Atmospheric Sciences [A]

A53C
 MC:Hall D  Friday  1340h

Year of Tropical Convection (YOTC): High-Resolution Modeling, in Situ Data, and State-of-the-Art Satellite Observations to Address the Challenge of Multiscale Organized Convection III Posters


Presiding:  D Waliser, Jet Propulsion Laboratory/Caltech; J Done, NCAR

A53C-0279

Horizontal and vertical structure of easterly waves across Central America and Mexico

Serra, Y serra@atmo.arizona.edu, Institute of Atmospheric Physics, University of Arizona 1118 E 4th St, P.O. Box 210081, Tucson, AZ 85721-0081, United States
* Kiladis, G george.kiladis@noaa.gov, NOAA/Physical Science Division, R/PSD1 325 Broadway, Boulder, CO 80305-3328, United States

Tropical easterly waves are most commonly associated with Atlantic and east Pacific hurricanes, whose origins the National Hurricane Center attributes primarily to easterly waves propagating off the coast of Africa. However, in addition to hurricanes, easterly waves also influence the summertime precipitation over the land areas surrounding the Caribbean Sea and Gulf of Mexico. There is also evidence to suggest that they influence precipitation patterns throughout the North American Monsoon region, including northern Mexico and the Southwest US. Thus, understanding the evolution of these waves as they pass over Central America and Mexico is critical for modeling summertime climate in many heavily populated areas and agricultural communities. In this study reanalyses are used to examine the coherence of the large-scale wave signature in vorticity, winds and streamfunction across this region. The temperature and moisture signals associated with the waves are also examined. These results are compared with local wave signals in winds, temperature and moisture observed at historic radiosonde locations throughout the region and archived at the Integrated Global Radiosonde Archive (IGRA). Regression analyses suggest that easterly waves reduce in horizontal scale as they enter the Caribbean Sea/Gulf of Mexico region. Twin vortices form on either side of the mountains of Central America and Mexico of even smaller scale and propagate northwestward along both the east and west coasts, while the larger scale wave pattern reestablishes itself and propagates westward along approximately 10°N into the east Pacific. IGRA sounding data support these results observed in the reanalyses. Strong convective signals remain coupled to the vortices that spin off of the wave troughs and are likely the mechanism by which easterly waves influence precipitation patterns throughout Mexico and into the Southwest US. Future work will explore the energetics and precipitation features of the cutoff vortices over land using a regional model.

A53C-0280

Online Visualization and Analysis of Merged Global Geostationary Satellite Infrared Dataset

Liu, Z Zhong.Liu-1@nasa.gov, CEOSR, George Mason University 4400 University Dr., Fairfax, VA 22030, United States
Liu, Z Zhong.Liu-1@nasa.gov, NASA Goddard Earth Sciences Data and Information Services Center, NASA/GSFC, Code 610.2, Greenbelt, MD 20771, United States
Ostrenga, D Dana.Ostrenga@nasa.gov, ADNET Systems, Inc., 164 Rollins Avenue Suite 303, Rockville, MD 20852, United States
Ostrenga, D Dana.Ostrenga@nasa.gov, NASA Goddard Earth Sciences Data and Information Services Center, NASA/GSFC, Code 610.2, Greenbelt, MD 20771, United States
* Leptoukh, G gregory.leptoukh@nasa.gov, NASA Goddard Earth Sciences Data and Information Services Center, NASA/GSFC, Code 610.2, Greenbelt, MD 20771, United States
Mehta, A amita.v.mehta@nasa.gov, JCET/UMBC, Suite 320 5523 Research Park Drive, Baltimore,, MD 21228, United States
Mehta, A amita.v.mehta@nasa.gov, NASA Goddard Earth Sciences Data and Information Services Center, NASA/GSFC, Code 610.2, Greenbelt, MD 20771, United States

The NASA Goddard Earth Sciences Data Information Services Center (GES DISC) is home of Tropical Rainfall Measuring Mission (TRMM) data archive. The global merged IR product, also known as, the NCEP/CPC 4-km Global (60°N - 60°S) IR Dataset, is one of TRMM ancillary datasets. They are globally-merged (60°N-60°S) pixel-resolution (4 km) IR brightness temperature data (equivalent blackbody temperatures), merged from all available geostationary satellites (GOES-8/10, METEOSAT-7/5 & GMS). The availability of data from METEOSAT-5, which is located at 63E at the present time, yields a unique opportunity for total global (60°N-60°S) coverage. The GES DISC has collected over 8 years of the data beginning from February of 2000. This high temporal resolution dataset can not only provide additional background information to TRMM and other satellite missions, but also allow observing a wide range of meteorological phenomena from space, such as, mesoscale convection system, tropical cyclones, hurricanes, etc. The dataset can also be used to verify model simulations. Despite that the data can be downloaded via ftp, however, its large volume poses a challenge for many users. A single file occupies about 70 MB disk space and there is a total of ~73,000 files (~4.5 TB) for the past 8 years. Because there is a lack of data subsetting service, one has to download the entire file, which could be time consuming and require a lot of disk space. In order to facilitate data access, we have developed a web prototype, the Global Image ViewER (GIVER), to allow users to conduct online visualization and analysis of this dataset. With a web browser and few mouse clicks, users can have a full access to over 8 year and over 4.5 TB data and generate black and white IR imagery and animation without downloading any software and data. Basic functions include selection of area of interest, single imagery or animation, a time skip capability for different temporal resolution and image size. Users can save an animation as a file (animated gif) and import it in other presentation software, such as, Microsoft PowerPoint. These capabilities along with examples will be presented in this poster. The prototype will be integrated into GIOVANNI and existing GIOVANNI capabilities, such as, data download, Google Earth KMZ, etc. will be available. Users will also be able to access other data products in the GIOVANNI family.

http://disc.gsfc.nasa.gov/hurricane/index.html

A53C-0281

Mesoscale Convective Systems Associated With 7 African Easterly Waves During NAMMA

* Zawislak, J jon.zawislak@utah.edu, University of Utah, Department of Meteorology 135 South 1460 East Rm. 819, Salt Lake City, UT 84112, United States
Zipser, E ed.zipser@utah.edu, University of Utah, Department of Meteorology 135 South 1460 East Rm. 819, Salt Lake City, UT 84112, United States

The 2006 NAMMA (NASA African Monsoon Multidisciplinary Analyses) Experiment was staged from the Cape Verde Islands during August and September 2006. The primary goals of the experiment were to distinguish the essential differences between developing and non-developing African easterly waves, and the effects of the Saharan Air Layer (SAL) on cloud microphysics and the waves. During the campaign, 7 easterly waves passed through the NAMMA domain. 2 waves became tropical cyclones near the west coast of Africa (Debby and Helene), 2 waves did not develop, while the role of the other 3 waves in the cyclogenesis of Ernesto, Florence and Gordon is not clear. The goal of this study is to 1) track mesoscale convective systems near small-scale vorticity maxima accompanying each large-scale wave, and 2) quantify the associated convective and rainfall characteristics, and compare those characteristics of the developing versus the non-developing waves. These quantities include cloud top height, area of the cold cloud, radar reflectivity profile where available, minimum 37 and 85 GHz brightness temperature, and area covered by intense ice scattering at these frequencies. 3-hourly instantaneous rain rates derived in the TRMM 3B42 algorithm are utilized to produce evolutions of the volumetric rainfall associated with each of the vorticity maxima.

