Thursday, January 26, 2012

Notes from the 92nd Annual Meeting of the American Meteorological Society

Greetings from New Orleans! Writing this on Wednesday evening, and it's mild and muggy here, ahead of what should be a soaker of a rainstorm tonight and tomorrow.

I had an opportunity to attend the Annual Meeting of the American Meteorological Society this week for work (still one more day of talks tomorrow, but just going for a couple work-related items). For those of you that have never heard of the conference or were always curious about it, or heck, even never knew there WAS an American Meteorological Society, here's a link to the conference agenda. Also, for those of you that were curious about the details contained in the talks, the AMS provides a wonderful service by putting all the presentations online after the conference adjourns. (Just a caveat...I'm not 100% sure if that's limited to dues-paying members of AMS or not, but check it out in a week or two). They've also been blogging all week on the AMS "Front Page." Also, you can just do a twitter search for the hash-tag #ams2012 to get some insights on the goings on, both inside and outside the conference.

Anyway, the conference is the pre-eminent meeting of meteorologists from all background here in the US. It's usually held in January or February (early January next year in Austin) and some of the more frequent locations for it include Long Beach, CA, Seattle, Atlanta, Phoenix, New Orleans, and Austin. 2020's will be in mid January in Boston...I squirm a little at that! I tried to blend the talks I went to between (mostly) things I found useful for my day to day forecasting and work, along with interests in the field that I have. The following are some notes I took over the last few days from some talks you may find interesting. I'll try and be brief where I can, and don't hesitate to post questions in the comments or on Facebook if something doesn't make sense.

Climate Change

- Attended a talk on global extremes, specifically related to a possibly warming climate (staying out of the global warming debate on this...there were a LOT of talks on that with the latest research and all, but I didn't have a chance to attend many). There was discussion about the US Climate Extremes Index, developed as a way to gauge how extreme the weather has been and if there are trends. There was a push to develop "fact sheets" on certain kinds of events with a lot of information that will answer obvious questions after major/extreme events.

Seasonal and US Weather

- Several talks from folks at AER Inc. that are refining a "snow advance index" they developed that links October snow cover advances in Eurasia to the state of the Arctic Oscillation (AO).

- A group from Lyndon State College have found a link between a preconditions for extreme precipitation events in winter in the Northeast (rain or ice) and subsequent cold air outbreaks in the Central and Southeastern US (in other words, perhaps a way to predict the state of the AO as it changes).

- Some research that throws up a major caution flag on utilizing sudden warming events in the stratosphere to predict the state of the AO and possible pattern changes in winter.

2011 Severe Weather Season

The most sobering and eye opening talks I attended had to do with the tornado outbreaks of 2011. Here are some highlights.

- Research was done into how it was possible for three waves of tornadic thunderstorms to occur on April 27th (the Alabama Superoutbreak), and it was hypothesized that a very strong 850 mb low level jet out of the Gulf allowed for quick replenishment of the atmosphere after each wave passed.

- The early morning line of thunderstorms in Alabama and Tennessee (which is known as a Quasi-Linear Convective System (QLCS)...good info on that here) caused many, many power outages that possibly had a negative impact on the ability for people to receive warnings when the main event struck that afternoon. There was also talk about how those sorts of lines of thunderstorms need to have the warning size reduced for tornadic spinups. There's frequently a large warning issued to cover most of the line, when it should be smaller warnings issued for the main threat areas.

- It was also hypothesized that some people may have believed that the AM event WAS the main event.

- Some modeling was done on predicting tornado tracks of past outbreaks vs. April 27th, and the 2011 one showed insanely extreme characteristics in the modeling ahead of other words, this was more than likely a once in a lifetime event.

- The NWS Service Assessment summary didn't really have much new to it. There's some discussion that Tornado Watches may have been issued almost too early. It was obvious that people seem to wait for confirmation of a threat before they believe what they hear about that possible threat.

- The talk at the conference seemed to indicate a strong consideration within and outside of the NWS for dynamic warnings (warnings that are issued by a meteorologist and then automatically refresh every 1-minute to reduce overwarning and free up manpower to work on confirmation, reports, analysis, etc.). There was also a push to produce more "familiar" placenames in the warnings so people hear the "trigger" they need to take action...this was also to remind TV meteorologists of the importance of that.

- During April 27th the NWS in Atlanta issued small polygon warnings for tornadoes, surrounding them with larger severe thunderstorm warnings to account for non-tornadic threats and to minimize overwarning. This seemed to be met with some consternation in the room, as that is a ton of work during an event like 4/27.

- No work is planned immediately to reduce the false alarm rate (FAR) of tornado warnings (major theme of the 2011 tornadoes was the public's warning fatigue in some of these places, namely Joplin, where sirens go off and warnings are issued, but "nothing ever happens here."). The NWS sees no benefit to that right away, but they are beginning work to make methodical changes to how business is done with the hopes of reducing FAR soon.

- NOAA Weather Radios have made a leap to SAME code technology, but there is word that a couple companies have developed prototypes for GPS based weather radios.

- The St. Louis Airport tornado last spring provided the impetus to force them to develop clearly marked shelter signs and a backup plan (that didn't exist) if they had to evacuate the airport control tower. Also, the FAA didn't disseminate tornado warnings to pilots (including two planes that were on the tarmac at the time the tornado hit) because those "aren't aviation specific products." And they still have not added them to their data suite...I'm beyond perplexed.

- One of the cooler talks related to the Joplin tornado in May, and it described how there was actually a seismic signature recorded as the tornado passed 1.7km from a station with a seismograph. These stations also come equipped with other sensors and there was a tremendously intriguing signature of a lot of turbulence that can't be explained as background noise once the tornado came within 10 km of the station. The moral of this talk was that there may be a way to detect tornadoes in progress when no visual confirmation can be confirmed.

- Consensus was that as bad as the Joplin tornado was, had it struck during a weekday or had the high school had its graduation as planned in the auditorium, it could have been far worse, as in death toll near 1,000.

- Mike Smith, who runs the Meteorological Musings blog and wrote a book (currently in my queue to read) called "Warnings" (about the amount of life saving work that's been done to improve the US warning system for weather) made what I thought was a poignant (but necessary) challenge at one of the talks regarding Joplin about the shortcomings of the warnings issued during that afternoon by the NWS in Springfield. There was much discussion about tornado sirens and how there needs to be standards regarding when they need to be activated, as their usefulness becomes diminished as people are overwarned continuously.

- Nate Johnson from WRAL in Raleigh gave a good talk on how social media has exploded in severe weather (specifically relating the North Carolina tornado outbreak in April as an example). It's actually become a vital tool now (not just for NWS, but for media) to get the message out and to gather up to date information.

All in all, there's a lot there, but these were serious talks that had to be made regarding the warning system, the sociological side of weather (what I like to call "sociometeorology"), and the methodology that we all go about trying to get the message to people. I found it to be very eye-opening, sobering, and it poses a significant challenge to folks in the NWS and the media going forward.