for five of the LTER sites and hyperspatial Global Fiducial Library (http://www.lternet.edu/technology/ltergis/) data for two LTER sites. Landsat Thematic Mapper data (http://landsat7.usgs.gov/index.php), AVHRR raw and NDVI data (http://edc.usgs.gov/products/satellite/avhrr.html) and MODIS data product (http://modis.gsfc.nasa.gov/) time series data for all LTER sites (http://www.lternet.edu/technology/satellite/) have also been included in the SDW. The SDW is not meant to be a static archive, but to continue well beyond the scope of NPACI. The recent inclusion of an additional 0.25 terabyte of MODIS data products are an example of datasets that will be added from time to time to the LTER Spatial Data Workbench.

The primary goal of the University of New Mexico’s FY04 NPACI effort has been to migrate the project beyond the NPACI framework. This has been accomplished by two means. First, the existing infrastructure within the Storage Resource Broker (SRB) at SDSC has been maintained for open access to the GRID computing environment. Second, the SDW has been migrated to a database as part of the LTER Network Information System (LTER-NIS) for long-term preservation of the project. The developing LTER Network Information System will ultimately provide integrated data access for spatial datasets of the SDW as well as those maintained at individual LTER sites. In addition, the incorporation of the SDW into the LTER-NIS will permit the direct integration of the SDW spatial data with the other "non-spatial" datasets of the LTER Network. The LTER-NIS will also include Science Environment for Ecological Knowledge (SEEK) processing environments which will permit the initial processing pipelines developed in the SDW to continue beyond NPACI. Both the LTER-NIS and SEEK efforts themselves are developing, so this integration has become a natural part of plans for both SEEK (http://seek.ecoinformatics.org/) and the LTER-NIS (http://intranet.lternet.edu/projects/informatics/).

Scripts have been written for generation of structured metadata - Ecological Metadata Language (EML) from existing metadata of the SDW and are now being implemented. All data are documented (http://www.lternet.edu/technology/sdw/) and are available through the web interface and ftp access on the LTER server (ftp.lternet.edu). The final efforts of the SDW will focus on the updated metadata documentation as Ecological Metadata Language (EML) and incorporation of the data within the Metacat server and SRB at the LTER Network Office. In addition, the collaborative partners Tony Fountain at the San Diego Supercomputer Center (SDSC) and Dylan Keon and Cherri Pancake at Oregon State University, Northwest Alliance for Computational Science and Engineering (NACSE) will add to the processing and visualization components of the Spatial Data Workbench.

Image 1:

AVIRIS_browse_image.jpg Credit: U.S. LTER Network Office Caption: This is a browse image of AVIRIS data that is displayed within the menu system of the Spatial Data Workbench data ccess page.

Image 2:

Browse_Image_of_AVIRIS_flightlines.jpg Credit: U.S. LTER Network Office. Caption: This is a view of the browse image layout for description of AVIRIS data within the Spatial Data Workbench.