|
19-12-2008, 13:02
|
#1
|
Registered User
Join Date: Mar 2006
Location: San Diego
Boat: Passport 47 CC
Posts: 468
|
Challenge: Interpreting Satellite Imagery
I know alot of folks here know about weather and interpreting wefax maps. I would like to access that expertise and learn how to interpret satellite imagery. Around 9:15 this AM, Pacific Coast Time, NOAA 17 made a pass. I finally got the hardware configured correctly and got one of the best images to date. Turns out the sound input to the Mac was too low and thus the software was not able to code one of the sat's sensors in yesterday's images.
Here are four images. The dual image is the raw data -- no processing. The image to the right of the Raw is a MultiSpectral Analysis that combines channels 3 and 4. I do not know what that means. (Right now, I simply parrot what I read while seeking education.). The image below Raw is an MSA with Precipitation. I do not know the algorithms NOAA uses to detemine how to color precipiatation. It has something to do with (1) density of cloud mass, (2) elevation of cloud mass, and (3) probability curves -- the result is a false colored image of where it might be raining. Th final image is from 6am I caught NOAA 15 for a short period of time -- it imaged a chunk of the Pacific.
Here are some questions I am throwing out.
Can anyone judge wind speed based on the trailings of clouds on the MSA image?
Is the wide swath of air moving from west to northwest -- is that the Jet Stream?
If I was off the coast of Baja and this image came in -- what would be thing to do?
Does that dark and red area mean heavy rain or heavy rain and heavy winds? To me it looks like that cloud mass wants to sit, but it is being pulled against its will in a northwest direction.
In regards to NOAA 15: What am I looking at in the middle where there are no clouds? A high? or a low? If you look at the top of NOAA 15, it looks like everything is spinning clockwise around an invisible mass. Is that a high? A low?
I know folks here really know wefax maps -- I hope you people that know wefax will take a moment to marry what you know of wefax onto these images.
The names of the picture files are like so: The starting number is the NOAA satellite. The initials indicate direction (SB= Southbound, just like the Allman Bros!!) The words say the filters. The ending numbers represent month day time (UTC).
Thanks
Michael
|
|
|
20-12-2008, 04:01
|
#2
|
Senior Cruiser
Join Date: Mar 2003
Location: Thunder Bay, Ontario - 48-29N x 89-20W
Boat: (Cruiser Living On Dirt)
Posts: 51,309
|
I’m not certain this tutorial from NOAA/NESDIS will help you (don’t have time to review the content, but it looks promising)
Satellite Interpretation Tutorials / Image Examples:
Satellite Tutorials and Image Examples (NOAA/NESDIS)
__________________
Gord May
"If you didn't have the time or money to do it right in the first place, when will you get the time/$ to fix it?"
|
|
|
20-12-2008, 10:28
|
#3
|
Moderator Emeritus
Join Date: Sep 2006
Posts: 4,663
|
Take a look at the analysis charts available from the NOAA, so you can compare them to the satellite images. Here is a webpage with a bunch of surface and upper-level charts:
Radiofax Charts - Pt. Reyes, CA
Here is the 500-millibar (upper-air) chart from that site:
You can see the jetstream (around the thick contour line), and it is well to the north of the cloud structures on the satellite image.
Here is another good site for analysis and forecasts:
Ocean Prediction Center
From this page, look at the Pacific surface analysis:
See the three high-pressure regions where the cloudless region is in the sat photo? This is the Pacific High. It shifts daily and seasonally.
Here is the tropical surface analysis (from the first link):
The "scalloped" figures are thunderstorm activity, and correlate nicely with the dark red areas on your sat image. If I were sailing in that location I would prepare for squalls -- but you wouldn't need a satellite image to tell you that!
__________________
Paul Elliott, S/V VALIS - Pacific Seacraft 44 #16 - Friday Harbor, WA
www.sailvalis.com
|
|
|
20-12-2008, 11:19
|
#4
|
Registered User
Join Date: Mar 2006
Location: San Diego
Boat: Passport 47 CC
Posts: 468
|
Great!
Good stuff! Thanks Gord for your efforts to locate tutorials.
Paul, I am going to bite the bullet and buy a small printer for the boat. I can see the way to interpret is to simply compare a wefax map with my photos. Duh. I am embarrassed to say it did not even occur to me to do so! Very nice annotation work -- that helped alot. I
I have read some posts from 5-7 years back on what looks to be defunct sailing sites and it looks like oceanic yacht races almost always have satellite imaging capability on the boats -- is that true? There is almost no discussion of such topic here on or on SCRA. Wished I would have copied those posts -- I was surprised to read how common place the equipment was.
Michael
Michael
|
|
|
20-12-2008, 11:38
|
#5
|
Moderator Emeritus
Join Date: Sep 2006
Posts: 4,663
|
Michael,
There are a bunch of ways to get wfax images, and satellite images for that matter. I don't have direct satellite reception on VALIS, but I used to receive them over HF radio and view them on my laptop. Lately I have been getting wfax and grib chart/data via satphone email, or occasionally via SailMail (HF radio). In my opinion, a printer is nice, but not necessary. I've been using the laptop screen.
I think that since the availability of gribs and wfax-style charts via sailmail/winlink/satphone email, fewer people are directly receiving the shore-station wfax broadcasts. Some of the hardcore racers (and some cruisers) are directly receiving the satellite transmissions in order to get the most up-to-date info, but I for one don't want to work that hard. I think the racers are receiving the QUICKSCAT real-time wind data, which is also available on-line and from some email servers. Here is the QUICKSCAT website: QuikSCAT
I am tempted to get some satellite-reception gear though, just to play with. It looks way-geeky-cool!
__________________
Paul Elliott, S/V VALIS - Pacific Seacraft 44 #16 - Friday Harbor, WA
www.sailvalis.com
|
|
|
20-12-2008, 12:07
|
#6
|
Registered User
Join Date: Mar 2006
Location: San Diego
Boat: Passport 47 CC
Posts: 468
|
Quote:
Originally Posted by Paul Elliott
Michael,
There are a bunch of ways to get wfax images, and satellite images for that matter. I don't have direct satellite reception on VALIS, but I used to receive them over HF radio and view them on my laptop. Lately I have been getting wfax and grib chart/data via satphone email, or occasionally via SailMail (HF radio). In my opinion, a printer is nice, but not necessary. I've been using the laptop screen.
I think that since the availability of gribs and wfax-style charts via sailmail/winlink/satphone email, fewer people are directly receiving the shore-station wfax broadcasts. Some of the hardcore racers (and some cruisers) are directly receiving the satellite transmissions in order to get the most up-to-date info, but I for one don't want to work that hard. I think the racers are receiving the QUICKSCAT real-time wind data, which is also available on-line and from some email servers. Here is the QUICKSCAT website: QuikSCAT
I am tempted to get some satellite-reception gear though, just to play with. It looks way-geeky-cool!
|
When I first asked questions about WEFAX here on CF several months back folks -- and you might have been one of them -- indicated that you can get imagery anywhere, but it is the wefax maps that count, and that you simply can get those through an SSB (inexpensive one at that too with the subsequent discussion of the Kaito unit). However, I just did not quite get that message -- and I went pretty much full circle reinventing the wheel. But that is ok, cause it re-awakened the teenage astronomer / geo scientist wannabe in me. The only reason why I want to get a printer is so I can physically manipulate the images -- hold them and feel them. I really don't need a printer for the boat -- it would simply accelerate my learning.
What is incredibly fascinating to me is how these satellites use hyperspectral sensors -- we have gone so way past anything from the Landstat series. Emissions and absorption spectra are recorded and sampled in microseconds!!
It really is not at all hard to get these images -- and it does not have to be that expensive, but I can see downside to the equipment while sailing. The antenna itself is sturdy, but it is not that sturdy. In a rollover or even a huge crashing wave from the rear, I can see it being jacked up pretty bad.
