Extreme values trend analysis

Hi everyone!
I've got a theoretical question: i'm analyzing an homogeneous rainfall time series, time resolution is 1 hour.
In order to find whether a statistically significant trend is present in extreme values of the precipitation time series i've extracted the 99-th percentile of annual daily maximum and annual hourly maximum precipitation.
Therefore i applied a Mann Kendall test on those values to see if the trend is present.
I repeated the same analyisis over Discharge data wich is related to the precipitation series.
Does anyone know if this can make sense for my purpose? Do you have some scientific papers which i can refer to, or other similar studies?
Thank you and have a nice day!!

This question seems to have no relation to R per say.
Perhaps you will get better responses on Cross Validated (stackexchange.com) or some other statistics focused forum.

1 Like

Yes, thank you for your advice