A quick bout of severe weather impacted the Sioux Falls (FSD) CWA on Sunday, April 15th as a deep upper low lifted northeast into the Northern Plains. This small event was my first on radar this season, and even included one decent supercell in southwest Minnesota. As a mini post-mortem, I’ve written up a brief for the radar analysis for that supercell here.
The views expressed in this article are those of the author and do not necessarily represent those of the National Weather Service.
Over the last few months, there have been several news articles floating around the meteorological community that glorify dual-polarization radar as the savior of tornado warning lead time, which is a serious misconception that needs to be addressed. I haven’t read all of these articles, but it seems that the root of this claim lies in dual-pol’s capability to more clearly detect tornadic debris via correlation coefficient. Unfortunately, when tornadic debris is detected this means the tornado is already on the ground, doing enough damage to loft sizable particles which fall into the Mie or geometric scattering regimes. By the time this signal is returned to the radar, processed at the RDA, and appears in AWIPS, we’re already seconds to minutes past tornado touchdown; read: negative lead time. Sure, a debris signature can provide more confidence that a damaging tornado is present, but waiting for that confirmation before issuing an initial tornado warning is not only unreasonable, it’s extremely dangerous. A wonderful series of training modules on the proper use of dual-pol radar is freely available from WDTB, and emphasizes many times over the potential peril in waiting for a debris signature before making a warning decision. Dual-pol radar does not – and can not – increase lead time for tornadoes.
Taxpayers may be wondering why, then, federal funds have been appropriated for this new radar technology, since the key phrase “increased tornado warning lead time” is traditionally used every time National Weather Service funding requires justification. For those that don’t know, it might be important to note that tornadoes are not the number one convective (stormy) weather related killer — it’s actually flooding (Ashley and Ashley 2008). Fortunately, dual polarization radar allows for better differentiation between hail and heavy rain (Scharfenberg et al. 2005), which will improve forecasters’ ability to detect and warn for flash flooding in a timely manner, potentially saving countless lives. Along the same line, hail with a diameter greater than two inches can cause significant property damage and poses a serious risk to exposed life, and can likely be distinguished from smaller hail much more easily using dual-pol signatures (Balakrishnan and Zrnic 1990). Additionally, icy roads and other hazardous winter driving conditions claim multiple lives each year, and are far more common and widespread than tornadoes — chances are, unless you’re a Joplin resident, you know many more people that have brushed with death on wintry roads than have experienced damage or injury from a tornado. During winter weather, dual-pol helps forecasters identify different precipitation types (Elmore 2011), allowing for the detection of freezing rain vs. snow when coarsely spaced surface observations are insufficient. As a result, forecasters can issue more effective winter weather advisories and warnings, and can communicate this information to keep the media, street and transportation departments, and other decision makers abreast of changing conditions.
Just as dual polarization radar is about more than detecting tornadoes, the National Weather Service is about more than issuing tornado warnings. High impact weather occurs year round, across the country, and the message that every tool and every dollar must directly increase tornado warning lead time diminishes our mission. There’s an abundance of ongoing tornado research, social science, and forecaster training aimed at issuing better and more effective tornado warnings which will be invaluable in our goal to save lives, but we’re tasked with protecting life and property in all situations, not just those that are tracked by GPRA goals. In addition, even if the National Weather Service pooled all of their resources and efforts into tornado forecasting, the science is simply not advanced enough to detect every tornado, much less with ever increasing lead time. So instead of sensationalizing and overselling our abilities, news articles should relay the true utility of this new technology: that dual polarization radar will be critical during severe weather to detect flash flooding and hail, and to give forecasters the confidence to re-issue or enhance the wording on a tornado warning — but not to increase tornado warning lead time. Promulgating an accurate message is necessary, because the more that this ‘boy who cried lead time’ syndrome is perpetuated, the more misinformed the public will become about the true abilities of NWS forecasters and our changing needs for funding and new technologies — spelling a troubled future when the promised increase in lead time doesn’t occur.
Ashley, S. T. and W. S. Ashley, 2008: Flood Fatalities in the United States. J. Appl. Meteor. Climatol., 47, 805-818.
Balakrishnan, N. and D. S. Zrnic, 1990: Use of Polarization to Characterize Precipitation and Discriminate Large Hail. J. Atmos. Sci., 47, 1525-1540.
Elmore, K. L., 2011: The NSSL Hydrometeor Classification Algorithm in Winter Surface Precipitation: Evaluation and Future Development. Wea. Forecasting, 26, 756-765.
Scharfenberg, K. A., D. J. Miller, T. J. Schuur, P. T. Schlatter, S. E. Giangrande, V. M. Melnikov, D. W. Burgess, D. L. Andra, M. P. Foster, and J. M. Krause, 2005: The Joint Polarization Experiment: Polarimetric Radar in Forecasting and Warning Decision Making. Wea. Forecasting, 20, 775-788.
The RUC model is set to be replaced by the Rapid Refresh (RR or RAP) on 20 March 2012. This is a delay from the original release date in January, so I don’t consider this date to be set in stone. Either way, the RR will have a similar vertical and horizontal resolution, but several updates/improvements that will affect model forecasts. Here’s what I consider to be the biggest changes:
Also of note is that the RR will have a larger (North American) domain, and thus will use the GFS for boundary conditions, instead of the NAM. For more complete (and updated) information on the RUC to RR transition, see the ESRL’s RR page here.
I’ve added a new page describing my first real storm chase across north central Kansas on 6 April 2006. Some of the pictures are a little iffy since I was using a point and shoot and most were shot from inside the car, but it was fun to remember the event and challenge my memory (no geocoded pictures!) with roads and locations. Check out the page here! I’ve also added a new album to Picasa with storm photos that I’ve taken over the years.
I’ll be presenting a talk entitled “HailSTONE 2011 and Oklahoma Record Hail” to the Weather MINDs group here in Sioux Falls on Tuesday, December 13th, at 7 PM. The meeting will be held at the National Weather Service office, and we’ll also be meeting ahead of time at Big Cheese Pizza around 5:30 PM, so feel free to stop by. For more information on HailSTONE or the record diameter hail stone that was identified by the project this summer in Oklahoma, visit hailstoneresearch.org or find the project on Facebook.
This page is under construction as of 17 Nov, and was developed with the intent to bring together my social media, research, and miscellaneous web content into one centralized location. Please visit any of these sites (FB, twitter, G+, etc.) as this site is developed.
