Severe, tornadic storms and upper-level forcing

Just thought I'd quickly share this one screen capture I grabbed yesterday during the unusual November tornado outbreak over the Midwest.  I find myself more and more often using Weather Underground's "Wundermap" tool to overlay weather model output on current observations, satellite and radar to try and both see how well the model is doing and make connections between model features and what's actually happening.  Here we see the composite radar image over the continental US overlaid on top of the ECMWF 300mb forecast for around 1 PM CST on Sunday.  The 300mb wind speeds are shown in color with the 300mb height lines shown as the white contour lines.

You can really see how these lines of convection in the radar imagery formed along the exit region of the upper-level jet streak.  This is a favorable location for divergence aloft, particularly when the jet streak has cyclonic curvature as it does here.  Furthermore, the surface cold front associated with this storm was actually back over central Illinois at this time.  It was upper-level support out ahead of the cold front---this strong jet streak coming around the base of the trough---that provided a lot of the forcing to get this convection going.  Of course, once these storms organize into line segments they can be self-reinforcing in generating enough lift to keep themselves going.  But once again, this emphasizes how understanding what's going on in the upper-levels (something our models forecast fairly well) can be both useful and important for understanding when and where convection is going to develop (something our models don't do nearly as well).

A return--with some thoughts on forecasting severe thunderstorms

Hello again everyone!  It has been a year since my last blog post, and having finished a lot of work in that time I finally decided to resume posting again.  Hopefully I'll continue more steadily again from here on out.  I continue to get comments and feedback on my old posts (thanks to Google cataloging everything so effectively...) and I appreciate all of the thoughts and comments I've received.  Keep them coming!

Today I just want to offer a few thoughts or notes on forecasting severe thunderstorms with some examples surrounding the Moore tornado.

• It has been well-established by many people that the National Weather Service did an excellent job at warning the people of Moore before the storm hit--as much as 36 minutes of lead time by some estimates, which is well above the national average of around 10 minutes of lead time.  I happened to be in central Oklahoma last weekend (though I left on Sunday evening when there were a different set of tornadic storms moving through) and it was my experience, just as when I was living down there, that the people of Oklahoma are extremely weather literate and aware.  It seemed everywhere I went, the people I encountered would remark about how we're, "in for some rough weather this afternoon" or to, "get that rental car back before a hailstorm moves in...or worse".  Everywhere TVs were tuned to the local news and the Weather Channel and I overheard non-meteorologists talking about "moderate risks" and "mesoscale discussions".  This weather event was not something that was completely unexpected at all, and I admire and credit the people of Oklahoma for taking such an active interest in their weather forecasts so that they can stay safe.
• As someone interested in mesoscale numerical modeling, I follow the work being done by the Hazardous Weather Testbed Spring Experiment teams who are looking at evaluating our model performance and nowcasting tools and abilities for severe weather events.  They have multiple blogs, for example the GOES-R Proving Ground group or the Experimental Forecast Program.  These groups have some fascinating new tools and observations that show what's on the forefront of our ability to predict severe convective weather events.  Below are some things that I noted from their tools surrounding the Moore event.
• To illustrate just how hard it is for our models, even at high resolution, to predict when and where individual thunderstorms will strike, below is a comparison of two model runs from the NSSL 4km WRF, one starting at 00Z on May 20 and another at 12Z on the 20th.  Both are simulations of  what the radar composite would look like for the hours leading up to 20Z--the time when the tornado was entering Moore.  You can see that the 00Z model run (the left column) seemed to pick up more on storms in southwestern Oklahoma and some in central Oklahoma, but it completely missed the development further to the northeast along the cold front.  The 12Z run (middle column) really picked up on the development along the cold front, but kind of missed the storms in central Oklahoma.  The right column shows the actual observed reflectivity composite for comparison.  It's often been noted in convective events that the 12Z model guidance doesn't always provide a better forecast than the 00Z guidance, even though it was run later with more recent information.  This shows just how hard it is for our deterministic models to forecast storm development, even only a few hours in advance.
  
• One way to try and work with the difficulties that any single model will have at trying to predict when and where storms will form is to use an ensemble--many models runs with slightly different initial conditions and/or different model formulations.  We can use these to estimate probabilities of certain events occurring by seeing how many of the ensemble members have that event occurring.  Below shows an example of forecast probabilities of updraft helicity (i.e., rotating updrafts) exceeding various thresholds from the Storm Prediction Center's Storm-Scale Ensemble of Opportunity for a three hour period including the time of the Moore tornado.  Not too bad--the ensemble suggests a high probability of  rotating updrafts throughout central Oklahoma.

          Of course, some ensembles can also be misleading or still not capture things well.  Below is an example prediction from the CAPS ensemble of where a certain parameter (the Significant Tornado Parameter) will be greater than 3 (indicating a likelihood of strong tornadoes) during the Moore event.  The highest probabilities are indicated far to the northeast of Moore.  However, low probabilities are still present in the Moore area, indicating that by this parameter a significant tornado was still possible in this area...

• There are also efforts underway to support a project called "Warn-on Forecast".  The idea is that once cumulus clouds start growing, we can first identify which particular clouds have the highest potential of growing upscale into stronger thunderstorms.  Tools are being developed to do this, including satellite products that estimate the rate of cloud growth by how fast the tops of the clouds are cooling, or lightning-based methods that will identify storms where lightning frequency is ramping up.  Once these growing storms have been identified, the plan would be to build a high-resolution ensemble of models centered on that storm.  We could then forecast the evolution of that storm and get uncertainty information from the different ensemble members.  This would let forecasters evaluate the ensemble to get probabilities for the storm producing large hail, strong winds or even tornadoes.  We could also get uncertainty in the path of the storm, allowing forecasters to make much more accurate warnings.  Furthermore, as our confidence in these forecasts grows, we could issue warnings for the storms before they even become severe, greatly increasing lead times.
• Unfortunately an active, real-time warn-on forecast system like this still a ways off, but some of the tools to support it do exist.  Cloud-top cooling from satellites and lightning mapping arrays are very real things and are actively being tested for their ability to identify storms that will grow.  Some groups are also working on using radar data to make high-resolution model analyses of the wind, temperature, pressure, and moisture fields surrounding these storms as they are developing.  This is the first step in trying to make these storm-scale ensembles.  Below is an example 1km analysis produced from radar data as the Moore tornado was developing.  Vorticity is contoured in black, and you can see that the analyzed wind fields have a lot of vorticity (rotation) in the right areas of these storms.

