According to WMO, gamma distribution is used to fit the non-zero values of rainfall and then the Gamma CDF is calculated. Then, CDF of the whole dataset is obtained by adding the probability of zero values and (1- P0)* Gamma CDF of non-zero values. From this plot, corresponding SPI is obtained. But several papers are there which fit different distributions to obtain SPI. Can anyone explain in detail what is the actual (conceptual) procedure?
Dear Swain,
Please check these useful researchgate links and PDF attachment.
https://www.researchgate.net/publication/227803277_Accepting_the_Standardized_Precipitation_Index_A_Calculation_Algorithm
https://www.researchgate.net/publication/235741629_On_the_use_of_Standardized_Precipitation_Index_SPI_for_drought_intensity_assessment
https://www.researchgate.net/publication/257197372_Computation_of_Drought_Index_SPI_with_Alternative_Distribution_Functions
https://www.researchgate.net/figure/229459231_fig2_Figure-2-SPI-equiprobability-transformation-In-this-example-a-3-month-precipitation
https://www.researchgate.net/post/Can_anyone_guides_me_how_to_calculate_Standardized_Precipitation_Index_SPI_in_excel_when_rainfall_is_zero_x0
Good luck
Article Accepting the Standardized Precipitation Index: A Calculatio...
Article On the use of Standardized Precipitation Index (SPI) for dro...
Article Computation of Drought Index SPI with Alternative Distributi...
Hi
According to many papers, the proper one is Gamma distribution function. If you want to calculate it easily, please try my free software that it is located in the bellow website:
www.agrimetsoft.com
It has a user-friendly software and you can apply and run it easily. It name is MDM (doi: 10.1007/s40333-017-0070-y ), also you can calculate other 7 rain-base indices by run it. Just try it!
Best Regards,
Nasrin
Hello Sir,
I have tried for Gamma distribution and found that 100 out of 150 stations are not following Gamma.
Does it mean I can not find SPI?
What if I find SPI drought index with the same results???
Any option to Gamma? Please.
we have worked on these manuscripts that can serve you in drought:
Article Systematization of the calculation of the Standardized Preci...
The Standardized Precipitation Index (SPI) is calculated using a statistical distribution that fits precipitation data. The most commonly used is the Gamma distribution because it models positively skewed data and only takes positive values, which matches rainfall patterns—frequent dry periods and rare extreme wet events.
First, long-term monthly precipitation data (preferably over 30 years) is collected. The Gamma distribution is fitted to the data, and cumulative probabilities are calculated. These probabilities are then transformed into standard normal values (Z-scores), resulting in SPI values with a mean of 0 and standard deviation of 1.
When zero precipitation values are present, especially in arid regions, the model adjusts by incorporating the probability of zero rainfall.
SPI can be calculated over different time scales (e.g., SPI-1, SPI-3, SPI-12) to assess short- and long-term droughts, making it a reliable and flexible tool for drought monitoring.
- Badges
- Science topic