I have calculated Standard Precipitation Index (SPI) values using "SPEI" package in R. Interestingly, I have got some zero values of SPI for about 10 months in the same year and this has happened for multiple years. I was wondering about the possibility of this to happen as I haven't seen anything like this in the papers that I've been through. Please do tell if there is someting wrong here.
You have to refer to SPI by categories and their probability of occurrence according to Edwards and McKee, 1997.
SPI > 2 , Probability (%) 97.72, Interpretation : Extremely wet
SPI = 1.50, Probability (%) 93.732, Interpretation : Severely wet
SPI = 1, Probability (%) 84.13, Interpretation : Moderately wet
SPI = 0, Probability (%) 50, Interpretation : Normal
etc...
Regards
SPI gives relative variation from the normal conditions as wet or dry conditions. So one can interpret that zero SPI means normal conditions are prevailing.
Ashutosh Chaturvedi I agree. But the thing is that the zero values aren't scattered. They occur only in the month of July (highest monrhly rainfall) multiple years.
Abhijit Mitra I realise this but how's this possible. I've verified that the input data is correct. The zero values follow a pattern as they occur only in the month of July (highest monthly rainfall) across multiple years.
if i remember correctly the SPEI package has an option to show you the fitted distribution for the month of july (your case) did you check that? does it make sense with observations?
Andrea Galletti I assumed it to be in a Gamma distribution as that's the default setting but I'll take a look just to make sure.
- Badges
- Science topic
- Similar topics
- Engineering
- Modeling