A compilation of propagation studies I have done in the past, as well as new ones on interesting topics in the future.
My studies are based on Mathematical principals used in Statistics. I am not a Physicist. I will use data from a number of sources, such as PropNET.org and WSPRNet.
Enjoy and I hope to share with you my curiosity of radio propagation.
73 Art Jackson KA5DWI/7
Hello everyone. After a few computer issues with my PropNET operation, I have been trying some other processes to see how they work on my older XP and Vista laptops and desktops. I have recently started up on WSPR (Weak Signal Propagation Reporter)
I received this Email today: Good evening Arthur, what a pleasure to see that you copied my WSPR beacon! I am running a very modest WSPR setup on this side: - Ultimate 2 kit (designed by Hans Summers, G0UPL) - with additional GPS receiver and active indoor-antenna - power output is ca. 200 mW into a - horizontal 28 MHz quarterwave on a Bamboo whip out of the kitchen window. My homepage shows some images: www.qsl.net/dl2wb Thank you very much for running your WSPR setup and uploading the spots! Let me know whether you would like to receive a QSL card that confirms your reception report (paper, PDF or other format). Best regards from Bavaria, Tom, DL2WB 200 mW into this antenna!!!!!!!
The 200 mW QRO rig.
Distance Per Watt
8921 kilometers 44605 km / 1 watt
5543 miles 22715 mi / 1 watt All I can say is WOW!! I captured this transmission with my 27 year old Yaesu FT-747GX fed with RG8X into a 42 year old 3-Element 10 Meter yagi. Caliche and copper must enhance signals here.... hi hi 73 Art Jackson KA5DWI/7
The 2011 10-Meter Spring/Summer Es Season By Art Jackson KA5DWI
The following documentation is a summary of the 2011 Spring/Summer Es season based upon PropNET captures and identified fragments received at the home QTH of Art Jackson, KA5DWI, maidenhead grid square EM12ju located near Fort Worth, Texas. All of the data analyzed and displayed in this summary was accumulated by the software PNP (formally known as PropNETPSK, written and maintained by Jeff Steinkamp, N7YG. All captures logged were of the mode PSK31 from other participants of PropNET (www.propnet.org) and others located near the 10-Meter operating frequency of 28.1188 MHz. No other frequency, bands or modes of communication were used in this synopsis. Data using PropNET captures of KA5DWI transmissions are used for comparison purposes only and captures of non-PropNET participants were used for probability measurements and non-Es propagation phenomena.
On occasion I will make references to past years in this summary. A 5-Year, followed by a 6-Year study of 10-Meter Es was written and published in 2010 and 2011 (Central States VHF Society Conference Proceedings). The data used in this analysis will be added to a 7-Year study in the future.
Any opinions and theories expressed in the summary are independent thoughts of my own and not to be construed from those members and officials of PropNET. My personal thanks go out to everyone involved and participated in PropNET for both the current and the past 8 years.
QTH - North Richland Hills, Texas (Grid Square EM12ju), a northeastern suburb of Fort Worth, Texas in a 35+ year old neighborhood surrounded by trees and in a creek valley. A ridge of hills between 1/3 and 1/2 mile away and no lower than 60 feet higher than the tops of my antennas runs from northeast to east of the QTH.
Primary Rig – 20+ year old Yaesu FT-747GX, running 10 watts.
Secondary Rig – Radio Shack HTX-10 (not used for transmitting, called Lurker mode in PropNET)
Primary – 3-Element Yagi @ 30 feet.
Secondary – 102 inch Whip Ground Plane @ 10 feet for severe weather conditions.
Transceiver Mode – Operated 24 hours a day for most the season.
Lurking Mode – Operated a few Sunday evenings and Monday mornings and while on vacation (7/07-7/13) or whenever severe weather was expected or occurring.
Data Statistics (April 20–August 28, 2011):
First identified partial or capture – 23:39 UTC on April 20, 2011 (KP3FT FK68qa)
Last identified partial or capture – 23:47 UTC on August 28, 2011 (KP3FT FK68qa)
Total Captures and identified partials (April 25 until August 14) – 9,453 PropNET Captures and Identified Partial/Fragments, plus 1,528 Non-PropNET Captures in which calls and grid square were identified.
Total KA5DWI Captured by other PropNET stations – 5,676 (Data not used in analysis)
The Year’s Challenges:
1. Computer – I continued to run PNP on a Windows XP Home P4 1.8-GHz 648-Meg RAM Dell Desktop. It did not run as well as it had in past years. The best I could determine was that running the newest version of PNP along with a memory and CPU hungry antivirus software were totally incompatible. On the third day of vacation, PNP failed to connect to LiveX, ran and read very poorly. I had a few short-lived power outages early in the season.
2. Software - I used PNP Version 18.104.22.168 for the season on the XP Desktop and at times the XP Pro 3 Gig Dell Desktop. After much pleasure with the 4.3.X.X version, the new version software was a disappointment. It may have been due to a CPU hog antivirus program, but at the same time changes in the settings of both the software, and what appeared to be quirkiness with LiveX added frustration to a disappointing year in Es conditions. I had to manually recalculate several days of the measured season due to several minor incidents. The changes made to the settings and operations for both the software and LixeX were to enhance the Robot operations of the software, and in my opinion adversely affected the collection of propagation data for my purposes. I had to audit this year’s data much more than I had in past ones.
3. Weather – The weather was rather violent to begin the spring. We had 3 hailstorms in April. Beginning in July, one of the longest and worst heat-waves that have happened in history began. The lack of wind caused very high low temperatures, but helped keep down static.
4. Antennas – One end of the off-centered reflector on my 3-Element Yagi bent down about 30 degrees, but did not seem to affect overall performance. It is time to finally repair it.
Transceive Mode - Yaesu FT-747GX @ 10 watts into 3-Element Yagi pointed 60-110 degrees (80 usual) azimuth.
Lurker Mode – Radio Shack HTX-10 into the 3-Element Yagi or the vertical in bad weather.
Results – 2011 Spring/Summer 10-Meter Es Season @ KA5DWI EM12ju
By the end of 2010, I had compiled the data into a 6-year study and prepared it for publication. This year I was able to present it at the Central States VHF Society Conference in Irving, Texas. With the increase in solar activity, I felt that a 7th year of participation would be interesting.
Daily Figures from the Beginning to End of the Es Season A Year That Was a Paradox
The Spring/Summer Es season of 2010 was one of the most “on average” of the 6 years of the study. The capture trend-line for the season was amazingly “right on”. The trend line had a slight shift to the left, indicating an early active season. Solar activity continued to increase during the fall and winter of 2010-11. Solar flux had been near 130 prior to this year’s season and at the beginning was around 110. As in 2010, this year’s season began very strong very early. The season seemed to be off to an exciting beginning, but it was short-lived. The expected peak that is annually experienced on 5/09 was weaker than normal and this trend continued until the second week of June.
The deficit between this year and the average of the prior 6 was as much as 39% of normal. The peak of the trend occurred on 7/17. At this point, the season appeared to be the worst measured. Out of the blue, things picked up, the deficit declined and the season finished strong. The 2011 Spring/Summer season was only the second worst in the 7 years measured.
The first chart shows how the daily activity of 2011 compared to the prior 6 years of the Es study that I have compiled. The “Regression” line is a best fit for the 6 years of data. The “Poly (2011Total)” is the polynomial line of best fit for this year. Notice that from the start of the season afterwards that activity is well below the average levels experienced for past years and the regression line. Good or at least normal activity levels did not occur until mid-June. The best part of the season was actually from mid-July to early August and is strongly evident by the trend-line (black dashes) and peaked about 2 weeks after the summer solstice.
It was all about the strength of the openings this season. The number of active hours once again (3 years running) were better than average. What is most obvious in this analysis was that the beginning and ending of the season was much better than past years. The peak of the season was very close to past experiences.
In most other years, the first 2 weeks of the season are slow. But like last year, it started off fairly active. Most days of the season, active hours were especially good and about 3 hours better than average. This year, peak daily volumes failed to be above normal and lows (valleys) were quite common. The peaks in activity were generally one day events. There were very few multiple highly active days and is prime reason for the short-fall. July 17 was the turning point and the salvation of a bad season. The seasonal median (midpoint) for activity was on July 2, much later than other years. The seasonal median (midpoint) for active hours was on June 29, also later than other years.
Despite lower results in activity, there was at least one PNP capture each day from May 3 until August 19. This was consistent with other years measured. The active hour measurement clearly shows the late peak of the season. Although better than average years, there was only one excellent day in which there was at least 21 hours was active.
3 Day Averages:
The “3-Day Average” is an average of daily captures two days prior to the day measured and the day measured. It shows about 16 short-lived peak periods for activity and about 12 for active hours. The slow growth of the season is quite clear as well as the later peak. The most highly active period again was the 3rd and 4th week of July and not the Summer Solstice. There were never any extremely high activity periods, nor with high active hours.
6 Day Average:
The 6-Day average is the average number of captures 5 days prior to and including the day measured. It shows approximately 11 peak activity days during the season (consistent with past years). This also shows that the season was much more inactive, along with the later peak. The daily hour open hour rate did not peak as sharply as it had in past years, but remained higher throughout the entire season.
After 3 years of measurements, statistics show that the Spring/Summer Es season begins 8 weeks prior to the Solstice and ends 10 weeks after. This year, the charts look like reflections of last year’s. The first half of the season was slow, the second half improved, and the exact opposite of what occurred last year.
In 7 years of measuring, the following chart confirms each year that the most active period for Es occurs mid-mornings, followed by a less active second peak during the late afternoon approaching sunset. This year the second peak was less pronounced and twilight hour activity was non-existent.
This chart represents the actual number of captures for each hour of the day from April 25 to August 28. It is shown as a 3-hour average to centralize the activity to a specific hour. For example, the 6 AM (6) local hour is the average of 5, 6 and 7 AM. The dual-peaked diurnal (daytime) pattern matched what occurred from 2006 through 2010.
Also, the number of captures that occur each open hour during the season closely duplicates the results of the capture activity chart above. These trends occur year after year. The more likely an opening is to occur, the better the quality of the opening. This year’s decline in activity was obvious on this chart as well. The only hour that improved this year was 7 AM. Non-PropNET Captures and Captures of This Station:
In the 4th Quarter of 2009, it was decided to change the operating frequency from 28.131 to 28.1188 MHz. Moving below .120 allowed us to capture some of the activity on that frequency. The routine setup by Dave Donnelly, KF6XA accepts user non-PropNET captures via LiveX, pulls QTH data from QRZ.com and shows them as NPN captures, as well as adds them to the database. The routine is very selective. “CQ Contest” and “CQ 10-10” does not work.
After a hot first season, this season’s total NPN activity was down as well, but impressive nonetheless. Only 2,200 captures were added. I captured 506 different calls (down 200+), 208 grid squares (only down 6) and 27 DXCC entities (way up!). I had 19 transatlantic Non-PropNET captures (also up).
I have added this data to Open Hour, Distance and Directional statistics. Total capture activity charts remain PropNET data only. The hourly data rates were not strongly affected (about ¾ hour).
Throughout the season, non-PropNET captures favored morning times. There was no discernable dual peak. Again, many felt that Es were evening phenomena. It was due to most Hams operating in the evening not the morning.
The following charts are the PropNET captures, non-PropNET captures, and PropNET captures of KA5DWI during the season. The dual peak-diurnal pattern is evident in these charts and is much sharper than past years. Captures of my station were also much lower this year by 30%. Results by Distance
As in all prior years, a vast majority of all 10-Meter PropNET captures occur at this QTH between 1251 and 1750 kilometers (770-1070 miles). This area encompasses Florida, Georgia, the Carolinas, onward to Michigan and the Southwest U.S. Much of thee captures longer than 2250 kilometers are Non-PropNET, except for KP3FT, Jeff in Puerto Rico. We lost Hawaii and gained back Puerto Rico.
New this year was a significant increase in Trans-Equatorial propagation over 5000 kilometers. Solar flux is still not high enough to support F2, and it is my opinion that Es are still a part of this propagation. From a presentation given by James Kennedy K6MIO/KH6 this July at the Central States VHF Society Conference, he associates much of our long distance multi-hop DX events in the Summer or Winter to a combination of Es and F2 influences. Distances by Hour of the Day:
As mentioned before, the vast majority of PropNET captures at this QTH occur near the 1251-1750 kilometer mark. The dual-peaked diurnal pattern is evident at these distances and was similar to past years. Captures between 751-1250 kilometers also peaked in the late morning hours. Both distance groups declined this year.
Once DX occurs past 1750 kilometers, it becomes apparent that it strongly favored late afternoon to evening peaks. I have classified trans-equatorial captures separately in the next chart as it deserves special attention. Due to the loss of a PNP station in Hawaii and a gain of one in Puerto Rico, these distances shifted in activity this season.
These are new charts this season. It is the Trans-Equatorial propagation experienced this season compared to last year. Notice the significant increase from 2010.
