I'm trying to identify a specific day or week of each year when a gradient has been achieved of an increase in 'delta'. I have one value for each day of the year from 2000 to 2018, and I want to extract the day or week of each year that the specific gradient increase occurs following the trough in the data.
I'm not sure if this is even possible but I'm trying to create a gradient rule so it can be applicable for each year, and not a specific value since the values are different every year. So maybe over that week the loess smoothed values increase by 0.07 per day... achieving a gradient of 0.01... or something similar? Given that the changes in value are not linear is this even possible and how could I extract that week from each year so I end up with 18 different values?
This is an example of the first few lines of my data:
This is an example of what the data for one year looks like - I want to extract the day/week each year when it starts to increase following the trough using some kind of gradient rule so I can transfer it across each year and keep it simple:
