Jump to content


Photo

Clear Sky Charts Accuracy

  • Please log in to reply
37 replies to this topic

#1 Bill McNeal

Bill McNeal

    Ranger 4

  • -----
  • Posts: 337
  • Joined: 07 Oct 2007

Posted 27 July 2013 - 01:34 AM

CSCs are very helpful and the best observing weather resource I've found,, but are they reliable? Anyone encounter errors or discrepancy in the weather forecast, or are they consistently accurate?

#2 Starman81

Starman81

    Vanguard

  • *****
  • Moderators
  • Posts: 2047
  • Joined: 06 Mar 2008
  • Loc: Metro Detroit, MI, USA

Posted 27 July 2013 - 01:42 AM

They are a great at-a-glance starting point for determining whether conditions will be favorable for observing, this is where they shine. The next step for me is to always check AccuWeather, specifically the hourly forecast. AccuWeather also does their own 'Astronomy' forecast which is somewhat useful as well because you can take a look at a week at a time, unlike CSC which is only 2 days (~45 hrs), though it's not that informative overall by itself but it is still helpful I am always looking to plan observing as much in advance as possible.

#3 frito

frito

    Apollo

  • -----
  • Posts: 1183
  • Joined: 05 Oct 2012
  • Loc: Fremont, CA

Posted 27 July 2013 - 01:50 AM

in my experince its a good guide, the weather is the weather. i've had great seeing when CSC said i'd have average and vise versa. transparency is somewhat harder to judge. overall I use it to tell me what the general condition forcast should be, if i'm really interested to know what is going on with my local forcast i will turn to convential weather sites as well as raw data feeds from the NOAA GOES sat's they are where the info ultimately comes from and the real time data is there to see

http://www.goes.noaa.gov/

#4 GlennLeDrew

GlennLeDrew

    Voyager 1

  • *****
  • Posts: 10823
  • Joined: 17 Jun 2008
  • Loc: Ottawa, Ontario, Canada

Posted 27 July 2013 - 04:06 AM

The CSC is based on the same computer model output used as the foundation for public and aviation forecasts. The latter products do involve forecasters' brains, whereas the CSC is *just* the computer generated prediction.

#5 Tony Flanders

Tony Flanders

    Voyager 1

  • *****
  • Posts: 11204
  • Joined: 18 May 2006
  • Loc: Cambridge, MA, USA

Posted 27 July 2013 - 05:22 AM

Anyone encounter errors or discrepancy in the weather forecast, or are they consistently accurate?


Surely you jest! Have you never seen or listened to a weather forecast before? Have you ever found one that is consistently accurate? Can't be done with the current state of technology, and very likely can't be done ever, for theoretical reasons.

The CSC is a fine and useful tool, but often a little wrong and occasionally very wrong, just like every other weather forecast.

#6 City Kid

City Kid

    Vanguard

  • *****
  • Posts: 2357
  • Joined: 06 May 2009
  • Loc: Northern Indiana

Posted 27 July 2013 - 05:35 AM

I have found them to be fairly accurate most of the time. Like others have said, we're talking about the weather here so it's impossible to be 100% accurate but they do a pretty good job. It seems that high, thin clouds are more often than not the source of inaccuracies when it's not as transparent as the CSC predicts.

#7 Cotts

Cotts

    Just Wondering

  • *****
  • Posts: 4886
  • Joined: 10 Oct 2005
  • Loc: Toronto, Ontario

Posted 27 July 2013 - 09:15 AM

I like CSC very much.

I also am a fan of Skippy sky which has a nice feature: it distinguishes between low, middle and high cloud.

If you live in the LakeErie, Lake Ontario region this site is fabulous. It has all the radars, satellites etc. that you could ask for.

It's all still a bit of a coin-toss though....

Dave

#8 Seldom

Seldom

    Viking 1

  • -----
  • Posts: 818
  • Joined: 05 Aug 2012
  • Loc: N of Cedar City Light Dome

Posted 27 July 2013 - 09:38 AM

You really need a bunch of weather sites to compare. I just added Skippy Sky to mine.

http://www.intellica.../JetStream.aspx is nice to know. So's this http://weather.gc.ca...ta/prog/regi...

