Hurricanes, High-rises, and Hubris

[Jose Garcia]

Culebra:

El Yunque (Palmer, Rio Grande):

Fajardo:

 

[Margie Kieper]

Hi all -- just a quick post to note the errata on the 1990 Hugo NDS report.

 

[Michael Laca]

Hi all -- just a quick post to note the errata on the 1990 Hugo NDS report.

Hi Margie - Exactly where did this "errata" sheet come from? I personally own the physical printed 1990 DOC survey (which I received in 1991) and that errata was neither printed in, or distributed with the report? Additionally, the date of the errata page says May, 1990 implying it was included at the time of the original document's publishing... if so, why wouldn't the authors just correct the errors, rather than going to print with known issues and adding a separate errata page in to correct them?

It's also quite amazing, that the errata covers the EXACT points of contention from our debate about the 1990 Survey Report?

Also, curiously, when our debate started a few weeks back, I had checked a readily available PDF version of this report (linked through Wikipedia to the Service Assessment area of the NWS's website) and found the document to be EXACTLY the same as my physical copy (i.e. no errrata page). Curiously enough, when I went to check that same PDF tonight, the errata page has mysteriously appeared in the PDF file linked through Wikipedia. A quick check of the history for the Hurricane Hugo page on Wikipedia revealed that you have been rather busy lately, making several changes to the data...

(cur) (last) 17:44, 10 March 2007 Mkieper (Talk | contribs) (→External links)
(cur) (last) 17:43, 10 March 2007 Mkieper (Talk | contribs) (→External links)
(cur) (last) 17:34, 10 March 2007 Mkieper (Talk | contribs) m (→Storm history)
(cur) (last) 17:33, 10 March 2007 Mkieper (Talk | contribs) (→Storm history)

...these included changes to the body copy such as the following:

A [[tropical wave]] moved off of [[Cape Verde]], [[Africa]], on [[September 9]], [[1989]]. Moving westward, it developed into Tropical Storm Hugo on [[September 11]], and became a [[tropical cyclone|hurricane]] on the 13th. Hugo rapidly intensified and briefly reached [[Saffir-Simpson Hurricane Scale|Category 5]] intensity while well out in the [[Atlantic]]. The NHC best track documents that it scraped the [[Caribbean]] as a Category 4 hurricane, when it passed over [[Guadeloupe]], the [[Leeward Islands]], [[St. Croix]], and then weakened to a Category 3 prior to landfalling over [[Vieques]], [[Culebra]], and the eastern tip of mainland [[Puerto Rico]], while undergoing an eyewall replacement cycle (ERC).

...and to the external links...such as the 1990 DOC Survey.

What I find exceptionally curious, is that the physical URL for the PDF file of the 1990 Survey document DID NOT change between the time I checked this document, at the start of our debate (when it didn't have the errata), and when I checked it again tonight (when it does have the errata)... which can only mean that the PDF file was overwritten on the server where it is stored, sometime after our debate began?!?

Do you have ANY thoughts on how these discrepancies might have occurred... as a database admin, I'm sure you must be able to offer some possible explanations?

-=Michael=-

 

[Peter Kirk]

I think the error page was only added recently and the May 1990 date just refers to the original publication date.

 

[Michael Laca]

I think the error page was only added recently and the May 1990 date just refers to the original publication date.

Hi Peter - That is definitely a possibility. That said, I downloaded the PDF from the Service Assessment area on the NWS site and checked the file properties for the creation and modified date... the PDF was posted on the NWS server on March 8, 1999 at 11:26 AM and this version (with the errata) was only posted this past Monday, March 19, 2007 at 10:59 AM. So, besides the fact that I personally find it too coincidental that the NWS just happened to update the PDF version of this 17 year old document this week, I really want to find out:

• When the actual errata page was added to the physical document.
• Who authored and/or authorized the errata, and why that is not mentioned on the page.
• Why, if it was added later than May, 1990, there is no mention on the errata page of what date it was added.
• Why if it is supposed to correct "errors" in the document (one of those errors presumably being the citation in the second paragraph of the "Executive Summary" on page "xi" that Hugo was a Category Four when it struck Puerto Rico) that it doesn't also correct a similar statement only four pages later in the first paragraph of Chapter 1, which reads: "When Hugo struck the Virgin Islands, Puerto Rico, and the Carolinas, it was classified as a Category 4." It seems to me that if someone were actually going to take the time to correct "errors" in the document, that they wouldn't have overlooked this other instance, especially since the second point in the errata covers another presumable "error" on the same page?
• Why if the document is incorrect in reporting Hugo's maximum sustained winds at 132MPH over Vieques and that the authors really meant to quote the value of 109MPH (from the table in Appendix A-3)... why then is the statement (just four paragraphs before) that reads "Maintaining 140MPH maximum winds, the hurricane destroyed or damaged more than 90 percent of the buildings on St. Croix..." not in question, when the table in Appendix A-3 clearly indicates that sustained winds on St. Croix were 132MPH? Also, assuming that the table in A-3 was intended to provide the wind values cited in the document, and the 132MPH at Vieques was misquoted because it was the peak gust then, in theory, the 140MPH value shown for St. Croix would also refer to a peak gust, yet the A-3 table has the peak gusts on St.Croix as 161MPH?
• Why does the errata go as far as to say "...109 mph makes Hugo a Category 2 when it affected Vieques, Puerto Rico", when the previous errata point says "Hugo was a category 3 when it struck Puerto Rico."?

All of this is just WAY too fishy!!! I have e-mailed the NWS for clarification on when and who authored the errata and when it was added to the document, as well as who made the modification to the PDF file last Monday. I'm also contacting the actual authors of the 1990 DOC survey for the final word on when the errata was added and why there are so many discrepancies.

-=Michael=-

 

[Jose Garcia]

Aircraft recon during Hugo

Hi Michael and all...

I wanted to bring some points about what Kieper said about the recon measurements during the passage of Hugo across our area the morning of Septemeber 18th 1989. I know this is a little bit late in the debate but early enough to be discussed as I wanted to be sure I made enough research to mention it.

First:

Kieper
"At 1208Z, a recon vortex fix in the center of Hugo’s eye, located at 18.3 -65.5, or about 6 nmi to the east of Fajardo, measures a pressure that has risen to 955 mbar, and observes the eye is approximately elliptical, stretching 30 miles along an axis of 60 degrees (that is, an axis stretching from WSW to ENE), and 20 miles wide along the perpendicular axis. Also noted is that the eye is open to the SE. These are all signs of weakening and of the in-progress ERC. The recon also notes that the highest flight level winds encountered on the leg leading up to the center fix, which traveled into the eye from the south, were winds from the west at 67 kt (i.e, zippo)."

This 955mb measurement by the aircraft seems to miss the minimal central pressure of the storm as the well known TJNR site at Ceiba reported a pressure of 946.1 millibars. Landfall is listed 52 minutes later over Fajardo (13z/9am) and with a pressure of 946mb, this is based on the Ceiba report, not the reconnoisance aircraft measurements. The maximun sustained winds at Ceiba of 90kts in the southern (or SW) eyewall of the storm is measured at 1158z/7:58am from inland (with all the well mentioned friction factors before) and the minimal pressure of 946.1mb is measured at 1250z/8:50am.

Then the aircraft measured a 955mb pressure at 8:08am and Ceiba measures a lower pressure 42 minutes later (8:50am) and further south, a pressure of 946.1mb. That is an 8.9mb difference, and Ceiba did not get the center of the eye. Notice that the measurement of 956mb in Luquillo is made only 10 minutes after Ceiba's report leaving clearly that the center of the eye was still east of Luquillo and north of Ceiba, thus over Fajardo, were the minimal pressure was probably experienced. The NHC lists the minimal pressure of Hugo in PR based on the 946mb report of Ceiba and they fail to clarify that the pressure could have been lower at the time of landfall because it seems quite clear that the center of the eye passed north of that location.

Do we have any other experiences with discrepancies between measurements in land and a/c obs?

The answer is yes, and it seems to happen a lot when TCs are interacting with land.

An example is hurricane Marilyn in 1995:

When the eye of the hurricane passed right between Culebra and St.Thomas the night of September 16th, there was a recon airplane doing fixes in the storm and measured a pressure of 960mb with an extrapolation of 957mb at 0305z/11:05pm. The ASOS at the international airport in St. Thomas measured a pressure of 956.7mb at 0422z/12:22am while the sustained winds were still of 60kts in that site, almost hurricane intensity. They were not in the eye but in the edge of the eyewall while the eye was passing right between both islands. On this case, NHC clarified the fact that the pressure was actually lower than what the measurement of the airplane was and they even leave space for discussion about this fact:

TPC's Preliminary Report on Hurricane Marilyn

"The ASOS measured a minimum pressure of 956.7 mb. This occurred at 0422 UTC when the airport was still experiencing 60 knot one-minute winds, apparently on the inside edge of the eyewall. The estimated minimum pressure for Marilyn at that time is 952 mb. This is lower than implied by the data obtained from the Hurricane Hunters. They reported extrapolated and dropsonde pressures of 957 and 960 mb, respectively, at 0305 UTC, and 954 and 958 mb for those techniques at 0600 UTC. This is reminiscent of Hurricane Andrew's landfall over Florida, where the minimum pressure obtained from surface observations was lower than analyzed using aircraft data. The reason for this discrepancy in Marilyn is not obvious."

They also increased the winds to 95kts in that report after operationally being of 90kts, I'm sure this storm was actually stronger but then again, that's another debate!

Another example is hurricane Andrew in 1992:

1993 TPC Preliminary Report:

"The final offshore "fix" by the reconnaissance aircraft came at 0804 UTC and placed the center of the hurricane only about 15 nautical miles, or roughly one hour of travel time, from the mainland. A dropsonde indicated a pressure of 932 mb at that time. The pressure had been falling at the rate of about 2 mb per hour, but the increasing interaction with land was expected to at least partially offset, if not reverse, that trend. Hence, a landfall pressure within a few millibars of 932 mb seemed reasonable.



Shortly after Andrew's passage, however, reports of minimum pressures below 930 mb were received from the vicinity of Homestead, Florida (Fig. 4). Several of the barometers displaying the lowest pressures were subsequently tested in a pressure chamber and calibrated by the Aircraft Operations Center (AOC) of NOAA. Two key observations came from a Mrs. Hall and Mr. Martens, sister and brother. They rode out the storm in residences about one-quarter mile apart. Mrs. Hall's home was built by her father and grandfather in 1945 to be hurricane-proof. Although some of the windows broke, the 22-inch thick concrete and coral rock walls held steady, allowing her to observe her barometer in relative safety. The AOC tests indicate that the minimum pressure at her home was near 921 mb. The barometer at her brother's home was judged a little more reliable and the reading there was adjusted to 923 mb. Based on the observations and an eastward extrapolation of the pressure pattern to the coastline, Andrew's minimum pressure at landfall is estimated to be 922 mb. This suggests that the trajectory of the dropsonde deployed from the aircraft did not intersect the lowest pressure within the eye."

I put this long paragraph because it explains quite well the whole situation about the pressure measurement discrepancies. Again the diference was of around 10 millibars on this storm. This seems to happen more frequently while hurricanes are interacting with land or maybe it just happens anytime and it can only be noticed when there are actual land measurements that differ from a/c recon, or maybe is a little of both. Investigations need to be done about this.

Hurricane Georges in 1998:

Again the airplane failed to measure the storm's peak winds in the hours prior to landfall in Puerto Rico as it was deepening. The NHC had the storm at 95kts with 967mb (measured by recon) and then Fajardo measured 96kts sustained as the northern eyewall pounded the same area blasted by Hugo 9 years before. The NWS radar at Cayey measured winds of 100kts at that time and combined with the 96kts and later 98kts in Fajardo they increased the winds to 100kts at time of landfall here. I would have loved to see a good measurement from Yabucoa and Humacao were the eastern eyewall directly hit these areas. Georges was probably also sligtly stronger than the 100kts.

