Developing a Comprehensive Set of Tools for Large-Scale Environmental Science Data Analysis

Stanford University

Abstract:
In environment science, geographic data is currently collected and
made available by federal, state, and local agencies along with some
local research projects. Although this data is freely available, it
is a tedious and painful process to download and use the data,
particularly if you need data from across several different data
sets. Our goal behind this project is to develop a cloud of tools
which the scientist can use for both data retrieval and data
analysis. The set of tools shall be available via a web service with
an easy to use web interface.

The first step towards this goal has been to support easy data
retrieval, preliminary analysis and data manipulation of satellite
imagery including spatial and spectral subsetting, reprojection and
mosaicing of images. The next step is leveraging the idea of cloud
computing to host the files on the high performance server and allow
scientists to access the data via the cloud instead of requiring
large amounts of data to be downloaded on the user's hard disk. This
can allow for the development of sophisticated tools that can
manipulate data directly on the cloud rather than those that operate
on locally stored data(which is currently done by tools such as
Matlab and Excel).

Host of Seminar:
Keith Jackson