Seems we owe a great debt to Canadian weather forecasters, but what I've been seeing for the past two weeks just makes me depressed. :(

#9 BrooksObs

BrooksObs

    Viking 1

  • -----
  • Posts: 875
  • Joined: 08 Dec 2012

Posted 27 July 2013 - 10:14 AM

I would give CSC about a 60% accuracy rate when it calls for clear evening skies for the NE United States.

I cannot comment precisely on elsewhere, but locate situations/terrain in the Northeast seems to strongly skew its prediction accuracy. I would note that CSC seems to me to be more accurate for daytime conditions than evening ones around here.

As somewhat of an amateur meteorologist myself, ANY "predictions" beyond 48 hours these days seems little more than a WAG. Often I can see decided flaws in the official NWS forecasts for my region. Incidentally, I've seen the NWS predict 8 clear nights here over the past two weeks. There were exactly none until last evening!

BrooksObs

#10 StrangeDejavu

StrangeDejavu

    Surveyor 1

  • *****
  • Posts: 1846
  • Joined: 20 Mar 2013
  • Loc: Florida

Posted 27 July 2013 - 10:26 AM

I have learned that the CSC is a good indicator, but ultimately one needs to rely on their own vision. I've been burned by the CSC twice when I was first starting out.

The first time, the CSC showed 5/5 Cloud Cover, 4/5 Transparency, 5/5 Seeing. My initial reaction of course was: :bigshock: I set up the light shade, bug sprayed up, let my Dob cooldown and that entire night, clouds filled every inch of the sky.

It goes both ways, though. The second time, the CSC showed mostly white blocks, so I didn't even bother going outside to check. It wasn't until I woke up to use the restroom at around 2:00am that I popped outside and saw 5/5 conditions. Distressed that I missed out on all that time, I quickly set up and took advantage of it until sunrise, where I went back to sleep. :p

It can be accurate some times, but i've learned to check the skies myself each night. You can usually stare at a bright star for 5-10 seconds and observe how bad it's "twinkling". The amount of "twinkle" tells you how steady atmospheric conditions are; little to no twinkle is preferred for DSO. I've yet to learn a good way to gauge Seeing, so I still turn to the CSC for that. :waytogo:

#11 John_G

John_G

    Messenger

  • -----
  • Posts: 474
  • Joined: 18 Jan 2010

Posted 27 July 2013 - 10:37 AM

I like it for giving me an idea on which way the clouds are rolling. I find it very reliable most nights.

#12 GlennLeDrew

GlennLeDrew

    Voyager 1

  • *****
  • Posts: 10823
  • Joined: 17 Jun 2008
  • Loc: Ottawa, Ontario, Canada

Posted 27 July 2013 - 12:04 PM

Mitch,
Your description of the way you use the CSC, where you *seem* (?) to rely primarily or exclusively on the time-series hourly blocks, prods me to issue this *important* note to all users:

Click on at least one of those cloud cover blocks to see a forecast map of your whole part of the country.

The time series blocks represent merely a *single pixel* taken from that image. You must see the bigger picture. Do this on a few occasions and you'll soon enough see why.

And by the way, those map images can be animated.

#13 StrangeDejavu

StrangeDejavu

    Surveyor 1

  • *****
  • Posts: 1846
  • Joined: 20 Mar 2013
  • Loc: Florida

Posted 27 July 2013 - 12:14 PM

Click on at least one of those cloud cover blocks to see a forecast map of your whole part of the country.


:bow:

#14 WarmWeatherGuy

WarmWeatherGuy

    Surveyor 1

  • *****
  • Posts: 1912
  • Joined: 27 Aug 2011
  • Loc: Orlando, FL 28° N, 81° W

Posted 27 July 2013 - 12:19 PM

CSCs are very helpful and the best observing weather resource I've found,, but are they reliable? Anyone encounter errors or discrepancy in the weather forecast, or are they consistently accurate?


If you're referring to ClearDarkSky.com then here is what I've found here in Florida. Whatever the predicted humidity the actual humidity will be 100%. If it says the humidity will be 60% then expect dew (always expect dew).

If you click on a little square you will see a map for that time. The map will show a contour of different levels of whatever the square was (cloud cover, seeing, etc.) and a big plus sign showing your location. If you are on the edge of a contour then the prediction is less likely to be accurate. If your plus sign is in the center of a region all the same color then the prediction is more likely to be right.

Also, you can animate the map or single step forwards and backwards in time. Usually the first part of the animation has already happened. You are seeing a prediction of what has already happened. That prediction was made before it happened though and if it is way off from what eventually happened then the rest is less likely to be accurate.

#15 Ed Wiley

Ed Wiley

    Apollo

  • -----
  • Posts: 1025
  • Joined: 18 May 2005
  • Loc: Kansas, USA

Posted 27 July 2013 - 12:20 PM

2+ on Skippy Sky!

Ed

#16 REC

REC

    Fly Me to the Moon

  • *****
  • Posts: 5339
  • Joined: 20 Oct 2010
  • Loc: NC

Posted 27 July 2013 - 01:10 PM

David, thanks for the tip....just added it to use along with CSC

Bob

#17 GeneT

GeneT

    Ely Kid

  • *****
  • Posts: 12782
  • Joined: 07 Nov 2008
  • Loc: South Texas

Posted 27 July 2013 - 03:09 PM

I find them very helpful, but often wrong. Last week, I had just received four new eyepieces. The Clear Sky Charts indicated that viewing would be good in San Antonio. I headed out to my pretty good dark sky site and set up. At sunset, the clouds and haze began to clear. I was getting ramped up. Then, 30 minutes after sunset, the clouds slowly thickened. An hour after sunset, the only thing I could see was Arcturus, through a sucker hole. I tested out the eyepieces to ensure no defects, and that there would be plenty of infocus on one of them.

Any other evening, I would have been really upset because the predictors all said the the evening skies would be clear, to include Clear Sky Charts. However, the evening was not a waste because I was able to give my new eyepieces a check ride.

#18 JayinUT

JayinUT

    I'm not Sleepy

  • *****
  • Posts: 3933
  • Joined: 19 Sep 2008
  • Loc: Utah

Posted 27 July 2013 - 04:57 PM

Clear Sky is just one of several sites I use. I feel strong if your going to be a serious visual observer you have to become an amateur weather forecaster for the sites you use. Time/experience/knowing how to use some websites helps one to know if a site is good to use or not. Here are the sites.
1 SkippySky Astronomy

2. NWS Activity Planner I enter my data that I want to capture with the latitude and longitude of my site's location and I get this information, link. This information is about 90% accurate.

3. Intellicast provides a quick look at info for that day and a few days ahead. Matches with 2 above for me.

4. Next Gen Weather Lab. I like this site because I can click on region on the left side, go to my region and select visible for that satellite image, vapor which for me is critical to know.

6. NWS Mesonet Reports. This is critical in making a determination of how a site is looking. I pick a site near to my observing site and can read wind, direction, temp., relative humidity and dew point, It helps me to know what to wear, will I need dew protection, do I camp over or come home etc., as I look at the current and historical data by examining the site and clicking on it etc.

7. wunderground.com is a forecast site that is relatively accurate.

#19 herrointment

herrointment

    Aurora

  • *****
  • Posts: 4777
  • Joined: 12 Mar 2011
  • Loc: North of Hwy. 64

Posted 27 July 2013 - 04:57 PM

I have always felt that the accuracy of the CSC was dependent on the users knowledge of current weather beyond what is presented in the chart and the users ability to interpret the data that is in the CSC.

It's a great personalized product and Mr. Danko offers it to you at no charge. I would hate to be without it.

#20 star drop

star drop

    contra contrail

  • *****
  • Moderators
  • Posts: 72055
  • Joined: 02 Feb 2008
  • Loc: Snow Plop, WNY

Posted 27 July 2013 - 11:23 PM

None of the forecast models really work in my area. The Clear Sky Chart gives me a heads up that the sky might be clear. The seeing is guaranteed to always be worse than predicted. Then it drops back to past experience and hourly checking by eye.

Example: In the winter the Doppler radar in Buffalo sees a few flurries thirty miles out in my direction. At my location twenty miles further out it might be clear, flurries or three inches of snow per hour for hours on end.

#21 Tony Flanders

Tony Flanders

    Voyager 1

  • *****
  • Posts: 11204
  • Joined: 18 May 2006
  • Loc: Cambridge, MA, USA

Posted 28 July 2013 - 05:40 AM

I find them very helpful, but often wrong.


