Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Nightly Cloud Cover Prediction

  1. May 18, 2013 #1
    Hi, I was wondering if I would, or would not, be able to predict the cloud cover for a night using a minimum of simple instruments during the day. If so, how and what instruments? Obviously I could just look at a satellite image or something, but I was wondering how I could do it without advanced instruments like that. This is for amateur astronomy. Thanks.
  2. jcsd
  3. May 19, 2013 #2


    User Avatar
    Gold Member

  4. May 19, 2013 #3


    User Avatar

    Staff: Mentor

    The biggest factor is humidity. A barometer will tell you if the weather is changing though.

    I'd just use www.cleardarksky.com It is an excellent astronomy forecast.
  5. May 19, 2013 #4
    russ_waters that is certainly an awesome internet tool, I will use that for sure. Really through just for educational purposes I'd like to develop the skill of forecasting, if its even possible without more sophisticated equipment. I'd heard of barometers being used for general weather forecasts, but wasn't sure if cloud cover specifically could be predicted based on a few local measurements. So you're saying if I measure pressure and humidity every hour I might be able to infer something?
  6. May 19, 2013 #5
    You really can't forecast much from local observations only. A rising/falling barometer, changes in wind direction and just observing the sky is where weather forecasting was in the 18th century. You really need to understand the motion of weather systems over a large area to give reasonably accurate forecasts. If you're mainly interested in astronomy, your most recent local weather forecast will be better than anything you do in your home or backyard.
    Last edited: May 19, 2013
  7. May 19, 2013 #6


    User Avatar

    Staff: Mentor

    If you will look at the satellite imagery you will see how cloud cover moves across the continents. That's not something you can predict from local observations - you can have a pretty weather where you are, but a thick sheet of clouds can be just 30 minutes away from you. Your local data often can tell you something is changing, but it won't give you any details about the future.
  8. May 19, 2013 #7
    Ok, sounds like it isn't possible just to use a few local instruments then. So, what would the absolute minimum be to make reasonable judgments about cloud cover? How large an area would I need data from? What kind of calculations could I do with a calculator or simple spreadsheet to assist this? Ultimately I was hoping there could be a way that myself and my high school astronomy students could do this ourselves (or as by ourselves as possible). Thanks.
  9. May 19, 2013 #8

    Are you in the US? If you're doing this as a hobby you can get data sent to your computer. I'm not sure exactly what kind of personal website you'll need. However you're competing against supercomputers and the vast data bases and expertise of NOAA. The EU has a similar agency.

    For a 24 hour forecast, you'll need to know about fronts within about 250-500 mi or 400-800 km from you. You'll need to know upper level winds and ground moisture. In the summer, in many places, a warm sunny day in an area with high ground moisture ( estimated from local humidity or directly) will produce thunder storms. These usually clear out with cooler temperatures at night, but not always. Local conditions can also produce low clouds over hills or ground fog in low lying areas. High clouds are usually associated with approaching fronts. You need to know the expected surface wind direction and changes associated with approaching weather systems in your area.

    If you're doing this for fun, fine. But if you need a reliable forecast, your local weather forecast is your best bet. It's never 100% accurate, but it will very likely be much more accurate on average than your best efforts.
    Last edited: May 19, 2013
  10. May 19, 2013 #9
    How far in advance do you want to know the cloud cover? If it is a matter of hours then it is usually safe to assume that the weather conditions now will be the same as in the future.

    There exist several techniques to forecast temperatures for a day or two into the future based on local observations but I am not aware of any for clouds. These techniques often assume that the air mass you're in doesn't change (but you can't tell if it will just from local obs)

    In order to make accurate weather forecasts for a day or more, you need lots of data, certainly on a continental scale and preferably on a global scale. This then needs to be assimilated into a complex numerical weather model and run on a supercomputer to generate the forecast. Even with all the resources of a national weather centre, forecasts typically not accurate further than a week in advance for numerous reasons and are often not reliable on very local scales either.

    Perhaps you would do better to simply interpret the weather charts produced by your national weather service.
  11. May 19, 2013 #10


    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    hahaha that made me laugh ! :)

    on more than one occasion, I have had the scope out for a night of viewing and I have see the sky go from perfectly clear to totally cloudy inside 30 minutes.... leaving me, as I put the scope away, muttering ... where the *O@E#H)* did that come from ??!!!

  12. May 20, 2013 #11
    Usually safe ;)
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook