Rain calibration corrections

We are now half way through the rain season. I am concerned the rain season will end before the Sky will be reasonably calibrated.
In the last week of storms, my Davis VP2 has recorded a rainfall total of 2.51" while the Sky recorded 7.07".
I have noticed that winds above 12-14 mph and/or gusts over 15-18 mph will often result in rainfall being recorded - even on a clear day.

Don’t worry. When your rain seasons stops my rain season starts. There will be plenty of seasons coming to keep the data flowing. I’ll be placing a station in Saigon where the rain season is all year. I’m looking forward to measuring the buckets full.

Todays rain event WF did better.
WF picks up the start of a rain event better than the VP2. Today’s rain stayed moderate without a long period of lite rain. So it never lost ground at the end.
I think when there are water droplets on the air it dampens the vibration of light rain but not moderate rain.

1 Like

Hi @stanleyfam Appreciate your reports. If you are experiencing false rain signals when the wind is above XX, then the cause is typically vibration in your mounting. Try to eliminate any possible vibration. Once the CL system for Rain is deployed, it will use data from your SKY as one of many inputs to ultimately determine accumulation values. If it does not corroborate with the other rain measurement signals, the algorithms will disqualify the SKY data and prevent ‘rain on a clear day’.

PS - we need to get down to PR for some wine tasting! :wine_glass:

what is there to celebrate?

Having a snow, sleet event today. Since the sleet started about 4 hours ago WF rain totals are very heavy. I have a heated VP2 which with snow (about 2.5 inches) and sleet has .34 inches of precipitation. The WF has .70 inches and that’s just the sleet.

1 Like

My Sky showed a somewhat high reading relative to my 2 area airports for our snow & sleet event. I had .69" vs .54" & .52" for the 2 area airports in fairly close proximity to me. To be honest, given that my Davis showed 0" precipitation (not heated) and no other area PWS showed any either, I’m not unhappy with the results. This is the first event of this type since I’ve had my Sky and I wasn’t sure what to expect.

So I’d rather have a bit of an overestimate than a ‘non-event’ reported by another system. IMO, for this kind of event, this beats a non-heated tipping bucket. :slight_smile:


We are having the Atmospheric River event here in SoCal today. My station is showing 2.22 inches for the day, and adjacent stations in the area are showing 3.34 to 3.97 inches. I live in an urban area, so the stations are bracketing me pretty well. Something seems really off here. Season to data I am a minimum of 20% below the stations in the area.

I know that light rain event show low outcome on the sensor, I have watched the rain come down, but the Sky is not showing any accumulation, but water is flowing on the ground. It also looks like heavy rain events are not reporting correctly as well, the intensity when compared with stations within a half mile of this station are showing a much higher intensity and related accumulation.

1 Like

I’m having the same issue with my WF under reporting rainfall compared to my VP2. I have the VP2 and the Weatherflow installed next to each other. The WF sits on top of the same pole about 1 meter above the VP2 and they don’t shadow each other.

Today the VP2 was reporting 4.2mm when the WF was reporting 1.8mm. So, pretty far off in these conditions. I’m routinely noticing this pattern the last month, but it’s summer here and we have been in drought conditions with only a few sporadic rain events the last 8 weeks.

1 Like

I’m having the opposite issue here with rain totals. We’re having a rain event here, and as has been happening lately, my Sky is underreporting. We’ve gotten at least .53” at most nearby reporting stations, including nearby airports, but my Sky is reporting only .39”. I looked at other WF stations on LI and a number are showing higher amounts. As I move farther away from my site though, I have less confidence in what the actual, accurate, rainfall is. With that said, I have no doubt my Sky underreports heavy rain events.

1 Like

I’d like to add my observations for SKY to the discussion:

I have SKY mounted level directly next to a two year old Acurite 5in1. I compare their rainfall measurements to those from 5 Harris Co. Texas Flood Control rain gages all within 2.0-2.5 miles of my location, in one case 3800 feet. I can track these gage readings on the web. https://www.harriscountyfws.org/

I assume that even if rainfall is "hyperlocal"there should be good consistency between my measurements and those of the nearby county gages. In fact, the Acurite measurement will be the same or very close to one or more of the county gages in almost every rain event, whereas the SKY will read quite a bit lower.

