|
Activity Number:
|
601
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Thursday, August 6, 2009 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statistics and the Environment
|
| Abstract - #305590 |
|
Title:
|
Stochastic Modeling of Stream Flow Measurements in the Mississippi and Missouri Rivers
|
|
Author(s):
|
Jonathan F. Joseph*+ and Haikady N. Nagaraja
|
|
Companies:
|
The Ohio State University and The Ohio State University
|
|
Address:
|
404 Cockins Hall, Columbus, OH, 43210,
|
|
Keywords:
|
Normal ; Lognormal ; Weibull ; Goodness-of-Fit ; Record Values ; Prediction
|
|
Abstract:
|
To describe patterns in annual peak stream-flow values, we collected data from five sites along the Mississippi and Missouri rivers measured over the past 80+ years using the USGS web resources. We fit parametric models to the data at each site and goodness-of-fit tests were performed. We also examined the serial dependence of the data from each site by using Box-Cox transformed variables. Sites with dependent data values were fit using autoregressive models. An examination of the record values was also performed. The number of records in the data from each site was compared to the expected number of record values. Finally, for each site, the 95th percentile of the peak flow rate in 2008 was estimated using our models. Our models are helpful in the prediction of future extreme events associated with the river peak stream-flows along these rivers.
|
