weather forecast with a barometer: feasible? reliable?

[DiTBho]

you need to define what "forecasting" means to you.

What you can predict with just atmospheric pressure and its changes is limited. Sure a relatively fast drop in pressure is usually not a good sign. But you don't know what's really going to happen. Some heavy wind, or rain, or a storm, or... a combination thereof. A slow drop may not mean really bad weather OTOH. It's probably going to be cloudy, but you can go with a low pressure for days on end without a drop of rain.

I think three hours of warning before a storm/rain is *ok* and enough to find shelter. I don't need precise forecast, only the weather forecast for the next three hours in terms of { find a shelter, continue } 

  weather forecast: func(barometer, altimeter, thermometer) -> mission { find a shelter, continue }
              find a shelter(wait for the end of the storm, *) -> mission { go back home, continue }
              *: unexpected, local eval on the environmental data of the moment

[DiTBho]

A humidity meter would be more useful. I doubt the results would be much use for you, and certainly less useful that a decent online forecast - or even using your eyes.

online forecast = internet?

There are areas where WiFi is not good and areas where you don't have WiFi coverage. Forget smartphones there, unless you use a Sat Garmin bridge. I don't own one. This is also the reason why I would like to bring with me an SDR engine (software defined radio): if you don't have WiFi coverage and mobile phone, well, your smartphone is useless, but the SDR could still capture a radio (even AM, FM, etc) signal for your PDA.

I don't want to bring my smartphone with me, however. I don't need to publish photos on social media and in case I have a small camera reflected in my backpack 

-

I need to think about the humidity meter.

[jpanhalt]

online forecast = internet?

The British Met Office broadcasts weather reports (https://www.gov.uk/government/organisations/met-office#:~:text=The%20Met%20Office%20is%20the,%2C%20industry%2C%20agriculture%20and%20commerce. )  I have not listened to them.

The equivalent US service is NOAA .  It broadcasts regularly on special frequencies.  I my area, which is remote, I have an inexpensive, battery operated "NOAA" radio. Our internet and electricity go down regularly with severe storms.

As for a barometer, it can be useful and is certainly more useful than humidity in my opinion.  I've only visited the UK once, which was for a week.  As I recall, the island was divided vertically.  On one side it was rain and fog and the other was fog and rain.  The weather forecast basically told you which side you were on.  (We did arrive on a sunny day and drove South around Salisbury, saw the Magna Carta, then to a small coastal town -- very nice town and tea room -- before going north to Bristol to spend a couple of days before returning to London.  Great visit, the weather was unimpressive compared to the violence we see in our Midwest.)

[james_s]

I visited the UK in late September 2011 and it was beautiful unusually warm sunny weather throughout most of the trip. Then on the last day it shifted to dark and gloomy with gusty winds and pissing down rain. I was told that was more typical of the season.

[CatalinaWOW]

I don't want to take away from your project if you are getting joy from it.  But you have access to the internet since you are on this forum, so you have access to the national weather services.  If you use those well to plan your ride you will be far better off.  You can even modify your route to get better wind conditions.  I know.  I do it.

You just get the local forecasts for a few well spaced locations in your route.  Do it the night before, or even better, in the morning before you depart.  Here in the US they are update roughly every six hours and are really quite good for the next 24 hours, often predicting the arrival of windy conditions within 15-30 minutes.

[tggzzz]

I visited the UK in late September 2011 and it was beautiful unusually warm sunny weather throughout most of the trip. Then on the last day it shifted to dark and gloomy with gusty winds and pissing down rain. I was told that was more typical of the season.

 "Season" implies climate.

The UK, because of the track of the jet stream, doesn't have climate. We have weather.

One turning point in the south and midlands has traditionally been Guy Fawkes' night, because that's around the time when we start wearing clothes for cold damp weather. Over the past decade or so, that's been less true

[DiTBho]

if you are getting joy from it.  But you have access to the internet since you are on this forum, so you have access to the national weather services.

Wrong assumption 

At the moment I'm not in the mountains, I'm in an urban area where the WiFi level is great. Friday afternoon, after work, I will take the car to reach another location on the outskirts, in open countryside, then the bicycle to reach the final destination in the mountains, and - already tested - there will be areas not covered by WiFi.

[james_s]

Is weather radio a thing there? Here we have NOAA weather stations that broadcast weather updates continuously on VHF, from my house I can pick up KHB60 on 162.550 MHz, there are a bunch of them in other locations.

[CatalinaWOW]

if you are getting joy from it.  But you have access to the internet since you are on this forum, so you have access to the national weather services.

Wrong assumption 

At the moment I'm not in the mountains, I'm in an urban area where the WiFi level is great. Friday afternoon, after work, I will take the car to reach another location on the outskirts, in open countryside, then the bicycle to reach the final destination in the mountains, and - already tested - there will be areas not covered by WiFi.

Actually correct assumption.  The weather reports you collect before you leave on Friday will give you better data through the weekend than a barometer/thermometer setup.  If you are going on a two week holiday it would probably be worth a trip into internet land once or twice.

[SiliconWizard]

you need to define what "forecasting" means to you.

What you can predict with just atmospheric pressure and its changes is limited. Sure a relatively fast drop in pressure is usually not a good sign. But you don't know what's really going to happen. Some heavy wind, or rain, or a storm, or... a combination thereof. A slow drop may not mean really bad weather OTOH. It's probably going to be cloudy, but you can go with a low pressure for days on end without a drop of rain.

I think three hours of warning before a storm/rain is *ok* and enough to find shelter. I don't need precise forecast, only the weather forecast for the next three hours in terms of { find a shelter, continue } 

For three hours of warning, if you have minimal experience with mountain weather, it should not be hard to predict just looking at the sky and other cues such as wind.
Yeah, that's how people did it for a very long time.

[james_s]

Mountain weather is notoriously unpredictable even with state of the art equipment, weather balloons, supercomputer simulations, etc. I live in a mountainous region and the forecast has always been a bit of a joke, if you check the 7 day forecast and then check it again 5 minutes later it is not uncommon for it to be different. When you're actually up in the mountains themselves the weather can change very rapidly, I've been on hikes where it changed from clear and sunny to cold and windy with pissing down rain, over just a matter of minutes.

[DiTBho]

ok, I have collected enough information to proceed alone with my project.

[tggzzz]

ok, I have collected enough information to proceed alone with my project.

Before relying on it, you would be wise to spend several months validating its predictions against what actually happens. But you know that.

In mountains, as others have noted, weather varies significantly, unpredictably, and very locally - e.g. raining on one side of the hill but not the other. Local advice can be very useful

[DiTBho]

you would be wise to spend several months validating its predictions against what actually happens

yup, that's the point: I need on field experiments

[SiliconWizard]

Note that specifically detecting a coming storm can be done using a dedicated type of sensor - and I guess the real danger is a thunderstorm, rain is just water.

There are professional and very expensive detectors for that, but an affordable, "hobbyist"-level one is the AS3935, for which you can find relatively inexpensive breakout boards: https://www.sparkfun.com/products/15441

Combining this with a barometer sensor and humidity sensor may get you something helpful after some field experimentation.
If anything, it should be fun.

[CatalinaWOW]

you would be wise to spend several months validating its predictions against what actually happens

yup, that's the point: I need on field experiments

The field experiments are fun, but you can get far more data by matching your algorithm against historical data.  Try multiple locations in your area of interest.  With a bit of poking a lot of this data is available. 


Try the Weather Underground, Ambient Weather, the national weather agencies and TV stations for starters.

[floobydust]

Note that specifically detecting a coming storm can be done using a dedicated type of sensor - and I guess the real danger is a thunderstorm, rain is just water.

There are professional and very expensive detectors for that, but an affordable, "hobbyist"-level one is the AS3935, for which you can find relatively inexpensive breakout boards: https://www.sparkfun.com/products/15441

Combining this with a barometer sensor and humidity sensor may get you something helpful after some field experimentation.
If anything, it should be fun.

Franklin Lightning Sensor IC AS3935 is a FAIL. Many issues but ultimately it false triggers on AM radio. It's tuned to 500kHz.  People had problems with them. It did win many awards lol.

Here's your weather forecast lol  https://earth.nullschool.net/#current/wind/isobaric/500hPa/overlay=temp/orthographic=-80,27,424