A53C-0282

Mass Divergence, Temperature and RH Anomalies in Regions of Enhanced Precipitation: Observations vs. GCMs

* Mitovski, T tn648214@dal.ca, Department of Physics and Atmospheric Science, Dalhousie University, Halifax, NS B3H 3J5, Canada
Folkins, I Ian.Folkins@dal.ca, Department of Physics and Atmospheric Science, Dalhousie University, Halifax, NS B3H 3J5, Canada

The purpose of our research is to compare diagnostics of modeled and observed vertical mass transport. The diagnostics are: dynamical (mass) divergence, temperature anomalies and RH anomaly regression in the regions of enhanced precipitation. The mass divergence provides an insight into the vertical mass transport. Here we are comparing the mass divergence estimated for 7 rings of stations for the rainy season to the same estimated from the third generation coupled global climate model (CGCM3-T63) and from the Geophysical Fluid Dynamics Laboratory Climate Model Version 2.1 (GFDL CM2.1) outputs. The second diagnostic comes from comparing observed to GCMs low level temperature anomalies. It is believed that the temperature anomalies are a result of mesoscale activity in the regions of enhanced precipitation [Folkins et al., 2007]. The low level cooling, a result of the stratiform heating mode [Mapes and Houze, 1995], is important for the excitation of small-scale gravity waves. The small-scale gravity waves contribute to the 'gregariousness' of deep convection by increasing the buoyancy of the neighbouring shallow cumuli [Mapes and Houze, 1993] and, consequently, the small-scale gravity waves create a positive feedback between existing deep convection and newborn shallow convective clouds. The last diagnostic is expressed through RH anomaly regression. The RH anomaly regressions are estimated for two days before and two days after maximum precipitation events from radiosondes and results are compared to regressions estimated from CGCM3 3-hourly output. Two distinct features are seen on the RH regression plot: growing cumuli clouds before the main event and a stratiform anvil after. In addition, there is also a 'pool' of dry mid-tropospheric air just after the maximum precipitation event which might be associated to mesoscale downdrafts.

A53C-0283 TI: Mesoscale Characteristics of Mei-Yu Precipitation Systems over South China, Taiwan and the South China Sea
* xu, w weixin.xu@utah.edu, Tropical Convection Group Department of Meteorology Univ. of Utah, 135S 1460E RM819, SLC, UT 84108, United States
Zipser, E edzipser@utah.edu, Tropical Convection Group Department of Meteorology Univ. of Utah, 135S 1460E RM819, SLC, UT 84108, United States
Liu, C liu.c.t@utah.edu, Tropical Convection Group Department of Meteorology Univ. of Utah, 135S 1460E RM819, SLC, UT 84108, United States

After the onset of the Asian summer monsoon in early to mid May over the South China Sea, a quasi- stationary front (called Mei-Yu in Taiwan and Pre-summer rainy season by Chinese meteorologists) occurs frequently and repeatedly over South China (SC), Taiwan and South China Sea (SCS). During this period, heavy rainfall and flash floods are often produced by the slow moving Mei-Yu frontal rainband, especially by the embedded active Mesoscale Convective Systems (MCSs). This work summarizes the mesoscale characteristics of precipitation systems within defined Mei-Yu rainbands over SC, SCS and Taiwan, based on a 10-year database of precipitation feature (PFs). The 3 hourly rainfall product derived from the Tropical Rainfall Measurement Mission (TRMM) is utilized to explore the definition, position and rainfall contribution of Mei-Yu rainbands in May-June from 1998 to 2008. A Mei-Yu rainband is defined by the criteria of: (1) a well-defined daily accumulated rainfall band, (2) lifetime > 3 days, (3) length > 10 degree longitude, and (4) a rainfall center > 50mm. Most of the rainbands happen after May 10th, with the largest occurrence center located along south China Coast. The distribution pattern of rainfall contributed by well-defined Mei-Yu rainbands during May11 and June24th dominates that of total rainfall in the period. Smooth distribution of Mei-Yu fronts doesn't correspond to smooth rainfall center: bulk eyes show up in the smooth Mei-Yu rainband center. Further studies show that more than 60 percent of total rainfall is contributed by Mei-Yu rainfall. The properties of PFs within the defined Mei-Yu rainband are analyzed using TRMM PF database. Distribution of PFs with different properties such as maximum height of 40 dBZ, minimum PCT of 85 GHz, radar area and volume, and lighting flash rate is mapped. This can indicate specific regions where Mei-Yu related active MCSs and strong thunderstorms tend to be. Vertical structure of the MCSs within the Mei-Yu rainband will be compared to that of the MCSs occur during the whole monsoon season. Relationship between ice scattering and lightening related to the Mei-Yu systems will also be explored.

A53C-0284

Observations of the Double Intertropical Convergence Zone using CloudSat

* Quantz, H M quantz@cira.colostate.edu, Colorado State University/CIRA, 1375 Campus Delivery, Fort Collins, CO 80523, United States
Vonder Haar, T H vonderhaar@cira.colostate.edu, Colorado State University/CIRA, 1375 Campus Delivery, Fort Collins, CO 80523, United States

Recent analysis of the most current coupled global climate models (CGCMs) used for the climate predictions in the IPCC AR4 suggests that the tropical mean climate is poorly simulated, particularly in the case of the Intertropical Convergence Zone (ITCZ). As shown in observational studies, the ITCZ in the Eastern Pacific splits and forms two distinct bands situated quasi-symmetrically about the equator for roughly two months during the boreal spring. However, most of the models examined had some degree of the double ITCZ (DITCZ) problem, which is characterized by the overproduction of DITCZ events during the simulation. This results in the production of excessive precipitation over much of the tropics in the models, often accompanied by insufficient precipitation along the equator. The Cloud Profiling Radar aboard the CloudSat satellite allows for vertical profiling of the clouds in the DITCZ, offering a novel look at the vertical cloud structure. This new data can be added to previous knowledge of sea surface temperature (SST) and wind data to provide a more complete picture of the poorly understood DITCZ phenomenon. In addition, vertical cloud observations will contribute to an improved understanding of the discrepancy between the observations and the model simulations in this area.

