Aerosol Data Assimilation Aerosol Data Assimilation with Lidar Observations with Lidar Observations and Ensemble Kalman and Ensemble Kalman Filter Filter T. Thomas Sekiyama (MRI/JMA, Japan) T. Y. Tanaka (MRI/JMA, Japan) A. Shimizu (NIES, Japan) T. Miyoshi (Univ. of Maryland, US) The Second GALION Workshop 22 September 2010, Geneva, Switzerland
18
Embed
Aerosol Data Assimilation with Lidar Observations and Ensemble Kalman Filter
Aerosol Data Assimilation with Lidar Observations and Ensemble Kalman Filter. T. Thomas Sekiyama ( MRI/JMA, Japan ) T. Y. Tanaka( MRI/JMA, Japan ) A. Shimizu( NIES, Japan ) T. Miyoshi( Univ. of Maryland, US ). The Second GALION Workshop 22 September 2010, Geneva, Switzerland. Agenda. - PowerPoint PPT Presentation
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Aerosol Data Assimilation Aerosol Data Assimilation with Lidar Observations with Lidar Observations
and Ensemble Kalman Filterand Ensemble Kalman FilterT. Thomas Sekiyama (MRI/JMA, Japan)T. Y. Tanaka (MRI/JMA, Japan)A. Shimizu (NIES, Japan)T. Miyoshi (Univ. of Maryland, US)
The Second GALION Workshop22 September 2010, Geneva, Switzerland
2
AgendaAgenda
• Introduction
• Data Assimilation System– Global Aerosol Model (MASINGAR)
– 4-Dimensional Ensemble Kalman Filter (4D-EnKF)
– Observational Data #1 (CALIPSO/CALIOP)
– Observational Data #2 (Asian Dust Network, AD-Net)
• Results on Asian Dust– Comparison with Independent Observations
– Comparison between CALIOP exp. and AD-Net exp.
• Summary and Future Work
3
IntroductionIntroduction
• Aerosol observation:Available data are limited or very sparse spatio-temporally! ( weather obs.)
• Model simulation:It’s useful, but not real! ( virtual reality)
• Data assimilation:It’s a fusion of observation and simulation with highly informative techniques to extract hidden information from data on hand.
4
Global Aerosol Model (MASINGAR)Global Aerosol Model (MASINGAR)
• The Model of Aerosol Species in the Global Atmosphere (MASINGAR) was developed by MRI/JMA.
• MASINGAR simulates dust (partitioned into 10-size bins), seasalt, and sulfate aerosols with a resolution of 2.8º by 2.8º.
• The meteorological field is assimilated with the JMA reanalysis (6-hourly).
• JMA is using MASINGAR to forecast Asian dust storms operationally.
8 stations of the NIES Asian Dust Network – only in East Asia– but, temporally
dense– aerosol extinction
coefficients (provided by the NIES team) are used.
Lidar data of 8 stations (indicated green circles) were used for this data-assimilation experiment.
12
AD-Net Results (comparison with CALIOP results)AD-Net Results (comparison with CALIOP results)
532nm extinction coefficients for non-spherical particles ( dust aerosol).
The X-axis shows date in April 2007.
(a) AD-Net ground-based lidar observation;
(b) free model-run result without assimilation;
(c) data assimilation result with CALIPSO data,
(d) with AD-Net lidar data.
13
Instrument Location of Toyama stationInstrument Location of Toyama station
14
532nm extinction coefficients for spherical particles ( dust excluded).
The X-axis shows date in April 2007.
(a) NIES ground-based lidar observation;
(b) free model-run result without assimilation;
(c) data assimilation result with CALIPSO data,
(d) with NIES lidar data.
AD-Net Results (comparison with CALIOP results)AD-Net Results (comparison with CALIOP results)
15
AOT on 24Apr2007: data assimilation resultsAOT on 24Apr2007: data assimilation results
16
SummarySummary
• CALIOP assimilation results were validated by independent dust observations in East Asia: ground-based lidars and weather reports of aeolian dust events.
• The assimilation system was successfully performed with CALIOP aerosol observations in springtime 2007.
• This assimilation system can potentially provide global aerosol reanalyses for various particle types and sizes.
17
SummarySummary
18
Future WorkFuture Work
• Dust forecasting– To apply this data assimilation system to the
operational dust prediction service of JMA. – The 4D-EnKF with lidar data makes it possible to
supply the initial conditions for aerosol forecasting.
• Predictability in the Chaotic system– Data assimilation with lidar data can
provide plenty of information to explore the scientific frontier.