Back to Hugo:

Kieper again
"At 1238Z the same recon flight on its way north records the highest flight-level wind speed in the stronger northern eyewall: 108 kt. This translates to surface-level winds of 92 kt (using 85% reduction) – strong Cat 2 winds. This could also be considered the northwestern eyewall given the elliptical shape of the eyewall recorded minutes earlier. This maximum is located about 25 nmi to the NNE of Luquillo at this time."

Less than an hour before, Ceiba measured winds of around 90kts, the same intensity than that based on aircraft recon in the northern quadrant of the storm. Then Kieper's arguments about the intensity of the storm in different quadrants fall down like a rock (and without parachute). It seems quite logical then that the airplane failed to sample the storm's peak winds at that time and that those winds were actually hiting directly over land in NE Puerto Rico and Culebra at that point. Again its hard for a/c recon to be reliable when a hurricane is and has been interacting with land for several hours time frame. The proof is given with the other storms mentioned and the differences between the obs between recon and land stations in Hugo. Also I want to mention NHC's failure to discuss this in their report along with the anemometer failure after the 148kts gust in Culebra mentioned here by Michael before.

Every time I get to analyze all the data and see what that storm did in here I get more certain about the intensity this storm had. If a good thing can come out of this long debate, it could be the fact that those who experienced this storm and those who have analyzed it from every aspect known, have gotten more certain about the real intensity of Hugo in Puerto Rico no matter how many erratas come more than 17 years after the storm.

Jose

 

[Tom Mitchell]

Great thread. I have learned a lot. That you all.

I was in a typhoon with gusts recorded at 74 meters/second. That's 165 mph. Gusts on the other side of the island (Miyako Island), where I happened to be staying (stuck actually), have been estimated to be 180 mph based on damage. I have also been in typhoons with winds/gusts measured at 100/120 mph. The winds in the video look to me like 100 sustained with gusts to 130 or higher. No science or charts. Just an eyeball. I have video, but it simply looks just like Mr. Laca's, with the exception of the neighbor's car being blown down the street on its side. Here is the English version of the Japanese report on it (be prepared for some funky English.) http://www.pwri.go.jp/eng/ujnr/joint/36/paper/33kikits.pdf

Jose, welcome! I was in PR in 1990 for a week. I loved it. I was befriended by a couple from San Juan and the took me all over the island, and fed me gallons of Medalla beer. I have had a desire to go back ever since. I wish I could find that beer over here.

Keep up the good thread folks.

 

[Jose Garcia]

Hi Tom...

I'm happy you had a good time here and you're welcome whenever you can . And the Medallas are now even better I can assure you!...

I've seen a lot of hurricane/typhoon footage from many places including Okinawa and Miyako Jima which are probably the most hited places in the world. I also saw the video of Hugo in Youtube a few days ago and I was impressed on how much the quality and detail of the video is lost. One cannot give an estimate on wind speeds of a video on a small screen with a greatly reduced quality of video and sound. I compare it to seeing the video with dvd quality and in the 52-inch flatscreen tv in my house and it makes a world of a difference, and believe me, still today along with the typhoon Omar video in Guam this is the most impressive footage I've seen.

Also the wind damage by the storm as it was already mentioned a lot before was a lot worst in Luquillo and Fajardo than that of Georges in 1998 in the same place. In Georges there was an actual measurement of sustained 110mph winds and later 113mph in Fajardo to give you an idea that Hugo was in fact way over 100mph sustained in that place. Also you can verify my analysis on Ceiba's report of 104mph sustained in Hugo about the direction of the wind and the topographical characteristics of the area were the measurement was done and its differences with the area were the video was taken which was in front of the sea and with onshore winds during the eyewall. Again, I've seen TC's in live and lots of videos, I have also seen a lot of post-storm damage surveys documentation conducted after major hurricanes and the damage by Hugo in the NE part of Puerto Rico, Culebra, Vieques and St. Croix just keeps standing out as that of a cat-4 intensity hurricane...

Best Regards,
Jose