IEEE Geoscience and Remote Sensing Magazine - December 2015 - 10

and process data closer to the archive and facilitating access to lower-level core web services fronted by middleware
services, the LP DAAC is enabling Earth science data users
to achieve more expansive data access and manipulation
in significantly less time and at a substantially lower cost
(Werpy & Torbert, 2014).
Improved data interaction, analysis, and visualization of
information derived from data will require a combination
of data and compute servers
which perform much of the
processing transparently, before returning responses to a
theSe teChNOLOGY
user request. This concept reADvANCeMeNtS MAKe It
quires migrating more of the
POSSIBLe tO eXtrACt PIXeL
data processing and interaction toward systems closer
vALUeS FOr eXACt POINt
to the data archives than to
LOCAtIONS FrOM
the data consumer. Tool deINDIvIDUAL BANDS WIthIN
velopers must consider how
PrODUCtS FOr SPeCIFIC
scientists need to interact
DAteS Or DAte rANGeS.
with large volumes of data
archived at various physical
locations. Ideally, tools can
be developed for earth scientists to access remote sensing archives for answers to science
questions instead of links to large volumes of data to download and process locally (Davis & Maddox, 2009).
To address these burgeoning data archive use requirements, technology researchers at the LP DAAC are developing a visual client platform, consisting of underlying modern, open standards services, with the client itself serving
as a "reference model" for internal and external developers
to access the platform. This permits the operational deployment of the client, and a means for experimentation and
Agile (Agile Alliance) software development of new capabilities by the LP DAAC, while also providing a means for
external developers to "mash-up" their own clients based
on this platform. This modular approach permits rapid
prototype evolution of the client based on dynamic user requirements, and simultaneously permits the integration of
a range of data sources and services in support of new and
emerging users and projects.
The following sections describe the services and tools
developed for a new paradigm of information discovery for
users. This new paradigm is meant to supplant traditional
data download and processing, reducing time to obtain answers to science questions from days, weeks or months to
seconds, minutes or hours - almost interactively.
Section 2 provides background information about the
LP DAAC MODIS archive, and the data used for this work.
Section 3 explains OPeNDAP at a high level, and the local implementation at the LP DAAC. Section 4 presents
the middleware services and applications developed to
expose the MODIS archive through web services. Section
5 describes an application for consuming services. Section
6 defines the advantages of creating and accessing "Analy10

sis Ready Data", with planned performance improvement
strategies outlined in Section 7.
2. LP DAAC MODIS ArChIve ACCeSS
The MODIS sensor has been consistently providing remotely sensed data products since 2000. From that time,
the demand for and the utility of MODIS data have
steadily increased. Most standard MODIS land products
use a global Sinusoidal grid tiling system. The grid tiles
are approximately 10 degrees in latitude by 10 degrees
in longitude. The tiling system uses a horizontal (x) and
vertical (y) tile coordinate scheme to determine the location of each tile within the grid. The tile coordinate
system starts in the upper-left corner of the grid at horizontal tile coordinate 0 and vertical tile coordinate 0, i.e.
(0,0). The horizontal tile coordinates proceed left to right
from 0 to 35 while the vertical tile coordinates proceed
from top to bottom from 0 to 17. The location of pixels within each tile are referenced in a similar manner,
where the location of a specified pixel is x, y distance
from the upper-left corner of the tile, i.e. (0,0). This gridded coordinate scheme is suitable for granule-based access methods and can also be leveraged when developing
pixel-based access methods.
For more information about LP DAAC data and MODIS
products, visit: https://lpdaac.usgs.gov.
Data archive access methods have progressively evolved
over the past fifteen years. Previously, MODIS granules
needed to be manually downloaded individually. With the
advent of File Transfer Protocol (FTP) and HTTP technologies, users are able to download thousands of granules at a
time. More recently, additional services were built on top of
existing tools allowing users to spatially subset data granules. Users are able obtain the data needed over specific
geographic area without downloading the entire granule
containing the area of interest. The latest technology advancements can now allow users to investigate and interact
with the data like never before. These advancements make
it possible to extract pixel values for exact point locations
from individual bands within products for specific dates
or date ranges. However, this capability requires data archives to be configured to receive requests for interactive
data analysis.
3. LP DAAC OPeNDAP IMPLeMeNtAtION
OPeNDAP, an acronym for "Open-source Project for a
Network Data Access Protocol", is a data transport architecture and protocol widely used by Earth scientists. The
protocol is based on HTTP and the current specification
is OPeNDAP 2.0 Draft. OPeNDAP includes standards for
encapsulating structured data, annotating the data with
attributes and adding semantics that describe the data.
The protocol is maintained by OPeNDAP.org, a publicly
funded non-profit organization that also provides free reference implementations of OPeNDAP servers and clients
(OPeNDAP, 2014).
ieee Geoscience and remote sensing magazine

december 2015

https://lpdaac.usgs.gov http://www.OPeNDAP.org

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