One problem with this warn-on forecast methodology is that it is inherently limited--we can only identify when and where the growing storms will be after they have already formed and are beginning to grow.  Forecasting convective initiation--that is, when and where storms will form before they've even formed is still incredibly difficult (see the NSSL WRF above for an exmaple).  But we're working on that too...

So, in brief, the meteorological community is working on ways to better predict severe convective events like what happened in Moore.  Forecasters already are doing an amazing job with the resources they have, but new tools are in the works to more accurately refine our forecasts to reduce false alarms and give people more lead time.  It's an exciting time to be researching in this field!

Thunderstorms in Seattle

I awoke this morning to the sound of what I thought were airplanes flying over my house periodically.  It's not uncommon--the main northerly approach to Sea-Tac airport goes right over my house.  And they don't call Seattle "Jet City" for nothing.  But the sound wasn't quite right--and it was occuring way too frequently.  Turns out it was something far more exciting--thunderstorms.

KATX 0.5 degree base reflectivity from 15Z, July 25, 2011.

Lacking my usual Gibson Ridge program this morning, I used an internal utility on the UW computer systems to get this radar image.  You can see a lot of yellows with embedded cores of reds, indicating rather intense rainfall in some locations.  In particular there looked to have been a rather intense cell in eastern Pierce County southeast of Tacoma.

But were these thunderstorms or just heavy downpours?  I did a quick plot of the National Lightning Detection Network data of all lightning strikes registered from 7:30 AM PDT to 9:30 AM PDT.  Here's the result:

NLDN Lightning strikes from 1430Z to 1630Z, July 25, 2011.

Sure enough there have been several lightning strikes in the Puget Sound lowlands, particularly over the Kitsap peninsula.  There has been a lot more activity on the eastern slopes of the Cascades, though. 

The visible satellite image from this morning also shows the billowing, tall clouds identifying this as deep convection.

GOES-W 1km visible satellite image from 15Z, July 25, 2011.

Because the sun had just come up in this image, the sun angle on the clouds was still low.  For this reason, taller clouds cast shadows to their wests.  You can really see the shadow to the west of the cloud in the area of that big storm over eastern Pierce County.  Interestingly that storm began developing right in the vicinity of Mount Rainier.  Perhaps the mountain helped promote the lift necessary to get the storm going...

So why thunderstorms today?  First, we got really warm yesterday.  Here's a meteogram image showing plots of various weather variables on top of the atmospheric sciences building in Seattle over the past day. Temperature is the third panel from the top.  You can see that we almost got up to 85 degrees Fahrenheit yesterday.

Meteogram from 17Z July 24,2011 to 17Z July 25, 2011 from the top of the UW Atmospheric Sciences building, Seattle.

All that heating warmed up the lowest layer of the atmosphere.  At night, the surface and the near-surface layer cooled off again (you can see in the above image that last night the temperature got down to the upper 50s).  However, the air above that retained some of its residual warmth from the day before.

Below is a somewhat complex, but still very informative plot.  This shows several soundings taken using a vertical profiling radar at Sand Point in Seattle.  Every hour, the profiler uses a vertical-pointing radar beam to derive the temperature structure of the lowest 1600 meters or so of the atmosphere--up to about 850mb.  On the plot, the height above ground is given on the y-axis and the "virtual" temperature (which is similar to temperature, only a degree or two different because it factors in the humidity) in Celsius is on the x-axis.  However, coordinates of the graph are tilted--temperature lines are slanted up and to the left and are shown by the light gray lines.  So, it's kind of like a skew-T chart.

The actual temperature profiles are the colored curves.  One is taken each hour, and the legend showing what hour each color represents is on the upper right. 

Virtual temperature soundings from the Sand Point profiler in Seattle for July 2, 2011.

The first sounding is the yellow one, which was at about 10Z or about 3 AM local time.  As time progressed, you can see that the temperature really cooled down, at least below 1000 meters.  The surface temperature drives this--notice that the temperatures at the surface for the first two profiles (the yellow and the light blue ones) are much colder than the air above.  The surface radiates away more energy than the atmosphere above it, so it cools faster.

However, by 12Z (around 5 AM) the sun started to come up (this is now the magenta profile).  You can see that the surface temperature is no longer much cooler than the air above it.  Immediately the surface starts absorbing the solar radiation and its cooling slows down.  It eventually starts warming up.  However, the air above the surface (but below 1000 meters) continues to cool.  Just as the atmosphere cannot cool as efficiently as the surface, the atmosphere also cannot warm up as efficiently as the surface.  It has to wait for warm air in the near-surface layer to mix upwards to really start warming up.  As such, even though the sun has come up, the air between around 500-1000 meters continues to cool even though the surface cooling slows down and actually begins to warm.

However, the cool air at the surface overnight had not deepened beyond 700-1000 meters by the time the sun came up.  You can see that the temperature profiles all abruptly get warmer above the 700-1000 meter depth.  The night was just not long enough for the cold air at the surface to deepen up beyond that height.  Therefore, above this level, the residual warmth from the day before remains.  Because we got so warm yesterday, these temperatures above 700-1000m really are very warm--I mean, it's still 23-25 degrees Celsius up there (around 75 degrees Fahrenheit) while at the surface it had cooled to the upper 50s. 

Why do I go through all of this description of the temperature profiles?  I want to point out that even though at the surface we really cooled off last night, residual heat from our unusually warm day yesterday still lingered just above the surface.  It's that still-warm layer that provided the instability for this morning's storms.

Helping this instability is a trough moving in from offshore.  Here's the 700mb temperature forecast for 15Z this morning:

UW 4km WRF 3-hour forecast of 700mb temperature at 15Z, July 25, 2011.

This forecast is for temperatures above the top of the temperature profile I was just showing.  Notice the cooler temperatures just off the west coast of Washington and the westerly, onshore winds.  These cooler temperature are associated with a trough that is now moving onshore.  So, combine these two ingredients:

1. A warm layer around 1000m above sea level that was left over from yesterday's very warm day, even though the surface cooled down overnight.
2. On top of that warm layer, colder air was moving in associated with a trough moving onshore.

And we have cold air moving in over warming air--the right combination to destabalize the atmosphere.  These are not surface-based storms--they are not drawing from warm air at the surface.  Instead they are elevated storms--they're drawing from warm air in a layer above the surface.

