The basic objectives of research presented in this report are characterizing and modeling short time increment (hourly) rainfall data from Indiana. Characteristics of hourly rainfall data from Indiana were investigated. Data from 74 stations were used in the study.

The homogeneity of Indiana hourly rainfall data was tested as a part of the study. Indiana hourly rainfall data is found to be statistically homogeneous. Several probability distributions were evaluated. Surprisingly, both the type I extreme value as well as the generalized extreme value distributions were found to be acceptable to characterize Indiana hourly rainfall data. The generalized extreme value distribution was used in this study.

The intensity-duration-frequency relationships for Indiana were investigated next. Relationships are developed so that rainfall depth for any location in Indiana can be accurately estimated for specified durations and frequencies.

There are several methods and procedures which have been developed to estimate rainfall depths. Results from these methods were compared to the results from in-situ data analysis. The results of this analysis are used to recommend the methods to use for rainfall estimation in Indiana.

Huff curves were developed for all the stations and analyzed. Although stations in the state were divided into three groups as north, central and south, the Huff curves from the three regions were very close to each other. Consequently, a single set of Huff curves is recommended for use for the state of Indiana.

Report Number



Hourly rainfall, Intensity-duration-frequency, curves, huff curves, probability distribution, SPR-2932

SPR Number


Performing Organization

Joint Transportation Research Program

Publisher Place

West Lafayette, IN

Date of this Version