ARLA/CLUSTER: Fw: ARLP018 Propagation de K7RA

ACViegas ct2ixq gmail.com
Sexta-Feira, 2 de Maio de 2014 - 21:42:51 WEST



-----Mensagem Original----- 
From: ARRL Web site 
Sent: Friday, May 02, 2014 9:33 PM 
To: ct2ixq  gmail.com 
Subject: ARLP018 Propagation de K7RA 

SB PROP @ ARL $ARLP018
ARLP018 Propagation de K7RA

ZCZC AP18
QST de W1AW  
Propagation Forecast Bulletin 18  ARLP018
>From Tad Cook, K7RA
Seattle, WA  May 2, 2014
To all radio amateurs 

SB PROP ARL ARLP018
ARLP018 Propagation de K7RA

Solar and geomagnetic activity slackened over the past week, with
average daily sunspot numbers declining from 202.7 to 73.4, and
average daily solar flux dropping from 160.4 to 122.6.

The latest outlook has solar flux at 125 on May 2, 130 on May 3, 135
on May 4 to 6, 140 on May 7 to 9, 150 on May 10 to 13, 145 on May
14, 140 on May 15 to 17, 135 on May 18, 130 on May 19 and 20, and
125 on May 21.  It then reaches a low of 115 on May 24 and 25, then
bounces back to 140 on May 31, and looking way, way out, 184 on June
9.  But June 9 is 37 days out, perhaps too far to rely on a forecast.

Predicted planetary A index is 5 on May 2, 8 on May 3 and 4, 5 on
May 5 to 13, 10 and 8 on May 14 and 15, and 5 on May 16 to 20, 10
and 8 on May 21 and 22, 5 on May 23 to 25, and 8 on May 26 and 27.

With April now over, we can look at the 3-month moving average of
daily sunspot numbers centered on March.  Even with the softening of
activity over the past week, the moving average is higher than ever
for this solar cycle, demonstrating that this second (or third) peak
is stronger than the first.

According to our moving average, the first peak was in the two
3-month periods centered on October and November 2011, when the
average daily sunspot numbers were 118.8 and 118.6.  There was
another slightly weaker peak centered on April and May of 2013, at
106.4 in both periods.

The averages following that period, centered on June 2013 through
March 2014 were 97.5, 85.6, 77.4, 91.2, 102.9, 123.7, 123.3, 138.5,
146.4 and 148.2.  This includes daily sunspot data from May 1, 2013
through April 30, 2014.  At the end of every month the most recent
month of data is added in, and the data from four months back is
dropped off, so we get a smoother set of numbers to look at.  When
you see those graphs of smoothed sunspot numbers, each point on the
graph represents a year of data.  The three-month moving average is a
bit more responsive.

F. K. Janda, OK1HH predicts quiet geomagnetic conditions May 2,
active to disturbed conditions May 3, quiet to unsettled May 4,
quiet to active May 5, quiet on May 6, quiet to unsettled May 7,
quiet May 8 and 9, active to disturbed May 10, quiet to active May
11 and 12, mostly quiet May 13 to 16, quiet to unsettled May 17 and
18, quiet to active May 19, mostly quiet May 20 and 21, quiet to
active May 22 and 23, mostly quiet May 24, quiet May 25 to 27, quiet
to unsettled May 28, and quiet to active May 29.

Jon Jones, N0JK notes that on May 1 there were interesting 6 meter
e-skip and TEP openings for both Europe and Eastern North America.
"It looks like the Es - TEP was open from 2000 UTC to 2230 UTC May
1.  I see VE1, VE9, W1, W2, W3, W8 and W9 spotted CE, LU and PY
during that period on 6 meters.  There were Es spotted from W1 to HI
and KP4, and VP9 to W4.  PY2AB worked W1AW/1 at 2046 UTC!!!

VE9DX AND PA2RU spotted the ZD8VHF/b at 2101 UTC.  Open from Europe
and the eastern USA and Canada to South Atlantic and America at the
same time."

Robert Miles, K9IL of Martin, Tennessee says for a 30 meter beacon
to check propagation, he likes to use a German RTTY station on
10.1005 MHz.  When the signals are S9, he knows that propagation to
Europe and sometimes long path to Japan should be good.

Bob Liddy, K8BL of Mentor, Ohio wrote on April 26:

"I had an experience with the 'Dead Band Syndrome' yesterday.