That was a look at our fun collection of thunderstorms here in Seattle.  It's a rare event out here, but it makes sense once you look at the setup.  I'm going to bring back those virtual temperature profiles at some point in the future to talk about the development of the capping inversion.  You can really see in those profiles how a capping inversion is developing this morning over the area.  But more on that in another blog.  For now I'll just enjoy the show.

Aha! An OBSERVED MCV--just the other day

I was pleasantly surprised to see the University of Wisconsin -- Cooperative Institute for Mesoscale Satellite Studies (CIMSS) blog post on July 6th.  It discusses a mesoscale convective vortex that just occured over the southwestern US on July 5th and 6th.  Moreover, they have a stunning GIF animation of visible and IR satellite images that shows how the storms first organized as a mesoscale convective complex then later evolved to a mesoscale convective vortex.

The blog post is at:
http://cimss.ssec.wisc.edu/goes/blog/archives/8409

And that GIF animation (which is the first image in the blog post) is here.

I encourage you to watch the animation to really get a good idea of the evolution of these sorts of storms.  The synoptic-scale flow aloft was relatively weak, indicating that this complex of storms organized away from any strong upper-level forcing.  What began as clusters of thunderstorms during the day (with very distinct outflow boundaries) congealed into an organized mass as night fell.  You can see in the overnight hours (when the animation switches to the lower-resolution IR images) how the cloud structure does indeed take on the characteristic round shape of a MCC. 

By the morning hours (when the animation returns to the higher-resolution visible images), the remaining circular cloud shield (and a fair amount of the convection going on) dissipates.  But, it leaves behind the warm-core vortex that lies at the heart of the complex.  Internal interactions within the cyclonic flow about this vortex are able to maintain some convection about the vortex center.  However, because this is a vortex, the convective structure has the characteristic "swirl" shape we see in MCVs.  Once again, not unlike a hurricane...

It's incredible how fragile MCVs actually are.  As I mentioned in my last post, studies indicate that on average only three or four MCVs are observed in the US each year.  Any strong upper level winds will tend to shear apart that warm-core vortex long before it gets very organized.  You can tell in the animation that the synoptic flow is weak, though, and atypical of a progressive, strong jet pattern.  The MCV itself drifts westward, which is not the typically direction we see storms move--this shows the relatively "mild" conditions aloft.

Anyhow, I just thought it was pretty amazing that the day after I do a post about MCVs (a relatively rare phenomenon), one happened to be observed.  So, please enjoy this fascinating type of storm...

The Mesoscale Convective Vortex (MCV)

For my last post in this series about different kinds of mesoscale convective systems (MCSs), I'm going to briefly talk about one of the rarer types of MCSs--the mesoscale convective vortex (MCV).  These sorts of storms usually form from parent mesoscale convective complexes (MCCs), so we have a full line of mesoscale convective classification:

MCS --> MCC --> MCV

In trying to trace the history of MCVs in the meteorological literature, it seems that the MCV was recognized very soon after Maddox's 1980 paper describing the features of MCCs. It was noted that in some cases, strong mesoscale low-pressure centers would develop in the wake of MCCs.  Johnston (1981) talks about mesoscale vorticity maxima induced by mesoscale convective compexes. Such maxima were theorized to have helped organize the storm that caused the 1977 Johnstown, Pennsylvania, floods by  Zhang and Fritsch (1987).  However, the first actual mention I could find of the term "mesoscale convective vortex" being used to describe this phenomenon came from a paper by Menard and Fritsch in 1989, where they describe a "mesoscale, convectively-generated vortex (MCV)" over Oklahoma and Arkansas.  So you can see that our understanding of this particular phenomenon is pretty young, having only been recognized over the past 20-30 years of reseach.

So what is a mesoscale convective vortex?  In my discussion of MCCs, I talked about how the large area of rising air in the circular MCC had lots of condensation going on.  That condensation of water vapor into rain releases a lot of latent heat, warming the core of the storm.  That warming causes the air to expand, reducing the pressure relative to the surrounding environment.  This, in effect, creates a "warm-core low" with a structure analogous to the structure of a hurricane.  Sometimes, after convection and rainfall associated with an MCC dissipates, this area of low-pressure can actually linger on.  The cyclonic flow associated with the low-pressure center can later help organize new areas of rising motion and convection, often with a charactaristic cyclonic "swirl".

Here's an example of a visible satellite image of this organization going on from the CIMSS satellite page:

GOES visible satellite image from 1615Z, Jul 8, 1997 showing a MCV over western Missouri.  From http://cimss.ssec.wisc.edu/goes/misc/970708.html.

Sometimes you'll see MCVs described as a "mesoscale vorticity center".  This is an apt description, and it means essentially the same thing--it just changes the acronym to MVC. You can see on that example image how the MCC has organized itself into a charactaristic swirl-shape.  The analogies between MCVs and tropical cyclones really start to make sense when you see this kind of organization.

Here's a radar image from Patrick Marsh's blog last year when an MCV was observed moving through the Houston area:

NEXRAD base reflectivity composite of an MCV near Houston at 2220Z on June 3, 2010.  From http://www.patricktmarsh.com/2010/06/day-154-mesoscale-convective-vortex-near-houston/.  Annotated by Patrick Marsh.

Patrick has annotated that radar image with streamlines showing the wind patterns and an L marking the center of this mesoscale vortex.

You can see in these two images why these features are considered to be "mesoscale".  Unlike the banded "swirls" or comma-shapes we see with strong synoptic-scale low-pressure centers, these bands do not mark the locations of frontal boundaries.  They are much more akin to hurricane rain bands than frontal bands.  Furthermore, the size of these vortexes is only at most a few hundred kilometers across--often less.  You can see above that these storms are about the size of a state or less.  This still puts them in the "mesoscale" category--smaller than the synoptic scale.

It turns out that, though mesoscale convective complexes are relatively common across the central US, only a very few actually give way to the eventual development of a mesoscale convective vortex.  Bartels and Maddox (1991) did a survey over seven and a half years of visible satellite data and only observed 24 cases of MCVs, working out to around three or four per year.  So, they're a fun feature to see when you can actually find them.

That's just a quick look at MCVs, a rarer, but still important type of mesoscale convective system.  In my next blog, I think I'll finally move away from talking about MCSs and get back to talking about the weather that's currently going on.

What is a mesoscale convective complex?

