HR: 0830h
AN: U41B-0001 [PDF]
TI: Bringing Selected MODIS Vegetation Indices Over to an Open GIS Consortium Web Map Server
Environment
AU: Dwyer, J L
EM: dwyer@usgs.gov
AF: USGS/EROS Data Center, EROS Data Center, Sioux Falls, SD 57198
AU: * Patrick, K W
EM: kpatrick@usgs.gov
AF: USGS/EROS Data Center, EROS Data Center, Sioux Falls, SD 57198
AB:
Over the past few years, the Open GIS Consortium, OGC, has been building specifications to enable open data use across
platforms and software packages. The OGC specifications have reached a maturity level that enables an alternative means for
distributing satellite remotely sensed data acquired by NASA's Earth Observing System, EOS, particularly data products from
the Moderate Resolution Imaging Spectroradiometer, MODIS. A spatially-seamless, time-series dataset of the MODIS vegetation
normalized difference vegetation index, NDVI, covering the conterminous United States was made accessible through a
world-wide-web interface to enable exploratory data visualization. User-specified spatial and/or temporal subsets of these
data may be retrieved electronically as map graphics. \\ The standard NDVI data are a 1 km grid product in the Sinusoidal
projection that is organized as a set of non-overlapping tiles, $10\deg$ latitude by $10\deg$ longitude, consisting of a
composite of observations for a 16-day period. To facilitate user-specified map compositions, or data extractions of the
standard NDVI products, required several preprocessing steps including data mosaicking, data reprojection and resampling, and
metadata compilation. Once the MODIS NDVI were compiled into a seamless mosaicked dataset and hosted on an OGC compliant
data server, the data became accessible through web-based mapping services and applications. \\ By hosting the MODIS NDVI
data over on an OGC compliant data server, the accessibility and usability of the data through a wider variety of users'
applications became possible. This not only enables the use of all the OGC tool sets, but also enables greater flexibility
for data visualizations and map rendering, independent of the end-users' system and software packages. Different data sets
from any number of sources are automatically co-registered, independent of the initial map projections of the originating
datasets. The net results are faster and more efficient web-based access to data spanning different sources, spatial scales,
and time periods. \\
DE: 1640 Remote sensing
DE: 1694 Instruments and techniques
DE: 1699 General or miscellaneous
SC: U
MN: 2003 Fall Meeting