Volker's company in Germany ( Weather satellite Receivers and NOAA Antennas - Weather Satellite Station, click on hardware on the left side and scroll to see the antenna at sea) has a different quad antenna design. It looks very sturdy compared to mine, but it requires a preamp. I included a picture of it in use on a boat in the Baltic.
Paul: I say go for it! It simply is too much fun -- and the software out there, particularly if you have a PC, offers some unbelievable image manipulations based on combining the raw visual image and the infrared image. There was a Christmas party down here on the dock -- and my Quadrafilar Helix antenna mounted on the pushpit was a definite geek turn on! Besides, you can see these images in 3D -- and the fact is, looking at the clouds and sea just fills you with a sense of beauty and wonderment.
Here is this morning latest pass from NOAA 17. Came in about 20 minutes ago.
Multispectral Analysis with and without precipitation, and a Day Light Sea Temperature Image.
Michael
|
|
|
20-12-2008, 12:50
|
#7
|
Registered User
Join Date: Oct 2008
Location: Princeton, NJ
Boat: Challenger Anacapa 42
Posts: 2,097
|
Michael, do you have info on the antenna in the posted pic? Anything would help... Thanks, Chris
|
|
|
20-12-2008, 13:48
|
#8
|
Moderator Emeritus
Join Date: Sep 2006
Posts: 4,663
|
I'll jump in again: The antenna is this: Weather Satellite Station - Wettersatelliten NOAA Receiver and Antenna, WRAASE electronic GmbH - Hardware - MX-137 (I just followed the link in Michael's post). It looks like a turnstile antenna, essentially two crossed dipoles. They are typically fed through a 90-degree phase-shift network, which gives a circularly-polarized pattern aiming up (and down). You can put the turnstile above a properly-spaced reflector to get a nice pattern for satellite work, but as shown it probably works fine for reasonably strong signals.
It also looks like they are using Rubber Duckies (short helically-loaded quarter-wave whip antennas) for the dipole elements, which keeps the dimensions smaller at the expense of higher loss and/or narrower bandwidth. The built-in preamp should help with the loss, and the bandwidth isn't an issue for this application.
This is all speculation from just looking at the photo, though.
__________________
Paul Elliott, S/V VALIS - Pacific Seacraft 44 #16 - Friday Harbor, WA
www.sailvalis.com
|
|
|
20-12-2008, 14:18
|
#9
|
Registered User
Join Date: Mar 2006
Location: San Diego
Boat: Passport 47 CC
Posts: 468
|
I looked at this antenna: Weather Satellite and Remote Sensing Ground Stations - Dartcom because alot of people really liked it. It looks solid, but Bob at National RF was local and less expensive by almost a couple of hundred dollars.
For a receuver, you can use an FT-817 by Yaesu or something similar. I have read that the images are fairly decent. The problem is that their bandwidth is too narrow as it is optimized for voice. APT satellites use a wide bandwidth, so if you use a scanner or a ham set up, you are going to clip the signal.
This guy describes the construction well: A QFH antenna for the weather satellite band
http://www.skycross.com/Technology/W..._Portables.pdf is utter geek to me, but I did read it only because I get easily annoyed when I dont understand something.
Hamtronics makes a receiver -- and they do the turnstile type as well. Both are built in the USA. APT-2CP Antenna But i think it is way too big for my boat.
|
|
|
20-12-2008, 14:22
|
#10
|
Registered User
Join Date: Mar 2006
Location: San Diego
Boat: Passport 47 CC
Posts: 468
|
This is my antenna. I put some pictures up in my album.
Michael
|
|
|
20-12-2008, 17:06
|
#11
|
Registered User
Join Date: Mar 2006
Location: San Diego
Boat: Passport 47 CC
Posts: 468
|
Quote:
Originally Posted by Paul Elliott
The built-in preamp should help with the loss, and the bandwidth isn't an issue for this application. .
|
Paul
I thought the bandwidth was a particular issue for APT satellites? I was going to get an Icom1500 for this purpose, but several folks said the bandwidth was a little too narrow for the signal.