As I mentioned in my last post, I want to go into some detail about the definition and structure of several classes of storms.  We use the term mesoscale convective system to describe any collection of storms (or other convective activity) that is organized on a scale larger than a single storm.  Today I want to focus on one particular type of mesoscale convective system--the mesoscale convective complex.

Much of this discussion is going to follow Robert Maddox's 1980 paper "Mesoscale Convective Complexes," which is the first paper to define and describe these types of storms.  You can read the full paper here (which is actually very understandable if you can follow the kinds of things I write in my blog).

In Maddox (1980), he describes a breakdown of different types of large-end mesoscale convective phenomena using this convenient chart which follows nicely from my last blog post:

Table -- Describing a classification of meso-alpha (large mesoscale) convective phenomena, broken down by geometric shape and location.  From Maddox (1980).

The main point from this chart is that these mesoscale convective complexes are "circular" in structure and they can occur in both the tropics and the midlatitudes.  Maddox proposed a definition that still stands today (at least it still matches the American Meteorological Society's definition of a mesoscale convective complex) though there have been proposed amendments over the years:

Table -- A definition of a mesoscale convective complex from Maddox (1980).

MCCs are mostly determined by the shape and size of their infrared satellite signature.  This has always struck me as somewhat arbitrary--what's so significant about the cloud returns being more than 100,000 sq. kilometers in area?  Why does the structure have to have a near-circular shape (eccentricity greater than 0.7)?  It turns out that these parameters themselves are indeed arbitrary--they exist as a way of describing MCCs in the context of the observations that we have and in as succinct of a way as possible. 

From Maddox's paper and several subsequent descriptions, there are several key points that describe MCCs uniquely and separate them from other features such as squall lines:

1. MCCs do not require significant large-scale forcing to form.  Whereas squall lines often form along fronts and are associated with deep troughing and strong upper-air support,  MCCs can form in areas where there is relatively weak support for convection--only mild lifting motion and no significant support aloft.
2. MCCs require a very warm and moist layer that extends from the surface relatively deeply up into the atmosphere.  Because MCCs typically form in areas with weak upper-level support (as mentioned above), they need a lot of latent heat energy to sustain themselves.  Basically, with structures like squall lines along a front, you have an ambient, synoptic-scale environment that works to force and maintain thunderstorms.  With MCCs, it's more like you have storms that work to change and modify the synoptic-scale environment.
3. Because of the strong dependence on latent heat release to maintain the structure of a MCC, they tend to share many charactaristics in common with a warm-core system as opposed to a cold-core system. Typically we think of hurricanes as the strongest warm-core-type storms.  However, MCCs also exhibit this kind of structure.

Let's look at a more direct comparisons between MCCs and squall lines.  Maddox provides some examples from 1979 of both in an attempt to illustrate the difference in structure.

Figures taken from Maddox (1980).  The left side illustrates a MCC on July 12, 1979 and the right side illustrates a squall line on June 20, 1979.  The top panels show infrared satellite images, the middle panels show radar returns and the lower panels show the surface analysis.

The left-hand side of the above image shows an MCC whereas the right-hand side shows a squall line event.  In the satellite images (the top panels), the shape difference is clear--the MCC is very round in shape and covers a large, continuous area.  The squall line, in contrast, has a much more linear shape and lots of variation in the infrared satellite returns--it's not just one big blob.

In the middle panels showing the radar returns, we can see that in the squall line, the bulk of the radar return (and therefore most of the precipitation) is confined to the immediate area of the squall line (ignoring the other precipitation off to the northwest associated with the surface low).  In the MCC, however, there's a rather large area of radar returns underneath the big circular infrared satellite feature.  Furthrmore, there are embedded areas of deeper convection in the MCC--some of which do look "linear".  But, the satellite return and other radar return areas are very large and aren't just confined to that deep convective area.  This sets the MCC apart from the squall line.

Finally, in the lower panels, for the squall line case on the right we can see that the squall line is out ahead of a deep surface low to the west.  The only way you would get a deep surface low like that is to have significant upper-level support (i.e., a strong jet streak with its exit region above).  We can see a lot of pressure contours, and they're fairly close together, indicating strong pressure gradients.  The squall line itself has moved out ahead of the main cold front, but the fact that its structure so well mirrors the main cold front hints that the original storms probably formed along the cold front before strong winds aloft started moving them out ahead of the front.  This is a classic scenario that I've talked about in this blog multiple times.

However, in the MCC case, we notice a very different synoptic setup.  There's no strong surface low nearby at all.  The pressure contours are rather few and far between, indicating weak pressure gradients.  There is a cold front analyzed off to the northwest, but it doesn't appear to be very strong.  Furthermore, the shape of the convective area doesn't really match well with the shape of the cold front.  This implies that while the front may have forced initial storm development, it was not the primary mechanism for organization in these storms.  You can see that in the MCC surface analysis, they've drawn outflow boundaries on the southern and eastern edge of the complex, with a mesoscale area of high pressure underneath the middle of the complex.  This contrasts greatly with the squall line case, and gets back to the difference I mentioned earlier--

Squall lines--their forcing and maintenance--are a product of the surrounding synoptic environment whereas MCCs tend to force and change the surrounding synoptic environment.

In the surface analyses above, the MCC analysis is dominated by features (like the outflow boundaries and meso-high pressures) created by the MCC itself.  In the squall line case, the dominant features (the surface low, the strong fronts and pressure gradients) were NOT created by the squall line.  They worked to create the squall line.  This is a fundamental difference.

Let's explore some of the other features of MCCs by looking at some potential MCC candidates from the other day.  Remember this radar image from Sunday night?

NEXRAD composite base reflectivity from 0228Z, June 27, 2011.

Now--do we call these features squall lines or MCCs?  On radar the deep convection does look rather linear, but we saw even in Maddox's example MCC that we could see linear-like convective elements within an MCC.  There's also extended areas of precipitation to the north and the west of these convective lines in the radar image.  However, the technical definition revolves around the infrared satellite image, so I'll bring that back too.

GOES Infrared satellite image from 0245Z, June 27, 2011.

Immediately we can see that the shapes of these storm complexes on the satellite image is more rounded--there's no linear feature or "comma shape" like we'd expect if there was a deep surface low or a squall line. I haven't measured the area of the satellite returns explicitly, but knowing that the state of Iowa has an area of about 147,000 square kilometers seems to imply that these are large enough to qualify as MCCs (the exact definition(s) of area needed is in the first table above).  So just from these satellite images, I'd call these MCCs.

