Dear Piyush

This is a question that is asked frequently on ResearchGate (see here for an example of a recent discussion: https://www.researchgate.net/post/How-to-obtain-short-duration-rainfall-from-daily-rainfall)

The simple answer is that you cannot accurately calculate rainfall intensities for short durations (i.e. less than 24 hours) if you only have a daily time series. However, you can attempt to disaggregate the daily extreme rainfall values. As I mention in the link above the process is as follows:

1. Carry out an extreme value analysis on your annual maximum (AMAX) daily rainfall time series for the period 1950 to 2018. Generally it is best to fit a General Extreme Value (GEV) distribution to your AMAX daily rainfall data. This will allow you to estimate daily rainfall for a range of annual exceedance probabilities (i.e. return periods)

2. Calculate the 24 hour extreme value rainfalls for the annual exceedance probabilities (AEPs) (i.e. return periods) you’re interested in e.g. 1% AEP (1 in 100 year return period) etc

3. Disaggregate the 24 hour rainfall data for different return periods to shorter time durations

I am making the assumption that the site you're interested in is somewhere in India?

If so you could non-dimensionalise a rainfall intensity - duration - frequency (IDF) curve for a location close to your site of interest relative to the daily rainfall e.g. if the 1 in 10 year 20 minute rainfall intensity is 30 mm and the 1 in 10 year 24 hour (i.e. daily) rainfall is 60 mm the ratio would be 30/60 = 0.5.

Following your statistical analysis (i.e. using a GEV distribution) on your daily rainfall at your site of interest you find you 1 in 10 year daily rainfall is 70 mm, you can estimate your 1 in 10 year 20 minute rainfall using the non-diemnsionalised ration i.e. 1 in 10 year 20 minute rainfall at your site of interest = 0.5 x 70 mm = 35 mm.

There are several papers which detail rainfall IDF curves for India see: Conference Paper Development of Sub-Daily Intensity Duration Frequency (IDF) ...

I hope that this is of help.

Darren

USING ARTIFICIAL NEURAL NETWORKS TO ESTIMATE MISSING RAINFALL DATA

1998, Journal of The American Water Resources Association

📷Ana Paula Barros

I do agree with Darren Lumbroso.

Hello everyone,

Darren Lumbroso's response is very explicit and educative. I strongly agree with him.

Watch the below video:

https://youtu.be/YDJkgMuEjgA

It would be helpful.