100 year return period flow

annmg

New member
Joined
Jul 8, 2019
Messages
2
For a dataset of river flow, I need to calculate the 100 year return period flow, and the 95% confidence interval. I first model the river flow extremes/maximums with a probability model, but what would be the next step?
 
Have you established the basic assumptions?

Independence of consecutive years?
Homogeneity?

What probability model? Poisson? Binomial? Other?
 
Hi, thanks for reply. Yes I have independent peak flows - recorded on 109 days out of a
14336 day dataset. The dataset is homogenous.
Probability model = Pareto wide tail.
 
Wow! I didn't expect that. You have restored my hope for humanity!

1) In case you REALLY want to get a better handle on the Pareto CIs. Here's some recreational reading: https://mspace.lib.umanitoba.ca/bitstream/handle/1993/32278/Min_Bo Yun.pdf?sequence=1

2) This one is a little more practical. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.429.3440&rep=rep1&type=pdf

3) Least Painful, see Example 9.11 in this text: http://lagrange.math.siu.edu/Olive/ich9.pdf

Are we getting anywhere?
 
Top