A53C-0285

Convective signals from surface measurements at DOE ARM Tropical Western Pacific sites

* Wang, Y yi.wang@pnl.gov, Pacific Northwest National Laboratory, 902 Battelle Boulevard P.O. Box 999, MSIN K9-24, Richland, WA 99354, United States
Long, C N chuck.long@pnl.gov, Pacific Northwest National Laboratory, 902 Battelle Boulevard P.O. Box 999, MSIN K9-24, Richland, WA 99354, United States
Mather, J H jim.mather@pnl.gov, Pacific Northwest National Laboratory, 902 Battelle Boulevard P.O. Box 999, MSIN K9-24, Richland, WA 99354, United States

Madden-Julian Oscillation (MJO) signals have been detected using highly sampled observations (1-minute resolution) from the U.S. Department of Energy (DOE) Atmospheric Radiation Measurement (ARM) Climate Research Facility (ACRF) Tropical Western Pacific (TWP) Manus and Nauru sites. With downwelling shortwave radiative fluxes and derived shortwave fractional sky cover, and the statistical tools of wavelet, cross wavelet, and spectrum power, we report finding major convective signals and their phase change from surface observations spanning the periods from 1996 to 2006 for Manus and from 1998 to 2006 for Nauru. Our findings are confirmed with the satellite-retrieved values of precipitation from the Global Precipitation Climatology Project (GPCP) and interpolated outgoing longwave radiation (OLR) satellite measurements from the National Oceanic and Atmospheric Administration (NOAA) for the same location and period, though these products inherently represent large spatial scale. During the 1997-98 strong El Niño years, the MJO signals over both sites are weak. This is consistent with our current understanding. Overall, the MJO signal is much stronger and more persistent over Manus than over Nauru. We further composite 21 MJO events spanning the extended boreal winter period (November through the following April) from 1996 to 2006 for the Manus site, chosen using the NOAA Climate Prediction Center's MJO index over 140E. We find a strong modulation of the diurnal cycle by the MJO over Manus. In particular, our major findings are: 1) Daily average cloudiness is in phase with the MJO peak while the diurnal amplitude is out of phase; 2) The peak MJO phase is associated with enhanced early morning and suppressed late afternoon deep convection; 3) The convective precipitation lags the enhanced cloudiness in the morning by 3-6 hours; 4) Strong pre- conditioning of low and mid-level cloudiness is detected at about a 5-day lead of the MJO peak; and 5) The surface westerly wind burst (WWB) is strongly modulated by the MJO phases with afternoon maximums at peak MJO phases.

http://www.arm.gov/sites/twp.stm

A53C-0286

Diurnal Cycle Variability of Rainfall Over the Indian Region: Perspectives From the TRMM Satellite

* Sahany, S sandeep@caos.iisc.ernet.in, Centre for Atmospheric and Oceanic Sciences, Indian Institute of Science, Bangalore, KTK 560012, India
Venugopal, V venu@caos.iisc.ernet.in
Nanjundiah, R S ravi@caos.iisc.ernet.in

Using the TRMM 3-hourly, 0.25x0.25 degree 3B42 rainfall product for nine years (1999-2007), we characterise the summer season (JJAS) diurnal cycle of rainfall over the Indian land and its neighbouring oceans (10S to 35N, 60E to 100E). Most previous studies have provided an analysis of a single or few years of satellite- or station-based rainfall data (e.g., Basu, 2007; Yang and Smith, 2006; Nesbitt and Zipser, 2003) and, to our knowledge, this is one of the first studies that aims to exhaustively characterise the diurnal scale statistical characteristics of rainfall over the Indian and surrounding regions. Using harmonic analysis, we extract, at each grid point, every year, the signal corresponding to time periods smaller than 1 day, i.e., the signal that relates to diurnal and sub-diurnal variability. Subsequently, the time of rainfall peak for this filtered signal, referred to as the "peak hour," is estimated, with care taken to eliminate spurious peaks arising out of Gibbs oscillations. Our analysis suggests that the mode of the peak hour (of the diurnal-scale rainfall) over a significant part of Indian land is at 12 UTC (i.e. 5:30PM local time), a finding similar to that reported in previous studies (e.g., Liu and Zipser, 2008; Krishnamurti and Kishtawal, 2000). The Himalayan foothills were found to have a mode of peak hour at 21 UTC (i.e. 2:30AM local time), whereas over the Burmese mountains the rainfall peaks at 9 to 12 UTC (i.e. 3:30 PM to 6:30 PM local time). In addition, over the Bay of Bengal, there is a stratified spatial structure of mode of the peak hour of diurnal rainfall at 6, 9 and 12 UTC from North central to South Bay. This finding, not reported before, could be seen to be consistent with southward propagation of the diurnal rainfall pattern (e.g., Hoyos and Webster, 2007; Zuidema, 2003). We also find that the Arabian sea (to the east of 65E and north of the Equator, along a region where it rains for more than 50% of the time) shows a peak hour around 9 to 12 UTC. Finally, most of the neighbouring oceans exhibit a noon to early evening peak, unlike the early morning peak reported by some of the earlier studies (e.g., Lau et. al., 2007; Bowman et. al., 2005).

A53C-0287

Using high resolution TRMM observations to understand the role of topography in spatial variations in precipitation and storm structure

* Nesbitt, S W snesbitt@illinois.edu, Department of Atmospheric Sciences, University of Illinois, 105 S Gregory St, Urbana, IL 61801, United States
Anders, A M amanders@illinois.edu, Department of Geology, University of Illinois, 1301 W Green St, Urbana, IL 61801, United States

TRMM has provided an unprecedented view of the spatial and temporal variation of precipitation processes since its launch in 1997. It has also improved our understanding of the extent of convective organization in the tropics, convective organization's role in modulating the rainfall diurnal cycle, as well as the impacts of organized convection in the tropical circulation. With nearly 11 years of data now in hand, sampling uncertainties have been reduced such that small spatial scale variations of precipitation characteristics and convective organization can be observed using the TRMM data record. We will present the 10-year high- resolution precipitation climatology from the TRMM precipitation radar (PR) and TRMM microwave imager (TMI) at resolutions as high as the 5 km scale. We will examine climatological high-resolution precipitation variations in rainfall frequency, accumulation, and vertical structure along topographic features to show (1) the extreme horizontal variations in precipitation and latent heating associated with tropical orographic precipitation, often with impacts on the monsoon systems of which it is manifested, and (2) examine the interplay between dynamic forcing for orogenic convective systems, resulting convective structures, latent heating, and the resulting diurnal cycle of convective systems. As a contribution to the Year of Tropical Convection (YOTC), the 10 year TRMM climatology dataset will be presented as a new benchmark precipitation climatology to not only examine the representation of the spatial distribution of precipitation near topography, but also as the basis for evaluation of improved representation of precipitation physics near topography in cloud-resolving to climate models.

