Robert L. de Zafra
2nd ITPA Student Symposium
April 28, 2005, 1-5pm
Wang Center, Room 103
Background and Objectives
The purpose of this symposium is to: 1) provide a formal and regular mechanism for ITPA graduate students to share their research interests and activities with other students and the faculty, 2) provide a forum for ITPA graduate students to practice their presentation skills in a more formal seminar / symposium setting, 3) provide a convenient and exciting way to meet course (i.e. OCN 694) and graduation (i.e. annual reviews) requirements, and 4) honor one of ITPA’s most eminent professors – one that has played a substantial role in ITPA’s formation and development.
Thursday, April 28, 2005
The symposium will be a half day event(1-5pm), and will be held at the Wang Center on the Stony Brook University campus, in Room 103. The accomodations involve a modest size conference room equipped with a computer and digital projector. All presentations will be loaded onto the computer prior to the symposium. Refreshments will be provided.
Abstract: The coastal region of the southern Alaska often experiences high wind speed events near the surface, which are problematic for the local fishing, shipping industry, and recreation. In this region, there is a complex spectrum of coastal flows, from offshore gap flows to terrain parallel enhancements (barrier jets). A field study, Southern Alaskan Region Jet (SARJET), was completed in late Sept-Oct 2004 to explore the evolution of the thermal and kinetic structure of these coastal winds and determine the primary forcings which govern their structure and intensity. The region near Juneau, Alaska was chosen because of the high frequency of barrier jets in this region and the high frequency of Synthetic Aperature Radar (SAR) image coverage. The Wyoming King-Air aircraft collected in situ meteorological data from 11 cases studies for a variety of barrier jets. This presentation will highlight the sampling of a barrier jet case observed on 12 October, 2004, in which a frontal system made landfall along the Alaskan coast. There were two flights conducted during this event, which captured the structural change over a four hour period. Early in the event, a gap flow of approximately 30 kts was observed exiting a large coastal gap. Later in the event, a barrier jet with maximum observed wind speeds in excess of 50 kts was located near the coast of the Fairweather Mountains. An observed warm anomaly along the coast was created by the adiabatic warming of descending offshore flow early in the event, which resulted in a different thermal structure from previously documented barrier jet cases. Preliminary results from an MM5 simulation is presented to diagnose the thermal and dynamic mechanisms of this event. Some of the motivating questions we intend to address in this case study are: 1) How does the adiabatic warming of the offshore flow influence the structure and intensity of the jet, 2) What mechanisms are responsible for the fairly rapid decrease in wind speeds on the offshore side of the jet, and 3) How did the changing ambient flow on the synoptic scale influence the barrier jet.
Abstract: There are clear deficiencies and uncertainties in bulk microphysical parameterizations in climate and weather models, which need to be addressed using in situ observations and high-resolution model simulations. This talk describes the synoptic, mesoscale, and precipitation structures observed and simulated over the central Oregon Cascades on 04-05 December 2001 during IMPROVE-2. The Fifth Generation Penn State/NCAR Mesoscale Model (MM5) was used to simulate the flow and precipitation features associated with the storm. Extensive model verification was performed utilizing the wealth of observational assets available during the experiment, including in-situ aircraft measurements, radiosondes, radar data, and surface observations. The 04-05 December 2001 storm system was characterized by strong (20 ms-1) cross barrier flow interacting with the Oregon Cascades. The model realistically simulated the three-dimensional thermodynamic and kinematic fields as well as the associated orographic precipitation enhancement. Deficiencies in the model simulations included a 1-2 oC warm bias in the boundary layer, a ~5 kt weaker cross-barrier flow than observed, and a 2-hour faster upper level trough. Despite a relatively skillful kinematic and thermodynamic forecast, the model overpredicted surface precipitation by 30-50% in localized areas along the windward slopes and underpredicted precipitation over a broad area to the lee of the Oregon Cascades. Some of the same deficiencies were noted in other IMPROVE-2 IOPs, so this talk will discuss ways to isolate the problems based on the field data.
