Aequinoctium problems (?)


Ernst Lobsiger
 

Dear All,

for some time I get ugly GEO images twice a day (see attached GOES16 images from yesterday).
I think that these are a result of sunrise and sunset near Aequinoctium (sun in the sat orbit plane).

Provided my above assumption is right: How long will this be going on?

Why do I only see this on Himawari-8 AHI and GOES-16/17/18 ABI but not on MSG2/3/4 Seviri?

Best regards,
Ernst


David J Taylor GM8ARV 🏴󠁧󠁢󠁳󠁣󠁴󠁿 🇪🇺
 

On 28/09/2022 13:15, Ernst Lobsiger via groups.io wrote:
Dear All,

for some time I get ugly GEO images twice a day (see attached GOES16 images
from yesterday).
I think that these are a result of sunrise and sunset near Aequinoctium (sun in
the sat orbit plane).

Provided my above assumption is right: How long will this be going on?

Why do I only see this on Himawari-8 AHI and GOES-16/17/18 ABI but not on
MSG2/3/4 Seviri?

Best regards,
Ernst
Ernst,

See:
https://www.satsignal.eu/software/GOES-ABI-Manager.html

for other examples. There are similar effects on SEVIRI, but that is on a
spinning, not a starting, platform. The NOAA messages will likely mention the
expected duration. You can subscribe here: ESPC.Notification@....

Duration would be similar to the Seviri effects.

Cheers,
David
--
SatSignal Software - Quality software for you
Web: https://www.satsignal.eu
Email: david-taylor@...
Twitter: @gm8arv


Ernst Lobsiger
 

Hello David,

thanks for confirming my assumptions. The point is that I have not
noticed that on my *.webm movies until recently. Your examples date
from 2018-March-30. So this was after spring Aequinoctium which is
normally around March-20. That's why the problematic bits are above
(north of) the equator. The recent Aequinoctium was 2022-September-23
so my example from 4 days later already shows the problem south of
the equator. But indeed I discovered nothing similar on my MSG2 and
MSG4 full globe *.webm movies (even nothing on 2022-September-23).

From what I see on GOES-XY and Himawari-8 movies I expect to
see this effect for at least two more weeks. Then the sun at local
midnight should be below the ABI and AHI fields of view. This is
probably given by the declination of the sun IIRC my astronomy.

Here are the *.webm movies of GEO sats for the last couple of days:

http://5.153.116.86:86

Best regards,
Ernst



geojohnt@...
 

Hello Ernst, David and All,

As I recall from casual observations GOES satellites are far more affected by sunlight intrusion into the radiometer than MSG.
It will be interesting to see if the new MTG satellites which are three axis stabilised (compared to MSG spin stabilised) suffer from 'more' light intrusion.

Whilst talking about the 'moving sun,' we are all coming up to the autumn Sun/satellite/dish co-linearity season.
A good time to see if anything is partially shadowing your dish location.

Regards,
John




-----Original Message-----
From: Ernst Lobsiger via groups.io <ernst.lobsiger@...>
To: MSG-1@groups.io
Sent: Wed, 28 Sep 2022 17:06
Subject: Re: [MSG-1] Aequinoctium problems (?)

Hello David,

thanks for confirming my assumptions. The point is that I have not
noticed that on my *.webm movies until recently. Your examples date
from 2018-March-30. So this was after spring Aequinoctium which is
normally around March-20. That's why the problematic bits are above
(north of) the equator. The recent Aequinoctium was 2022-September-23
so my example from 4 days later already shows the problem south of
the equator. But indeed I discovered nothing similar on my MSG2 and
MSG4 full globe *.webm movies (even nothing on 2022-September-23).

From what I see on GOES-XY and Himawari-8 movies I expect to
see this effect for at least two more weeks. Then the sun at local
midnight should be below the ABI and AHI fields of view. This is
probably given by the declination of the sun IIRC my astronomy.

Here are the *.webm movies of GEO sats for the last couple of days:

http://5.153.116.86:86

Best regards,
Ernst



Ernst Lobsiger
 

Dear All,

to keep my aging brain going I did a little math.
Yesterday I had the sun disturbing ABI at 08:20
(west) and 09:20 (east) UTC. Taking the radius of
mother Earth Re=6378km and the radius of a GEO
orbit Ro=42164km and estimating the problem zone
at about -30° latitude (south) I get an angle left to
right as seen from GOES-18 approximately with:

Alfa = 2 * atan((Re * cos(-30)) / Ro) = 14.9°

As we all know mother Earth is turning 15° per
hour, which fits with 08:20 to 09:20 UTC nicely.

Cheers,
Ernst


Ernst Lobsiger
 

On Thu, Sep 29, 2022 at 02:15 AM, <geojohnt@...> wrote:
 
Whilst talking about the 'moving sun,' we are all coming up to the autumn Sun/satellite/dish co-linearity season.
A good time to see if anything is partially shadowing your dish location.
 
Regards,
John
 
John and Robert,

in a long MSG-1 thread I tried to explain what the sun can do. See:

https://groups.io/g/MSG-1/message/33160

As a reminder at the time of solar outage your antenna should not
show shadows of nearby trees or buildings. Below table applies to
Robert Moore's antenna position. Solar outage is expected strongest
2022-10-12 11:01:15 and 2022-10-13 11:00:44. This is also the time
where Robert's system most probably looses lock. If your minimum SNR
is earlier or later then remember that the sun moves 15° per hour or
0.25° per minutes. This allows for an independent dish azimuth check.
Maybe Robert can tell us at what date/time the first shadows of his famous
trees reach the dish. It will show the free space under his antenna beam.

For those who have a Python environment with pyorbital installed
I attach the small bit that made the table below. It's very easy to
adapt for other antenna positions and later solar outages in 2023.


Cheers,
Ernst

P.S: All times are UTC of course ...


Robert Moore's Pos Longitude -3.244306 Latitude 53.279917
---------------------------------------------------------
Robert Moore's Dish Opt Azimuth 163.6, Opt Elevation 27.9
---------------------------------------------------------
2022-10-01 11:07:19 Sun Azimuth 163.6, Sun Elevation 32.3
2022-10-02 11:06:44 Sun Azimuth 163.6, Sun Elevation 31.9
2022-10-03 11:06:10 Sun Azimuth 163.6, Sun Elevation 31.5
2022-10-04 11:05:36 Sun Azimuth 163.6, Sun Elevation 31.1
2022-10-05 11:05:02 Sun Azimuth 163.6, Sun Elevation 30.7
2022-10-06 11:04:28 Sun Azimuth 163.6, Sun Elevation 30.3
2022-10-07 11:03:54 Sun Azimuth 163.6, Sun Elevation 29.9
2022-10-08 11:03:22 Sun Azimuth 163.6, Sun Elevation 29.5
2022-10-09 11:02:49 Sun Azimuth 163.6, Sun Elevation 29.1
2022-10-10 11:02:16 Sun Azimuth 163.6, Sun Elevation 28.8
2022-10-11 11:01:46 Sun Azimuth 163.6, Sun Elevation 28.4
2022-10-12 11:01:15 Sun Azimuth 163.6, Sun Elevation 28.0
2022-10-13 11:00:44 Sun Azimuth 163.6, Sun Elevation 27.6
2022-10-14 11:00:14 Sun Azimuth 163.6, Sun Elevation 27.2
2022-10-15 10:59:45 Sun Azimuth 163.6, Sun Elevation 26.9
2022-10-16 10:59:16 Sun Azimuth 163.6, Sun Elevation 26.5
2022-10-17 10:58:48 Sun Azimuth 163.6, Sun Elevation 26.1
2022-10-18 10:58:21 Sun Azimuth 163.6, Sun Elevation 25.7
2022-10-19 10:57:55 Sun Azimuth 163.6, Sun Elevation 25.4
2022-10-20 10:57:30 Sun Azimuth 163.6, Sun Elevation 25.0
2022-10-21 10:57:04 Sun Azimuth 163.6, Sun Elevation 24.7
2022-10-22 10:56:40 Sun Azimuth 163.6, Sun Elevation 24.3
2022-10-23 10:56:16 Sun Azimuth 163.6, Sun Elevation 23.9
2022-10-24 10:55:53 Sun Azimuth 163.6, Sun Elevation 23.6
2022-10-25 10:55:31 Sun Azimuth 163.6, Sun Elevation 23.2
2022-10-26 10:55:10 Sun Azimuth 163.6, Sun Elevation 22.9
2022-10-27 10:54:50 Sun Azimuth 163.6, Sun Elevation 22.5
2022-10-28 10:54:29 Sun Azimuth 163.6, Sun Elevation 22.2
2022-10-29 10:54:11 Sun Azimuth 163.6, Sun Elevation 21.9
2022-10-30 10:53:53 Sun Azimuth 163.6, Sun Elevation 21.5
2022-10-31 10:53:37 Sun Azimuth 163.6, Sun Elevation 21.2


geojohnt@...
 

Hello Ernst,

Thanks for the further co-linearity comments.

One thing I do using David's WXtrack Solar Outage calculation is to update my WXtrack Keps more regularly during this period when using the readout.

It can make several seconds difference to the time of actual max co-linearity if your Keps were old.
Not that a few seconds will make much difference to the accuracy of the dish alignment(?)

Another thing I do when I think I have/had nearby trees/obstructions, during this period, is to calculate the Suns position 2 degrees and 1 degree east (and west) of the dish pointing and again watch for shadowing from nearby objects.

And for beginners setting up their dish a useful Arne van Belle tool is to run a piece of string from the middle of the LNB to the top centre of the dish. The shadow should cross the centre of dish at the time of max co-linearity and during this period.
This string shadow tool can be used at any time of the year to check azimuth - you just need that days solar crossing time at you location for E 10-A.

Regards,
John



 


-----Original Message-----
From: Ernst Lobsiger via groups.io <ernst.lobsiger@...>
To: MSG-1@groups.io
Sent: Thu, 29 Sep 2022 18:18
Subject: Re: [MSG-1] Aequinoctium problems (?)

On Thu, Sep 29, 2022 at 02:15 AM, <geojohnt@...> wrote:
 
Whilst talking about the 'moving sun,' we are all coming up to the autumn Sun/satellite/dish co-linearity season.
A good time to see if anything is partially shadowing your dish location.
 
Regards,
John
 
John and Robert,

in a long MSG-1 thread I tried to explain what the sun can do. See:

https://groups.io/g/MSG-1/message/33160

As a reminder at the time of solar outage your antenna should not
show shadows of nearby trees or buildings. Below table applies to
Robert Moore's antenna position. Solar outage is expected strongest
2022-10-12 11:01:15 and 2022-10-13 11:00:44. This is also the time
where Robert's system most probably looses lock. If your minimum SNR
is earlier or later then remember that the sun moves 15° per hour or
0.25° per minutes. This allows for an independent dish azimuth check.
Maybe Robert can tell us at what date/time the first shadows of his famous
trees reach the dish. It will show the free space under his antenna beam.

For those who have a Python environment with pyorbital installed
I attach the small bit that made the table below. It's very easy to
adapt for other antenna positions and later solar outages in 2023.


Cheers,
Ernst

P.S: All times are UTC of course ...


