Rain Sensitivity coded Backwards?
I noticed when we were forecasted to get 2 inches of rain, the system still wanted to water the next day. Found that odd so did some digging and found out it seems the logic is backwards for this setting. unless i am understanding it wrong.
Default Value: 80%
Lower Value: 10%
If i'm understanding this right, the manual says the lower the value, the more water your system will use. Therefore when i say 80% of the forecasted value about 1.5 inches will be put down on the grass give or take the other factors. Therefore it should be in a water surplus mode the day after but it is not. When i put it down to 10% or (0.2 inches of rain taking into account) it shows its in a water surplus mode therefore holding excess water.
This logic/results of the logic seems backwards. Any help further explaining this??
Does having forecast correction on negate the NOAA values, so maybe that is why? if that is the case then it probably is right since it just started raining now.
-
As you observed rain sensitivity only changes the amount of QPF (or observed RAIN if you have a PWS from WUnderground) considered by the algorithm. Water surplus is triggered only when the available water in soil is enough for plants without any added irrigation. I'm not sure why water surplus was triggered but more information can be gained by looking at Field Capacity graphs from zones and also on device logs (Settings > About > View Log) at the bottom there is the simulation log.
Forecast correction doesn't negate NOAA values, it will work like this (assuming that you don't have a PWS from WUnderground) :
1. Let's say you have a morning program at 6AM, this program looks into next days forecast and sees 0.2 inches of QPF (for the days until next scheduled run).
2. Based on this QPF it computes a running time that's 80% from summer reference duration (we ignore soil available water and some other variables)
3. As the time/day passes the NOAA corrects the forecast used by program with better information increasing the QPF value to 0.8 for the interval.
4. At next run (after 2 days in your case) the program will see that it watered too much because it only knew about 0.2 inches of QPF instead of best forecast of 0.5.
So it will compute the delta and will subtracts the overwatering that he did on the past from the current computed duration updating the soil available water and other variables.
So basically Forecast Correction allows the program to not only look into future (forecast) but also in the past, which works even with only a forecast weather service as the forecasts sometime improve.
With a PWS from WUnderground,CIMIS, FAWN it will work on observed RAIN instead of QPF (will correct the QPF with RAIN)
-
Thanks for the explanation, was always curious as to the logic behind the forecast correction. It does work well as i've looked at the log many times.
Think i figured some of it out, just happen to catch the system at an awkward time, in between the corrections.
But still confused on why when I decreased the percentage the amount of watering decreased and visa versa, when I would think it would be the opposite according to the manual. I would think it should be an inverse relationship, which the results didn't show that in this case.
I will mess with it again when we have another day of predicted rain, probably next week to see if the same thing happens.
-
Had another day with a lot of rain forecasted (1.2 inches) and it actually rained over 1.5 inches, but still wants to water with the sensitivity up at 85%. When I Lower the sensitivity to 15% it say water surplus for the next few days as I would think would be true.
The relationship still seems off, that when I decrease the sensitivity the amount of watering times decreases as well. It should be inveresed.
Rain Sensitivity at: 15%:
Rain Sensitivity at 90%:
Is this correct? Can you explain the logic here? I don't see why it wants to put down about another inch of water by the end of the week. Seems like it thinks it didn't rain when it is at 90% and it did rain when it is at 15% which is backwards.
-
Hi Kenny,
I don't see the same thing, when I simulate with reduced rain sensitivity the watering times grow, see screenshots.
I'm not sure why you see it in reverse, I can open a private ticket if you wish so I can investigate your unit directly.
Rain Sensitivity 10%
Rain Sensitivity 100%
Please sign in to leave a comment.
Comments
9 comments