But lets look at some analyses of the surrounding environment.  Let's go back to 18Z (around the middle of the day) to when these storms were just forming.  Here was the radar image then:

NEXRAD composite radar reflectivity from 18Z, June 26, 2011.

Scattered storms were beginning to develop throughout the Dakotas and into northern Nebraska.  Other storms were beginning to fire in southeastern Nebraska.  Now, let's look at the 300mb chart for this time:

SPC mesoanalysis 300mb winds (barbs and blue colors), divergence (pink) and height (black) for 18Z, June 26, 2011.

Is there a deep trough moving in?  Not really.  Are we in the exit region of a strong jet streak?  Kind of--you can see the blue shadings highlighting the core of the jet winds off to the west.  There's some divergence aloft (shown by the pink contours) but this doesn't seem to be tied to the jet streak.  This divergence is right above the area where storms were developing on radar.  This makes sense--the rising motion in the convection implies that air above has to get out of the way, so we expect divergence aloft above the storms.  But the fact that this divergence is so localized to just the areas where the storms are seems to indicate that this is a storm-induced phenomenon--the divergence is not significantly coming from the overall synoptic pattern.  This agrees nicely with what I said about MCCs before--they modify the surrounding environment more than it forces them.

Continuing to the time when the MCCs had matured, here's the new 300mb map.

SPC mesoanalysis 300mb winds (barbs and blue colors), divergence (pink) and height (black) for 2Z, June 27, 2011.

We can see the divergence aloft (once again, the pink contours) associated with the MCCs in eastern and western Iowa--they're nowhere near any significant jet at 300mb.  There are some wiggles in the height contours, but no troughs or anything supporting these storm clusters.  Once again, the storms are modifying the environment. 

Also in this 300mb image, I find it interesting how the jet aloft has changed surrounding the storms that were developing in central South Dakota at this time.  Notice how the jet streak seems to curve up and around the top of the strong area of divergence associated with those storms in central South Dakota.  If you follow that curvature, you'd see that the jet streak is starting to become anticyclonically curved--it curves clockwise.  Interestingly, one of the features of a warm-core storm is that it promotes an area of anti-cyclonic flow (and "higher" pressure) aloft.  Is this anticyclonic curvature of the jet evidence of a warm-core structure in the storms in central South Dakota?  Remember, in my list of features of an MCC, I said that these storms often exhibited warm-core charactaristics.  So it could be...

Finally, some more interesting facts about MCCs:

1. Most MCCs develop in the afternoon hours, then strengthen through the evening and perisist overnight until the next morning.  Thus they provide a lot of nocturnal rainfall.
2. In fact, MCCs are thought to be one of the largest contributors to annual rainfall for much of the central US (several studies have looked at this, including Fritsch (1986), McAnelly and Cotton (1986, 1989) and Kane et al. (1987)).  Fritsch (1986) claims that most states in the plains and midwest receive 30-70% of their annual rainfall from MCC events.
3. Because of their heavy rainfall contribution (remember--MCCs have precipitation over a much larger area than squall lines and usually occur when there is a very deep moist layer), MCCs are responsible for many flash flood events across the central US.
4. Severe weather is definitely possible within a MCC.  Severe wind and hail are frequently found with the stronger convective elements.  If those elements can organize along one of the MCC's outflow boundaries, the wind threat can get even greater.  Tornadoes are also possible from stronger convective elements within the MCC.  Here's the SPC storm reports for that group of MCCs on Sunday night:

SPC storm reports for June 26, 2011.

And with that I'm going to wrap up my discussion of what makes a mesoscale convective complex.  It's a peculiar type of storm that is very frequent in the late spring/early summer months.  While superficially it can look like a squall line, it's a unique type of organized storm that:

• Doesn't need strong forcing aloft
• Actually works to change the synoptic pattern aloft instead of being driven by the synoptic pattern
• Usually forms with a very deep layer of moisture in the low-levels
• Has warm-core charactaristics
• Brings a lot of rainfall, usually at night, to the central part of the US
• Has a very recognizable circular-type structure on infrared satellite images.

Now you'll know what to look for the next time that the term "MCC" comes up.

Pattern shift -- when will the convective season start?

The first half of February has been one major snowfall after another.  This includes several snowfalls in places that aren't used to seeing a lot of snow--like central Oklahoma.  The bitter cold weather and several snow events in the central US have been due to a persistent pattern aloft since the beginning of February.

Fig 1 -- Hemispheric plot of 500mb heights (shaded) and mean sea level pressure (contoured) valid 00Z, Feb 8, 2011.

The above pattern is similar to what we've had in place for the past two weeks.  Note the very high amplitude ridge off the west coast of the US. This is contrasted by the general troughing (the cooler colors extending further south) across much of the central US.  Short waves (which are the source of many of our storms) will tend to follow the jet stream, and the jet stream (from our thermal wind arguments) will tend to follow the area with the strongest height gradient (assuming height is a proxy for temperature, which it isn't the best, but still...).  This means that all of our short waves will tend to follow the sharp gradient in colors on the map above, which is right around where the blues transition through greens and into yellows.  This implies a storm track where the shortwaves dive south along the eastern slopes of the Rockies, dig across the southern Plains and then move north up the east coast.  The result? Snow-producing storms a lot further south than they normally are and no northward advection of warm air into the central US--so it stays cold.

But a pattern change is currently happening.  Here in Seattle we're getting our first trough to come on shore in the last two weeks--this means that that ridge that has been just parked over the west coast is finally moving inland...

Fig 2 -- Hemispheric plot of 500mb heights (shaded) and mean sea level pressure (contoured) for 12Z, Feb 12, 2011.

Note how there's no longer a ridge off the west coast--instead the ridge has been shunted further south and inland, with its axis (we're looking at the color shadings here, by the way) now over Nevada.  That large-scale troughing over the eastern US no longer extends as far south as it used to.  Warm air looks to be returning to the central US...hooray!

But note another difference in these pictures.  There is a persistent low-pressure center in the north Atlantic that's usually referred to as the Icelandic low (even though it can meander away from Iceland).  We can see it in both the images above as that big, heavily-contoured surface low in the north Atlantic.  Notice how in the first image back on February 8th, the Icelandic low is not as deep as it is in the current image.  In fact, it seems that as the trough over the eastern US is lifting out, some of the "energy" associated with that trough might be deepening the Icelandic low.  The fluctuations in the strength of the Icelandic low are a phenomenon that is called the "North Atlantic Oscillation", and its dynamics are still being researched.  However, there is an index that measures the NAO which basically looks at the strength of the Icelandic low.  Here's a graph of how the NAO has varied over the past few months (we're only really concerned with the top frame).

