IEEE Geoscience and Remote Sensing Magazine - December 2015 - 9

to access remote sensing archives for immediate answers to
science questions instead of links to large volumes of data to
download and process.
Exposing data and metadata to web-based services enables machine-driven queries and data interaction. Also,
product quality information can be integrated to enable
additional filtering and sub-setting. Only the reduced content required to complete an analysis is then transferred to
the user.
1. INTRODUCTION
oday's search and discovery tools for satellite-based
remote sensing data are often catalog-based and rely
on query systems which use individual granule-level
metadata for those queries. Though these traditional
catalog systems are often robust, progress has been limited in enabling interactive metadata mining to aid in
the data search and extraction process. Remote sensing
users face challenges in deriving pertinent information
from the vast data holdings of satellite-based land remote sensing data archives.
Large catalog-based search and discovery systems
such as the U.S. Geological Survey (USGS) Earth Explorer (EE) system and the National Aeronautics and Space
Administration (NASA) Earth Observing System Data
and Information System (EOSDIS) Reverb system provide comprehensive access to these data holdings, but
do little to exploit the full utility of the underlying data
and metadata. These catalog-based systems are extremely flexible, but are manually intensive and often require

IMAGE LICENSED BY INGRAM PUBLISHING

DECEMBER 2015

IEEE GEOSCIENCE AND REMOTE SENSING MAGAZINE

a high level of user expertise to extract the desired data
from the vast holdings of remote sensing data with
greatly varying attributes and characteristics. Exposing
scene-based data and metadata to external, web-based
services can enable machine-driven queries to aid in
the search and discovery process. Furthermore, services
which expose additional scene-based content data (such
as product quality information) are now available and
can provide a "deeper look" into remote sensing data
archives too large for efficient manual search methods
(Steinwand & Davis, 2014).
One of NASA's 12 EOSDIS Distributed Active Archive Centers (DAAC), the
EXPOSING SCENE-BASED
LP DAAC processes, archives,
DATA AND METADATA TO
and distributes land remote
EXTERNAL, WEB-BASED
sensing data for the public remote sensing community. The
SERVICES CAN ENABLE
LP DAAC operates as a partMACHINE-DRIVEN QUERIES
nership between the USGS
TO AID IN SEARCH AND
and NASA. In addition to
DISCOVERY.
several other remote sensing
data products, the LP DAAC
archives and distributes land
data products collected by the
Moderate Resolution Imaging Spectroradiometer (MODIS)
sensors located onboard NASA's Terra and Aqua satellites.
Large-scale archive data providers such as NASA's LP
DAAC, located at the USGS Earth Resources Observation
and Science (EROS) Center have recently begun implementing electronic services for web-based Hyper Text
Transfer Protocol, or HTTP, interfaces to manipulate, organize, modify, and deliver Earth science data and information to users. This web architecture provides the foundation for streamlining user exploration and exploitation
of data. However, missing critical components need to be
developed to increase capabilities and performance, expand the scope, and achieve the potential of these services. Middleware services represent a class of data services
able to communicate capabilities more clearly and effectively with science data users while also efficiently interacting with lower-level web services unseen by users. A
multi-layer services architecture with a middleware layer
of services can efficiently coordinate interactions between
users and lower-level core web services. This layering of
services can simplify data selection, integration, delivery,
and analysis activities. The LP DAAC utilizes core services
to provide basic access to archives, and data manipulation
and processing.
Augmentation of existing capabilities is underway
through the development and implementation of middleware to enable workflows necessary to empower science
data users to perform meaningful data discovery and analysis more efficiently. The middleware layer acts as the "glue"
that allows multiple services to work together in processing workflows. By moving the algorithms which organize
9

Table of Contents for the Digital Edition of IEEE Geoscience and Remote Sensing Magazine - December 2015