Sign of weather modification or sampling/recording/transmiting error at NOAA images?

I would like to know weather expert opinion on this images. I'm sorry for external links - I can't create ATSmedia account for some reason. But I downloaded this pictures on HDD and will post them somewhere later.

So let me know if you expect sinus like curves in clouds:




Other images can be found on this address www.chmi.cz

Above linked images are from this batch, line CZ:

29.11.2009 14:59UTC (NOAA16):
EU: | NM | RGB124 | RGB345 | b1 | b2 | b3 | b4 | b4BT |
CE: | NM | RGB124 | RGB345 | b1 | b2 | b3 | b4 | b4BT |
CZ: | NM | RGB124 | RGB345 | b1 | b2 | b3 | b4 | b4BT |
CZ2: | NM | RGB124 | RGB345 | b1 | b2 | b3 | b4 | b4BT |

Any ideas?

These are also interesting:




EDIT to add:
Same area with animation possibility - just press number of frames to start it:
eumetsat.org
Can somebody do automatic screen capture of the animation? My notebook can't handle such big resolution (just 1024x768 display).

[edit on 29-11-2009 by zeddissad]

I'm not really sure what you're asking about in those images but there are indeed such a thing as wave clouds.
www.weatherquestions.com...

Originally posted by Phage
I'm not really sure what you're asking about in those images but there are indeed such a thing as wave clouds.
www.weatherquestions.com...

I think this is something different. You are referring to this effect probably wikipedia.org. It is quite common at this part of year here and I can see it perfectly on this satellite servers almost every day, behind the mountains chains - but it do not produce sinus curves as on pictures above.

EDIT to add:
Here are satellite images of this:

Orographic lee cirrus and rotor cloud. Strong flow over a mountain ridge can create a variety of phenomena. Here we see a rotor cloud above east coast of Spain, and a lee cirrus plume further downwind. Besides these, you can also see here wave clouds in central Spain, and a small vortex east of Straits of Gibraltar. The animation (0.9 MB) shows its short life.

chmi.cz

[edit on 29-11-2009 by zeddissad]

reply to post by zeddissad

Ok. When you said "sinous" I thought you mean an actual sine wave type shape. I think I see what you are talking about.

The AVHRR images are not actual images, they are representations of radar data. I think the zigzag lines are a result of data processing. Do you see similar patterns in the infrared or visible images of the area?

Originally posted by Phage
reply to post by zeddissad

 

Ok. When you said "sinous" I thought you mean an actual sine wave type shape. I think I see what you are talking about.

The AVHRR images are not actual images, they are representations of radar data. I think the zigzag lines are a result of data processing. Do you see similar patterns in the infrared or visible images of the area?

Sorry for spelling. Yes I was talking about "sine wave type shape" at north-western part of image.
Hmmm... I'm not sure if I fully understand you. Radar is for me ground based emitter/receiver - and I know images produced by such weather-radars (rain "density"). Technology implemented in AVHRR is for sure derived from basic "radar" idea. Let speak wiki - here it can be reliable source:

AVHRR instruments measure the reflectance of the Earth in 5 relatively wide (by today's standards) spectral bands. The first two are centred around the red (0.6 micrometer) and near-infrared (0.9 micrometer) regions, the third one is located around 3.5 micrometer, and the last two sample the thermal radiation emitted by the planet, around 11 and 12 micrometers, respectively.

wikipedia.org

So these are IMO visible red or infrared images from satellite "camera".




I'm not sure, but I think AVHRR have now fully digital chain of processing - this type of sine wave I would expect with old analogue technology.

EDIT: typo

[edit on 29-11-2009 by zeddissad]

Here is actual operational status of NOAA-16 craft:

Component Description Status
BASEPLATE AVHRR baseplate Green
CHANNEL 1 Spectral channel #1, 0.58-0.68 micro meters. Green
CHANNEL 2 Spectral channel #2, 0.725-1.0 micro meters. Green
CHANNEL 3A Spectral channel #3A, 1.58-1.64 micro meters. Green
CHANNEL 3B Spectral channel #3B, 3.55-3.93 micro meters. Green
CHANNEL 4 Spectral channel #4, 10.3-11.3 micro meters. Green
CHANNEL 5 Spectral channel #5, 11.5-12.5 micro meters. Green
ELECTRONICS AVHRR Electronics Green
OPTICS AVHRR Optical mirrors and telescopes Green
PSU Power Supply Unit Green
RADIANT COOLER AVHRR Radiant Cooler Green
SCAN MOTOR AVHRR Scan Motor Yellow

It also list actual frequency - not all NOOA crafts are equipped identically.

So b4 image is according to wikipedia: "the thermal radiation emitted by the planet, around 11 and 12 micrometers", while b3 image is infrared and b2 visible red.

reply to post by zeddissad

Yes, I was wrong. I said radar instead of radiometric. But these are, as I said, processed images.

