Transcript Document

WWRP Nowcasting Working Group
Contributors: T. Keenan (BMRC), Stéphane Sénési (Météo -France),
Shingo Yamada (JMA), Jim Wilson (NCAR), Paul Joe (MSC), Jarmo
Koistinen (FMI), Peter Li (HKO), Augusto Pereira (USP), Andrew
Crook (NCAR), Chris Collier (U. Salford) and Isztar Zawadzki (McGill
U.)
•
•
•
•
Background
Objectives
Overall Strategy
Work plan
11 June 2004
WWRP/NWG-COST 722
Joint Session
Background
•Propose a plan for promoting and advancing nowcasting within the WWRP.
•An outcome of the 6th WWRP SSC recommendation
Definition: Follow that provided by Conway (1998): nowcasting is forecasting
with local detail, by any method, over a period from the present to a few hours
ahead; this includes a detailed description of the present weather.
Nowcasting is now expanded to include the blending of extrapolation
techniques, statistical techniques, heuristic[1] techniques and numerical weather
prediction (NWP).
In the foreseeable future it is most likely that it will be possible to make
deterministic and probabilistic nowcasts for end-users of the 0-6h time period
with sufficient time and space specificity to effectively mitigate losses or
enhance benefits.
[1] Heuristic is defined as forecast rules based on such things as experiment, numerical
simulation, theory, expert systems, fuzzy logic and forecaster rules of thumb.
11 June 2004
WWRP/NWG-COST 722
Joint Session
•A review of conferences and research efforts indicates there is a relative lack of
international focus on nowcasting programs
•Lack of coordinated research with established verification and optimization of
information interchange between researchers, forecasters and end-users.
•Limiting the transfer of nowcasting capabilities to the operational sphere.
•Improvement in nowcasting services has significant social and economic benefits.
•WWRP Role?.
•Include research into basic understanding, defining of limits of predictability, a
collaborative and systematic approach to the nowcasting process, interaction with the
user base, and relevant social impact studies.
The objectives of the proposed WWRP Nowcasting Working Group
(NWG) are:
•To advance the science of nowcasting, including meteorology
processes and predictability.
• To promote, and aid the implementation of nowcasting in the
WWRP framework and within National Meteorological Services
and their end users.
11 June 2004
WWRP/NWG-COST 722
Joint Session
Extrapolation
Convergence line
Detection & characterization
Statistical
Data Fusion
System
Radar retrievals
Model derived forecast
fields
Nowcast
NWP
Forecast Rules
11 June 2004
Forecast Input
rules
Forecaster
e.q
collision
e.q.boundary
convergence
line
storm
likely
Input, initiation
meteorological
situation
WWRP/NWG-COST 722
Joint Session
Courtesy, Jim Wilson NCAR
Important nowcasting applications and end uses include:
•Transport weather including aviation, rail and road
•Emergency Services including Law and Order Agencies
•Hydrology
•Agriculture industry
• Construction Industry
• Recreational Activities
•Power industry
•Local authorities monitoring air pollution
• Air quality and particle dispersion
•Homeland security
11 June 2004
WWRP/NWG-COST 722
Joint Session
Aviation Impact of Convective Weather
Courtesy, Jim
Wilson NCAR
11 June 2004
WWRP/NWG-COST 722
Joint Session
Stages of disaster prevention activities and relevant
weather information
Met. Information
SRF(NWP)
>Outlook
Local Government
Stand by
Citizen
Keep in mind
1-2 days
before
Precip.
Intensity
11 June 2004
VSRF
Nowcast
(Anal+NWP)
(Anal)
>caution
>warning
Ready to
take action
preparation
3-6 hours
before
Action for
disaster
prevention
evacuation
Cancel
warning
Action for
recovery
Back to normal
(recovery)
1 hour
before
Is the skill of our
disaster
forecast fulfilling
(Flood/Landslide) their needs?
WWRP/NWG-COST 722
Joint Session
Courtesy, Shingo Yamada JMA
•Promote nowcasting science (RDP), transfer of technology and science
through Forecast Demonstration Projects (FDP) and assist in the
development of nowcast test beds.
•Organization of international nowcasting conferences and specialist
symposia, training workshops and collaboration with WMO and CBS
•Inter comparison and systematic testing of these algorithms on common and
representative databases (including consideration of extreme and high
impact events) would provide valuable assessment and help define optimal
paths for development.
•A tendency exists within National Meteorological Services (NMS’s) to
develop nowcasting capabilities in isolation
11 June 2004
WWRP/NWG-COST 722
Joint Session
Symposia
•No major international conference with a specific nowcasting focus since
the 1984 Nowcasting 2 Symposium in Norrkoping, Sweden.
•Nowcasting science and applications are scattered in sessions of related
theme conferences
•Initial major conference aimed at
(1) a review of the science and recommendations on the way to advance
the science, and
(2) plans for a series of symposia concentrating on the specific science
and system issues and the way they can be incorporated into RDP, FDP
and testbed activities.
•Possible Technical Symposia
•Analysis and conceptual models
•Nowcasting algorithms
•Blending NWP with other techniques
•Consistent time integration of nowcast and other forecast products
•Quality assurance for nowcasting data
•Alerting, nowcast production systems, dissemination processes and
systems and end user interface
•Man-machine interfaces and the role of a meteorologist in the
nowcasting process
•Verification and validation
11 June 2004
WWRP/NWG-COST 722
Joint Session
Nowcasting RDP
Potential Science Issues for RDPs
•Use of high resolution NWP models for nowcasting
•Value adding to NWP (analysis/forecasts) for nowcasting, including merging techniques
for combining extrapolation and NWP forecasts.
•Limits of nowcasting predictability
•Nowcasting extreme events (rain, wind, lightning….)
