The
MAP ’06
Project: A NASA Contribution to Tropical Cyclone Research
Introduction:
During the summer of 2006 the Earth-Sun
Exploration Division (now
the Earth Sciences
Division) of Goddard
Space Flight Center(GSFC) and the Science
and Mission Systems Office at Marshall
Space Flight Center brought together resources from
NASA and from corporate partners to study tropical cyclones.
The “MAP ’06 Project,” so named for its
affiliation with NASA’s MAP program, applied NASA’s
advanced satellite remote sensing technologies and earth
system modeling capabilities to improve our understanding
of tropical cyclones that develop in and move across the
Atlantic basin. MAP
'06 implemented the most recent version of the Goddard
Earth Observing System (GEOS) fifth-generation global atmospheric
model and the Grid point Statistical Interpolation (GSI)
analysis system under development as a collaboration between
NOAA's National Centers
for Environmental Prediction (NCEP) and the Global
Modeling and Assimilation Office (GMAO) at GSFC. In
addition, the capability to initialize the Weather Research
and Forecast (WRF) regional model using GEOS-5 was developed
and implemented. The project began in the early
portion of the 2006 hurricane season and continued through
late autumn.
Project
Objectives:
The primary objective of MAP ’06 was the application
of NASA’s advanced satellite remote sensing technologies
and earth system modeling capabilities to improve understanding
and prediction of tropical cyclones. Specific science
questions addressed included:
- Can satellite data assimilation improve prediction (out
to 5 days) of easterly waves, tropical cyclogenesis and
hurricanes?
- How sensitive are tropical cyclone predictions to the types and distributions
of assimilated satellite data?
- What are the impacts of increased horizontal resolution in the model
and data assimilation system on tropical cyclogenesis and subsequent
track and intensity forecasts?
- What role do aerosols play in tropical cyclogenesis in the Atlantic
basin?
- What roles do land – ocean – atmosphere interactions
play in the structure and evolution of easterly waves as they propagate
off the western African coast?
Global
Modeling
A unique aspect of the project was the use of
new satellite data sets by scientists at the Global
Modeling and Assimilation Office (GMAO). The GEOS-5 atmospheric
model is a weather-and-climate capable model which has been
integrated with the NCEP/GMAO GSI to produce a new atmospheric
data assimilation system (DAS), the GEOS-5 DAS. The global
model forecasts for MAP '06 were at 1/4 degree with 72
vertical levels. Initial conditions during the early phase
of the project were acquired from the NCEP Global Forecast
System (GFS). Analyses from the GEOS-5 DAS were available
for model initialization in late August. During
the course of the project, depending on the availability
of computational resources, tests of the data assimilation
system were conducted at 1/4 degree resolution. Five-day
global model forecasts were conducted up to twice per
day. A prototype version of GEOS-5
that incorporated the direct effect of aerosols on radiation
was incorporated into the project later in the summer
to study the impact of sub-Saharan dust on tropical cyclone
development in the eastern Atlantic.
Regional Modeling
The Weather
Research and Forecasting (WRF) Model is a next-generation
mesocale numerical weather prediction system designed to
serve both operational forecasting and atmospheric research
needs. It can be operated across a wide range of scales
ranging from 100’s of meters in cloud resolving mode
and 100’s of kilometers on the synoptic scale. Use
of WRF in MAP '06 linked the activities of the MSFC Short-term
Prediction Research and Transition Center (SPoRT) with
regional modeling activities at GSFC and global modeling
capabilities at the GMAO. The intent was to initialize WRF
with GEOS-5 analyses produced by the GSI analysis system
in the tropical Atlantic basin at resolutions of 2-km. This
was done on an on-demand basis as events of interest
unfolded in the tropical Atlantic. The regional modeling
component of the project investigated the impact of high-resolution
(1-km) sea surface temperature composites generated from
the MODIS instrument on air-sea interactions from the perspective
of tropical cyclogenesis and subsequent tropical cyclone
track and intensity prediction. Another area of interest
was land-atmosphere interactions over western Africa. MAP
'06 testedt a prototype version of WRF dynamically coupled
with the Goddard high resolution Land Information System
(LIS). In addition, MAP '06
provided a framework to test alternative versions of WRF
developed through NASA projects designed to study cloud-microphysical
processes.
Potential Field Support for NAMMA-06
The project coincided with the NASA
African Monsoon Multidisciplinary Activities (NAMMA) Extended
Observing Period (EOP). MAP '06 worked with NAMMA-06 mission
scientists to contribute the global model forecasts for field
support of research aircraft in the study cyclogenesis off
the west coast of Africa during Special Observing Period
3 (SOP3: August 15 – September 15). MAP '06 experiments
have the potential to advance our knowledge of the impact
of aerosols and land-sea-ocean interactions on tropical cyclogenesis
in concert with the NAMMA-06 program. Goddard scientists
used satellite data in conjunction with the GEOS-5 analyses
and forecasts to identify direct forcing by dust and black
carbonaceous aerosols prior to and during cyclogenesis. The
WRF regional model was run over areas at high resolution
(2-km) where the global model suggested cyclogenesis might
occur. Special attention was given to air-sea-atmosphere
interactions using the MODIS SST fields and the LIS coupled
to WRF.
Technology and Visualization:
Northrop
Grumman Information Technology, GSFC, and Intel joined
forces to deliver StormGrid, a unique computational solution
for NASA's tropical weather forecasting research. Northrop
Grumman IT, through its Business Integration Competency Center
(BICC) laboratory located at its Colshire facility in McClean,
VA, in collaboration with Intel, provided a high performance
computing (HPC) capability designed to leverage 10 Gig optical
network access and distributed data servers. StormGrid demonstrated
a highly accessible on-demand HPC solution for delivering
services in support of NASA's "MAP
'06" tropical cyclone prediction effort based on the
GEOS-5 global forecast model and the high resolution WRF
regional forecast model.
Northrop Grumman's StormGrid provided an HPC system, a Linux cluster
consisting of 256 Itanium processors, capable of generating
terabytes of research data per day. This research data was
seamlessly transferred back to NASA Goddard's scientific
data portal via the 10-40Gb/s National Lambda Rail ultrafast
optical network to Goddard as
it was produced. The StormGrid project leveraged Northrop
Grumman's internal research in lambda optical networks, distributed
SAN's, HPC, and grid computing systems. NASA used the
data from the model runs to analyze hurricane formation with
the goal of improving future hurricane forecast systems.
StormGrid augmented Agency computing resources provided
by the NASA
Center for Computational Sciences (NCCS) at Goddard and Project
Columbia at the NASA
Ames Research Center.
|
