Talk:2005 Atlantic hurricane season statistics

ACE Irony
The wording of the first few lines of ACE section's commentary has bugged me more and more as the season became more impressive. It seems like it's apologetic for the low ACE numbers when the season has climbed up the rankings to 2nd place and possibly 1st place all time ACE (after final reporting). But any edit I would attempt would need to rethink the whole structure of the paragraph. Hopquick 13:38, 1 January 2006 (UTC)
 * The sentence that bugs me is: "The ACE value shows the 2005 season in a slightly less impressive light than the storm counts because only three long-lasting Cape Verde hurricanes - Maria, Irene, and Emily - formed, a lower number than in other seasons of similar activity" -- Perhaps it could be: "The record high ACE value shows the 2005 season in an impressive light because only three long-lasting Cape Verde hurricanes - Maria, Irene, and Emily - formed, a lower number than in other seasons of similar activity..." It just seems like the explaination that 2005 has a low ACE count for the kind of year it's been is almost moot now that it is so close to #1 and will probably be record breaking in spite of only 3 Cape Verde hurricanes. Happy new year.  Hopquick 13:44, 1 January 2006 (UTC)


 * The ACE value plus few long duration hurricanes explanation shows 2005 in an impressive light. However, the ACE value alone, without the few long duration hurricanes explanation, does show the season in a slightly less impressive light. So I think it is more correct as it is. I wouldn't rule out there being a better way to express it though. Happy new year. crandles 16:38, 1 January 2006 (UTC)


 * I added something about possibly passing 1950 despite the low number of long-lasting storms. --AySz88 ^ -  ^  17:46, 1 January 2006 (UTC)


 * The wording of this section, which I have never really liked, seems to be getting more and more strained as the 2005 ACE continues to rise. I think it should be rewritten from scratch.  This is what I would write; what do you guys think?


 * ACE measures the intensity and duration of tropical storms. Generally, the storms with the highest ACE values are Cape Verde type storms, which can remain active for weeks as they travel across the Atlantic without hitting land.  Compared to other high activity seasons, 2005 has seen relatively few Cape Verde systems.  In spite of this lack, 2005 has taken one of the top spots in this catagory due to the record breaking number of storms and the extreme intensity of its strongest hurricanes. --71.142.176.115 03:35, 3 January 2006 (UTC)

I for one like it --Perfection 07:53, 3 January 2006 (UTC)


 * I have to say I agree with the dislikes. I have made a suggested modification which rewords this in terms of ACE/storm and a comparison to previous seasons, which I hope sounds acceptable... Nashikawa 22:06, 4 January 2006 (UTC)


 * Ooops perhaps I should have read the suggested change above rather than quickly scanning the comments before I did the modification.... Apologies if I have been precipitate! I am happy for someone else to edit modify anyway, although I do think the average ACE/storm is a useful indicative tool here. Nashikawa 22:11, 4 January 2006 (UTC)

The ACE record, rounding, and propagation of error
I thought of putting this into the Zeta ACE section but I think it warrants its own headline. Since we're so close to 1950 now and various calculations have come up with differing but approximately equal numbers, I think we really need to look into what is needed to break the 1950 record. What exactly is the most accurate way of depicting an ACE number? We regularly go to 3 significant figures, but no real basis for this. The actual "measured" quantity on which the calculation is based is kts, which is given to 2 s.f.s (45, 55, etc.)  Or is it?. From what I've seen, it's actually given in 5-kt intervals. So a value of 45kt could be anything from 42.51 kt to 47.49 kt! Thus, it is +/- 2.5 kt in either direction. With the lowest value calculated in the ACE as 35kt, that is approximately a 7% error. ACE is calculated by squaring the windspeed, i.e. 35*35, so the error in ACE would be 7%+7%=14%. (An actual windspeed of 37.5 yields an ACE = 0.1406. (0.1406-0.1225)/0.1225= 14%. ).  Now, this error varies depending on the windspeed (it's less of a percentage difference with higher windspeeds). I don't want to clutter the page up with all the numbers, but here's a small sample: At 50kt, the error is 10%, at 75kt, the error is 7%, at 100kt, 5%, at 150kt 3%.

Now, we could go back to each storm in 1950 and 2005 to figure out the exact +/- for each ACE calculation, but I think we could estimate that the ACE number has a variation of about 10%. We could then go into confidence intervals and other statistical formulas, but the simplest way is for the ranges not to overlap. Thus, in order for 2005 to beat 1950, we'd need about a 20% difference, or +48, approximately 290 before we really can say that 2005 beat 1950.

--PK9 21:59, 3 January 2006 (UTC)

NO!!!! While one reading could easily be out by 14% or more, the chances of all reading being out in the same direction is very very low. crandles 23:23, 3 January 2006 (UTC)


 * I have to strongly disagree with this way of thinking. The NHC intentionally releases the storm reports and even the best track data post-season in 5 knot increments for a reason.  Each advisory period used in ACE calculations cover a 6 hour period, during which time a storm's strength is likely to fluctuate considerably!  While you may be able to argue the point that there may be some statistical error involved, it's the sort of thing that is supposed to work itself out by averaging over time.  To put it simply, as far as the NHC is concerned, when we say that Zeta has a speed of 55kt for the last advisory period, the official strength is 55kt, not 55kt +/- 10%.  ACE goes on these official numbers.  Folks may argue with each other until they are blue in the face over how accurate statistics may or may not be, but in the end it's pointless.  We have official figures, and we should use them. Arkyan 23:50, 3 January 2006 (UTC)


 * I agree that the windspeeds fluctuate during the course of the 6-hour period, but that is not the point. The point is the NHC gives an "official" number than is only accurate to +/- 2.5 kt.  One cannot argue that every hurricane has an average windspeed during every 6 hour interval that falls exactly on -0 or -5 kt!.   If the average windspeed as calculated by the NHC is 53.3859 kt then they round it to 55.  To then calculate an ACE value of 0.3025 +/- 0.0001 is mathematically ludicrous.   -PK9 01:04, 4 January 2006 (UTC)


 * In calculations, it is inappropriate to round to significant figures until the final result is reached, and adding numbers allows sigificant figures to increase. There are two sig figs from the wind speed (the second digit is the one-digit estimate and therefore is a sig fig).  The season ACE is made by adding many numbers (and all have the ones place as a sig fig or left-of-a-sig-fig), so the season ACE has 3 sig figs - the three to the left of the decimal.
 * As long as 2005 gets one ACE more than 1950, I think we're allowed to say (with the last digit as an estimate) that 2005 passed 1950, and it follows the sig fig rules. --AySz88 ^ -  ^  02:44, 4 January 2006 (UTC)


 * Okay. I'll avoid actually approaching this from a point of statistics and mathematics and just try to state my point more simply.  ACE is by definition arbitrary.  It does not include figures for storms below 35 knots, yet it is unreasonable to assume those tropical depressions were not contributing to the total energy for the season (they are tropical cyclones and they are obviously expending energy yet are not included in Accumulated Cyclone Energy calculations).  This is an arbitrary decision.  The NHC chooses to release figures in 5 knot increments because it is felt that the differing methods of maximum sustained winspeed estimation can vary, and an accurate example can never be found, and the 5 knot interval is their compromise.  This is an arbitrary decision.  Your choice of a 20% minimum difference in ACE values to declare the season the new recordholder is also an arbitrary decision.  Again, in the end the numbers can be nitpicked and critisized, and one may make a convincing argument that by some calculation year X actually had more tropical cyclone energy than year Y.  However, the ACE values are official and even if by some definition or estimation 1950 may have actually been more energetic, 2005 will have had more ACE (by the next advisory) and thus will be the irrefutable recordholder for highest ACE, barring a sudden and total demise of Zeta. Arkyan 02:49, 4 January 2006 (UTC)


 * A couple of refuting comments. First, AySz88, the wind speed does not truly have 2 s.f.s, because average 6-hour wind speeds of 46 or 73 kt are not possible.  Secondly, ACE is calculated by squaring the wind speed measurement (a multiplication) so the number of s.f.s would need to be conserved (even giving you windspeed to 2 s.f.s, the ACE for that interval should only be reported to 2 s.f.s).   After that, you can start keeping decimal places for adding.  Second point, Arkyan, yes, ACE is arbitrary, but my propagation of error calculation was not my arbitrary decision; it's a result of the NHC's arbitrary decision.  Obviously, it's not realistic to expect a 48+ difference to declare something the record holder, but statistics will tell you that right now, 1950 has close to a 50% chance of having a true ACE higher than that of 2005.  That's where confidence intervals come in; scientifically its generally preferred to have a 90% or a 95% chance of the true values being different to stay there is a "statistically significant" difference. --PK9 20:31, 4 January 2006 (UTC)


 * Hm. I think the ones digit of the wind speed, the 'estimate' digit, is still a sig fig, even if it's rounded to the nearest 5 knots (that's allowed since it's an estimate).  That's 2 sig figs (3 if it's 100+).  When you square that number, as long as it's less than 315 knots (ow), the ones tenths place value is still on the sig fig side for every 6-hour interval of every storm.  Then one sums the 6-hour intervals, resulting in a number that has sig figs out to at least the ones tenths place.  If you do that for each individual storm, you would round, but we're not stopping there.  The values are all summed and that produces a 3 4-sig-fig number (hundreds, tens, ones, tenths).
 * Anyway, I'm thinking that the NHC won't really be picky about being more than 50% confident that 2005 beat 1950, and it definitely isn't going to matter in the ranking by sig figs. --AySz88 ^ -  ^  02:07, 5 January 2006 (UTC)

Much of this discussion has been on the number of significant figures. I'd like to go back to the subject of propagation of error. While it is correct that any single ACE value can have a relative error up to ~14% (0.1255±0.0175 for the lowest wind speed), the propogation of these errors after adding up an entire season (i.e., hundreds of values) is small (<1%). When adding the single ACE numbers, the absolute error grows much slower than the sum, and the relative error for an entire season is certainly much less than the 10% proposed above. My rough calculations for the 2004 ACE is ±0.33%, and for the 2003 ACE is ±0.57%. --Spiffy sperry 22:48, 5 January 2006 (UTC) -- calcs edited on 21:34, 25 January 2006 (UTC)


 * You're confusing propagation of error with variance/standard deviation. Variance is dependent on sample size, decreasing with a larger sample, so you are correct in saying the season ACE would have a very small standard deviation.  Propagation of error is dedicated to finding the range of possible values, so that only increases with each calculation.  Basically, as stated before, over the course of the season the error will average out and the value will most likely be accurate.  However, because each single ACE number is +/- 14%, it is theoretically possible for every value to be +14%, which is why the range becomes so big.  It is very unlikely that that would be the case, which is where confidence intervals come in.  Statistically, you determine a range of numbers in which you are 90% or 95% sure that the true value falls within. --PK9 07:00, 9 January 2006 (UTC)


 * Sorry, but I’m not confusing anything. Frankly, the concept of variance/standard deviation is not particularly helpful in this context (which is, I assume, the accuracy of the value of ACE for an entire season).  One could say that the st. dev. of ACE for the 2004 storms is 20.1 or the st. dev. of each 6-hr ACE value is 0.5.  But this only says something about the dispersion of the data (0.8-70 for the storms, 0.1225-2.125 for the 6-hr values).  It says nothing of the error in the sum of data for all storms.  You are correct in stating that prop. of error increases with each calculation, but my point is that it increases very slowly with respect to the sum.  The theoretical possibility for every value having a large possitive error is simply unrealistic for independent variables and not contemplated in prop. of error calculations.  To suggest that the 1950 ACE was 243 ± 24 is simply wrong.  For what it's worth, I think it's ± 0.8.  (Disclaimer: All values in this post and my Jan. 5 post were calculated using accepted mathematical formulas.  They are not estimations.) --Spiffy sperry 23:03, 11 January 2006 (UTC)


 * All right, I guess the best way is to go through an example. I took Epsilon's ACE table since it was conveniently at the top of the page and put it into Excel.  As calculated, the ACE was 13.4275.  This is assuming the average windspeed for each advisory fell exactly as reported on the ACE table.  I then made a theoretical second storm whose average windspeed was exactly 2.49 kt less than Epsilon's at every advisory, i.e., the first advisory would have had an actual windspeed of 37.51 kt, the second 42.51 kt, etc.  Since the windspeed is reported to the nearest 5kt, it would appear identical to Epsilon (40, 45, etc).  However, the actual ACE for this storm, using the "actual" windspeeds, is 12.37.  I did the same thing with a storm whose average windspeed was 2.49 kt higher than Epsilon, resulting in an ACE of 14.53.  What does this mean?  12.37-14.53 is the range of possible "actual" ACE values for Epsilon, officially reported as 13.4275.  This is +/- 8% (or about 1.1), due to the fact that Epsilon spent much of its life at 65kt (ACE +7.8%/-7.5%).  Add this up for all 27 storms, and we have a +/- of around 20.  (I think I overestimated the error at +/- 10%, it's probably closer to 7% given Epsilon's calculations).  --PK9 23:45, 17 January 2006 (UTC)


 * I'm pretty sure that ± values are given to 95% confidence, not 100% confidence. I can say with about (100-2-37)% confidence that not all the values for Epsilon are too low, for example.  I'm not a student of stat, but I'll give this a go using just my mathlete skills:
 * Let's say the average error per advisory is 1.25 kt and there have been 270 advisories. The total number of ways the error can be distributed is 2^270 permuatations.  The number of ways to distribute x too-highs and 270-x too-lows is 270Cx.  I think we want to add together 270Cx starting with x = 135 and going a step at a time (134, 133, 132...), doubling everything after 135 to account for the reverse situation, until we can include 95% of 2^270, which is about 1.8*1081.
 * So I think we need to find maximum x for which $$._{270}C_{135} + 2 \sum_{i=x}^{135} ._{270}C_i < .95 (2^{270})$$
 * It happens to be about 123, so with 95% confidence 147-123= 24 net advisories out of 270 are erroneous by an average of 1.25 knots. If this was right, the error for the entire season is (with 95% probability) less than what you calculated to be the maximum error for Epsilon alone.
 * I reserve the right to have entirely screwed up this calculation. --AySz88 ^ -  ^  03:08, 18 January 2006 (UTC)

I find it kind of unsettling that the records are stated clearly but the actual total ACE figure is not even listed in the section, just in the table. That needs to be fixed. -- Hurricane Eric archive -- my dropsonde 03:47, 7 January 2006 (UTC)

Link to ACE
I suggest that the link to the ACE article be from the paragraph that follows the ACE section, rather than the section's title itself. This appears to me as a more elegant way of linking to articles, and perhaps it is already a wiki convention (or else, I'd argue it should be). --BenG 23:35, 3 January 2006 (UTC)


 * Just to clarify, I meant the link should be from the first paragraph of the ACE section. --BenG 23:38, 3 January 2006 (UTC)
 * It is a style guideline already, so I'll go ahead and switch it. Tito xd (?!? - help us) 23:44, 3 January 2006 (UTC)


 * Thanks. --BenG 23:47, 3 January 2006 (UTC)

Which record has been broken by the biggest margin?
3 Category 5 hurricanes versus the previous record of 2 is a case of beating the previous record by 50%. Pretty impressive!

But are there other records that have been beaten by a higher percentage? I am wondering about the number of 6 hourly tropical advisaries after the season has ended. There is 29 or 30 for Epsilon (does the season end just after 30 Nov 23:59:59 UTC time?) and 26 or 27 for Zeta. What was the record before the 2005 season? crandles 15:45, 6 January 2006 (UTC)


 * Lili had 47. So that record is only beaten by 17-21%. crandles 15:54, 6 January 2006 (UTC)


 * Costiest hurricane/season $75Bn vs S43.7Bn = beaten record by ~72%. I missed that, guess I was partly thinking about whether the record provide any significant evidence of a global warming impact and for that you don't want a noisy signal. crandles 16:17, 6 January 2006 (UTC)


 * $75bn+ is just Katrina. Wilma was $14.4bn, Rita $9.4bn, Dennis "$4-6bn" and Ophelia $1.6bn, so the season is over $100bn. - Cuivienen 23:04, 6 January 2006 (UTC)


 * So the cost record is beaten by 55 billion, and the storms record is beaten by 6. 55 billion > 6. Jdorje 23:23, 6 January 2006 (UTC)


 * It's an apple/oranges kinda argument there. 70.177.68.209 00:02, 9 January 2006 (UTC)