•Initiation, maintenance and prediction of convection. Heavy precipitation, monsoon
rainfall, nocturnal convection
•Microphysical and dynamical processes important in low cloud, fog, icing, and mixed
phase precipitation
•Development and propagation of snow bands
•Air quality and dispersion modelling
11 June 2004
WWRP/NWG-COST 722
Joint Session
Potential System and Forecast Process Issues for RDPs
•Integrated observing system development with observational error
characterization
•Representation of weather in objects. To define characteristics of weather
elements and as a basis for the exchange of weather information. Examples are
the representation of information from cell detection, tracking, growth and
decay algorithms, and the exchange of such information with other applications,
and NWP e.g., SIGOONS(Sig. Wea. Object-Oriented Nowcasting System)/OPIC
(Object for Con. Nowcasting) approaches used at Meteo France
•Heuristic methods, such as statistical models, conceptual models, including
data mining and artificial intelligence and fuzzy logic methods.
•Human interfaces which display and enable interaction with information
observations, analyzed fields, extracted objects and NWP outputs. Required in
a unified manner that allows timely implementation of man-machine procedures
for field and forecast modification.
•Information integration from multiple sources e.g., examination and
intercomparison of techniques used in AWIPS/GFE (USA), SYNERGIE
(France), and NINJO systems
•End-to-End
11 June 2004
WWRP/NWG-COST 722
Joint Session
The European COST projects provide opportunities for developing
RDPs
•COST Action 717 is examining the use of radar observations in
hydrological and NWP models. A related activity is EU Framework V
Project CARPE DIEM.
•COST Action 722 is concentrating on fog and low cloud
forecasting(physical and statistical) with verification.
•COST Action 728 will address measuring and forecasting atmospheric
icing on structures
•US nocturnal convection is a major forecasting problem that has
received little attention.
It is possible that this could be developed into a WWRP RDP as part
of USWRP efforts.
11 June 2004
WWRP/NWG-COST 722
Joint Session
Nowcasting FDP
FDP’s focused on the following seem possible: •Convective weather. This was the focus for S2000 but potential also
exists in the USA with NASA/NWS SPORT project.
• Hydrology is considered another potential target for an FDP. An FDP
should be planned over a specified river basin to demonstrate the river
management organizations benefits to use precipitation nowcasts in the
course of flood control and water resource management.
• Air quality monitoring and prediction. Would make use of latest
observational techniques, NWP and dispersion models.
• Aviation weather with emphasis on terminal area applications. Specific
user with well defined costs structure related to convection, wind shifts,
visibility, fog and low cloud, winter Icing and snow forecasting.
• Road weather
• Satellite and other non-radar based techniques for convective
nowcasting systems
11 June 2004
WWRP/NWG-COST 722
Joint Session
The Beijing 2008 Olympic Games is presently the best candidate for a
new FDP. It will be the first FDP with formal involvement of the NWG.
Other opportunities that should be explored are:
•Flash flood awareness and civil design in association with the RISKEOS (ESA R&D framework) project . Includes users, France and Italy.
•NWC SAF (Eumetsat) system. To demonstrate utility of Meteosat
Second Generation (MSG) satellite based nowcasting projects. e.g., but
RDT automatic monitoring of convection form MSG (tracking and
characterisation) extended to include end-users
11 June 2004
WWRP/NWG-COST 722
Joint Session
Test beds
•Test beds aimed at accelerating science and system issues in
an end-to-end setting involving developers, nowcasters and
end-users.
•The WWRP test bed is essentially an international focus for
activities encompassing both RDP and FDP philosophies.
• Given the range of nowcast issues and varying climatic
regimes it is expected test beds will have a regional or national
focus but with a theme of international significance
•Test beds should make use of existing opportunities in
various countries that have a particular focus, interest and
infrastructure to support long-term research and operational
testing on nowcasting techniques
•High quality observational networks covering the areas where
the meteorological phenomena occur with high probability
and established end-users
11 June 2004
WWRP/NWG-COST 722
Joint Session
Candidate test bed sites include:
A. Huntsville Alabama (NWS/NASA SPORT) for convective weather;
B. Linking with the short range precipitation prediction initiative of the
NWS HPC
C. Toronto and Montreal (EC/McGill) framework for convective and
winter weather, including WSDDM and AVISA/AIRS for icing;
D. Brisbane, Australia (Bureau of Meteorology) for subtropical
convection and hydrological activities.
11 June 2004
WWRP/NWG-COST 722
Joint Session
North Alabama Nowcasting
Short-term Forecasting Test Bed
11 June 2004
WWRP/NWG-COST 722
Joint Session
Courtesy, S. Goodman, NASA
11 June 2004
WWRP/NWG-COST 722
Joint Session
Information exchange
•NWG should consider publishing a guide "Global Guide to
Nowcasting" which will be a useful reference to nowcasters/system
developers worldwide.
•Website which facilitates exchange of knowledge and experience on
nowcasting among NMSs
Nowcasting Fellowship
•WMO/ESCAPE Typhoon Committee has established a Research fellowship
scheme to enable scientists of various NMS’s to visit other groups to undertake
tropical cyclone research.
•Establishment of a similar nowcasting research fellowship focused on young
scientists
11 June 2004
WWRP/NWG-COST 722
Joint Session
Work Plan and Timetable
Establish NWG -2004.
Conferences and Workshops
1. Brazil 2003 Training Workshop
2. Nowcasting Science and Technical Conference (1st half 2005)
3. System Developer Workshop (2005)?
4. Asian 2006 Workshop?
WWRP NWG Web Site (2004)?
Develop Test Bed /RDP Facility (2005-2006)
11 June 2004
WWRP/NWG-COST 722
Joint Session