Hi Tuppad,

If you understand the basic steps below, the coding would be easier:

1. Let X be your precipitation values for individual years

i.e x1 =1901 precip

x2 = 1902 precip

etc

2. Let Xmean =longterm mean of precip for a particular station (mean of the 100 years: mean of X1: X99)

3. Let Stdev = Standard deviation of precip for a particular station

4. Then, SPI = (X - Xmean)/Stdev

The above can be computed in Excel easily. However, to use R

Here is a package dedicated for SPI (find below link of the package and example link)

https://cran.r-project.org/web/packages/spi/spi.pdf

https://rstudio-pubs-static.s3.amazonaws.com/202762_d1269e6d029a443a889782fd9dc4f3a0.html

Regards.

Hi Pushups,

I totally agree with Akintomide Afolayan Akinsanola. Moreover, if your rainguage stations are within a basin, I suggest you estimate mean areal precipitation and (ofcourse if that is within the scope of your study) before you compute the spi.

You can also try the following to compute spi;

Install and load the library SPEI.

spi1