Abstract: Accurate Quantitative Precipitation Forecasting (QPF) has been a long-standing goal of the operational forecasting community. However, this goal has proved elusive since mesoscale precipitation features largely affect precipitation accumulation. Mesoscale band formation in extratropical cyclones has been a popular focus for study in improving QPF, since these features can dramatically affect the basic intensity, timing, and subsequent accumulation of precipitation. Numerous theoretical and observational studies have established that mesoscale bands are primarily forced by frontogenesis in the presence of small moist symmetric stability and sufficient moisture. This current study aims to refine our understanding of the physical processes responsible for mesoscale band formation and further explore their predictability through high resolution and ensemble modeling. Initial work has focused on the 25 December 2002 (Christmas Day) snowstorm, where the formation of an intense mesoscale snowband in the comma-head region of a rapidly deepening cyclone along the northeast U.S. coast contributed to extreme snowfall. Comparisons between versions of the fifth-generation Penn State/NCAR Mesoscale Model (MM5) and the Weather Research and Forecasting (WRF) model will be presented. Model simulations were run at 12 km horizontal resolution and initialized at 0000 UTC 25 December 2002—approximately 19 h before band development. Preliminary results show both the MM5 and WRF accurately predicted rapid cyclogenesis with forecast surface cyclone tracks within 50 km of the observed track; however, the models underpredicted the cyclone intensification, especially the MM5. Despite various strengths of the surface cyclone, both models exhibited band development in central New York within 2 h of the observed time, suggesting some degree of predictability in this case. However, the initial band position in all three models was as much as 100 km too far southeast, and band dissipation occurred 2–3 h prematurely in all three model simulations. Furthermore, both models underpredicted precipitation within the banded region. Detailed analysis of the physical processes responsible for the simulated precipitation bands will be shown. To further explore the predictability of the 25 December 2002 case, a number of simulations using different boundary layer physics and convective parameterizations were run. Most model simulations predicted band formation, suggesting high predictability in this case, although the timing, location, and intensity of the simulated bands varied. It is hypothesized the predictability of banded events will be highly case-dependent. Future research will draw on a number of cases to better understand the nature of band predictability.
Abstract: The maintenance of the Southern Hemisphere (SH) winter monthly-mean split/non-split flow anomaly has been examined by two model diagnostics in this study. A stationary wave model simulation illustrates that the split/non-split jet is maintained by the anomalous vorticity flux. The anomalous heat flux acts to weaken the split flow, but its effects are overwhelmed by those of the vorticity flux. The model also shows that the
anomalous tropical heating is not important for the maintenance of the split/non-split flow. The organization of high-frequency eddies by the low-frequency split/non-split jet has been studied by a storm track model simulation. Two sets of linear runs of a linear GCM with random initial perturbations superimposed onto the split and non-split jet basic state respectively have been conducted to establish the statistics of the storm tracks. The modeled storm track anomalies that are caused by the split/non-split jet match the storm track anomalies that are associated with the split/non-split jet observed in the ECMWF reanalysis data, thus demonstrating that the low-frequency split/non-split jet does organize the high-frequency eddies. Our results establish that there is a two-way feedback between eddies and mean flow anomalies in the low frequency variability of the SH winter split jet.
Abstract: The leading mode in the interannual variability of the winter Pacific storm track is a strengthening (weakening) of the storm track, accompanied by weakening and broadening (strengthening and narrowing) of the Pacific jet stream. A previous study by Harnik and Chang has shown that a stronger but narrower jet inhibits eddy growth by meridional confinement of the eddies, suggesting that a weaker storm track is not inconsistent with a stronger but narrower jet. However, the question remains as to what leads to the changes in jet structure. The maintenance of the anomalous jet structures is examined using stationary wave model experiments. Forcing due to anomalous tropical diabatic heating, as well as convergence of transient vorticity and heat fluxes, are diagnosed from reanalysis data and imposed to force a stationary wave response. When all the forcings are included, the model largely reproduces the observed changes in jet structure. The stationary wave model results are further examined by diagnosing the response to each of the individual forcings, as well as partitioning the eddy forcings into local and remote forcings. Preliminary results suggest that different forcings are responsible for the maintenance of the narrow strong jet and broad weaker jet. For the strong jet case, much of the response come from eddy vorticity forcing, with diabatic heating also contributing. For the weak jet case, eddy heat fluxes give the largest response, with very little response due to diabatic forcing. In addition, evidence that suggests that part of the Pacific jet response could be forced by eddy fluxes over the Atlantic are found — in support of a recent suggestion by Nakamura, Izumi, and Sampe — and perhaps offering an explanation to the recent finding that storm track variations in the Pacific and Atlantic are significantly correlated.
Abstract: The leading EOF of the wintertime geopotential height field in the stratosphere tends to descend through the lower troposphere and to the surface on a time scale of a few months. The first principal component at 10 mb, also known as the Northern Annular Mode, shows significantly greater variability under solar maximum than under minimum conditions. The imprint of this variability is traced to lower altitudes using composite analyses in the east and west phases of the Quasi-Biennial Oscillation. We specifically examine the impact on the Icelandic Low and the Azores High in the lower troposphere.
Abstract: The magnitude and configuration of tropical upwelling of tropospheric air into the stratosphere plays an important role in determining tropical tropopause properties and in stratosphere-troposphere exchange. It is likely of great importance in understanding observed variations in stratospheric water vapor. Here, we present several idealized models of tropical upwelling, which clarify the roles of the nonlinear Hadley circulation and extratropical wave driving in determining the circulation. In particular, we show that the Hadley circulation and wave-driven circulation interact to determine the nature of tropical upwelling. We are able to explain several observed features such as maximum upwelling in the summer hemisphere and the annual variation of the upwelling.