The second chart is in 2,500 kilometer increments of the “TE” propagation. The shorter distance may be more Es than F2, the farther distances are more F2 like. In order by distance; northern South America, southern South America, and last New Zealand-Australia.
Hourly Directional Trends:
A way to determine what directions are best by the time of day is to separate all captures into directional half-groups.
The next chart divides the captures easterly (0-179.9 degrees azimuth) and those westerly (180-359.9 degrees azimuth). Each group is broken into the hourly percent of the daily total. This method eliminates skewing due to the differences in volume by each directional group. As noted in this year’s chart, easterly directions are best as the sun rises. Westerly opportunities peak 1 hour after easterly ones. The westerly second peak occurs quicker. The decline is slower after sunset to the west.
As then sun rises, northern activity rises quickly. By Noon, southerly ones catch up and continued to rise and drop quickly after sunset. Northern paths strongly favored morning hours, southerly ones favored afternoon.
Outside Affects on Es
I never want to say that I have all the answer to Es, but I am always curious to find what might affect them. I have determined that meteors do strongly influence them, and I also have identified that solar events have an effect.
This year was interesting again. In 2010 activity dropped significantly when solar flux began to rise. This year it appeared to do the opposite. As solar activity was high at the beginning of the season, PNP activity was above normal. When solar activity decreased, activity was way below normal. It never approached normal, or better than until solar activity increased in the later stages of the season.
There were slight increases in the K and A-Indices during rises in solar flux. There were no major long term extremes.
It goes to show you that it is hard to pinpoint solar influences on Es propagation.
Daily Capture Maps of KA5DWI EM12ju
There were very few “big” days this year. There were few good ones. I captured 50 non-local PropNET stations this year, down 4 from last year. Here are maps of outstanding days and overall coverage.
August 4, 2011. Best day for PNP & NPN captures.
June 12, 2011. The second best day this season.
July 27, 2011. The third best day this season
2011 Spring/Summer Es Season 10-Meter Grid Squares Captured
I looked forward to see how a rise in the solar cycle affected Es. Overall it was a below average year and if not for the latter half surge, it would have been the worst. It appeared that the increase in solar flux may have had a southerly influence and put a lid on east-west propagation. Did the rise in the solar cycle deter Es?? The worst season was at the solar cycle minimum!! There is still a mystery for someone to figure out. Good luck.
73s, Art Jackson KA5DWI 2011 KA5DWI Capture List
Once again, Rich-WD4RBX, Bob-WB8ILI, Jesse-K4EPS, Jerry-K8VGL, and Robert-K4RKM top the list.
How about some 2-Meter Tropo.
I am a longtime 2-Meter SSB operator (early 80's). I have over 200 Grid squares and 43 states using CW and SSB.
I noticed last night that there was a large Upper-Air High Pressure ridge over us, so I unplugged the PropNET rig and plugged in the FT726R and UI-View to see how good the Tropo was. Not too bad. This is the time of the year to be on watch.
The maps tell the tale.
First is a normal map.
This is the early morning map with the Tropo well evident.
Here is a recap of what you can expect during this year's Spring/Summer Es Season.
Es appear around April 21th to the 25th now and then. The first full scale opening appears near May 9-11. By the end of May, Es are in full swing and will not decline until mid-August. Activity-wise, it peaks very near the Summer Solstice. It rises quickly and takes its time end.
This chart represents the daily active hours you can expect during the season. For about 6 weeks, beginning the third week of May you can expect as much as 12 hours of Es each day. Three days (6/13, 6/18 and 7/4) average 16 hours.
The next 4 charts show the probability of working 10-Meter Es each week of the season. The dates listed are the beginning of each week.
Es season will begin in earnest by the end of April. It will rise steadily until the summer solstice, then slowly decline towards the end of August. From the 2nd week of May through mid-August, conditions are excellent. All of the following charts are based on 7 years collecting 10-Meter PSK31 signals on PropNET here in North Texas.
This chart represents what portion of the day will have Es activity.
These are hourly probabilities in 3-week segments from April 25 - August 28.
The following charts a single week hourly probabilities throughout the entire season.
Your results may vary, but understand that after 7 years I have discovered that Es have their own schedule and will follow very distinctive patterns. Do not look towards severe weather as a cause, only as a possible, but not probable localized enhancement. The only outside affect I have found are meteor showers.
The following charts and data are based on 10-Meter PropNET and Non-PropNET PSK31 captures of North American PropNET participants from April 25 through August 15 2010 - 2013.