There are four modes of data processing and transmission. Two of interest:

3.1.3.1 Automatic Picture Transmission (APT)

Any two of the six AVHRR/3 channels can be command-selected for processing. This data undergoes the following:

1. Resolution reduction by using every third scan line of AHVRR data.
2. Geometric correction to reduce the perspective effect due to the Earth's curvature and the satellite altitude. Details are contained in Section 4.2.
3. Digital to Analog - The digitally processed APT data are converted to a 2080 Hz bandwidth analog signal, amplitude modulated onto a 2.4 kHz carrier, and bandwidth limited to 4160 Hz in preparation for transmission by the VHF transmitters.

3.1.3.3 High Resolution Picture Transmission (HRPT)

MIRP produces the HRPT and the LAC outputs by combining unprocessed AVHRR/3 data with TIP data. The basic frame rate and data rate of the HRPT is compatible with the AVHRR/3. Therefore, only buffering is required to construct the HRPT frame from the AVHRR/3 data. However, since the TIP frame rate is only one third that required to fill the HRPT frame, the TIP data must be repeated three times. This is accomplished by repeating the five TIP minor frames (104 words each) in each of three HRPT frames. As in GAC, two parity bits are added to the 8-bit TIP word to form the 10-bit MIRP word. The HRPT output is supplied to the S-band transmitter input control for real-time transmission.

If the images are APT, the skipping of two scan lines and geometric corrections could be the cause of the zigzag effect.

www2.ncdc.noaa.gov...



[edit on 11/29/2009 by Phage]

reply to post by Phage


Resonance between scanline and mapping/geometric correction definitively can produce such effect. I would expect they will solve such trivial problems when they send something to space.

1. Resolution reduction by using every third scan line of AHVRR data.
2. Geometric correction to reduce the perspective effect due to the Earth's curvature and the satellite altitude. Details are contained in Section 4.2.
3. Digital to Analog - The digitally processed APT data are converted to a 2080 Hz bandwidth analog signal, amplitude modulated onto a 2.4 kHz carrier, and bandwidth limited to 4160 Hz in preparation for transmission by the VHF transmitters.

Errors in transmission we can rule out - it will produce degradation of decoded output but not such zig-zag pattern.

As I understand it, the site linked is buying processed images from Eumetsat. So I suppose that APT is not used here. I would expect APT receiver on ship, not at EU weather center.

EDIT to add:
There is also another possibility
"SCAN MOTOR AVHRR Scan Motor Yellow" = something noncritical with mechanics. Vibration in mechanics of mirror can maybe initiate such effect. But then I would expect it in all channels - not only in two of them.

[edit on 29-11-2009 by zeddissad]

reply to post by zeddissad

I disagree that dropping two thirds of the available scan lines will not produce the zigzag.

I try not to make assumptions about other aspects of the data processing when I don't know the details but it looks like data processing to me.

I think this is an HRPT image.
imkhp2.physik.uni-karlsruhe.de...

Originally posted by Phage
reply to post by zeddissad

 

I disagree that dropping two thirds of the available scan lines will not produce the zigzag.
...

I'm sorry for my poor English but I think sentence:

Resonance between scanline and mapping/geometric correction definitively can produce such effect.

Mean, that "dropping two thirds of the available scan lines" in combination with mapping/geometric correction, CAN be reason of this effect - I agree with you (after 10 years of professional video editing it was first think which came into my mind).

Image, you posted, is composite image roughly in same resolution as I posted. There is no sign that it was transmited via better technology. Still some combination of image processing methods is probably in place. You are right - we don't know.

It's Monday now so I'll send them email - maybe they will explain it.

My common sense say me that it is just some bad work by overpaid nut in some agency. My "conspiracy me" is saying proof!

reply to post by Phage


Now I'm sure that images which I presented are also gained via HRPT (High Resolution Picture Transmission).
chmi.cz
Page is in Czech so trust me or use some translation tool. We still can't rule out some mapping issue but we can say, that sine pattern is NOT related to intentional scan lines omission.

EDIT: syntax and syntax


[edit on 30-11-2009 by zeddissad]

[edit on 30-11-2009 by zeddissad]

Just for reference. Here is picture with error in transmition apparent in lower right part. This error is apparent in all channels - it is consistent with serial mode of communication. It also show IMO real scan line direction - real path of recording ray.

what you guys are seeing in the NW portion of the image (NW in relation to the image borders... i dont know the exact relation to true or magnetic north lets be clear) is a result of the resolution deteriorating as the imaged object is farther away in one portion of the image than in the other.

for instance, if my resolution is .5km (DMSP) at 830km away with my scan being perpendicular to the surface of a sphere, on the edges of my image where the earth "falls away" or curves away, my resolution rapidly decreases. in other words, in an area that my scanner thinks is 0.5km wide, i may be covering 2 or 3 or 5km of space. when the image is corrected and the data displayed, it causes a pixelation effect that can be seen when you examine the images in the OP closely.

another thing to consider, the color or value of the pixel is an average, so when there are blotchy clouds you can end up with pixelated wave-like patterns.