A Combined Wind and Pressure TC Intensity Calculation Method

[aspen]

A few years ago, I decided to create a little formula to give a measure of intensity of a tropical cyclone that uses both maximum wind speeds and minimum central pressure. I was inspired by Irma and Maria, where the former had a lower pressure (908 mbar), while the latter had the higher winds (originally 160 kt before being readjusted to 155 kt in post-season analysis). Since then, I’ve been using the formula for every Atlantic and EPac system, as well as to rank the strongest systems of the year worldwide and generate an overall score for each season, similar to ACE.

The formula is (P x W)/100. W is max winds in mph, and P is 1010 minus the minimum pressure in mbar. The reason I chose 1010 is because it’s a nice interval of 5 number, and very few storms are weak enough to peak at 1010 mbar or higher. Those would technically have a negative score, but I count them as 0. Also, the whole formula is divided by 100 so the resulting score isn’t often in the range of 500-20,000 per storm.

As mentioned before, this can function similar to ACE. However, while ACE is an accumulative score, this is an instantaneous score of the system’s peak intensity, no matter how long it remained at that peak or how long the storm lasted in general. A combined score for the season does give an indicator of how active it was, just like ACE. The 1977-2017 average combined seasonal score for the Atlantic basin is 406.67 units. An average season is between 85% and 115% of this mean value (345.7 to 467.7 units), above average is up to 150% (610 units), hyperactive is above 150%, and below average is less than 85%. Also, I include “medicanes” and South Atlantic TCs in the Atlantic’s total score, because there’s nowhere else to put them and they do contribute a little bit.

EXAMPLES
—Atlantic 2016 total score: 432.5 units (106% average)
—Atlantic 2017 total score: 865.45 units (212% average)
—Atlantic 2018 total score: 525.9 units (129% average)
—Atlantic 2019 total score: 558.2 units (137% average)

While these don’t match up exactly to the ACE values for each season, when compared to the number of named storms and hurricanes, it can reveal how much of various types of storms formed. For example, 2016 has a lower total intensity score because it had plenty of weak systems, while 2017’s is so high due to a near-record amount of major hurricanes and Cat 5s.

The individual intensity scores can be used as a good ranking of the strongest systems per period of time or per basin. Below are two lists: the top ten strongest in the Atlantic overall, and the top ten in the last decade.

1.) Wilma ‘05 (236.8 units)
2.) Gilbert ‘88 (225.7 units)
3.) Labor Day (218.3 units)
4.) Allen (210.9 units)
5.) Rita ‘05 (207 units)
6.) Camille ‘69 (192.5 units)
7.) Katrina ‘05 and Mitch ‘98 (189 units)
8.) Dorian ‘19 (185 units)
9.) Dean ‘07 (183.75 units)
10.) María ‘17 (178.5 units)

1.) Dorian ‘19 (185 units)
2.) María ‘17 (178.5 units)
3.) Irma ‘17 (172.8 units)
4.) Michael ‘18 (145.6 units)
5.) Lorenzo ‘19 (136 units)
6.) Igor ‘10 (133.3 units)
7.) Matthew ‘16 (125.4 units)
8.) Joaquin ‘15 (122.45 units)
9.) Jose ‘17 (111.6 units)
10.) Florence ‘18 (109.5 units)

[1900hurricane]

Just out of curiosity, why mi/hr instead of kt? Just personal preference?

Also spoiler alert, Patrica '15 wins the global overall.

[aspen]

Just out of curiosity, why mi/hr instead of kt? Just personal preference?

Also spoiler alert, Patrica '15 wins the global overall.

Yeah, it’s just a personal preference. You could use kt and while all the totals will be slightly lower, it doesn’t change the functions of this formula.

Patricia is 296.7 units, nearly twice as high as the entire 2013 Atlantic season, which was only 128.8 units. If you believe the 858 mbar estimate for Haiyan, it comes in as a close second at 296.4 units.

[1900hurricane]

Not sure if it is your intention or not, but wind ends up with a far greater weight than pressure. With Patricia '15 for example, the wind of 215 mi/hr far outweighs the contribution possible by the 138 mb pressure departure from 1010 mb.

[aspen]

Not sure if it is your intention or not, but wind ends up with a far greater weight than pressure. With Patricia '15 for example, the wind of 215 mi/hr far outweighs the contribution possible by the 138 mb pressure departure from 1010 mb.

That was unintentional; I didn’t really realize it until you pointed it out.

Maybe I should revise the entire formula to work with winds in kt and give both winds and pressure equal weight. Maybe I should set 25 kt/30 mph as the lower wind limit.

[aspen]

I have an idea for an updated formula: P will still be 1010 minus the minimum central pressure, but W would be the wind speed in kt minus 25 kt (30 mph), which is the lowest wind speed used for a TC. The whole formula will also be divided by 75 instead of 100. These changes give wind and pressure almost equal weight and only slightly change the resulting scores.

For example, Dorian sits at 185 units with a pressure of 910 mbar and 185 mph. Using 910 mbar and 160 kt and putting them in the new formula, its score is only slightly reduced to 180 units. If the formula was still divided by 100, Dorian’s score would plummet down to 135 units.

[1900hurricane]

What about something like this? You could make a ratio of winds and pressure compared to the max known cases, multiply each ratio by 50, and then add the two numbers up to get a total out of a possible 100.

50(Vmax - 25 kt)/(185 kt - 25 kt) + 50(1010 mb - Pmin)/(1010 mb - 870 mb)

My example uses kt, but you could easily switch it to whatever unit you want to use. In a more generalized form, it looks like this:

50(Vmax - Floor Case)/(Peak Case - Floor Case) + 50(Floor Case - Pmin)/(Floor Case - Peak Case)

[aspen]

What about something like this? You could make a ratio of winds and pressure compared to the max known cases, multiply each ratio by 50, and then add the two numbers up to get a total out of a possible 100.

50(Vmax - 25 kt)/(185 kt - 25 kt) + 50(1010 mb - Pmin)/(1010 mb - 870 mb)

My example uses kt, but you could easily switch it to whatever unit you want to use. In a more generalized form, it looks like this:

50(Vmax - Floor Case)/(Peak Case - Floor Case) + 50(Floor Case - Pmin)/(Floor Case - Peak Case)

I don’t want to use hard upper limits for cyclone intensity. Lower estimates are okay because something below them is probably not even a TC, but there’s nothing that says Patricia and Tip’s record intensities are the peak of what a TC can be (it’s quite possible Patricia, Haiyan, and/or Hagibis could’ve broken Tip’s min pressure of 870 mbar). Also, I want to preserve as much of the original score range as possible.

I’ve slightly tweaked my original formula to ((1015-P)(W-25))/75. This finally allows for those >1010 mbar depressions to get a score, and it’s not far off from my original forums using mph instead of kt. For example, Dorian was 185 units but is now 189; Patricia was 296.7 units but is now 305.1; This year’s Arthur was 11.4 units but is now 8.

I’m starting to go through the tedious process of finding the 1977-2017 average, but I might alter the formula again to try and make it somewhat exponential (score increases at a greater rate for more intense storms).

[1900hurricane]

The upper limits aren't hard, so if there is something higher than 185 kt or deeper than 870 mb, it's possible to get a score over 100. And to have the numbers closer to what you were seeing before (topping out around 300), just multiply each term by 150 instead of 50. Patricia tops out at 99.3 multiplying each term by 50 and summing them, but 297.9 if each term were multiplied by 150.

If you are going to make an average that includes TDs, just beware that TDs don't have pressure estimates in NHC best track until part way through the 1991 season. If I exclude those storms, I end up with an average of 24.4 units out of a total 642 NAtl TCs in that date range.

[GrayLancer18]

Are you me?

I thought of this same formula to come up with the rankings of my hurricane personas, except that I used 1013mb as the standard pressure to subtract with the TC's min pressure.

I got Patricia at 303, Tip at 272, Wilma at 242, Maria at 184 and Dorian with 191 units.

At one time I thought of adding into the equitation the amount of days the TC was active to give more credit to the long tracking ones and bring it nearer to an ACE-like measurement
but thought it would complicate things.

[aspen]

Here are the current numbers for the Atlantic and EPac seasons:

Arthur: 8.28
Bertha: 3.03
Cristobal: 7.93
Dolly: 2.69
1E: 0.62
Amanda: 1.66
Boris: 1.38

Atl total: 21.93 units
EPac total: 3.66 units


This is all pretty self-explanatory: just like with ACE, the combined storm intensity scores are far higher in the Atlantic (exactly 6 times higher) than that of the EPac. The Atlantic is so far well ahead of most of the recent seasons. Here is how the last ten ranked by June 29th of that year:
—2010: <66.62 units (Alex had not peaked at 95kt/946 mbar yet)
—2011: 9.10 units (all from Arlene)
—2012: 56.62 units
—2013: 12 units
—2014: 0 units
—2015: 12.07 units
—2016: 30.83 units (mainly from Alex)
—2017: 17.39 units
—2018: 10.34 units (all from Alberto)
—2019: 1.24 units (all from Andrea)

Out of all of the seasons from 2010-20, 2020 currently has the fourth-highest combined intensity score by June 29th, ranked only behind other seasons with enhanced early activity (or, in 2010’s case, one unusually strong early-season storm). There doesn’t appear to be a particularly strong correlation between this score prior to July and overall season activity (2019 is less than 2013, but we all know how they ended up), but it’s still worthy of noting regardless.

Also, I need a better name for this score/intensity estimation, badly.

[USTropics]

Here are the current numbers for the Atlantic and EPac seasons:

Arthur: 8.28
Bertha: 3.03
Cristobal: 7.93
Dolly: 2.69
1E: 0.62
Amanda: 1.66
Boris: 1.38

Atl total: 21.93 units
EPac total: 3.66 units


This is all pretty self-explanatory: just like with ACE, the combined storm intensity scores are far higher in the Atlantic (exactly 6 times higher) than that of the EPac. The Atlantic is so far well ahead of most of the recent seasons. Here is how the last ten ranked by June 29th of that year:
—2010: <66.62 units (Alex had not peaked at 95kt/946 mbar yet)
—2011: 9.10 units (all from Arlene)
—2012: 56.62 units
—2013: 12 units
—2014: 0 units
—2015: 12.07 units
—2016: 30.83 units (mainly from Alex)
—2017: 17.39 units
—2018: 10.34 units (all from Alberto)
—2019: 1.24 units (all from Andrea)

Out of all of the seasons from 2010-20, 2020 currently has the fourth-highest combined intensity score by June 29th, ranked only behind other seasons with enhanced early activity (or, in 2010’s case, one unusually strong early-season storm). There doesn’t appear to be a particularly strong correlation between this score prior to July and overall season activity (2019 is less than 2013, but we all know how they ended up), but it’s still worthy of noting regardless.

Also, I need a better name for this score/intensity estimation, badly.

I like your premise here, it's sort of a snapshot of ACE without getting into the complications of time duration. In typical engineering/scientist fashion, you are required to use an acronym. Something like the Combined Intensity Method score, or the CIM score.

[aspen]

Here are the current numbers for the Atlantic and EPac seasons:

Arthur: 8.28
Bertha: 3.03
Cristobal: 7.93
Dolly: 2.69
1E: 0.62
Amanda: 1.66
Boris: 1.38

Atl total: 21.93 units
EPac total: 3.66 units


This is all pretty self-explanatory: just like with ACE, the combined storm intensity scores are far higher in the Atlantic (exactly 6 times higher) than that of the EPac. The Atlantic is so far well ahead of most of the recent seasons. Here is how the last ten ranked by June 29th of that year:
—2010: <66.62 units (Alex had not peaked at 95kt/946 mbar yet)
—2011: 9.10 units (all from Arlene)
—2012: 56.62 units
—2013: 12 units
—2014: 0 units
—2015: 12.07 units
—2016: 30.83 units (mainly from Alex)
—2017: 17.39 units
—2018: 10.34 units (all from Alberto)
—2019: 1.24 units (all from Andrea)

Out of all of the seasons from 2010-20, 2020 currently has the fourth-highest combined intensity score by June 29th, ranked only behind other seasons with enhanced early activity (or, in 2010’s case, one unusually strong early-season storm). There doesn’t appear to be a particularly strong correlation between this score prior to July and overall season activity (2019 is less than 2013, but we all know how they ended up), but it’s still worthy of noting regardless.

Also, I need a better name for this score/intensity estimation, badly.

I like your premise here, it's sort of a snapshot of ACE without getting into the complications of time duration. In typical engineering/scientist fashion, you are required to use an acronym. Something like the Combined Intensity Method score, or the CIM score.

Making an acronym for this is surprisingly difficult. Combined Intensity Score, Combined Cyclone Intensity Score, Integrated Cyclone Intensity Score, and more have yielded rather undesirable acronyms. I’m trying for something short, easy to remember, and pronounced like an actual word like ACE.

My best result so far is Integrated/Instantaneous TC Intensity Method (ITIM, pronounced “item”). The first I is interchangeable because it integrates both maximum winds and minimum central pressure, and like you said, it’s essentially an instantaneous version of ACE. I’m pretty sure I’ll end up changing this too.

Edit: perhaps I could lean more into its relation to ACE and name it Instantaneous Cyclone Energy, or ICE. Where ACE is the total energy over the duration of the cyclone’s life, ICE is the energy/intensity score when the cyclone peaked.

[1900hurricane]

I just call it the Aspen Index in the file I made to compute it.

[aspen]

I’ve finally completed the calculation of mean ICE from 1977-2016 and the ranges of below-average, average, above-average, and hyperactive seasons.

Mean 1977-2016 ICE: 386.28 units
Average: 328.34 to 444.22 units (85-115% mean)
Below-average: <328.33 units (<85% mean)
Above-average: 444.22 to 579.42 units (115-150% mean)
Hyperactive: >579.43 units (>150% average)


Now the seasonal ICE totals from the last ten Atlantic hurricane seasons, as well as 2005 for comparison:

2005: 1,300.35 (336.63% mean; extremely hyperactive)
2010: 764.54 (197.92% mean; hyperactive)
2011: 469.41 (121.52% mean; above-average)
2012: 453.52 (117.41% mean; above-average)
2013: 116.35 (30.12% mean; extremely below-average)
2014: 275.73 (71.38% mean; below-average)
2015: 268.27 (69.45% mean; below-average)
2016: 433.85 (112.31% mean; high-average)
2017: 874.57 (226.41% mean; hyperactive)
2018: 525.24 (135.97% mean; above-average)
2019: 550.06 (142.40% mean; above-average)

[aspen]

The Atlantic’s ICE has risen significantly due to the record pace of storms forming in July. With Hanna likely at its peak intensity, here’s the ICE for all of the Atlantic’s storms so far and the current season total.

Arthur: 8.28
Bertha: 3.03
Cristobal: 7.93
Dolly: 2.69
Edouard: 1.66
Fay: 5.86
Gonzalo: 7.45
Hanna: 31.86
TOTAL: 68.76 (8/1/0)

2020 has now risen to be ahead of all of the recent active seasons (since 2016) in terms of its total Instantaneous Cyclone Energy, even more than 2018, which mainly got so high due to 41-unit Cat 2 Hurricane Chris. The ICE of 2016-19, 2010, 2008, 2005, 2001, 1999, 1998, 1996, and 1995 through July 25th are shown below for comparison.

2019: 13.52 (2/1/0, +1 TD)
2018: 66.48 (3/2/0)
2017: 20.15 (4/0/0)
2016: 30.83 (4/1/0)
2010: 69.38 (2/1/0, +1 TD)
2005: 297.85 (6/3/2)
2001: 5.17 (1/0/0, +1 TD)
1999: 3.86 (1/0/0, +1 TD)
1998: 0 (0/0/0)
1996: 80.28 (3/2/1)
1995: 39.11 (3/1/0)