Transcript Slide 1

Linked Environments for Atmospheric Discovery
Transforming the sensing and numerical prediction of high impact local
weather through dynamic adaptation
Vision
Revolutionize the ability of scientists, students, and operational practitioners to
observe, analyze, predict, understand, and respond to hazardous or severe
weather by interacting with it dynamically and adaptively
Motivation
1)
2)
Weather technologies are hard to use in sophisticated ways because they’re complicated and linked together
using cumbersome software - creating a HUGE DIVIDE BETWEEN THE HAVES AND THE HAVE NOTS
High impact local weather is VERY DYNAMIC while our tools, cyber environments and learning modalities are
highly STATIC
GOALS
#1: Lowering the barrier for using complex end-to-end weather technologies
Democratize the availability of advanced weather technologies for research and education
Empower application in a grid context
Facilitate rapid understanding, experiment design and execution
#2: Dynamic Adaptation to Weather
Models and hazardous weather detection systems responding to observations and their own output
Models and hazardous weather detection systems driving the collection of observations
IT infrastructures providing on-demand, fault tolerant services
What can LEAD do for you?
Researchers
Access to:
Educators
Access to:
Powerful tools for Assimilation, Prediction, Mining and Visualization
Supercomputing resources - TeraGrid
Data – real-time and recent data resources
LEAD Learning Communities – tools and colleagues
Classroom demonstrations – hands on access to tools
Data –real-time and recent data resources
Students
Access to:
Resources to learn about and visualize the weather
With approval – create experiments access tools
Data –real-time and recent data resources
Learn more at: http://portal.leadproject.org
LEAD is sponsored by the National Science Foundation
Linked Environments for Atmospheric Discovery
http://portal.leadproject.org
1 Traditional NWP Methodology is Static
Analysis/
Assimilation
STATIC
OBSERVATIONS
2 LEAD Provides Dynamic Capabilities
through a Web Portal and Workflows
Prediction/
Model
Graphical
Domain
Selection
The Process is Entirely
Prescheduled and Serial;
It Does NOT Respond to the
Weather!
End Users
NWS
Private Companies
Students
3
How? : Built on (Web)
Services
[But in a Grid Framework]
Product
Generation,
Display,
Dissemination
Service A
(Forecast
Model)
Drop down menu
for selection of runs
Service B
(Terrain
Preprocessor)
Service C
(Interpolator
Service)
4 How? :
Workflow
Generation
Pre-configured
workflows
automatically
generated, which can
be modified by users
via a composition tool
– launch manually or
via trigger (dynamic)
LEAD is sponsored by the National Science Foundation
Many
others…