Fig 3 -- NAO index for the previous four months (top panel). From the CPC NAO page.

We note that ever since the end of January, we've been in a "positive" phase of the NAO--the NAO index measured positive values which correspond to higher than usual pressures in the Icelandic low--just like we saw, the low has not been as deep as it is getting now.  However, we can also see that the recent trend in the NAO over the past week or so has been downward--the low pressure center in the north Atlantic is getting deeper again.  The red lines at the end of the graph indicate various model projections of how the NAO may evolve. It's interesting to note that several of them bring the NAO back up again, which could be evidence of more troughing over the eastern US as a result.  But, these are just models...

Speaking of models, I've started seeing a lot of buzz in various weather discussion areas online about the potential showing up in our models for the first severe weather event of the plains this season.  What's the buzz about?  Well, most of what I've seen has been about the model forecasts for next Friday, which at this point have a surface low moving across the northern plains with rather warm air (compared to the frigid air we've been seeing in the central US) brought up in the warm sector.

Fig 4 -- GFS 144 hour forecast of sea-level pressure (contours), temperature (shaded) and  winds (barbs) for 12Z, Friday, Feb. 18, 2011.

Those are 50 degree temperature projected to get all the way up to northern Illinois by the end of the week--pretty impressive considering how bitterly cold it has been recently.  Of course, in the image above, there's also a relatively strong cold front associated with this low.  At the forecast time shown here, the front stretches from near Sioux Falls, SD through Nebraska, western Kansas and down into western Oklahoma.  Considering the GFS forecast for dewpoint temperatures at the same time:

Fig 5 -- GFS 144 hour forecast of dewpoint temp (shaded) and winds (barbs) for 12Z, Friday, Feb. 18, 2011.

We see marginal dewpoints, maybe in the low 50s at best, across eastern Oklahoma and down into Texas.  So there is some northward moisture advection with this low and with convergence along the cold front, maybe we'll see some storms.  But this is still at 144 hours out--no one should trust any model on details this far out.  We'll see how this evolves in the forecasts for this week...

So when does the convective season usually start in the plains?  I did a quick survey of the past ten years on the SPC's severe weather event archive and documented the date of the first events in the year that had significant severe weather reports of any type in the plains.  I did not include events that were primarily in the southeastern US and also used some discretion as to when the first event was supposed to be--for example, the random early January severe weather outbreaks we often see were not included.  Averaging across the ten years, the average date I came up with was...

March 2nd.

Though we did see first events occurring as early as February 5th and as late as March 23rd.  There also have been month-long gaps between the first and second severe weather events on the plains.  So it's a bit rough.  But, on average--March 2nd seems to be the time when our severe weather kicks up.  But we're already within the envelope of possible dates based on past records--so it's time to start thinking ahead for this year's convective season.

The Puget Sound Convergence Zone and Convection

Ok--I was going to do a post about the large scale pattern in upper-air features over the past two weeks and how that will hopefully start breaking down by the end of this week.  But then on the way back from the University today, I started getting text messages from people...

"Was that hail that just fell?"  "Hail?  Graupel?"  "Is it supposed to snow today?"

Hmm...we were in the upper 40s today in terms of temperature at the surface...I was pretty sure we weren't expecting any kind of frozen precipitation...

But sure enough, when I turned on this evening's news, their lead story was about the downpour of hail that hit Seattle this afternoon.  Interesting...  So I fired up the radar and cycled back to the time of the report.  The radar showed this:

Fig 1 -- KATX 0.5 degree base reflectivity at 0048Z, Feb. 8, 2011.

The little cell I point out with the red arrow is the closest I could come to something looking remotely like convection going on.  A cross-section through the radar returns in that cell is also not very impressive:

Fig 2 -- KATX base reflectivity cross section through the cell identified in figure 1.

There is a small core in the middle of the cell with some vertical coherency.  The reports said the "hail" was only pea-sized or smaller (hardly worth calling "hail" in my opinion...) and as such we wouldn't expect to see THAT large of a reflectivity return from the hail compared with the surrounding rain.  The top of the core only makes it up to an estimated 5000-6000 feet too--not very tall.  But apparently this was vigorous enough convection to produce some small hail.  It's too bad our local sounding site is way out at Quileute out on the Pacific coast so our upper-air profile over Seattle is a bit of a mystery...

So how do we get convection in Seattle?  This place usually does not get the strong fronts and temperature gradients nor the instability aloft needed to produce deep convection.  So we live with shallow convection.  Still...there has to be a mechanism to provide the lift necessary to get convection going.  This is where the Puget Sound convergence zone comes in.

To describe this phenomenon, we first need to focus on the geography of the Puget Sound area.  To the east of the city of Seattle and Puget Sound lie the Cascade Mountains, running north to south.  However, to the west of Puget Sound lies the Olympic Peninsula and the relatively high, but isolated Olympic Mountains.

Fig 3 -- Topographic map of western Washington.  Higher elevations are shown in the tans and browns.  The orange colors indicate urban areas.  To the west of Seattle lie the Olympic Mountains.

Puget Sound convergence zones set up when the winds in the low levels of the atmosphere have a strong westerly component.  Let's consider what happens to low level winds out of the west once they reach the western shore of Washington.  Immediately they are confronted by the isolated high terrain of the Olympic Mountains.  All that air rushing in has two choices--it can either rise up over the mountains or go around them.  Now, we do see a fair amount of lift on the windward side of the Olympics--the Hoh rainforest lies on the western side of the Olympics for that very reason.  But generally the atmosphere suppresses large-scale rapid vertical motions like that (there are many reasons for this, but we'll just accept that for now).  Instead, much of the air is forced to go around the Olympic Mountains.

Fig 4 -- Schematic of westerly flow separating to go around the Olympic Mountains.  A lee-side low is formed.

However, when all that air splits to go around the mountains, what happens on the eastern side of the mountains?  All the air flowing around the mountains creates a sort of void in the atmosphere immediately behind the mountain range.  This is signified by a lowering of the pressure there.

But what does the atmosphere do in response to a lowering of pressure?  Air will rush in to try and fill this "void" that has been left on the eastern side of the mountains.  As a result, the wind that split to go around the mountains will be sucked back toward the low pressure that has formed on the lee side of the mountains.