Abstract: While it is widely recognized that the MJO is quasiperiodic, several factors (quality and length of data records, the lack of a sufficiently explanatory and rich theoretical understanding of the MJO, and the challenge of interpreting traditional statistical analyses performed on quasiperiodic data) make it difficult to assess the interplay of the slowly evolving mean state and the intraseasonal perturbations of the MJO. As a result there have been very few studies to date that attempt to take account of changes in the background state, either observationally or theoretically, to examine intraseasonal evolution. Two primary strategies are used here to elucidate some features of the MJO that have not been widely studied. The focus is on the moist fields, which are expected to play a key role in those aspects of asymmetrical response of the MJO about climatological mean states. The first strategy is to construct maps estimating the typical rectification response (change in the mean state) that appears in intraseasonally analyzed moisture data. To some degree, the rectification of the MJO appears as an intermittent means of mediating the slow changes between summer/winter background states on which intraseasonal anomalies themselves propagate. In the process of analyzing the quasiperiodicity of MJO-related precipitation, it became evident that a statistical strategy also exists for predicting the amplitude of future MJO events even when the exact timing of future events is far from certain. The second strategy used here for addressing the challenge of nonlinearity of the MJO response to certain basic states is to re-examine simple models for how the intraseasonal time scale of the MJO emerges. In the literature, a host of competing conceptual frameworks have been proposed based on differing mechanisms. Rather than focus on settling the debate over what the dominant mechanism should be, it seems fruitful to examine what the consequences of mean-anomaly interaction might be. It is possible that the quasiperiodicity and occasional breaks in consistent eastward propagation can be explained by hypothesizing relatively simple forms of nonlinearity. However, it remains to be seen whether or not such an approach can credibly provide a coherent link between the observed character of the rectification and the proposed interaction of the mean state and the MJO.
Abstract: The 41-year NCEP/NCAR and ECMWF reanalysis data (Jan1958-Dec1998) are used to study the trend in the tropical vertical circulation. Results show that the Walker circulation in the tropical Pacific region is weakened with the global warming trend during the summer season. The GFDL GCM Model results are also used to verify this conclusion. Possible reasons for the weaker Walker circulation trend are given.
Abstract: It’s a common problem for model simulations to produce a spurious double ITCZ in the tropical Pacific region south of the equator. A comparison of observation from CMAP and simulation from CAM3 is presented. Based on the pentad CMAP data set, tropical Pacific precipitations are filed into three categories: the ENSO type, the double ITCZ type and the ordinary type. The ENSO type and the double ITCZ type represent two ‘extreme’ patterns in the tropical Pacific precipitation. The ENSO events have the greatest effect on tropical precipitation in both eastern and west-middle tropical Pacific regions. The double ITCZ events largely affect the precipitation in the eastern tropical Pacific during the boreal springtime, with a considerably dry background in the whole tropical Pacific region. In summary, the ‘double ITCZ’ problem in the CAM3 model simulation is: the model misses out the real double ITCZ in the far east Pacific region (110ºW-90ºW) in the boreal spring time and creates a spurious double ITCZ in the middle and middle-east pacific region almost throughout the year.
Abstract: The goal of this study is to analyze large-scale heat and moisture budgets under different synoptic regimes to identify how atmospheric models can better parameterize the effects of convective processes. In this study, apparent heat source (Q1) and apparent moisture sink (Q2) are analyzed over the Atmospheric Radiation Measurement (ARM) Southern Great Plains (SGP) site for the year 2000. The vertical profiles of Q1 and Q2 will be supplemented with satellite and ground based observations to assess how the budges of heat and moisture vary with cloud type. Preliminary results using cluster analysis identify four composites of heating rates from analysis of twenty precipitation events between March and July 2000. The four composites are classified as convection in a moist environment, convection in a dry environment, large-scale condensational processes, and frontal subgrid scale. Synoptic regime associated with these four categories will be analyzed. For this period, it is found that the radiative heating rates are in general, small as compared to heating due to condensation/evaporation and subgridscale processes. This analysis will be extended to include the years 1999 through 2001. Future work also includes assessment of Q1 and Q2 from model output.
Abstract: Large-scale Community Atmospheric Model (CAM), and Meso-scale Weather and Research Forecast (WRF) Model are used to simulate the deep clouds related with a cold front over Southern Great Plain. CAM tends to overestimate the horizontal cloud cover but to underestimate the clouds in vertical. The meso-scale modeling using WRF model showed improvements in the horizontal cloud morphology, but the middle level clouds are still underestimated. Further examination of the bottom and top level of convective clouds suggests that convective processes are not well parameterized, which may lead to the underestimation of the middle clouds. Cloud-resolving modeling without cumulus parameterization showed improvements in middle level clouds.