I believe the program uses the current observations data. Specifically, the ‘icon’ and ‘icon_url’ fields because they provide a text value of what the weather is currently.
The following are considered adverse weather:
adverse_codes = ["flurries","sleet","rain","sleet","snow","tstorms","nt_flurries","nt_sleet","nt_rain","nt_sleet","nt_snow","nt_tstorms"]
The following will reset (clear??) the rain delay:
reset_codes = ["sunny", "clear", "mostlysunny", "partlycloudy"]
Maybe there are more fields values that originally anticipated?
Now, I’d like to ask if the program could consider using the same information from the hourly forecast. The reason being is that I don’t want to water the grass if I know there is a X percent probability of rain in the next Y hours. Of course, I’d like X and Y to be user defined. 🙂 I believe the hourly forecast has a field called ‘pop’ for probability of precipitation?
Thanks!
Brad