Robert Moore's Pos Longitude -3.244306 Latitude 53.279917
---------------------------------------------------------
Robert Moore's Dish Opt Azimuth 163.6, Opt Elevation 27.9
---------------------------------------------------------
2022-10-01 11:07:19 Sun Azimuth 163.6, Sun Elevation 32.3
2022-10-02 11:06:44 Sun Azimuth 163.6, Sun Elevation 31.9
2022-10-03 11:06:10 Sun Azimuth 163.6, Sun Elevation 31.5
2022-10-04 11:05:36 Sun Azimuth 163.6, Sun Elevation 31.1
2022-10-05 11:05:02 Sun Azimuth 163.6, Sun Elevation 30.7
2022-10-06 11:04:28 Sun Azimuth 163.6, Sun Elevation 30.3
2022-10-07 11:03:54 Sun Azimuth 163.6, Sun Elevation 29.9
2022-10-08 11:03:22 Sun Azimuth 163.6, Sun Elevation 29.5
2022-10-09 11:02:49 Sun Azimuth 163.6, Sun Elevation 29.1
2022-10-10 11:02:16 Sun Azimuth 163.6, Sun Elevation 28.8
2022-10-11 11:01:46 Sun Azimuth 163.6, Sun Elevation 28.4
2022-10-12 11:01:15 Sun Azimuth 163.6, Sun Elevation 28.0
2022-10-13 11:00:44 Sun Azimuth 163.6, Sun Elevation 27.6
2022-10-14 11:00:14 Sun Azimuth 163.6, Sun Elevation 27.2
2022-10-15 10:59:45 Sun Azimuth 163.6, Sun Elevation 26.9
2022-10-16 10:59:16 Sun Azimuth 163.6, Sun Elevation 26.5
2022-10-17 10:58:48 Sun Azimuth 163.6, Sun Elevation 26.1
2022-10-18 10:58:21 Sun Azimuth 163.6, Sun Elevation 25.7
2022-10-19 10:57:55 Sun Azimuth 163.6, Sun Elevation 25.4
2022-10-20 10:57:30 Sun Azimuth 163.6, Sun Elevation 25.0
2022-10-21 10:57:04 Sun Azimuth 163.6, Sun Elevation 24.7
2022-10-22 10:56:40 Sun Azimuth 163.6, Sun Elevation 24.3
2022-10-23 10:56:16 Sun Azimuth 163.6, Sun Elevation 23.9
2022-10-24 10:55:53 Sun Azimuth 163.6, Sun Elevation 23.6
2022-10-25 10:55:31 Sun Azimuth 163.6, Sun Elevation 23.2
2022-10-26 10:55:10 Sun Azimuth 163.6, Sun Elevation 22.9
2022-10-27 10:54:50 Sun Azimuth 163.6, Sun Elevation 22.5
2022-10-28 10:54:29 Sun Azimuth 163.6, Sun Elevation 22.2
2022-10-29 10:54:11 Sun Azimuth 163.6, Sun Elevation 21.9
2022-10-30 10:53:53 Sun Azimuth 163.6, Sun Elevation 21.5
2022-10-31 10:53:37 Sun Azimuth 163.6, Sun Elevation 21.2


Ernst Lobsiger
 

Dear All,

to better understand the light intrusion problem I wrote a little
Python script that does it all. My simple math usually predicts
the offending datetimes +/- 1 slot. I attach the oldest frames I
found on my systems when the problem started on top of the disc.

I think it's safe to say that Himawari-8 AHI is affected most,
GOES-XY ABI is affected a little bit less and Meteosat-NM seviri
does not show the effect (at least I have not found such frames).
We can expect the effect to be present 1 month before and after
Aequinoctium which is in fall and spring. So 4 month out of 12
show problems with e.g. 2-4 GOES-XY frames out of 144 per day.

Cheers,
Ernst



P.S.
I append the script that runs in (pytroll) plus an output sample


Ernst Lobsiger
 

All,

it's me once more. Searching for more accuracy the "equation_of_time" came to my mind. While GOES satellite slots now fitted well with my
predictions Himawari-8 had big offsets. It turned out that as a long standing *BUG* I have taken the Himawari-8 longitude as 144.7° while
it is actually 140.7°. This *BUG* is also found at least in the comments of respective python scripts of my PyTROLL/Satpy Starter Kit V 3.0.

A much improved prediction script using a NOAA approximation of the "equation_of_time" and a corrected Himawari-8 table is attached.
    
Cheers,
Ernst