 * You have to convert to percentage because only then are the units comparable. You can't compare dollars to hurricanes to kt^2/1000. -PK9 00:42, 10 January 2006 (UTC)


 * Another possibility that might be more interesting is the number of standard deviations further from the norm the current record is compared to the previous record. This might deal with how noisy the information is. Unfortunately, I don't think I want to do all the calculations. crandles 16:21, 11 January 2006 (UTC)


 * I was joking. However, my point is that even converting to percentage does not allow comparisons across units, because not all values are counted incrementally.  Is a decrease of 6 millibars for maximum storm strength a difference of 0.675% or 4.8% or 50% or 150%?  Does making a percentage out of such a value even make sense? Jdorje 06:22, 13 January 2006 (UTC)


 * For the Tropical Advisories-79/87 for Alberto (2000).HurricaneCraze32 17:16, 11 June 2006 (UTC)

Metric
The tables in this article are incomplete without metric (especially the big table of storms). Jdorje 17:27, 17 January 2006 (UTC)


 * I counsel against that for now. There is no room for more data in that table as-is.  Perhaps if the storm strength column got set back to just being a number or "TS" and "TD", that would help.  (If there is a consensus for that change I will gladly do it BTW.)  Another idea is to split the table into a storms/maximum-intensity table and a storms/landfall-and-damage table.  This second idea is viable because of the templating currently used to construct the tables.  However, adding related representations of the data (such as knots and km/hr for wind speeds; and in. Hg for pressure) will be grunt work.  In any case, a consensus for doing such a revision and of what if should look like would be appreciated. --EMS | Talk 17:56, 25 January 2006 (UTC)


 * Metric is essential. If that means the table has to be split, then so be it. Jdorje 17:59, 25 January 2006 (UTC)

Gamma's ACE
Since Gamma briefly reached tropical storm strength south of Hispaniola, shouldn't it have a higher ACE value?Icelandic Hurricane 22:06, 24 January 2006 (UTC)
 * It was already adjusted upwards from 1.05 to 1.33. --AySz88 ^ -  ^  01:07, 25 January 2006 (UTC)

Oh. I didn't notice it was 1.05 ACE before.Icelandic Hurricane 21:44, 25 January 2006 (UTC)

Wilma's ACE
Now that I'm confident in my ACE calculations, I'm not sure we have the right value for Wilma. After the final report was issued on Jan. 12, the value was changed from 38.6 to 38.4, but I calculate 38.9. Can someone verify this? --Spiffy sperry 22:09, 25 January 2006 (UTC)
 * My bad, fixing... --Ajm81 23:16, 25 January 2006 (UTC)
 * I think we are still wrong on Wilma. Oct 17 0600Z (reaches TS strength) through to Oct 26 0000Z (Becomes extratropical). Sustained wind readings are:

(TD), 35, 40, 45, 55, 60, 65, 75, 130, 150, 160, 140, 135, 130, 130, 130, 130, 130, 125, 120, 120, 110, 100, 85, 85, 85, 85, 90, 95, 110, 100, 105, 110, 100, 90, 75 (next reading is 60, but extratropical) I make that an ACE of 39.015. Page updated, but I'd be grateful if someone checked the numbers.FrankDynan (talk) 01:08, 21 September 2008 (UTC)

Cindy's ACE
I was wondering if Cindy's ACE was updated considering the status change? User:tdwuhs:tdwhs
 * Yes, it was updated. You can find out yourself by checking the page history of the article, usually it'll be noted in an edit summary. :) --AySz88 ^  -  ^  03:30, 27 January 2006 (UTC)

TD ACE?
Technically, don't Tropical Deppressions have an ACE value? Why don't they include it?Icelandic Hurricane 23:18, 2 February 2006 (UTC)


 * No, depressions do not have an ACE value. &mdash; jdorje (talk) 23:49, 2 February 2006 (UTC)


 * But couldn't you calculate it?Icelandic Hurricane 12:57, 3 February 2006 (UTC)


 * It is 0. &mdash; jdorje (talk) 18:44, 3 February 2006 (UTC)


 * Look at the definition of ACE. Only tropical cyclones (i.e., tropical systems of storm strength or higher) get a value. &mdash; jdorje (talk) 18:45, 3 February 2006 (UTC)


 * Oh.Icelandic Hurricane 20:38, 3 February 2006 (UTC)

Stan
I fixed the Stan entry to reflect the correct (pre-TCR) values of unknown damages and 80-100 deaths. Any other values are either speculation or counts of the impact of the entire weather system, not of stan itself. While the stan article does mention those totals, it is not correct to include them in the season totals. &mdash; jdorje (talk) 19:20, 3 February 2006 (UTC)


 * We should wait for the TCR before anything is changed. CrazyC83 00:03, 4 February 2006 (UTC)


 * Has there been any other information other than the year-end summary? As someone noted in one of the above sections, the year-end summary doesn't say whether the Stan caused a thousand+ deaths indirectly, only that the NHC is counting 80-100 direct deaths for Stan. The NHC apparently never counts indirect deaths. --AySz88 ^  -  ^  21:19, 4 February 2006 (UTC)


 * Nothing official. There are some unofficial sources: Jeff Masters wrote in his blog that he talked to a Guatemala official, who said there was $1-2 billion in damages, but that's almost certainly including the non-tropical rains.  I know of nothing new from the NHC. &mdash; jdorje (talk) 21:48, 4 February 2006 (UTC)

Individual storms table
The Individual storms table needs references for the death and damage totals for each storm, particularly those that do not match what is given in the TCR. --Ajm81 00:18, 16 March 2006 (UTC)

Table Rewrite
I've begun to reorganise and redo the table. This is what I have currently; I've done enough to make it obvious what the rest of the table would look like. Comments? —Cuivi é nen 01:15, 16 June 2006 (UTC)
 * Awesome. TimL 01:38, 16 June 2006 (UTC)


 * One of the problems I have with the current table is that it has too much color. Surely we only need the name in color (peak intensity), and the name of the landfall location in color (intensity at landfall)? Tito xd (?!?) 01:39, 16 June 2006 (UTC)
 * Or just the landfall windspeed column. TimL 01:48, 16 June 2006 (UTC)
 * Well, then it looks inconsistent to not color the max windspeed column. I could agree to Titoxd's idea of coloring only the landfall locations and the name, though remember that this version has already cut down the amount of color greatly. —Cuivi é nen 23:43, 16 June 2006 (UTC)
 * Also, I'm considering removing ACE from the table as the ACE table covers it better, and I would prefer not to introduce trivia such as ACE to a table intended to be a summary. Removing ACE also gives room for km/h wind conversions in the table. —Cuivi é nen 23:45, 16 June 2006 (UTC)
 * Well perhaps just color both of the max windspeed columns? (and not the name). TimL 00:27, 17 June 2006 (UTC)

New Format
Removing the ACE and adding km/h conversions to the above table gives this format. I've done only the header:

1Because of the devastating nature of Hurricane Katrina, assessment of the death toll is ongoing. Therefore, a distinction is not being drawn between direct and indirect deaths. It is estimated that approximately two-thirds of deaths caused by Katrina were indirect, resulting from poor conditions in New Orleans after the disaster.

Is it an improvement? —Cuivi é nen 23:57, 16 June 2006 (UTC)


 * Couple comments...
 * I agree, ACE isn't needed in this table. It is just redundant to the ACE section.
 * How about inHg for the pressure as well? (Since we have mph and km/h).
 * Some kind of border for the cells would ease readability, especially for the landfalls area.
 * Splitting up deaths may not always be possible.
 * When the table is redone, proper references should be added. This would be most important for figures updated since the TCRs, such as deaths and possibly damage.
 * --Ajm81 00:29, 17 June 2006 (UTC)


 * I am hesitant to add inHg because, unlike mph, it is hardly ever used. Personally, I prefer the borderless style, and I don't see there being any difficulty telling which landfall belongs to which storm. (Plus, borders are always ugly.) Splitting up deaths is possible for all storms except Katrina, for which I thought all would be listed across both columns using colspan="2" and a note made at the bottom of the table. References will of course come later. —Cuivi é nen 03:21, 17 June 2006 (UTC)