Yep, that's it in a nutshell.

Last week, I had just received four new eyepieces. The Clear Sky Charts indicated that viewing would be good in San Antonio.


Well, that's a special case. A human forecaster, knowing that you had just bought new equipment, would have forecast clouds regardless of what the weather models said. But you can hardly expect that from a computer . . .

#22 sg6

sg6

    Viking 1

  • *****
  • Posts: 795
  • Joined: 14 Feb 2010
  • Loc: Norfolk, UK.

Posted 28 July 2013 - 06:49 AM

A useful guide.
They are not produced for the spot that you will be observing from. Not sure of the US but that is usually not comprehended here in the UK. :foreheadslap:

What is clear? Listening to a program a few weeks back and one of the weather presenters or the the Met Office happened to drop the comment that less then 1mm of rainfall was not considerded as rain. So it can be cloudy, "rain" and you can get wet, but if less then 1mm fell then it wasn't rain and so I suspect you didn't get wet. :help:

Does Clear mean a particular time? If it is not clear at say 10:00pm and you go to sleep, but is clears at say 02:00 then was it a clear night? :question:

What to a Met Office is "clear"? I can often see stars but only the brighter ones. Does high thin cloud/moisture that acts as a slight veil not lumpy cloud rate as clear?

I am still amazed by the number of people that simply do not look outside themselves. I have read many complaints of: "The forecast said cloud, then when I looked out before bed it was clear and I missed an evenings viewing." are many and numerous.

#23 BrooksObs

BrooksObs

    Viking 1

  • -----
  • Posts: 875
  • Joined: 08 Dec 2012

Posted 28 July 2013 - 07:36 AM

If one wishes to talk about forecasters' misuse of terminology these days, I can offer an interesting situation that speaks to their growing uncertainty about what is going on.

It used to be that if a clear night was predicted, the forecast would distinctly employ the term "Clear". Now, almost without exception, I see this has been replaced by "Mostly Clear", or "Generally Clear", even when clear skies can unquestionably be anticipated. It's kinda hedging your bets, I suppose.

Likewise, I find only one TV channel (a very local one) that is willing to acknowledge that the presence of considerable cirrus renders the sky no longer "Clear". The other all seem to disregard the fact. I've also seen the major forecasting channels report local conditions as clear even though the entire sky is already solid with cirrus and there is no possibility of change in many hours to come.

I have a good friend who is an on-camera professional meteorologist and we do have some very interesting exchanges sometimes!

BrooksObs

#24 GlennLeDrew

GlennLeDrew

    Voyager 1

  • *****
  • Posts: 10823
  • Joined: 17 Jun 2008
  • Loc: Ottawa, Ontario, Canada

Posted 28 July 2013 - 02:44 PM

There is a distinction between forecasts for the aviation community and the public. In the latter case, thin cirrus cloud, through which stars can be seen, is treated as clear.

But the computer models upon which the CSC is based does consider cirrus. And forecasters are very tuned into all cloud types and coverage. Just be aware of the simplifications made in the public forecast.

#25 Madratter

Madratter

    Fly Me to the Moon

  • *****
  • Posts: 6737
  • Joined: 14 Jan 2013

Posted 28 July 2013 - 05:28 PM

I find them very useful. BUT, you have to remember that weather predictions are based on a very course grid of input data. There aren't weather stations every mile, etc.

Weather is also a chaotic system meaning that small changes in inputs make for significant changes in the outputs. That is exacerbated by the first problem, but even with very fine grids would still be problematic.

Don't forget a computer actually has to be able to finish the forecast in a useful amount of time as well. So even if we had perfect input information, you couldn't use it.

On top of that, from a global point of view a miss on where the cloud line is going to be by 5 miles is a very accurate forecast. But if you happen to be near the line, as I was the other night, the forecast can be completely wrong for you.

We aren't stirring livers anymore, but the tendency is to have rather unrealistic expectations about weather predictions.

I can tell you that aviation forecasts even 2 hours out are imperfect enough to have operational impacts.

This stuff is REAL hard to get completely right.

EDIT: Just to relate something else, THE hardest course I took in college was physical oceanography. It was dealing with things like the physics of ocean currents, etc. The physics of weather is very similar.

The guys that build these models are very very smart and I have a great deal of respect for them.






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics