What does everyone think about a feature that maps lightning strikes? For me, it’s somewhat useful to know a lightning strike has occurred close to my home, but what would be really useful is to know WHERE the strike was on a map. Did it happen close to my wife’s office? Close to my kid’s school? Did it happen in a part of a storm that is moving towards my house? Or away? Just knowing that there was a lightning strike without knowing where it was doesn’t really help me and leaves me wanting more.
7 Likes
The WeatherFlow/Tempest system is not accurate enough to give you what you want. You will get better and usefull results using https://map.blitzortung.org.
2 Likes
This is true, but I would add that the app now displays Radar data which isn’t native to WeatherFlow either. I don’t think it’s unreasonable for WeatherFlow to provide a 3rd party map of lightning strikes so that I didn’t have to switch between apps like I used to do for radar data. And to be fair, the distance to the strike data that you see in the app is not coming from your sensor, the sensor only reports a value of an “estimated distance to the front of the storm”. What you see in the app is a value provided by WeatherFlow, so they must already be triangulating the strike locations from 3rd party sources anyways.
Really helpful input; thanks for that. Do you know if there is any way to determine if data we see in the app is coming from the Tempest device I have at my home vs. data that’s coming from another source? I just purchased my Tempest, and I’m often wondering if something as simple as the outside temp reported on the app is what the actual outside temp is at my location vs. the reported temp for my zip code provided by other sources. The reason I wonder is that we’re told to locate the Tempest in a clear area without obstructions – which means the sun is shining on it all day. Tempest customer service has told me that the unit ‘calibrates’ itself to account for this, but I honestly wonder if this is really accurate.
I have another outside temp instrument that’s older but still works, and its manufacturer instructs to locate the device in the shade so that it will give you true air temperature. And sure enough, the reported temp from this instrument was usually 5 or more degrees cooler than what the Tempest reports. To break the tie, I placed a third thermometer I had from inside the house outside, and it also was reporting lower temperatures than the Tempest by 5 or more degrees. Then I reported the difference to Tempest customer services, and immediately after the temps coming from the Tempest started to more closely match my other instrument.
So I’m suspicious if the readings I’m getting on the Tempest app are really from the device at my house, or from data coming from other local weather reporting sources.
You need the unit in a clear area so that the solar panels can get sufficient illumination and, obstructions can affect the wind performance.
For the effects of direct sunshine, https://help.tempest.earth/hc/en-us/articles/360052101413-Tempest-System-FAQs says:
”Do I need a solar radiation shield?
Nope! The Tempest thermometer is located inside the built-in radiation shield. The unique design of this radiation shield itself allows for accurate ambient temperature readings the majority of the time. Air moves through the radiation shield freely while blocking radiative heat sources. Under certain conditions, especially full sun combined with calm or very light winds, solar radiation can lead to elevated readings inside the shield compared to ambient conditions outside. In those cases, Tempest will automatically apply an empirically-derived compensation, using its own solar radiation and wind readings, to keep temperature readings accurate. No additional shielding is necessary!”
Another thing, make sure your station height is the height above ground level (not height above sea level). Pretty sure that is a factor in a lot of the calibrations.
1 Like
Do you know if there is any way to determine if data we see in the app is coming from the Tempest device I have at my home vs. data that’s coming from another source?
Are you familiar with programming at all? The hub broadcasts the raw sensor data over your local network using UDP. Here is a simple python program I wrote that does it. GitHub - MABeatty1978/Simple-WeatherFlow-Python-Listener: Simple Python UDP listener for the WeatherFlow Tempest Weather System
You could use that to display the sensor data and match it up with the app.
But….if you just want to take my word for it. All of the data you see in the app is data from your sensor, with the exception of the lightning strike distance.
Thanks for the reply; really appreciate it.
Unfortunately, I know nothing about programming. More than happy to take your word for it.
What makes me nervous is that often my Tempest is giving me temp and humidity readings that are pretty far off of the temp and humidity readings I get from two other separate sensors I placed outside at my house to compare to the Tempest. However, while the Tempest readings are pretty far off of my other local sensors (whose readings match), the Tempest readings more often match the temp and humidity readings that are reported through other weather data sources (Weather Channel app, local news, etc.). So, since the Tempest also reports the weather forecast and lots of other things in the app that I know don’t come from the device, I was nervous that whenever Tempest’s systems sensed that the device’s readings were way off, the app was just reporting offsite data to me.
But again, happy to take your word for it. Gives me the assurance that I didn’t just waste my money.