Arlene is now included to give an idea of what this format looks like. —Cuivi é nen 02:12, 18 June 2006 (UTC)
 * Minor thing: Date needs wikification due to date preferences for individal users. Therefore, I'd spell out date active in whole " June 8 – June 13 ", which would produce, depending on preferences, "June 8 – June 13" or "8 June – 13 June"... NSLE 02:15, 18 June 2006 (UTC)
 * Done. —Cuivi é nen 13:47, 18 June 2006 (UTC)


 * One suggestion: try and work out a way to get the damage expressed in "then" and "now" USD. For 2005 obviously there isn't a significant advantage in having 2005 and 2006 damages expressed, but it will become useful as time goes on (and if we expand this format to other seasons).--Nilfanion (talk) 14:09, 18 June 2006 (UTC)


 * Well, you can always just split the column into, say, 2005 USD and 2010 USD the way the death toll is split. At this point, though, that's not useful. —Cuivi é nen 14:30, 18 June 2006 (UTC)
 * That works for me - though I think a clarification that the damage figures are in millions 2005 USD is sensible.--Nilfanion (talk) 14:58, 18 June 2006 (UTC)

Nitpcik Alert! :) I would like to see a border (line) where the landfall colums start, I think it would help guide the eye around the table and mark off the two really distinct areas of onformation being conveyed about each storm (overall statistics vs landfalling statisitics). TimL 14:45, 18 June 2006 (UTC)

Regarding the date preferences, I tested it and they don't seem to kick in for just the day and the month, so given as the old way was more compact I'd like to see it go back to that. TimL 14:51, 18 June 2006 (UTC)
 * Do you not have date preferences set? I think it's supposed to work with just dates. —Cuivi é nen 14:55, 18 June 2006 (UTC)


 * It does work; you have to remember to bypass your cache after changing your prefs.--Nilfanion (talk) 14:58, 18 June 2006 (UTC)
 * Well it only works partilly. I have mine set to the 5th option and it has no effect. TimL 15:06, 18 June 2006 (UTC)

Complete! I'll insert it into the main article if no one objects. —Cuivi é nen 21:49, 18 June 2006 (UTC)


 * Do we still want a summary section like the old table? (With highest winds, lowest pressure, and death/damage totals) --Ajm81 00:37, 19 June 2006 (UTC)
 * I think a summary table should be separate if included at all. (After all, all of that information is in the 2005 Atlantic hurricane season infobox except ACE, which isn't in the new table.) I've added the new new format of the table to the article. —Cuivi é nen 14:24, 19 June 2006 (UTC)


 * 2 things:
 * 1)Can you put the border lines back-its very hard to read.
 * 2)Hg is a good idea.HurricaneCraze32 00:41, 23 June 2006 (UTC)


 * inHG is not a good thing. inHg is only "official" in the US, and even US weather authorities use mbar over inHg. Even The Weather Channel mentions inHg only tangentially. (And, unlike measurements such as miles, inHg is not a commonly understood measurement in the US anyway.) To insert borders at only some parts of the (which is what is necessary to make it easier to read) is beyond my skill. Someone else will have to do it. —Cuivi é nen 15:54, 25 June 2006 (UTC)

Unnamed Subtropical Storm
Shouldn't the Unnamed subtropical storm have an ACE? Nicole did in 2004.--Lionheart Omega 03:47, 20 July 2006 (UTC)
 * Subtropical time doesn't count towards ACE. See this discussion. --Ajm81 04:16, 20 July 2006 (UTC)
 * Well, at this site, Sub-Tropical Storm Nicole is counted toward the ACE total. Also,here, the time storms were sub-tropical was also counted toward ACE. A response please.--Lionheart Omega 01:58, 21 July 2006 (UTC)
 * The way I understand it, subtropical time is excluded by the definition of ACE. Some basins do not clearly/reliably differentiate between tropical and subtropical, so the NCDC includes subtopical values when comparing ACE across basins. User:Spiffy sperry could probably explain it better. --Ajm81 19:15, 21 July 2006 (UTC)
 * I still say we should do it the way it is shown on the website. I'll see what Hurricanehink thinks.--Lionheart Omega 23:21, 21 July 2006 (UTC)
 * The message by the NCDC is clear, subtropical ACE is not part of ACE per se, so no reason to add it. Tito xd (?!?) 02:04, 22 July 2006 (UTC)
 * Hmm, this is a bit unusual. There are reasons to support either way. Maybe someone should just email them and ask? Hurricanehink ( talk ) 02:35, 22 July 2006 (UTC)
 * Can you show me this message Titoxd?--Lionheart Omega 21:31, 22 July 2006 (UTC)
 * I'm back! I've been busy, but I can now put an ACE count with tropical & subtropical time down, since no on4e gave me proof onethe NCDC policy. Unrelated, you can put 30 knots into the ACE equation and get .3. --Lionheart Omega 17:10, 24 December 2006 (UTC)
 * From the NCDC: Note that the official NOAA definition of the ACE Index does not include those periods when the system is characterized as sub-tropical. However, we are including this information to allow for comparison between individual cyclones across all basins, many of which contain records which inadequately distinguish between tropical and subtropical cyclones. --Ajm81 02:53, 25 December 2006 (UTC)