Michael
|
|
|
21-12-2008, 00:14
|
#12
|
Moderator Emeritus
Join Date: Sep 2006
Posts: 4,663
|
Quote:
Originally Posted by MV
Paul
I thought the bandwidth was a particular issue for APT satellites? I was going to get an Icom1500 for this purpose, but several folks said the bandwidth was a little too narrow for the signal.
Michael
|
The receiver audio bandwidth can be a problem since this circuitry is designed for voice, and the demodulated satellite data probably needs a broader frequency response. I have seen mention of hacking a voice receiver to tap off the demodulator output before it gets filtered by the audio section. Other radios have a "data" output available without the audio filtering. I don't know the specs for the Icom 1500, nor for the satellite data, however.
On the other hand, even a narrow-bandwidth antenna will still be plenty wide enough for this type of data. In fact, the narrower bandwidth will probably help reject interference that might otherwise cause a problem. I see that the antenna I was referring to has a built-in filter for interference rejection. This antenna, even with extra filtering, probably has a bandwidth of several MHz. I would guess that the demodulated satellite data has a bandwidth of 15 KHz or so. An voice-optimized audio channel will have a bandwidth of perhaps 5 KHz.
__________________
Paul Elliott, S/V VALIS - Pacific Seacraft 44 #16 - Friday Harbor, WA
www.sailvalis.com
|
|
|
05-04-2009, 14:51
|
#13
|
Registered User
Join Date: Mar 2009
Location: Virginia and Puerto Rico
Boat: Capri 25 and bareboat charters
Posts: 55
|
Michael
I’m not a meteorologist, but have worked on some of these satellite sensor systems, so let me give it a shot at some of your questions.
Here are four images. The dual image is the raw data -- no processing. The image to the right of the Raw is a MultiSpectral Analysis that combines channels 3 and 4. I do not know what that means. (Right now, I simply parrot what I read while seeking education.).
A channel represents a particular range of frequencies, or wavelengths. I believe the photos you posted are from the NOAA Advanced Very High Resolution Radiometer (AVHRR), which is a six-channel radiometer detecting visible through infrared light.
There’s not a uniform naming system for these channels, but on this radiometer Channel 1 is used the visible part of the spectrum, what you would see with the unaided eye. Other channels are used for different portions of the infrared.
Channels 3 and 4 represent wavelengths from about 3 to 11 microns, a portion of the infrared spectrum sensitive to water vapor in fairly cool regimes that represent high clouds. Through a lot of complex analysis it’s been found that the channels can be combined in various ways, added, subtracted, transformed, etc, to provide even better sensitivity to some factors that affect weather. Combining channels 3 and 4 utilizes data from both of those channels to provide information both on the amount of water vapor and the height of clouds, which indicates that precipitation is likely.
The image below Raw is an MSA with Precipitation. I do not know the algorithms NOAA uses to detemine how to color precipiatation. It has something to do with (1) density of cloud mass, (2) elevation of cloud mass, and (3) probability curves -- the result is a false colored image of where it might be raining.
It’s exactly as you said. The algorithms use the various channels to infer cloud height, thickness, water vapor, etc. From this, a probability of precipitation is calculated from a set of curves and a color is assigned to designate the various probabilities. It's the same as with the colored weather radars, but this is using infrared sensors instead of radar.
It’s a bit of voodoo, since the algorithms are not pure physics or pure meteorology, it’s a combination of whatever mixture of data has been shown to work best when compared with actual weather records. And these algorithms are being constantly worked on and improved.
Can anyone judge wind speed based on the trailings of clouds on the MSA image?
The short answer is no, because the MSA images are measuring heat and water vapor content, rather than wind. But, as you note, one might infer wind speed from the way the clouds trail. However, those trails would be at pretty high altitudes, and may not correlate well with surface winds.
