Trend Analysis: Mann Kendall

#1
I doing trend analysis on rainfall using mann-kendall.
I have daily rainfall data for 60 years.

My p value is close to 0 which indicates that there is a trend in data.
But the slope (sen's slope) is also 0 or in some cases 0.000008. How does one interpret these results?
Shouldn't the trend have a direction?

Attaching one of the results.
Using XLStat
 

bugman

Super Moderator
#2
Your p-value will almost always be small when you have such a large data set.
The sen slope estimator is used to estimate the annual rate of change, wheras your Kendalls tau (0.028? I think) is showing a very weak positive increase.

I would also be inclined to run a seasonal Kendalls model on this data set since it is rainfall.