Ambiguous Sentence
"Katrina, Rita and Wilma are the three most intense storms ever in a single Atlantic hurricane season."

Could someone please clarify that statement for me? As I read it, it's not true. Allen was stronger than Katrina and the Labor Day Storm and Gilbert were stronger than both Katrina or Rita so I'm lost. -- § Hurricane E  RIC § archive 04:39, 27 May 2007 (UTC)

Dead link
During several automated bot runs the following external link was found to be unavailable. Please check if the link is in fact down and fix or remove it in that case!


 * http://www.weathermatrix.net/tropical/2005records.htm
 * In 2005 Atlantic hurricane season statistics on 2011-05-25 07:55:18, 404 Not Found
 * In 2005 Atlantic hurricane season statistics on 2011-06-12 00:28:17, 404 Not Found

--JeffGBot (talk) 00:28, 12 June 2011 (UTC)

Orphaned references in 2005 Atlantic hurricane season statistics
I check pages listed in Category:Pages with incorrect ref formatting to try to fix reference errors. One of the things I do is look for content for orphaned references in wikilinked articles. I have found content for some of 2005 Atlantic hurricane season statistics's orphans, the problem is that I found more than one version. I can't determine which (if any) is correct for this article, so I am asking for a sentient editor to look it over and copy the correct ref content into this article.

Reference named "costliest": From 2004 Atlantic hurricane season:  From 1998 Atlantic hurricane season:  

I apologize if any of the above are effectively identical; I am just a simple computer program, so I can't determine whether minor differences are significant or not. AnomieBOT ⚡ 22:13, 1 September 2014 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 3 external links on 2005 Atlantic hurricane season statistics. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Corrected formatting/usage for http://www.ncdc.noaa.gov/oa/climate/research/2005/hurricanes05.html
 * Added archive https://web.archive.org/web/20110607002113/http://www.usno.navy.mil/NOOC/nmfc-ph/RSS/jtwc/atcr/1983atcr/pdf/wnp/11.pdf to http://www.usno.navy.mil/NOOC/nmfc-ph/RSS/jtwc/atcr/1983atcr/pdf/wnp/11.pdf
 * Added archive https://web.archive.org/web/20060318022305/http://www.nhc.noaa.gov/pdf/TCR-AL052005_Emily.pdf to http:///

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 06:22, 22 May 2017 (UTC)

External links modified
Hello fellow Wikipedians,

I have just modified 3 external links on 2005 Atlantic hurricane season statistics. Please take a moment to review my edit. If you have any questions, or need the bot to ignore the links, or the page altogether, please visit this simple FaQ for additional information. I made the following changes:
 * Added archive https://web.archive.org/web/20060524032544/http://www.nhc.noaa.gov/pdf/TCR-AL122005_Katrina.pdf to http:///
 * Added archive https://web.archive.org/web/20060524183736/http://www.nhc.noaa.gov/pdf/TCR-AL182005_Rita.pdf to http:///
 * Added archive https://web.archive.org/web/20060524183813/http://www.nhc.noaa.gov/pdf/TCR-AL252005_Wilma.pdf to http:///

When you have finished reviewing my changes, you may follow the instructions on the template below to fix any issues with the URLs.

Cheers.— InternetArchiveBot  (Report bug) 12:42, 17 June 2017 (UTC)