Why water today what we'll get tomorrow?


#1

Our miserably rainy spring continues here in the Boston area. I honestly think my poor little Rachio has not had a single chance to perform a full watering schedule since installing it weeks ago. That’s great and all that I’m not wasting well pump time, but it must feel lonely. Like a poor minion without a villain to look up to and perform a job.

I looked just now at the moisture levels and predicted water schedule for a zone to see it plans on running June 3rd and dumping ~0.36" of water into the lawn to bring it from 33% to 97%. It must be so excited to perform its job! :sunglasses:

But… wait a minute. The graph keeps going up for the two consecutive days afterwards. :neutral_face:

Ah… KTAN is forecasting 0.88" of rain over those two days. :frowning:

We’re still just over 48 hours from when it is supposed to run so anything is possible, but when I see something like that it starts to cast doubt into the smarts of a device. Would dropping my Allowed Depletion from 50% to a lower figure help me avoid situations like this?

It feels a little silly to pump the lawn full when mother nature may take care of it soon after anyway. I’d rather let that forecast play out and then let the Rachio make up the delta if necessary. I know I can manually enable Rain Delay, but I’m hoping the closer to the 3rd we get it starts taking the 4th/5th into account and does the smart thing on its own by looking ahead a bit. :slight_smile:


#2

Poor thing saw we received 0.29" of rain yesterday and pushed the schedule out to June 9th. He’s never going to get a chance at this rate! Come on summer-ish weather, where are you? :cry:


#3

That is New England weather! July is coming. I lived in Boston and then western Mass for seven years. July can be absolutely oppressive. As bad as Houston.


#4

Yup, if you don’t like the weather just wait 15 minutes. The funny thing is Winter the last few years is really delaying its arrival. We’ve been having unseasonably warm weather into January before February punches the hope out of you.


#5

@scorp508 If your lawn is go to drop any amount below the AD line in the next day, we do decide to water, so that’s probably why it was watering on the 3rd, even though it was going to rain the following day! You could lower that allowed depletion level, but it has some other implications as well (longer run times) and I would never take it below 20%, as your plants begin to stress more permanently at that point!

Looking forward to hearing how it works when you guys stop getting dumped on :umbrella:

McKynzee :rachio:


#6

Understood, but the logic is… challenging to accept given the product likes to tout how much water it saves. Looking at the data above looks like an entirely wasteful use of resources it was planning to do. It may align with the formulas and the code, but it’s one of those things you look at as a human and think… wait a minute, that just seems odd.

It is going to the store and buying two gallons of milk today even though there’s half a gallon left in the fridge and the milk man is scheduled to come tomorrow to make a delivery. :slight_smile: Unless I’m planning to eat a heck of a lot of cereal today there’s no reason to run out and buy some.

I would completely understand if it looked at the forecast and then stared into the mirror and said “Self, I know we’re approaching the point of allowed depletion. However… we’re a smart controller! See that possible rain coming? We’re doing to just to a little top-off to get the zones through another 24 hours of evaporation instead of filling to capacity. I know I know, we like long watering times to promote root growth, but we may have some free water coming our way! If we spit a little less on the lawn today we may be able to increase our water saved (or well pump time saved) number with help from mother nature!” At that point it may run each zone for just a couple minutes to get through another 24 hours, see what if any rain drops, then catch up if necessary afterwards.


#7

@scorp508, living in the Arizona desert, I’ve heard something of that thing you call rain but haven’t experienced much firsthand. Seriously though, I might be a little concerned about runoff if Rachio watered and then the next day we had the rare downpour - our clay soil doesn’t drain well.

@mckynzee, what if Rachio still watered if the lawn may drop below the AD line the next day, but limited the maximum watering amount if predicted rainfall will likely replenish the field? Rachio could water just enough to keep the moisture level above the AD value on the following day. For example, in the above if Rachio would have watered 0.12" on June 3 instead of 0.36", then the balance on June 4 would have been: 0.12 + 0.12 - 0.13 = 0.11" (June 4 ET value). Then even if no rainfall would have occurred on June 4 (weather prediction turns out to be dead wrong), the balance would still not drop below the AD on June 4.

Other strategies might work, but maybe this could be a good compromise between opening the door to unexpectedly low moisture levels and field runoff.


#8

@scorp508 - I get your point, but I can probably see a weather pattern that will cause the Rachio to water just a little bit several times in a row because it is being teased about “free” water falling from the sky. At some point, if one believes in watering deeply to encourage root depth, Rachio will need to fill the field. I think this much “smarts” may have been “a bridge too far” for V2 and depending on where Rachio is in V3 development and if this was included in the original design for V3 it may be best suited for V4 - IMHO.


#9

Exactly what I was trying to portray in my last paragraph. :cheers: :cheers:


#10

I imagine the algorithm could “give up” on predicted rain if the weather prediction was faulty once or twice. Probabilistically it seems possible but unlikely that there would be a persistent chain of faulty rain predictions that would push the field past capacity. If so that meteorologist might want to consider another day job.


#11

@scorp508, come summer, I am going to have exactly the OPPOSITE problem than what you’ve been talking about in this thread — We get summer storms predicted every afternoon, so Rachio thinks it’s going to get some rain almost every day. But they rarely happen in my area. So my perennial/annual beds sometimes don’t water based on the predicted rain for a few days, and I may have to go in and force a manual watering.

When trying to automate this, we are all at the mercy of how good the forecasts are. As Rachio matures, I’m hoping they will be able come up with some good viable ideas to overcome the unpredictability.


#12

I believe the OP did say ( in a roundabout way) that he preferred the device to be smart enough to make up the delta from forecast-caused rain delays when the expected rain does not materialize. Who cares if the controller lives in a constant state of equalizing deltas after-the-fact? At least that way the objective of factoring in forecasted rain is 100% met, while using the absolute minimum amount of water required. Seems like a no-brainer to me. Programming the algorithm this way would also ensure that manual watering mentioned by @scorp508 would never be necessary.


#13

^— ding ding ding.


#14

@scorp508, I feel your pain. I installed my Rachio in March, and my Rachio still has not done a “automatic” water on the flex daily schedule. And we have just gone into winter. At least I am saving water…


#15

On the daily flex schedule the water balance displays precipitation for “today” based on the downloaded forecast to estimate the moisture level and decide whether to water today. On 8 June the forecast was for no rain and Rachio accordingly scheduled some irrigation at 4.30pm. However there had been rain in the morning which was not recognised in the Moisture levels More Detail calculation screen until midnight - after watering had taken place (actual rain comes from my PWS, forecasts from somewhere a bit further away so actuals are likely to diverge). The amount of rain was not huge (about half the daily estimated transpiration) so no great harm done. But what if there had been a local cloudburst in the morning? Does Rachio not recognise actual current day rain prior to initiating watering on daily flex or is it recognising it but not displaying it on the screen?


#16

I’m pretty sure that Rachio checks everything about an hour before a flex schedule is supposed to run to see if it needs to run that day. If it rains after that, Rachio is not going to know about it until the next day.

See this post from @franz:


#17

That’s fair enough. I wasn’t sure if Rachio only took the forecast. My flex programs water in the afternoon and I haven’t seen the moisture level detail tables show an actual today figure say half an hour before watering is due (assuming the actual is greater than the forecast). The actual only seems to appear after midnight. If Franz says they are taking an actual reading and comparing it with forecast to make flex watering decisions on the current day then I’m not fussed that it doesn’t appear in the display until the next day. Bottom line, I haven’t had the box long and I’m still learning!


#18

I could be wrong, but I think that there are some PWS’s that only report precip once every 24 hours, so in that case, Rachio wouldn’t pick it up immediately anyway.

The more you learn, the more I think you will like it!


#19

Normally a PWS reports rain with the same frequency it reports other data. The frequency is set by the owner - typically 5 to 10 minutes although you . choose longer or shorter intervals as you wish. I have mine set to 5 minute intervals and it reports three rain items to CWOP: rain in the last hour, rain in the last 24 hours, and rain since last midnight (the first two incorrectly at present!).

Oh yes I’m already loving the Rachio!:heart_eyes: