Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

There is no way to credibly make this kind of claim. We don't have the data to quantify the recurrence rate of low-probablity events like this, nor can we be sure the the underpinning mechanism isn't changing with time.


If we have a baseline of 500 year flood level at 0.2% and it's well over that then yes there is a basis for making this claim.

I wouldn't make this claim though...


> If we have a baseline of 500 year flood level at 0.2%

That's exactly the point. We don't have that baseline, despite acting as if we do. We would need data going back 5,000 years to do that.


You're right.

But he's going off of what is established, whether that is faulty or not.

So his claim is based on something faulty, but it is based on something.

Your qualm is not with him or his statements, rather with the standards for current 10 year/100 year/500 year estimates.


I don't think I understand what you're saying. To be able to say things about 0.2%/year events, we would need many thousands of years of data on an unchanging system.


This is not true. You can assume a uniform prior and run your model through bayes law and come up with an estimate.


That's called "making things up", but sure.


And why would a judge (And not someone in environmental sciences) be making that claim anyway?!


He's the chief county administrator.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: