| gsgs |
| (Registered User) |
| Mon Sep 08 2008 06:18 PM |
damage predictions
|
are there hurricane damage-estimates (forecasts), which are being updated
when new storm-data come in ?
(Post moved to the appropriate Forum.)
| Clark |
| (Meteorologist) |
| Mon Sep 08 2008 09:22 PM |
|
Re: damage predictions
|
There generally are not, at least not publicly available, just expectations for various weather features (rain, wind, etc.).
| gsgs |
| (Registered User) |
| Tue Sep 09 2008 06:34 AM |
|
Re: damage predictions
|
there should be a general list, how much windspeed is expected to cause how much damage
in which region.
This could be combined with the forecast-models.
I remember, I saw a quick estimate of the Gustav damage before the first reports
about actual damages came in
expected (local) damage is more important, more interesting for the mitigation,
evacuation plans than predicted paths and windspeeds
| cieldumort |
| (Moderator) |
| Tue Sep 09 2008 07:09 AM |
|
|
Re: damage predictions
|
In a way, your question gets to another topic of discussion I have long advocated: Replacing the Saffir-Simpson scale with either a major overhaul, or something entirely new. I have my own share of ideas regarding this, and it has everything to do with IMPACT forecasting, rather than an almost myopic focus on maximum sustained wind speed (which occurs when using Saffir-Simpson).
You are right to put your finger on the crux of the problem: Currently, public forecasts (which are based on getting word of an expected Saffir-Simpson level out) by its very nature, focuses not on impact, but on some really rather arcane rating based on some core of wind that is unlikely to occur in roughly 99% of the total affected area.
For impact forecasts, I have long appreciated TWC. They truly *get it*. Leading up to most significant landfalling tropical cyclones, Dr. Lyons typically creates an impact forecast for "real people," if you will. Four of the key components of his forecasts cover:
Rain/Flooding potential
Wind damage potential
Wave threat potential
Storm surge potential
and are updated to reflect what is going on and what is expected.
You can read more about what Dr. Lyons has had to say about this subject here.
| gsgs |
| (Registered User) |
| Tue Sep 09 2008 09:03 AM |
|
Re: damage predictions
|
yes, looks similar to what I want.
What's TWC ?
They still need the windspeed-analysis, but only as a tool to predict damage.
BTW. why limit to 5 categories representing the windspeeds ?
Just give the speed. Or if folks can't remember their relations and significance,
extent it to fractions like Cat2.7 , see earthquakes.
I'd like several forecasters giving subjective expected damage predictions,
(expectation value,deviation) and an analysis who gives best predictions
as with the "competition" here with the several forecasters, forecasting-models.
And prediction-markets trading damage-contracts, with the opportunity to
"insure" by shorting damage-contracts. Well, there are catastrophe-bonds,
but only for big institutions
hmm... maybe I should start a different thread
| cieldumort |
| (Moderator) |
| Wed Sep 10 2008 01:20 AM |
|
|
|
Re: damage predictions
|
What's TWC ?
The Weather Channel (Click on the link I provided above for more on TWC & Dr. Lyons)
They still need the windspeed-analysis, but only as a tool to predict damage.
Yes, but the "analysis" as it is currently conveyed (by way of the Saffir-Simpson Scale) gives a horrendously unclear picture of what that damage is going to be like. (Ignores things such as the total duration of damaging winds, the radius of damaging winds, radius of exceptionally damaging winds, etc.)
BTW. why limit to 5 categories representing the windspeeds ?
Just give the speed. Or if folks can't remember their relations and significance,
extent it to fractions like Cat2.7 , see earthquakes.
Rating tropical cyclone windspeeds is not at all as simple as the Richter Scale. A "Cat 2.7" would never work, whatsoever. It would only be a worsening of an already flawed system.
I'd like several forecasters giving subjective expected damage predictions,
(expectation value,deviation) and an analysis who gives best predictions
as with the "competition" here with the several forecasters, forecasting-models.
Good luck with that! Hurricane forecasting is simply not that precise, and likely won't be for decades to come. Forecasters might have contests that could be tracked, but you're likely to get more snake oil than not, until forecasting improves markedly.
And prediction-markets trading damage-contracts, with the opportunity to
"insure" by shorting damage-contracts. Well, there are catastrophe-bonds,
but only for big institutions
I believe what you seek for the rest of us to be able to trade here already exists. (Oil futures, grain futures, etc.) While not "damage contracts," these are pretty darn close, when you think about it. As close as it's going to be given the state of the science -- although, I think more exactly of what you want might be a nifty item on Intrade.