Objectives
The Point of Rocks gage (USGS 01638500) is located on the Potomac River in Maryland. The gage has a drainage area of 9,600 square miles and is located approximately 55 river miles upstream of Washington D.C. The location of the gage is shown in Figure 1.
Annual peak discharge data is available starting in water year (WY) 1889. A distribution fitting analysis was performed to determine the most appropriate probability model to fit to the annual maximum discharges. HEC-SSP includes 19 analytical distributions and three fitting methods. Also, 5 goodness of fit tests are included to evaluate model performance. The 5 goodness of fit tests used in HEC-SSP are Kolmogorov-Smirnov, Chi-Squared, Anderson-Darling, Bayesian Information Criterion, and Akaike Information Criterion.
In this tutorial, the user will manually calculate the goodness of fit statistics for the log-normal distribution and compare the results with those from HEC-SSP.
The following major tasks will serve as an outline for the workshop:
- Copy data from HEC-SSP and compute distribution parameters for the log-normal distribution.
- Compute the Kolmogorov-Smirnov test statistic.
- Compute the Chi-Squared test statistic.
- Compute the Anderson-Darling test statistic.
- Compute the Bayesian information criterion.
- Compute the Akaike information criterion.
Spend approximately 20 minutes per task to accomplish Tasks 1 - 3. If time allows, complete Tasks 4-6.
Continue to Task 1. Getting Started in Excel.
An Excel spreadsheet with calculations for Tasks 1-6 is provided at the end of the tutorial.