A53C-0288

The Challenges of Interpreting Microwave-Sounded Height- Registered Water near Tropical Cyclones

* Fishbein, E Evan.Fishbein@jpl.nasa.gov, Jet Propulsion Laboratory, California Institute of Technology, 4800 Oak Grove Dr,, Pasadena, CA 91109-8099, United States
Fetzer, E Eric.Fetzer@jpl.nasa.gov, Jet Propulsion Laboratory, California Institute of Technology, 4800 Oak Grove Dr,, Pasadena, CA 91109-8099, United States
Kahn, B Brian.Kahn@jpl.nasa.giv, Jet Propulsion Laboratory, California Institute of Technology, 4800 Oak Grove Dr,, Pasadena, CA 91109-8099, United States
Lambrightsen, B Bjorn.Lambrightsen@jpl.nasa.gov, Jet Propulsion Laboratory, California Institute of Technology, 4800 Oak Grove Dr,, Pasadena, CA 91109-8099, United States
Teixeira, J Joao.Teixeira@jpl.nasa.gov, Jet Propulsion Laboratory, California Institute of Technology, 4800 Oak Grove Dr,, Pasadena, CA 91109-8099, United States

Satellite-derived coincident profiles of water, and liquid water (cloud properties) provide some of the best constraints on the energetics of tropical cyclones. The problem addressed is how accuracy depends on amount and distribution of cloud liquid water. We focus on products derived from the AMSU/MHS sounders using algorithms from the AIRS/AMSU/HSB investigations. We use error analysis and inter omparisons with total precipitable and liquid water from AMSR-E and MODIS. Accuracy degrades as cloud liquid water amount increases and the vertical distribution of water vapor depends on climatological constraints on either partitioning between phases or vertical correlation.

A53C-0289

"Electrically-Hot" Convection and Tropical Cyclone Development in the Eastern Atlantic

* Leppert, K leppert@nsstc.uah.edu, Department of Atmospheric Science, University of Alabama Huntsville, Huntsville, AL 35899, United States
Petersen, W A walt.petersen@nasa.gov, NASA, NASA-MSFC/NSSTC, Huntsville, AL 35805, United States
Williams, E earlew@ll.mit.edu, Massachusetts Institute of Technology, Massachusetts Institute of Technology, Cambridge, MA 02139, United States

The depth and intensity of convective-scale "hot" towers in intensifying tropical disturbances has been hypothesized to play a role in tropical cyclogenesis via dynamic and thermodynamic feedbacks on the larger meso-to-synoptic scale circulation. In this investigation we investigate the role that widespread and/or intense lightning-producing convection (i.e., "electrically-hot towers") resident in African Easterly Waves (AEW) may play in tropical cyclogenesis over the eastern Atlantic Ocean. NCEP reanalysis data for the months of July to November for the years 2004, 2006, and 2007 are analyzed for the domain of 5° N - 15° N and 50° W - 30° E. Specifically, NCEP data for individual AEWs are partitioned into northerly, southerly, trough, and ridge phases using the 700 hPa meridional winds. Subsequently, information from National Hurricane Center storm reports were divided up into developing and non-developing waves (i.e. tropical cyclogenesis). Finally, composites were created of developing and non- developing waves using the NCEP variables, but with the inclusion of lightning flash count and infrared brightness temperature information. The Zeus and World Wide Lightning Location Network lightning data were used for the lightning information, and the IR brightness temperature data was extracted from the NASA global-merged infrared brightness temperature dataset. Results indicate that developing AEW composites have greater low-level positive vorticity (9.0E-06 s-1 vs. 4.5E-06 s-1), slightly greater upward vertical motion (-0.035 Pascals s-1 vs. -0.028 Pascals s-1), slightly higher upper-level divergence(2.5E-06 s-1 vs. 1.8E-06 s-1), a higher mid-level (i.e. 600 hPa) moisture anomaly (0.6 g kg-1 vs. -0.15 g kg-1), cooler average brightness temperatures (273.5K vs. 278.4K), and more lightning strikes (1018 vs. 641 strikes) when compared against the non-developing composites. These results collectively indicate that AEWs producing tropical cyclones may have increased convective activity- and in particular, convective activity producing more lightning. To complement the composite work, we are further investigating the degree to which the aforementioned behavior is observed in individual cases. Work is also proceeding to identify whether the higher lightning strike/cooler brightness temperature behavior in the AEW tropical cyclone-developing composites is the result of more widespread convection producing a larger frequency of lightning, or just a few more vertically developed and electrically-intense convective towers.

A53C-0290

Temporal Relations of Column Water Vapor and Precipitation

* Holloway, C E c.e.holloway@reading.ac.uk, Dept. of Meteorology / NCAS-Climate, University of Reading Earley Gate, Reading, RG6 6BB, United Kingdom
Neelin, J neelin@atmos.ucla.edu, Dept. of Atmospheric and Oceanic Sci., UCLA 405 Hilgard Ave., Los Angeles, CA 90095, United States

Empirical studies using satellite data and radiosondes have shown that precipitation increases with column water vapor (CWV) in the tropics, and that this increase is much steeper above some critical CWV value. Here, eight years of 1-minute resolution microwave radiometer and optical gauge data at the Atmospheric Radiation Measurement (ARM) site on Nauru Island are analyzed to better understand the relationships between CWV, column liquid water (CLW), and precipitation at small time scales. CWV is found to have large autocorrelation times compared with CLW and precipitation. Before precipitation events, CWV increases on both a synoptic-scale time period and a subsequent shorter time period consistent with mesoscale convective activity---the latter period is associated with the highest CWV levels. Probabilities of precipitation increase greatly with CWV: 10--12 hr after high CWV, there are still significantly higher probabilities. Even in periods of high CWV, probabilities of initial precipitation in a 5-minute period remain low enough that there tends to be a lag before the start of the next precipitation event. This suggests that CWV can be a useful predictor for precipitation in stochastic convective parameterizations.

A53C-0291

Relationship between ENSO and northward propagating intraseasonal oscillation of East Asian summer monsoon and its interdecadal change

Ha, K kjha@pusan.ac.kr, Pusan National University, Division of Earth Environmental System College of Natural Science Pusan National University, Busan, 609-735, Korea, Republic of
* Yun, K kssh@pusan.ac.kr, Pusan National University, Division of Earth Environmental System College of Natural Science Pusan National University, Busan, 609-735, Korea, Republic of
Seo, K khseo@pusan.ac.kr, Pusan National University, Division of Earth Environmental System College of Natural Science Pusan National University, Busan, 609-735, Korea, Republic of

We investigate the interdecadal change in the relationship between El Nino/Southern Oscillation (ENSO) and the northward propagating intraseasonal oscillation (NPISO) in the East Asian summer monsoon (EASM) defined by Yun et al. [2008]. To find the long-term change in the relationship, top net long-wave radiation (i.e., outgoing long-wave radiation) obtained from ERA-40 for 44 year from 1958 to 2001 year are used. An empirical orthogonal functions (EOF) analysis of the 30-60 day bandpass filtered data is applied in the domain over EASM. The NPISO is identified using the first two leading EOF modes and its activity is estimated by the variance of their associated time series. Its interdecadal variability is detected by a change before and after late 1970s. After late 1970s, the lagged relationship between the ENSO and NPISO as a strong positive correlation is found during the late summer, while a weak positive correlation before late 1970s is shown during the early summer. It is seen that the interdecadal change is linked to a significant interdecadal change associated with the Indian Ocean and western North Pacific high (WNPH). It was approved with some results: After late 1970s, the correlation between the Indian Ocean SST anomalies and WNPH is significantly increased. The NPISO-WNPH relationship is also closely related after this period. Therefore, the Indian Ocean warming may contribute the interdecadal change in the relationship between ENSO and the NPISO after late 1970s.

A53C-0292

Global circulation and precipitation response to tropical heating in the Madden-Julian oscillation

* Han, S sdhan@pusan.ac.kr, Pusan National University, Jangjeon-dong, Busan, 609-735, Korea, Republic of
Seo, K khseo@pusan.ac.kr, Pusan National University, Jangjeon-dong, Busan, 609-735, Korea, Republic of

This study investigates the global circulation and precipitation response to tropical heating associated with the Madden-Julian oscillation (MJO) during the northern winter and summer. In order to study the causes of these circulation and precipitation anomalies, the GFDL dynamical core primitive equation model is used. Initialized with the observed three-dimensional climatological horizontal wind and temperature, the model is forced with latent heating representing deep convection, stratiform cloud forcing and shallow convection. The extratropical response similar to the observed emerges in two weeks of integrations and this is compared with the theory of Rossby wave dynamics. A series of sensitivity tests are performed and discussed to assess the roles of the following factors: (a) deep convective heating vs stratiform heating and shallow convective heating, (b) background baroclinicity (horizontal temperature gradient), (c) vertical wind shear, (d) interaction with higher-frequency transients, and (e) El Nino and La Nina basic state.

A53C-0293

Interannual Variability of the Onset and Withdrawal Dates of the Baiu Season

* Hashimoto, T tetsu@kauai.sci.kumamoto-u.ac.jp, Department of Environmental Sciences, Kumamoto University, Kurokami 2-39-1, Kumamoto, 860-8555, Japan
Tomita, T tomita@sci.kumamoto-u.ac.jp, Department of Environmental Sciences, Kumamoto University, Kurokami 2-39-1, Kumamoto, 860-8555, Japan

Baiu is a local name indicating the rainfall or the rainy season in early summer in Japan. Since the dates of onset and withdrawal are determined empirically by staff in the Japan Meteorological Agency, more objective determination is desired for scientific research, particularly, for the seasonal prediction of Baiu. This work proposes a new definition using the gradient of equivalent potential temperature (∇ θe) for the onset, withdrawal, and the period of Baiu. In addition, the physical mechanisms of their interannual variability are discussed through case studies. By the northward shift of region with large ∇ θe, the average date of the Baiu onset is determined around May 23 near Japan, and the interannual variability is characterized by the periodicity of two to three years. The earliest onset date was recoded in 1998, while the latest ones were in 1981, 1986 and 1987. On the other hand, the average date of the Baiu withdrawal was around July 17 with the earliest date in 1997 and the latest one in 1982. The interannual variability of the withdrawal dates is manifested by the 3-yr and interdecadal variations. The results of case studies clearly show that the large-scale pressure system around the Baiu front, i.e., the subtropical Pacific high, the Okhotsk high to the north, and the thermal low over the Asian continent collaboratively control the formation and northward shift of large ∇ θe. That is, the dates of onset and withdrawal are modified in such a pressure system. For instance, the early onset is forced by both of the Pacific and Okhotsk highs straddling the Baiu front. Further discussions will be introduced in December.

A53C-0294

The Role of Convective Moistening in the Formation and Progression of the MJO

* Thayer-Calder, K katetc@atmos.colostate.edu, Colorado State University, Department of Atmospheric Science, Fort Collins, CO 80525, United States
Randall, D A randall@atmos.colostate.edu, Colorado State University, Department of Atmospheric Science, Fort Collins, CO 80525, United States

This study compares two models which differ primarily in their convection parameterizations but produce extremely different MJO signals. The Community Atmosphere Model (CAM) version 3.0 from NCAR uses the Zhang and McFarlane (1995) scheme for deep convection and does not produce an MJO. The Super Parameterized version of the CAM (SP-CAM) replaces the convection parameterizations with a two dimensional cloud resolving model (CRM) in each gridcell (Khairoutdinov and Randall 2001) and produces an extremely vigorous MJO. Data from ERA-40 reanalysis and TOGA-COARE are also used in our study. Our analysis supports the Discharge Recharge Cycle proposed by Bladé and Hartmann (1993), which asserts that intraseasonal oscillations are the result of cyclic moistening and drying of the troposphere above the Indian Ocean and western Pacific. The CAM is unable to produce high-humidity regions in the mid- to lower-troposphere because of unrealistic aspects of its convection parameterization. The SP-CAM produces an overly moist column due to excessive winds and evaporation during strong convective events. In the real tropics and the SP-CAM, convection within a high-humidity environment produces intense heating and forces the large-scale circulation that is the signature of the MJO. This circulation expedites the drying and re- stabilization of the troposphere, and the cycle begins again. Our analyses show that a model must accurately represent convective processes that moisten the entire tropical troposphere in order to produce a simulation of the MJO.

A53C-0295

Convectively coupled equatorial waves in a simple multicloud model.

* Khouider, B khouider@uvic.ca, University of Victoria, Mathematics and Statistics PO BOX 3060 STN CSC, Victoria, BC V8W 3R4, Canada
Majda, A J jonjon@cims.nyu.edu, Courant Institute, 251 Mercer Street, New York, NY 10012,

Despite the recent progress in super-computing, current general circulation models (GCM) do not represent adequately the tropical variability associated with organized convection, especially the MJO. In this talk I will discuss a recent multicloud model parametrization for organized convection developed recently in collaboration that takes into account the three cloud types characterizing tropical convection: congestus, deep, and stratifrom, and the inherent role of moisture in the progressive deepening of convection, as seen in observations and CRM simulations of organized convective systems. The multicloud models use three vertical modes of heating profiles, corresponding to the three cloud types. Linear theory for the case of a simple beta-plane model reduced to the first two baroclinic modes, of vertical structure, revealed instabilities at the synoptic scales of Kelvin waves, mixed Rossby-gravity and inertio-gravity waves, corresponding to most of the observed spectral power of organized tropical convection as it is reported by Wheeler and Kiladis (1999), with similar reduced phase speeds and horizontal and vertical structures. The multicloud model is currently being implemented in the next generation NCAR GCM: the high order methods modeling environment (HOMME) using the vertical normal modes of Kasahara and Puri. Preliminary results revealing important variability associated with organized convection and equatorial waves in the multicloud-GCM will be presented.

A53C-0296

Equatorial Superrotation in the IPESD Multi-scale MJO Model

* Biello, J A biello@math.ucdavis.edu, Department of Mathematics, University of California, 1 Shields Ave, Davis, CA 95616, United States
Majda, A J jonjon@cims.nyu.edu, Center for Atmosphere Ocean Science, Courant Instutite of Mathematical Sciences, New York University 251 Mercer St, New York, NY 10012, United States
Moncrieff, M W moncrief@ucar.edu, Mesoscale and Microscale Meteorology Division, National Center for Atmospheric Research, P.O. Box 3000, Boulder, CO 80307-8960,

The derivation of the meridional momentum flux arising from a multi-scale zonal velocity field in the IPESD multi-scale models of the equatorial troposphere is presented. It is shown that, due to the balance dynamics on the synoptic scales, the synoptic scale component of the meridional momentum flux convergence must always vanish at the equator. Plausible MJO models are presented along with their planetary scale meridional momentum fluxes. These models are driven by synoptic scale heating fluctuations that have vertical and meridional tilts. Irrespective of the sign of the synoptic scale meridional momentum flux (direction of the tilts) in each of the four MJO examples, the zonal and vertical mean meridional momentum flux convergence from the planetary scales always drives westerly winds near the equator: this is the superrotation characteristic of actual MJOs. We demonstrate that equatorial superrotation occurs when the planetary scale flow due to the upscale momentum flux from synoptic scales reinforces the horizontally convergent flow due to planetary scale mean heating.

A53C-0297

Optimal width of hot spots for driving deep moist convective systems

* Robinson, F J frank.robinson@yale.edu
sherwood, s steven.sherwood@yale.edu
Gerstle, D daniel.gerstle@yale.edu

In a recent study, Robinson et al. (2008) proposed a resonance mechanism for regulating the intensity of convection independently of classical instability measures, over heterogeneous surfaces. They predicted that the most efficient driving for deep moist convection occurs when the surface heating is confined to a scale close to the product of the environmental buoyancy frequency, the characteristic heating time scale and the thickness of the thermal boundary layer. The theory was tested with 2-D models and against lightning observations reported over islands, but the tests were qualitative since the models did not predict cloud electrification or lightning and were highly idealized. The robustness of this "resonance mechanism" is clarified by simulating in three dimensions and with more realistic surface heating and ice microphysics. Moreover, to establish more quantitatively the role of this mechanism in explaining observed behavior, we compare the more-realistic WRF model simulations against TRMM- observed effective radar reflectivities for a range of island sizes. Results suggest dry dynamics is more important than parcel theory in controlling convective vigor. This work is funded by NSF grant 'Physical and Dynamical Meteorology', award number NSF 078550

A53C-0298

Mesoscale convective systems and critical clusters

* Peters, O ole@santafe.edu, Department of Atmospheric and Oceanic Sciences and Institute of Geophysics and Planetary Physics, UCLA, 405 Hilgard Ave, Los Angeles, CA 90095-1565, United States
Neelin, J D neelin@atmos.ucla.edu, Department of Atmospheric and Oceanic Sciences and Institute of Geophysics and Planetary Physics, UCLA, 405 Hilgard Ave, Los Angeles, CA 90095-1565, United States
Nesbitt, S W snesbitt@uiuc.edu, Department of Atmospheric Sciences University of Illinois at Urbana-Champaign, 105 S. Gregory St., Urbana, IL 61801-3070, United States

Size distributions and other geometric properties of mesoscale convective systems (MCS), identified as clusters of adjacent pixels exceeding a precipitation threshold in satellite radar images, are examined with respect to a recently identified critical range of water vapor. Satellite microwave estimates of column water vapor and precipitation show that the onset of convection and precipitation in the tropics can be described as a phase transition, where the rain rate and likelihood of rainfall suddenly increase as a function of water vapor. This is confirmed in Tropical Rainfall Measuring Mission radar data used here. Percolation theory suggests that cluster properties should be highly sensitive to changes in the density of occupied pixels, which here translates into a rainfall probability, which in turn sensitively depends on the water vapor. To confirm this we categorize clusters by their prevalent water vapor. As expected, mean cluster size and radius of gyration strongly increase as the critical water vapor is approached from below. In the critical region we find scale-free size distributions spanning several orders of magnitude. Large clusters are typically from the critical region: at low water vapor most clusters are small, and super-critical water vapor values are too rare to contribute much. The perimeter of the clusters confirms previous observations in satellite, field and model-data of robust non-trivial scaling. The well-known area-perimeter scaling is fully compatible with the quantitative prediction from the plausible null-model of gradient percolation, where the accessible hull is a fractal object with dimension 4/3.

A53C-0299

Diurnal cycle of precipitation in a global cloud-resolving model

* Sato, T t_sato@ccsr.u-tokyo.ac.jp, University of Tokyo, Center for climate system research, Kashiwanoha, Kashiwa, 277- 8568, Japan
Miura, H miurah@jamstec.go.jp, Frontier Research Center for Global Change, Japan Agency for Marine-Earth Science and Technology, Yokohama, Yokohama, 236-0001, Japan
Miura, H miurah@jamstec.go.jp, Department of Atmospheric Science, Colorado State University, Fort Collins, Fort Collins, 80523-1371, United States
Satoh, M satoh@ccsr.u-tokyo.ac.jp, Frontier Research Center for Global Change, Japan Agency for Marine-Earth Science and Technology, Yokohama, Yokohama, 236-0001, Japan
Satoh, M satoh@ccsr.u-tokyo.ac.jp, University of Tokyo, Center for climate system research, Kashiwanoha, Kashiwa, 277- 8568, Japan
Takayabu, Y N yukari@ccsr.u-tokyo.ac.jp, Institute of Observational Research for Global Change, Japan Agency for Marine- Earth Science and Technology, Yokosuka, Yokosuka, 237-0061, Japan
Takayabu, Y N yukari@ccsr.u-tokyo.ac.jp, University of Tokyo, Center for climate system research, Kashiwanoha, Kashiwa, 277- 8568, Japan

This study summarizes the diurnal cycle of precipitation that is simulated by a global cloud resolving model (GCRM) named NICAM (Nonhydrostatic ICosahedral Atmospheric Model) which does not use cumulus parameterizations due to high horizontal resolution. Thirty-day integration by NICAM successfully simulates precipitation diurnal cycle associated with land/sea breeze and thermally-induced topographic circulations as well as the horizontal propagation of diurnal cycle signals. A first harmonic of the diurnal cycle of precipitation in 7 km-mesh run agrees very well with the satellite observations in its geographical distributions although the amplitude is slightly overestimated and peak time is 1.5 hour later than that observed over land. Sensitivity experiments, changing the horizontal resolution, suggest that prominent resolution dependence is discernible in the precipitation diurnal cycle in NICAM. The coarser resolution (14 km mesh) run induces about three hour later peak than that in 7 km mesh run. The 3.5 km mesh run realistically produces peak time (around 15 LT) and amplitude similar to those observed in TRMM PR observations. Meanwhile, the resolution dependences in phase and amplitude are negligibly small over ocean domains. The different sensitivity against the horizontal resolution attributes to the different structures and life cycles of convective systems between land and ocean. The NICAM simulation revealed that the diurnal cycle of rainfall over the maritime continent is strongly coupled with the land-sea breeze systems controlling a convergence/divergence pattern in the lower troposphere around the islands. Additionally, the analysis on the cold pool events suggests that the cold pool is often formed over the open ocean where the precipitation intensity is high, and cold pool propagation is related to the diurnal cycle of precipitation as well as the land-sea breeze.

A53C-0300

Development and Testing of the Dynamic Formulation of Relaxation Timescale for use in Cumulus Convection Parameterizations

* ALAPATY, K kiran.alapaty@science.doe.gov, US Department of Energy, Office of Science SC 23.1, Germantown Bldg. 1000 Idependence Ave., SW, Washington, DC 20585-1290, United States

To avoid artifacts resulting from using a specificied value of relaxation timescale, the new dynamical empirical formulation is developed and tested using observations and a single column version of CAM. The new formulation is based on the ratio of two characteristic scales associated with cumulus convection: (1) depth scale, and (2) representative vertical velocity scale. The depth scale is considered as the thickness of an ensemble of convective clouds in a grid column or thickness of a convective cloud obtained from observations. The velocity scale is estimated using the convective available potential energy either from an observed sounding data or from numerical model grid column parameters. The new formulation is implemented into the Zhang-McFarlane cumulus convection scheme to study the importance of dynamically varying relaxation timescale as opposed to the specified constant value. Various data available from the Tropical Western Pacific-International Cloud Experiment (TWP-ICE) and other observations are used to evaluate the new formulation and also to drive the single column model. Estimated values of relaxation timescale obtained from using the TWP-ICE measurements for 25 days range from 10 minutes to about 2 hours for shallow clouds; and about 3 hours to about 18 hours for deep clouds, consistent with those reported in literature. Analysis of results obtained from using the new dynamical formulation for relaxation timescale in the 1-D CAM simulations for 10 days indicate: (1) the new dynamical formulation works well for this case study (and also for two other studies); (2) estimated relaxation timescale ranges from about 15 minutes to a maximum of about 8 hours for various stages of modeled deep convection; (3) simulated precipitation, and temperature and moisture profiles are closer to observations than those obtained using a constant value for the relaxation timescale; and (4) a partial shift of precipitation from deep clouds to shallow and grid-scale clouds, consistent with a larger relaxation timescale estimated by the new formulation. These preliminary results also indicate that the new formulation has a potential to improve MJO signal and reduce tropical biases in climate model simulations.

A53C-0301

Kelvin Wave and Madden-Julian Oscillation simulated by a Spectral Element Atmospheric Model under Aqua-Planet Conditions

* Kim, Y yj.kim@nrlmry.navy.mil, Naval Research Laboratory, 7 Grace Hopper Ave. Stop 2, Monterey, CA 93943, United States
Giraldo, F X fxgirald@nps.edu, Naval Postgraduate School, Department of Applied Mathematics, 833 Dyer Road, Bldg. 232, Spanagel 262, Monterey, CA 93943, United States
Flatau, M maria.flatau@nrlmry.navy.mil, Naval Research Laboratory, 7 Grace Hopper Ave. Stop 2, Monterey, CA 93943, United States
Liou, C chisann.liou@nrlmry.navy.mil, Naval Research Laboratory, 7 Grace Hopper Ave. Stop 2, Monterey, CA 93943, United States
Peng, M S melinda.peng@nrlmry.navy.mil, Naval Research Laboratory, 7 Grace Hopper Ave. Stop 2, Monterey, CA 93943, United States

The Naval Research Laboratory (NRL) Spectral Element Atmospheric Model (NSEAM) coupled with full physics is used to investigate the organization and propagation of Kelvin waves and Madden-Julian Oscillation under the aqua-planet conditions. It is reported that the model simulation is highly sensitive to the horizontal viscosity, distribution of model vertical levels, and precipitation physics, based on the analysis of simulated convective precipitation in terms of time-longitude plots and the spectral diagrams designed by Wheeler and Kiladis. The initial results confirm the large variability of model simulations associated with convective processes and their coupling to large-scale wave motion, as observed by other global atmospheric models with the aqua-planet setting. The physics of the model is calibrated to capture the essential interaction between the dynamics and physics of the atmosphere, thereby improving the simulation of equatorially trapped atmospheric waves. The main features simulated by the new model with calibrated physics are similar to those predicted by the simplified theory and found in limited observations in view of the speed and spectrum of the eastward propagating Kelvin waves and the signature of the Madden-Julian Oscillation.

A53C-0302

Wave Accumulation and Tropical Cyclone Genesis

* Done, J done@ucar.edu, National Center for Atmospheric Research, NCAR/MMM PO Box 3000, Boulder, CO 80307, United States
Holland, G gholland@ucar.edu, National Center for Atmospheric Research, NCAR/MMM PO Box 3000, Boulder, CO 80307, United States
Suzuki-Parker, A gtg985n@mail.gatech.edu, Georgia Institute of Technology,, School of Earth and Atmospheric Sciences, 311 Ferst Drive, Atlanta, GA 30332, United States

Climatological conditions suitable for tropical cyclone genesis have been known for several decades. They are frequently met over the tropical North Atlantic during the hurricane season, yet the multitude of cloud clusters that do not develop into tropical cyclones suggests these conditions are not sufficient. A hierarchical modeling approach is used to study the transient dynamical process in which interactions of easterly waves with the background flow can change the kinematic character of the waves; thereby resulting in regional accumulation of wave energy and a reduction in the longitudinal and vertical scale of the waves; thereby increasing relative vorticity and the likelihood of genesis. Interactions between waves and the background flow are first studied in the simplified framework of a shallow water model, with emphasis on identifying the important spatial (zonal and meridional) and temporal scales. The importance of the vertical shear of the zonal and meridional flow for wave modulation is then studied in idealized simulations using the Advanced Research WRF model. Finally, a high-resolution simulation (12km horizontal grid spacing) of the tropical North Atlantic region for the entire 2005 hurricane season using the NCAR Nested Regional Climate Model (NRCM) is used to study the importance of the wave accumulation process for tropical cyclone genesis in a full dynamical and physical model. A composite dynamical environment leading up to genesis time, based on the seven tropical cyclones that formed over the tropical Atlantic, differs significantly from the mean environment. Specifically, we find coherent regions in which easterly winds increase towards the east and vertically prior to genesis, consistent with the theory of wave accumulation.

A53C-0303

Relating Convective and Microphysical Properties to Large-scale Dynamical and Thermodynamical processes within Tropical Cyclones

* Mehta, A V amita.v.mehta@nasa.gov, NASA-UMBC Joint Center for Earth Systems Technology (JCET), 613.1 NASA-GSFC, Greenbelt, MD 20771, United States
Smith, E A eric.a.smith@nasa.gov, NASA-GSFC Laboratory for Atmospheres, 613.1 NASA-GSFC, Greenbelt, MD 20771, United States
Tripoli, G J tripoli@aos.wisc.edu, Department of Atmospheric and Oceanic Sciences, University of Wisconsin, 1225 West Dayton St., Madison, WI 53706, United States

Abstract It is well known that precipitating convection within tropical cyclones result from a complex interactions among large-scale, storm-scale, cloud-scale, and micro-scale processes. For improved representation of these processes within tropical cyclone models, it is crucial to first understand how micro-scale and cloud- scale properties within tropical cyclones are related to large-scale processes, one of the key objectives of the Year of Tropical Convection (YOTC) program. In this study, a combination of cloud resolving model (CRM) simulations, TRMM Microwave Imager (TMI) measurements, NOAA Optimum Interpolation sea surface temperatures (SST), and Global Forecasting System (GFS) analysis are used to address this issue. The University of Wisconsin Nonhydrostatic Modeling System (UWNMS), a CRM, is used to simulate hurricanes Dennis (9-10 July 2005), Katrina (29-30 2005), and Gustav (30-31 August 2008) at 2-km resolution, nested within 1ºx1º GFS analyses. The UWNMS-generated thermodynamic and hydrometeor profiles are used in a radiative transfer model to calculate brightness temperatures (Tbs) at TMI frequencies. The UWNMS-based Tbs and TRMM-based Tbs are compared to validate overall consistency of the CRM simulations. The cloud-scale profiles of hydrometeors, vertical wind, temperature, and wind shear from UWNMS are analyzed to study their characteristics as functions of SST and GFS-based large-scale regimes represented by parameters including horizontal moisture divergence, vertical moisture flux at 500 hPa, potential vorticity, large-scale wind shear, and Convective Available Potential Energy among others, throughout mature stage of these major hurricanes. Results of this study show how cloud-scale processes are related to large-scale processes within these tropical cyclones.

A53C-0304

Assessment of prediction skill of intraseasonal variation from dynamical, statistical, and combined models

* Kim, H hyemi.kim@eas.gatech.edu, Georgia Institute of Tech., EAS, 311 Ferst Dr., Atlanta, GA 30332, United States
Kang, I kang@climate.snu.ac.kr, Seoul National University, 501-402, Seoul, 151-742, Korea, Republic of
Webster, P J pjw@eas.gatech.edu, Georgia Institute of Tech., EAS, 311 Ferst Dr., Atlanta, GA 30332, United States

We consider intraseasonal variation (ISV) prediction by statistical and dynamical models. For the fair comparison, the real-time multivariate Madden-Julian Oscillation (MJO) (RMM) index for the boreal winter is used as a predictand. The statistical prediction results are compared by reassessing the multi linear regression (MLR), wavelet, and singular spectrum analysis (SSA) model. The correlation score for RMM1 (RMM2) falls away to 0.5 between 16-17 (15-16) days for MLR, 7-8 (9-10) days for wavelet, and 8-9 (9-10) days for SSA model. As both wavelet and SSA model have a discontinuity at the boundary of data, the skill of the real-time forecast shows a steep decrease at the beginning of the forecasts. To examine the skill of dynamical prediction, serial integration is performed with Seoul National University AGCM and CGCM over the entire boreal winter period. The ocean-atmosphere coupling acts to improve the simulation ability of MJO variability, the eastward propagation, and the phase relationship between convection and SST. The skill score of RMM1 (RMM2) falls out to 0.5 at 18-19 (22-23), 15-16 (17-18), and 16-17 (15-16) for CGCM, AGCM, and MLR. This result demonstrates that dynamical prediction does not lag statistical prediction in skill and is even better when ocean-atmosphere coupling is included. The dependency of prediction skill on the initial phase and amplitude of the MJO is investigated. The score is better when the MJO is initialized during an active period than during a quiescent period for both systems. Based on different characteristics of prediction skill for each phase and amplitude and for individual models, predictions are combined using available information extracted using the better of the two predictions. By simple selection, the prediction skill is clearly improved in strong MJO cases. Using another combination process based on Bayesian concepts, two independent predictions are combined by minimizing the forecast error that is known from historical information. It shows the superior to both of the predictions over the entire forecast lead days.