Fig 5 -- The lee-side low draws the winds inward on the lee side of the mountains.  As the winds from the north and the south meet, a convergence zone is formed.

Of course, air is being sucked in toward the low pressure from both the north and the south.  When these two wind streams meet, there's a region of rather strong convergence (indicated by the dotted line in the figure above).  This convergence can provide enough lift to get shallow convection over Seattle and form small hailstorms like we saw today.  So that's the Puget Sound convergence zone in a nutshell.

So did we have those kinds of conditions today?  You bet we did.  But with a slight twist.  Here's the forecast 925 mb (low-level) chart for 00Z this evening (near the time when the hail over Seattle occurred):

Fig 6 -- UW 4km WRF 12 hour forecast of 925 mb  temperature (shaded) and winds (barbs) at 00Z, Feb. 8, 2011.

We can see here that off the Washington coast, the winds were out of the northwest.  This isn't straight out of the west, but you can imagine that a similar effect occurs.  In this case, the convergence zone would form further south than indicated in the diagram above and would also be oriented in a northwest-to-southeast direction.

Fig 7 -- Same as in figure 5 but adjusted for more northwesterly winds.  The convergence zone is located further south and oriented from northwest to southeast.

So we'd expect to see a convergence in our low-level winds south of Seattle somewhere, though the location tends to meander with time.  What I showed above is a model forecast for 925mb winds.  Do we see this convergence reflected in the observations?  Here's the surface map of observations from 0100Z this evening (about the time of the hail in Seattle):

Fig 8 -- Surface METAR observations from western Washington at 0100Z, Feb. 8, 2011. 

I've added several blue arrow roughly paralleling the wind barbs in those areas.  We see a clear flow of wind around the Olympic Mountains and decent convergence at the surface right through the southern part of Puget Sound (the dashed red line).  Also, the lowest pressure in the area (1022.8 mb) is the observation at Shelton, which I circled in orange.  This is pretty close to where we would expect see that lee-side low pressure form.  So--this is a classic convergence zone case.

Furthermore, we can look at the radar radial velocities from this time to see convergence there.

Fig 9 -- KATX 0.5 degree radial base velocities at 0101Z, Feb. 8, 2011.  Arrows showing the rough direction indicated by these colors demonstrate convergence.

The KATX radar is in the northern part of this image.  If we remember that green colors indicate air moving toward the radar and red colors indicate air moving away from the radar, we can see convergence right along that line we were expecting to find it in figure 8.  So the radar velocities also show some convergence there.

It gets more complex though--a second convergence zone seems to be forming across the northern Puget Sound area and the Strait of Juan de Fuca.  Can we have dual convergence zones?  Interesting possibility.  The surface winds don't show as clear of convergence there.  Could this be convective instability released in a direction parallel to a jet over open water (similar to a lake-effect snow band coming off of the Strait of Juan de Fuca)?  Perhaps.  More investigation would be needed.  But that's about all I can cover in one blog without going too long... 

MD for the NW

It's not often at all that you see something like this out here in the Pacific Northwest:

Fig 1 -- Mesoscale Discussion 2128 from the Storm Prediction Center on Tuesday, Dec. 14, 2010.

A mesoscale discussion for this region?  Not only that, but a discussion about an isolated tornado threat for the Puget Sound lowlands down through the Willamette Valley? Very intriguing.  To see the full text of the mesoscale discussion, you can visit the SPC's page for it here.  They note that there probably won't be a watch because the threat is so limited.  However, there already has been a tornado report near Salem, OR.

Right now, there's just some scattered showers moving through the region.

Fig 2 -- 0.5 degree base reflectivity from KRTX radar at 2207Z, Dec. 14, 2010.

Some stronger convective elements are present, particularly in the storm east of Portland.  Some of the showers coming over the Coast Range to the west are also showing stronger cores.  Added orographic lift as the southwesterly winds aloft are forced to rise over the low mountains there seems to be helping to get some of the showers convecting.  A view looking south from the top of the Atmospheric Science building at the University of Washington in Seattle shows the rain to the south (but clear skies over Seattle!).

Fig 3 -- Southward view toward Mount Rainier (though you can't see it) from the roof of the UW Atmospheric Sciences building at 2:14 PM PST.  From the UW Northwest Observations website.

So why this isolated tornado threat for the area?  The big story is the wind shear.  Here's this morning's 12Z sounding out of Salem, OR.

Fig 4 -- 12Z sounding from Salem, OR, on Dec. 14, 2010.  From the SPC website.

Focusing on the winds, it's clear there's a lot of wind shear, both directional and speed-wise, going on in the lowest levels.  Winds go from southerly at 10 knots at the surface to westerly at 40 knots at 850mb.  That's a fairly large amount of wind shear, particularly for this region of the country.  In terms of instability, the lapse rate is conditionally unstable--it's steeper than a moist adiabat but not quite as steep as a dry adiabat.  The absence of any strong temperature inversion also helps make this a tempting sounding for instability.  There's clearly a lot of moisture and with just a little warming at the surface, a fair amount of CAPE could be generated.  There are many ways to get warming at the surface.  One way is through warm air advection--wouldn't you know that the surface to 850mb directional wind shear represents a veering of winds with height.  That's a sign of warm air advection going on.  You can also see in the sounding above that the SPC analysis tools have  picked up on this as well--the vertical bar chart labeled "Inferred Temperature Advection" shows red bars (indicating warm air advection) at low levels with blue bars (indicating cold air advection) in the mid levels.  Warming below and cooling above will tend to destabilize the lapse rate.

Of course another thing that will help warm the surface is clear areas where sunlight can get through.  Here's a look at the latest visible satellite image over Washington.

Fig 5 -- Visible satellite image over Washington state at 2215Z, Dec. 14, 2010.  From the College of DuPage website.

There's still a lot of stratus cloud cover over southwestern Washington and northern Oregon (you could even see that on the webcam image above).  But, Puget Sound is definitely clear.  There's also some isolated patches of clearing on the northern Oregon coast between areas of very convective-looking clouds.  This makes sense as strong convective updrafts usually have compensating downdrafts (or subsidence) that tend to clear the air around the storm.  This clearing lets sunlight get through and can further help to warm things up.

So, perhaps we'll see some more destabilization this afternoon--there's still a few hours of sunlight left.  And with that wind shear, we'll have to watch extra carefully to see if we spot any rotation.

Convection in Seattle

This morning, as I was waiting at my bus stop, I was somewhat startled by a sudden flash and the familiar, but so strangely foreign sound of thunder here in Seattle.  The ensuing downpour was unlike any I have seen in Seattle so far.  The severe thunderstorm warning that followed for Snohomish County to the north was even more unusual.  So I thought I'd grab some images from this event for a quick discussion.

In the synoptic setup, a large, cut-off upper-level low is rotating around in the northeastern Pacific.

Fig 1 -- GOES-W infrared satellite image from 1500Z, Dec 8, 2010.  From the HOOT website.

Note along the northern Oregon coast how you can see little "blips" of clouds firing off in a line extending southwestward from Puget Sound.  Convective clouds like that all forming along a line point to a frontal boundary pushing onshore and providing the lift for a more convective rain.  Here are a few glimpses at this rain as it came through on radar:

Fig 2 -- 0.5 degree base reflectivity from KATX radar at 1556Z, Dec. 8, 2010.

Note the high reflectivity values in the core of this precipitation--red colors showing dBZ values well in the 50dBZ range.  That's some very, very heavy rain--far heavier than you would usually expect in a non-convective system.  Also notice (where I inserted the little purple arrow) how the leading edge of the precipitation area is forming up into a relatively tight band with a strong reflectivity gradient on the leading edge.  This is usually indicative of strong winds being brought down to the lower levels helping to orgnaize the convection.  And that spike shooting out to the southeast?  That's where the radar is pointed directly at the rising sun.  The sun emits energy at all sorts of wavelengths--including some in the 10cm wavelength band that the radar uses.  So, when the sun is low in the sky and the radar scans toward it, we get strong, continuous returns from the direction of the sun.

Now about 20 minutes later:

Fig 3 -- 0.5 degree base reflectivity from KATX radar at 1615Z, Dec. 8, 2010.

Here we can see that that leading line has really gotten together, now evident as a narrow band of high reflectivities stretching down the Cascade foothills.  This almost looks like a sqaull-line structure--a leading edge of deep convection and strong radar returns followed by a region of weak reflectivity and then an area of enhanced trailing stratiform precipitation.  A wind gust of 58 mph in Everett behind that leading line prompted the severe thunderstorm warning for Snohomish County.  Let's take a look at the base velocity image for this time:

Fig 4 -- 0.5 degree base velocity from KATX radar at 1615Z, Dec 8, 2010.

Remember in interpreting radar velocity images that cool colors (the greens) represent velocities toward the radar and warmer colors (the reds) inidicate velocities away from the radar.  In that leading line, particularly in the area within the severe thunderstorm warning, notice the shift in wind directions.  To the east of the line, that tongue of green represens an area of air moving toward the radar.  To the west of the line, the reds indicate air moving away from the radar.  Therefore we see a zone of convergence right underneath our leading line--exactly what we'd expect to find under deep convection.  Convergence near the ground forces air to rise and, if the air is unstable aloft, that rising motion can blossom into thunderstorms.

Also of note in the above image are the strong inbound velocities in the southern blob of the radar returns and the strong outbound velocities in the northern blob. This is the radar sampling a jet at low-levels aloft--remember as the radar beam moves away from the radar, it tends to increase in height above the ground.  So at the southern and northern parts of the domain, the radar beam is relatively high off the ground.  Those high velocities like that point to a strong south-to-north jet over the area.  Do we see this in analyses?

Fig 5 -- UW 12-km WRF model analysis for 12Z run of winds at the 4000 ft. level.  From the UW model page.

The above image repersents a 12Z analysis of winds at the 4000 foot level from the UW-12km WRF model.  There is a nice, south-to-north oriented jet of wind in the 45+ knot range in the central Puget Sound region.  So everyone agrees.  Like I mentioned before, strong winds in the low levels like this can enhance wind shear which helps the convection to stay organized and develop into something severe.

Another observation to look at is a meteogram from the roof of the UW Atmospheric Sciences building:

Fig 6 -- 24-hour meteogram ending at 1919Z from the roof of the UW Atmospheric Science building.  From the UW NW Obs. page.

Our front with its line of convection moved through at around 16Z, and that time immediately stands out on te right side of the meteogram.  We can go through each of the variables individually. Note the wind speeds peaked with sustained winds over 20 knots and a gust measured to near 30 knots right as the leading edge came through.  At the same time, the wind changed direction from being southeasterly to southwesterly.  Very typical wind changes at the leading edge of a convective line.

At the same time, the temperature dropped sharply--from 54 degrees down to around 46 degrees.  Some of this cooling could be due to colder air behind a cold front, but I think a lot of the cooling has to do with convective downdrafts out of this line.  Warm air rising and cold air sinking provide the dynamical mechanism that gives thunderstorms their strength.  Thus, in very strong thunderstorms, we'd expect a lot of sinking cold air in the downdraft region of the storm. 

Now, the plot above plots relative humidity as a moisture variable instead of dewpoint temperature or something more specific.  Relative humidity is a function of the temperature--so as the temperature drops, even with the same amount of moisture in the air we'd expect the relative humidity to rise (we cool closer to saturation).  Therefore it's difficult to tell much from the relative humidity trace--yes, it rises sharply when the line comes through, but that's probably mostly due to the temperature falling rapidly at the same time.  However, with all that rain falling, a case could be made for an increase in the ambient miosture content of the air near the surface...

Finally, we see that the pressure had been falling steadily all night until it jumped higher as the line came through and has been slowly rising since then.  This further supports this line being associated with a cold front, as I've talked about in previous posts how cold fronts tend to lie in local pressure troughs.  Finally, the precipitation accumulation jumped as the rain began to fall--all to be expected.

You can see from the solar radiation plot that much of the morning has been cloudy, but then there was a very sharp transition to rather clear skies once the rain stopped--lots of solar radiation suddenly reaching the surface.  With all the rising motion in the convective line and behind the cold front, compensating subsidence (sinking motion) promoted an area of clear skies and sunny weather for at least part of the morning (though now more clouds seem to be moving in.

So there was a quick look at this morning's convection in Seattle.  Just to confirm that there was thunder, here's Vaisala's map of the National Lightning Detection Network from this morning, showing there was indeed lightning in the Puget Sound area...a magical sight.

Fig 7 -- Lightning strikes from the NLDN for the previous two hours as of 16Z, Dec 8, 2010.  From Vaisala's Public Lightning Data page.