Exploring new approaches to prediction
across scales
Understanding of the Earth system is a prerequisite to predicting its
behavior, the latter being however of a more direct use to many components
of society. In that context, the priorities within the laboratory deal
with improving climate models, exploring new approaches to prediction
across scales and global and local weather prediction. The NCAR
highlight deals with the WRF/ARW. Furthermore, the laboratory presents
two highlights on CCSM and on the international experimental endeavor
THORPEX.
THORPEX
| |
 |
The
top panel shows the spatial distribution of averaged CO2
at 1 m above the ground over about a 1 km x 1km area and the bottom
panel shows the CO2 along the
distance across Como Creek (140 m). The plot demonstrates that nighttime
CO2 accumulation was transported
toward low lying grounds, even toward the 2-m-wide creek along the
main slope; however, the spatial distribution of CO2
was also affected by low respiration from the water and local mixing
generated by the thermal contrast between the water and its surrounding
land, leading to low CO2 right
above the water (not included in the top panel).
High resolution figure |
Over the last several decades, the skill of numerical weather prediction
is generally considered to have increased at an average rate of about
one day per decade. Thus, the skill of today’s four-day forecast
is equivalent to the skill of a three-day forecast of a decade earlier.
The rate of improvement is even slower for the forecast variables needed
most by society, such as the prediction of heavy rainfall. This relatively
slow, linear rate of improvement is not sufficient to keep pace
with the demand for accurate weather information in the world today, where
an exponentially growing world population places an ever-increasing
number of people in areas at risk for weather disasters. In view of this
situation, the approximately 180 nations and territories of the World
Meteorological Congress initiated THe
Observing System Research and Predictability Experiment (THORPEX)
under the World
Weather Research Program to conduct research that will accelerate
the rate of improvement in the predictive skill of high-impact weather
and to improve the utilization of weather information for the benefit
to society, the economy and environmental stewardship. The time-scale
of interest to THORPEX is the 1 to 14-day forecast with collaborative
efforts aimed at seasonal prediction. The goals of THORPEX match the NCAR
Strategic Plan in the area of improving the prediction of the Earth System.
THORPEX began its implementation phase in 2005 and the Earth and Sun Systems
Laboratory (ESSL) staff, Christopher Snyder, Rebecca Morss and David Parsons,
have been involved in the national and international leadership of the
program. The North
American and U.S. Project Office for THORPEX is currently located
in TIIMES and staffed by Parsons and Pamela Johnson.
During the past year, several THORPEX activities have been initiated
under the leadership of staff at ESSL. These activities include: i) The
scientific plans development for a major international research campaign
called THORPEX Pacific
Asian Regional Campaign (T-PARC) that will have a field phase in 2008.
T-PARC is the community’s first scale interaction experiment aimed
at medium range weather prediction. The goals of T-PARC are to increase
understanding of the factors that limit our ability to predict both high-impact
weather over the densely populated regions of East Asia’s Pacific
Rim and the downstream effects of these processes on weather events over
North America; ii) Conducted a pilot field experiment in August
and September, 2006 to investigate the roots of hurricane genesis and
to demonstrate the driftsonde technology developed by the Earth Observing
Laboratory (EOL). T-PARC plans to make extensive use of driftsonde. This
work was done in collaboration with Centre National d'Etudes des Telecommunications
(CNES, the French Space Agency), during the African Monsoon Multi-disciplinary
Analysis (AMMA) Campaign; iii) Worked with Computational
& Information Systems Laboratory (CISL) staff to have NCAR as
an archive center for the THORPEX Interactive
Grand Global Ensemble (TIGGE). The intent of TIGGE is to make all
members of the world’s operational forecast ensembles available
for research, thus moving weather research away from determistic modeling
toward a greater focus on probabilistic prediction that better matches
the non-linear character of the atmosphere; iv) Held two community
workshops with significant university participation. The first workshop
concentrated on developing a research plan for collaboration between the
climate and weather research communities in the area of tropical convection.
The second workshop focused on developing a social science agenda aimed
at understanding and improving how society utilizes weather information.
During the next year, activities will focus on several areas. First,
the findings of the two workshops discussed earlier will be written up
and submitted for publication in the Bull. Amer. Meteor. Soc.
to publicize the findings. Staff will work with university visitor(s)
to analyze the output of the Nested Regional Climate Model to advance
the understanding of why both weather and climate models fail to make
adequate predictions of the Madden-Julian Oscillation. This work is a
first step toward acting on the recommendations of the workshop on tropical
convection. We will also analyze the data from the pilot driftsonde effort
to study hurricane genesis campaign. We will also continue to be a community
focal point for T-PARC and TIGGE planning.
return to top |