Thanks again,
Josh
While this makes some sense, with the amount of Tempest stations that show up on a map, why can’t they be configured to “triangulate” where the strike occurs? Accuracy would vary by area depending on the amount of stations for specific areas but the site you referenced has far less sensors monitoring lightning than Tempest does.
I’m not a programmer/developer but this seems like something that could be done just by the mere number of Tempest stations collecting such data.
The AS3935 lightning sensor in the Tempest has a best-case range of about 32 miles, with a resolution of three miles best case. Having three stations close enough to detect a strike is probably not possible with existing station densities in many areas, and the combined distance error between those stations would not give a decent strike location even if there were several stations in range.
Blitzortung linked above is a crowd-sourced private lightning detection network, that works over large distances using time-of-flight measurements and stations with clocks synchronized by GPS down to nanoseconds. Blitzortung actually included an AS3935 chip in one of their earlier stations as an experiment, and dropped it from later designs because they didn’t find it useful. I’ve written this several times that the AS3935 is NOT a scientific instrument, and any attempt to try and treat it as one is doomed to failure.
As a reference point, my Blitzortung station has detected 3,748 strikes in the past hour, with the nearest being 795 km and the furthest 6,232 km away.
One last point to consider is that a part of WeatherFlow’s “day job” that pays their bills is a commercial lightning sensor network that doesn’t use the Tempest’s lightning data AFAIK. I would not count on them ever implementing a competing lightning sensor network with free access when they already have a paid product that’s paying their bills…
1 Like
I appreciate the well thought out reply @vreihen My initial thought was that it had something to do with the onboard sensor that would not fully accommodate something like a strike map but then thought that an abundances of stations nearby might be able to provide enough accuracy.
1 Like
Here’s what I’d settle for…
I understand the Tempest sensor itself isn’t the best at plotting strikes on a map.
I also like the lightning feed in Tempest One, but it’s way overkill for most homeowners. Homeowners can’t afford to pay $$$$ for Tempest One.
I don’t need “the entire lightning feed” in Tempest One. I just need to look at a map and see nearby lightning to my house where my Tempest is installed.
It would be great if all Tempest owners, as a thank you for sharing your Tempest data with WeatherFlow (since WeatherFlow is able to use your data to power their forecasts and resell it to enterprises), gets a complimentary, fixed map that shows the Tempest Lightning Data network within 40KM of the address of your Tempest. It basically simulates what the Tempest Lightning Sensor could do if it could plot lightning data well on a map, but it also leverages the other Tempest Lightning Network sensors.
I’d easily settle for a static map of the address of my Tempest weather station plus a 40KM radius and be able to quickly place and see when lightning is traveling near me. If I ever needed “the entire lightning feed” and I had the funds, then I could invest in Tempest One.
That’s the only drawback I’ve seen with Tempest at the moment that has kept it from dominating the weather station market. Some of the functionality of its app is underwhelming, but to get at some additional datapoints, Tempest’s other solutions are way overkill for the consumer. It would be great to see a few “middle of the road” tweaks to the Tempest app, then you’ll truly be in a position to sweep the industry off its feet.
Nathan–
I think your proposed solution is sound and more than fair both to Tempest as well as its customers. Hopefully the Tempest folks will take it seriously.
By saying the lack of mapped lightning strikes are the only drawback you’ve seen with the Tempest, does this just include the understood assumption that the rain meter is and always will be inaccurate? Because if not, I think this is the only other major drawback. I feel like Tempest customers have just accepted that Tempest will never accurately measure precipitation, and we’ve just let it go. Is that the case?
Thanks for the comments. I made another post on here chatting about the rain measurements.
I used to work for AEM/Earth Networks/Davis Instruments (for 20 years), so I’ve been in this industry for a long time (my contract with them ended in September. I’m officially on the job hunt if anyone hears of anything). I didn’t get to keep my Davis gear after departing AEM, so I’m officially “all-in” on using a Tempest now and trying to recommend Tempest stations to others. I also plan to keep posting solid feedback to help Tempest dominate this market.
Rain measurements are the other area I’ve noticed my Tempest is still pretty wild on and inaccurate, although the rain alert measurements are nice.
For accurate rainfall, I’ve kept my CoCoRaHS rain gauge around as a supplement to the rain alerts from my Tempest.
For the time being, it would be fun if Tempest pulled in the CoCoRaHS feed and was able to show nearby comparisons of rain data from CoCoRaHS stations so that Tempest customers have some form of comparison to look at inside the app.
Long-term, I could see what other kind of feedback I could provide to Tempest on improving the rain sensor technology, and if Tempest ever needs a new technical consultant aboard their team to help design future sensors, they can always let me know. 