The first thing that you have to do is to convert the table of annual maximum (statistical) values of rainfall depth into Intensity Values (mm/hr) just divide the total depth (precipitation) by the rainfall duration (i.e. for 1971 during 1 hr you have a rainfall max. value of 16.3 mm, then intensity for this year and this specific duration will be 16.3 mm/hr).
I am not sure about the negative values in your table. It must be or negligible rainfall depth or measurement data errors.
Once you have a table of annual maximum Intensity values (where the columns will be duration ranges/steps), you can perform a statistical analysis in order to calculate parameters such as the mean and standard deviation for each duration range.
After that, you can perform an analysis using the Gumbel distribution (a valid option but not the only one) that is used to model the distribution of the maximum of a number of samples of various distributions (in your case maximum rainfall intensity values for a sample obtained during 23 years).
I won´t go in depth into Gumbel Distribution parameters, just say that in hydrology the Gumbel distribution is widely used to analyze such variables as monthly and annual maximum values of daily rainfall. If required I will be able to attach and excel spreadsheet with a numerical example and calculations to obtain manually IDF tables or equations for different return periods (years) that are associated to Frequency parameter (1/T).
At the end, with a table of Duration vs. (statistical) Rainfall Intensity Values for different Return Periods you can use CivilStorm or SewerGEMS (the software will be able to generate graphs with IDF curves or Intensity Equations) and then use it for example to calculate runoff flow (peak) under modified rational method.
I expect this explanation could be useful for you.
Regards from Colombia,
Juan Carlos Gutierrez Araujo.
Bentley – Latin America
Juan Carlos Gutiérrez Araújo I.C. M.Sc.
Regional Sale Engineer - Bentley Latin America