Since the beginning of the ARRL Centennial QSO Party, I've been on
the WARC Bands looking for QSO Points and trying to finish up WAS on
those 3 Bands.  I'll often tune across these Bands and try to get an
idea of what activity was taking place, if any.

On 4/25, I tuned across 12M and only heard a couple of weak signals,
so I decided to roll the dice and see what might happen.  At 2200Z, I
put out a CQ on 12M CW using 400W and my homebrew 12/17 rotatable
VEE dipole.

Little did I imagine that I'd be working WAC in less than 40 minutes
by calling CQ on a seemingly Dead Band!  At 2200Z, I was immediately
called by ZL4PLM for Oceania.  This was followed by NN4R at 2201 for
North America, at 2203 by G7BXU for Europe, at 2212 by JJ1IRS for
Asia, at 2220 by PY1CMT for South America and at 2239 by EA8YV for
Africa.  Also, HZ1BH/QRP called in at 2225Z for the other end of
Asia!

The solar numbers at the time were SFI-125, A-9 and K-2.  These
didn't look like great DX numbers, but the stations were there and
heard me well enough to answer my CQs.  Working WAC on a supposedly
'Dead Band' in less than 40 minutes by calling CQ definitely amazed
me!  It makes me wonder what activity would be like if folks would
give some Bands a try even though a quick scan turned up nothing."

Interesting, and good advice, Bob.

Dan Eskenazi, K7SS asked about the typical beam width of a coronal
mass ejection.  He was thinking about that monster flare on July 23,
2012 that fortunately missed us.  I wrote about it in the ARLP012
bulletin (see
http://www.arrl.org/w1aw-bulletins-archive/ARLP012/2014 ).

I suspected that CME beam widths vary quite a bit, and I put the
question to a couple of experts.  Carl Luetzelschwab, K9LA mentioned
an old Journal of Geophysical Research paper analyzing 241 CMEs from
1979 to 1981 (the peak of solar cycle 21) for angular span.  They
ranged from 2 degrees to over 120 degrees, with the average at 45
degrees, and median at 30 degrees.

Carl just sent me more information from an early 1980s analysis of
1209 CMEs, showing average angular width at 47 degrees, and median
at 44 degrees.

Robert Steenburgh, KA8JBY is a NOAA Space Weather Forecaster, and he
said that on July 23, 2012 there were three CMEs, "the largest had
an angular width of 190 degrees, followed by a 132 degree event."

He referred me to Cactus, a system for computer aided CME tracking
which analyzes data from LASCO, the "Large Angle and Spectrometric
Coronagraph Experiment."

See http://lasco-www.nrl.navy.mil/index.php .

For information on Cactus, see http://sidc.oma.be/cactus/ .

http://sidc.oma.be/cactus/catalog/LASCO/2_5_0/2012/07/latestCMEs.html
has July 2012 data on CMEs, and this has a great deal of detailed
information.  Anything in color on this page is a clickable link to
data on each CME, including a movie you can download.  You can go to
July 23 and watch the movies for each of the three CMEs on that
date.  As you will see, each of these were huge events.  These were
aimed directly at Earth's orbit, but Earth was not in position on
that date.  A few days difference (I don't know whether it was too
early or too late) and our fragile electrical grid and vulnerable
telecommunications infrastructure would have been in very deep
trouble.

For more information concerning radio propagation, see the ARRL
Technical Information Service at
http://arrl.org/propagation-of-rf-signals.  For an explanation of the
numbers used in this bulletin, see
http://arrl.org/the-sun-the-earth-the-ionosphere.  An archive of past
propagation bulletins is at
http://arrl.org/w1aw-bulletins-archive-propagation.  More good
information and tutorials on propagation are at http://k9la.us/.

Monthly propagation charts between four USA regions and twelve
overseas locations are at http://arrl.org/propagation.

Instructions for starting or ending email distribution of ARRL
bulletins are at http://arrl.org/bulletins.

Sunspot numbers for April 24 through 30 were 71, 73, 46, 84, 79, 80,
and 81, with a mean of 73.4.  10.7 cm flux was 130.1, 124.7, 120.7,
118.1, 120.8, 120.1, and 123.6, with a mean of 122.6.  Estimated
planetary A indices were 10, 9, 6, 4, 6, 6, and 18, with a mean of
8.4.  Estimated mid-latitude A indices were 8, 7, 8, 4, 7, 7, and
12, with a mean of 7.6.
NNNN
/EX



Mais informações acerca da lista CLUSTER