Comparison of two dataframe or time series and display the final results as maps or time series

[surajitdb]

Say, we have daily precipitation time series like CHIRPS/ PERSIANN-CDR from GEE. What I want is to have a total count if a particular condition is satisfied considering two dataframe or time series. For example, a particular pixel of daily CHIRPS precipitation has the following values: df1=[1, 0, 2, 3, 4, 5, 0, 6, 0,10] for say 2001-01-01, 2001-01-02, ...., 2001-01-10 days. Similarly, for the same days, the daily precipitation for the corresponding pixel of another gauged/measured based precipitation are:
df2= [1.5, 0, 2.4, 3.1, 4.5, 5, 0, 6.2, 0.5, 10.9]. Now, we want to know the total count for 10-day precipitation satisfying a particular condition considering both df1 and df2. So, total_count= =COUNTIFS(df1[1]:df1[10], ">0",df2[1]:df2[10], ">0"), Also, total_count1= =COUNTIFS(df1[1]:df1[10], ">0",df2[1]:df2[10], "=0") and total_count2= =COUNTIFS(df1[1]:df1[10], "=0",df2[1]:df2[10], ">0") . This style was from Microsoft Excel. This could be done easily in Excel when we consider areal average values, but is there way out to do the same spatially over a given length of time on daily basis?

Once I have the total count I want to calculate certain categorical statistics spatially. So, the output would be the third data frame .
So, to start with we will extract the time series of daily precipitation like CHIRPS (extending beyond 5000 data points possible?) for a given area defined by lat/long bouns or a shapefile over a given time period and also upload raster files of gauged-based daily precipitation to carry out the task.

Please suggest to carry out this task efficiently on GEE.

[giswqs]

This can be done using ee.Reducer and ee.Filter. There are a lot of ee.Filter for performing conditional operations. See

• https://developers.google.com/earth-engine/guides/reducers_reduce_columns
• https://developers.google.com/earth-engine/apidocs/ee-filter-gt

[surajitdb]

Thank you, Dr Wu, for the response. I will look into it and get back if I get stuck somewhere.