Starting out

[Sue Kinzelman]

I'd like yo buy a weather station and share the data with others. Where do I start? I see a lot of them for sale on amazon but how to know which ones would be good? I have wifi that reaches out to most of my yard/fields.

[googl...@tedlum.com]

So, I'm a HUGE Amazon buyer, but not when it comes to weather stations, UNLESS there is an Amazon vendor with a lower price on a station or part. DO NOT try to use Amazon to "select" a station, but do check Amazon for pricing on a station you've already selected. You will typically find the best station catalogs and pricing from dedicated sellers like https://www.scientificsales.com .

Davis stations are the most popular and most prolific stations in this market segment. They are no less durable than any of the others in the class and have been exceptionally well supported directly by the US manufacturer for at least the last 20 years. They also tend to be the most expensive, so are beyond the budget of some. I would start here to level-set your expectations, and go with Davis if it's within your budget. If your budget constraints won't allow that then look to lower cost alternatives. Ambient weather is another popular and respected choice.

Keep in mind that many stations in the market are "Branded" generic Chinese stations - meaning, there is a single Chinese manufacturer and multiple companies who simply slap their own name and brand on them. These tend to be the least expensive but are also fairly "closed", meaning you get ONLY what functionality, weather analysis, and connectivity THEY provide; they don't allow you to export the data in a way that you can use it with other software or share it in ways that were not built into the device. Unfortunately, people who are just entering this market "don't know what they don't yet know" so are more easily taken in by the exciting list of features these stations seem to provide, and are left to find what they don't and can't do after the fact... i.e. many can send data to Wunderground, but not to CWOP, etc..      

Keep in mind that this tends to be a long-term purchase of something that is intended to be exposed to the harsh outdoor environment, with it's high winds, thermal cycling extremes, and degrading UV radiation. So, I would choose a vendor that is not only here today but who is likely to be here 20 years from now, because, since there is a relatively high up-front cost you will need ongoing support to maximize the value of your longer term investment... though, the less you spend the shorter your time horizon.

Wifi is a non-issue with regards to the station itself. All stations should have their own proprietary radio that connects the outdoor station hardware with the indoor console. Now, how the indoor console connects with other things, like the Internet for the sharing of data is another matter and likely does involve Wifi, but there should not be a concern with Internet quality outside of the house itself.

I hate writing long complicated responses that just throw a wet blanket on peoples aspirations, so I'm going to stop here and encourage you to ask us any specific questions that will help you figure out what you need to know to make the right choice. Hopefully this will at least give you a starting point in that. Two things you want to be clear on is 1) Siting and quality of the data, and 2) access to the data and the ability to analyze, share, and archive that data.

[Karen Y]

A lot depends on your budget. If you have several hundred dollars to invest, any of the Davis Instruments stations will give you many years of service. I currently have a wireless Davis Vantage Pro2 connected through an inexpensive computer to our home wifi. That has worked well for me for over 15 years. Prior to that, I had used a Davis Weather Monitor II since the early 90s. The Davis Weather Monitor II still works after 30+ years. It is now in our vacation home, although no longer connected to the internet (only because I would need to figure out how to upgrade the wifi interface, if that is even possible). Other less expensive weather stations work well, but I can’t speak from personal experience to their longevity.

One thing I would look for in any station is the “radiation shield” that Davis has as an option (it may be standard now - not sure). It shields the temperature/humidity sensor from direct sunlight. If you don’t have a location that is in shade all day long, year-round, you will be dealing with temperature spikes during the time It is in direct sunlight. Weather Underground, at least, doesn’t like that, and I found my original station disappearing from their service regularly at those times of the day until I got the radiation shield.

Also, I would try to buy a station where you can separate the rain gauge from the rest of the station. You want your rain gauge easily accessible and only a few feet off the ground for easy cleaning and less wind currents, whereas you want the wind sensor up higher where it can measure wind speed without the ground-level obstructions.

[googl...@tedlum.com]

Karen Y  nails it as far as "Siting and quality of the data", is concerned.

To put this into hard numbers I'd recommend the:

Davis 6253 - Wireless Vantage Pro2 with 24-Hour Fan Aspirated Radiation Shield and WeatherLink Console. $933

This will give you all the basic parameters with detachable wind instruments and the fan aspirated shield.

If you have the money you can add Solar radiation and UV with the:

Davis 6263 - Vantage Pro2 Plus Wireless Weather Station w/WeatherLink Console, 24hr Fan Aspirated Radiation Shield, UV & Solar Sensors. $1,330

but these parameters are not strictly necessary unless you are interested in them or want to calculate evapotranspiration , which would be useful for agricultural purposes if you operate a farm, for example.

If the $933 is too steep you can fall back to:

Davis 6252 - Wireless Vantage Pro2 Weather Station with Standard Radiation Shield and WeatherLink Console. $730.

Which eliminates the active radiation shield and $200 from the price tag at the expense of roughly +/- 2.0°F in temperature accuracy in "average" circumstances.

If all of those are beyond your budget you'd need to fall back to an all-in-one Vantage Vue to stick with Davis. Frankly, I'd probably turn to another vendor before sacrificing the ability to separate the wind instruments from the rain gauge since there are cheaper detachable alternative available from other vendors and the #1 priority is proper siting as long as you can afford even the cheapest NOT-all-in-one station. There are a number of reasons you will come to regret siting your rain gauge 33 feet in the air, if that's what you end up having to do.

Davis stations are modular and the fan aspiration and Solar & UV options can always be added later. There will be an additional cost for one of the WeatherLink interfaces which is required to connects to an external computer or to the internet, which will be the same regardless of the station model.

I know these are big numbers! I purchased the equivalent to the 6263 18 years ago for practically the same amount and now have 5-minute samples of weather data from the last 18 years - 1,884,960 samples. That's about $0.20 per day, or $74.22 per year - the station has been operating non-stop for 6,545 days.

[Sue Kinzelman]

oh my - thank you for all that information everyone.

I was thinking of a budget of $300-$400.. Is that too unreasonable? Do I need to consider spending a lot more?

I am in New Mexico and just about anything I put out on my property gets destroyed by the sun. (So I would need a radiation shield?) I'm a serious gardener and interested in weather.

--
You received this message because you are subscribed to a topic in the Google Groups "wxqc" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/wxqc/hgReb-qDQNg/unsubscribe.
To unsubscribe from this group and all its topics, send an email to wxqc+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/wxqc/22c3b2f9-0fef-42d2-850d-341bef31ebbfn%40googlegroups.com.

[John Tollini]

A +1 for Davis. I have a Vantage Pro2+ that has been up lakefront for 8 years. The only issue i have had to address was the bearings wore out in the anemometer.  $20 and 5 minutes later and all good again. 

The radiation shield is to protect the temperature sensors from direct sun, which leads to wildly inaccurate temps. So it's not protecting the sensor suite from the sun per se, just improving accuracy. 

If the entry price for Davis is out of your price range, check out  ambientweather.com They have much more budget friendly options than Davis. I use their MeteoBridge to get my Davis system online, so I have followed their forums for years.

--
You received this message because you are subscribed to the Google Groups "wxqc" group.
To unsubscribe from this group and stop receiving emails from it, send an email to wxqc+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/wxqc/c8ee5aa9-7ed7-4a52-a367-f1c50b4bbad8n%40googlegroups.com.

[googl...@tedlum.com]

Well, at least you didn't say $100-$150, at which point you'd be down in the bargain basement and probably end up disappointed with all the options.. sooner or later. Your initial range is not unreasonable but isn't going to get you into the range of most Davis stations. Since these tend to be long-term investments I like to think of it in terms of TCO (Total Cost of Ownership) which is your initial investment plus maintenance going forward. Cheaper stations will save you upfront but don't necessarily have a lower TCO because your maintenance costs may be higher and they may reach their end-of-life sooner, and that assumes you're comparing apples-to-apples. If you're comparing an $800 Davis Vantage Pro to an all-in-one $400 station then it's apples-to-oranges before you even look at TCO. You don't need to spend twice as much up front, but I think you'll find it costs you less on the back end if you do.

For what it's worth, I just did a query of the database and of the 13266 stations in there 9157 (69%) are Davis stations, so that loosely implies that when faced with the same question 9,157 went with Davis and 4,109 went a myriad of other directions.

Davis typically handles "repairs" by advance shipping what is usually a completely new part for a fixed "repair" cost, which is usually a small fraction of the new cost of the part and all you have to do is swap it out and return the original part to avoid being billed full price. I don't know of any other vendor that is as reasonable, predictable and still there after 30 years... hope I didn't just jinx it :-) You have to face the fact that regardless of durability over a few decades time you'll need to replace something eventually, there is very little that lasts forever.

The radiation shielding is all about preventing temperature errors due to radiation. The shield itself is plastic and subject to UV degradation - a part of mine only lasted 15 years. Heat is transferred through convection, conduction and radiation. Since you want to measure the air temperature you're only interested in measuring the temperature of the air that contacts the sensor, which without a fan is driven by convection. But radiation from the sun, or nearby heated surfaces, can directly heat the sensor independent of the air temperature through radiation, which results in inaccurate readings. So shielding is put in place to try to block the radiation, all of which is imperfect, but all of which is better than nothing at all. Fan aspiration actively draws air past the sensor and at the same time serves to carry away some of the heat from radiation that gets through the shield, so you get more accurate readings with active ventilation. The amount of the error depends on the level of radiation that's present in the first place. If you leave it to convection (passive ventilation) then heat from radiation will have a greater impact. If this part of the station is going to be out in the open with no natural shade, then you definitely want a station with a good quality shield that is fan aspirated. If this part of the station will be naturally shielded from direct radiation then shielding and ventilation are much less of a concern.

This goes for radiation cooling as well. How is it that dew turns to frost when the air temperature is above freezing? Radiation cooling. On clear nights - no cloud layer - the cold vacuum of space effectively sucks the heat from the ground through direct radiation, so surfaces can be colder than the air around them, including poorly shielded temperature sensors. This is evident on nights with broken cloud layers as temperature variations, as cloud movement randomly blocks radiation cooling.

Hope that helps?

Ted

[tucso...@gmail.com]

Hello Sue

I live in Tucson, AZ and have used Davis weather stations for 30 years. I can say that they will hold up to the intense sunlight that you will receive in New Mrxico.

As far as a recommendation, I'd say go with the Davis Vantage Pro 2 +, non-aspirated. If that exceeds your budget then go with the Vantage Vue. I've used both and they work just fine.

If you want to track Evapotranspiration for your gardening, you do NOT want the wind sensor at 33 feet. 5 to 6 feet is the recommended height for Agricultural Wind and evapotraspiration measurements.

Thanks,

James B

DW4536

[Don Curtis]

For what it's worth, Scientific Sales has the Vantage Vue available now for $489.

You received this message because you are subscribed to the Google Groups "wxqc" group.

To unsubscribe from this group and stop receiving emails from it, send an email to wxqc+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/wxqc/48278c45-5926-4f3a-835e-6d5d16d8163en%40googlegroups.com.

[googl...@tedlum.com]

At Nov 16, 2023, 9:43:10 PM tucso...

... you do NOT want the wind sensor at 33 feet. 5 to 6 feet is the recommended height for Agricultural Wind ...

I'm wondering what source you are quoting, since that is a significant deviation from established norms?

Ten meters is a fairly well established and well accepted norm, and compliance needs to be maintained within a mesonet for the same reason you can't have half the stations reporting altimeter and the other half reporting MSLP.

This paper, "Modeling the Variation of Wind Speed with Height for Agricultural Source Pollution Control", acknowledges the 10m standard and concludes, "... Wind data measured by weather stations at a height of 10m can be easily interpolated to any height between 0 m and 10m using models with the estimated parameters in this research ...".

Regardless, if one sets the height to be something other than the standard 10m then their wind data is not suitable for inclusion in a mesonet dataset, and should not be transmitted. If you need to do something different than standard for your own purposes then you need to keep that data to yourself, since it will taint any aggregate to try to make it a part of.

[tucso...@gmail.com]

I'll see if I can find the sources for Agricultural Wind and Evapotranspiration.

I read your paper and did not see where Agricultural Wind is measured at 10m.

I do agree that Meteorological Wind is measured at 10m.

Thanks,

James B

DW4536

[tucso...@gmail.com]

Hello Ted

Well here's a reference from Davis, spacifically for their GroWeatther Sensor Suite. The GroWeather calculations are used in ALL Davis sensor suites for calculating Evapotranspiration.  

""  Site the sensor suite in a location with good sun exposure throughout the day if the sensor suite is wireless or includes solar radiation or UV radiation sensors.

 For agricultural applications (important for evapotranspiration (ET) calculations): 

• Install the sensor suite and anemometer as a single unit with the radiation shield 5' (1.5 m) above the ground and in the middle of the farm between similar crop types (i.e. two orchards, two vineyards or two row crops), if possible. 

• Avoid areas exposed to extensive or frequent applications of agricultural chemicals which can degrade the sensors. • Avoid installing over bare soil. The ET formula works best when the sensor suite is installed over well-irrigated, regularly mowed grass. 

• If the last three guidelines cannot be met, install the weather station at the edge of the primary crop of interest  "" 

With the top of the radiation shield at 5 ft., that will place tthe anemometer close to 2 meters. The extension for the anemometer looks like the same extension used on my 12 year old Vantage Pro 2+.

Also looking at the references for Davis' Evapotranspiration (ET) calculation, Davis uses the Penman-Monteith Equation for calculating ET0, which calls for a wind height measurement at 2.1 meters.

In a perfect world everybody would conform to CWOP specs and siting requirements, but, as you know, it's not a perfect world. Many folks send in MSLP not Alt. Pressure, many if not most have thier anemometers a approx. 2 meters not 10 meters as required by CWOP and very few calibrate their sensors yearly. Regardless.  NOAA will not use the parameter you send if they think it's not a reasonable value....or that's how I understand it.

Hope this helps,

Thanks,

James B

DW4536

On Friday, November 17, 2023 at 2:05:33 PM UTC-7 googl...@tedlum.com wrote:

[googl...@tedlum.com]

Well, that's interesting. It sounds like their ET calculation may be incompatible. If you use a station in that way you're using it for a dedicated application and you really can't turn around and also submit that data, because it's not fit for purpose, it's fit for a different purpose.

Ten meters is actually an international standard. It's chosen in order to place the sensor above common physical obstructions. There is a constant drag coefficient as you change the elevation, but as you start getting under 10 meters you get a lot more variability not related to pure friction, like buildings and such.

Compliance is always an issue, but not a valid excuse to abandon standards. This argument is a slippery slope/thin end of the wedge fallacy. It's okay to make it worse because it's already bad, is somewhat flawed.

The reality is that MADIS does not omit any data. It enriches it with a quality score which a consumer can use or completely ignore at their discretion. The quality scoring itself can only look for outliers, so if enough stations are non-compliant good data starts looking like the outlier. This is why you can't afford to let the exception become the rule.

Also, ignorance can be remediated with education, but deliberate noncompliance is an insidious issue,

[tucso...@gmail.com]

Hello Ted

I agree.  Looking at the Davis Vantage Pro sensor suite design, with the short extension for the anemometer and of course the Vantage Vue, it seems Davis designs for validating the ET calculation ....assuming an approx. 5 foot tripod. Many other manufacturers have their anemometers close to the other sensors. So it's not just a Davis issue.

My anemometer is at 6 feet and to honest, I can't tell the difference between my winds and the other stations in my local area..... nobody usually agrees with anybody elso  Wind speed variability is much, much more complicated then can be described in theoretical mathematics. Also 10 meters is not a magic height. I and many others live at the foot of a 9000+ ft. mountain with lots of hills and valleys The altitude variation across Tucson is Approx. 1000 ft. So i'm not even sure the 10 meter rule would apply here.

My big complaint is those sending in MSLP. Here in Tucson all stations are above 2000 ft. and many are above 2500 ft with some over 3000 ft. At these altitudes there is a conciderable difference in MSLP and Alt. Pressure and it's clearly visible in the data. But, CWOP and NOAA do nothing about it. For a long time now, I've had the opinion that the number of participants on tthese weather networks is much more important than the quiality of the data.

Well enough ranting,

Thanks for the discussion Ted,

James B

Tucson AZ

DW4536

[Don Curtis]

Here in Colorado EVERY station is above 3,500 feet, my own sits at 4,600 feet. 

😉

You received this message because you are subscribed to the Google Groups "wxqc" group.

To unsubscribe from this group and stop receiving emails from it, send an email to wxqc+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/wxqc/1219ae4e-cf2c-4606-8ed8-6ae776dfa89fn%40googlegroups.com.

[googl...@tedlum.com]

Specific to the ET calculation, I think the reason is correlation of the parametric readings. When you’re building a calculated value from a bag of parametric values it’s important to be sampling the same environment. You’d have a similar problem with a calculated dew point if there was too much distance between the temp sensor and humidity sensor. You would expect there to be a lot of difference between winds at 10 meters and the movement of air that’s protected by a grove of trees or a stand of corn, and what’s relevant for ET is the airflow over the leaves of the plant - whatever is happening at 10-meters is not likely to be relevant, although it likely will be relative.

So, ET air movement is really completely different than meteorological surface winds; the only thing they have in common is the instruments they can be measured with, but they are two different parameters. Just as indoor air temperature and outdoor air temperature are both measured with a thermometer, but obviously represent different things. What matters is what the thing is vs. what you're passing it off as.

What's magic - not really - about 10-meters is it's considered to be more "undisturbed". Yes, it will be disturbed by terrain features, but not as much by man-made structures. I can take a dataset and a topographic map and build a model that has a good fit. But if I start taking lower readings and start building man-made structures the model becomes impossibly complex and "R" (goodness of fit) goes to hell.

Also, due to surface friction you'd need to apply a correction value if you take wind readings at an elevation other than 10-meters. Going back to the, "Modeling the Variation of Wind Speed with Height for Agricultural Source Pollution Control" paper, and using the Power Law model, wind speed of 4.75 m/s taken at 2-meters (6.6') would need to be transformed to ~8.1 m/s in order to submit it as meteorological surface winds. This is because another thing that's magic about 10-meters is the drag transfer function (skin, form and gravity wave drag). Just like pressure: we never report Station Pressure, we always at least add an altitude correction.

Now, if you're in the rocky mountains you've got a whole other thing going on: dynamic viscosity. Air is much less dense at 3,000 feet than it is at MSL, so the drag has less of an effect on the lower viscosity air. But since the station elevation is part of the metadata we can further transform wind using elevation, temperature, and humidity as long as we were provided with TRUE meteorological surface wind taken at, or at least corrected to, 10-meters. Ten meters is magic in that some transfer functions assume that constant where the height of the measurement is germane to the calculation. If you supply something different then the transfer function becomes garbage in, garbage out.

Sending MSLP to CWOP is largely an education issue, and then a software one. First someone needs to be educated to understand their software, which they trust, is getting it wrong. Then there's the matter of how to hack it so it's right, because the problem is not usually the result of a wrong configuration value that can simply be changed. All of which is moot unless there is a desire and willingness to do something about it. All we can really do is education outreach and hope for the best.

[tucso...@gmail.com]

Hello Ted

I think relevant is the key word here. In the Tucson area, it looks like the NWS and DOD report wind at 10m (Meteorological Wind). The County and Forestry Service report at 6.1 m. Looking at the wind speeds reported on Weather Underground, most private stations report wind probably down around  2m, as I do, either because that's where their hardware has the anemometer or, as in my case, I'm interested in ET and my wife wants to know what the wind is when she goes out for her daily walk. She wants to know what wind will be hitting her in the face and cares less what the wind is at 10m

Regardless, I've doubled the magnitude of the wind velocity my server reports (change from 2m to 10m) and I'll watch it for a few months to see if that is a reasonable change. So far it looks like a good change on the CWOP network, but, looks odd on Weather Underground network. My Davis consoles still show the 2m wind speed, so my wife is satisfied.

Thanks,

[tucso...@gmail.com]

Hello Ted

Well, it's seems that the relationship between wind speed and anemometer hieght shown in "Modeling the Variation of Wind Speed with Height for Agricultural Source Pollution Control" does not hold for the data that I'm seeing when I doubled my repored wind speed, estimating a anemometer change from 2m to 10m. It works well for speeds up to around 10mph, but, not for speeds 20 to 30mph. Possibly measured wind spped is a function of both anemometer height and wind speed.  The fact that drag goes as the square of the velocity may have some bearing here. Too bad the authors did not look at higher wind speeds.

Thanks,

James B

DW4536

  

On Sunday, November 19, 2023 at 3:57:51 PM UTC-7 googl...@tedlum.com wrote:

[googl...@tedlum.com]

That paper compares the fit of 5 different models, none of which is simple multiplication, and concludes the power law model demonstrates the best fit. Plus, the models are not expected to account for the myriad of man-made chaos you're probably dealing with, all of which will create direction dependent variation... wind direction aligned with the direction of a residential street will flow more consistently down that channel, but when at right angles it hits the front of houses and picks up speed ( Venturi effect) going around the sides or is deflected up as it goes over the roofs, only to create a vortex pattern behind the obstruction as the velocity suddenly drops on the back side of the restriction. The whole point of the 10-meters is to rise above as much of the chaos as possible. You can only correct for drag mathematically, you can't correct for the chaos anywhere near as easily.

See https://en.wikipedia.org/wiki/Wind_profile_power_law  . 

[Don Curtis]

There's a very easy answer to all this. 

FOLLOW THE STANDARD or don't post wind data. 

You received this message because you are subscribed to the Google Groups "wxqc" group.

To unsubscribe from this group and stop receiving emails from it, send an email to wxqc+uns...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/wxqc/4d9b1c9a-973f-4202-90d0-6bbd73e2c1b8n%40googlegroups.com.