Remote Sensing Data Access – A review of online resources for hyperspectral imagery

Hyperspectral CubeIn our previous post – Remote Sensing Data Archives – we explored some of the many general online data discovery tools for obtaining remote sensing imagery. We now sharpen our focus to the field of hyperspectral remote sensing, aka imaging spectrometry, and delve into resources for accessing this particularly versatile type of imagery.

Hyperspectral imaging emerged on the remote sensing scene in the 1980s, originating at the Jet Propulsion Laboratory (JPL) with the development and deployment of the Airborne Imaging Spectrometer (AIS), followed soon thereafter by the Airborne Visible Infrared Imaging Spectrometer (AVIRIS). Since then hyperspectral imaging has evolved into a robust remote sensing discipline, with satellite and airborne sensors contributing to numerous applications in Earth observation, and other similarly sophisticated sensors being used for missions to the moon and Mars.

The premise behind hyperspectral imaging is that these sensors measure numerous, relatively narrow, contiguous portions of the electromagnetic spectrum, thereby providing detailed spectral information on how electromagnetic energy is reflected (or emitted) from a surface. To give this some perspective and provide an example, this can equate to measuring the visible portion of the spectrum using 50 or more narrow bands as opposed to three broad bands (i.e., red, green and blue) that we typically see with cameras and our eyes. Because objects (plants, soil, water, buildings, roads, etc…) reflect light differently as a function of their composition and structure, this enhanced spectral resolution offers more information with which to identify and map features on the Earth’s surface.

For those interested in hyperspectral remote sensing, and curious to see what can be achieved using this type of data, let’s look at some of the archives that are available:

  • Hyperion – The Hyperion sensor (220 bands; 400-2500nm; 30m resolution) is located on NASA’s EO-1 satellite, and although deployed in 2000 as part of a one-year demonstration mission, the satellite and its onboard sensors have shown remarkable stamina, continuing to collect data today. Archive data from Hyperion are available through both Earth Explorer and GloVis, and new data can be requested through an online Data Acquisition Request (DAR).
  • HICO – The Hyperspectral Imager for Coastal Ocean sensor (128 bands; 350-1080nm; 90m resolution) was installed on the International Space Station (ISS) in 2009 and is uniquely configured for the acquisition of ‘dark’ targets, specifically coastal aquatic areas. The sensor was initially developed and sponsored by the Office of Naval Research, with continuing support now provided through NASA’s ISS Program. Archive data from HICO, as well as requests for new data, are available through the HICO website hosted by Oregon State University; however, interested users must first submit a short proposal to become part of the HICO user community.
  • CHRIS – The Compact High Resolution Imaging Spectrometer (18-62 bands; 410-1020nm; 17-34m resolution) is the main payload on ESA’s Proba-1 satellite, which was launched in 2001. As with the EO-1 satellite, Proba-1 was only intended to serve as a short-lived technology demonstrator, but has managed to continue collecting valuable science data for more than a decade. Data from CHRIS are available to registered users, obtained via submittal and acceptance of a project proposal, through ESA’s Third Party Missions portfolio on Earthnet Online.
  • AVIRIS – The Airborne Visible Infrared Imaging Spectrometer (224 bands; 400-2500nm, 4-20m resolution) has been supporting hyperspectral projects for more than two decades, and can be credited as a true pioneer in the field. AVIRIS is most commonly flown onboard a Twin Otter turboprop or ER-2 jet, but has also been configured to operate from several other airborne platforms. Images from 2006-2011 are available through the AVIRIS Flight Data Locator, with plans to soon expand this archive to include additional imagery from 1992-2005 (currently available through request from JPL).
  • NEON – The National Ecological Observatory Network is a continental-scale network of 60 observation sites located across the United States, where a standardized set of field and airborne data are being collected to support ecological research. Remote sensing data are being acquired via the Airborne Observation Platform, which includes a high-resolution digital camera, waveform LiDAR, and imaging spectrometer. The NEON project is adapting an open data policy, but data acquisition and distribution tools are currently still in development. Thus, initial “prototype” data, which includes a sampling of hyperspectral imagery, are being made available through the NEON Prototype Data Sharing (PDS) system.
  • TERN – The Terrestrial Ecosystem Research Network is an Australian equivalent of NEON, providing a distributed network of observation facilities, datasets, map products and analysis tools to support Australian ecosystem science. Within this larger project is the AusCover facility, which leads the remote sensing image acquisition and field validation efforts for TERN. Current hyperspectral datasets available through AusCover include both airborne data and a comprehensive collection of Hyperion imagery. Data are accessible through the TERN Data Discovery Portal and the AusCover Visualization Portal.

These aren’t the only hyperspectral instruments in operation. There are new instruments, such as the Next Generation AVIRIS (AVIRIS-NG), Hyperspectral Thermal Emission Spectrometer (HyTES) and Portable Remote Imaging Spectrometer (PRISM), which all recently conducted their first science missions in 2012. There are a growing number of hyperspectral programs and instruments operated by government agencies and universities, such as the NASA Ames Research Center and the Carnegie Airborne Observatory (CAO). There are various airborne sensors operated or produced by commercial organizations, such as the Galileo Group, SpecTIR, HyVista and ITRES. And there are also a number of new satellite-based sensors on the horizon, including HyspIRI (NASA), EnMAP (Germany), PRISMA (Italy) and HISUI (Japan).

It’s an exciting field, with substantial growth in both sensor technology and analysis methods continuing to emerge. As the data becomes more and more available, so too does the potential for more researchers to get involved and new applications to be developed.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s