A specific formula was used to distinguish Sporadic Es propagation from F2 or Trans-Equatorial propagation. Only PropNET participant data from North America and Hawaii was used in the study. Any questionable data was removed to the best of my ability.
This information is for the use Amateur Radio public. Please reference PropNET Studies as the source.
Narrative will be added as time permits.
Enjoy. This stuff is fascinating and I believe it answers many questions about Sporadic Es propagation from year to year.
After one of the longest lulls in the Solar cycle ever recorded, the sun finally started to get more active in 2011. Unfortunately the increased activity in Solar Flux peaked in 2012. Many look for a second peak in 2013-14.
Once the 2012 Es season ended, solar flux hit levels approaching 180. By the spring of 2012 it had begun a decline.
On the other hand, the actual number of sunspots has not declined that much.
The number of new sunspot regions have actually grown year to year since 2010.
Despite the growth in new regions, these do not last very long.
Their ability to produce intense flares has been impeded.
Solar activity actually has a steady increase since 2010. Due to mistiming of the solar poles, there has been a less extreme rise in solar flux. New sunspot regions form, create instability but do not produce the high MUF to support F2 propagation.
Es ACTIVITY BY DAY:
In 4 years, North American PropNET participants captured almost 300,000 packets from other members or non-participants transmitting specific algorithms.
These are the average daily captures by day six weeks prior to and after the Summer Solstice. The trend line produced by the data shows that peak activity occurs at or near the solstice.
The overall trend was highly affected by the volumes of 2010 and 2011.
2010 - The last year of low solar activity.
2011 - Solar activity picks up. The season had a lull early, and then became active toward the latter half.
2012 - Solar activity has peaked. What activity there is occurs in the first half with a large lull during the solstice period.
2013 - Solar activity remained high. The trend line was almost a duplicate of 2012.
Activity has drastically deceased since 2010.
The number of open hours each season has declined, but at a slower level.
Es ACTIVITY BY HOUR:
As has been seen in most years, Es activity has a dual-peaked diurnal. In 2010, it was a morning peak. As solar activity increased, it became afternoon active.
11:00 UTC represents sunrise on the East Coast of North America. Activity peaks 2 hours prior to noon in the center of the continent and again 2 hours before the sun begins to set.
When solar activity was low, the morning hours are most active. As solar activity increases, it shifts to afternoon.
This chart is based on an equal percentage breakdown. The highest hourly percentage of day occurred late afternoon when solar activity was its highest.
As solar activity increased, active hours of Es declined.
As the season begins if an hour opens, intensity increases until the solstice and steadily declines afterward.
This chart shows that once solar activity increases, intensity drops of at beginning and end of the season.
As noted on the chart, low solar activity has a great influence.
F2 and Trans-Equitorial:
As solar activity rises so does TEP open. This was very prevalent in the last 3 years.
Population Changes in 4 Years:
PropNET captures were about 5 times greater than Non-PropNET captures.
Changes in the mixture of PropNET to Non-PropNET changes since 2010 have occurred, but are very slight overall. For the past two seasons, PropNET to Non-PropNET are at 80% to 20%. The only real changes that were noticeable were around year 2011 and were no more than 5%.
The decline became more extreme after 2011
Non-PropNET declines more evenly in 4 years.
The mix of Non-PropNET to PropNET does not change much.
The overall decrease since 2010 for PropNET captures is about 3.5% greater than Non-PropNET..
Comparing North To South:
I wondered if what happens in the southern half of the continent is different than the northern half. I split the continent at the grid squares FMx7, EMx7, DMx7 and CMx7 to the north and measured the data by hour by year. To my surprise there is little difference, other than there is more activity in the south.
There are more active southern participants in PropNET.
When the totals are broken down by percentage of day, the north sees more Noon activity, the south sees more sunrise and sunset activity. Not much difference, but a little more noticeable the last couple of years.
More Directional Comparisons:
So how does Es propagation differ by direction. Once again consistant North to South, East to West. Only Northeast to Southwest, Northwest to Southeast show any differences.
There are little differences in these directions.
Most activity is to the North.
Only Eastward conditions are more volumes found.
Northeast and Southwest show morning activity.
Northwest and Southeast show afternoon and evening activity.
As was expected, solar activity has a profound affect on Es propagation after the minimum has occurred. The peak should occur by the winter. It will be interesting to see how these numbers change the next few years. I think it will be on the way up.