Some random examples: 2/11/19 acurite .43, county .44, .44, .44, .48, .64, SKY .16

2/20/19 acurite .04, county .04, ,04, ,04,.04,.04, SKY .00

2/21/19: acurite .16, county .16, .16, .16,.20, .20, SKY .10

2/26/19 acurite. 1.89, county 2.35, 1.76, 1.52, 1,52, 1.64, SKY .89

In light rain or drizzle, SKY struggles to register any totals at all. In a heavier rain SKY’s total will be about 50% of the others.This pattern has held true ever since I installed SKY several months ago.

I’ve read most of the above discussion and am confused as to whether I may have a defective SKY, or if I just need to wait for the anticipated firmware calibration. Any comments on this?

Everything else seems to be working OK and I"m very pleased with the overall concept and execution of AIR and SKY; but the rain measurement problem does keep me from showing off the units the way I otherwise would.

1 Like

As others have mentioned mine also seems to be registering low in both light rain events and heavy ones as well. We had a thunderstorm a few weeks back that my Sky recorded about half of what other nearby gauges did. It was a widespread storm system and not localized. Then this past week we’ve had rain off and on for several days, one day my Sky recorded .27 whereas nearby units ranged from .50 to .80. It seems like the sensitivity might be a bit too low since both light and heavy rain is reporting anywhere from 30-60% too low in my case and seems like this is a common issue.

Do we have any ETA on a possible timeline for the deployment of the auto-calibration system or any adjustments to bump the sensitivity up some?

1 Like

Most likely your SKY is not defective, it just needs a better calibration.

We have been hard at work for several months on significant improvements to the way rain accumulation is calculated as well as the way SKY rain sensors are calibrated. No ETA yet but we are getting closer every day. Thanks for your patience!


Thanks, I just didn’t know if there was a timeline for an initial release of this since it’s been close to a year now since this topic was started. I love the auto-calibration for other features, my UV index has fixed itself now… I’d say this is the main issue for me since we get a lot of rain here and having it consistently reporting 30-50% too low with no fix in sight is a big drawback for me. If you need additional beta testers I certainly would be happy to test out the auto-calibration and report results. Thanks.


Thanks. We have enough testers for now but appreciate the offer! We are hopeful that the rain CL process will receive a significant update for all users before the end of March.


So we’re just ending a rain/sleet/snow event here in NY.

I’ve been running relatively low on general rain events compared to local tippers, but for this event the nearby airports both recorded .42” of precip. My results:

Sky- .45”
My Davis VP2- .13”

So for whatever reason, the Sky did very nicely for this particular event.

1 Like

@WFmarketing @WFsupportx

Today, at 10:25, the station detects non-existent precipitation

1 Like

Looking forward to that and hopefully it happens by then since our rain season (if we actually get one) is normally done in April.

Both the WF server data and the UDP data are pretty much way off the actual rain amount. It continues that the WF doesn’t pick up light and heavy rain very well so it’s amounts are consistently low we got 6" one day and IIRC WF only showed around 2", there has been a few days that the rain rates were just right and my station was very close to others. UDP still doesn’t pick up light rain but is high moderate rain and really high for heavy rain, the other day we had about .5" of rain but my UDP data was over 2".

You should get identical totals from the UDP broadcasts and the data that makes it to the WF servers, assuming you’re not missing any of the UDP broadcasts on however you’re listening to the network for data. A wired connection is usually preferred for the system you’re listening with.

I never get/got identical totals. UDP is always higher (at least since they starting doing stuff for false rain), today I have .04 and .126 of false rain (bird landings) and similar differences when it rains (difference is much higher if there’s any hard rain, less so if it’s mostly moderate and light rain). Pretty much everything on my network is wired except for mobile devices, my WF hub and some Pi zero w’s once I put them into use (currently my server is listening to the WF UDP). The data on WF servers is almost always less, while my UDP data is almost always higher than other local stations.