EEVblog Electronics Community Forum
Electronics => Projects, Designs, and Technical Stuff => Topic started by: nardev on January 29, 2013, 06:27:06 pm
-
I'm preparing to build small weather station and as it seems, i will have a lot of empty space in my box and around.
So far, this is the plan:
Anemometer
Wind direction
Barometric pressure
Relative humidity
Ambient light
Temperature
Bucket Rain Gauge
1. I'm wondering, what else would be suitable to track/record?
2. Do you have any idea how to track harmful radiations and what are those?
3. Did anybody have experience with measuring air pollution via some sensor/sensors?
4. Have you ever heard of simple and automated way to measure snow height/24h?
5. Is there any way to measure visibility? (maybe futuristic but... i would like if it is possible)
-
You can use a geiger counter to measure CPM (counts per minute) of ionizing radiation. In most places this will be a very boring number, because it almost never changes. In some areas, there is natural radon emission from radioactive decay processes underground, so the reading will change depending on which way the wind blows, and the atmospheric pressure causing more or less radon gas to seep out of the ground.
Unfortunately they are not very cheap, $100 for a kit you have to assemble. https://www.adafruit.com/products/483 (https://www.adafruit.com/products/483)
no doubt cheaper options on ebay though.
You can try to measure atmospheric pollution with sensors that measure various pollutant gases, but unless your pollution is extremely bad, they may not detect much. Here is one example: http://www.pololu.com/catalog/product/1482 (http://www.pololu.com/catalog/product/1482)
I guess "atmospheric visibility" could be done if you point a webcam out the window and there is some distant mountainside or other feature that doesn't change, you could compare the image contrast to a sample image captured on a clear day, to calculate an atmospheric haze index. Probably need to do it at the same time every day so the sunlight is the same, and maybe the illumination difference from cloudy or sunny conditions would prevent it from working.
-
If you're measuring ambiant light, you may want to look at measuring both direct and diffuse (what comes direct from the sun and what comes from everything that's not directly from the sun)
-
If you're measuring ambiant light, you may want to look at measuring both direct and diffuse (what comes direct from the sun and what comes from everything that's not directly from the sun)
Good tip. thnx
-
You can use a geiger counter to measure CPM (counts per minute) of ionizing radiation. In most places this will be a very boring number, because it almost never changes. In some areas, there is natural radon emission from radioactive decay processes underground, so the reading will change depending on which way the wind blows, and the atmospheric pressure causing more or less radon gas to seep out of the ground.
Unfortunately they are not very cheap, $100 for a kit you have to assemble. https://www.adafruit.com/products/483 (https://www.adafruit.com/products/483)
no doubt cheaper options on ebay though.
You can try to measure atmospheric pollution with sensors that measure various pollutant gases, but unless your pollution is extremely bad, they may not detect much. Here is one example: http://www.pololu.com/catalog/product/1482 (http://www.pololu.com/catalog/product/1482)
I guess "atmospheric visibility" could be done if you point a webcam out the window and there is some distant mountainside or other feature that doesn't change, you could compare the image contrast to a sample image captured on a clear day, to calculate an atmospheric haze index. Probably need to do it at the same time every day so the sunlight is the same, and maybe the illumination difference from cloudy or sunny conditions would prevent it from working.
Thank you for the tip.
I didn't know that geiger kit can be so cheap. Thank you for the tip, i believe it will be part of 2. phase.
Concerning visibility. Unfortunately, it would be useless to try to measure with webcam. I have been through various scenarios how to process images for that, and it would not be useful at all.
Anyway, what i find out is that i need "Forward-Scatter Visibility Sensor". But the problem is that it seems it's not equipment that some can just easily get.
I fond this sweet article: http://ops.fhwa.dot.gov/weather/mitigating_impacts/interactive_ess.htm (http://ops.fhwa.dot.gov/weather/mitigating_impacts/interactive_ess.htm)
Also, few other gas sensors: http://www.futurlec.com/Gas_Sensors.shtml (http://www.futurlec.com/Gas_Sensors.shtml)
-
Nice setup so far. Well, you already heard about the radiation (visible) measurement.
you could buy a sensor or build one yourself. If you have any access to a "professional" weather station around, you could maybe check your DIY stuff against their sensors.
When you measure the solar radiation from the sky and reflected from the ground, you can calculate the albedo for the ground. We use a Kipp Zonen CNR4 and a CNR1 which both have a layout like that and measure Visible and IR radiation.
About the pollution it can be tricky. There is some sensor made by Sanyo I think which can measure dust traces, but I guess it's not that accurate. Regarding the forward scatter visibility sensor.... there are some patents online where you can check how they work (check out for the Vaisala ones), but they are pretty expensive. The cheapest ones we found are about 1000Euro. Standard systems from Campbell Sci or Biral are about 2000-3000Euro.
Visibility is nice, but tends to be pretty expensive and isn't measured too much.
How do you want to measure rel. humidity and temperature? There are some nice sensors from Sensirion which are not too expensive for the accuracy. Or you can build a psychrometer with two PT100. That would be the most accurate way to measure it, but need some maintenance from time to time.
The bucket rain gauge is nice, but make sure to keep it clean and that no leaves or insects get into it.
What to measure else: It depends on interests I think, but a leaf wetness grid is interesting and easy to make. Just two conductive grids which decrease their resistance to each other when water is building up on them (from fog or dew)
Last but not least. How do you want the station to be running? Datalogger, Computer, DIY with micrcontroller? Is it near your house or will it need to be battery/solar powered?
-
Thank you for replies.
Most likely, i'll use BeagleBone as controller. I have been thinking about Arduino Mega, Rapsberry Pi or so. But so far, BeagleBone is my choice.
Concerning insects, it was my worry too. Also what i'm afraid of is thunder.
For insects, I'm not sure about those who fly :D Others could be easily blocked by chemicals around joints holding base box.
For thunders, hmm, i'll have to think out something. the location is very near airport, (~800m away) and it's ~25m tall building. Building is well grounded but i never had any experience with this before. I guess i need to read more about it.
-
Pollution readings are moslt a concern for those in Beijing, currently rather a problem.
-
We operated a station on a steel tower in the Amazonian rainforest for some years..... We didn't have any problems with thunderstorms there, even if it was that exposed.
With our station here nearby it's also no problem. It is on an open field with a 10m Alumiumium mast. There are no problems either.
Sure, you could have bad luck, but I guess it's not that much of an issue as far as you'll not mount it directly on an fully exposed area somewhere.
Be careful to make sure that you have a good grounded connection between your sensors, your board and earth. You'll often have to handle mV with some sensors.
As you said the location is near an Airport.... a small regional one for hobbyist or a bit bigger? You can easily get reference data online for that location. Just search for METAR data. There are some sites which also decode them for you. They give you visibility, temperature, cloud height and weather conditions. Maybe they also provide more data... You can have a look here http://en.allmetsat.com/ (http://en.allmetsat.com/)
I was also thinking about a private, basic, weather station, but I considered the Rhaspberry Pi, as that one has a fully functional linux and can easily be integrated in a network. If you are going with an Arduino or so, I would consider DIY sensors, at least for the analog ones. You can easily make an illumination sensor. For rH&T I would suggest digital sensors via I2C. I already mentioned the SHT sensors. They come in a cheaper basic version and a high accuracy, expensive, version, but can be swapped against each other I guess.
-
I definitely planned to use data from an airport as referenced. (I'm quite lucky, three of my good friends are working as air control :) )
BeagleBoard has much more ports. Has Digital and Analog ports. Also needs less power etc.
I have Raspberry pi too, great platform, but it will be second option now.
Because i mentioned, i just find nice table comparing Rpi and BeagleBone
http://digitaldiner.blogspot.com/2012/10/arduino-uno-vs-beaglebone-vs-raspberry.html (http://digitaldiner.blogspot.com/2012/10/arduino-uno-vs-beaglebone-vs-raspberry.html)
(http://1.bp.blogspot.com/-tBGrQRdfv3k/UIoS8luRf-I/AAAAAAAABJ0/mylfuLYOEME/s400/Screen+shot+2012-10-25+at+9.33.34+PM.png)
Main purpose and goal is to make it good platform to learn something. Not only me but maybe someone after me too.
(*)So i would rather go with good sensors that require serial or I2C connection, make it all log properly, build simple php API on the platform for broadcasting the data etc...
And of course, as i'm making it i would like to have as precise as possible sensors that my pocket can provide under already mentioned condition(*).
-
The June 2011 issue of Elektor had a radiation measuring project that used a cheap diode.
It might also be interesting to measure sound. Perhaps listening to specific frequencies to track bird or bat activity.
-
I'd look at monitoring UV radiation.
As far as insects go, I have to clean the cobwebs out of my weather station every so often. I rarely get insects in the rain gauge collector.
-
Yes, definitely UV level tracking. Preferably with at least two spectral measurement points for long and shortwave UV.
Considering that if the ozone layer ever does pack it in, most surface plant life would be dead within weeks, if not days. And you definitely would want to know about temporary extremes so you could avoid going outside.
Another thing to look into might be this:
http://webflash.ess.washington.edu/ (http://webflash.ess.washington.edu/)
WWLLN World Wide Lightning Location Network (wwlln.net)
Quote:
-------------------
How it works
We welcome offers of hosting a new WWLLN sensor to add to the list above. All hosts receive all the world-wide data for their own research on monthly CDs. In return, each host provides the computer and meets any local expenses like power, Internet, and maintenance. However, do not think that a sensor on your own campus is going to give you lightning location data on its own. Only the whole network does that.
Each lightning stroke location requires the time of group arrival (TOGA) from a least 5 WWLLN sensors. These sensors may be several thousand km distant from the stroke. The geographical arrangement of the sensors is important: a lightning stroke which is enclosed by sensors is much more accurately located than one which is not so enclosed. Clearly a uniform spacing of sensors around the Earth is the ideal. Since the Earth is round, there are no edges: every lightning stroke is surrounded by sensors, but not necessarily by the sensors which sense it. Typically only about 15 to 30% of strokes detected by one sensor are detected by 5 or more. These strokes are usually the stronger ones. Recent research indicates our detection efficiency for strokes about 30 kA is approximately 30% globally.
To cover the whole world by sensors spaced uniformly about 1000 km apart would require roughly 500 sensors. If spaced 3000 km apart, we would need “only” around 50 to 60 sensors. Presently we have 40 WWLLN sensors, and we are in the process of expanding to 60 sensors within the next year or two.
-------------
Something related: the atmospheric electrostatic potential. There's a volts/meter standing charge gradient, that varies.
Google: measuring the atmospheric charge gradient
Another lightning related project that could be quite fun - a lightning range detector. It could include a light pulse detector, radio energy detector, and a sound detector. Try to correlate the flashes and bangs in software, and calculate distance based on velocity of sound in air.
Having both a log of local atmospheric charge gradient, and lightning times & ranges would be cool.
-
The June 2011 issue of Elektor had a radiation measuring project that used a cheap diode.
It might also be interesting to measure sound. Perhaps listening to specific frequencies to track bird or bat activity.
Wou, sound pollution :D That is nice idea :D
-
Yes, definitely UV level tracking. Preferably with at least two spectral measurement points for long and shortwave UV.
Considering that if the ozone layer ever does pack it in, most surface plant life would be dead within weeks, if not days. And you definitely would want to know about temporary extremes so you could avoid going outside.
Another thing to look into might be this:
http://webflash.ess.washington.edu/ (http://webflash.ess.washington.edu/)
WWLLN World Wide Lightning Location Network (wwlln.net)
Quote:
-------------------
How it works
We welcome offers of hosting a new WWLLN sensor to add to the list above. All hosts receive all the world-wide data for their own research on monthly CDs. In return, each host provides the computer and meets any local expenses like power, Internet, and maintenance. However, do not think that a sensor on your own campus is going to give you lightning location data on its own. Only the whole network does that.
Each lightning stroke location requires the time of group arrival (TOGA) from a least 5 WWLLN sensors. These sensors may be several thousand km distant from the stroke. The geographical arrangement of the sensors is important: a lightning stroke which is enclosed by sensors is much more accurately located than one which is not so enclosed. Clearly a uniform spacing of sensors around the Earth is the ideal. Since the Earth is round, there are no edges: every lightning stroke is surrounded by sensors, but not necessarily by the sensors which sense it. Typically only about 15 to 30% of strokes detected by one sensor are detected by 5 or more. These strokes are usually the stronger ones. Recent research indicates our detection efficiency for strokes about 30 kA is approximately 30% globally.
To cover the whole world by sensors spaced uniformly about 1000 km apart would require roughly 500 sensors. If spaced 3000 km apart, we would need “only” around 50 to 60 sensors. Presently we have 40 WWLLN sensors, and we are in the process of expanding to 60 sensors within the next year or two.
-------------
Something related: the atmospheric electrostatic potential. There's a volts/meter standing charge gradient, that varies.
Google: measuring the atmospheric charge gradient
Another lightning related project that could be quite fun - a lightning range detector. It could include a light pulse detector, radio energy detector, and a sound detector. Try to correlate the flashes and bangs in software, and calculate distance based on velocity of sound in air.
Having both a log of local atmospheric charge gradient, and lightning times & ranges would be cool.
OMG i didn't know ideas can go this far :)
-
I'd look at monitoring UV radiation.
As far as insects go, I have to clean the cobwebs out of my weather station every so often. I rarely get insects in the rain gauge collector.
Did you try to treat rain gauge collector with some insecticide?
-
I'd look at monitoring UV radiation.
As far as insects go, I have to clean the cobwebs out of my weather station every so often. I rarely get insects in the rain gauge collector.
Did you try to treat rain gauge collector with some insecticide?
Any insecticide will wash off in the first rain. The spiders tend to weave webs inside the raingauge where, contrary to what you'd expect, it''s quite dry. Too many webs and dead insects and the tipping bucket can stick. A quick clean if necessary every 6 months does the job.
-
Just to add one very good web shop for weather station equipment :)
http://www.ambientweather.com/ (http://www.ambientweather.com/)
-
You could try to measure the Schumann resonances: http://en.wikipedia.org/wiki/Schumann_resonances (http://en.wikipedia.org/wiki/Schumann_resonances) Some hobbyists published projects how they did it, and there are professional physicists who did it (e.g. http://www.ann-geophys.net/27/3497/2009/angeo-27-3497-2009.pdf (http://www.ann-geophys.net/27/3497/2009/angeo-27-3497-2009.pdf) ). But can't be measured near power lines, cars, motors etc.
-
Another thing to measure.
http://www.unihedron.com/projects/darksky/ (http://www.unihedron.com/projects/darksky/)
-
I wanted to use a webcam to detect when the local rubbish burner was spewing pollution over my house but I know little about image processing I'm afraid.
A webcam sounds too complicated for this. Maybe using something like in smoke detectors? LED and photodiode, isolated from ambient light, but the air can flow around.
-
Could also look at atmospheric currents (alongside voltage gradients) - I don't know how the two compare in practice but I watched the current going up today as a shower approached. Could be interesting to explore.
-
Grab 2 ionisation smoke detectors and use the one as is for pollution detection, and take the other and remove the Alpha source and use it as a field detector with a short external well insulated stub antenna.
-
Pollen detection
Air quality detection
Oxygen sensor
Carbon monoxide sensor
Soil humidity
Time dat, sunrise, sundown, phase of moon.
-
Pollen detection
Air quality detection
Oxygen sensor
Carbon monoxide sensor
Soil humidity
Time dat, sunrise, sundown, phase of moon.
Do you have any idea how to catch this: "Pollen detection"??? Seems to be nice idea since some people are allergic to pollen.
-
Temperature is (as far as I remember) measured by dry and wet temperature, when the wet is with some kind of watered fabric on the meter. Also, cloud base if you are able to. Lightning detection is nice. This can be either acoustic, or even inductive. And there is snow, which is different from rain.
(http://www.cookislands.org.uk/image/Tai's%20weather%20rock.jpg)
-
You could install a tiny solar panel and measure the solar energy reaching the surface of the earth, or to put it another way the amount of daylight. I wonder if some kind of colourometer could be used to detect when it seems "grey" outside, which is known to have an affect on human beings. Could even be used to determine the amount of cloud cover vs. blue sky, or the colour of the clouds.
We already have solar panel but goal is to get bigger for battery charge :D but we can log that too.
-
Snow depth. An ultrasonic sensor should probably work.
-
Earth magnetic flux and seismic events.
Why do you care about weather anyway?
-
There is an interesting method employed at some telescope sites to measure cloudiness using the generated signal from a Peltier element in reverse. Check this out: http://www.noao.edu/staff/gillespie/projects/cloud-detector.html (http://www.noao.edu/staff/gillespie/projects/cloud-detector.html)
It could be an interesting application for a weather station as well.
The description from the original inventor in an email:
---clip---
Attached is information on an operational cloud sensor system at Fairborn
Observatory, forwarded from Louis Boyd.
-----Original Message-----
From: Lou Boyd [mailto:boyd@...]
Sent: Thursday, October 12, 2000 6:24 PM
To: Lucas, Gene; mpml@egroups.com
Subject: Re: {MPML} Re: Automated observatory -- Cloud Sensor approaches
The cloud detector Gene described has been in constant use at the
Fairborn observatory since 1984. It's output is combined with a wind
speed detector, a raindrop detector, a humidity sensor, and a sky
brightness sensor to make the decision to allow the roofs to open.
Actual opening time is determined by calculating the solar zenith
angle. The roofs open at 96 degrees. The telescopes take data at 102
degrees or greater solar zenith angle.
The cloud detector consists of two 8" dia 1/4" thick aluminum plates
painted with white epoxy paint on the outside surfaces. That has good
emissivity around 10 microns but is fairly reflective of sunlight. There
is a 6" diameter 3/16" thick O ring to keep moisture away from the
peltier device. There are six nylon screws with stainless acorn nuts to
clamp the plates together and hold them against the peltier device
coupled with silicone heat sink compound. The output of the peltier
device is a voltage which varies with the difference in temperature of
the plates but is very insensitive to the average temperature. The
voltage goes to an op amp as the difference between clouds and no clouds
is in the millivolt range and then to an A-D converter which is
montiored by the roof control computer. Actually a simple comparitor
would be sufficient as I use a fixed threshold, but it's informative to
check the variations over the night.
The cloud detector does not detect high thin clouds but will detect any
clouds capable of producing precipitation. As Gene mentioned it can be
fooled by snow, but the raindrop detector senses snow too. Statistical
tests on the photometric data from the telescopes is about the best
detector of thin clouds which exists. I've tried better 10 micron
detectors which are very sensitive to water vapor but that didn't
correlate very well to the visible wavelength transparency of thin
clouds.
--
Lou Boyd - Director
Fairborn Observatory
-
With temperature and humidity you can calculate the dew point. Then use it in this formula: Cloud Base Altitude = ((((temperature - dew point) / 4.5) * 1000) + measure station altitude) to find the height of the base of the cloud layer. Or to find that there are no clouds. Insolation (sun brightness) sensors made things easy when the calculations were done on slide rules, but these days computers make it possible to calculate the values from simpler data.
Here's my weather page: http://manilaweather.webatu.com/index.htm (http://manilaweather.webatu.com/index.htm)
I have sensors for temperature, humidity, barometric pressure, wind speed and direction, and rainfall. That's all it takes.
-
Earth magnetic flux and seismic events.
Why do you care about weather anyway?
It's a good project to learn:
1. construct casing and the rest of hardware
2. how some sensors are dependent and how do they function
3. how things are hooked into one system (electronics)
4. how to use "internet of things" kind a thing, APIs or make web interfaces etc..
5. always upgradeable
6. connects students with different interests...
etc..
-
smog detectin ?
air quality detection ?
lightning ranging (analog devices has a very neat chip to do that )
uv detection
sunrise, moonrise
moonphase
tides
humidity
seismic activity
distance from earth to sun
tilt angle of earth
don't forget calender
-
smog detectin ?
air quality detection ?
lightning ranging (analog devices has a very neat chip to do that )
uv detection
sunrise, moonrise
moonphase
tides
humidity
seismic activity
distance from earth to sun
tilt angle of earth
don't forget calender
Thank you for this list, i particularly like tides, moonphase and distance from earth.. haven't think about that at all.. could be interesting to calculate that..
-
This sounds like it is going to be one sweet weather station when it is finished! Much more exiting than my old wind speed/direction, rain, and temp station that I still have sitting outside :P
-
sunrise, moonrise
moonphase
distance from earth to sun
tilt angle of earth
Why do you need to measure this? Can be calculated, until something big enters the solar system.
-
sunrise, moonrise
moonphase
distance from earth to sun
tilt angle of earth
Why do you need to measure this? Can be calculated, until something big enters the solar system.
Of course that wouldn't be measured but will be calculated and aggregated with the rest of data.
-
But measuring the tilt angle of earth might be interesting, probably with some telescope, webcam and star detection software. You never know, some experiments at the LHC might change it:
http://www.lhcfeed.com (http://www.lhcfeed.com)
-
Any progress to report on your weatherstation?
-
G'day Nardev!!
How is it all coming along? A mate and I are also building a weather station and are tossing up a few ideas about what to use for the anemometer and wind direction, one idea to pick up speed is to use a savonius rotor on an optical encoder, another way was to drive a constant current through a known resistance and measure the voltage driving it. But that gets effected by ambient temp and sunlight and so on. Also gets a bit power hungry!
Would you mind if I asked what you ended up using to measure wind speed and direction?
-
Look up a basic solar radiation sensor. We use them for calculating evapotransporation for plant watering. It's now becoming required on many sites.
-
G'day Nardev!!
How is it all coming along? A mate and I are also building a weather station and are tossing up a few ideas about what to use for the anemometer and wind direction, one idea to pick up speed is to use a savonius rotor on an optical encoder, another way was to drive a constant current through a known resistance and measure the voltage driving it. But that gets effected by ambient temp and sunlight and so on. Also gets a bit power hungry!
Would you mind if I asked what you ended up using to measure wind speed and direction?
Sorry for so late response, this was the choice on the end: https://www.sparkfun.com/products/8942 (https://www.sparkfun.com/products/8942)
-
That looks like it's all done for you except to read the values. I read your initial post and I thought the intent was to DIY. I'm not criticizing your choice. I'm interested about how/why you decided. I've wanted to really make my own wx station for a long time. I've never liked the pre-made stations for various reasons. Price, dependability, fragility, durability.
For example the cloud detector mentioned here sounds really interesting and might be a multipurpose detector. i.e. it could do double, or triple duty. Clouds, snow, rain, maybe even particulate matter (a.k.a. smog).
Those pictures on sparfkun look strange??
-
That looks like it's all done for you except to read the values. I read your initial post and I thought the intent was to DIY. I'm not criticizing your choice. I'm interested about how/why you decided. I've wanted to really make my own wx station for a long time. I've never liked the pre-made stations for various reasons. Price, dependability, fragility, durability.
For example the cloud detector mentioned here sounds really interesting and might be a multipurpose detector. i.e. it could do double, or triple duty. Clouds, snow, rain, maybe even particulate matter (a.k.a. smog).
Those pictures on sparfkun look strange??
No, the pictures are ok. Yes it's just plug, read and bit calculate before you have your data. Not so straight forward because you get only mechanics. it's not redial/digital read.
Concerning the smog, you can get various gas sensors cheaply but what i really wanted and i couldn't make is visibility sensor. even on airports they use simple methods of looking through window and looking marks alongside runway.
-
No, the pictures are ok.
They look sub ok.
Yes it's just plug, read and bit calculate before you have your data. Not so straight forward because you get only mechanics. it's not redial/digital read.
Just not very interesting.
Concerning the smog, you can get various gas sensors cheaply but what i really wanted and i couldn't make is visibility sensor. even on airports they use simple methods of looking through window and looking marks alongside runway.
I think you're incorrectly reaching a conclusion. You're saying that because airports don't use a sensor then a visibility sensor must not be simple to do. First you have to define what visibility means. For a runway it's specific and below 1/4 mile I think they can shut the runway down. Is it RVR (runway visible range)? Of course for VFR (visual flight rules) it's got to be clear. For IFR different rules apply plus there needs to be an ILS and other things I can't remember.
Visibility over a wide area is just a wild ass guess basically because it might be clear in spots and thick in others.
-
NO
I explored that field about measuring of visibility and it's quite tricky!!!
There are two most common approaches. I forgot the names basically one is laser beam that is reaching aligned pointers over the certain distance, mostly used in airports but has certain limitations and is quite complicated to install, needs a lot of equipment etc.
Second one is stand with two arms. Basically in one arm is a sensor and in another is a light source. And the sensor is calculating how reduced light source has been and according to that it's calculating visibility. And of course that is not goo enough because fog can be concentrated easily in on very small area and not evenly distributed.
I think that i have very very very good sources to find out what is used and why in an airports. I live next to one international airport SJJ and i have 4 VERY GOOD friends working in a control tower and regional flight control so i think that i know what i'm talking about!!!
-
I explored that field about measuring of visibility and it's quite tricky!!!
Tricky is relative, you have to define what visibility means.
You want RVR or visibility for a region.
There are two most common approaches.
If I recall there's only one system approved for RVR, plus the tower can override.
I think that i have very very very good sources to find out what is used and why in an airports.
Living next to airport and having friends working in ATC has nothing to do with the technology in use at the airport.
My point was that it's more fun to think up new and novel approaches to measuring visibility. I would try using sound waves and that's just one idea. Don't go high tech, go low tech. Simpler is better.
-
Well, some of those guys are in fact tech support and maintain their equipment... I think that it's enough good source to make decision.