Jason Persoff
EF3
I think the SPC has one of the cooler graphics around about this very issue.
Now what this graph shows are some interesting inflection points. An inflection point is a point in the graph where the data stop being "smooth" and move up or down abruptly.
Looking at the yellow (2004), there is an undeniable change in the slope of the curve starting mid-May with the curve then resuming a normal rise starting the first week of June. A second point is clear in early September (the so-called second season). But I would advise some degree of caution in this regard: autumn, '04, was--as anyone with a pulse will recall--the busiest year for hurricanes in recent memory. The second season spike could be explained by the higher-than-average numbers of tornadoes brought about by land-falling hurricanes. By most people's recollections, myself included, 2004 was a more "typical" year. However, this graph shows that last year was one of the busiest tornado years on record.
Looking at this year (red line), there is no doubt that the season was slow in May compared to 10-30 year averages. And, as anyone who was out this May will confirm, in fact there was virtually no inflection point until the beginning of June. Note that the June inflection point is very steep and actually has now brought 2005 in line with the 10-30 year averages. Surprisingly, these data show that this year is a more "typical" year in terms of tornado number, though the inflection point was shifted rightward into June.
In 2003 (blue), I chased the last two weeks of May. You can see exactly when I arrived on the Plains--there is a plateau in tornado number--and when I left--in June there is a sudden new inflection point. This confirms that my frustration from the timing of my trip in '03 wasn't just because I chased the wrong storms: instead there were definitely fewer storms during my trip
.
Now, look at the 10 and 30 year averages (harder to see as easily-->look for the faint gray and faint blue lines). They show a much more smooth line structure (as you'd imagine--far more many points to make up the line), but show--very clearly--that there is an inflection point which shows a brisk uptake in tornado number starting on 5/1 and spanning until 7/1 before returning to its normal asymptote. Also note that the "second season" does not appear to be as noticeable on the longer term records.
Based on the longer term data, there is no doubt the highest spike in tornado activity occurs starting in early May and begins to wane by late June. These graphs are not location specific, and since we're looking at nationwide reports, many of these tornadoes may not have occurred exclusively in the Alley.
I think the important thing to those of us who can only chase during a single window period every year is that we be out there when the highest numbers of tornadoes are likely to be out and about. Per these data, our "typical" chase season of May/June is perfect. And I seem to remember Rich Thompson (I think) compiling an article about this in the old StormTrack magazine (when it was in print!).
The reason I am pointing this out is that this year was tough for most of us who came during the lull before a spectacular June. That has forced a lot of us to rethink our vacation timing. But after looking at the data, all things being equal, most of us should probably continue to take our vacations at the same time we usually do--whether you prefer early May, late May, or early-mid June. The data are clear: if you're out there enough, it will always end up averaging out.
And I for one, have no intention ever of giving up this hobby. Aye, Nature, she's a frustrating friend, but I always saddle back up for more. And next year is bound to return the curve back to the middle. Keep the faith
Jason
Now what this graph shows are some interesting inflection points. An inflection point is a point in the graph where the data stop being "smooth" and move up or down abruptly.
Looking at the yellow (2004), there is an undeniable change in the slope of the curve starting mid-May with the curve then resuming a normal rise starting the first week of June. A second point is clear in early September (the so-called second season). But I would advise some degree of caution in this regard: autumn, '04, was--as anyone with a pulse will recall--the busiest year for hurricanes in recent memory. The second season spike could be explained by the higher-than-average numbers of tornadoes brought about by land-falling hurricanes. By most people's recollections, myself included, 2004 was a more "typical" year. However, this graph shows that last year was one of the busiest tornado years on record.
Looking at this year (red line), there is no doubt that the season was slow in May compared to 10-30 year averages. And, as anyone who was out this May will confirm, in fact there was virtually no inflection point until the beginning of June. Note that the June inflection point is very steep and actually has now brought 2005 in line with the 10-30 year averages. Surprisingly, these data show that this year is a more "typical" year in terms of tornado number, though the inflection point was shifted rightward into June.
In 2003 (blue), I chased the last two weeks of May. You can see exactly when I arrived on the Plains--there is a plateau in tornado number--and when I left--in June there is a sudden new inflection point. This confirms that my frustration from the timing of my trip in '03 wasn't just because I chased the wrong storms: instead there were definitely fewer storms during my trip
.Now, look at the 10 and 30 year averages (harder to see as easily-->look for the faint gray and faint blue lines). They show a much more smooth line structure (as you'd imagine--far more many points to make up the line), but show--very clearly--that there is an inflection point which shows a brisk uptake in tornado number starting on 5/1 and spanning until 7/1 before returning to its normal asymptote. Also note that the "second season" does not appear to be as noticeable on the longer term records.
Based on the longer term data, there is no doubt the highest spike in tornado activity occurs starting in early May and begins to wane by late June. These graphs are not location specific, and since we're looking at nationwide reports, many of these tornadoes may not have occurred exclusively in the Alley.
I think the important thing to those of us who can only chase during a single window period every year is that we be out there when the highest numbers of tornadoes are likely to be out and about. Per these data, our "typical" chase season of May/June is perfect. And I seem to remember Rich Thompson (I think) compiling an article about this in the old StormTrack magazine (when it was in print!).
The reason I am pointing this out is that this year was tough for most of us who came during the lull before a spectacular June. That has forced a lot of us to rethink our vacation timing. But after looking at the data, all things being equal, most of us should probably continue to take our vacations at the same time we usually do--whether you prefer early May, late May, or early-mid June. The data are clear: if you're out there enough, it will always end up averaging out.
And I for one, have no intention ever of giving up this hobby. Aye, Nature, she's a frustrating friend, but I always saddle back up for more. And next year is bound to return the curve back to the middle. Keep the faith
Jason
