Notes from the CSTAR Quarterly Conference Call on 25 May (dilatory edition)

I apologize for the much delayed posting of these notes. I recently realized they had not been posted.  JB

A conference call with the principal investigators, the collaborative investigators, students, and other NWS participants with the ongoing CSTAR project was held on 25 May. Some notes from the call are shown below.

Participants included Dr Lackmann, Keith Sherburn, Jason Davis, and Jordan Dale (NCSU), Mike Cammarata and Hunter Coleman (CAE), Larry Lee (GSP), Barrett Smith and Jonathan Blaes (RAH), and Jim Hudgins (RNK).

QPF associated with tropical cyclone boundary interaction
•    Monthly calls continue on the first Wednesday of the month at 9 AM.
•    The project continues to focus on examining the predictability of inland surface boundaries and their impact on tropical cyclone QPF. This project will examine the interaction of tropical cyclones with boundaries to identify what processes determine if a boundary will form and then determine the resulting distribution of precipitation and wind.
•    Based on the input and available data, Tropical Strom Ernesto from August/September 2006 was selected as the first cases.
•    Tropical Storm Ernesto which can be described as a modest tropical cyclone with a strong boundary will be compared with Tropical Storm Hanna which can be described as a modest tropical cyclone with a weak boundary.
•    Jordan made multiple visits to the Raleigh WFO in March and April reviewing data for Ernesto on the Weather Event Simulator (WES). During these visits, Jordan printed hard copies of surface plots and subsequent manual surface analyses of pressure and temperature were completed for every 6 hours from 00Z 8/31 to 23Z 9/01 using METAR observations from SC, NC, VA, and a few offshore locations. These analyses were then compared to MSAS 2-D surface frontogenesis and theta-e plots along with radar and satellite imagery to better confirm the frontal location.
•    Initial results show that an MCS occurring around 00Z 8/31 was a Predecessor Rain Event (PRE) that formed along a cold front. The front then stalled along the coast after the MCS dissipated. As Ernesto approached the coastline, the coastal front strengthened and became entrained into the circulation between 00Z and 06Z on 9/01. As Ernesto moved inland, it transitioned into an extratropical cyclone with an occluded front developing by 18Z on 9/01. We speculate the presence of a frontal boundary may have provided enhanced forcing for lift, particularly to the east of the TC center where a region of enhanced radar reflectivities was present. Storm total accumulations in excess of 10 inches occurred in this region.
•    The next steps for the project include creating frontogenesis and theta-e plots using RUC data and comparing them to the MSAS plots, examining upper air charts for the presence of any jet features, examining GFS, NAM forecast accuracy for this event, and determining the role of the boundary by conducting modeling runs.
•    Conference calls for the TC boundary QPF project were held on 4 April and 9 May.  Notes are available on the CIMMSE blog.

Tropical cyclone inland winds
•    Monthly conference calls continue on the second Wednesday of the month at 11 AM.
•    Preliminary results were presented at the recent NWS Tropical Cyclone Webinar held on 27 April. Topics covered during the presentation included the over prediction of wind speeds in the NDFD, inland decay of wind after landfall, suggested wind reduction factors, and collaboration needs.
•    A set of unified training slides were produced for ER WFOs to incorporate into their seasonal tropical cyclone training.
o    Demonstrate the problem with examples of poor collaboration
o    Illustrate the inexact nature of the TCM product by noting the limitations the NHC deals with in creating the TCM guidance
o    Provide the NDFD verification from this project which notes a consistent high bias in wind forecasts
o    Provide initial guidance on quality land reduction factors and some sort of TCM product correction factor
•    Bryce has begun to setup WRF-LES simulations for Hurricane Irene using the configuration below:
o    Initialized 12 hours prior to landfall
o    Nests:  8.1 km / 2.7 km / 900 m / 300 m / 100 m
o    Account has been created on NCAR’s supercomputing system; working with NCAR folks on getting WRF-LES up and running.
•    Working to complete the NDFD verification with a focus on the 48, 36, 24 hour and just prior to landfall forecasts. Will post all images to the blog so forecasters can reflect/analyze.
•    Bryce continues to examine storm decay and will be plotting radial decay for all storms as a function of time and will analyze by quadrant to see if there is a better way than the currently used linear interpolation to develop hourly grids. Also plan to derive surface roughness based on land use to examine frictional decay resulting from this surface roughness.
•    May examine unaltered TCM wind forecasts and compare them to HWind analyses to provide a measure of land reduction factors or TCM error.
•    Plan on examining the rankine vortex used in the TCM wind tool.  Is there a better model for wind speeds as a function of distance from storm center?
•    Conference calls for the TC wind project were held in 9 March and 11 April. Notes are available on the CIMMSE blog.

High Shear Low CAPE (HSLC) convection
•    Monthly conference calls continue on the fourth Thursday of the month and are well attended.
•    Keith Sherburn continues to work on a parameter based climatology. The desire is to develop composite parameters to better anticipate HSLC events and tornado or no tornado events in particular.
•    Keith has investigated high impact events (those with at least 80 severe reports) to determine differences between widespread tornado events against primarily wind events; the main discriminating factor seemed to be deep layer shear orientation with respect to an associated boundary. Tornado events had a larger cross boundary component to winds and shear, while wind events had more of an along boundary component.
•    Using archived RUC analyses and the SPC’s convective modes database, Keith has created mean soundings for four convective modes: discrete supercells (DISC; 56 cases), supercells embedded within clusters (CLUS; 61), supercells embedded within lines (LINE; 57), and linear non-supercells (NONS; 126). It was determined that the depth of instability (depth from the level of free convection to the equilibrium level) was more shallow in CLUS and NONS cases when compared to DISC and LINE cases. Furthermore, the mid- and upper-level winds in LINE and NONS cases became somewhat unidirectional, while curvature continued throughout the entire hodograph in DISC and CLUS cases. Subtle differences in wind speeds were noted, as well.
•    The null data has been acquired from the SPC, and a null database has been created. In total, the database has 114 null points. Preliminary calculations indicate that effective shear, low-level lapse rates, and composite parameters such as the significant tornado parameter and vorticity generation parameter show the greatest discrimination between tornado events and all nulls.
•    The CIN discrimination mentioned during the call regarding events vs. nulls was recently found to be inconsequential.
•    Jason Davis continues to work on a radar based climatology of HSLC events including an analysis of tornado vs. no tornado cases, examining convective mode, along with radar interrogation strategies such as azimuthal shear and mesocyclone change algorithms.
•    Jason has been working on a portion of the radar climatology that will involve quantifying the strength of the radar-observed rotation for numerous tornadic and non-tornadic circulations during HSLC events.
•    He will be recording the strength of the rotation over the lifetime of the circulation and at multiple radar elevation angles. This will be done with azimuthal shear, produced by the WDSS-II application from NSSL (the same variable that is used in the rotation track plots that some people may be familiar with).
•    The hope is to find differences in the strength, lifetime, depth and/or rate of strengthening between tornadic and non-tornadic circulations, and look for variations in these parameters by convective mode.
•    Jason is currently working on finalizing a method for automatically tracking these circulations. He has also been working on applying an algorithm from a recent paper that attempts to correct azimuthal shear values for their range from the radar.
•    Over the summer, it is hoped to apply the method for tracking these circulations and recording information about them to many cases, and also study radar reflectivity structures as well.
•    WFO GSP continues to finalize a case study template for participating WFOs to use to produce nearly a dozen comprehensive cases studies of different events based on the proposed “high priority” case study responsibility table.
•    Justin developed an excellent comprehensive methodology for creating the case studies. The methodology provides excellent references and guidance for a variety of aspects of the case study including obtaining RUC proximity soundings.
•    FFC has largely completed a null database which has required a tremendous amount of work.
•    We plan to have three HSLC abstracts submitted to the Severe Local Storms Conference in Nashville; one from GSP, one from Keith and another from Jason.
•    Conference calls for the HSLC project were held in 2 February and 26 April. Notes are available on the CIMMSE blog.

Tropical cyclone initialization
•    Dr Lackmann has agreed to create and present a webinar for an operational audience on how TC initialization is accomplished and how this impacts operational NWP. The date of the training is still to be determined but it will likely be in early July. The goal is to provide a baseline of knowledge (bogusing, vortex relocation, etc) to forecasters and share with them the strengths and limitations of each system in a way that would be helpful for forecasters.

High-Resolution Mid-Atlantic Forecasting Ensemble (HME)
•    No update.

Other items
•    A fall workshop is still planned in Raleigh on 15-16 November.
•    The next quarterly call is scheduled for late August.

This entry was posted in CSTAR. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s