A meteorologist could use these images to recreate weather patterns and predict surface winds, but that’s a meteorological calculation rather than a direct readout from these images.
To measure surface winds on the ocean, we rely on wave patterns. Besides reflecting light, the earth also radiates in the microwave (radio) spectrum, and ocean waves polarize this microwave energy, so by measuring how it changes with time we can get an idea of the surface (waves) that are causing this change.
NOAA has a Windsat on orbit precisely for that mission, but it’s still early technology not yet mature. In the future I expect to see wind speed measured fairly accurately along with precipitation, at least over the oceans, and transmitted from space.
Is the wide swath of air moving from west to northwest -- is that the Jet Stream?
Not necessarily. The wide swath may be just a dry air mass that has little emissivity, so the algorithms leave it blank.
In regards to NOAA 15: What am I looking at in the middle where there are no clouds? A high? or a low? If you look at the top of NOAA 15, it looks like everything is spinning clockwise around an invisible mass. Is that a high? A low?
That’s a meteorological question, but the empty area doesn’t necessarily mean a cloud-free region. These images are thresholded, so there may be clouds that are warm (low) enough to have been thresholded out, or may not have shown up on the visible sensor for a number of reasons. But it does mean an area without a large amount of water vapor, so chances are that it's a cloud-free zone.
__________________
The pessimist complains about the wind; the optimist expects it to change; the realist adjusts the sails.
- William Arthur Ward
|
|
|
05-04-2009, 18:08
|
#14
|
Registered User
Join Date: Sep 2008
Location: NC
Boat: Prout Quest 33
Posts: 43
|
I'm currently Active Duty in the USAF, as a Weather Forecaster.
Lets see if I can quickly answer some of your questions.
To truly interpret your satellite date correctly you have much to learn. You first need to start small and learn basic characteristic's of High's Low's and how the different air masses interact with each other.
You cannot gain a correct determination on wind speed from cloud movement, and if you could it wouldnt relate to your surface winds. As you move up in altitude your winds incease greatly, vearing and backing.
You also need to learn about each of your Jet Streams if you want to find them, the PFJ can be a pain as it likes to split into muliple fingers. The Sub-tropical has a tendance to move up into florida, each jet moving north to south with the change in season.
To Identify cloud hight's you need to learn how to Identify different types of clouds from your Cirus to your Happy Q's. Each has a general altitude.
In regards to NOAA 15: What am I looking at in the middle where there are no clouds? A high? or a low? If you look at the top of NOAA 15, it looks like everything is spinning clockwise around an invisible mass. Is that a high? A low?
To find out what that is, I'd have to compair to a Vis shot, compair it to any other chats I had. Weather Forcasting and Meteorology isn't a perfect science. You also want to learn about Low Pressure systems and how fronts work off them.
I love this science, your constantly learning to things. Good luck!
__________________
""See, this is a sign of your tragic space dementia. All paranoid and crotchety, it breaks the heart.""
: Ship like this, be with ya 'til the day you die.
: Yes Sir. Because it's a deathtrap.
: That's not... you are very much lacking in imagination.
: I imagine that's so, sir.
|
|
|
10-04-2009, 11:39
|
#15
|
Registered User
Join Date: May 2008
Location: Augusta, GA
Boat: Ranger 22, currently saving for a larger cruising boat
Posts: 550
|
See if you can get your hands on a copy "The Annapolis Book of SEAMANSHIP" I borrowed one from the county library and it has a section about weather.
__________________
Sailing and exploration are necessary for life to endure
|
|
|
|
Thread Tools |
Search this Thread |
|
|
Display Modes |
Rate This Thread |
Linear Mode
|
|
Posting Rules
|
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
HTML code is Off
|
|
|
|
Advertise Here
Recent Discussions |
|
|
|
|
|
|
|
|
|
|
|
|
Vendor Spotlight |
|
|
|
|
|